Rene te Witt bw.

indd 1 24-11-11 10:35

 Printing of this thesis was financially supported by:

ISBN: 978-94-6169-173-6

© R. te Witt, 2011
All rights reserved. No part of this thesis may be reproduced or transmitted in any form or
by means without prior permission of the author, or where appropriate, the publisher of the
article.

Layout and printing: Optima Grafische Communicatie, Rotterdam, The Netherlands

Coverdesign with permission of Reid, Geleijnse & van Tol

Rene te Witt bw.indd 2 24-11-11 10:35

 Clinical Microbiological Diagnostics 2.0
Klinisch microbiologische diagnostiek 2.0

Proefschrift

ter verkrijging van de graad van doctor aan de

Erasmus Universiteit Rotterdam
op gezag van de
rector magnificus

Prof.dr. H.G. Schmidt

en volgens besluit van het College voor Promoties.

De openbare verdediging zal plaatsvinden op

vrijdag 13 januari 2012 om 13.30 uur

René te Witt
geboren te Rotterdam

Rene te Witt bw.indd 3 24-11-11 10:35

 Promotiecommissie

Promotor: Prof.dr.dr. A. van Belkum

Overige leden: Prof.dr. H.A. Verbrugh

Prof.dr. P.H.M. Savelkoul
Prof.dr. H.H. Hooijkaas

Copromotor: Dr. W.B. van Leeuwen

Rene te Witt bw.indd 4 24-11-11 10:35

 Voor Dirk-Jan (20-12-1974 – 22-11-2008)

Rene te Witt bw.indd 5 24-11-11 10:35

Rene te Witt bw.indd 6 24-11-11 10:35
 Contents
Introduction 9
Chapter 1 General introduction and outline of the thesis. 11
Chapter 2 Specific diagnostic tests for atypical respiratory tract pathogens. 33

Biomarkers 57
Chapter 3.1 Neopterin and procalcitonin are suitable biomarkers for exclusion 59
of severe Plasmodium falciparum disease at the initial clinical assessment of
travellers with imported malaria.
Chapter 3.2 Procalcitonin- and neopterin levels do not accurately distinguish 71
bacterial from viral infections in ill-returned febrile travellers.

Culture 79
Chapter 4 In vitro evaluation of the performance of Granada selective 81
enrichment broth for the detection of group B streptococcal colonization.

Nucleic acid amplification 93

Chapter 5 Molecular diagnostics and genotyping of methicillin-resistant 95
Staphylococcus aureus (MRSA): an update.

Bacterial typing 107

Chapter 6 Comparison of the DiversiLab system, pulsed-field gel 109
electrophoresis and multi-locus sequence typing for the characterization of
epidemic reference MRSA strains.
Chapter 7 Good performance of the SpectraCellRA system for typing of 119
methicillin-resistant Staphylococcus aureus (MRSA).

Quality control 131

Chapter 8 External Quality Assessment of molecular diagnostics and 133
genotyping of methicillin-resistant Staphylococcus aureus.

Discussion and future perspectives 143

Chapter 9 New clinical microbiological diagnostics. 145
Chapter 10 Summarizing discussion and future perspectives. 167
Chapter 11 Nederlandse samenvatting. 181

Rene te Witt bw.indd 7 24-11-11 10:35

 Appendices 195
Dankwoord 197
Curriculum Vitae 204
List of publications 205
PhD portfolio 206

Rene te Witt bw.indd 8 24-11-11 10:35

 Introduction

Rene te Witt bw.indd 9 24-11-11 10:35

Rene te Witt bw.indd 10 24-11-11 10:35
 Chapter 1
General introduction
and outline of the thesis

René te Witt

Rene te Witt bw.indd 11 24-11-11 10:35

Rene te Witt bw.indd 12 24-11-11 10:35
 General introduction and outline of the thesis

Historical aspects

Chapter 1
Life has changed since the Dutch botanist Anthonie van Leeuwenhoek (1632-1723) revealed
the diversity and ubiquity of the microbial world through the discovery of microscopy.
Van Leeuwenhoek can be considered to be the first genuine microbiologist. Microscopic
evidence provided support for the emerging germ theory of disease in the 19th century. In
the 1880s, Robert Koch defined his postulates for determining whether or not a microorgan-
ism is the etiological agens of a disease. Since the 19th century, advances in knowledge have
included the discovery of viruses, chlamydiae, mycoplasmata and rickettsiae as new classes
of microorganisms that cannot (yet) be grown in pure culture, but require living cells for re-
production. The spectrum of bacterial, fungal and protozoan pathogens has been expanding
with improved culture techniques and the development of advanced imaging techniques.
However, the most revolutionary advance in biomedical science since Van Leeuwenhoek, is
due to the discovery of nucleic acids in 1871 by Miescher, which lead to the discovery of DNA
as the source of genetic information and as the basis for characterization of an organism in
1953 by Watson, Crick and Wilkins (1-2).

Introduction to infection disease diagnostics

Today, infectious diseases remain a global burden to human health. Infectious diseases can
be classically omnipresent (tuberculosis, cholera, malaria and gastro-enteritis) or display
annual epidemics (norovirus, influenza and seasonal colds). They can present as incidental
emerging infectious diseases (avian influenza, SARS and hemorrhagic fevers) or they can
be caused by emerging antibiotic-resistant pathogens (methicillin-resistant Staphylococcus
aureus (MRSA), Extended Spectrum Beta-Lactamase (ESBL) producers, carbapenem-resistant
bacteria, multidrug-resistant Acinetobacter species and other multidrug resistant bacteria).
And, finally, infectious diseases can present as pandemic infections (AIDS, the recent H1N1
outbreak and the already mentioned tuberculosis) (For examples, see Figure 1).
Current diagnosis of most bacterial infections is still based on conventional microscopy
and culture-based methods. Substantial advantages of conventional diagnostics are high-
throughput and ease-of-use. Essentially, laboratories can process direct and easy tests on hun-
dreds to thousands of samples on a daily basis, using inexpensive culture media and simple
techniques. This strategy also provides bacterial isolates, which can be further characterized if
needed, i.e. more detailed identification of the species, analysis of their antibiotic susceptibil-
ity profiles and/or epidemiological type for outbreak analysis and surveillance studies.
However, this strategy has limitations. Culture has limited sensitivity and is time-consum-
ing. Many new diagnostic techniques and strategies have been developed during the past
two decades, all with the objective to save time and to improve accuracy.

13

Rene te Witt bw.indd 13 24-11-11 10:35

 Chapter 1

Figure 1. Incidence rates of examples of infectious diseases.

A. Worldwide TBC incidence rate, 2010 (Data from the World Health Organization (WHO) (www.who.int).
B. Worldwide Cholera incidence rate, 2009 (Data from the World Health Organization (WHO) (www.who.int).
C. Worldwide HIV incidence rate, 2006 (Data from the World Health Organization (WHO) (www.who.int).
D. Worldwide malaria incidence rate, 2004 (Data from the World Health Organization (WHO) (www.who.int).
E. European MRSA incidence rate, 2007 (Data from the European Antimicrobial Resistance Surveillance Network (EARS-NET)
(www.ecdc.europa.eu/en/activities/surveillance/EARS-NET).

14

Rene te Witt bw.indd 14 24-11-11 10:35

 General introduction and outline of the thesis

Biomarkers

Chapter 1
While fever in a patient can presage an infection that can be rapidly progressive and fatal
(sepsis, meningitis), it may also be a manifestation of a self-limiting, more trivial infection (Sal-
monellosis) or even sterile inflammation (fever of unknown origin, allergy etc.). Nevertheless,
the initial clinical evaluation should focus on infections that can be treated, may be transmis-
sible and that may cause serious sequelae or even death. The clinician must decide in a short
time whether the infection is likely to be caused by bacteria, viruses or parasites and whether
the patient should be admitted for intensified empirical treatment with antibiotics. Results
of routine (microbiological) laboratory tests may provide additional clues for the diagnosis.
Traditionally, an elevated number of circulating leukocytes (leukocytosis) is considered to be
suggestive of bacterial infection, whereas a decreased number of leukocytes (leukopenia)
may suggest viral infection. Unfortunately, a number of bacterial infections, such as uncom-
plicated typhoid fever, brucellosis and rickettsial infections are associated with a normal to
low white blood cell count. As a consequence, the clinician cannot completely rely on these
traditional parameters for decision making in the acute care setting. In order to reduce em-
pirical, broad-spectrum antibiotic treatment, a rapid identification of the cause of the clinical
symptoms is needed. Concerns about the accuracy of microbiological diagnostic techniques
have led to the hypothesis that the use of biologic markers could improve fast recognition of
a true infection in patients in an early state, thereby facilitating decisions on patient manage-
ment. Host-specific biomarkers can provide both diagnostic and prognostic information for
individual patient care. Similarly, biomarkers can indicate (ab)normal responses to therapeu-
tic intervention. Biomarkers can also help to advance basic knowledge of pathogenesis.
Several studies have been performed on the relevance of biomarkers in the distinction
between bacterial and viral infections prior to or on admission of a patient to the hospital. In
certain clinical studies procalcitonin (PCT) demonstrated to be good predictive biomarker for
bacterial infection (sepsis and pneumonia) (3-6), whereas neopterin showed to be predictive
for viral infection (7). Moreover, PCT was proven to be an accurate monitor for the efficacy of
antibiotic treatment (6, 8).
PCT is the propeptide of calcitonin, which plays an important role in the regulation of bone
calcium and phosphate metabolism (9). A specific protease cleaves PCT to calcitonin (10).
Normally all the procalcitonin is cleaved and none is released into the bloodstream. PCT lev-
els are therefore very low (<0.1 ng/ml) in healthy humans (11). Inducers for PCT synthesis are
inflammatory cytokines, Interleukin-1 (IL-1) and Tumour Necrosis Factor-α (TNF-α), but also
bacterial membrane structures or bacterial cell wall products such as lipopolysaccharides
(LPS) or peptidoglycans (PG) (12). Bacterial infections may induce PCT in measurable con-
centrations in serum only after triggering an inflammatory response (either local or systemic)
and this response should have reached a certain severity or threshold. Consequently, local
bacterial infections may not always induce PCT. Therefore, it is not possible to reliably use

15

Rene te Witt bw.indd 15 24-11-11 10:35

 Chapter 1

PCT measurement to exclude any type of bacterial infection (11, 13). Still, PCT has a number
of properties which support its use in clinical routine diagnostics. For example, PCT levels in
blood samples are stable. It can be tested in routinely collected blood samples which can
be transported to the laboratory in the usual way, without suffering any significant drop in
concentration of PCT (14).
The production of neopterin is induced mainly by the inflammatory factor interferon-
gamma (INF-γ) and is co-stimulated by TNF-α and bacterial endotoxins (15-16). In this way,
neopterin may be of clinical use as a marker of pro-inflammation and bacterial infection
(17-18).
Soluble triggering receptor expressed on myeloid cells-1 (sTREM-1) may be another
interesting infectious disease biomarker. sTREM-1 is a cell-surface receptor identified on
neutrophils and a subset of monocytes (19). sTREM-1 may be an appropriate candidate as an
infection biomarker as concentrations were elevated in patients with bacterial sepsis (20-21).
Furthermore, sTREM-1 can possibly predict mortality as concentrations significantly declined
in survivors as compared to non-survivors of bacteraemia (21). Of course, there are many
more biomarkers that may be of clinical use as a marker of inflammation and bacterial, viral or
parasite infection. Most studies focus on tuberculosis (22-23), sepsis or bacteraemia (24-25),
respiratory tract infections (26-28) or fungal (mainly Aspergillus spp.) infections (29).

Microscopy

Detection of human pathogens by direct microscopy of the clinical specimen is the most
ancient, but still highly relevant, diagnostic method in medical microbiology. Many different
cell wall structure-specific staining procedures are available (Gram, Ziehl-Neelsen etc.), pro-
viding an initial classification of a potential pathogen together with morphology (shape and
positioning). However, in the majority of cases, final identification still relies on phenotypic
(biochemical and serological) properties of the bacterium after culturing on appropriate
media.

Culture

A conventional and commonly used diagnostic technique in clinical settings is culture on and
in a variety of solid and liquid media. These media can be universal (supporting the growth of
most clinically relevant bacterial species e.g. blood agar, chocolate agar, Müller Hinton broth),
specific (selective agar or broth specifically for Gram-positive cocci or anaerobic bacteria etc.)
or even species-specific (i.e. MRSA enrichment broth, chromogenic agars) via the addition
of specific substrates, growth inhibitors and/or certain antibiotics. In the last decades, many

16

Rene te Witt bw.indd 16 24-11-11 10:35

 General introduction and outline of the thesis

different commercial chromogenic agars have been developed for specific pathogens, such

Chapter 1
as diagnostic agar for Candida (30) and Staphylococcus aureus ID agar or CHROMagar MRSA
for MRSA identification directly from clinical samples (31-33).
The introduction of automated laboratory systems for culture, e.g. the blood culture sys-
tems Bactec (BD Diagnostics, Erembodegem, Belgium) and BacT/Alert (bioMérieux, Marcy
l’Etoile, France), has increased the quality and efficiency of clinical diagnostic laboratories.
Culture methods require time to allow bacteria to grow, generally 24-72 hours. This results
in a delay in more specified anti-microbial treatment of the patient. For fastidious or slow-
growing bacteria, such as Mycobacterium tuberculosis, the diagnostic time is lengthened even
further to several weeks, while bacteria that cannot be grown on conventional culture media
and under conventional conditions, obviously, will remain undetected. Furthermore, logistic
factors may influence bacterial growth. Samples need to be transported to the laboratory
prior to analysis. During transport, samples may deteriorate or change e.g. due to changes in
temperature, enzymatic effects and/or lysis of cells. This may have dramatic effects on bacte-
rial quality (specificity) and quantity (sensitivity) in the clinical sample. Moreover, transport
time will further delay the diagnostic procedure.
An example of a bacterial pathogen that is mainly detected by culture is Group B strepto-
coccus (GBS). GBS has been considered as a human pathogen since 1938. During the course
of pregnancy and the postpartum period, GBS may cause a variety of serious infections in
both mother and neonate (34-35). Neonatal early-onset GBS disease (GBS-EOD) presents
in the first week of life. Mortality is high among preterm infants, with case-fatality rates of
approximately 20% and as high as 30% among those <33 weeks’ gestation, compared with
2-3% among full-term infants (34). In most cases neonates acquire GBS during delivery from
mothers colonized with GBS in the rectovaginal tract (up to 35%) (36-38). Approximately 1%
of neonates born from colonized mothers develop GBS-EOD and up to 40% of the surviving
neonates suffer serious sequelae, such as mental retardation or seizures (34).
For diagnosis of GBS, several culture techniques have been developed. These techniques
focus on the sampling site, increasing of the sensitivity, the moment of sampling and the
culture technique. However, little attention has been paid to possible consequences of
transport conditions with respect to time, temperature and transport medium. Revised CDC
guidelines state specifically that GBS isolates remain viable in Amies transport medium for
several days at room temperature. However, a decline of up to 30% for the recovery of GBS
isolates is observed over a period of 1-4 days, particularly at elevated temperatures (39-40).
Even when appropriate transport media are used, the sensitivity of culture is most optimal
when the specimen is stored at 4°C before culture and processed within 24 hours after sam-
pling (34, 39-43).
More sensitive methods than the currently available and recommended transport and
culture methods would improve the effectiveness of the screening-based approach and
would lead to improved patient management. A new transport and enrichment broth, called

17

Rene te Witt bw.indd 17 24-11-11 10:35

 Chapter 1

Granada Tube (GT) broth (bioMérieux) was introduced recently. An orange pigment is pro-
duced in this broth in the presence of GBS (44-45). Means such as this medium may help to
further improve diagnostic accuracy and the ultimate prevention of GBS disease in the target
population.

Serology

In diagnostic laboratories, microscopy and culture are usually supplemented with serological
techniques for (in)direct diagnostics of infections, by detecting specific antibodies or anti-
gens. Both methods can speed up detection and diagnosis of fastidious or non-culturable mi-
croorganisms. Serological techniques may be antibody-based, e.g. Enzyme Linked Immuno
Sorbent Assay (ELISA) or specifically geared towards the detection of bacterial antigens.
However, there are also numerous drawbacks. Serological results can be unreliable due to
cross-reactivity. Furthermore, because of the frequent need of a convalescent serum sample
after 1 to 4 weeks, results are available too late to understand infection dynamics and to have
an impact on immediate patient management.
Optimal serological testing for individual patient care depends on several factors: age of
the patient, timing of serum collection, whether paired (acute and convalescent) sera are
available for confirmation of seroconversion, and, not in the least, availability of appropriate
equipment and experience of the laboratory personnel. Despite these potential pitfalls, com-
mercial antigen detection and serological assays represent a huge market and are globally
used for diagnosis of viral infections and fastidious or non-culturable bacterial species, such
as Legionella spp. and Borrelia burgdorferi. Especially since the introduction of automated,
high-throughput antigen and/or antibody detection systems such as VIDAS (bioMérieux)
and Luminex (Luminex, Austin, Texas, USA) and robotic workstations such as Hamilton STAR
(Hamilton, Reno, Nevada, USA), serological testing has become more accessible (cheaper,
faster, easier and high-throughput).

Identification and antimicrobial susceptibility testing

After cultivation of a microorganism, the next step comprises of identification. Most clinical
laboratories use biochemical and/or immunological identification assays, which are based on
species-specific phenotypic characteristics of a microorganism. However, the results can be
influenced by in vitro test conditions, such as incubation time, - temperature, - conditions (O2,
anaerobic incubation, CO2, etc.) and composition of the culture medium. These factors may
lead to misinterpretation of the results and subsequent misidentification of the etiological
agens (46). Automated bacterial identification and antimicrobial susceptibility testing (AST)

18

Rene te Witt bw.indd 18 24-11-11 10:35

 General introduction and outline of the thesis

platforms such as the VITEK (bioMérieux), PHOENIX (BD Diagnostics) and Microscan (Siemens

Chapter 1
Healthcare Diagnostics, Munich, Germany) are well established in diagnostic bacteriology.
All these systems are based on a conventional concept, using enzymatic (or biochemical)
properties of bacteria for the identification and micro-dilution for AST.

Antibiotic resistance

Penicillin resistance in S. aureus appeared very soon after the general clinical use of penicillin
in 1943, and the mechanism of resistance was the production of β-lactamase (47). The first
case of MRSA was reported in 1962, only 2 years after methicillin was introduced to treat
penicillin-resistant S. aureus. The original MRSA strains circulated within healthcare set-
tings, acquiring resistance to a range of antibiotics. Infections occurred only in individuals
with risk factors such as hospitalizations or residence in long-term care facilities. However,
since the 1990’s, there has been a rise in community-acquired MRSA (CA-MRSA) infections
in previously healthy individuals without healthcare exposure. These CA-MRSA strains did
not escape from hospitals, but arose in the community (48-50). The efficiency of person-to-
person transmissibility is one of the key components of its successful spread, combined with
increased virulence (51).
However, the largest current threat in antimicrobial resistance is the rapid and widespread
dissemination of multidrug resistant Gram-negative bacteria. There are frequent reports in
the literature of new β-lactamases and other novel resistance mechanisms as well as mobile
genetic elements that may spread easily from cell to cell. Laboratory detection is difficult
because multiple resistance mechanisms may be present and resistance may be “hidden”
– resistance genes may be present without the characteristic phenotype. Reliance on carbap-
enems for the treatment of extended-spectrum β-lactamases (ESBLs) has led to increasing
carbapenem resistance and the detection of various carbapenemases. The newly described
New Delhi metallo-β-lactamase 1 (NDM-1) and its subsequent global dissemination is of
great concern as there are few effective antimicrobials beyond carbapenems for these highly
resistant organisms (52). More recently, carbapenem multiresistance has been described in
a number of bacteria (53). Even with lowered Clinical and Laboratory Standards Institute
(CLSI) breakpoints, detection of carbapenemases is problematic, with different susceptibility
results obtained with different methods and error rates when using the reference broth mi-
crodilution method. Nucleic acid amplification assays are essential for detecting these types
of resistance mechanisms.

19

Rene te Witt bw.indd 19 24-11-11 10:35

 Chapter 1

Nucleic acid amplification

Each species harbours a unique nucleic acid signature, which can be used as target sequence
for the detection and identification of a specific microorganism in clinical samples. Addition-
ally, clinically relevant characteristics of that particular strain, such as its genes encoding
for resistance determinants and its array of virulence factors can also be detected. DNA is a
stable molecule in the absence of nucleases or restriction enzymes, which makes it very suit-
able as a diagnostic target, especially since it can be isolated relatively easily from a variety
of biological matrices. Over the past decades, a wide variety of nucleic acid amplification
techniques has been developed and implemented in microbiological diagnostic laborato-
ries. Clear benefits over traditional culture-based assays are, in particular, reduced time to
identification and improved sensitivity and specificity. Polymerase chain reaction (PCR) is the
most commonly used platform. PCR was developed by Kary Mullis and colleagues in 1987
(54). Since then, applications of PCR have expanded to clinical -, food -, environmental - and
essentially all biological research areas.
The most significant optimisation in PCR diagnostics has been the development of a
closed-tube, real-time detection system, thereby significantly reducing the risk of PCR prod-
uct contamination (55-57). Real-time PCR continuously monitors the amplification process,
in contrast to the end point detection of the amplification product in conventional PCR. Real-
time PCR provides sensitive and quantitative detection of PCR products in a rapid turnaround
time and strongly contributes to the strict requirements of clinical diagnostics. Real-time PCR
also allows for the quantification of the bacterial target DNA by using an internal calibration
curve. Nowadays, molecular quantification by real-time PCR is very common and important
in virology to establish and monitor the viral load during antiviral therapy to improve
patient management (HIV) (58). The determination of bacterial load potentially allows for
the monitoring of antimicrobial therapy response or discrimination between infection and
colonization (59). Results can be obtained on the same day of specimen collection, allowing
adequate focusing of antibacterial therapy and reduction of unnecessary use of antibiotics.
Real-time PCR depends on different detection strategies. Fluorescent technologies
employed are either non-specific or specific. Non-specific fluorescence uses dyes such
as SYBR Green or SYBR Gold, which efficiently intercalate into the PCR product during the
annealing phase of amplification. Specific fluorescence technologies use probes to detect
target sequence amplification. A number of different fluorescent probe chemistries have
been developed for real-time PCR assays, based on 2 main principles, hydrolysis probes (e.g.
TaqMan probes, molecular beacons, Scorpion probes or minor groove binding probes) and
hybridisation probes (e.g. Fluorescence Resonance Energy Transfer probes (FRET)). Different
methods of fluorescence detection are shown in Figure 2.

20

Rene te Witt bw.indd 20 24-11-11 10:35

 General introduction and outline of the thesis

Chapter 1
Figure 2. Methods of fluorescence detection.
R=reporter, Q=quencher, D=donor, A=acceptor.

A wide variety of alternative, commercial, amplification-based technologies has been de-

scribed for the detection of an extended number of bacteria. These tests are too complex to
be developed for an in-house format and involve nucleic acid sequence based amplification
(NASBA), strand displacement amplification (SDA), multiplex ligation-dependent probe am-
plification (MLPA), loop-mediated isothermal amplification (LAMP) and many others (Table
1). SDA allows for 1010-fold isothermal amplification of a DNA target in less than 15 minutes
(60). SDA was the first nucleic acid amplification technology to be coupled with real-time
fluorescence-based detection for routine diagnostic application in the clinical laboratory. The
isothermal nature of the reaction process offers distinct advantages with regard to the cost
and simplicity of the instrumentation.

Table 1. Commonly used nucleic acid amplification techniques in clinical microbiology.

Amplification technique Strengths Weaknesses Example(s) References
Polymerase chain reaction - Efficient - Limited in multiplex Methicillin resistant (54-57, 61-62)
(PCR) - Sensitivity Staphylococcus aureus
Nucleic acid sequence-based - Efficient - Low specificity Respiratory tract infections (62-64)
amplification - Isothermal
(NASBA)
Strand displacement - Isothermal - Low specificity Sexually transmitted diseases (60, 62, 65)
amplification - Rapid - Short target sequences
(SDA)
Ligase chain reaction - Specific - High risk of contamination HIV, sexually transmitted (66-67)
(LCR) - Low reproducibility diseases
Loop-mediated isothermal - Rapid - Complex Clostridium difficile, (68-71)
amplification - Specific Mycobacterium tuberculosis
(LAMP) - Isothermal
Multiplex ligation-dependent - Multiplex - Costs Respiratory tract infections, (72-74)
probe amplification - Post-amplification analysis sexually transmitted diseases
(MLPA) - Slow

21

Rene te Witt bw.indd 21 24-11-11 10:35

 Chapter 1

However, in contrast to the obvious advantages of nucleic acid amplification techniques,

there is a risk of false-positivity or false-negativity. False-positive results are primarily due to
contamination of reactions by spill-over of other samples or products of previous amplifica-
tions. Nucleic acid amplification is highly sensitive and will pick up even the tiniest amounts
of a contaminant. This means that contamination prevention has a high priority in the di-
agnostic laboratory by using different laboratories for PCR mix setup, nucleic acid isolation,
PCR setup, amplification and detection. Next to this, contamination can also be prevented by
using Uracil-DNA Glycosylase (UNG) prior to the thermal cycling in combination with the use
of dUTP in the PCR amplification. Another factor causing false-positivity may be the result of
cross-reactivity, indicating a low specificity of the diagnostic test.
False-negative results may be the consequence of insensitivity of the test; only high
quantities of the target will lead to a positive result. Variability of the target sequence
(e.g. SCCmec variance in MRSA) may also result in false-negativity. In such cases, the assay
should be redesigned. Another cause of false-negative results is the presence of inhibitors
of nucleic acid amplification in the clinical sample. Blood, urine and feces are well-known
examples of complex biological samples, with many different inhibitors (61-62). The inhibi-
tory compounds of the clinical sample interact with one or more of the components of the
PCR reaction mixture. For instance, anticoagulants, added to blood samples, chelate Mg2+
ions, hereby reducing DNA-polymerase effectiveness (63). Blood compounds such as hae-
moglobin, lactoferrin, heme and IgG may also act as inhibitors (64). As a consequence of
inhibition, extensive nucleic acid purification is often required to generate a high quality and
high yield of nucleic acids (65). Morata et al demonstrated that incorporation of additional
washing steps into their DNA extraction protocol removed inhibitors to allow successful PCR,
although it resulted in a lower DNA yield (66). Purification adds to the time and expense of
sample preparation, as well as to the loss of nucleic acids. Therefore a balance needs to be
found between time, costs and purity (67).
Nucleic acid amplification assays have a number of limitations that restrict their applicabil-
ity. Contamination of samples and PCR reagents remains an issue. These assays require care-
ful validation. Validation includes sampling, DNA extraction, amplification and interpretation
of the results. Furthermore, it implies the inclusion of positive and negative controls, testing
of inhibition controls, pre-treatment step(s) to ensure good extraction of nucleic acids from
bacterial cells and regular testing of all reagents, such as primers, probe(s), polymerase etc.

22

Rene te Witt bw.indd 22 24-11-11 10:35

 General introduction and outline of the thesis

Bacterial typing

Chapter 1
After detection and identification of (antibiotic-resistant) bacteria, identification at the strain
level (also known as typing) may be necessary in order to trace potential transmission routes
and decide whether infection prevention measures are mandatory. In addition, typing is es-
sential for elucidation of (inter)national dissemination of bacterial clones.
Since S. aureus is one of the most important human pathogens and its antibiotic resistance
is continuously increasing worldwide, many of the most relevant typing techniques have
been developed and validated for (methicillin-resistant) S. aureus.
Currently, a wide variety of genetic typing methods are in use in the diagnostic laboratory,
but pulsed-field gel electrophoresis (PFGE) of macro restriction fragments of genomic DNA
is preferred because of its high discriminatory power (68-70). However, PFGE is technically
demanding, with limited portability due to low reproducibility (70).
The introduction of a commercially available automated rep-PCR system, the DiversiLab
system (bioMérieux) offers advances in standardization and reproducibility of the procedure
over manual fingerprint-generating systems (71). However, although two independent stud-
ies concluded that the DiversiLab system is a rapid and reproducible technique, it clearly
lacks resolution for typing of Gram-positive bacteria. DiversiLab analysis does not differenti-
ate genetically among epidemiologically unique MRSA strains, which is needed for adequate
outbreak analysis (72-73).
Multilocus variable number of tandem repeat analysis (MLVA) is a high-throughput ge-
notyping technique that can be used for hospital- and (inter)national genotyping, but the
discriminatory power depends on the number and types of loci analyzed (74-75).
Sequence-based approaches, such as spa sequence typing and multilocus sequence typing
(MLST), have resulted in large sequence databases (76-78). The determination of sequence
polymorphism of the spa gene encoding the staphylococcal surface protein A (spa sequence
typing) has become the most popular MRSA typing system, thanks to high-throughput
capacity and an excellent reproducibility, which allows portability of data and comparison
worldwide (79). However, spa sequence typing has a moderate discriminatory power and
misclassification of particular sequence types is possible due to recombination. MLST has
great interlaboratory reproducibility thanks to standardised nomenclature, but comes with
limited discriminatory power and low throughput. Therefore, MLST is more suited for long-
term epidemiological studies, such as population structure analysis, phylogeny and long-
term surveillance.
Whole genome sequencing (WGS) provides a complete inventory of micro-evolutionary
changes within bacterial genomes, but this approach is as yet impractical for routine diagnos-
tic laboratories due to high costs and mandatory technical expertise. Mwangi and colleagues
used WGS to identify steps in the evolution of multidrug resistance in S. aureus isolates recov-
ered from the bloodstream of a patient. Sequencing the first vancomycin susceptible isolate

23

Rene te Witt bw.indd 23 24-11-11 10:35

 Chapter 1

and the last vancomycin non-susceptible isolate identified genome wide only 35 point muta-
tions in 31 loci (80). In a recent paper, Harris et al described a new high-throughput genomics
approach based on full-genome sequencing, which provides a high-resolution view of the
epidemiology of MRSA with the potential to trace person-to-person transmission within a
hospital environment (81).
Raman spectroscopy (SpectraCellRA, River Diagnostics, Rotterdam, The Netherlands) has
been described as a promising tool for phenotypical identification and typing of microorgan-
isms (82-83). This vibrational spectroscopic technique does not require any labels and is fast,
non-invasive and highly reproducible. Its high throughput and ease of use enables it suitable
for use in routine diagnostic laboratories.
Currently available molecular typing methods are described in Table 2.

Table 2. Comparison of the main currently available molecular typing methods.

Method Principle/target Strengths Weaknesses Example(s) References
Pulsed-field gel Restriction - High discriminatory - Technically demanding Methicillin resistant (82-83, 98-99)
electrophoresis polymorphism of power - Slow Staphylococcus aureus
(PFGE) the whole genome - Limited portability (MRSA), Pseudomonas
- Multiple nomenclatures aeruginosa
Multilocus Sequence - Phylogenetic structure - Limited discriminatory Candida albicans, (90-91, 100)
sequence typing determination of of core genome power MRSA
(MLST) allelic variants of - Portability - Low throughput
housekeeping genes - Standard - Expensive
nomenclature
spa-sequence Polymorphism - Rapid - Moderate (methicillin resistant) (92, 101)
typing of number and - High throughput discriminatory power S. aureus
(forS. aureus sequence of repeat - Portability - Misclassification of STs*
typing only) elements of the - Standard due to recombination
hypervariable gene nomenclature
rep-PCR typing Polymorphism - Rapid - Limited discriminatory MRSA, multi-drug (85-86,
(Diversilab) in chromosomal - High throughput power for Gram-positive resistant bacteria, P. 102-103)
inter-repeat element - Portability cocci aeruginosa
spacers - No validated
interpretation criteria
- No standard
nomenclature
Multilocus VNTR Polymorphism - Rapid - Limited discriminatory Mycobacterium (104-108)
analysis in number of - High throughput power tuberculosis, Bacillus
(MLVA) chromosomal - Standard anthracis, MRSA
VNTR** elements nomenclature
Raman Overall molecular - High discriminatory - No standard MRSA, P. aeruginosa (96, 109)
spectroscopy composition as power nomenclature
spectroscopic - Rapid
fingerprints - High throughput
- Portability
* ST: Sequence Type
** VNTR: Variable Number of Tandem Repeat.

24

Rene te Witt bw.indd 24 24-11-11 10:35

 General introduction and outline of the thesis

Quality control

Chapter 1
A challenge for high-quality molecular-based diagnostics is to monitor the performance of
the participating laboratories and to standardize assays between laboratories. There are a
limited number of commercially standardized molecular assays available, compared with the
high variety of ill-controlled “in-house” real-time PCR assays. Different diagnostic microbio-
logical laboratories may use different assays with different levels of sensitivity and specificity.
One way of tackling this problem is by implementing external quality assessment (EQA)
programs. This can be done by engaging both manufacturers and diagnostic laboratories in
well-defined and robust EQA programs organized by independent institutes.

Depending on the technological and financial possibilities, a clinical microbiological labora-

tory should optimize its diagnostic strategy by applying a combination of conventional with
real-time amplification tests for the detection of microbiological agents. When implementing
such a strategy, a balance between performance criteria (sensitivity, specificity) and conve-
nience criteria (clinical utility, turn-around time and costs) will have to be defined. In the end,
this should result in the optimization of clinical patient management.
During the past ten years, a shift from classical technologies to refined molecular tech-
niques could be observed in clinical microbiology. Rapid identification has been established,
however, AST is still slow. Furthermore, there is a need for EQA and further automisation and
communication of data into the clinic followed by action in the clinic. This thesis will cover
various aspects of the abovementioned challenges.

25

Rene te Witt bw.indd 25 24-11-11 10:35

 Chapter 1

Outline of the thesis

The scope of this thesis was to explore new applications of modern diagnostic tools within
the setting of a clinical microbiology laboratory.

Introduction

Currently available microbiological tests for the detection of Mycoplasma pneumoniae, Chla-
mydophila pneumoniae and Legionella spp., the clinical performance of these tests and their
future in the clinical microbiology laboratory are outlined in Chapter 2.

Biomarkers

The first part of Chapter 3 describes procalcitonin and neopterin as suitable biomarkers
for exclusion of severe Plasmodium falciparum disease at the initial clinical assessment of
travellers with imported malaria. The diagnostic accuracy of procalcitonin -, neopterin – and
sTREM-1-levels was evaluated in a cohort of 104 travellers with imported malaria (26 patients
with non-P. falciparum malaria, 64 patients with uncomplicated P. falciparum malaria and 14
patients with severe P. falciparum malaria).
The second part of Chapter 3 focuses on the diagnostic relevance of procalcitonin- and
neopterin levels as biomarkers for a bacterial or a viral infection in a cohort of 69 ill-returned
febrile travellers after a stay in the (sub)tropics (Dengue virus [n=33], Salmonella enterica
serovar [para]Typhi [n=17] and Rickettsia africae [n=19]).

Culture

Chapter 4 describes the in vitro evaluation of the performance of Granada broth for detec-
tion of Group B Streptococcal (GBS) colonization, both direct and after transport. Simulating
conditions in everyday practice, we have compared the sensitivity of Granada tube broth
(GT) (bioMérieux) with that of classical Amies transport medium (AT) (bioMérieux) in vitro.
A total of 1,485 GT and 1,485 AT were tested with 33 well-characterized GBS strains in three
different concentrations, five different incubation times and three different temperatures.

Nucleic acid amplification

An update on molecular diagnostics and typing of MRSA is provided in Chapter 5.

26

Rene te Witt bw.indd 26 24-11-11 10:35

 General introduction and outline of the thesis

Bacterial typing

Chapter 1
Chapter 6 discusses typing of MRSA. We have analyzed a representative selection of the HAR-
MONY MRSA strain collection originating from 11 European countries with the DiversiLab
system, pulsed-field gel electrophoresis (PFGE) and multi-locus sequence typing (MLST).

Chapter 7 evaluates the application of SpectracellRA (Raman spectroscopy) for typing of

MRSA isolates. We have retrospectively tested a well-documented MRSA collection, consist-
ing of 113 MRSA isolates, originating from 54 households.

Quality control

Chapter 8 focuses on External Quality Assessments (EQA) of molecular diagnostics and

genotyping of MRSA. Two multicentre EQA studies for molecular detection and genotyping
of MRSA were produced and arranged in cooperation with Quality Control for Molecular
Diagnostics (QCMD) in Glasgow. Firstly, eleven samples containing various amounts of inacti-
vated MRSA strains, methicillin-susceptible S. aureus (MSSA), methicillin-resistant coagulase-
negative staphylococci (MRCoNS) or Escherichia coli, were distributed to 82 laboratories for
MRSA detection. Secondly, a panel for MRSA typing, consisting of 10 samples (2 identical, 3
genetically related and 5 unique MRSA strains) was distributed to 19 laboratories.

Discussion and future perspectives

Chapter 9 provides an overview of diagnostic tools (Matrix assisted laser desorption

ionisation-time of flight mass spectrometry [MALDI-TOF MS], electronic nose and Raman
spectroscopy) that are currently being developed and implemented for fast and reliable
identification or typing of bacteria in clinical laboratories. Furthermore, the most important
candidates for multiplex analysis are discussed.

Finally, in Chapter 10, the preceding chapters are discussed, accompanied by a perspective
on future developments within the field of diagnostic microbiology.

27

Rene te Witt bw.indd 27 24-11-11 10:35

 Chapter 1

References
1. Watson, J. D. and Crick, F. H. (1953). Genetical implications of the structure of deoxyribonucleic acid. Nature
171(4361): 964-7.
2. Watson, J. D. and Crick, F. H. (1953). Molecular structure of nucleic acids; a structure for deoxyribose nucleic acid.
Nature 171(4356): 737-8.
3. Chirouze, C., et al. (2002). Low serum procalcitonin level accurately predicts the absence of bacteremia in adult
patients with acute fever. Clin Infect Dis 35(2): 156-61.
4. Hoen, B. (2009). [Differentiating bacterial from viral meningitis: contribution of nonmicrobiological laboratory
tests] Diagnostic differentiel entre meningite bacterienne et meningite virale : apport des examens non microbi-
ologiques. Med Mal Infect 39(7-8): 468-72.
5. Pfafflin, A. and Schleicher, E. (2009). Inflammation markers in point-of-care testing (POCT). Anal Bioanal Chem
393(5): 1473-80.
6. Simon, L., et al. (2004). Serum procalcitonin and C-reactive protein levels as markers of bacterial infection: a
systematic review and meta-analysis. Clin Infect Dis 39(2): 206-17.
7. Ip, M., et al. (2007). Value of serum procalcitonin, neopterin, and C-reactive protein in differentiating bacterial from
viral etiologies in patients presenting with lower respiratory tract infections. Diagn Microbiol Infect Dis 59(2): 131-6.
8. Schuetz, P., et al. (2010). Procalcitonin for guidance of antibiotic therapy. Expert Rev Anti Infect Ther 8(5): 575-87.
9. Russwurm, S., et al. (1999). Molecular aspects and natural source of procalcitonin. Clin Chem Lab Med 37(8): 789-97.
10. Snider, R. H., Jr., Nylen, E. S. and Becker, K. L. (1997). Procalcitonin and its component peptides in systemic inflam-
mation: immunochemical characterization. J Investig Med 45(9): 552-60.
11. Reinhart, K., Karzai, W. and Meisner, M. (2000). Procalcitonin as a marker of the systemic inflammatory response to
infection. Intensive Care Med 26(9): 1193-200.
12. Muller, B. and Becker, K. L. (2001). Procalcitonin: how a hormone became a marker and mediator of sepsis. Swiss
Med Wkly 131(41-42): 595-602.
13. Reinhart, K., Meisner, M. and Brunkhorst, F. M. (2006). Markers for sepsis diagnosis: what is useful? Crit Care Clin
22(3): 503-19, ix-x.
14. Meisner, M., et al. (1997). Procalcitonin--influence of temperature, storage, anticoagulation and arterial or venous
asservation of blood samples on procalcitonin concentrations. Eur J Clin Chem Clin Biochem 35(8): 597-601.
15. Muller, M. M., et al. (1991). Neopterin in clinical practice. Clin Chim Acta 201(1-2): 1-16.
16. Werner, E. R., et al. (1991). Biochemistry and function of pteridine synthesis in human and murine macrophages.
Pathobiology 59(4): 276-9.
17. Millner, M. M., et al. (1998). Neopterin concentrations in cerebrospinal fluid and serum as an aid in differentiating
central nervous system and peripheral infections in children. Clin Chem 44(1): 161-7.
18. Delogu, G., et al. (1995). Serum neopterin and soluble interleukin-2 receptor for prediction of a shock state in
gram-negative sepsis. J Crit Care 10(2): 64-71.
19. Bouchon, A., Dietrich, J. and Colonna, M. (2000). Cutting edge: inflammatory responses can be triggered by TREM-
1, a novel receptor expressed on neutrophils and monocytes. J Immunol 164(10): 4991-5.
20. Gibot, S., et al. (2004). Plasma level of a triggering receptor expressed on myeloid cells-1: its diagnostic accuracy in
patients with suspected sepsis. Ann Intern Med 141(1): 9-15.
21. Gibot, S., et al. (2005). Surface triggering receptor expressed on myeloid cells 1 expression patterns in septic shock.
Intensive Care Med 31(4): 594-7.

28

Rene te Witt bw.indd 28 24-11-11 10:35

 General introduction and outline of the thesis

22. McNerney, R. and Daley, P. (2011). Towards a point-of-care test for active tuberculosis: obstacles and opportunities.
Nat Rev Microbiol 9(3): 204-13.

Chapter 1
23. Walzl, G., et al. (2011). Immunological biomarkers of tuberculosis. Nat Rev Immunol 11(5): 343-54.
24. Kibe, S., Adams, K. and Barlow, G. (2011). Diagnostic and prognostic biomarkers of sepsis in critical care. J Antimi-
crob Chemother 66 Suppl 2: ii33-40.
25. Riedel, S., et al. (2011). Procalcitonin as a marker for the detection of bacteremia and sepsis in the emergency
department. Am J Clin Pathol 135(2): 182-9.
26. Gilbert, D. N. (2011). Procalcitonin as a biomarker in respiratory tract infection. Clin Infect Dis 52 Suppl 4: S346-50.
27. Fowler, C. L. (2011). Procalcitonin for triage of patients with respiratory tract symptoms: a case study in the trial
design process for approval of a new diagnostic test for lower respiratory tract infections. Clin Infect Dis 52 Suppl
4: S351-6.
28. Ruuskanen, O., et al. (2011). Viral pneumonia. Lancet 377(9773): 1264-75.
29. Chen, T. H., et al. (2010). Cytotoxic lignan esters from Cinnamomum osmophloeum. Planta Med 76(6): 613-9.
30. Cooke, V. M., et al. (2002). New chromogenic agar medium for the identification of Candida spp. Appl Environ
Microbiol 68(7): 3622-7.
31. Hedin, G. and Fang, H. (2005). Evaluation of two new chromogenic media, CHROMagar MRSA and S. aureus ID, for
identifying Staphylococcus aureus and screening methicillin-resistant S. aureus. J Clin Microbiol 43(8): 4242-4.
32. Kipp, F., et al. (2005). Evaluation of two chromogenic agar media for recovery and identification of Staphylococcus
aureus small-colony variants. J Clin Microbiol 43(4): 1956-9.
33. Perry, J. D., et al. (2004). Development and evaluation of a chromogenic agar medium for methicillin-resistant
Staphylococcus aureus. J Clin Microbiol 42(10): 4519-23.
34. Verani, J. R., et al. (2010). Prevention of perinatal group B streptococcal disease--revised guidelines from CDC,
2010. MMWR Recomm Rep 59(RR-10): 1-36.
35. Muller, A. E., et al. (2006). Morbidity related to maternal group B streptococcal infections. Acta Obstet Gynecol
Scand 85(9): 1027-37.
36. Bergseng, H., et al. (2007). Real-time PCR targeting the sip gene for detection of group B Streptococcus coloniza-
tion in pregnant women at delivery. J Med Microbiol 56(Pt 2): 223-8.
37. Valkenburg-van den Berg, A. W., et al. (2006). Prevalence of colonisation with group B Streptococci in pregnant
women of a multi-ethnic population in The Netherlands. Eur J Obstet Gynecol Reprod Biol 124(2): 178-83.
38. Campbell, J. R., et al. (2000). Group B streptococcal colonization and serotype-specific immunity in pregnant
women at delivery. Obstet Gynecol 96(4): 498-503.
39. Rosa-Fraile, M., et al. (2005). Specimen storage in transport medium and detection of group B streptococci by
culture. J Clin Microbiol 43(2): 928-30.
40. Stoner, K. A., Rabe, L. K. and Hillier, S. L. (2004). Effect of transport time, temperature, and concentration on the
survival of group B streptococci in amies transport medium. J Clin Microbiol 42(11): 5385-7.
41. Schrag, S., et al. (2002). Prevention of perinatal group B streptococcal disease. Revised guidelines from CDC.
MMWR Recomm Rep 51(RR-11): 1-22.
42. Ostroff, R. M. and Steaffens, J. W. (1995). Effect of specimen storage, antibiotics, and feminine hygiene products on
the detection of group B Streptococcus by culture and the STREP B OIA test. Diagn Microbiol Infect Dis 22(3): 253-9.
43. Hakansson, S., et al. (2008). Group B streptococcal carriage in Sweden: a national study on risk factors for mother
and infant colonisation. Acta Obstet Gynecol Scand 87(1): 50-8.
44. Martinho, F., et al. (2008). Evaluation of liquid biphasic Granada medium and instant liquid biphasic Granada
medium for group B streptococcus detection. Enferm Infecc Microbiol Clin 26(2): 69-71.

29

Rene te Witt bw.indd 29 24-11-11 10:35

 Chapter 1

45. Heelan, J. S., et al. (2005). Evaluation of a new selective enrichment broth for detection of group B streptococci in
pregnant women. J Clin Microbiol 43(2): 896-7.
46. Hengstler, K. A., Hammann, R. and Fahr, A. M. (1997). Evaluation of BBL CHROMagar orientation medium for detec-
tion and presumptive identification of urinary tract pathogens. J Clin Microbiol 35(11): 2773-7.
47. Kirby, W. M. (1944). Extraction of a Highly Potent Penicillin Inactivator from Penicillin Resistant Staphylococci.
Science 99(2579): 452-3.
48. McDougal, L. K., et al. (2003). Pulsed-field gel electrophoresis typing of oxacillin-resistant Staphylococcus aureus
isolates from the United States: establishing a national database. J Clin Microbiol 41(11): 5113-20.
49. Tenover, F. C. and Goering, R. V. (2009). Methicillin-resistant Staphylococcus aureus strain USA300: origin and
epidemiology. J Antimicrob Chemother 64(3): 441-6.
50. Tenover, F. C., et al. (2008). Characterization of Staphylococcus aureus isolates from nasal cultures collected from
individuals in the United States in 2001 to 2004. J Clin Microbiol 46(9): 2837-41.
51. Li, M., et al. (2009). Evolution of virulence in epidemic community-associated methicillin-resistant Staphylococcus
aureus. Proc Natl Acad Sci U S A 106(14): 5883-8.
52. Kumarasamy, K. K., et al. (2010). Emergence of a new antibiotic resistance mechanism in India, Pakistan, and the
UK: a molecular, biological, and epidemiological study. Lancet Infect Dis 10(9): 597-602.
53. Tato, M., et al. (2010). Carbapenem Heteroresistance in VIM-1-producing Klebsiella pneumoniae isolates belonging
to the same clone: consequences for routine susceptibility testing. J Clin Microbiol 48(11): 4089-93.
54. Mullis, K. B. and Faloona, F. A. (1987). Specific synthesis of DNA in vitro via a polymerase-catalyzed chain reaction.
Methods Enzymol 155: 335-50.
55. Heid, C. A., et al. (1996). Real time quantitative PCR. Genome Res 6(10): 986-94.
56. Gibson, U. E., Heid, C. A. and Williams, P. M. (1996). A novel method for real time quantitative RT-PCR. Genome Res
6(10): 995-1001.
57. Holland, P. M., et al. (1991). Detection of specific polymerase chain reaction product by utilizing the 5’----3’ exo-
nuclease activity of Thermus aquaticus DNA polymerase. Proc Natl Acad Sci U S A 88(16): 7276-80.
58. Holzl, G., et al. (2003). Entirely automated quantification of human immunodeficiency virus type 1 (HIV-1) RNA in
plasma by using the ultrasensitive COBAS AMPLICOR HIV-1 monitor test and RNA purification on the MagNA pure
LC instrument. J Clin Microbiol 41(3): 1248-51.
59. McCulloch, E., et al. (2011). Improved early diagnosis of Pseudomonas aeruginosa by real-time PCR to prevent
chronic colonisation in a paediatric cystic fibrosis population. J Cyst Fibros 10(1): 21-4.
60. Walker, G. T., et al. (1992). Strand displacement amplification--an isothermal, in vitro DNA amplification technique.
Nucleic Acids Res 20(7): 1691-6.
61. Mercier, B., et al. (1990). Direct PCR from whole blood, without DNA extraction. Nucleic Acids Res 18(19): 5908.
62. Wiedbrauk, D. L., Werner, J. C. and Drevon, A. M. (1995). Inhibition of PCR by aqueous and vitreous fluids. J Clin
Microbiol 33(10): 2643-6.
63. Abu Al-Soud, W. and Radstrom, P. (2000). Effects of amplification facilitators on diagnostic PCR in the presence of
blood, feces, and meat. J Clin Microbiol 38(12): 4463-70.
64. Al-Soud, W. A. and Radstrom, P. (2001). Purification and characterization of PCR-inhibitory components in blood
cells. J Clin Microbiol 39(2): 485-93.
65. Boom, R., et al. (1990). Rapid and simple method for purification of nucleic acids. J Clin Microbiol 28(3): 495-503.
66. Morata, P., Queipo-Ortuno, M. I. and de Dios Colmenero, J. (1998). Strategy for optimizing DNA amplification in a
peripheral blood PCR assay used for diagnosis of human brucellosis. J Clin Microbiol 36(9): 2443-6.

30

Rene te Witt bw.indd 30 24-11-11 10:35

 General introduction and outline of the thesis

67. Kreader, C. A. (1996). Relief of amplification inhibition in PCR with bovine serum albumin or T4 gene 32 protein.
Appl Environ Microbiol 62(3): 1102-6.

Chapter 1
68. Goering, R. V. and Duensing, T. D. (1990). Rapid field inversion gel electrophoresis in combination with an rRNA
gene probe in the epidemiological evaluation of staphylococci. J Clin Microbiol 28(3): 426-9.
69. Ichiyama, S., et al. (1991). Genomic DNA fingerprinting by pulsed-field gel electrophoresis as an epidemiological
marker for study of nosocomial infections caused by methicillin-resistant Staphylococcus aureus. J Clin Microbiol
29(12): 2690-5.
70. van Belkum, A., et al. (1998). Assessment of resolution and intercenter reproducibility of results of genotyping
Staphylococcus aureus by pulsed-field gel electrophoresis of SmaI macrorestriction fragments: a multicenter study.
J Clin Microbiol 36(6): 1653-9.
71. Healy, M., et al. (2005). Microbial DNA typing by automated repetitive-sequence-based PCR. J Clin Microbiol 43(1):
199-207.
72. Tenover, F. C., et al. (2009). Comparison of typing results obtained for methicillin-resistant Staphylococcus aureus
isolates with the DiversiLab system and pulsed-field gel electrophoresis. J Clin Microbiol 47(8): 2452-7.
73. Babouee, B., et al. (2011). Comparison of the rep-PCR system DiversiLab with spa typing and pulsed-field gel
electrophoresis for the clonal characterization of methicillin-resistant Staphylococcus aureus. J Clin Microbiol.
74. Schouls, L. M., et al. (2009). Multiple-locus variable number tandem repeat analysis of Staphylococcus aureus:
comparison with pulsed-field gel electrophoresis and spa-typing. PLoS One 4(4): e5082.
75. Pourcel, C., et al. (2009). Improved multiple-locus variable-number tandem-repeat assay for Staphylococcus aureus
genotyping, providing a highly informative technique together with strong phylogenetic value. J Clin Microbiol
47(10): 3121-8.
76. Maiden, M. C., et al. (1998). Multilocus sequence typing: a portable approach to the identification of clones within
populations of pathogenic microorganisms. Proc Natl Acad Sci U S A 95(6): 3140-5.
77. Ibarz Pavon, A. B. and Maiden, M. C. (2009). Multilocus sequence typing. Methods Mol Biol 551: 129-40.
78. Frenay, H. M., et al. (1996). Molecular typing of methicillin-resistant Staphylococcus aureus on the basis of protein
A gene polymorphism. Eur J Clin Microbiol Infect Dis 15(1): 60-4.
79. Friedrich, A. W., et al. (2008). A European laboratory network for sequence-based typing of methicillin-resistant
Staphylococcus aureus (MRSA) as a communication platform between human and veterinary medicine--an update
on SeqNet.org. Euro Surveill 13(19).
80. Mwangi, M. M., et al. (2007). Tracking the in vivo evolution of multidrug resistance in Staphylococcus aureus by
whole-genome sequencing. Proc Natl Acad Sci U S A 104(22): 9451-6.
81. Harris, S. R., et al. (2010). Evolution of MRSA during hospital transmission and intercontinental spread. Science
327(5964): 469-74.
82. Willemse-Erix, D. F., et al. (2009). Optical fingerprinting in bacterial epidemiology: Raman spectroscopy as a real-
time typing method. J Clin Microbiol 47(3): 652-9.
83. Maquelin, K., et al. (2003). Prospective study of the performance of vibrational spectroscopies for rapid identifica-
tion of bacterial and fungal pathogens recovered from blood cultures. J Clin Microbiol 41(1): 324-9.

31

Rene te Witt bw.indd 31 24-11-11 10:35

Rene te Witt bw.indd 32 24-11-11 10:35
 Chapter 2
Specific diagnostic tests
for atypical respiratory
tract pathogens

René te Witt
Willem B. van Leeuwen
Alex van Belkum

Infectious Disease Clinics of North America, 2010; 24: 229-248

Rene te Witt bw.indd 33 24-11-11 10:35

 Chapter 2

Abstract

This chapter reviews the microbiological diagnostic tests that are currently available for the
detection of M. pneumoniae, C. pneumoniae and Legionella spp., their clinical performance
and their future in the clinical microbiology laboratory. When implementing a strategy, a bal-
ance between performance criteria (sensitivity, specificity) and convenience criteria (clinical
utility, turn-around time and costs) will have to be defined. In the end, this should result in the
optimization of clinical patient management.

34

Rene te Witt bw.indd 34 24-11-11 10:35

 Specific diagnostic tests for atypical respiratory tract pathogens

Introduction

The term “atypical pneumonia” most commonly refers to pneumonia caused by Chlamydophi-
la pneumoniae, Mycoplasma pneumoniae or Legionella species. Although Bordetella pertussis/

Chapter 2
parapertussis, Franscissella tularensis, and several respiratory parasites, fungi and viruses are
also part of the spectrum of causative agents, these organisms will not be discussed further
in this chapter. C. pneumoniae, M. pneumoniae and Legionella spp. are increasingly recognized
as frequent and important pathogens in (acute) respiratory tract infections (RTI), such as
community-acquired pneumonia and exacerbations of chronic bronchitis (1-2).
For the detection of these pathogens, serology is still considered as the gold standard (2).
However, serological results are often unreliable and usually available too late to have an im-
pact on patient management. Optimal serological testing for individual patient care depends
on the age of the patient, timing of serum collection, whether paired (acute and convalescent)
sera are obtained, availability of appropriate equipment and experience of the laboratory
personnel. The latest developments in diagnostic strategies include the application of Nucleic
Acid Amplification Techniques (NAATs), such as Nucleic Acid Sequence Based Amplification
(NASBA), (real-time) Polymerase Chain Reaction (PCR), Strand Displacement Amplification
(SDA), Multi Ligation-dependent Probe Amplification (MLPA) and others in the detection of an
extended number of agents responsible for RTI. Advantages of real-time PCR over traditional
PCR include a more rapid turn-around time and the complete lack of post-amplification analy-
sis. Results can be obtained within the same day of specimen receipt, allowing appropriate
focusing of therapy and reduction of unnecessary antibiotic therapy in RTI (3-4).
Viruses cause a large proportion of RTI and are responsible for extensive morbidity and
mortality. The availability and use of these new molecular diagnostic tools in virology has
contributed tremendously to a better understanding of the viral etiology of RTI. This, how-
ever, depends on the populations studied, the geographical location and seasonality. Next to
the increasing importance of viral agents, the role of the bacterial pathogens C. pneumoniae,
M. pneumoniae and Legionella spp. in RTI is becoming more clear (5-7). Novel diagnostic tests
are more frequently presented and these also cover bacterial agents of RTI (8).
The overall diagnostic approaches for these bacterial pathogens are depicted in Table 1
and will be described in separate sections in this chapter. The quality of different types of
specimens commonly collected to detect pathogens causing RTI and different processing
routes, have been compared and are important features in the diagnostic setting. These will
also be discussed in forthcoming sections.

Depending on the technological and financial possibilities, a microbiological laboratory

should optimize its diagnostic strategy by applying a combination of classical and/or real-
time amplification tests for the detection of viruses and the atypical bacterial agents. When
implementing such a strategy, a balance between performance criteria (sensitivity, speci-

35

Rene te Witt bw.indd 35 24-11-11 10:35

 Chapter 2

Table 1. Diagnostic approaches for the detection of bacterial atypical RTI pathogens.
Pathogen Clinical sample “New” test Gold standard
Legionella spp. Urine Antigen test
Respiratory samples NAAT Culture
Serum IgG/IgM
C. pneumoniae Respiratory samples NAAT Culture
Serum IgG/IgM
M. pneumoniae Respiratory samples NAAT Culture
Serum IgG/IgM

ficity) and convenience criteria (clinical utility, turn-around time and costs) will have to be
defined. In the end, this should result in the optimization of clinical patient management.
This chapter reviews the microbiological tests that are currently available for the detection of
M. pneumoniae, C. pneumoniae and Legionella spp., their clinical performance and their future
in the clinical microbiology laboratory.

Mycoplasma pneumoniae

Historically, serology has been the most common laboratory method for the diagnosis of M.
pneumoniae infections. An infection is defined as a fourfold increase in antibody titre in acute
and convalescent sera and is still considered the “gold standard” for reliably defining infection
(9). Although culture and PCR are also used, non-clinical persistence of the viable organism
or dead cells for variable periods of time following acute infection remains a challenge in the
assessment of the significance of a positive culture or PCR assay result.

Culture

The time required for microbiological culturing, ranging from a few days to 3 weeks, renders
cultivation impractical for patient management. Hence, culture expertise is not widely avail-
able, except in specialized research and reference laboratories.

Serological testing

M. pneumoniae has both lipid and protein antigens which can induce antibody responses
that can be detected after about one week of infection, peaking at 3-6 weeks, followed by a
gradual decline. This response facilitates several different types of serological assays, based
on these different antigens and technologies. Serology is a useful epidemiologic tool in
circumstances where the likelihood of RTI caused by M. pneumoniae is high, but it is not
optimally suited for direct management of individual patients. Its main disadvantage is the

36

Rene te Witt bw.indd 36 24-11-11 10:35

 Specific diagnostic tests for atypical respiratory tract pathogens

need for both acute and convalescent paired sera collected 2-4 weeks apart from each other
that should be tested simultaneously for IgM and IgG to confirm sero-conversion indicative
for infection. In a recent study, it was demonstrated that the percentage of persons with
acute infection who demonstrate a positive IgG response in the acute phase was <50% (10).

Chapter 2
This low sensitivity could be explained by the presence of specific IgG from previous infec-
tions or slow IgG increase in the individuals tested. However, when convalescent sera were
tested, the number of IgG positive specimens increased to 82%. A single measurement of IgM
might indicate an acute infection if the test is performed after at least 7 days after onset, but
the test could be false-negative if the test is performed before. This same study revealed that
only 14 out of 27 (52%) of acute-phase sera tested positive by various IgM assays. However,
this number increased to 39 (88%) when convalescent sera were tested. IgM antibodies can
persist for several weeks to months. Another study showed that the IgM ImmunoCard (Merid-
ian Bioscience, Cincinnati, Ohio) had a sensitivity of only 32% for the detection of acute M.
pneumoniae infection in sero-positive children suffering from pneumonia (11). Again, the
sensitivity increased to 89% when paired sera were analyzed. These findings suggest that
diagnosis of RTI should not be based on a single IgM measurement in patient’s serum, as was
suggested in several other studies (12-13). It is important to realize and to emphasize that
the antibody response may also be delayed in some infections or can even be absent if the
patient is immuno-suppressed or immuno-deficient.
M. pneumoniae is a mucosal pathogen and, for that reason, IgA is produced in an early state
of the infection. Measurement of serum IgA either alone or in combination with IgM may
therefore be an alternative for the diagnosis of an acute infection. Unfortunately, very few
commercial assays include reagents for IgA detection. The limited number of studies that in-
volved detection of IgA, generally documented improved detection of acute infection, espe-
cially in adults (14-16). However, one study measured IgG, IgM and IgA antibodies in healthy
blood donors and in patients with various infections caused by microorganisms other than M.
pneumoniae, using various commercial enzyme immuno assays (EIA) (15). It was documented
that 23% of the blood donors and 54% of the patients with various non-Mycoplasma infec-
tions were positive for IgA, raising doubts about its value to accurately diagnose a current
M. pneumoniae infection. Talkington et al showed that single-use EIAs were better able to
identify seropositive samples than several plate-type EIAs (10). However, plate-type EIAs may
be more efficient and cost effective in laboratories that need high-throughput.

Molecular testing

Owing to the relative insensitivity and prolonged time needed for the detection of M.
pneumoniae by culture, the need for paired acute and convalescent sera collected 2-4 weeks
apart for optimal serological diagnosis and other problems inherent to serological assays as
described before, PCR gained considerable interest very soon after its introduction in the late

37

Rene te Witt bw.indd 37 24-11-11 10:35

 Chapter 2

80s of the previous century. The first reports of PCR suited for detection of M. pneumoniae
appeared in 1989 (17-18). Since then, more than 200 papers describing the use of classical
PCR for detection of M. pneumoniae in human infections were published. Gene targets used
in various PCR-protocols for M. pneumoniae detection include the ATPase operon, the P1
adhesin, 16S rRNA gene, the elongation factor Tu coding gene (tuf gene) and the repetitive
element repMP1 (17-21) (see Table 2). The sensitivity of PCR is very high, theoretically cor-
responding to a single organism. However, its major advantage is the exclusive gain of time.
Several real-time PCR assays have also been described (21-27). Comparison of (real-time)
PCR with culture and/or serology has yielded various results that are not always concordant.
As would be expected, molecular-based assays often demonstrate superior sensitivity
for detection of acute infection over serology and culture (13, 23, 28), although this is not
always the case (25, 29). Positive PCR results in culture-negative persons without evidence
of respiratory tract disease suggest inadequate assay specificity, persistence of the organism
after infection or asymptomatic carriage. Quantitative studies may be required before we
can draw final conclusions on the usefulness and clinical applicability. Positive PCR results
in sero-negative patients could be the result of an inadequate immune response, to earlier
successful antibiotic treatment or sampling prior to specific antibody response. Negative
PCR results in culture-proven or sero-positive patients might be caused by the presence of
inhibitors or technical problems with the assay and its targetgene. If antibiotics have been
administered, PCR results may be negative even though serology is positive.
Because M. pneumoniae is only one of a variety of fastidious and/or slow-growing patho-
genic microorganisms responsible for RTI with clinically similar manifestations, there has
been considerable interest and effort to develop multiplex PCR assays for single step detec-

Table 2. Original references on NAAT for atypical RTI pathogens.

Gene target Authors Reference
M. pneumoniae
ATPase operon Bernet et al (18)
P1 adhesin Jensen et al (17)
16S rDNA van Kuppeveld et al (19)
tuf Luneberg et al (20)
repMP1 Dumke et al (21)
Legionella spp.
16S rDNA Jonas et al (46)
23S-5S Herpers et al (47)
5S rDNA Kessler et al (48)
mip Lindsay et al (49)
C. pneumoniae
ompA Kaltenboeck et al (50)
pstI Campbell et al (51)
pmp4 Mygind et al (52)
16S rDNA Gaydos et al (53)

38

Rene te Witt bw.indd 38 24-11-11 10:35

 Specific diagnostic tests for atypical respiratory tract pathogens

tion of multiple causative agents. Most assays now include gene targets for M. pneumoniae,
C. pneumoniae, L. pneumophila and, occasionally, other organisms (26, 30-40). Still, monoplex
assays seem to have higher sensitivity and specificity than multiplex assays (41).
NASBA, which is a technique based on isothermal RNA amplification, has also been applied

Chapter 2
for the detection of M. pneumoniae in clinical samples (42). Initial studies have shown that the
performance of NASBA is comparable to PCR in terms of sensitivity. A multiplex NASBA assay
targeting M. pneumoniae, C. pneumoniae and L. pneumophila has also been described (43).
Other techniques such as multiplex reverse transcription PCR have been described recently
(44) and will be developed further over the years to come (45).
In 2002 and 2004 the diagnostic performance of laboratories for the molecular detection
of M. pneumoniae was investigated (41). For these two quality control exercises with a 2-year
interval, specimens were spiked with M. pneumoniae. In 2002, only 2 out of 12 participants
obtained 100% correct results, 2 out of 12 produced false-positive results and 10 out of 12
had between 0 out of 9 and 8 out of 9 correct positive results. In 2004, correct results were ob-
tained in 15 out of 18 participating laboratories and no false-positive results were reported.
Multiplex PCR and NASBA formats scored less samples positive than the monoplex reactions.
This shows that the quality of the molecular tests is improving and that one should carefully
consider whether single tests or multiplex tests will be used.

Diagnostic efficacy in different clinical samples

Clinical samples suitable for M. pneumoniae PCR include nasopharyngeal and oropharyngeal
secretions, sputa, bronchoalveolar lavages (BALs) and throat swabs. Many patients with M.
pneumoniae infection do not produce significant amounts of sputum; especially children fail
to do so. Michelow et al evaluated nasopharyngeal and oropharyngeal samples obtained
from children with serologically proven M. pneumoniae pneumonia and reported that either
specimen type was equally effective for bacterial detection by PCR (29). Still, combining
results from both clinical sites provided the most significant diagnostic yield. One group of
investigators found sputa to be superior sources over nasopharyngeal aspirates and throat
swabs in young adults with serologically proven M. pneumoniae infection (46). However, oth-
ers found no difference in the detection efficiency of M. pneumoniae using PCR in the various
anatomic sites (47). Gnarpe et al compared nasopharyngeal swabs and oropharyngeal swabs
for the detection of M. pneumoniae by PCR (48). A total of 7 patients were sero-positive for
M. pneumoniae and, of these, 6 were positive from oropharyngeal swabs and only 2 were
positive from nasopharyngeal swabs. Another study compared nasopharyngeal swabs to
oropharyngeal swabs in children and found no significant difference in the detection of M.
pneumoniae by PCR (47). The authors did note that nasopharyngeal swabs were more likely
to be problematic than oropharyngeal swabs because of the presence of PCR inhibitors or
lack of respiratory epithelial material. Honda et al applied capillary PCR to sputum, BALs

39

Rene te Witt bw.indd 39 24-11-11 10:35

 Chapter 2

and oropharyngeal swabs (49). The highest rate of detection was shown for oropharyngeal
swabs.
Loens et al combined the results of two studies on LRTIs. From 25 patients both an oro-
pharyngeal swab and a sputum were available for NAAT analysis and culture (43, 50). In
both studies, sputa were the preferred specimens. Räty and co-workers collected sputum, a
nasopharyngeal aspirate and an oropharyngeal swab from 32 young military employees suf-
fering from pneumonia during an M. pneumoniae outbreak and applied PCR (46). This study
also concluded that sputum is the best sample to detect M. pneumoniae. Dorigo-Zetsma et
al confirmed this finding (51). Care should be taken when applying NAAT to sputum samples
since inhibitors occur frequently in sputum and these may be difficult to eliminate (52).
In conclusion, if a sputum sample is available, this might be the most optimal specimen
for M. pneumoniae detection by culture and NAAT. A nasopharyngeal swab, nasopharyngeal
aspirate or oropharyngeal swab might be the second best option for analysis by NAAT.

Legionella spp

Legionella pneumophila was identified as an important causative agent of severe RTI. The
first cases of Legionnaires’ disease or Legionellosis were discovered among attendants of an
American Legion convention in Philadelphia in 1976 (53). Pontiac fever is caused by the same
bacterium. This produces milder respiratory illness without pneumonia.
Legionella spp. are water borne bacterial species that can be found in tap water to which
the first outbreak was attributed, but also in water from cooling towers, air-conditioning,
ventilators and humidifiers (54).

Culture

Culture-based isolation of Legionella spp. from body fluids is still considered the gold standard
for the diagnosis of Legionellosis. Culture requires special enriched media, adequate process-
ing of specimens and technical expertise. Several days are required to obtain a positive result,
with most Legionella spp. colonies being detected within 4-7 days. Legionella species other
than L. pneumophila may grow at a slower rate and may, therefore, be detectable only after 10
days of incubation (55-56). Some Legionella spp. have unusual colony morphology and may
easily be overlooked. The standard medium to culture Legionella is buffered charcoal yeast
extract (BCYE) agar supplemented with α-ketoglutarate, with or without selective antimicrobial
agents. The antibiotics most commonly added are polymyxin to control commensal flora, an-
isomycin against yeasts and cefamandole or vancomycine against Gram-positive bacteria (57).

40

Rene te Witt bw.indd 40 24-11-11 10:35

 Specific diagnostic tests for atypical respiratory tract pathogens

Serological testing

Indirect immunofluoresence assay (IFA) was used to detect antibodies in patients from the
Philadelphia outbreak which turned out to be instrumental in determining the cause of the

Chapter 2
illness. Since then, a number of serological tests has been developed and evaluated (58-68).
Of the various antibody detection methods that are available, IFA and enzyme-linked immu-
nosorbent assays (ELISA) are the most commonly used. Nowadays, many laboratories prefer
ELISA assays over IFA testing, because they are less subjective, more accurate than IFA testing
and have the potential for automated high-throughput performance. The reported sensitivi-
ties vary from 41% to 94% (56). A recent study showed sensitivities of 64%, 61% and 44% for
ELISA, IFA and rapid microagglutination assays, respectively (69). In the ELISA assay, half of
the patients showed a seroconversion in IgM and the other half showed a seroconversion in
IgG. Other studies have shown that the early immune respons primarily involves IgM and that
IgM tests must be included for optimal sensitivity (70-73). Seroconversion may take several
weeks, which is a major limitation of serological testing. In most cases, a fourfold increase in
antibody titre is detected within 3-4 weeks, but in some cases this may take more than 10
weeks (74). Obviously, this severely compromises the timeline of serological testing. Acute-
phase IFA antibody titres of ≥256 during pneumonia were once considered sufficient for a
presumptive diagnosis, but this has been shown to be unreliable, given the high prevalence
of Legionella antibody positivity in persons without clinical evidence for Legionellosis (75).
Another disadvantage of serological testing is its incapability to accurately detect all Legio-
nella species and serogroups. In addition, a diagnosis by a fourfold IgG or IgM titre increase
can only be made retrospectively and cannot support patient management. Therefore, there
is a urgent need for additional tests for Legionella diagnosis in the early stage of disease.

Antigen detection

The detection of Legionella antigen in urine was developed shortly after the first outbreak
in Philadelphia (76). Legionella antigen can be detected in urine as early as 24 hours after
onset of symptoms and antigen shedding persists for days to weeks. The detected antigen is
a component of the lipopolysaccharide portion of the Legionella cell wall (77-78). The urinary
antigen tests combine reasonable sensitivity (80-95%) and high specificity (95-100%) with
very rapid results (79-80). Nowadays, it is the most commonly used laboratory test for Legio-
nella diagnosis (81-82). In Europe, the proportion of cases diagnosed by the urinary antigen
detection has increased rapidly from 15% in 1995, 33% in 1998 and 74% in 2004 to more than
90% in 2006 (83).
Commercial kits that use both radioimmunoassay (RAI) and EIA methodologies have been
available for several years and have similar performance characteristics (55). Agglutination
assays have also been introduced, but these have not yet provided an acceptable level of

41

Rene te Witt bw.indd 41 24-11-11 10:35

 Chapter 2

sensitivity and specificity (84). In addition, immunochromatographic assays (IA) have been
developed that have similar sensitivity and specificity as compared to EIA (85). The major-
ity of IA is most sensitive for the detection of the Pontiac monoclonal antibody type of L.
pneumophila serogroup 1 (up to 90%), less sensitive for other monoclonal antibody types of
L. pneumophila serogroup 1 (to 60%) and poorly sensitive (to 5%) for other L. pneumophila
serogroups and other Legionella species (86-87). An important feature of these assays is its
high specificity (>99%).
The sensitivity of urinary antigen detection appears to be associated with the clinical
severity of the disease (88). Yzerman et al tested two enzyme immunoassays, Binax Legionella
Urinary Antigen EIA (Binax, Portland, Maine) and Biotest Legionella Urin Antigen EIA (Biotest,
Dreieich, Germany) and one immunochromatographic assay, Binax NOW Urinary Antigen Test
(Binax, Portland, Maine), using urine samples from outbreak related Legionellosis patients.
For patients with mild Legionellosis, the test sensitivities ranged from 40-53%, whereas for
patients with severe Legionellosis, the sensitivities reached 88-100%. These findings have
implications for the diagnostic process in patients with mild pneumonia and suggest that
patients with mild pneumonia may be underdiagnosed if only the urine antigen test is used.
The use of concentrated urine samples increased sensitivity without decreasing the speci-
ficity. Since this concentration step is timely and laborious, some laboratories only use this
approach in case of equivocal results or strong clinical indication.
Another association between test sensitivity and certain defined subpopulations has been
described by Helbig et al (85). The clinical utility of Legionella urinary antigen assays for the
diagnosis of Legionnair’s disease has been assessed by using samples from 317 culture-
proven cases. The sensitivities of the Binax EIA and Biotest EIA urinary tests were found to
be 94% and 94% for travel-associated infection and 87% and 76% for community-acquired
infection, but only 44% and 46% for hospital-acquired infection.
Several new immunochromatographic urinary antigen tests for the detection of L. pneu-
mophila serogroup 1 have been developed (79, 89). The Binax NOW urinary antigen test, in
concordance with the findings of previous studies, has excellent sensitivity and specificity.
The performance of some new tests is below the acceptable level for diagnostic assays (Table
3) (89).

Molecular testing

The first assay designed to detect DNA of L. pneumophila, was based on RIA in which a ra-
diolabeled ribosomal probe specific for all strains of Legionella spp. was applied. Researchers
reported a varying sensitivity and specificity for this assay (90-91). The use of this probe for
the detection of Legionellosis at one hospital resulted in 13 false-positive cases and the assay
was removed from the market soon after it falsely recorded this pseudo outbreak (92).

42

Rene te Witt bw.indd 42 24-11-11 10:35

 Specific diagnostic tests for atypical respiratory tract pathogens

Table 3. Performance of urinary antigen detection tests.

Urinary antigen test Sensitivity Specificity Reference
NCU CU
SAS Legionella Test (SA Scientific, San Antonio, Texas) 82,9% NT 99,0% (87)
Binax NOW Urinary Antigen Test (Binax, Portland, Maine) 91,4% NT 100,0% (87)

Chapter 2
Binax Legionella Urinary Antigen EIA (Binax, Portland, Maine) 97,1% NT NT (88)
Biotest Legionella Urin Antigen EIA (Biotest, Dreieich, Germany) 91,4% NT NT (88)
Binax NOW Urinary Antigen Test (Binax, Portland, Maine) 94,3% NT NT (88)
Binax Legionella Urinary Antigen EIA (Binax, Portland, Maine) 86,5% NT NT (93)
Biotest Legionella Urin Antigen EIA (Biotest, Dreieich, Germany) 76,0% NT NT (93)
Biotest EIA (Biotest, Dreieich, Germany) 94,6% NT 100,0% (94)
Bartels EIA (Bartels Inc. Trinity Biotech Company, Wicklow, Ireland) 74,1% 91,5% 100,0% (95)
Biotest Legionella Urin Antigen EIA (Biotest, Dreieich, Germany) 51,7% 91,5% 100,0% (95)
Biotest Legionella Urin Antigen EIA (Biotest, Dreieich, Germany) 71,0% 74,0% NT (96)
Binax Legionella Urinary Antigen EIA (Binax, Portland, Maine) 69,0% 79,0% NT (96)
Binax NOW Urinary Antigen Test (Binax, Portland, Maine) 72,0% 81,0% NT (96)
Rapid U Legionella Antigen Test (Diamondial, Sees, France) 71,2% NT 96,6% (97)
SD Bioline Legionella Urinary Antigen Test 31,5% NT 98,9% (97)
(Standard Diagnostics Inc., Kyonggi-do, Korea)
Binax NOW Urinary Antigen Test (Binax, Portland, Maine) 91,8% NT 100,0% (97)
NCU: non-concentrated urine
CU: concentrated urine
NT: not tested

Legionella PCR is available in an increasing number of laboratories that use a variety of

in-house or commercial assays. Diagnostic PCR assays target specific Legionella pneumophila
DNA regions within 16S rRNA genes (93-100), the 23S-5S spacer (101), 5S rDNA (102-103)
or the macrophage inhibitor potentiator (mip) gene (104-107). The application of PCR to
non-respiratory samples seems particularly attractive for patients who do not produce
sputum. Legionella DNA can be detected in urine, serum and leukocyte samples obtained
from patients with Legionellosis, with sensitivities ranging from 30% to 86% (102, 108-110).
The sensitivity of the detection of Legionella DNA in serum is relatively low (50-60%) in
Legionellosis patients, but was shown to be higher (70-90%) in patients with more severe
disease (108-109). When testing samples from the lower respiratory tract (bronchoalveolar
lavage), PCR has repeatedly shown to have a sensitivity equal to or higher than culture (99-
100, 111). Indeed, PCR has been considered by some authors to be the test-of-choice for
patients who produce sputum (111). However, a number of false positive results has been
reported, both with commercially available tests and with in-house tests (55, 99). A problem
with the interpretation of these false positive results is the question whether these are truly
false positive or whether it was the reference method that failed. It is difficult to solve this
issue and, at present, there is only one study available were the authors have determined the
exact sensitivity and specificity of Legionella PCR in patients with pneumonia of unknown
aetiology (112). Diederen et al demonstrated that variation of DNA targets influences the test

43

Rene te Witt bw.indd 43 24-11-11 10:35

 Chapter 2

performance. PCR designed on the detection of 16S RNA had a sensitivity and specificity of
86% and 95%, as for PCR designed to detect the mip gene, this was 92% and 98%, respec-
tively. In conclusion, laboratory workers and clinicians must be cautious when interpreting
results and should not hesitate to question the results, especially when these results are
unexpected based on clinical presentation and local epidemiology. Only one study describes
a multicenter comparison of molecular methods for detection of Legionella spp. (113). The
authors compared the methods of 9 laboratories for 12 sputum samples with L. pneumophila
or Legionella longbeachae and conclude that PCR targeting the mip gene is L. pneumophila
specific and 16S rRNA gene amplification is genus-specific.

Relevance of sample type

Legionella spp. can be isolated from a variety of sample types, although sputum and bron-
choalveolar lavage are the samples of choice. Sputum samples are generally considered to
be optimal for isolation of L. pneumophila in patients with RTI. Culture results depend on
the severity of illness, with the lowest result (15-25%) for mild pneumonia and the highest
result (>90%) for severe pneumonia (111). A major limitation of culturing the pathogen from
sputum is the fact that less than 50% of Legionella patients produce sufficient amounts of
sputum (111, 114). Most patients with Legionellosis produce non-purulent sputum. Obvious-
ly, laboratories that reject sputum samples containing limited mucosal polymorphonuclear
leukocytes may reject potential positive samples. Ingram and Plouffe demonstrated that
up to 84% of L. pneumophila positive samples would have been discarded by using sputum
purulence screens and they recommended acceptance of all aspects of sputum suspected for
Legionella culture (115). Estimated sensitivities of sputum culture range from <10% to 80%
and vary between individual laboratories (55, 111).
In conclusion, the urinary antigen detection is currently the most helpful rapid test for the
diagnosis of Legionella infection. The use of the rapid urinary antigen tests reduces mortality
and avoids unnecessary or inappropriate use of antibiotics in patients with CAP. However,
combining test results from more than one sampling site appears to improve the diagnostic
accuracy. Especially in this diagnostic segment, molecular testing may play an important
(future) role.

44

Rene te Witt bw.indd 44 24-11-11 10:35

 Specific diagnostic tests for atypical respiratory tract pathogens

Chlamydophila pneumoniae

Culture

Chapter 2
For C. pneumoniae, culture on cell lines has traditionally been considered as a reference and
standard diagnostic method. However, due to the important limitations in the cultivation
of C. pneumoniae (technical complexity, limited viability of the bacteria, slow growth and
variable diagnostic success) performance of culture remains restricted to specialized labo-
ratories. Hence, the use of culture as a diagnostic tool is suboptimal and not often recom-
mended. Still, the most common method for diagnosis of C. pneumoniae infection is serology.
Assays available for detection of C. pneumoniae specific antibodies include Micro Immuno
Fluorescence (MIF) tests, ELISAs and EIAs, each of which exists in a variety of in-house and
commercial variations.
Recommendations by the Centre for Disease Control (CDC) and the practical guidelines of
the Infectious Disease Society of America (IDSA) defined the main criteria for the diagnosis
of acute C. pneumoniae infection as a single IgM titre of ≥1:64 or a fourfold increase in the
IgG titre in acute serum and convalescent serum, measured 4 weeks apart from each other.
The use of single IgG or IgA titres is discouraged because of the relatively high overall serop-
revalence in healthy populations (116). However, several studies deviated widely from these
guidelines and there are several inherent limitations to the serodiagnosis of C. pneumoniae
infection (117-121).

Serological testing

Serological testing at best offers a retrospective diagnosis. In primary infections, IgM antibodies
appear 2-3 weeks and IgG antibodies appear 6-8 weeks after infection, whereas in reinfections,
IgM may be absent or of low titre and IgG appears earlier, within 2-3 weeks after infection.
The MIF assay has been repeatedly demonstrated to be insensitive and it showed a poor
correlation with the detection of C. pneumoniae by culture or PCR, particularly in children.
Only 1%-3% of culture positive children in a study by Hammerschlag met the serological
criteria for acute infection (122). Wellinghausen et al. reported that 17 patients with CAP, had
seronegative results with MIF (119).
Generally, the specificity of serological testing may be suboptimal as well. Serological
evidence of acute infection was found in 19% of healthy adults who had negative results
by culture and PCR (122). This lack of specificity may result from serological cross-reactivity
with other Chlamydia species, as well as Mycoplasma spp. or Bartonella spp. In addition, other
limitations of the MIF test relate to a lack of standardized reagents, technical complexity and
subjective endpoint determination, all of which result in significant intra- and interlaboratory
variation of test performance. One study evaluated the interlaboratory reliability of the MIF

45

Rene te Witt bw.indd 45 24-11-11 10:35

 Chapter 2

test for measurement of C. pneumoniae specific IgA and IgG titres for 392 serum samples,
using reagents and antigens obtained from a common source. The investigators observed
agreement between IgA and IgG titres to be as low as 55% and 38%, respectively (123).
EIAs may overcome some of the limitations of the MIF test by being more objective in
the interpretation of the results and less technically demanding. Hermann et al. compared
7 commercial EIAs or ELISAs with 4 MIF assays for detection of specific IgG antibodies (124).
The authors used serum samples from 80 healthy subjects and reported sensitivities and
specificities ranging from 42% to 100% and from 88% to 100%, respectively. The SeroCP
ELISA (Savyon Diagnostic, Ashdod, Israel) and Quant EIA (Savyon, Ashdod, Israel) showed the
best sensitivities, 96% and 92%, respectively, followed by the Vircell ELISA (Viva Diagnostica,
Köln-Hürth, Germany) and the Labsystems EIA (Labsystems, Helsinki, Finland), which each
had a sensitivity of about 75%.

Molecular testing

For reasons of the above-mentioned issues with culture and serology, PCR can be an interest-
ing alternative for diagnosis of C. pneumoniae infections. The first reports of PCR application
for the detection of C. pneumoniae appeared in 1990 and 1992 (125-126). Since then, more
than 250 studies describing the use of PCR for detection of C. pneumoniae in human infec-
tions have been published. Gene targets used in various types of PCR for C. pneumoniae
include the ompA gene (127), pstI (126), pmp4 (128) and 16S rDNA (129). Real-time PCR assays
have also been described (128, 130-134). Multicenter studies that use a large and diverse
repertoire of clinical specimens and compare data independently are likely to provide impor-
tant insights into the performance of new assays. Two such studies describing multicenter
comparisons of the performance of various NAAT test for detection of C. pneumoniae in re-
spiratory specimens have been published. Both studies revealed significant variations of test
performance from laboratory to laboratory. Cherneskey et al compared a C. pneumoniae PCR
kit from Abbot (Abbott Laboratories, North Chicago, Illinois) with 5 conventional PCR assays,
using specimens spiked with pre-extracted DNA (135). Loens et al used spiked respiratory
specimens to compare the performance of several in-house PCR assays (41). Correct results
were produced in 12 out of 16 and 13 out of 18 tests in 2002 and 2004, respectively. Both of
these studies revealed significant intercenter discordance of detection rates, using different
or even the same tests. Both multiplex PCR and NASBA formats scored a smaller number of
positive samples than the monoplex tests.

Relevance of sample type

The choice of a specimen from the respiratory tract has an impact on the sensitivity of C.
pneumoniae isolation and detection by culture and PCR. In a study in which the authors

46

Rene te Witt bw.indd 46 24-11-11 10:35

 Specific diagnostic tests for atypical respiratory tract pathogens

enrolled 260 previously healthy children (3-12 years), it was shown that the nasopharynx
might be superior to the throat as a source of materials to be used for isolation of C. pneu-
moniae. Of 34 children from whom C. pneumoniae was isolated, nasopharyngeal swabs were
positive for all children, but oropharyngeal swabs were positive for only 50% of the same

Chapter 2
group (136). During a C. pneumoniae outbreak, Boman et al collected sputum, oropharyngeal
swabs and nasopharyngeal swabs from 116 patients presenting with RTI (137). When the
authors compared the performances of PCR, culture and antigen detection for samples from
three different niches in 61 patients for whom all samples were available, 20 patients were
positive for C. pneumoniae, for whom 7 nasopharyngeal swabs, 10 oropharyngeal swabs
and 20 sputum samples were considered to be true positives. Sensitivities of PCR, culture
and antigen detection by EIA for sputum samples were 95%, 100% and 80%, respectively.
Sensitivities for the other types of sampling sites were much lower (25%-50%). The clinical
relevance of sputum for the detection of C. pneumoniae was confirmed by Kuoppa et al (131).
In this study, a sputum sample, a nasopharynx aspirate and an oropharyngeal swab from 35
patients suspected of having a C. pneumoniae infection were examined by PCR. The majority
of all samples had C. pneumoniae DNA copies below 1x104 genome copies per ml, but the
majority of the sputum samples contained higher inocula of C. pneumoniae DNA, with an
average of 8.6x105 copies/ml. However, these results are in contrast with those obtained by
Verkooyen et al, who examined sputum, nasopharyngeal swabs, oropharyngeal swabs and
throat wash specimens from 156 hospitalized CAP patients by PCR and culture (138). The
highest sensitivity in this study was obtained by applying PCR on nasopharyngeal swabs
(51.3%). Surprisingly, none of the sputum samples tested positive.
Gnarpe et al compared PCR results for C. pneumoniae from nasopharyngeal swabs and
oropharyngeal swabs in 66 patients presenting with RTI (48). Of a total of 18 patients positive
for C. pneumoniae, in 15 patients the oropharyngeal swab was the only positive specimen,
whereas for 3 patients both the oropharyngeal swab and the nasopharyngeal swab yielded
a positive PCR result.
In conclusion, sputum may be the preferred specimen for detection of C. pneumoniae by
NAAT.

Concluding remarks

Historically, atypical agents of RTI were merely detected on the basis of microbiological
culture. Without exception, sensitivity and specificity of such cultures did not meet criteria of
excellence. Hence, a variety of antigen- or antibody mediated tests was developed. Unfortu-
nately, none of these individual tests was in itself sufficient to reliably identify the causative
infectious agents. Also the more recent availability of a variety of NAATs did not (yet) solve
this problem. In today’s clinical microbiology laboratories no widely accepted gold standard

47

Rene te Witt bw.indd 47 24-11-11 10:35

 Chapter 2

for the detection of M. pneumoniae, C. pneumoniae and Legionella spp. is available. Techni-
cians and physicians have to rely on a combination of test results that, together with clinical
presentation of the patient, may lead to a presumptive identification of a causative agent at
best. Although great diagnostic improvements have been made over the last twenty years,
no final tool is as yet available. Future biomarker discovery is still required before the ultimate
test for the diagnosis of atypical RTI agents will become available.

48

Rene te Witt bw.indd 48 24-11-11 10:35

 Specific diagnostic tests for atypical respiratory tract pathogens

References
1. File, T. M., Jr., Tan, J. S. and Plouffe, J. F. (1998). The role of atypical pathogens: Mycoplasma pneumoniae, Chlamydia
pneumoniae, and Legionella pneumophila in respiratory infection. Infect Dis Clin North Am 12(3): 569-92, vii.
2. Blasi, F. (2004). Atypical pathogens and respiratory tract infections. Eur Respir J 24(1): 171-81.

Chapter 2
3. Charles, P. G. (2008). Early diagnosis of lower respiratory tract infections (point-of-care tests). Curr Opin Pulm Med
14(3): 176-82.
4. Nicolau, D. P. (2004). Treatment with appropriate antibiotic therapy in community-acquired respiratory tract infec-
tions. Am J Manag Care 10(12 Suppl): S381-8.
5. Boersma, W. G., et al. (2006). Reliability of radiographic findings and the relation to etiologic agents in community-
acquired pneumonia. Respir Med 100(5): 926-32.
6. Oosterheert, J. J., et al. (2005). Predicted effects on antibiotic use following the introduction of British or North
American guidelines for community-acquired pneumonia in The Netherlands. Clin Microbiol Infect 11(12): 992-8.
7. van der Eerden, M. M., et al. (2005). Comparison between pathogen directed antibiotic treatment and empirical
broad spectrum antibiotic treatment in patients with community acquired pneumonia: a prospective randomised
study. Thorax 60(8): 672-8.
8. Reijans, M., et al. (2008). RespiFinder: a new multiparameter test to differentially identify fifteen respiratory viruses.
J Clin Microbiol 46(4): 1232-40.
9. Gavranich, J. B. and Chang, A. B. (2005). Antibiotics for community acquired lower respiratory tract infections (LRTI)
secondary to Mycoplasma pneumoniae in children. Cochrane Database Syst Rev (3): CD004875.
10. Talkington, D. F., et al. (2004). Analysis of eight commercial enzyme immunoassay tests for detection of antibodies
to Mycoplasma pneumoniae in human serum. Clin Diagn Lab Immunol 11(5): 862-7.
11. Ozaki, T., et al. (2007). Utility of a rapid diagnosis kit for Mycoplasma pneumoniae pneumonia in children, and the
antimicrobial susceptibility of the isolates. J Infect Chemother 13(4): 204-7.
12. Nir-Paz, R., et al. (2006). Evaluation of eight commercial tests for Mycoplasma pneumoniae antibodies in the
absence of acute infection. Clin Microbiol Infect 12(7): 685-8.
13. Beersma, M. F., et al. (2005). Evaluation of 12 commercial tests and the complement fixation test for Mycoplasma
pneumoniae-specific immunoglobulin G (IgG) and IgM antibodies, with PCR used as the “gold standard”. J Clin
Microbiol 43(5): 2277-85.
14. Yoo, S. J., Oh, H. J. and Shin, B. M. (2007). Evaluation of four commercial IgG- and IgM-specific enzyme immunoas-
says for detecting Mycoplasma pneumoniae antibody: comparison with particle agglutination assay. J Korean Med
Sci 22(5): 795-801.
15. Csango, P. A., Pedersen, J. E. and Hess, R. D. (2004). Comparison of four Mycoplasma pneumoniae IgM-, IgG- and
IgA-specific enzyme immunoassays in blood donors and patients. Clin Microbiol Infect 10(12): 1094-8.
16. Souliou, E., et al. (2007). Laboratory diagnosis of Mycoplasma pneumoniae respiratory tract infections in children.
Eur J Clin Microbiol Infect Dis 26(7): 513-5.
17. Jensen, J. S., et al. (1989). Detection of Mycoplasma pneumoniae in simulated clinical samples by polymerase chain
reaction. Brief report. Apmis 97(11): 1046-8.
18. Bernet, C., et al. (1989). Detection of Mycoplasma pneumoniae by using the polymerase chain reaction. J Clin
Microbiol 27(11): 2492-6.
19. van Kuppeveld, F. J., et al. (1994). 16S rRNA based polymerase chain reaction compared with culture and serologi-
cal methods for diagnosis of Mycoplasma pneumoniae infection. Eur J Clin Microbiol Infect Dis 13(5): 401-5.

49

Rene te Witt bw.indd 49 24-11-11 10:35

 Chapter 2

20. Luneberg, E., Jensen, J. S. and Frosch, M. (1993). Detection of Mycoplasma pneumoniae by polymerase chain reac-
tion and nonradioactive hybridization in microtiter plates. J Clin Microbiol 31(5): 1088-94.
21. Dumke, R., et al. (2007). Sensitive detection of Mycoplasma pneumoniae in human respiratory tract samples by
optimized real-time PCR approach. J Clin Microbiol 45(8): 2726-30.
22. Hardegger, D., et al. (2000). Rapid detection of Mycoplasma pneumoniae in clinical samples by real-time PCR. J
Microbiol Methods 41(1): 45-51.
23. Templeton, K. E., et al. (2003). Comparison and evaluation of real-time PCR, real-time nucleic acid sequence-based
amplification, conventional PCR, and serology for diagnosis of Mycoplasma pneumoniae. J Clin Microbiol 41(9):
4366-71.
24. Ursi, D., et al. (2003). Detection of Mycoplasma pneumoniae in respiratory samples by real-time PCR using an inhibi-
tion control. J Microbiol Methods 55(1): 149-53.
25. Pitcher, D., et al. (2006). Real-time detection of Mycoplasma pneumoniae in respiratory samples with an internal
processing control. J Med Microbiol 55(Pt 2): 149-55.
26. Morozumi, M., et al. (2006). Assessment of real-time PCR for diagnosis of Mycoplasma pneumoniae pneumonia in
pediatric patients. Can J Microbiol 52(2): 125-9.
27. Di Marco, E., et al. (2007). Development and clinical validation of a real-time PCR using a uni-molecular Scorpion-
based probe for the detection of Mycoplasma pneumoniae in clinical isolates. New Microbiol 30(4): 415-21.
28. Nilsson, A. C., Bjorkman, P. and Persson, K. (2008). Polymerase chain reaction is superior to serology for the diagno-
sis of acute Mycoplasma pneumoniae infection and reveals a high rate of persistent infection. BMC Microbiol 8: 93.
29. Michelow, I. C., et al. (2004). Diagnostic utility and clinical significance of naso- and oropharyngeal samples used
in a PCR assay to diagnose Mycoplasma pneumoniae infection in children with community-acquired pneumonia. J
Clin Microbiol 42(7): 3339-41.
30. Tong, C. Y., et al. (1999). Multiplex polymerase chain reaction for the simultaneous detection of Mycoplasma
pneumoniae, Chlamydia pneumoniae, and Chlamydia psittaci in respiratory samples. J Clin Pathol 52(4): 257-63.
31. Corsaro, D., et al. (1999). Multiplex PCR for rapid and differential diagnosis of Mycoplasma pneumoniae and Chla-
mydia pneumoniae in respiratory infections. Diagn Microbiol Infect Dis 35(2): 105-8.
32. Grondahl, B., et al. (1999). Rapid identification of nine microorganisms causing acute respiratory tract infections by
single-tube multiplex reverse transcription-PCR: feasibility study. J Clin Microbiol 37(1): 1-7.
33. Welti, M., et al. (2003). Development of a multiplex real-time quantitative PCR assay to detect Chlamydia pneu-
moniae, Legionella pneumophila and Mycoplasma pneumoniae in respiratory tract secretions. Diagn Microbiol
Infect Dis 45(2): 85-95.
34. Ginevra, C., et al. (2005). Development and evaluation of Chlamylege, a new commercial test allowing simulta-
neous detection and identification of Legionella, Chlamydophila pneumoniae, and Mycoplasma pneumoniae in
clinical respiratory specimens by multiplex PCR. J Clin Microbiol 43(7): 3247-54.
35. Khanna, M., et al. (2005). The pneumoplex assays, a multiplex PCR-enzyme hybridization assay that allows simulta-
neous detection of five organisms, Mycoplasma pneumoniae, Chlamydia (Chlamydophila) pneumoniae, Legionella
pneumophila, Legionella micdadei, and Bordetella pertussis, and its real-time counterpart. J Clin Microbiol 43(2):
565-71.
36. McDonough, E. A., et al. (2005). A multiplex PCR for detection of Mycoplasma pneumoniae, Chlamydophila pneu-
moniae, Legionella pneumophila, and Bordetella pertussis in clinical specimens. Mol Cell Probes 19(5): 314-22.
37. Raggam, R. B., et al. (2005). Single-run, parallel detection of DNA from three pneumonia-producing bacteria by
real-time polymerase chain reaction. J Mol Diagn 7(1): 133-8.

50

Rene te Witt bw.indd 50 24-11-11 10:35

 Specific diagnostic tests for atypical respiratory tract pathogens

38. Stralin, K., et al. (2005). Design of a multiplex PCR for Streptococcus pneumoniae, Haemophilus influenzae, Myco-
plasma pneumoniae and Chlamydophila pneumoniae to be used on sputum samples. Apmis 113(2): 99-111.
39. Geertsen, R., et al. (2007). A multiplex PCR assay for the detection of respiratory bacteriae in nasopharyngeal
smears from children with acute respiratory disease. Scand J Infect Dis 39(9): 769-74.
40. Wang, Y., et al. (2008). A multiplex PCR-based reverse line blot hybridization (mPCR/RLB) assay for detection of

Chapter 2
bacterial respiratory pathogens in children with pneumonia. Pediatr Pulmonol 43(2): 150-9.
41. Loens, K., et al. (2006). Two quality control exercises involving nucleic acid amplification methods for detection of
Mycoplasma pneumoniae and Chlamydophila pneumoniae and carried out 2 years apart (in 2002 and 2004). J Clin
Microbiol 44(3): 899-908.
42. Loens, K., et al. (2003). Detection of Mycoplasma pneumoniae by real-time nucleic acid sequence-based amplifica-
tion. J Clin Microbiol 41(9): 4448-50.
43. Loens, K., et al. (2008). Development of real-time multiplex nucleic acid sequence-based amplification for detec-
tion of Mycoplasma pneumoniae, Chlamydophila pneumoniae, and Legionella spp. in respiratory specimens. J Clin
Microbiol 46(1): 185-91.
44. Kumar, S., et al. (2008). Detection of 11 common viral and bacterial pathogens causing community-acquired
pneumonia or sepsis in asymptomatic patients by using a multiplex reverse transcription-PCR assay with manual
(enzyme hybridization) or automated (electronic microarray) detection. J Clin Microbiol 46(9): 3063-72.
45. Atkinson, T. P., Balish, M. F. and Waites, K. B. (2008). Epidemiology, clinical manifestations, pathogenesis and labora-
tory detection of Mycoplasma pneumoniae infections. FEMS Microbiol Rev.
46. Raty, R., Ronkko, E. and Kleemola, M. (2005). Sample type is crucial to the diagnosis of Mycoplasma pneumoniae
pneumonia by PCR. J Med Microbiol 54(Pt 3): 287-91.
47. Reznikov, M., et al. (1995). Comparison of nasopharyngeal aspirates and throat swab specimens in a polymerase
chain reaction-based test for Mycoplasma pneumoniae. Eur J Clin Microbiol Infect Dis 14(1): 58-61.
48. Gnarpe, J., et al. (1997). Comparison of nasopharyngeal and throat swabs for the detection of Chlamydia pneu-
moniae and Mycoplasma pneumoniae by polymerase chain reaction. Scand J Infect Dis Suppl 104: 11-2.
49. Honda, J., et al. (2000). Clinical use of capillary PCR to diagnose Mycoplasma pneumonia. J Clin Microbiol 38(4):
1382-4.
50. Loens, K., et al. (2008). Evaluation of different nucleic acid amplification techniques for the detection of M.
pneumoniae, C. pneumoniae and Legionella spp. in respiratory specimens from patients with community-acquired
pneumonia. J Microbiol Methods 73(3): 257-62.
51. Dorigo-Zetsma, J. W., et al. (2001). Molecular detection of Mycoplasma pneumoniae in adults with community-
acquired pneumonia requiring hospitalization. J Clin Microbiol 39(3): 1184-6.
52. Loens, K., et al. (2003). Molecular diagnosis of Mycoplasma pneumoniae respiratory tract infections. J Clin Microbiol
41(11): 4915-23.
53. Fraser, D. W., et al. (1977). Legionnaires’ disease: description of an epidemic of pneumonia. N Engl J Med 297(22):
1189-97.
54. Winn, W. C., Jr. (1988). Legionnaires disease: historical perspective. Clin Microbiol Rev 1(1): 60-81.
55. Fields, B. S., Benson, R. F. and Besser, R. E. (2002). Legionella and Legionnaires’ disease: 25 years of investigation. Clin
Microbiol Rev 15(3): 506-26.
56. Den Boer, J. W. and Yzerman, E. P. (2004). Diagnosis of Legionella infection in Legionnaires’ disease. Eur J Clin
Microbiol Infect Dis 23(12): 871-8.
57. Doern, G. V. (2000). Detection of selected fastidious bacteria. Clin Infect Dis 30(1): 166-73.

51

Rene te Witt bw.indd 51 24-11-11 10:35

 Chapter 2

58. Lennette, D. A., et al. (1979). Serology of Legionnaires disease: comparison of indirect fluorescent antibody, im-
mune adherence hemagglutination, and indirect hemagglutination tests. J Clin Microbiol 10(6): 876-9.
59. Holliday, M. G. (1980). The diagnosis of Legionnaires’ disease by counterimmunoelectrophoresis. J Clin Pathol
33(12): 1174-8.
60. Yonke, C. A., et al. (1981). Evaluation of an indirect hemagglutination test for Legionella pneumophila serogroups 1
to 4. J Clin Microbiol 13(6): 1040-5.
61. Soriano, F., Aguilar, L. and Gomez Garces, J. L. (1982). Simple immunodiffusion test for detecting antibodies against
Legionella pneumophila serotype 1. J Clin Microbiol 15(2): 330-1.
62. Wreghitt, T. G., Nagington, J. and Gray, J. (1982). An ELISA test for the detection of antibodies to Legionella pneu-
mophila. J Clin Pathol 35(6): 657-60.
63. Thompson, T. A. and Wilkinson, H. W. (1982). Evaluation of a solid-phase immunofluorescence assay for detection
of antibodies to Legionella pneumophila. J Clin Microbiol 16(1): 202-4.
64. Harrison, T. G. and Taylor, A. G. (1982). A rapid microagglutination test for the diagnosis of Legionella pneumophila
(serogroup 1) infection. J Clin Pathol 35(9): 1028-31.
65. Elder, E. M., et al. (1983). Microenzyme-linked immunosorbent assay for detection of immunoglobulin G and im-
munoglobulin M antibodies to Legionella pneumophila. J Clin Microbiol 17(1): 112-21.
66. Herbrink, P., et al. (1983). Detection of antibodies against Legionella pneumophila serogroups 1 to 6 and the
Leiden-1 strain by micro ELISA and immunofluorescence assay. J Clin Pathol 36(11): 1246-52.
67. Sampson, J. S., et al. (1983). Kinetic-dependent enzyme-linked immunosorbent assay for detection of antibodies
to Legionella pneumophila. J Clin Microbiol 18(6): 1340-4.
68. Barka, N., Tomasi, J. P. and Stadtsbaeder, S. (1986). ELISA using whole Legionella pneumophila cell as antigen. Com-
parison between monovalent and polyvalent antigens for the serodiagnosis of human legionellosis. J Immunol
Methods 93(1): 77-81.
69. Yzerman, E. P., et al. (2006). Sensitivity of three serum antibody tests in a large outbreak of Legionnaires’ disease in
the Netherlands. J Med Microbiol 55(Pt 5): 561-6.
70. Hartigan, D. A. (1981). Comparison of specific immunoglobulin G, M and agglutinating antibodies against Legio-
nella pneumophila. Scand J Infect Dis 13(4): 269-72.
71. Zimmerman, S. E., et al. (1982). Immunoglobulin M antibody titers in the diagnosis of Legionnaires disease. J Clin
Microbiol 16(6): 1007-11.
72. De Ory, F., et al. (2000). Detection of specific IgM antibody in the investigation of an outbreak of pneumonia due
to Legionella pneumophila serogroup 1. Clin Microbiol Infect 6(2): 64-9.
73. Rojas, A., et al. (2005). Value of serological testing for diagnosis of legionellosis in outbreak patients. J Clin Microbiol
43(8): 4022-5.
74. Monforte, R., et al. (1988). Delayed seroconversion in Legionnaire’s disease. Lancet 2(8609): 513.
75. Plouffe, J. F., et al. (1995). Reevaluation of the definition of Legionnaires’ disease: use of the urinary antigen assay.
Community Based Pneumonia Incidence Study Group. Clin Infect Dis 20(5): 1286-91.
76. Berdal, B. P., Farshy, C. E. and Feeley, J. C. (1979). Detection of Legionella pneumonophila antigen in urine by
enzyme-linked immunospecific assay. J Clin Microbiol 9(5): 575-8.
77. Kohler, R. B., et al. (1981). Rapid radioimmunoassay diagnosis of Legionnaires’ disease: detection and partial
characterization of urinary antigen. Ann Intern Med 94(5): 601-5.
78. Williams, A. and Lever, M. S. (1995). Characterisation of Legionella pneumophila antigen in urine of guinea pigs and
humans with Legionnaires’ disease. J Infect 30(1): 13-6.

52

Rene te Witt bw.indd 52 24-11-11 10:35

 Specific diagnostic tests for atypical respiratory tract pathogens

79. Diederen, B. M. and Peeters, M. F. (2007). Evaluation of the SAS Legionella Test, a new immunochromatographic
assay for the detection of Legionella pneumophila serogroup 1 antigen in urine. Clin Microbiol Infect 13(1): 86-8.
80. Koide, M., et al. (2006). Detection of Legionella species in clinical samples: Comparison of polymerase chain reac-
tion and urinary antigen detection kits. Infection 34(5): 264-8.
81. Joseph, C. A. (2004). Legionnaires’ disease in Europe 2000-2002. Epidemiol Infect 132(3): 417-24.

Chapter 2
82. Formica, N., et al. (2001). The impact of diagnosis by Legionella urinary antigen test on the epidemiology and
outcomes of Legionnaires’ disease. Epidemiol Infect 127(2): 275-80.
83. Diederen, B. M. (2008). Legionella spp. and Legionnaires’ disease. J Infect 56(1): 1-12.
84. Leland, D. S. and Kohler, R. B. (1991). Evaluation of the L-CLONE Legionella pneumophila Serogroup 1 Urine Antigen
Latex Test. J Clin Microbiol 29(10): 2220-3.
85. Helbig, J. H., et al. (2003). Clinical utility of urinary antigen detection for diagnosis of community-acquired, travel-
associated, and nosocomial legionnaires’ disease. J Clin Microbiol 41(2): 838-40.
86. Harrison, T., et al. (1998). A multicenter evaluation of the Biotest Legionella urinary antigen EIA. Clin Microbiol Infect
4(7): 359-365.
87. Dominguez, J., et al. (2001). Assessment of a new test to detect Legionella urinary antigen for the diagnosis of
Legionnaires’ Disease. Diagn Microbiol Infect Dis 41(4): 199-203.
88. Yzerman, E. P., et al. (2002). Sensitivity of three urinary antigen tests associated with clinical severity in a large
outbreak of Legionnaires’ disease in The Netherlands. J Clin Microbiol 40(9): 3232-6.
89. Diederen, B. M. and Peeters, M. F. (2006). Evaluation of two new immunochromatographic assays (Rapid U Le-
gionella antigen test and SD Bioline Legionella antigen test) for detection of Legionella pneumophila serogroup 1
antigen in urine. J Clin Microbiol 44(8): 2991-3.
90. Doebbeling, B. N., et al. (1988). Prospective evaluation of the Gen-Probe assay for detection of Legionellae in
respiratory specimens. Eur J Clin Microbiol Infect Dis 7(6): 748-52.
91. Wilkinson, H. W., Sampson, J. S. and Plikaytis, B. B. (1986). Evaluation of a commercial gene probe for identification
of Legionella cultures. J Clin Microbiol 23(2): 217-20.
92. Laussucq, S., et al. (1988). False-positive DNA probe test for Legionella species associated with a cluster of respira-
tory illnesses. J Clin Microbiol 26(8): 1442-4.
93. Jonas, D., et al. (1995). Enzyme-linked immunoassay for detection of PCR-amplified DNA of Legionellae in bron-
choalveolar fluid. J Clin Microbiol 33(5): 1247-52.
94. Reischl, U., et al. (2002). Direct detection and differentiation of Legionella spp. and Legionella pneumophila in clini-
cal specimens by dual-color real-time PCR and melting curve analysis. J Clin Microbiol 40(10): 3814-7.
95. Rantakokko-Jalava, K. and Jalava, J. (2001). Development of conventional and real-time PCR assays for detection
of Legionella DNA in respiratory specimens. J Clin Microbiol 39(8): 2904-10.
96. Wellinghausen, N., Frost, C. and Marre, R. (2001). Detection of Legionellae in hospital water samples by quantitative
real-time LightCycler PCR. Appl Environ Microbiol 67(9): 3985-93.
97. Stolhaug, A. and Bergh, K. (2006). Identification and differentiation of Legionella pneumophila and Legionella
spp. with real-time PCR targeting the 16S rRNA gene and species identification by mip sequencing. Appl Environ
Microbiol 72(9): 6394-8.
98. van Der Zee, A., et al. (2002). Novel PCR-probe assay for detection of and discrimination between Legionella
pneumophila and other Legionella species in clinical samples. J Clin Microbiol 40(3): 1124-5.
99. Cloud, J. L., et al. (2000). Detection of Legionella species in respiratory specimens using PCR with sequencing
confirmation. J Clin Microbiol 38(5): 1709-12.

53

Rene te Witt bw.indd 53 24-11-11 10:35

 Chapter 2

100. Templeton, K. E., et al. (2003). Development and clinical evaluation of an internally controlled, single-tube multi-
plex real-time PCR assay for detection of Legionella pneumophila and other Legionella species. J Clin Microbiol 41(9):
4016-21.
101. Herpers, B. L., et al. (2003). Real-time PCR assay targets the 23S-5S spacer for direct detection and differentiation
of Legionella spp. and Legionella pneumophila. J Clin Microbiol 41(10): 4815-6.
102. Murdoch, D. R., et al. (1996). Use of the polymerase chain reaction to detect Legionella DNA in urine and serum
samples from patients with pneumonia. Clin Infect Dis 23(3): 475-80.
103. Kessler, H. H., et al. (1993). Rapid detection of Legionella species in bronchoalveolar lavage fluids with the Envi-
roAmp Legionella PCR amplification and detection kit. J Clin Microbiol 31(12): 3325-8.
104. Ratcliff, R. M., et al. (1998). Sequence-based classification scheme for the genus Legionella targeting the mip gene.
J Clin Microbiol 36(6): 1560-7.
105. Koide, M. and Saito, A. (1995). Diagnosis of Legionella pneumophila infection by polymerase chain reaction. Clin
Infect Dis 21(1): 199-201.
106. Lindsay, D. S., Abraham, W. H. and Fallon, R. J. (1994). Detection of mip gene by PCR for diagnosis of Legionnaires’
disease. J Clin Microbiol 32(12): 3068-9.
107. Wilson, D. A., et al. (2003). Detection of Legionella pneumophila by real-time PCR for the mip gene. J Clin Microbiol
41(7): 3327-30.
108. Diederen, B. M., et al. (2007). Evaluation of real-time PCR for the early detection of Legionella pneumophila DNA in
serum samples. J Med Microbiol 56(Pt 1): 94-101.
109. Diederen, B. M., et al. (2007). Sensitivity of Legionella pneumophila DNA detection in serum samples in relation to
disease severity. J Med Microbiol 56(Pt 9): 1255.
110. Helbig, J. H., et al. (1999). Diagnostic relevance of the detection of Legionella DNA in urine samples by the poly-
merase chain reaction. Eur J Clin Microbiol Infect Dis 18(10): 716-22.
111. Murdoch, D. R. (2003). Diagnosis of Legionella infection. Clin Infect Dis 36(1): 64-9.
112. Diederen, B. M., et al. (2008). Utility of real-time PCR for diagnosis of Legionnaires’ disease in routine clinical
practice. J Clin Microbiol 46(2): 671-7.
113. Bencini, M. A., et al. (2007). Multicenter comparison of molecular methods for detection of Legionella spp. in
sputum samples. J Clin Microbiol 45(10): 3390-2.
114. Sopena, N., et al. (1998). Comparative study of the clinical presentation of Legionella pneumonia and other
community-acquired pneumonias. Chest 113(5): 1195-200.
115. Ingram, J. G. and Plouffe, J. F. (1994). Danger of sputum purulence screens in culture of Legionella species. J Clin
Microbiol 32(1): 209-10.
116. Tuuminen, T., et al. (2000). Prevalence of Chlamydia pneumoniae and Mycoplasma pneumoniae immunoglobulin G
and A antibodies in a healthy Finnish population as analyzed by quantitative enzyme immunoassays. Clin Diagn
Lab Immunol 7(5): 734-8.
117. Liu, G., et al. (2005). Chlamydia pneumoniae and Mycoplasma pneumoniae in young children from China with
community-acquired pneumonia. Diagn Microbiol Infect Dis 52(1): 7-14.
118. Oosterheert, J. J., et al. (2005). Impact of rapid detection of viral and atypical bacterial pathogens by real-time
polymerase chain reaction for patients with lower respiratory tract infection. Clin Infect Dis 41(10): 1438-44.
119. Wellinghausen, N., et al. (2006). Low prevalence of Chlamydia pneumoniae in adults with community-acquired
pneumonia. Int J Med Microbiol 296(7): 485-91.
120. Lauderdale, T. L., et al. (2005). Etiology of community acquired pneumonia among adult patients requiring hospi-
talization in Taiwan. Respir Med 99(9): 1079-86.

54

Rene te Witt bw.indd 54 24-11-11 10:35

 Specific diagnostic tests for atypical respiratory tract pathogens

121. Michelow, I. C., et al. (2004). Epidemiology and clinical characteristics of community-acquired pneumonia in
hospitalized children. Pediatrics 113(4): 701-7.
122. Hammerschlag, M. R. (2000). Chlamydia pneumoniae and the lung. Eur Respir J 16(5): 1001-7.
123. Littman, A. J., et al. (2004). Interlaboratory reliability of microimmunofluorescence test for measurement of Chla-
mydia pneumoniae-specific immunoglobulin A and G antibody titers. Clin Diagn Lab Immunol 11(3): 615-7.

Chapter 2
124. Hermann, C., et al. (2002). Comparison of eleven commercial tests for Chlamydia pneumoniae-specific immuno-
globulin G in asymptomatic healthy individuals. J Clin Microbiol 40(5): 1603-9.
125. Holland, S. M., Gaydos, C. A. and Quinn, T. C. (1990). Detection and differentiation of Chlamydia trachomatis,
Chlamydia psittaci, and Chlamydia pneumoniae by DNA amplification. J Infect Dis 162(4): 984-7.
126. Campbell, L. A., et al. (1992). Detection of Chlamydia pneumoniae by polymerase chain reaction. J Clin Microbiol
30(2): 434-9.
127. Kaltenboeck, B., Kousoulas, K. G. and Storz, J. (1992). Two-step polymerase chain reactions and restriction endo-
nuclease analyses detect and differentiate ompA DNA of Chlamydia spp. J Clin Microbiol 30(5): 1098-104.
128. Mygind, T., et al. (2001). Evaluation of real-time quantitative PCR for identification and quantification of Chlamydia
pneumoniae by comparison with immunohistochemistry. J Microbiol Methods 46(3): 241-51.
129. Gaydos, C. A., Quinn, T. C. and Eiden, J. J. (1992). Identification of Chlamydia pneumoniae by DNA amplification of
the 16S rRNA gene. J Clin Microbiol 30(4): 796-800.
130. Tondella, M. L., et al. (2002). Development and evaluation of real-time PCR-based fluorescence assays for detection
of Chlamydia pneumoniae. J Clin Microbiol 40(2): 575-83.
131. Kuoppa, Y., et al. (2002). Quantitative detection of respiratory Chlamydia pneumoniae infection by real-time PCR. J
Clin Microbiol 40(6): 2273-4.
132. Reischl, U., et al. (2003). Rapid and standardized detection of Chlamydia pneumoniae using LightCycler real-time
fluorescence PCR. Eur J Clin Microbiol Infect Dis 22(1): 54-7.
133. Apfalter, P., et al. (2003). Comparison of a new quantitative ompA-based real-Time PCR TaqMan assay for detection
of Chlamydia pneumoniae DNA in respiratory specimens with four conventional PCR assays. J Clin Microbiol 41(2):
592-600.
134. Hardick, J., et al. (2004). Real-time PCR for Chlamydia pneumoniae utilizing the Roche Lightcycler and a 16S rRNA
gene target. J Mol Diagn 6(2): 132-6.
135. Chernesky, M., et al. (2002). Comparison of an industry-derived LCx Chlamydia pneumoniae PCR research kit to
in-house assays performed in five laboratories. J Clin Microbiol 40(7): 2357-62.
136. Block, S., et al. (1995). Mycoplasma pneumoniae and Chlamydia pneumoniae in pediatric community-acquired
pneumonia: comparative efficacy and safety of clarithromycin vs. erythromycin ethylsuccinate. Pediatr Infect Dis J
14(6): 471-7.
137. Boman, J., et al. (1997). Rapid diagnosis of respiratory Chlamydia pneumoniae infection by nested touchdown
polymerase chain reaction compared with culture and antigen detection by EIA. J Infect Dis 175(6): 1523-6.
138. Verkooyen, R. P., et al. (1998). Evaluation of PCR, culture, and serology for diagnosis of Chlamydia pneumoniae
respiratory infections. J Clin Microbiol 36(8): 2301-7.

55

Rene te Witt bw.indd 55 24-11-11 10:35

Rene te Witt bw.indd 56 24-11-11 10:35
 Biomarkers

Rene te Witt bw.indd 57 24-11-11 10:35

Rene te Witt bw.indd 58 24-11-11 10:35
 Chapter 3.1
Neopterin and procalcitonin are suitable
biomarkers for exclusion of severe Plasmodium
falciparum disease at the initial clinical
assessment of travellers with imported malaria

René te Witt
Marlies E. van Wolfswinkel
Pieter L. Petit
Jaap J. van Hellemond
Rob Koelewijn
Alex van Belkum
Perry J.J. van Genderen

Malaria Journal, 2010; 9: 255-62

Rene te Witt bw.indd 59 24-11-11 10:35

 Chapter 3.1

Abstract

Most clinicians in developed, non-malaria endemic countries have limited or no experience

in making clinical assessments of malaria disease severity and subsequent decisions regard-
ing the need for parenteral therapy or high-level monitoring in febrile patients with imported
malaria. In the present study, the diagnostic accuracy of plasma soluble Triggering Receptor
Expressed on Myeloid cells 1 (sTREM-1), neopterin and procalcitonin levels as biomarkers for
severe Plasmodium falciparum disease was evaluated in 104 travellers with imported malaria
(26 patients with non-P. falciparum malaria, 64 patients with uncomplicated P. falciparum
malaria and 14 patients with severe P. falciparum malaria).
sTREM-1, neopterin and procalcitonin were determined in serum using commercially avail-
able ELISA or EIA tests. The diagnostic performance of these biomarkers for severe disease
was compared with plasma lactate, a well-validated parameter for disease severity in patients
with malaria, as reference. Severe malaria was defined according to the modified WHO criteria.
No significant differences in sTREM-1 levels were detected between the different patient
groups. Patients with severe P. falciparum malaria had significantly higher neopterin and
procalcitonin levels on admission when compared to patients with uncomplicated P. falci-
parum malaria or non-P. falciparum malaria. Receiver Operating Characteristic (ROC) curve
analysis showed that neopterin had the highest Area-Under-the-ROC curve (AUROC 0.85)
compared with plasma lactate (AUROC 0.80) and procalcitonin (AUROC 0.78). At a cut-off
point of 10.0 ng/ml, neopterin had a positive and negative predictive value of 0.38 and 0.98
whereas procalcitonin, at a cut-off point of 0.9 ng/ml, had a positive and negative predictive
value of 0.30 and 1.00.
Although the diagnostic value of neopterin and procalcitonin is limited, the high negative
predictive value of both neopterin and procalcitonin may be helpful for a rapid exclusion
of severe malaria disease on admission. This may be a valuable tool for physicians only oc-
casionally dealing with ill-returned travellers from malaria-endemic regions and who need to
decide on subsequent oral anti-malarial treatment or timely referral to a specialized centre
for high-level monitoring and intensified parenteral treatment.

60

Rene te Witt bw.indd 60 24-11-11 10:35

 Neopterin and procalcitonin for exclusion of severe Plasmodium falciparum disease

Introduction

Travellers from industrialized countries and inhabitants of malaria-endemic regions clearly

represent two distinct worlds of malaria (1). The global burden of malaria is largely carried by
the world’s malaria-endemic regions with as many as 500 million cases annually and a death
toll of 1 to 3 million children each year. Severe malaria in areas of endemicity is associated
with a mortality of 15 to 40% (2-3). In many malaria-endemic regions, strict triage for admis-

Chapter 3.1
sion to ICU facilities must be applied because the ICU capacity is usually limited. Recently, a
5-point Coma Acidosis Malaria (CAM) score based on only acidosis (base deficit) and cerebral
malaria (measured with Glasgow Coma Scale) was introduced, which could identify adult
patients with severe malaria who were at high risk of death (4).
In striking contrast, in non-endemic industrialized countries malaria is only seen as an
occasionally imported disease (5) and is usually associated with a low case-fatality rate (6-
7). Even in the pre-artesunate era, the mortality of severe malaria in non-endemic regions
was significantly lower when compared with regions of malaria endemicity (6-8), probably
reflecting the availability of adequate supportive care facilities in industrialized countries.
Industrialized countries, however, have to face other –more trivial- problems. For instance,
the expertise on diagnosis and treatment of malaria is usually focussed in some specialized
hospitals and institutes but many ill-returning travellers may present to non-specialized
hospitals or even general practitioners. Making a proper diagnosis of malaria may be
troublesome under these circumstances, for instance, by lack of experience in the examina-
tion of malaria thick and thin blood smears and in the assessment of parasite load. These
non-specialized centres therefore often rely on rapid diagnostic tests for the diagnosis of
malaria (9). Although sensitive in diagnosing P. falciparum malaria, these rapid tests do not
provide any information about the severity of the infection. Moreover, although artesunate,
which is now considered the parenteral drug of choice for treatment of severe falciparum
malaria, is available as an orphan drug in The Netherlands, it is currently only in stock in
some specialized centres but certainly not available in every Dutch hospital. Some of these
general hospitals do not even have any drug in stock for the treatment of malaria (10). To
prevent unnecessary delay in diagnosis of severe malaria and institution of proper parenteral
treatment, a simple, well-validated, laboratory-based biomarker that predicts or excludes se-
vere disease accurately would be of great help for those clinicians occasionally dealing with
febrile travellers returning from malaria endemic regions. These clinicians have to decide on
subsequent oral anti-malarial treatment or a timely referral to a specialized centre for high-
level monitoring and intensified parenteral treatment. In the present study, the diagnostic
accuracy of plasma soluble Triggering Receptor Expressed on Myeloid cells 1 (sTREM-1),
neopterin and procalcitonin was evaluated as potential markers for malaria disease severity
in travellers with imported malaria. These bio-substances are all involved in the systemic
pro-inflammatory response of the host to invading pathogens. Some of these biomarkers

61

Rene te Witt bw.indd 61 24-11-11 10:35

 Chapter 3.1

are already in use for the diagnosis and follow-up of sepsis or used in treatment algorithms,
resulting in a successful reduction of antibiotic use and duration (11-12).

Materials and Methods

Study population

The Harbour Hospital is a 161-bed general hospital located in Rotterdam. It also harbours
the Institute for Tropical Diseases, which serves as a national reference centre. In the period
1999-2008 almost 500 cases of imported malaria were diagnosed (13). For the majority of
these cases, demographic, clinical and laboratory data and serum samples were available. For
the present study, a representative sample of this cohort was taken and analysed.

Definitions

Patients were classified as having severe P. falciparum malaria if they met one or more of the
WHO criteria for severe malaria, as modified by Hien et al (14):
• A score on the Glasgow Coma Scale (GCS) of less than 11 (indicating cerebral malaria).
• Anaemia (haematocrit <20%) with parasite counts exceeding 100,000/μl (roughly cor-
responding to 2% parasitaemia) on a peripheral blood smear.
• Jaundice (serum bilirubin >50 μmol/l) with parasite counts exceeding 100.000/μl on a
peripheral blood smear.
• Renal impairment (urine output <400 ml/24 h and serum creatinine >250 μmol/l).
• Hypoglycaemia (blood glucose <2.2 mmol/l).
• Hyperparasitaemia (>10% parasitaemia).
• Systolic blood pressure <80 mm Hg with cold extremities (indicating shock).

Study design

In previous studies (6, 13, 15) these severity criteria were also used to define severe malaria in
non-immune travellers. In the present study the occurrence of severe malaria was considered
a primary end-point. This contrasts with the design of many studies in patients with severe
malaria in regions of malaria endemicity where the severity criteria are used as an entry cri-
terion. In the present study, plasma lactate was used as a surrogate parameter for acid-base
disbalance and reference biomarker. It was evaluated in a previous study in non-immune
travellers with imported malaria (15). The diagnostic performance of sTREM-1, procalcitonin
and neopterin for malaria disease severity was compared with that of plasma lactate, which is
routinely measured at the Institute for Tropical Diseases in ill-returning travellers.

62

Rene te Witt bw.indd 62 24-11-11 10:35

 Neopterin and procalcitonin for exclusion of severe Plasmodium falciparum disease

Procedures

On admission, blood samples were taken for analysis of the red blood cell count, haematocrit,
white blood cell count, platelet count, serum electrolytes, total bilirubin, serum creatinine,
liver enzymes, and blood glucose. In addition, a serum sample was taken on admission which
was stored at -20°C until analysis. For the determination of plasma lactate, a separate blood
sample was drawn on admission without congestion and placed on melting ice after which it

Chapter 3.1
was immediately analysed after isolation of plasma. Malaria was diagnosed by QBC (Quanti-
tative Buffy Coat) analysis, by a rapid diagnostic antigen test for malaria (Binax NOW® Malaria
Test, Binax Inc., Maine, USA) and by conventional microscopy of stained thick and thin blood
smears. In case of P. falciparum infections, parasite density was determined. When the parasi-
taemia was less than 0.5% infected erythrocytes, parasites were counted per 100 leucocytes
in thick smears. When the parasitaemia was equal or higher than 0.5% infected erythrocytes,
infected erythrocytes were counted in thin blood smear and expressed as a percentage of
the total erythrocytes. The number of parasites per microliter was subsequently calculated
from these data.
sTREM-1 and neopterin levels were determined in serum samples using commercially
available ELISA tests (R&D Systems, Abingdon, UK; DRG, Marburg, Germany, respectively).
Procalcitonin levels in serum samples were determined using a commercially available EIA
test (VIDAS BRAHMS Procalcitonin, bioMérieux, Lyon, France). All tests were performed ac-
cording to manufacturer’s instructions. Detection limits were 3.88 pg/ml for sTREM-1, 0.2 ng/
ml for neopterin and 0.05 ng/ml for procalcitonin, respectively. According to the manufactur-
ers, normal serum values are <100 pg/ml for TREM-1, <3 ng/ml for neopterin and <0.1 ng/ml
for procalcitonin.

Statistical methods

For comparison between groups, the Mann-Whitney U-test was used and p-values of <0.05
were considered statistically significant. The diagnostic performance of each biomarker
was reported as sensitivity, specificity, positive and negative predictive value for severe P.
falciparum malaria and their corresponding 95% confidence intervals. Of each test a Receiver
Operating Characteristic (ROC) curve, a graphical plot of sensitivity (true positive rate) versus
1-specificity (false positive rate), was constructed as a summary statistic and the area under
the ROC curve (AUROC) and its corresponding 95% confidence intervals were calculated.
Youden’s index J (J=sensitivity+specificity-1) was used to choose the most appropriate cut-
off point for each biomarker. All statistical analyses were performed using SPSS 15.0.

63

Rene te Witt bw.indd 63 24-11-11 10:35

 Chapter 3.1

Results

Patient characteristics

In total 104 travellers with imported malaria were included in this study, of which 26 patients
were diagnosed with a non-P. falciparum infection (Plasmodium malariae n=2; Plasmodium
ovale n=5; Plasmodium vivax n=19) and 78 patients were diagnosed with P. falciparum infec-
tion. The general characteristics of all patients are shown in Table 1.

Table 1. General characteristics and laboratory results on admission of patients with various species of malaria. Data are given as median (range).
Non-P. falciparum P. falciparum
Uncomplicated Severe
(n=26) (n=64) (n=14)
Demographics
Male/female 20/6 51/13 6/8
Age, years 40 (17-62) 40 (11-67) 40 (26-57)
Continent of acquisition
Africa 12 (46%) 60 (94%) 12 (86%)
Asia 9 (35%) 3 (5%) 1 (7%)
South America 5 (19%) 1 (2%) 1 (7%)
Vital signs on admission
Body temperature, °C 38.8 (36.1-41.5) 38.7 (36.1-40.6) 38.8 (36.8-40.6)
Pulse rate, beats per minute 90 (60–130) 95 (68-120) 108 (78-140)
Systolic blood pressure, mm Hg 123 (100-196) 120 (95-185) 118 (80-160)
Laboratory data on admission
Parasite load, throphozoites/μl ND 5,502 (1.0-385,000) * 205,600 (80,500-860,000)
Plasma lactate, mmol/l 1.4 (0.7-3.0) * 1.5 (0.5-4.4) * 2.6 (0.9-5.8)
Haemoglobin, mmol/l 8.2 (6.1-10.1) 8.7 (5.3-11.1) * 7.6 (3.8-10.2)
Leucocytes, x 109/l 5.2 (1.9-9.3) 5.5 (1.8-11.3) 6.6 (3.2-18.5)
Platelets, x 109/l 93.0 (10.0-205.0) * 78.5 (16.0-247.0) * 27.0 (3.0-152.0)
C-reactive protein, mg/l 86.5 (18.0-208.0) * 109.0 (5.0-278.0) * 190.0 (91.0-265.0)
Serum creatinine, μmol/l 94.0 (66.0-149.0) 103.5 (63.0-208.0) 102.5 (70.0-199.0)
Total bilirubin, μmol/l 24.0 (6.0-84.0) * 25.0 (7.0-164.0) * 54.0 (20.0-269.0)
p-values as compared with severe P. falciparum malaria.
* indicates p<0.01
ND: not done

Characteristics of patients with severe malaria

Thirteen patients fulfilled the criteria for severe malaria at initial presentation. Another
patient did not fulfil these criteria on admission, but the clinical course deteriorated shortly
hereafter with impaired consciousness and hyperparasitaemia. Procalcitonin and neopterin
levels were already increased on admission in this particular patient. Eventually, at admission
to the ICU, all 14 patients fulfilled one or more of the severity criteria (GCS<11, n=1; anaemia

64

Rene te Witt bw.indd 64 24-11-11 10:35

 Neopterin and procalcitonin for exclusion of severe Plasmodium falciparum disease

with a parasite count exceeding 100,000 trophozoites per μl, n=2; icterus with a parasite
count exceeding 100,000 trophozoites per μl, n=8; acute oliguric renal insufficiency, n=0;
hypoglycaemia, n=0; hyperparasitaemia, n=5 and shock, n=1, respectively). Five patients had
an impaired conscious level but a GCS above 11; eight patients had a parasitaemia >5%,
respectively. The first arterial blood gas analysis on ICU showed a median bicarbonate level
of 22 mmol/l (range 17 to 26 mmol/l) and a median base deficit of 2 (range -3 to 8). Median
GCS was 15 (range 9 to 15). One patient needed mechanical ventilation. Eleven patients

Chapter 3.1
received exchange transfusion as an adjunct therapy. No case fatalities were observed. The
laboratory results on admission of travellers with imported severe P. falciparum malaria were
further characterized by significantly lower platelet counts and haemoglobin levels and by
significantly higher plasma lactate, bilirubin and C-reactive protein levels and erythrocyte
sedimentation rates, respectively (Table 1).

Analysis of biomarkers for severe malaria

sTREM-1

No statistically significant differences were observed in sTREM-1 levels in serum, between

patients with severe P. falciparum malaria, uncomplicated P. falciparum malaria and non-P.
falciparum malaria (Figure 1A).

Neopterin

Neopterin levels on admission were significantly higher in travellers with severe P. falciparum
malaria when compared to travellers with uncomplicated P. falciparum malaria (p<0.0001) and
travellers with non-P. falciparum malaria (p<0.0001) (Figure 1B). ROC curve analysis showed an
AUROC of 0.85 (95% Confidence Interval 0.76-0.94), suggesting a good accuracy (Figure 2). As
shown in Table 2, at a cut-off point of 10.0 ng/ml, neopterin had an excellent sensitivity and
negative predictive value but a poor specificity and positive predictive value for severe disease.

Table 2. Descriptive statistics of diagnostic accurary of neopterin, procalcitonin as compared with lactat for the diagnosis of severe falciparum
malaria on admission. The Youden’s index was used to choose an appropriate cut-off value.
Neopterin Procalcitonin Lactate
95% Confidence Interval 95% Confidence Interval 95% Confidence Interval
Optimal cut-offvalue 10 ng/ml 0.9 ng/ml 1.8 mmol/l
Youden’s index 0.60 0.53 0.58
Sensitivity 0.93 0.64-1.00 1.00 0.70-1.00 0.92 0.60-1.00
Specificity 0.67 0.54-0.78 0.53 0.40-0.66 0.66 0.51-0.78
Positive predictive value 0.38 0.23-0.56 0.30 0.17-0.47 0.39 0.22-0.59
Negative predictive value 0.98 0.86-1.00 1.00 0.87-1.00 0.97 0.83-1.00

65

Rene te Witt bw.indd 65 24-11-11 10:35

 Chapter 3.1

400 75
A B

Neopterin (ng/ml)
*
TREM-1 (pg/ml)

300 *
50

200

25
100

0 0
non- uncomplicated severe non- uncomplicated severe
P. falciparum P. falciparum P. falciparum P. falciparum P. falciparum P. falciparum
malaria malaria malaria malaria malaria malaria

50 7.5
C D
Procalcitonin (ng/ml)

** **
Lactate (mmol/l)

40 **
5.0
30

20
2.5
10

0 0.0
non- uncomplicated severe non- uncomplicated severe
P. falciparum P. falciparum P. falciparum P. falciparum P. falciparum P. falciparum
malaria malaria malaria malaria malaria malaria

Figure 1. Concentrations of potential biomarkers for disease severity in malaria patients on admission.
Individual data are shown with the median value of each biomarker; TREM-1 (panel A), neopterin (panel B), procalcitonin (panel C) and plasma
lactate (panel D). Significant differences in biomarker concentrations between patient groups (black square = non-P. falciparum malaria; black
triangle up = uncomplicated P. falciparum malaria; black triangle down = severe P. falciparum malaria) with P values < 0.0001 and < 0.005 are
indicated by * and **, respectively.

Procalcitonin

Procalcitonin levels were significantly higher in travellers with severe P. falciparum malaria
when compared to travellers with uncomplicated P. falciparum malaria (p=0.0022). However,
no significant differences were noted in comparison to travellers with non-P. falciparum infec-
tions (p=0.17) (Figure 1C). ROC curve analysis showed an AUROC of 0.78 (95% CI 0.66-0.91),
compatible with a fair accuracy (Figure 2). At a cut-off point of 0.9 ng/ml, procalcitonin had
an excellent sensitivity and negative predictive value, whereas specificity and positive pre-
dictive value for severe P. falciparum malaria was poor (Table 2).

66

Rene te Witt bw.indd 66 24-11-11 10:35

 Neopterin and procalcitonin for exclusion of severe Plasmodium falciparum disease

1,0
1.0

0,9
0.9

0,8
0.8

0,7
0.7

0,6
0.6

Chapter 3.1
sensitivity

0,5
0.5

0.4
0,4

0,3
0.3
Area Under RO- 95% confidence
Curve (AUROC) interval
0,2
0.2 Neopterin 0.85 0.76 - 0.94

Procalcitonin 0.78 0.66 - 0.91

0,1
0.1 Lactate 0.80 0.65 - 0.96

0,0
0.0
0.0
0,0 0.1
0,1 0.2
0,2 0.3
0,3 0.4
0,4 0.5
0,5 0.6
0,6 0.7
0,7 0.8
0,8 0.9
0,9 1.0
1,0

1- specificity
Figure 2. Receiver Operating Curves (ROC) characteristics of the diagnostic performance of neopterin, procalcitonin and lactate for severe
P. falciparum malaria.
The ROC curve is a graph of sensitivity (true positive fraction) plotted against 1-specificity (false positive fraction). The performance of a
diagnostic variable can be quantified by calculating the area under the ROC curve (AUROC). The ideal test would have an AUROC of 1, whereas a
random guess would have an AUROC of 0.5.

Plasma lactate

Plasma lactate levels were significantly higher in travellers with severe P. falciparum malaria
when compared to travellers with uncomplicated P. falciparum malaria (p=0.0012) and travel-
lers with non-P. falciparum malaria (p=0.0040). ROC curve analysis of plasma lactate levels
showed an AUROC of 0.80 (95% CI 0.65-0.96) compatible with a good accuracy (Figure 2). At a
cut-off point of 1.8 mmol/l, lactate had an excellent sensitivity and negative predictive value,
but a poor specificity and positive predictive value for severe P. falciparum malaria (Table 2),
respectively.

Combinations of various biomarkers for severe falciparum disease

Analysis of various combinations of newly tested biomarkers and the use of different cut-off
levels did not result in better discrimination of patients with severe P. falciparum malaria.

67

Rene te Witt bw.indd 67 24-11-11 10:35

 Chapter 3.1

Discussion

Severe malaria is disreputable for its high case-fatality rate, but the outcome of severe P.
falciparum infections has significantly improved since the introduction of artesunate as first
line treatment of severe malaria, in particular in developing countries (2). In industrialized
countries such as The Netherlands, the case-fatality rate of imported malaria is low and fatal
cases are only occasionally reported. In the present study, the biomarkers sTREM-1, neopterin
and procalcitonin were evaluated for their potential to be used as a marker for severe malaria
disease upon admission. This contrasts with the design of many studies in regions of malaria
endemicity where severe malaria is usually the entry criterion. For reasons of comparability,
the same set of criteria for severe malaria was strictly applied for the diagnosis of severe ma-
laria in this study, even though the study population comprised of presumably non-immune
travellers and some authors even suggest a threshold of 5% in stead of 10% parasitized
erythrocytes to define hyperparasitaemia in non-immune individuals.
The quantification of sTREM-1 levels on admission did not result in proper discrimination of
severe P. falciparum malaria from uncomplicated P. falciparum malaria and non-P. falciparum
malaria. In contrast, travellers with severe P. falciparum malaria had significantly higher levels
of neopterin and procalcitonin on admission as compared with travellers with uncomplicated
P. falciparum malaria or non-P. falciparum malaria, respectively. These findings correspond
with the results of several other studies performed in semi-immune malaria patients living
in malaria-endemic regions (16-18). When the ROC curve characteristics of neopterin and
procalcitonin were compared to that of plasma lactate, the AUROC of neopterin appeared
superior whereas the AUROC of procalcitonin appeared inferior to that of lactate, suggesting
that neopterin provided the most accurate diagnostic performance for severe P. falciparum
malaria in this cohort of travellers.
Unfortunately, the applicability of these tests in the initial clinical assessment of patients
with severe P. falciparum malaria will probably be limited by the poor positive predictive
value of neopterin and procalcitonin indicating that neither test can serve as a valuable tool
for the diagnosis of severe P. falciparum malaria. For illustration, applying a procalcitonin
level > 0.9 ng/ml or a neopterin level > 10.0 ng/ml as a guide to intensified monitoring and
treatment would result in more than 20 of 64 patients with uncomplicated P. falciparum
malaria receiving more intensive monitoring and treatment than strictly necessary. On the
other hand, the high negative predictive value of both neopterin and procalcitonin suggests
that these tests can still be of value by providing a tool for exclusion of severe disease. With
either a procalcitonin level of less than 0.9 ng/ml or a neopterin level of less than 7.9 ng/
ml in serum on admission as a cut-off point for severe P. falciparum malaria, no patient with
severe disease would have been denied access to high-level monitoring and intensive treat-
ment. In a previous study, in which a semi-quantitative ‘point-of-care’ procalcitonin test as
a diagnostic tool for severe P. falciparum malaria was evaluated prospectively, all 6 patients

68

Rene te Witt bw.indd 68 24-11-11 10:35

 Neopterin and procalcitonin for exclusion of severe Plasmodium falciparum disease

with severe P. falciparum malaria had procalcitonin values classified as either “moderate” or
“high” (corresponding to a procalcitonin level ≥ 2 ng/ml), but never as “normal” or “low (18).
This is compatible with the findings of the current retrospective serum sample-based study
in which procalcitonin was measured quantitatively.
Although severe or fatal malaria rarely results from infections with the non-sequestering
Plasmodium species vivax, ovale and malariae, increased neopterin and procalcitonin serum
levels were also observed in the majority of these patients, although levels were lower than

Chapter 3.1
compared with severe P. falciparum malaria patients. Although speculatively, these observa-
tions suggest that the mechanism whereby neopterin and procalcitonin levels increase in
malaria, is not specific for severe P. falciparum malaria alone. Therefore, it may not accurately
reflect the pivotal pathophysiological events in complicated P. falciparum malaria, such as
the sequestration of infected red blood cells in the microcirculation of vital organs and
disturbance of microcirculatory flow. Whereas an increased plasma lactate level conceivably
reflects a significant reduction in microcirculatory flow in vital organs, the elevated neopterin
and procalcitonin levels are probably the result of activation of a common inflammatory host
response evoked by infection with the respective Plasmodium parasites. In fact, some reports
even suggest that P. falciparum malaria per se is not associated with a stronger host response
than P. vivax or P. ovale malaria, but that the parasite burden of the causative Plasmodium
species may also modulate the extent of the host inflammatory response (19).
In conclusion, although neither neopterin nor procalcitonin can probably serve as a useful
single diagnostic tool for severe P. falciparum malaria, the high negative predictive value of
both neopterin and procalcitonin may be helpful for a rapid exclusion of severe P. falciparum
malaria on admission. This may be a valuable tool – particularly if available as a rapid diagnos-
tic test - for physicians only occasionally dealing with ill-returned travellers and who need to
decide on subsequent oral anti-malarial treatment or a timely referral to a specialized centre
for high-level monitoring and intensified parenteral treatment.

Acknowledgments

We acknowledge Dr. Martijn Huisman for his help in data analysis.

The Vidas™ kits for procalcitonin testing used in this study were supplied free of charge by
bioMérieux. There are no conflicts of interests to disclose.

69

Rene te Witt bw.indd 69 24-11-11 10:35

 Chapter 3.1

References.
1. Wellems, T. E. and Miller, L. H. (2003). Two worlds of malaria. N Engl J Med 349(16): 1496-8.
2. Dondorp, A., et al. (2005). Artesunate versus quinine for treatment of severe falciparum malaria: a randomised trial.
Lancet 366(9487): 717-25.
3. (2000). Severe falciparum malaria. World Health Organization, Communicable Diseases Cluster. Trans R Soc Trop
Med Hyg 94 Suppl 1: S1-90.
4. Hanson, J., et al. (2010). A simple score to predict the outcome of severe malaria in adults. Clin Infect Dis 50(5):
679-85.
5. van Genderen, P. J., Hesselink, D. A. and Bezemer, J. M. (2008). Imported malaria is falling in Netherlands and
Europe. BMJ 337: a1026.
6. van Genderen, P. J., et al. (2010). Efficacy and safety of exchange transfusion as an adjunct therapy for severe
Plasmodium falciparum malaria in nonimmune travelers: a 10-year single-center experience with a standardized
treatment protocol. Transfusion 50(4): 787-94.
7. Christen, D., Steffen, R. and Schlagenhauf, P. (2006). Deaths caused by malaria in Switzerland 1988-2002. Am J Trop
Med Hyg 75(6): 1188-94.
8. Bruneel, F., et al. (2003). The clinical spectrum of severe imported falciparum malaria in the intensive care unit:
report of 188 cases in adults. Am J Respir Crit Care Med 167(5): 684-9.
9. Stauffer, W. M., et al. (2009). Diagnostic performance of rapid diagnostic tests versus blood smears for malaria in
US clinical practice. Clin Infect Dis 49(6): 908-13.
10. Oudijk, J. M., et al. (2009). [Availability of antimalarial agents in Dutch hospitals]
Beschikbaarheid van malariamiddelen in Nederlandse ziekenhuizen. Ned Tijdschr Geneeskd 153: A462.
11. Schultz, M. J. and Determann, R. M. (2008). PCT and sTREM-1: the markers of infection in critically ill patients? Med
Sci Monit 14(12): RA241-7.
12. Schuetz, P., et al. (2010). Procalcitonin for guidance of antibiotic therapy. Expert Rev Anti Infect Ther 8(5): 575-87.
13. van Wolfswinkel, M. E., et al. (2010). Hyponatraemia in imported malaria is common and associated with disease
severity. Malar J 9: 140.
14. Tran, T. H., et al. (1996). A controlled trial of artemether or quinine in Vietnamese adults with severe falciparum
malaria. N Engl J Med 335(2): 76-83.
15. van Genderen, P. J., et al. (2005). Evaluation of plasma lactate as a parameter for disease severity on admission in
travelers with Plasmodium falciparum malaria. J Travel Med 12(5): 261-4.
16. Brown, A. E., et al. (1992). Urinary neopterin in volunteers experimentally infected with Plasmodium falciparum.
Trans R Soc Trop Med Hyg 86(2): 134-6.
17. Biemba, G., et al. (2000). Markers of inflammation in children with severe malarial anaemia. Trop Med Int Health
5(4): 256-62.
18. Hesselink, D. A., et al. (2009). Procalcitonin as a biomarker for severe Plasmodium falciparum disease: a critical
appraisal of a semi-quantitative point-of-care test in a cohort of travellers with imported malaria. Malar J 8(1): 206.
19. Hemmer, C. J., et al. (2006). Stronger host response per parasitized erythrocyte in Plasmodium vivax or ovale than
in Plasmodium falciparum malaria. Trop Med Int Health 11(6): 817-23.

70

Rene te Witt bw.indd 70 24-11-11 10:35

 Chapter 3.2
Procalcitonin- and neopterin levels do not
accurately distinguish bacterial from viral
infections in ill-returned febrile travellers

René te Witt
Robert-Jan Hassing
Pieter L. Petit
Alex van Belkum
Perry J. van Genderen

Submitted to Transactions of the Royal Society of Tropical Medicine and Hygiene;

accepted for publication

Rene te Witt bw.indd 71 24-11-11 10:35

 Chapter 3.2

Abstract

The diagnostic performance of procalcitonin and neopterin as markers for bacterial and
viral cause of fever was evaluated in a cohort of 69 febrile travellers with known etiological
agents. Our aim was to establish a decision rule to minimize empirical antibiotic treatment.
Compared with C-reactive protein (CRP) and leukocyte (differential) counts, procalcitonin
and neopterin had a disappointing diagnostic accuracy. Refraining antibiotics in case of
combined presence of lymphocytosis and/or CRP ≤ 10 mg/l would result in an 85% reduction
in unwanted antibiotic treatment in patients with viral disease but in adequate antibiotic
coverage of all patients with bacterial disease.

72

Rene te Witt bw.indd 72 24-11-11 10:35

 Procalcitonin and neopterin levels to distinguish bacterial from viral infections

Introduction

The initial clinical evaluation of ill-returned travellers with fever should focus on infections
that are treatable, transmissible and that may cause serious sequelae or even death. As a
dogmatic rule, malaria remains the most important infection to exclude in anyone who re-
turns with fever after visiting a tropical destination, necessitating investigation without delay
(1). After exclusion of malaria, the clinician must subsequently decide whether the infection

Chapter 3.2
is likely to be caused by bacteria or viruses and whether the patient should be admitted
for intensified treatment including empirical treatment with antibiotics. The most commonly
differential diagnoses in this subpopulation of ill-returning travellers involve Dengue fever,
enteric fever and rickettsial disease. Discrimination between these diseases may be difficult
because of overlapping geographical endemicity patterns and clinical features. Results of
routine laboratory findings may provide additional clues for a diagnosis. Traditionally, leuko-
cytosis is considered to be suggestive of bacterial infection, whereas leukopenia may suggest
viral infection. Unfortunately, in a number of bacterial infections, including uncomplicated
typhoid fever, brucellosis and rickettsial infections, leukocytosis is not a characteristic feature
of the disease. As a consequence, the clinician can not rely on these traditional parameters
for decision making in the acute care setting. In order to reduce the need for empirical an-
tibiotic treatment, we evaluated the diagnostic power of the biomarkers procalcitonin and
neopterin to accurately predict a bacterial or a viral cause of fever in a cohort of ill-returned
febrile travellers with a confirmed diagnosis of viral or bacterial disease. The performance of
these biomarkers was compared to that of the parameters leukocyte differential count and
C-reactive protein (CRP), respectively.

Materials and Methods

In this pilot study we evaluated procalcitonin and neopterin levels in stored serum samples
taken on admission of 69 patients with a confirmed diagnosis (Dengue virus infection
n=17, enteric fever [Salmonella enterica serovar Typhi n=9; Paratyphi A n=7 and B n=1] and
Rickettsia africae infection n=19) in the period January 2005 to December 2009, respectively.
Procalcitonin levels were determined using a commercially available EIA test (VIDAS BRAHMS
Procalcitonin, bioMérieux, Lyon, France). Neopterin levels were determined using a commer-
cially available ELISA test (DRG, Marburg, Germany). Normal serum values are <3 ng/ml for
neopterin and <0.1 ng/ml for procalcitonin. All other laboratory parameters were previously
established on admission using routine procedures.

73

Rene te Witt bw.indd 73 24-11-11 10:35

 Chapter 3.2

Results

As shown in Table 1, there was a considerable overlap in leukocyte differential counts between
patients with viral or bacterial disease. The bacterial indicators leukocytosis, segmentosis and
an increased proportion of neutrophil band forms did not discern bacterial from viral causes.
However, the viral indicators leukopenia, lymphocytosis and atypical lymphocytes were more
frequently observed in patients with viral than in patients with bacterial disease. The PPV
of these indicators varied between 61 and 80 percent. The corresponding likelihood ratios
ranged from 2.1 to 5.3 indicating that presence of these predictors resulted in a 2 to 5-fold
increase in odds favouring viral cause of fever. Determination of CRP levels also performed
well. At a level above 11 mg/L, CRP was associated with a likelihood ratio of 3.7 for bacterial
disease. The good performance of CRP for bacterial disease became even clearer in Receiver

Table 1. The diagnostic performance of the viral and bacterial predictors for a viral and bacterial cause of fever in ill-returning febrile patients
with a known etiological agent.
Parameter on Viraldisease Bacterialdisease Predictor Statistics
admission
Dengue Typhoidfever Rickettsioses Sensitivity Specificity PPV NPV LR P-value
(n=33) (n=17) (n=19)
Leukocytes, n (%)
<4.0 (x109/l) 14 (42) 2 (12) 2 (11) Viral 44 89 78 64 3.9 P=0.005
4.0 - 10.0 (x109/l) 17 (52) 15 (88) 16 (84)
>10.0 (x109/l) 1 (3) 0 (0) 1 (5) Bacterial 3 97 50 47 n.a. n.s.
Segments, n (%)
<45 (%) 13 (52) 3 (19) 2 (12)
45 – 75 (%) 11 (44) 11 (69) 14 (82)
>75 (%) 8 (32) 2 (13) 1 (6) Bacterial 9 96 75 44 n.a. n.s.
Lymphocytes, n (%)
<20 (%) 3 (12) 4 (25) 3 (18)
20 – 50 (%) 14 (56) 10 (63) 14 (82)
>50 (%) 8 (32) 2 (13) 0 (0) Viral 32 94 80 65 5.3 P=0.014
Band forms, n (%)
0 – 5 (%) 20 (80) 11 (69) 16 (94)
>5 (%) 5 (20) 5 (31) 1 (6) Bacterial 18 80 55 43 n.a. n.s.
Atypical lymphocytes,
n (%) 11 (44) 12 (75) 12 (71)
Absent 14 (56) 4 (25) 5 (29) Viral 56 73 61 69 2.1 P=0.033
Present
C-reactiveprotein, n (%)
< 11 (mg/l) 22 (79) 0 (0) 7 (41)
≥ 11 (mg/l) 6 (21) 17 (100) 10 (59) Bacterial 79 79 82 76 3.7 P<0.0001
Procalcitonin, n (%)
<0.14 (ng/ml) 27 (87) 3 (21) 16 (89)
≥0.14 (ng/ml) 4 (13) 11 (79) 2 (11) Bacterial 41 87 76 59 3.1 P=0.002
Neopterin, n (%)
<3.0 (ng/ml) 14 (44) 1 (6) 8 (44)
≥3.0 (ng/ml) 18 (56) 16 (94) 10 (56) Viral 56 26 41 39 n.a. n.s.

74

Rene te Witt bw.indd 74 24-11-11 10:35

 Procalcitonin and neopterin levels to distinguish bacterial from viral infections

Operating Characteristics (ROC) curve analysis where the AUROC was 0.83 (95% Confidence
Interval 0.71-0.92) (Figure 1).
In contrast, the performance of the biomarkers procalcitonin and neopterin for a respective
bacterial or viral cause of fever was inferior to that of CRP. Even though procalcitonin levels
were significantly higher in bacterial than in viral infections (0.66 ± 2.21 ng/ml vs. 0.07 ± 0.13
ng/ml, p=0.0355) and procalcitonin levels above 0.14 ng/ml were associated with a likelihood
ratio of 3.1, ROC curve analysis indicated a poor accuracy of procalcitonin for bacterial cause

Chapter 3.2
of fever (AUROC 0.61 [95% CI 0.47-0.73]). The performance of neopterin for viral cause of fever
was even worse. Neopterin levels were significantly higher in patients with bacterial infec-
tions than in viral infections (10.2 ± 6.4 ng/ml vs. 4.9 ± 4.3 ng/ml, p=0.0084). Furthermore,
neopterin had an AUROC of 0.33 (95% CI of 0.20-0.48), suggesting a very poor accuracy.

Figure 1.
AUROC analysis

100

80
Sensitivity
60

40

20 Area under 95 % confidence

ROC curve interval

CRP 0.833 0.712 – 0.918

Procalcitonin 0.605 0.468 – 0.731
Neopterin 0.339 0.203 – 0.475

0
0 20 40 60 80 100
100-Specificity

Figure 1. Receiver Operating Curves (ROC) characteristics of the diagnostic performance of CRP and procalcitonin for bacterial infection and
neopterin for viral infection. The ROC curve is a graph of sensitivity (true positive fraction) plotted against 1-specificity (false positive fraction).
The performance of a diagnostic variable can be quantified by calculating the area under the ROC curve (AUROC). The ideal test would have an
AUROC of 1, whereas a random guess would have an AUROC of 0.5.

Discussion

The value of procalcitonin and neopterin in clinical decision making was shown in several
studies (1-7), however not in all (8). In the present study we found a disappointing diagnos-
tic accuracy of both procalcitonin and neopterin for a bacterial and viral cause of fever in
ill-returning travellers, especially when the respective AUROCs were considered a measure
of diagnostic performance. Moreover, in this study neopterin even had a more accurate
performance for bacterial than for viral disease. These findings may be explained by the ob-

75

Rene te Witt bw.indd 75 24-11-11 10:35

 Chapter 3.2

servation that S. Typhi and Paratyphi and Rickettsia species are intracellular bacteria and that
other intracellular bacteria like tuberculosis and melioidosis also lead to increased neopterin
levels (6).
Of the more traditional leukocyte differential count-based parameters, only presence of
leukopenia, lymphocytosis and presence of atypical lymphocytes were suggestive of viral
disease. Its use in clinical practice is hampered by its considerable overlap with the findings in
the bacterial infections. When the indication for empirical antibiotic treatment was based on
a single parameter, at least 11 of 33 patients (maximally 18 of 33 patients) with viral disease
would receive unnecessary antibiotic treatment whereas a minimum of 7 of 36 patients
(maximally 35 of 36 patients) with documented bacterial disease were withheld antibiotic
treatment. Interestingly, when empirical antibiotic treatment was withheld on admission on
the basis of combined presence of lymphocytosis and/or CRP ≤ 10 mg/l, only 5 of 33 patients
with Dengue would receive unnecessary antibiotic treatment but all patients with enteric
fever and rickettsial diseases would receive antibiotic treatment; an almost 85% reduction of
unnecessary empirical antibiotic treatment.
Since the present findings were based on a relatively small number of selected patients
with possibly inherent study limitations like selection bias and lack of power, the suggested
clinical decision rule for empirical antibiotic treatment in returning febrile travellers without
leukocytosis on admission needs validation in properly designed prospective studies.

Acknowledgments

Rob Koelewijn is acknowledged for collection of patient materials and database manage-
ment. The Vidas kits for procalcitonin testing used in this study were supplied free of charge
by bioMérieux.

76

Rene te Witt bw.indd 76 24-11-11 10:35

 Procalcitonin and neopterin levels to distinguish bacterial from viral infections

References.
1. Simon, L., et al. (2004). Serum procalcitonin and C-reactive protein levels as markers of bacterial infection: a
systematic review and meta-analysis. Clin Infect Dis 39(2): 206-17.
2. Chirouze, C., et al. (2002). Low serum procalcitonin level accurately predicts the absence of bacteremia in adult
patients with acute fever. Clin Infect Dis 35(2): 156-61.
3. Hoen, B. (2009). [Differentiating bacterial from viral meningitis: contribution of nonmicrobiological laboratory
tests]

Chapter 3.2
Diagnostic differentiel entre meningite bacterienne et meningite virale : apport des examens non microbi-
ologiques. Med Mal Infect 39(7-8): 468-72.
4. Pfafflin, A. and Schleicher, E. (2009). Inflammation markers in point-of-care testing (POCT). Anal Bioanal Chem
393(5): 1473-80.
5. Ip, M., et al. (2007). Value of serum procalcitonin, neopterin, and C-reactive protein in differentiating bacterial from
viral etiologies in patients presenting with lower respiratory tract infections. Diagn Microbiol Infect Dis 59(2): 131-6.
6. Fuchs, D., et al. (1984). Neopterin as an index of immune response in patients with tuberculosis. Lung 162(6):
337-46.
7. te Witt, R., et al. (2010). Neopterin and procalcitonin are suitable biomarkers for exclusion of severe Plasmodium
falciparum disease at the initial clinical assessment of travellers with imported malaria. Malar J 9: 255.
8. Hesselink, D. A., et al. (2009). Procalcitonin as a biomarker for severe Plasmodium falciparum disease: a critical
appraisal of a semi-quantitative point-of-care test in a cohort of travellers with imported malaria. Malar J 8(1): 206.

77

Rene te Witt bw.indd 77 24-11-11 10:35

Rene te Witt bw.indd 78 24-11-11 10:35
 Culture

Rene te Witt bw.indd 79 24-11-11 10:35

Rene te Witt bw.indd 80 24-11-11 10:35
 Chapter 4
In vitro evaluation of the performance
of Granada selective enrichment broth
for the detection of group
B streptococcal colonization

René te Witt
Paul M. Oostvogel
Rachid Yahiaoui
Yingbin Wu
Alex van Belkum
Anouk E. Muller

European Journal of Clinical Microbiology and Infectious Diseases, 2011; article in press

Rene te Witt bw.indd 81 24-11-11 10:35

 Chapter 4

Abstract

A broth for the screening of group B streptococcocal (GBS) carriage during pregnancy is
about to be introduced. Simulating conditions in everyday practice, we have compared the
sensitivity of this Granada tube broth (GT) with that of classical Amies transport medium (AT)
in vitro.
A total of 1,485 GT and 1,485 AT were tested with 33 well-characterized GBS strains in three
different concentrations, five different incubation times and three different temperatures.
After initial incubation at room temperature (RT) or 4°C, GT were placed at 37°C. GT were
scored for the presence of orange pigment. GT and AT were subcultured on blood agar (BA).
Pigment was observed in 98% of GT incubated at 37°C. GBS could be cultured in 91%, 73%,
and 55% of GT incubated at 37°C, RT, or 4°C, respectively. For AT, these percentages were
only 20% at 37°C, 52% at RT, and 59% at 4°C. When GT initially incubated at RT or 4°C were
subsequently incubated at 37°C, the sensitivity improved significantly.
We conclude that GT is a more sensitive GBS transport and culture medium than the
conventional method, especially for low inocula and prolonged transport/incubation times.
GT does not exclude the presence of GBS, and should always be incubated at 37°C and
subcultured on solid agar for optimal sensitivity.

82

Rene te Witt bw.indd 82 24-11-11 10:35

 Evaluation of Granada for detection of GBS

Introduction

Group B streptococci (GBS) has been known as a human pathogen since 1938. In the course
of pregnancy and the postpartum period, GBS may cause a variety of serious infections in
both the mother and neonate (1-2). Neonatal early-onset GBS disease (GBS-EOD) presents in
the first week of life and is usually acquired during delivery by neonates born from mothers
colonized with GBS in the rectovaginal tract. Up to 35% of pregnant women are colonized
with GBS in the rectovaginal tract, most often without symptoms (3-5). Approximately 1%
of neonates born to colonized mothers develop GBS-EOD and up to 40% of the surviving
neonates suffer serious sequelae, such as mental retardation or seizures (2). Mortality is high

Chapter 4
among preterm infants, with average case-fatality rates of approximately 20% (2). These rates
vary from as high as 30% among those children born before 33 weeks of gestation to 2-3%
among full-term infants (2).
Schrag et al. demonstrated that a prevention strategy based on routine screening for GBS
carriage prevents more cases of GBS-EOD than an approach based on risk-factor assessment
(6). Therefore, screening for GBS carriage during pregnancy is the key in many guidelines
to prevent GBS-EOD (2, 7). Consequently, sensitive and specific GBS cultivation is essential
and laboratory procedures need to be streamlined. Little attention has been paid to the
consequences of transport and transport delay of swabs to the laboratory. Revised Centers
for Disease Control and Prevention (CDC) guidelines state specifically that GBS isolates re-
main viable in Amies transport medium for several days at room temperature. However, the
recovery of isolates declines over a period of 1-4 days, particularly at elevated temperatures.
Even when appropriate transport media are used, the sensitivity of culture is highest when
the specimen is stored at 4°C before culture and processed within 24 h of collection (2, 8-11).
More rapid and, especially, more sensitive methods than the currently available and
recommended transport and culture methods would improve the effectiveness of the
screening-based approach. A new transport and enrichment broth, called Granada tube
broth (GT) (bioMérieux, Marcy l’Etoile, France) is about to be introduced. In this broth an
orange pigment is produced in the presence of GBS (12-13).
The aim of this study was to investigate whether the use of GT would improve the sensi-
tivity of GBS cultures in comparison with the current gold standard under various culture
conditions. We also investigated the reliability of GT after prolonged transport times.

83

Rene te Witt bw.indd 83 24-11-11 10:35

 Chapter 4

Materials and methods

Bacterial isolates

A subset of 33 colonizing and invasive GBS isolates representing all seven of the important
subtypes was obtained from a reference collection (Table 1) (14). All isolates were previously
identified as GBS using both biochemical and molecular methods (14). Strains were stored
at -80°C. Prior to testing, strains were subcultured twice on Columbia III agar with 5% sheep
blood (bioMérieux) for 18-24 h at 37°C.

Table 1. Group B streptococci (GBS) isolates used (ref. 14).

Serotype Invasive Colonizing
Ia 2 3
Ib 0 5
II 0 4
III 3 2
IV 0 4
V 3 2
VI 0 5

Variables and culture

Phase I

In the first phase of the study, we studied the viability of the GBS strains using GT and Amies
transport medium (AT) under different circumstances (five different incubation times [1, 2, 3,
4 and 7 days], three different temperatures [37°C, room temperature (RT) (±20°C), and 4°C]
and three different inocula). Of each strain, a suspension of 0.5 McFarland (~ 1.5x108 colony-
forming units [CFU]/ml) was prepared in sterile saline. This suspension was diluted until three
concentrations were obtained: 1.5x106, 1.5x104 and 1.5x102 CFU/ml, respectively. All suspen-
sions were subcultured on blood agar (BA) for purity checking and growth control. For every
strain, 45 GT and 45 AT were inoculated with 100 μl of suspension. After incubation, GT were
scored for the presence of orange pigment by two individuals and subcultured on BA for
the detection of growth. The presence of orange pigment was checked after 1, 2, 3, 4 and 7
days. All changes in colour were compared to a negative control tube, which was processed
similarly to the inoculated tubes. After incubation, both GT (using 10 μl of broth) and AT were
subcultured onto BA using the four-quadrant technique, to allow semi-quantification of the
number of GBS colonies. Growth was recorded after 1 and 2 days of incubation at 37°C and
was graded as negative (no growth), weakly positive (growth in the first quadrant), or positive
(growth in the second, third or fourth quadrants).

84

Rene te Witt bw.indd 84 24-11-11 10:35

 Evaluation of Granada for detection of GBS

Phase II

Phase II of the study was performed in parallel to phase I. GT initially incubated at 4°C or RT
in phase I were subsequently incubated at 37°C for 1 and 2 days and were checked for the
presence of orange pigment as described previously. After this second incubation period,
BA were inoculated and both GBS growth detection and quantification were performed as
described above.

Results

Chapter 4
Phase I

Overall, orange pigment was detected in 98% of all GT after incubation at 37°C for 1-7 days,
with the lowest detection rate at day 3 for 102 CFU/ml (90%). After incubation at RT, orange
pigment was observed after 3 days of incubation and in 10-40% of the GT (Figure 1). At 4°C,
in none of the GT was orange pigment detected.
Comparisons of GT with AT for growth detection for the different incubation times, dif-
ferent incubation temperatures, and different inocula are shown in Figure 2 and Table 2.
Data for semi-quantification via the four-quadrant technique showed no differences (data
not shown).

Table 2. Overall detection of GBS after subculture of Granada tube broth (GT) and Amies transport medium (AT) on blood agar (BA) after
incubation for 1, 2, 3, 4, or 7 days at 37°C, room temperature (RT) or 4°C using three different inocula (study phase I). A p-value <0.05 is
considered to be statistically significant.
106 CFU/ml GT culture-positive (%) AT culture-positive (%) p-value
37°C 94 43 <0.0001
RT 98 98 NS
4°C 100 100 NS
104 CFU/ml
37°C 93 13 <0.0001
RT 80 55 NS
4°C 63 75 NS
102 CFU/ml
37°C 84 5 <0.0001
RT 41 4 0.0002
4°C 3 3 NS

85

Rene te Witt bw.indd 85 24-11-11 10:35

 Chapter 4

10^6CFU/ml

100%
90%
80%
70% 37°CPigment

60% 37°CCulture
RTPigment
50%
RTCulture
40%
4°CPigment
30% 4°CCulture
20%
10%
0%
D1 D2 D3 D4 D7

Time

Figure 1A. 10^6 CFU/ml

10^4CFU/ml

100%
90%
80%
70% 37°CPigment

60% 37°CCulture
RTPigment
50%
RTCulture
40% 4°CPigment
30% 4°CCulture
20%
10%
0%
D1 D2 D3 D4 D7

Time
Figure 1B. 10^4 CFU/ml

10^2CFU/ml

100%
90%
80%
70% 37°CPigment

60% 37°CCulture
RTPigment
50%
RTCulture
40% 4°CPigment
30% 4°CCulture
20%
10%
0%
D1 D2 D3 D4 D7

Time

Figure 1C. 10^2 CFU/ml

Figure 1. Detection of orange pigment and group B streptococci (GBS)-positive cultures in Granada tube broth (GT) after incubation for 1, 2, 3,
4, or 7 days at 37°C, room temperature (RT), or 4°C using three different inocula (study phase I). In none of the GT at 4°C, was orange pigment
detected.

86

Rene te Witt bw.indd 86 24-11-11 10:35

 Evaluation of Granada for detection of GBS

Positive culture, 10^6 CFU/ml, phase I

100%

90%
80%
Percentage positive culture

70% 37°C GT

60% RT GT
4°C GT
50%
37°C AT
40% RT AT
30% 4°C AT
20%
10%

0%
D1 D2 D3 D4 D7

Chapter 4
Incubation time

Figure 2A. 10^6 CFU/ml

Positive culture, 10^4 CFU/ml, phase I

100%

90%

80%
Percentage positive culture

70% 37°C GT

60% RT GT
4°C GT
50%
37°C AT
40% RT AT
30% 4°C AT
20%

10%
0%
D1 D2 D3 D4 D7
Incubation time

Figure 2B. 10^4 CFU/ml

Positive culture, 10^2 CFU/ml, phase I

100%

90%
80%
Percentage positive culture

70% 37°C GT
60% RT GT
4°C GT
50%
37°C AT
40% RT AT
30% 4°C AT
20%
10%
0%
D1 D2 D3 D4 D7
Incubation time

Figure 2C. 10^2 CFU/ml

Figure 2. Detection of GBS after subculture of GT and Amies transport medium (AT) on blood agar (BA) after initial incubation for 1, 2, 3, 4, or 7
days at 37°C, RT, or 4°C using three different inocula (study phase I).

87

Rene te Witt bw.indd 87 24-11-11 10:35

 Chapter 4

Phase II

Data on the observation of orange pigment after subsequent incubation at 37°C and the
detection of GBS after subculture on BA are shown in Figure 3.
No differences were observed in the presence of orange pigment and positive cultures
between 1 and 2 days subsequent incubation (data not shown). Furthermore, no differences
were observed between the different virulence types or between the different serotypes
(data not shown).

Discussion

Granada agar has long been used in Europe to detect GBS in pregnant women. Most strains of
beta-hemolytic GBS produce an orange carotenoid pigment on this agar, usually within 24h
of incubation. Variable results on the sensitivity of this culture medium have been reported,
with some authors considering it to be unacceptably low (15-18). However, our results can-
not be compared with the results of these studies as we tested Granada broth instead of agar.
On average, we observed the production of orange pigment in 98% of all GT that were
incubated at 37°C for 1-7 days, with the lowest detection rate at day 3 for 102 CFU/ml (90%).
At room temperature, orange pigment was observed only after 3 days of incubation and in
only 10-40% of the GT. At 4°C, in none of the GT was orange pigment detected. As the overall
sensitivity for GT for GBS culture was 95%, independent of the incubation-temperature and
the production of orange pigment, our results show that the absence of orange pigment
does not exclude the presence of GBS. Furthermore, because the GBS pigment is linked to
hemolysin activity, less or non-hemolytic strains may not be detected with GT (19). Therefore,
it may be reasonable to subculture all GT (both positive and negative) on BA. This allows the
recognition of non-hemolytic GBS strains and other GBS that may not produce the orange
pigment. This is important for identification, susceptibility testing and potential typing of
the cultured straiIn our simulation of transport conditions, GBS could be cultured only in
20%, 52%, and 59% of AT incubated at 37°C, RT, or 4°C, respectively. These percentages cor-
respond with those found by Rosa-Fraile et al., Stoner et al., and with revised CDC guidelines
(2, 10-11). The sensitivity of GT was significantly higher than that of AT, especially for low
inocula and extended transport/incubation time. This is important, since specimens may be
exposed to high temperatures during transport, especially when swabs are obtained outside
the hospital.
Direct incubation of GT at 37°C resulted in the highest yield of GBS. No difference in the
growth of GBS was observed between transport at RT or at 4°C. Without incubation at 37°C
after transport, the growth of GBS was seen in 73% and 55% of GT transported at RT or 4°C,

88

Rene te Witt bw.indd 88 24-11-11 10:35

 Evaluation of Granada for detection of GBS

10^6CFU/ml

100%
90%
80%
70%
60% RTPigment
RTCulture
50%
4°CPigment
40% 4°CCulture
30%
20%
10%
0%
1 2 3 4 5

Time

Chapter 4
Figure 3A. 10^6 CFU/ml

10^4CFU/ml

100%
90%
80%
70%
60% RTPigment
RTCulture
50%
4°CPigment
40% 4°CCulture
30%
20%
10%
0%
1 2 3 4 5

Time
Figure 3B. 10^4 CFU/ml

10^2CFU/ml

100%
90%
80%
70%
60% RTPigment
RTCulture
50%
4°CPigment
40% 4°CCulture
30%
20%
10%
0%
1 2 3 4 5

Time
Figure 3C. 10^2 CFU/ml

Figure 3. Detection of orange pigment and GBS-positive cultures in GT after subsequent incubation at 37°C for 24 h after an initial incubation
for 1, 2, 3, 4, or 7 days at RT or 4°C using three different inocula (study phase II).

89

Rene te Witt bw.indd 89 24-11-11 10:35

 Chapter 4

respectively. When GT were subsequently incubated at 37°C, after transport at RT or 4°C, the
sensitivity increased to almost 100%.
The study is limited in that it is performed in vitro and that the application of purified
isolates of GBS in GT and onto AT may not reflect GBS survival on swabs containing vaginal
and/or rectal flora. Nonetheless, we can conclude that GT is a highly sensitive transport and
culture medium to detect GBS in pregnant women. The survival of GBS was significantly bet-
ter in GT when compared to AT.
GT may not be suitable for the direct detection of GBS, as the absence of orange pigment
does not conclude in the absence of GBS. Therefore, GT should always be subcultured for
optimal use. Furthermore, GT should always be incubated at 37°C to improve its sensitivity.
GT may especially be suited for the transport of swab specimens from general practitioners
and midwifery practices to the laboratory, which may take 2-3 days. Specifically, if a woman
has low-density GBS colonization, extended transport times of swab specimens at RT or
higher could reduce the culture sensitivity for AT but possibly not for GT.

Acknowledgements

The Granada tubes, Amies transport media, and Columbia III agar with 5% sheep blood used
in this study were provided by bioMérieux free of charge.

Conflicts of interest

Alex van Belkum is an employee of bioMérieux. There are no conflicts of interest to disclose.

90

Rene te Witt bw.indd 90 24-11-11 10:35

 Evaluation of Granada for detection of GBS

References
1. Muller, A. E., et al. (2006). Morbidity related to maternal group B streptococcal infections. Acta Obstet Gynecol
Scand 85(9): 1027-37.
2. Verani, J. R., et al. (2010). Prevention of perinatal group B streptococcal disease--revised guidelines from CDC,
2010. MMWR Recomm Rep 59(RR-10): 1-36.
3. Bergseng, H., et al. (2007). Real-time PCR targeting the sip gene for detection of group B Streptococcus coloniza-
tion in pregnant women at delivery. J Med Microbiol 56(Pt 2): 223-8.
4. Campbell, J. R., et al. (2000). Group B streptococcal colonization and serotype-specific immunity in pregnant
women at delivery. Obstet Gynecol 96(4): 498-503.
5. Valkenburg-van den Berg, A. W., et al. (2006). Prevalence of colonisation with group B Streptococci in pregnant
women of a multi-ethnic population in The Netherlands. Eur J Obstet Gynecol Reprod Biol 124(2): 178-83.

Chapter 4
6. Schrag, S. J., et al. (2002). A population-based comparison of strategies to prevent early-onset group B streptococ-
cal disease in neonates. N Engl J Med 347(4): 233-9.
7. Money, D. M., Dobson, S. and Canadian Paediatric Society, I. D. C. (2004). The prevention of early-onset neonatal
group B streptococcal disease. J Obstet Gynaecol Can 26(9): 826-40.
8. Hakansson, S., et al. (2008). Group B streptococcal carriage in Sweden: a national study on risk factors for mother
and infant colonisation. Acta Obstet Gynecol Scand 87(1): 50-8.
9. Ostroff, R. M. and Steaffens, J. W. (1995). Effect of specimen storage, antibiotics, and feminine hygiene products on
the detection of group B Streptococcus by culture and the STREP B OIA test. Diagn Microbiol Infect Dis 22(3): 253-9.
10. Rosa-Fraile, M., et al. (2005). Specimen storage in transport medium and detection of group B streptococci by
culture. J Clin Microbiol 43(2): 928-30.
11. Stoner, K. A., Rabe, L. K. and Hillier, S. L. (2004). Effect of transport time, temperature, and concentration on the
survival of group B streptococci in amies transport medium. J Clin Microbiol 42(11): 5385-7.
12. Martinho, F., et al. (2008). Evaluation of liquid biphasic Granada medium and instant liquid biphasic Granada
medium for group B streptococcus detection. Enferm Infecc Microbiol Clin 26(2): 69-71.
13. Heelan, J. S., et al. (2005). Evaluation of a new selective enrichment broth for detection of group B streptococci in
pregnant women. J Clin Microbiol 43(2): 896-7.
14. van Elzakker, E., et al. (2009). Epidemiology of and prenatal molecular distinction between invasive and colonizing
group B streptococci in The Netherlands and Taiwan. Eur J Clin Microbiol Infect Dis 28(8): 921-8.
15. Gil, E. G., et al. (1999). Evaluation of the Granada agar plate for detection of vaginal and rectal group B streptococci
in pregnant women. J Clin Microbiol 37(8): 2648-51.
16. Overman, S. B., et al. (2002). Evaluation of methods to increase the sensitivity and timeliness of detection of
Streptococcus agalactiae in pregnant women. J Clin Microbiol 40(11): 4329-31.
17. Perry, J. D., et al. (2006). Evaluation of a new chromogenic agar medium for isolation and identification of Group B
streptococci. Lett Appl Microbiol 43(6): 615-8.
18. Rosa-Fraile, M., et al. (1999). Use of Granada medium to detect group B streptococcal colonization in pregnant
women. J Clin Microbiol 37(8): 2674-7.
19. Tazi, A., et al. (2008). Comparative evaluation of Strepto B ID chromogenic medium and Granada media for the
detection of Group B streptococcus from vaginal samples of pregnant women. J Microbiol Methods 73(3): 263-5.

91

Rene te Witt bw.indd 91 24-11-11 10:35

Rene te Witt bw.indd 92 24-11-11 10:35
 Nucleic acid amplification

Rene te Witt bw.indd 93 24-11-11 10:35

Rene te Witt bw.indd 94 24-11-11 10:35
 Chapter 5
Molecular diagnostics and genotyping of
methicillin-resistant Staphylococcus aureus
(MRSA): an update

René te Witt
Alex van Belkum
Willem B. van Leeuwen

Expert Reviews of Molecular Diagnostics, 2010; 10: 375-80

Rene te Witt bw.indd 95 24-11-11 10:35

Rene te Witt bw.indd 96 24-11-11 10:35
 Molecular diagnostics and genotyping of methicillin-resistant Staphylococcus aureus (MRSA): an update

The organism

Staphylococcus aureus is a significant human pathogen that can cause a wide variety of
diseases due to its ability to acquire and express various virulence factors and antimicrobial
resistance determinants. Colonization is an important step in the pathogenesis of S. aureus
infection. Approximately 20-30% of the general human population is persistently colonized
with S. aureus, most frequently in the anterior nares, although other body sites, such as the
perineum, skin and throat, may also be colonized. Another 30% of the general population is
intermittently colonized and the remaining 40-50% appear to be not susceptible to S. aureus
carriage (1-2).
Persistent nasal carriers of S. aureus have a three to six times higher risk of health-care
associated infections with S. aureus than non-carriers (3-5). More than 80% of health-care
associated S. aureus infections are endogenous (6-8). Recently it has been shown that the

Chapter 5
number of surgical-site S. aureus infections acquired in the hospital can be reduced by rapid
screening and decolonization of nasal carriers of S. aureus on admission (9).
Following the introduction of methicillin in 1959, methicillin-resistant S. aureus (MRSA)
has quickly and widely emerged as a major nosocomial burden worldwide (10). Today,
MRSA continues to be a major issue in hospitals but has also emerged as a problem in the
community (11). Methicillin resistance in S. aureus is encoded by the mecA gene, which is
embedded within a mobile staphylococcal cassette chromosome (SCC) element known as
SCCmec. MRSA can emerge from methicillin-susceptible S. aureus (MSSA) upon site-specific
integration of SCCmec into the orfX locus in the chromosome. To date, nine major types of
SCCmec have been recognized in S. aureus (12).

Phenotypic detection and identification of MRSA

Rapid, high-throughput diagnostic tools are needed to detect infections caused by MRSA
strains and to contain their spread. The traditional screening cultures require at least 48 hours
until a negative test result for MRSA can be reported. These methods have been superseded
by the introduction of selective agar cultivation methods and latex agglutination tests for
the specific detection of the product of the mecA gene, Penicillin Binding Protein 2a (PBP 2a).
Most of these tests have been evaluated in different studies (13-26). Sensitivity and specificity
for agar cultivation ranged from 40-98% and 80-100%, respectively. For latex agglutination
tests, both sensitivity and specificity were almost 100%.
The first commercial example of a rapid screening test for detection and identification of
MRSA, the BacLite Rapid MRSA assay (Acolyte Biomedia, United Kingdom) is documented
as a sensitive (90.4%) and specific (95.7%) test for the detection of MRSA nasal colonization

97

Rene te Witt bw.indd 97 24-11-11 10:35

 Chapter 5

within 5 hours (27). The test measures adenylate kinase activity in a selective broth during a
4h growth episode.
Recently, Stenholm et al described the results of a new culture-based method for MRSA
screening (28). This new method employs the two-photon excited fluorescence (TPX) technol-
ogy to provide a quantitative S. aureus-specific fluorescence signal in a single separation-free
process. Different fluorescence signal progressions are recorded for MRSA and MSSA when
bacterial growth under conditions of antibiotic pressure is monitored online by continuous
or repeated measurements. When monoclonal antibodies were used, the assay was 100%
sensitive and 100% specific for screening for MRSA from pure cultured samples and results
were available within 8 to 12 hours.
However, most phenotypic techniques still require 4-24 hours of cultivation before detec-
tion and identification of S. aureus and testing for methicillin resistance can be initiated and
may therefore not be suitable for fast screening of patients.

Genotypic detection and identification

Nucleic Acid Amplification Techniques (NAATs) offer clear benefits over traditional culture-
based assays, in particular a reduced time to identification and an improved specificity and
sensitivity. Over the past decade, a range of commercial and in-house developed NAATs has
been introduced (For an overview, see Table 1) with 2 main strategies: identification of MRSA
via detection of S. aureus-specific genes, such as spa, nuc and fem in combination with the
detection of the mecA gene or via detection of the SCCmec region. However, sensitivity and
specificity of these assays may be compromised due to the presence of methicillin-resistant
coagulase negative staphylococci (MRCoNS), variability within the used S. aureus specific
genes or variability within the SCCmec cassette. This may lead to false-negative results (e.g.
gene mutation or SCCmec variants) or false-positive results (e.g. deletion of the mecA gene
from the SCCmec region) (29-32).

The detection of the mecA gene by (real-time) PCR is widely recognized as the gold standard
for identification of MRSA. A variety of in-house assays performed on several different instru-
ments has been described in the literature with variable results (Table 1). Most of these are
performed on an overnight selective enrichment broth. The major limitations of this PCR
strategy for direct testing in clinical samples were the necessity of enrichment and the in-
ability to directly link identification of S. aureus with mecA gene detection, because of the
confounding effect caused by MRCoNS. In 2004, Huletsky et al described a novel real-time
PCR assay (33). This assay is able to distinguish between MRSA and mixtures of MSSA and
MRCoNS and therefore suitable for direct identification of MRSA in clinical specimen. The
assay became commercially available as the IDI-MRSA assay and is currently marketed as the

98

Rene te Witt bw.indd 98 24-11-11 10:35

 Molecular diagnostics and genotyping of methicillin-resistant Staphylococcus aureus (MRSA): an update

Table 1. Commercial molecular tests for MRSA detection and identification.

Commercial
Testkit (supplier) Test principle Sensitivity (%) Specificity (%) TAT (1) Reference
Culture
Genotype MRSA (Hain) PCR and reverse line blot detection of mecA and 91 99 4h (51)
S. aureus specific sequence
Direct from clinical specimen (swabs)
HyplexStaphyloResist PCR-ELISA for detection of mecA, coa and 92 90 4h (52)
(Alpha Omega)
conservedS. aureus housekeeping gene
GenoType MRSA direct PCR and reverse line blot detection 95 99 4h (53)
(Hain) of SCCmec/orfX
GeneOhm IDI-MRSA Multiplex Real-time PCR for detection 95 94 2h (33, 54-55)
(BD) of SCCmec/orfX
GeneXpert MRSA assay Multiplex Real-time PCR for detection 90 97 80’ (35)
of SCCmec/orfX

Chapter 5
In-house
Enrichment
NA (2) Multiplex PCR (agarose gel) detecting femB and 100 100 3h (56-57)
mecA
NA (2) Real-time PCR detecting mecA 100 100 2h (58)
NA (2) High-throughput real-time PCR detecting mecA 100 100 2,5h (59)
and nuc
NA (2) Real-time PCR detecting mecAand SA442 100 100 4h (60)
NA (2) Isothermal signal amplification (CytAMP) 83 93 3,5h (61)
detecting coa and mecA
NA (2) Real-time PCR detecting nuc and mecA 93 90 2,5h (62)
Direct from clinical
specimen (swabs)
NA (2) Real-time PCR detecting mecA and femA 100 91 6h (63)
NA (2) S. aureusImmunocapturing (protein A) and 100 64 6h (64)
NA (2) quantitative PCR detectingmecA, femA and femB
NA (2) Real-time PCR detecting SCCmec/orfX 93 (98)* 100 (100)* 3,5h (65)
*) Percentage sensitivity and specificity between brackets are after overnight selective enrichment.
(1) TAT: turn-around-time
(2) NA: not applicable

BD GeneOhm MRSA assay (BD GeneOhm, San Diego, California). When testing nasal swabs,
the detection limit was suggested to be 325 CFU per swab. However, in an External Quality
Assessment (EQA) performed by Quality Control for Molecular Diagnostics (QCMD, Glasgow,
Scotland) in 2009, the limit of detection was found to be 103-104 CFU/ml (34). Also, when
evaluating the performance of this test, Bartels et al described how a common variant of
SCCmec type IVa was not detected by the BD GeneOhm assay (30).
Rossney et al evaluated a commercial platform based on the BD GeneOhm principal for di-
rect detection of MRSA in clinical samples, the GeneXpert MRSA assay (Cepheid Diagnostics,

99

Rene te Witt bw.indd 99 24-11-11 10:35

 Chapter 5

Sunnyvale, California) (35). In this fully automatic, but low-throughput assay, cells are lysed
using ultrasonics and DNA from samples is extracted, amplified and detected in separate
chambers of single-use disposable cartridges which contain freeze-dried beads with all
reagents required for the real-time process. Sample preparation time is minimal and the PCR
assay time is 75 minutes maximum. The authors reported an average limit of detection of 610
CFU/ml with a sensitivity, specificity, positive predictive value and negative predictive value
of 90%, 97%, 86% and 98%, respectively.
In the QCMD EQA study of 2009, eleven samples containing various amounts of inacti-
vated MRSA cells, methicillin-susceptible S. aureus (MSSA), MRCoNS or Escherichia coli, were
distributed to 82 laboratories (34). When compared with previous EQA studies on molecular
diagnostics of MRSA, a statistically non-significant decrease was observed in the overall test
sensitivity. However, a minor improvement was observed for the ‘specificity’ and the ‘true-
negative’ samples. In this EQA, one normal MSSA strain and two MSSA samples harbouring a
SCCmec cassette lacking the mecA gene were included. There was a marked difference in the
percentage of correct results for the MSSA strain containing the mecA gene compared to the
two strains lacking it.
In conclusion, the quality of direct molecular diagnostic tests still needs improvement.
Every assay should be evaluated and continuously monitored to determine the assay’s use-
fulness. Furthermore, positive results should always be confirmed by a culture method or a
second molecular test.

Genotyping

After MRSA detection, genetic typing may be necessary in order to assess whether trans-
mission of MRSA occurred and whether infection-preventive measures are mandatory. In
addition, genetic typing is essential for elucidation of (inter)national dissemination of MRSA
clones. Currently, many different genotyping methods are in use in the diagnostic laboratory,
but pulsed-field gel electrophoresis (PFGE) of SmaI macro restriction analysis of genomic
DNA is preferential (36). However, PFGE is a technically demanding method, with limited
portability due to lack of reproducibility.
The development of a commercially available automated rep-PCR assay, the DiversiLab
system (bioMérieux, Marcy l’Etoile, France) offers advances in standardization and reproduc-
ibility over manual fingerprint generating systems (37). However, although two independent
studies concluded that the DiversiLab system is a rapid and reproducible technique, it also
lacks resolution. DiversiLab analysis does not differentiate genetically and epidemiologically
unique MRSA strains, which is needed for adequate outbreak analysis (38-39).
Multi-locus variable number of tandem repeat analysis (MLVA) is a high-throughput
genotyping technique that can be used for hospital, national and international genotyping

100

Rene te Witt bw.indd 100 24-11-11 10:35

 Molecular diagnostics and genotyping of methicillin-resistant Staphylococcus aureus (MRSA): an update

of MRSA, but the discriminatory power depends on the number and types of loci analyzed
(40-41).
Sequence-based approaches, such as spa sequence typing and multi-locus sequence
typing (MLST), have resulted in large sequence databases for MRSA. The determination of
sequence polymorphism of the spa gene encoding the staphylococcal surface protein A
(spa sequence typing) has become the most popular MRSA typing system, thanks to high-
throughput capacity and an excellent reproducibility, which allows portability of data and
comparison worldwide (42). MLST defines variation within a very small sample of the genome
and often cannot distinguish between closely related isolates.
Full-genome sequencing provides a complete inventory of micro-evolutionary changes,
but this approach is impractical for routine diagnostic laboratories. In a recent paper, Harris et
al described a new high-throughput genomics approach based on full-genome sequencing,
which provides a high-resolution view of the epidemiology of MRSA with the potential to

Chapter 5
trace person-to-person transmission within a hospital environment (43).

New techniques

New tools for identification and typing of MRSA include different applications of spectrosco-
py, such as PCR/Electrospray Ionization-Mass Spectrometry (44-45) and matrix-assisted laser
desorption ionization-time of flight mass spectrometry (MALDI-TOF MS) (46). MALDI-TOF MS
is cost-effective, analyzes samples in minutes and requires little hands-on-time.
Also Raman spectroscopy has been described as a promising tool (47-48). Raman spec-
troscopy is a fast and reproducible typing technique, which provides strain-specific optical
fingerprints in a few minutes instead of several hours to days, as is the case with genotyping
methods. Its high throughput and ease of use make it suitable for use in routine diagnostic
laboratories. Efforts to develop these technologies for the analysis of single cells are currently
in full progress and may in the end compete effectively with the currently preferred nucleic
acid based technologies.
Instead of detecting S. aureus itself, new strategies may be to look at the host immune
response or to look at genetic variation in the host.
The rapid detection of antibody levels against S. aureus with Luminex technology showed
that antibody levels were associated with the presence of toxin genes in infectious S. aureus
isolates (49). And Emonts et al (50) showed that persistent carriage of S. aureus is influenced
by and associated with genetic variation in host inflammatory response genes.
Both approaches can be useful in fast screening for (susceptibility to) (methicillin-resistant)
S. aureus carriage or infection.

101

Rene te Witt bw.indd 101 24-11-11 10:35

 Chapter 5

Concluding remarks

MRSA is responsible for a large and still growing number of both health-care and commu-
nity-associated infections, resulting in increased morbidity and excessive healthcare costs.
Screening of individuals combined with an aggressive infection control program has become
the standard for management of these infections. Rapid screening methods that allow reli-
able detection of MRSA within hours are now available. The short time-to-result is a clear
advantage that has provided a tool for successful infection control strategies. However, every
assay should be evaluated against the local MRSA diversity before being introduced in the
diagnostic microbiological laboratory. Continuous evolution of SCCmec, constrains continu-
ously monitoring of the assay performance and positive results of direct MRSA testing should
always be confirmed by a culture method or a second molecular test. For laboratories with
high false positivity rates or in regions with low prevalence of MRSA, confirmation is essential.
In conclusion, the quality of molecular diagnostic tests and (geno)typing techniques is still
under discussion. Adequate internal and external quality control and international standard-
ization for MRSA diagnostics should be developed over the years to come. To improve the
performance and quality of molecular detection, identification and typing of MRSA, both
laboratories and manufacturers should be encouraged to participate in EQAs.
Genetic and functional studies in large populations are warranted to clarify the contribu-
tion of new strategies such as different applications of spectroscopy and spectrometry,
determination of genetic variability in humans or rapid antibody – and/or antigen-detection.

Potential conflicts of interest

Alex van Belkum is member of the scientific advisory board of Cepheid (Cepheid Diagnostics,
Sunnyvale, California).

102

Rene te Witt bw.indd 102 24-11-11 10:35

 Molecular diagnostics and genotyping of methicillin-resistant Staphylococcus aureus (MRSA): an update

References
1. Wertheim, H. F., et al. (2005). The role of nasal carriage in Staphylococcus aureus infections. Lancet Infect Dis 5(12):
751-62.
2. Nilsson, P. and Ripa, T. (2006). Staphylococcus aureus throat colonization is more frequent than colonization in the
anterior nares. J Clin Microbiol 44(9): 3334-9.
3. Luzar, M. A., et al. (1990). Staphylococcus aureus nasal carriage and infection in patients on continuous ambulatory
peritoneal dialysis. N Engl J Med 322(8): 505-9.
4. Kluytmans, J. A., et al. (1995). Nasal carriage of Staphylococcus aureus as a major risk factor for wound infections
after cardiac surgery. J Infect Dis 171(1): 216-9.
5. Nouwen, J., et al. (2006). Staphylococcus aureus carriage patterns and the risk of infections associated with con-
tinuous peritoneal dialysis. J Clin Microbiol 44(6): 2233-6.
6. Weinstein, H. J. (1959). The relation between the nasal-staphylococcal-carrier state and the incidence of postop-
erative complications. N Engl J Med 260(26): 1303-8.
7. von Eiff, C., et al. (2001). Nasal carriage as a source of Staphylococcus aureus bacteremia. Study Group. N Engl J Med

Chapter 5
344(1): 11-6.
8. Wertheim, H. F., et al. (2004). Risk and outcome of nosocomial Staphylococcus aureus bacteraemia in nasal carriers
versus non-carriers. Lancet 364(9435): 703-5.
9. Bode, L. G., et al. Preventing surgical-site infections in nasal carriers of Staphylococcus aureus. N Engl J Med 362(1):
9-17.
10. Kluytmans, J. and Struelens, M. (2009). Meticillin resistant Staphylococcus aureus in the hospital. Bmj 338: b364.
11. Tenover, F. C., et al. (2006). Characterization of a strain of community-associated methicillin-resistant Staphylococ-
cus aureus widely disseminated in the United States. J Clin Microbiol 44(1): 108-18.
12. International Working Group on the Classification of Staphylococcal Cassette Chromosome, E. (2009). Classifica-
tion of staphylococcal cassette chromosome mec (SCCmec): guidelines for reporting novel SCCmec elements.
Antimicrob Agents Chemother 53(12): 4961-7.
13. Han, Z., et al. (2007). Evaluation of mannitol salt agar, CHROMagar Staph aureus and CHROMagar MRSA for detec-
tion of meticillin-resistant Staphylococcus aureus from nasal swab specimens. J Med Microbiol 56(Pt 1): 43-6.
14. Compernolle, V., Verschraegen, G. and Claeys, G. (2007). Combined use of Pastorex Staph-Plus and either of two
new chromogenic agars, MRSA ID and CHROMagar MRSA, for detection of methicillin-resistant Staphylococcus
aureus. J Clin Microbiol 45(1): 154-8.
15. Nahimana, I., Francioli, P. and Blanc, D. S. (2006). Evaluation of three chromogenic media (MRSA-ID, MRSA-Select
and CHROMagar MRSA) and ORSAB for surveillance cultures of methicillin-resistant Staphylococcus aureus. Clin
Microbiol Infect 12(12): 1168-74.
16. Stoakes, L., et al. (2006). Prospective comparison of a new chromogenic medium, MRSASelect, to CHROMagar
MRSA and mannitol-salt medium supplemented with oxacillin or cefoxitin for detection of methicillin-resistant
Staphylococcus aureus. J Clin Microbiol 44(2): 637-9.
17. Zadik, P. M., et al. (2001). Evaluation of a new selective medium for methicillin-resistant Staphyloccocus aureus. J
Med Microbiol 50(5): 476-9.
18. Davies, S. and Zadik, P. M. (1997). Comparison of methods for the isolation of methicillin resistant Staphylococcus
aureus. J Clin Pathol 50(3): 257-8.

103

Rene te Witt bw.indd 103 24-11-11 10:35

 Chapter 5

19. Hamdad, F., et al. (2006). Detection of methicillin/oxacillin resistance and typing in aminoglycoside-susceptible
methicillin-resistant and kanamycin-tobramycin-resistant methicillin-susceptible Staphylococcus aureus. Microb
Drug Resist 12(3): 177-85.
20. Monno, R., et al. (2003). Comparative evaluation of test assays for detection of methicillin resistance in Staphylo-
coccus aureus. Clin Microbiol Infect 9(6): 574-5.
21. Rohrer, S., et al. (2001). Improved methods for detection of methicillin-resistant Staphylococcus aureus. Eur J Clin
Microbiol Infect Dis 20(4): 267-70.
22. van Leeuwen, W. B., et al. (1999). Rapid detection of methicillin resistance in Staphylococcus aureus isolates by the
MRSA-screen latex agglutination test. J Clin Microbiol 37(9): 3029-30.
23. Yamazumi, T., et al. (2001). Comparison of the Vitek gram-positive susceptibility 106 card, the MRSA-Screen latex
agglutination test, and mecA analysis for detecting oxacillin resistance in a geographically diverse collection of
clinical isolates of coagulase-negative staphylococci. J Clin Microbiol 39(10): 3633-6.
24. Levi, K. and Towner, K. J. (2003). Detection of methicillin-resistant Staphylococcus aureus (MRSA) in blood with the
EVIGENE MRSA detection kit. J Clin Microbiol 41(8): 3890-2.
25. Arbique, J., et al. (2001). Comparison of the Velogene Rapid MRSA Identification Assay, Denka MRSA-Screen As-
say, and BBL Crystal MRSA ID System for rapid identification of methicillin-resistant Staphylococcus aureus. Diagn
Microbiol Infect Dis 40(1-2): 5-10.
26. van Leeuwen, W. B., Kreft, D. E. and Verbrugh, H. (2002). Validation of rapid screening tests for the identification of
methicillin resistance in staphylococci. Microb Drug Resist 8(1): 55-9.
27. Johnson, G., et al. (2006). Evaluation of BacLite Rapid MRSA, a rapid culture based screening test for the detection
of ciprofloxacin and methicillin resistant S. aureus (MRSA) from screening swabs. BMC Microbiol 6: 83.
28. Stenholm, T., et al. (2009). Methicillin-resistant Staphylococcus aureus screening by online immunometric monitor-
ing of bacterial growth under selective pressure. Antimicrob Agents Chemother 53(12): 5088-94.
29. Ibrahem, S., et al. (2009). Carriage of methicillin-resistant Staphylococci and their SCCmec types in a long-term-
care facility. J Clin Microbiol 47(1): 32-7.
30. Bartels, M. D., et al. (2009). A common variant of staphylococcal cassette chromosome mec type IVa in isolates
from Copenhagen, Denmark, is not detected by the BD GeneOhm methicillin-resistant Staphylococcus aureus
assay. J Clin Microbiol 47(5): 1524-7.
31. Jansen, W. T., et al. (2006). Novel mobile variants of staphylococcal cassette chromosome mec in Staphylococcus
aureus. Antimicrob Agents Chemother 50(6): 2072-8.
32. Ender, M., Berger-Bachi, B. and McCallum, N. (2007). Variability in SCCmecN1 spreading among injection drug
users in Zurich, Switzerland. BMC Microbiol 7: 62.
33. Huletsky, A., et al. (2004). New real-time PCR assay for rapid detection of methicillin-resistant Staphylococcus
aureus directly from specimens containing a mixture of staphylococci. J Clin Microbiol 42(5): 1875-84.
34. Te Witt, R., et al. External quality assessment of the molecular diagnostics and genotyping of meticillin-resistant
Staphylococcus aureus. Eur J Clin Microbiol Infect Dis.
35. Rossney, A. S., et al. (2008). Evaluation of the Xpert methicillin-resistant Staphylococcus aureus (MRSA) assay using
the GeneXpert real-time PCR platform for rapid detection of MRSA from screening specimens. J Clin Microbiol
46(10): 3285-90.
36. Ichiyama, S., et al. (1991). Genomic DNA fingerprinting by pulsed-field gel electrophoresis as an epidemiological
marker for study of nosocomial infections caused by methicillin-resistant Staphylococcus aureus. J Clin Microbiol
29(12): 2690-5.

104

Rene te Witt bw.indd 104 24-11-11 10:35

 Molecular diagnostics and genotyping of methicillin-resistant Staphylococcus aureus (MRSA): an update

37. Healy, M., et al. (2005). Microbial DNA typing by automated repetitive-sequence-based PCR. J Clin Microbiol 43(1):
199-207.
38. Te Witt, R., Kanhai, V. and van Leeuwen, W. B. (2009). Comparison of the DiversiLab system, Pulsed-Field Gel Electro-
phoresis and Multi-Locus Sequence Typing for the characterization of epidemic reference MRSA strains. J Microbiol
Methods 77(1): 130-3.
39. Tenover, F. C., et al. (2009). Comparison of typing results obtained for methicillin-resistant Staphylococcus aureus
isolates with the DiversiLab system and pulsed-field gel electrophoresis. J Clin Microbiol 47(8): 2452-7.
40. Schouls, L. M., et al. (2009). Multiple-locus variable number tandem repeat analysis of Staphylococcus aureus:
comparison with pulsed-field gel electrophoresis and spa-typing. PLoS One 4(4): e5082.
41. Pourcel, C., et al. (2009). Improved multiple-locus variable-number tandem-repeat assay for Staphylococcus aureus
genotyping, providing a highly informative technique together with strong phylogenetic value. J Clin Microbiol
47(10): 3121-8.
42. Friedrich, A. W., et al. (2008). A European laboratory network for sequence-based typing of methicillin-resistant
Staphylococcus aureus (MRSA) as a communication platform between human and veterinary medicine--an update
on SeqNet.org. Euro Surveill 13(19).

Chapter 5
43. Harris, S. R., et al. Evolution of MRSA during hospital transmission and intercontinental spread. Science 327(5964):
469-74.
44. Hall, T. A., et al. (2009). Rapid molecular genotyping and clonal complex assignment of Staphylococcus aureus
isolates by PCR coupled to electrospray ionization-mass spectrometry. J Clin Microbiol 47(6): 1733-41.
45. Wolk, D. M., et al. (2009). Pathogen profiling: rapid molecular characterization of Staphylococcus aureus by PCR/
electrospray ionization-mass spectrometry and correlation with phenotype. J Clin Microbiol 47(10): 3129-37.
46. Stevenson, L. G., Drake, S. K. and Murray, P. R. Rapid identification of bacteria in positive blood culture broths by
matrix-assisted laser desorption ionization-time of flight mass spectrometry. J Clin Microbiol 48(2): 444-7.
47. Willemse-Erix, D. F., et al. (2009). Optical fingerprinting in bacterial epidemiology: Raman spectroscopy as a real-
time typing method. J Clin Microbiol 47(3): 652-9.
48. Maquelin, K., et al. (2003). Prospective study of the performance of vibrational spectroscopies for rapid identifica-
tion of bacterial and fungal pathogens recovered from blood cultures. J Clin Microbiol 41(1): 324-9.
49. Verkaik, N. J., et al. Immunogenicity of toxins during Staphylococcus aureus infection. Clin Infect Dis 50(1): 61-8.
50. Emonts, M., et al. (2008). Host polymorphisms in interleukin 4, complement factor H, and C-reactive protein associ-
ated with nasal carriage of Staphylococcus aureus and occurrence of boils. J Infect Dis 197(9): 1244-53.

105

Rene te Witt bw.indd 105 24-11-11 10:35

Rene te Witt bw.indd 106 24-11-11 10:35
 Bacterial typing

Rene te Witt bw.indd 107 24-11-11 10:35

Rene te Witt bw.indd 108 24-11-11 10:35
 Chapter 6
Comparison of the DiversiLab system,
pulsed-field gel electrophoresis
and multi-locus sequence typing
for the characterization of epidemic
reference MRSA strains

René te Witt
Vikash Kanhai
Willem B. van Leeuwen

Journal of Microbiological Methods, 2009; 77: 130-3

Rene te Witt bw.indd 109 24-11-11 10:35

 Chapter 6

Abstract

We have analyzed a representative selection of the HARMONY methicillin-resistant Staphy-

lococcus aureus strain collection originating from 11 European countries (Cookson, B.D. et al.
2007, J. Clin. Microbiol. 45: 1830-1837) with the DiversiLab System, pulsed-field gel electro-
phoresis (PFGE) and multi-locus sequence typing (MLST). Simpson’s diversity indices were
0.905, 0.877 and 0.860 for PFGE, MLST and DiversiLab, respectively. All methods displayed
concordant classification of the MRSA strains, although with divergent resolution and repro-
ducibility.

110

Rene te Witt bw.indd 110 24-11-11 10:35

 Comparison of the DiversiLab system, PFGE and MLST for typing of MRSA

Introduction

Methicillin-resistant Staphylococcus aureus (MRSA) isolates emerged soon after the introduc-
tion of methicillin and has become an increasing problem in both healthcare settings and in
the community worldwide (1). The impact on patient morbidity, mortality and healthcare-
associated costs is significant (2-6). Therefore, controlling the spread of this pathogen by
screening patients, personnel and the environment remains a high priority in infection
control programs. Tracing the source and transmission routes of MRSA relies on typing meth-
ods as tools for the genetic characterization of isolates (7). Pulsed-field gel electrophoresis
(PFGE) has been accepted as the reference method for molecular strain typing of MRSA. PFGE
is known to be highly discriminatory and therefore frequently used for outbreak analysis
(8). However, this strategy is labor-intensive, time-consuming and technical instability has
an adverse effect on reproducibility (9). Therefore, results are not interchangeable. Other
molecular fingerprint-generating strain typing methods such as manual rep-PCR (9-10) and
randomly amplified polymorphic DNA (RAPD) analysis generate similar problem (11). Multi-
locus sequence typing (MLST) defines unambiguous strain types and results can easily be
exchanged between different laboratories (12). MLST has a moderate resolution to delineate

Chapter 6
genetically unique strains and has been proposed for population structure analysis or phy-
logeny studies (13). MLST is a relatively expensive method and therefore not an option for
many clinical laboratories (14).
Repetitive sequence-based PCR (rep-PCR) uses primers that target non-coding repeti-
tive sequences interspersed in bacterial and fungal genomes (15-17). When separated by
electrophoresis, the amplified DNA fragments constitute a genomic fingerprint that can be
employed for subspecies discrimination and strain delineation of bacteria and fungi (18). The
development of a commercially available, automated rep-PCR assay system, the DiversiLab
System (bioMérieux, Marcy l’Etoile, France), offers advances in standardization and reproduc-
ibility over manual fingerprint generating systems (19).

The aim of the current study was to compare the performance (discriminatory power and
reproducibility) and the feasibility (interchange of data, rapidity and cost) of the DiversiLab
system and two worldwide used S. aureus typing methods, PFGE and MLST.

Materials and methods

A representative selection of ninety-three MRSA-isolates of the HARMONY collection was

cultured on trypticase soy (TSA) II agar with 5% sheep blood for 18 hours at 37°C. All strains
were tested in triplicate. From each isolate DNA was extracted using the UltraClean Microbial
DNA Isolation Kit (Mo Bio Laboratories, Solana Beach, CA) according to manufacturer’s in-

111

Rene te Witt bw.indd 111 24-11-11 10:35

 Chapter 6

structions. The presence and concentration of genomic DNA was estimated with NanoDrop®
ND-1000 Spectrophotometer (Isogen, IJsselstein, The Netherlands) and DNA concentration
was adjusted to 25-50 ng/μl for each sample.
All DNA samples were amplified using the DiversiLab Staphylococcus aureus kit for DNA
fingerprinting (bioMérieux) following the manufacturer’s instructions. Briefly, 2 μl of genomic
DNA (final concentration 25-50 ng/μl), 0.5 μl (or 2.5 U) of AmpliTaq® polymerase (Applied
Biosystems, Foster City, CA, USA), 2 μl kit-supplied primer mix and 2.5 μl of 10X PCR buffer
(Applied Biosystems) were added to 18 μl of the kit-supplied rep-PCR master mix (MM1) for
a total of 25 μl per PCR mixture. Thermal cycling parameters were as follows: initial denatur-
ation at 94°C for 2 min; followed by 35 cycles of denaturation at 94°C for 30 sec, annealing at
45°C for 30 sec, extension at 70°C for 90 sec; and a final extension at 70°C for 3 min. Separa-
tion and detection of rep-PCR products was done by micro-fluidic chips of the DiversiLab
System (bioMérieux) and analysis was performed with DiversiLab software version v.r.3.3.40.
The resulting DNA fingerprint patterns were viewed on a personal DiversiLab website as
electropherograms. The reports included a dendrogram with similarity matrix, scatterplot
and a virtual gel image of the fingerprint for each sample.
The DiversiLab software used the Pearson correlation coefficient to analyze and calculate
genetic similarity coefficients among all samples. The unweighted pair-group methods of
averages (UPGMA) was employed to automatically compare the rep-PCR profiles and create
corresponding dendrograms (19). Reports included computer-generated virtual gel images,
scatterplots and selected demographic fields to aid interpretation of the data. Guidelines
have been suggested by the manufacturer for determining the strain-level discrimination
(typing). Cluster analysis, based on peak height and presence or absence of the peaks, was
done by DiversiLab software. Percentage similarity for S. aureus was set at 98% similarity.

Results

The 93 MRSA isolates were differentiated in 7 clusters and 8 unique rep-PCR types by the
DiversiLab system, comprising 15 different fingerprints in total (see Figure 1 and Table 1).
The technical reproducibility of DiversiLab as determined with the average similarity of the
triplicate samples was >99% (95-100%). Data for PFGE and MLST results were retrieved from
a manuscript by Cookson et al. (20). PFGE (SmaI digestion) resolved 28 pulsotypes and MLST
differentiated 16 sequence types in 6 clonal complexes (CCs). Rep-PCR cluster VII included
half of the total collection (n=47), consisting of 7 different STs and 12 different pulsotypes.
Next to this, 2 other large rep-PCR clusters were found. Cluster II with 10 isolates, consisting
of 1 ST and 3 pulsotypes and cluster III, with 14 isolates, composed of 3 STs and 6 pulsotypes.
The discriminatory power calculated as Simpson’s Index of Diversity (DI) was 0.860, 0.877
and 0.905 for DiversiLab, MLST and PFGE, respectively.

112

Rene te Witt bw.indd 112 24-11-11 10:35

 Comparison of the DiversiLab system, PFGE and MLST for typing of MRSA

Chapter 6

Figure 1. Pearson cluster analysis of rep-PCR generatedfingerprintsobtainedfor the 93 MRSA isolates.

PFGE and MLST data are included in the dendrogram. Clonalcomplexes are depicted in colouredboxes. A Similarity Index cutoff value of 95% was
used to define genetic classification.

113

Rene te Witt bw.indd 113 24-11-11 10:35

 Chapter 6

Table 1. MRSA strain typing results obtained with rep-PCR, PFGE and MLST.
rep PCR MLST PFGE
Cluster (n) ST (n) CC type (n)
I (4) 36 (3) 30 27 (3)
II (10) 45 (10) 45 22 (4), 23 (1), 24 (5)
III (14) 5 (3), 222 (1), 228 (10) 5 13 (6), 14 (2), 15 (2), 17 (1), 18 (1), 19 (2)
IV (6) 22 (6) 22 25 (6)
V (2) 5 (1), 80 (1) 5, 80 21 (1), 28 (1)
VI (2) 228 (2) 5 13 (1), 16 (1)
VII (47) 8 (9), 155 (1), 239 (9), 241 (2), 246 (1), 8 1 (6), 2 (1), 3 (24), 4 (2), 5 (1), 6 (1), 7 (5), 8 (3), 9 (1),
247 (22), 254 (3) 10 (1), 11 (1), 12 (1)
unique 1 30 30 26*
unique 2 Nd Nd Nd
unique 3 5 5 13
unique 4 Nd Nd Nd
unique 5 247 8 3
unique 6 5 5 20*
unique 7 Nd Nd Nd
unique 8 36 30 27
MLST: multi-locus sequence typing; PFGE: pulsed-field gel electrophoresis (after smaI restriction); ST: sequence type; CC: clonal complex; nd: not
determined
*: unique PFGE types

Discussion

The primary aim of this study was to assess the performance and feasibility of the DiversiLab
system for discrimination of MRSA isolates in a microbiological diagnostic routine setting and
to compare evaluation criteria with these obtained with PFGE and MLST. The HARMONY iso-
lates were collected to represent circulating nationwide epidemic nosocomial isolates from
1981 to 1998 in 11 European countries. These isolates represent the major clones causing
hospital-acquired MRSA outbreaks in Europe (13).
PFGE (SmaI digestion) resolved 28 pulsotypes and MLST differentiated the collection into
16 STs representing 6 CCs. The HARMONY collection is divided into 3 genetic homogenous
clonal clusters and 2 clonal clusters showing genetic heterogeneity. CC 30, CC 45 and CC 22
are very homogenous clusters, also defined with DiversiLab and PFGE. CC 5 and CC 8 are het-
erogeneous clusters, subdivided into several types with both MLST and PFGE, but not with
rep-PCR. These observations confirmed the result of a previous S. aureus population structure
analysis (21). DiversiLab results are fully concordant with the CC classification as defined with
MLST, but displayed less concordance with STs and pulsotypes. The discriminatory power
of the DiversiLab system was comparable to MLST, while PFGE was the most discriminatory
technique. The acceptable level of discrimination will depend on a number of factors, such as

114

Rene te Witt bw.indd 114 24-11-11 10:35

 Comparison of the DiversiLab system, PFGE and MLST for typing of MRSA

the epidemiological question, but a DI >0.900 might be desirable if the typing results are to
be interpreted with confidence (22).
The reproducibility of the DiversiLab system was excellent. The feasibility of the DiversiLab
system with respect to labour intensity and rapidity was good. The most labour-intensive
step is the extraction, which requires almost half a working day hands-on time, depending
on the number of isolates being tested. Some technical skill is necessary, especially for the
preparation of the micro-fluidic chips used for separation and detection of PCR products.
Occasionally, air bubbles form as DNA is loaded onto the chip. Costs of PFGE, MLST and Diver-
siLab are comparable and are approximately €46, €60 and €48 per sample, respectively. The
rapidity of DiversiLab is better than MLST and PFGE. Total turn-around-time is one working
day with a maximum of 36 samples, which is less than PFGE (3 to 5 days) and MLST (ap-
proximately 3 days). Web-based software is user-friendly, provides standardized comparisons
among the isolates and generates customized reports.
In summary, DiversiLab is a rapid and non labour-intensive technique, but lacks resolution
to differentiate genetically and epidemiologically unique MRSA strains, needed for outbreak
analysis. It may be useful as a library typing system for long-term epidemiological studies.

Chapter 6
Acknowledgements

The DiversiLab kits for S. aureus characterization used in this study were supplied free of
charge by bioMérieux. There is no conflict of interests.

115

Rene te Witt bw.indd 115 24-11-11 10:35

 Chapter 6

References
1. Shorr, A. F. (2007). Epidemiology and economic impact of meticillin-resistant Staphylococcus aureus: review and
analysis of the literature. Pharmacoeconomics 25(9): 751-68.
2. Cosgrove, S. E., et al. (2003). Comparison of mortality associated with methicillin-resistant and methicillin-
susceptible Staphylococcus aureus bacteremia: a meta-analysis. Clin Infect Dis 36(1): 53-9.
3. Taylor, M. D. and Napolitano, L. M. (2004). Methicillin-resistant Staphylococcus aureus infections in vascular surgery:
increasing prevalence. Surg Infect (Larchmt) 5(2): 180-7.
4. Abramson, M. A. and Sexton, D. J. (1999). Nosocomial methicillin-resistant and methicillin-susceptible Staphylo-
coccus aureus primary bacteremia: at what costs? Infect Control Hosp Epidemiol 20(6): 408-11.
5. Kim, T., Oh, P. I. and Simor, A. E. (2001). The economic impact of methicillin-resistant Staphylococcus aureus in
Canadian hospitals. Infect Control Hosp Epidemiol 22(2): 99-104.
6. Rubin, R. J., et al. (1999). The economic impact of Staphylococcus aureus infection in New York City hospitals. Emerg
Infect Dis 5(1): 9-17.
7. Maquelin, K., et al. (2007). Current trends in the epidemiological typing of clinically relevant microbes in Europe. J
Microbiol Methods 69(1): 222-6.
8. Tenover, F. C., et al. (1995). Interpreting chromosomal DNA restriction patterns produced by pulsed-field gel
electrophoresis: criteria for bacterial strain typing. J Clin Microbiol 33(9): 2233-9.
9. van Belkum, A., et al. (1998). Assessment of resolution and intercenter reproducibility of results of genotyping
Staphylococcus aureus by pulsed-field gel electrophoresis of SmaI macrorestriction fragments: a multicenter study.
J Clin Microbiol 36(6): 1653-9.
10. Deplano, A., et al. (2000). Multicenter evaluation of epidemiological typing of methicillin-resistant Staphylococcus
aureus strains by repetitive-element PCR analysis. The European Study Group on Epidemiological Markers of the
ESCMID. J Clin Microbiol 38(10): 3527-33.
11. Bart-Delabesse, E., et al. (2001). Comparison of restriction fragment length polymorphism, microsatellite length
polymorphism, and random amplification of polymorphic DNA analyses for fingerprinting Aspergillus fumigatus
isolates. J Clin Microbiol 39(7): 2683-6.
12. Enright, M. C. and Spratt, B. G. (1999). Multilocus sequence typing. Trends Microbiol 7(12): 482-7.
13. Enright, M. C., et al. (2002). The evolutionary history of methicillin-resistant Staphylococcus aureus (MRSA). Proc
Natl Acad Sci U S A 99(11): 7687-92.
14. Olive, D. M. and Bean, P. (1999). Principles and applications of methods for DNA-based typing of microbial organ-
isms. J Clin Microbiol 37(6): 1661-9.
15. Koeuth, T., Versalovic, J. and Lupski, J. R. (1995). Differential subsequence conservation of interspersed repetitive
Streptococcus pneumoniae BOX elements in diverse bacteria. Genome Res 5(4): 408-18.
16. Stern, M. J., et al. (1984). Repetitive extragenic palindromic sequences: a major component of the bacterial ge-
nome. Cell 37(3): 1015-26.
17. Versalovic, J., Koeuth, T. and Lupski, J. R. (1991). Distribution of repetitive DNA sequences in eubacteria and ap-
plication to fingerprinting of bacterial genomes. Nucleic Acids Res 19(24): 6823-31.
18. Versalovic, J. and Lupski, J. R. (2002). Molecular detection and genotyping of pathogens: more accurate and rapid
answers. Trends Microbiol 10(10 Suppl): S15-21.
19. Healy, M., et al. (2005). Microbial DNA typing by automated repetitive-sequence-based PCR. J Clin Microbiol 43(1):
199-207.

116

Rene te Witt bw.indd 116 24-11-11 10:35

 Comparison of the DiversiLab system, PFGE and MLST for typing of MRSA

20. Cookson, B. D., et al. (2007). Evaluation of molecular typing methods in characterizing a European collection of
epidemic methicillin-resistant Staphylococcus aureus strains: the HARMONY collection. J Clin Microbiol 45(6): 1830-7.
21. Melles, D. C., et al. (2004). Natural population dynamics and expansion of pathogenic clones of Staphylococcus
aureus. J Clin Invest 114(12): 1732-40.
22. Hunter, P. R. and Gaston, M. A. (1988). Numerical index of the discriminatory ability of typing systems: an applica-
tion of Simpson’s index of diversity. J Clin Microbiol 26(11): 2465-6.

Chapter 6

117

Rene te Witt bw.indd 117 24-11-11 10:35

Rene te Witt bw.indd 118 24-11-11 10:35
 Chapter 7
Good performance of the SpectraCellRA
system for typing of methicillin-resistant
Staphylococcus aureus (MRSA)

René te Witt
Norbert Vaessen
Sybren Lekkerkerk
Damian C. Melles
Liesbeth van der Zwaan
Willemien Zandijk
Juliëtte A. Severin
Margreet C. Vos

Manuscript in preparation

Rene te Witt bw.indd 119 24-11-11 10:35

 Chapter 7

Abstract

After methicillin-resistant Staphylococcus aureus (MRSA) detection, typing remains necessary

in order to assess whether transmission of MRSA occurred and to what extend infection
prevention measures need to be taken. In this study, the SpectraCellRA system (SCRA; River
Diagnostics, Rotterdam, The Netherlands) was evaluated for typing of MRSA strains. SCRA
determines clonal relationships between bacterial isolates by Raman spectroscopy.
We have analyzed a well-documented MRSA collection of 113 MRSA strains, collected from
54 patients and their household members. The epidemiological relationship between the
MRSA strains within one household was used as the “gold standard” as the a priori chance of
being MRSA carrier is very low in The Netherlands. Discrepant findings between the result
of SCRA and the epidemiological data were analysed using pulsed-field gel electrophoresis
(PFGE) of SmaI macro restriction fragments of genomic DNA.
Results of SCRA analysis corresponded with epidemiological data for 108 of 113 strains, a
concordance of SCRA and the gold standard of 95.6%. However, when results were analyzed
at the household level, results of SCRA were correct for 49 out of 54 households; a concor-
dance of 90.7%. This indicates that the discriminatory power of SCRA may be too high for
adequate outbreak analysis. Reproducibility was found to be 100%.
We conclude that the SpectraCellRA system is a fast, easy to use and highly reproducible
typing platform for outbreak analysis that can compete with the currently used typing tech-
niques.

120

Rene te Witt bw.indd 120 24-11-11 10:35

 Good performance of the SpectraCellRA system for typing of methicillin-resistant Staphylococcus aureus (MRSA)

Introduction

Methicillin-resistant Staphylococcus aureus (MRSA) continues to be a major problem in hos-

pitals. Following laboratory detection of MRSA, typing remains necessary in order to assess
whether transmission of MRSA occurred and what kind of infection prevention measures are
needed. An ideal typing system for epidemiological surveillance of MRSA should be user-
friendly, fast and reliable, combined with high-throughput capacity (1). However, next to
these the two most important requirements for such a typing system are high discriminatory
power and reproducibility (1). Currently, different typing methods are in use in the diagnostic
laboratory, with pulsed-field gel electrophoresis (PFGE) of SmaI macro restriction analysis
of genomic DNA still being considered as the gold standard for typing of bacteria because
of its high discriminatory power (2-3). However, PFGE is technically demanding with limited
portability due to low inter-center reproducibility, with a long time-to-result (4). Therefore,
alternative typing techniques such as staphylococcus aureus protein A typing (spa typing),
multi-locus sequence typing (MLST) or multi-locus variable number of tandem repeats analy-
sis (MLVA) are used for outbreak analysis.
Raman spectroscopy (SpectraCellRA [SCRA], River Diagnostics, Rotterdam, The Nether-
lands) has been described as an adequate tool for typing of bacteria (5). This vibrational spec-
troscopy-based technique does not require any labels or dyes and is fast. Its high throughput
and ease of use enables SCRA as a valuable tool for outbreak analysis in routine diagnostic

Chapter 7
laboratories.
Following detection of MRSA in many, if not all, Dutch hospitals the patient (primary case)
is isolated to prevent transmission. No eradication therapy is offered during the in-hospital
period due to lower success rates during this period. After discharge, eradication therapy is
started as soon as the patient has no wounds, no antibiotics or drains or when there is no
urgent need for treatment. In the Erasmus MC (Rotterdam, The Netherlands), all household
members of the patient are tested for MRSA colonization before therapy. All positive house-
hold members (secondary cases) are then treated simultaneously with the index patient.
Since the prevalence of MRSA is still low in The Netherlands (6-7), there is a very low a priori
chance for these household members of acquiring MRSA from another source. Furthermore,
Mollema et al showed in 2010 that all isolates from MRSA positive household members had
the same PFGE type as the isolate from their index person (8). Therefore, the presence of MRSA
carriers within one household is considered to be a consequence of household transmission
The primary aim of this study was to assess whether the SpectraCellRA system indeed can
determine clonal relationships between MRSA strains isolated from household members
infected or colonized with MRSA and determined the performance of this typing system.
PFGE was used as a second typing method for verification purposes in case of discrepancies
between SCRA typing and the gold standard of epidemiological data and initial PFGE results.

121

Rene te Witt bw.indd 121 24-11-11 10:35

 Chapter 7

Materials and methods

Bacterial strains

A total of 113 MRSA strains from a well-documented set of 54 households were included
in the analysis (8). Strains were collected in 2005, 2006 and 2007 from primary cases (index
patient, n=54) and corresponding secondary cases (household members of index, n=59).
Households consisted either of 4 members (n=3), 3 members (n=14), 2 members (n=22) or 1
member (n=15).
Cultures for MRSA were performed at the diagnostic laboratory of the Department of Medi-
cal Microbiology and Infectious Diseases (Erasmus MC). Results were confirmed by using Ac-
cuprobe (Gen-Probe Incorporated, San Diego, USA) and the MRSA-screen latex agglutination
test (Denka Seiken Co., Tokyo, Japan). Furthermore, all isolated MRSA strains were sent to the
National Institute for Public Health and the Environment (RIVM; Bilthoven, The Netherlands)
for confirmation and PFGE-typing.
For all MRSA isolates, epidemiological data including household information, date of
admission and sampling date and PFGE results were available. We defined all secondary
cases as having acquired their strain from the index case of their household. Epidemiologi-
cal household relations and PFGE results were defined as the gold standard. Verification of
discrepant results between SCRA and the gold standard (epidemiological data + initial PFGE)
was performed by new PFGE of SmaI macro restriction analysis of genomic DNA, as previously
described (2). Furthermore, we analysed discrepant household members for possibilities of
“exo-household” (from a source other than the household) acquisition of MRSA by looking
into the epidemiological data.

SpectraCellRA analysis

Clonal relationships among the MRSA isolates were tested by SCRA. Cultures, sample prepa-
ration and SCRA measurements were performed according to the Operators Manual (version
1.7) (5) .
Briefly, isolates were inoculated on blood agar (BD Diagnostics). After incubation for 18-20
h, isolates were sub-cultured for 20 h on Trypticase Soy Agar (TSA agar) plates. Biomass was
taken from this culture to fill a 1 μl loop. This biomass was suspended in 5 μl of distilled water
and 3 μl of this suspension was transferred onto a MicroSlide sample carrier.

Similarities between spectra

Pearson correlation coefficients (R2 values) were calculated between replicate measurements
of the same isolate and between spectra of different isolates. The R2 values between repli-

122

Rene te Witt bw.indd 122 24-11-11 10:35

 Good performance of the SpectraCellRA system for typing of methicillin-resistant Staphylococcus aureus (MRSA)

cates account for any signal variance due to differences in culturing, sample preparation, or
actual Raman measurements. To be able to discriminate 2 isolates, the R2 value between the
isolates had to be lower than the R2 values between replicate measurements of these isolates.

Hierarchical cluster analysis (HCA)

SCRA type analysis of sets of spectra was performed using the pair wise similarities as a
distance matrix in combination with Ward’s cluster algorithm. This results in a dendrogram
in which each node represents the lowest correlation coefficient (or similarity) between all
isolates combined in the cluster defined by this node.

Reproducibility

To establish reproducibility, 3 reference isolates were processed on 5 different, independent

days during the study by 2 different technicians. The reference isolates were obtained from
a reference MRSA strain collection that was used to study interlaboratory reproducibility of
PFGE (4). Reference isolates 811 and 814 are genetically related isolates, reference isolate 806
was chosen as a unique isolate. Furthermore, 26 MRSA isolates from the study were analyzed
in duplicate on 2 different days by 2 different technicians. Reference isolates and duplicates
were performed as full biological replicates; isolates were processed from the freezer on dif-

Chapter 7
ferent days of the study.

Workflow and costs

Estimation per test was made on the basis of total turn-around time, hands-on-time and costs
(in Euro’s) incurred by SCRA analysis. These parameters were then compared with those of PFGE.

Results

Similarities between spectra

The distribution of the R2 values between replicates (red bars) and between isolates (blue
bars) is given In Figure 1. When the R2 value for two isolates is below 0.9995 (region 1), these
isolates are non-related in the SCRA analysis. When two isolates have a R2 value above 0.9996
(region 3), these isolates are indistinguishable in the SCRA analysis. R2 values in between
these values (region 2) indicate a possible relatedness between isolates. A similarity cut-off is
chosen in region 2, based on the presumption that 95% of all replicates must have a R2 value
above this cut-off. This similarity cut-off in this study was set to 0.9996.

123

Rene te Witt bw.indd 123 24-11-11 10:35

 Chapter 7

Region3
Region1

Region2

Figure 1. Histogram of R2 values found for our study.

R2 values between replicates are indicated by red bars, R2 values between isolates are indicated by blue bars. Region 1 indicates non-related
isolates in the SCRA analysis. When isolates are in region 3, these isolates are indistinguishable in the SCRA analysis. Isolates in between these
regions (in region 2), are possibly related. The cut-off is chosen in region 2, based on the presumption that 95% of all replicates must have a R2
value above this cut-off. This similarity cut-off is set to 0.9996.

Hierarchical cluster analysis (HCA)

Among the 113 MRSA isolates, 59 SCRA types could be distinguished (Table 1). Based on the
epidemiological data, results of SCRA analysis were correct for 108 of 113 strains. This results
in a concordance of 95.6%.
When SCRA types were analyzed at the level of household clusters, results of SCRA analysis
were concordant for 49 out of 54 households (90.7%).
All MRSA isolates within household cluster 14 (n=3) were non-typeable using PFGE. How-
ever, SCRA resulted in 1 SCRA type.

Discrepancy analysis

Five discrepant results were observed between SCRA analysis and epidemiological data.
PFGE-typing results as reported by the RIVM of these discrepancies were fully concordant
with epidemiological data. PFGE was performed on all the isolates from the discrepancy-
households 7, 19, 20, 47 and 52. Household clusters 7, 19, 20 and 47 consisted of 1 primary
case and 1 secondary case each. PFGE showed 2 identical isolates within each of these clus-

124

Rene te Witt bw.indd 124 24-11-11 10:35

 Good performance of the SpectraCellRA system for typing of methicillin-resistant Staphylococcus aureus (MRSA)

Table 1. Results of SCRA analysis.

Household Household PFGE cluster as
SCRA type
cluster members (n) reported by the RIVM
1 3 2 155
2 2 57 55
3 4 11 18
4 3 10 18
5 3 21 27
6 2 69 37
7 2 48, 67 68
8 2 41 55
9 3 28 109a
10 2 43 587
11 3 9 23, 28
12 1 37 28
13 2 62 293
14 3 68 NT
15 1 34 263
16 1 3 15
17 3 16 23
18 4 39 22, 68, 665
19 2 47, 58 55
20 2 56, 59 55
21 2 52 50a, 675
22 1 19 22

Chapter 7
23 1 60 23
24 2 22 24
25 1 51 25
26 3 50 26
27 2 65 27
29 2 35 50a
30 1 73 30
31 3 25 31
32 3 6 32
33 1 54 381
34 1 76 34
35 2 30 35
36 2 29 199
37 2 5 37
38 2 71 38
39 1 31 39
40 3 75 40
41 1 32 41
42 2 77 42
43 3 82 43
44 1 27 44
45 4 14 27, 28
46 2 12 46

125

Rene te Witt bw.indd 125 24-11-11 10:35

 Chapter 7

Table 1. Contineud
Household Household PFGE cluster as
SCRA type
cluster members (n) reported by the RIVM
47 2 4, 18 47
49 2 36 49
50 1 17 50
52 3 40, 45 52
54 2 61 54
55 2 55 55
57 3 49 15
58 1 79 58
59 1 80 59
Reference 806 5 29
Reference 811 5 51
Reference 814 5 50
NT: Non-typeable with PFGE.

ters, whereas SCRA analysis resulted in 8 unique isolates. Household cluster 52 consisted of
3 members. For this household, PFGE showed 3 identical isolates, whereas SCRA resulted in
2 identical isolates and 1 unique isolate. Overall concordance of SCRA analysis was 95.6%
on isolate level and 90.7% on household cluster level. Analysis of the discrepant household
members for exo-household acquisition of MRSA did not result in any possible transmission
of another known source or contact with a person with known risk factors as described in the
Dutch national guidelines (WIP).

Reproducibility

Mean correlation coefficients and standard deviation (SD) were calculated for the 5 indepen-
dent measurements of each reference isolate. For sample 806, 811 and 814 mean correlation
coefficients (SD) were 0.9998 (±0.0), 0.9998 (±0.0) and 0.9999 (±0.0). Furthermore, all dupli-
cate measurements of the 26 isolates resulted in identical results. Therefore, reproducibility
was 100%.

Workflow

Total hands-on-time for 24 samples was around 3 h for SCRA and 7 h for PFGE. Total turn-
around-time was 36-48 h for SCRA and 96 h for PFGE, with a maximum of 72 samples for SCRA
and 50 samples for PFGE.
SCRA analysis requires a subculture on TSA agar on day 0 (15 min. for inoculation, 15
min. for documentation, 18-20 h incubation). On day 1, secondary subcultures on TSA were
performed (15 min. for inoculation, 20 h incubation). Then, isolates were processed and

126

Rene te Witt bw.indd 126 24-11-11 10:35

 Good performance of the SpectraCellRA system for typing of methicillin-resistant Staphylococcus aureus (MRSA)

measured on day 2 followed by analysis of the results (30 min. for preparation of the slides, 60
min. for measurements and 30 min. for analysis).
Costs of SCRA and PFGE are comparable and are approximately €50 per sample, including
personnel expenses and consumables.

Discussion

Results of SCRA analysis at the isolate level were concordant with the gold standard of epi-
demiology for 95.6% (108/113). However, when our results were analyzed at the household
cluster level, results were concordant for 90.7% (49/54). In other words, when using SCRA
in practice, for 9% of the small clusters SCRA would have given discrepant results. Analysis
of discrepant results using PFGE showed equal results as epidemiological data. When the
discrepant households were analyzed for possibilities of exo-household acquisition of MRSA,
no possible transmission of another known source or contact with a person with known risk
factors were found. This indicates that the discriminatory power of the SCRA system might be
too high for adequate outbreak analysis of small clusters.
During transmission and acquisition of MRSA, micro-evolution may take place (9-10).
PFGE may not be able to detect this micro-heterogeneity, where Raman spectroscopy does.
However, since a household can be considered as a close community, we can assume that

Chapter 7
transmission and detection have occurred in a short time period. Therefore, it is very likely
that the discrepancy between SCRA and the gold standard is not due to micro-evolution of
the organism.
Multiple independent measurements of 3 reference isolates and the duplicate measure-
ments of 26 MRSA strains resulted in a reproducibility of 100%. This indicates that SCRA is
stable over a longer period of time, which has been published before (5). Typeability (as
indicated by household cluster 14) was 100% for SCRA and <100% for PFGE.Our findings are
similar to those found in another study, where the observed concordance between SCRA
and the gold standard of epidemiological data and PFGE was 97% (11). Reproducibility in
our study was better (100% vs. 95%). In both studies isolates were tested as full biological
replicates at different points in the study.
PFGE is still being considered as the gold standard for typing of bacteria because of its high
discriminatory power (2-3). However, PFGE is technically demanding with limited portabil-
ity due to low inter-center reproducibility (4). Therefore, alternative typing techniques are
currently used for outbreak analysis. However, although spa typing, MLST and MLVA have a
good portability due to standard nomenclature, the discriminatory power of these methods
is too limited for adequate outbreak analysis (12-13). Furthermore, these methods cannot
be used in routine diagnostic laboratories easily because of the required technical expertise.
A commercially available automated rep-PCR system, the DiversiLab system (bioMérieux,

127

Rene te Witt bw.indd 127 24-11-11 10:35

 Chapter 7

Marcy l’Etiole, France), offers advances in ease-of-use and reproducibility of the procedure
over manual typing systems (14). However, although three independent studies concluded
that the Diversilab system is a rapid and reproducible technique, it clearly lacks resolution for
typing of gram-positive bacteria such as MRSA (15-17).
The feasibility of the SCRA system with respect to hands-on-time (~ 3 h) and time-to-result
(36-48 h) was good and better than PFGE. Time-to-result may be improved by applying 1
subculture instead of 2 subcultures. In this way, results will be available the next day, what is
enormous valuable for prevention control.
Mistyping of strains by using the SCRA system for outbreak analysis is a risk. Isolates
epidemiologically assigned to one cluster could now be considered as unique isolates and
as a single event each, which has serious consequences for infection prevention measures.
Prevention measures to a single event, or 2 separate single events, are different than the
response on an outbreak (2 or more MRSA cases belonging to one cluster). For 90-95% of
cases a considerable amount of time for implementing prevention measures is gained. For
the remaining 5-10%, further analysis will be necessary.
Many studies on typing of MRSA focus on a small set of MRSA isolates and the relevant
epidemiological data of the isolates are often (partially) unknown. We had the unique op-
portunity to use the epidemiological relationships within households as the gold standard,
together with PFGE.
Although the high discriminatory power of SCRA may lead to discrepancies, and accompa-
nying consequences, in 5-10% of epidemiologically related MRSA clusters, we conclude that
the SpectraCellRA system is a highly reproducible, easy-to-use and fast typing platform that
can compete with the currently used typing techniques.

Acknowledgements

Kees Maquelin is acknowledged for technical support.

The Unit Infection Prevention of the department of Medical Microbiology and Infectious
Diseases is acknowledged for data collection.

128

Rene te Witt bw.indd 128 24-11-11 10:35

 Good performance of the SpectraCellRA system for typing of methicillin-resistant Staphylococcus aureus (MRSA)

References
1. van Belkum, A., et al. (2007). Guidelines for the validation and application of typing methods for use in bacterial
epidemiology. Clin Microbiol Infect 13 Suppl 3: 1-46.
2. Ichiyama, S., et al. (1991). Genomic DNA fingerprinting by pulsed-field gel electrophoresis as an epidemiological
marker for study of nosocomial infections caused by methicillin-resistant Staphylococcus aureus. J Clin Microbiol
29(12): 2690-5.
3. Goering, R. V. and Duensing, T. D. (1990). Rapid field inversion gel electrophoresis in combination with an rRNA
gene probe in the epidemiological evaluation of staphylococci. J Clin Microbiol 28(3): 426-9.
4. van Belkum, A., et al. (1998). Assessment of resolution and intercenter reproducibility of results of genotyping
Staphylococcus aureus by pulsed-field gel electrophoresis of SmaI macrorestriction fragments: a multicenter study.
J Clin Microbiol 36(6): 1653-9.
5. Willemse-Erix, D. F., et al. (2009). Optical fingerprinting in bacterial epidemiology: Raman spectroscopy as a real-
time typing method. J Clin Microbiol 47(3): 652-9.
6. Bode, L. G., et al. (2011). Sustained low prevalence of meticillin-resistant Staphylococcus aureus upon admission
to hospital in The Netherlands. J Hosp Infect 79(3): 198-201.
7. Wertheim, H. F., et al. (2004). Low prevalence of methicillin-resistant Staphylococcus aureus (MRSA) at hospital
admission in the Netherlands: the value of search and destroy and restrictive antibiotic use. J Hosp Infect 56(4):
321-5.
8. Mollema, F. P., et al. (2010). Transmission of methicillin-resistant Staphylococcus aureus to household contacts. J
Clin Microbiol 48(1): 202-7.
9. Feil, E. J. (2004). Small change: keeping pace with microevolution. Nat Rev Microbiol 2(6): 483-95.

Chapter 7
10. Aziz, R. K. and Nizet, V. (2010). Pathogen microevolution in high resolution. Sci Transl Med 2(16): 16ps4.
11. Wulf, M. W., et al. (2011). The use of Raman spectroscopy in the epidemiology of methicillin-resistant Staphylococ-
cus aureus of human- and animal-related clonal lineages. Clin Microbiol Infect.
12. Struelens, M. J., et al. (2009). Laboratory tools and strategies for methicillin-resistant Staphylococcus aureus
screening, surveillance and typing: state of the art and unmet needs. Clin Microbiol Infect 15(2): 112-9.
13. Malachowa, N., et al. (2005). Comparison of multiple-locus variable-number tandem-repeat analysis with pulsed-
field gel electrophoresis, spa typing, and multilocus sequence typing for clonal characterization of Staphylococ-
cus aureus isolates. J Clin Microbiol 43(7): 3095-100.
14. Healy, M., et al. (2005). Microbial DNA typing by automated repetitive-sequence-based PCR. J Clin Microbiol 43(1):
199-207.
15. Babouee, B., et al. (2011). Comparison of the rep-PCR system DiversiLab with spa typing and pulsed-field gel
electrophoresis for the clonal characterization of methicillin-resistant Staphylococcus aureus. J Clin Microbiol.
16. Te Witt, R., Kanhai, V. and van Leeuwen, W. B. (2009). Comparison of the DiversiLab system, Pulsed-Field Gel Electro-
phoresis and Multi-Locus Sequence Typing for the characterization of epidemic reference MRSA strains. J Microbiol
Methods 77(1): 130-3.
17. Tenover, F. C., et al. (2009). Comparison of typing results obtained for methicillin-resistant Staphylococcus aureus
isolates with the DiversiLab system and pulsed-field gel electrophoresis. J Clin Microbiol 47(8): 2452-7.

129

Rene te Witt bw.indd 129 24-11-11 10:35

Rene te Witt bw.indd 130 24-11-11 10:35
 Quality control

Rene te Witt bw.indd 131 24-11-11 10:35

Rene te Witt bw.indd 132 24-11-11 10:35
 Chapter 8
External Quality Assessment
of molecular diagnostics
and genotyping of methicillin-resistant
Staphylococcus aureus

René te Witt
Alex van Belkum
William G. MacKay
Paul S. Wallace
Willem B. van Leeuwen

European Journal of Clinical Microbiology and Infectious Diseases, 2010; 29: 295-300

Rene te Witt bw.indd 133 24-11-11 10:35

 Chapter 8

Abstract

Two multicentre External Quality Assessments (EQA) for molecular detection and genotyping
of methicillin-resistant Staphylococcus aureus (MRSA) were arranged.
Firstly, eleven samples containing various amounts of inactivated MRSA strains, methicillin-
susceptible S. aureus (MSSA), methicillin-resistant coagulase-negative staphylococci (MR-
CoNS) or Escherichia coli, were distributed to 82 laboratories. Samples containing 102 or 103
MRSA cells were correctly scored in only 16% and 46% of the datasets returned, respectively.
Two of the used MSSA strains contained a SCCmec cassette lacking the mecA gene. There was
a marked difference in the percentage of correct results for these two MSSA strains (37% and
39%), compared to the MSSA strain lacking the SCCmec cassette (88%). The MRCoNS sample
was correctly scored negative by 40 out of 45 (89%) commercial tests used and by 23 out of
33 (70%) in-house assays. The E. coli sample was incorrectly reported as positive in 9 (11%) of
the datasets.
Secondly, a panel for MRSA genotyping, consisting of 10 samples (2 identical, 3 genetically
related and 5 unique strains) was distributed to 19 laboratories. Seventy-three percent of the
datasets recorded all samples correctly. Most pulsed-field gel electrophoresis (PFGE) proto-
cols proved to be suboptimal, resulting in inferior resolution in the higher or lower fragment
regions.
Performance of molecular diagnostics for MRSA showed no significant changes since
our first EQA in 2006. The first molecular typing results are encouraging. Both assessments
indicate that programme expansion is required and that major diagnostic and typing perfor-
mance discrepancies still continue to exist between diagnostic microbiology laboratories.

134

Rene te Witt bw.indd 134 24-11-11 10:35

 External Quality Assessment of molecular diagnostics and genotyping of methicillin-resistant Staphylococcus aureus

Introduction

Adequate infection control of methicillin-resistant Staphylococcus aureus (MRSA) strongly

depends on the speed and quality of (molecular) identification and characterisation strate-
gies used by the clinical microbiological laboratory (1-2).
Over the past 4-5 decades, cultivation assays have been primarily used for detection and
subsequent identification of MRSA. However, cultivation requires prolonged incubation pe-
riods, and, in general, clinically relevant methicillin resistance still needs to be confirmed by
the detection of the mecA gene or its product. Nucleic acid amplification techniques (NAATs)
offer benefits over traditional culture-based assays, in particular a reduced time to identifica-
tion and an improved specificity and sensitivity. Over the past decade, a range of commercial
and in-house developed NAATs has been introduced. Sensitivity and specificity of these
assays may be compromised, as a result of inhibition or false-positivity due to the presence
of methicillin-resistant Coagulase Negative Staphylococci (MRCoNS) or variability within the
mec-resistance cassette. This may lead to false-negative results (new Staphylococcal Cassette
Chromosome mec [SCCmec] variants) or false-positive results (deletion of the mecA gene)
(3-6). After MRSA detection, genetic typing may be necessary to assess whether local cross
infection occurs and whether preventive measurements are mandatory.
Currently, many different genotyping methods are in use in the diagnostic laboratory, but
pulsed-field gel electrophoresis (PFGE) of SmaI digested genomic DNA still remains the most
frequently used method (7). Only when outbreaks are properly defined, adequate infection
control measurements can be implemented.
The current multicentre External Quality Assessment (EQA) study determined the perfor-

Chapter 8
mance of molecular assays to detect MRSA and genotyping techniques to differentiate MRSA
strains. The studies were coordinated by Quality Control for Molecular Diagnostics (QCMD)
in Glasgow, Scotland.

Materials and methods

EQA for molecular MRSA detection and identification

In August 2009, the EQA MRSA panel for MRSA detection and identification was distrib-
uted to 80 participating laboratories in 15 countries, along with detailed sample processing
instructions. Participants were given 6 weeks to examine the samples and to report their
results to QCMD by using an online data collection system. The QCMD MRSA panel consisted
of six samples containing 106, 105 (n=2), 104, 103 and 102 CFU/ml MRSA bacterial cells, one
methicillin-susceptible S. aureus (MSSA) sample, one sample containing MRCoNS, two sam-
ples containing MSSA harbouring a SCCmec cassette lacking the mecA gene and one sample

135

Rene te Witt bw.indd 135 24-11-11 10:35

 Chapter 8

containing Escherichia coli (Table 1). The production laboratory quantified the contents of
the samples on the basis of colony counting, optical density measurements and real-time
molecular amplification results. All bacterial samples were heat-inactivated for 10 minutes
at 100°C.

Table 1. Composition of the QCMD 2009 panel for MRSA detection and identification.
Target value
Sample Sample content CFU/ml Sample status
MRSA09-01 MRCoNS c 1x10E7 Negative
MRSA09-02 MSSA d 5x10E6 Negative
MRSA09-03 MSSA b 5x10E6 Negative
MRSA09-04 MRSA a 5x10E5 Positive
MRSA09-05 MRSA a 5x10E6 Positive
MRSA09-06 MRSA a 2.5x10E2 Positive
MRSA09-07 MRSA a 2.5x10E4 Positive
MRSA09-08 MRSA a 5x10E5 Positive
MRSA09-09 MSSA e 1x10E7 Negative
MRSA09-10 MRSA a 2.5x10E3 Positive
MRSA09-11 E. coli f 5x10E6 Negative
MRSA a: Methicillin-resistant Staphylococcus aureus N315.
MSSA b: Methicillin-sensitive Staphyloccocus aureus ATCC 29213.
MRCoNS c: Methicillin-resistant Coagulase-negative Staphylocci 634.
MSSA d: Methicillin-sensitive Staphyloccoous aureus (mecA negative 92-1866 [SCCmec III]).
MSSA e: Methicillin-sensitive Staphyloccoous aureus (mecA negative 93-2886 [SCCmec I]).
E. coli f: Eschericia coli ATCC 35218.

EQA for MRSA genotyping

The EQA panel for MRSA genotyping was distributed to 19 participants in 8 countries in Au-
gust 2009. The panel consisted of 10 samples of viable MRSA strains in Müller Hinton broth.
Genetic relatedness of the MRSA panel was originally determined with PFGE (8). The current
panel consisted of two identical strains, three genetically related strains and 5 unique strains
(Table 2). Genotype and subtype were reported by the production laboratory. A different
letter signified the detection of a different genotype, whereas a different number signified
the detection of a different subtype. All data were reported in relation to the reference strain
in panel sample MRSATP09-01.
The QCMD Neutral Office analysed the data, which was anonymously released to all partici-
pants in a detailed EQA final report.

136

Rene te Witt bw.indd 136 24-11-11 10:35

 External Quality Assessment of molecular diagnostics and genotyping of methicillin-resistant Staphylococcus aureus

Table 2. Genotyping results per panel member and technology type of the QCMD 2009 panel for MRSA genotyping.
Total datasets PFGE a AFLP b SPA c
Expected n=15 n=11 n=2 n=2
Sample genotype n % n % n % n %
MRSATP09-01 A 15 100 11 100 2 100 2 100
MRSATP09-02 B 11 73.3 8 72.7 1 50.0 2 100
MRSATP09-03 C 14 93.3 11 100 1 50.0 2 100
MRSATP09-04 D 14 93.3 11 100 1 50.0 2 100
MRSATP09-05 A 13 86.7 10 90.9 1 50.0 2 100
MRSATP09-06 E 13 86.7 11 100 2 100 0 0.0
MRSATP09-07 F 12 80.0 9 81.8 1 50.0 2 100
MRSATP09-08 G 14 93.3 11 100 1 50.0 2 100
MRSATP09-09 G1 11 73.3 11 100 0 0.0 0 0.0
MRSATP09-10 G2 11 73.3 11 100 0 0.0 0 0.0
PFGE a: bioRadGenePath Group 1 reagent kit (n=2), in-house PFGE (n=8) and Double Locus Sequence Typing (n=1).
AFLP b: In-house AFLP (n=2).
SPA c: In-house spa typing (n=2).

Results

EQA for molecular MRSA detection and identification

Out of the 80 participants, 68 (85%) responded. Twelve participants did not return results.
Five of these withdrew officially, indicating ‘assay not offered’ (n=2), ‘internal issues’ (n=1) and
‘other’ (n=2) as the reason for withdrawal.

Chapter 8
The following commercial amplification assays were used for MRSA detection: BAG
Healthcare hyplex Staphyloresist (n=1) (BAG Healthcare, Lich, Germany), BD Diagnostic
GeneOhm MRSA Assay (n=13) (BD Diagnostics – GeneOhm, San Diego, California), BD Di-
agnostics GeneOhm Staph SR Assay (n=5) (BD Diagnostics – GeneOhm), Cepheid IDI MRSA
(n=1) (Cepheid, Sunnyvale, California), Cepheid Xpert MRSA Test (n=11) (Cepheid), Cepheid
Xpert MRSA/SA Test (n=2) (Cepheid), Roche LightCycler MRSA Advanced Test (n=4) (Roche
Diagnostics, Basel, Switzerland), TIB MOLBIOL LightMix Kit MRSA (n=2) (TIB MolBiol, Berlin,
Germany) and Hain Lifescience GenoQuick MRSA (n=2) (Hain Lifescience, Nehren, Germany).
This diversity overlaps with the spectrum of currently available commercial tests. All results
are summarised in Table 3.
Results for the panel samples with 106 MRSA cells (MRSA09-05), 105 MRSA cells (MRSA09-04
and MRSA09-08) and 104 MRSA cells (MRSA09-07) were reported correctly in 100%, 100%,
99% and 98% of the datasets, respectively. The samples containing lower amounts, MRSA09-
10 (103 CFU/ml) and MRSA09-06 (102 CFU/ml), were reported correctly in only 46% and 16%
of the datasets, respectively. No statistically significant differences in sensitivity or specificity
could be seen between the different tests or between commercial and in-house testing. MR-

137

Rene te Witt bw.indd 137 24-11-11 10:35

 Chapter 8

CoNS sample MRSA09-01 was correctly reported as MRSA negative by 87% (40 out of 46) of
the commercial PCR tests and in 70% (23 out of 33) of the in-house PCR assays. MSSA sample
MRSA09-03 was correctly reported as MRSA negative by 89% (41 out of 46) of commercial PCR
tests and in 85% (28 out of 33) of datasets generated by in-house PCR assays, respectively.
The MSSA samples containing the SCCmec cassette but lacking the mecA gene (MRSA09-02
and MRSA09-09) were both incorrectly reported as positive by commercial PCR tests in 87%
(40 out of 46). For in-house assays these samples were reported incorrectly in 24% (8 out of
33) and 30% (10 out of 33), respectively. These percentages of incorrect results underscore
the need for improved specificity of these MRSA tests and therefore positive results should
always be confirmed by a culture method or a second molecular test. For laboratories with
high false positivity rates or in regions with low prevalence of MRSA, confirmation is essential
(9). For the E. coli sample, commercial PCR results were reported correctly in 87% (39 out of
45), whereas in-house PCR tests recorded correct results in 97% (29 out of 30).

Table 3. Number of correct qualitative results per panel member and technology type on the QCMD 2009 panel for MRSA detection and typing.
PCR NASBA e Other f
Sample Sample content Total Conventional Real-time
datasets Commercial a In-house b Commercial c In-house d
n=82 n=1 n=3 n=45 n=30 n=1 n=2
n % n % n % n % n % n % n %
MRSA09-05 MRSA 82 100 1 100 3 100 45 100 30 100 1 100 2 100
MRSA09-04 MRSA 82 100 1 100 3 100 45 100 30 100 1 100 2 100
MRSA09-08 MRSA 81 98.8 1 100 3 100 44 97.8 30 100 1 100 2 100
MRSA09-07 MRSA 80 97.6 1 100 3 100 45 100 28 93.3 1 100 2 100
MRSA09-10 MRSA 38 46.3 0 0.0 2 66.7 22 48.9 13 43.3 0 0.0 1 50.0
MRSA09-06 MSSA 13 15.9 0 0.0 0 0.0 6 13.3 6 20.0 0 0.0 1 50.0
MRSA09-03 MSSA 72 87.8 1 100 3 100 40 88.9 25 83.3 1 100 2 100
MRSA09-09 MSSA 30 36.6 1 100 3 100 5 11.1 20 66.7 1 100 0 0.0
MRSA09-02 MSSA 32 39.0 1 100 3 100 5 11.1 22 73.3 1 100 0 0.0
MRSA09-01 MRCoNS 66 80.5 0 0.0 2 66.7 40 88.9 21 70.0 1 100 2 100
MRSA09-11 E. coli 73 89.0 1 100 1 33.3 39 86.7 29 96.7 1 100 2 100
Commercial a: BAG Healthcare hyplexStaphyloresist (n=1).
In-house b: Details not presented.
Commercial c: BD Diagnostics details not provided (n=7),BD Diagnostics GeneOhm MRSA Assay (n=13), BD Diagnostics GeneOhm Staph SR
Assay (n=5), Cepheid IDI MRSA (n=1), Cepheid Xpert MRSA Test (n=11), Cepheid Xpert MRSA/SA Test (n=2), Roche LightCycler MRSA Advanced
Test (n=4), TIB MolBiolLightMix Kit MRSA (n=2).
In-house d: Details not presented.
NASBA e: Details not presented.
Other f: HainLifescienceGenoquick MRSA (n=2).

EQA for MRSA genotyping

Out of the 19 potential participants, 14 (74%) responded. Four of the non-responders with-
drew officially indicating ‘panel used for research’ (n=1) and ‘assay not offered’ (n=3). The

138

Rene te Witt bw.indd 138 24-11-11 10:35

 External Quality Assessment of molecular diagnostics and genotyping of methicillin-resistant Staphylococcus aureus

majority of datasets were generated by PFGE (n=11), with the remainder generated by AFLP
(n=2) and spa typing (n=2). Only eight participants (73%) scored all samples correct: all with
PFGE (Table 2).

Discussion

To maintain high-quality clinical care, quality control of molecular diagnostics is very impor-
tant. The primary aim of our EQA programme was to assess the proficiency of laboratories
in the molecular detection and characterisation of MRSA strains. Here, we conclude that the
molecular detection of MRSA in samples with high CFU counts is reliable, which can and
has been implemented in various laboratory settings with confidence. All tests performed
equally well. However, for direct molecular diagnostics, we have to conclude that current
commercial and in-house tests do not meet the requested quality criteria. The sensitivity of
many tests is relatively low and the specificity needs to be improved. The FDA submission
in 2004 of the IDI-MRSA test had a detection limit of 325 CFU/swab (10). This is much more
sensitive than documented in this study. Pre-enrichment of clinical samples leads to concen-
trations of MRSA exceeding 109 CFU/ml, which is higher than the concentrations of MRSA
likely to be found in a patient sample, and those in this EQA panel. However, pre-enrichment
reduces one of the major advantages offered by NAATs, namely rapidity. In this year’s panel,
only inactivated cells were present. As a consequence, pre-enrichment was not possible. This
may have influenced the results of some laboratories. For the years to come, viable cells will
be distributed, which is more similar to the real clinical situation.

Chapter 8
This is the fourth year that QCMD has offered the MRSA DNA EQA Programme and the
number of participants has increased from 51 in 2006, 61 in 2007 and 74 to 2008 to 80 in
2009 (11). Over the years, we observe a statistically non-significant decrease in overall test
sensitivity. The most pronounced discrepancies were observed in the low concentration
panel samples (103 and 102 CFU/ml). Conversely, the percentage of correct results showed
an overall improvement for the “specificity” samples (containing MSSA and MRCoNS or E.
coli) and the “true negative” samples (Table 4). Again, this was not statistically significant.
Still, these incorrect results underscore the need for improved specificity of molecular MRSA
tests and therefore positive results should always be confirmed by a culture method or a
second molecular test. For laboratories with high false positivity rates or in regions with low
prevalence of MRSA, confirmation is necessary.
In 2009, two MSSA samples harbouring a SCCmec cassette, but lacking the mecA gene,
were included. There was a marked difference in the percentage of correct qualitative results
for the MSSA strain containing the mecA gene compared to the two strains lacking it. These
data show that confirmation on results generated using assays that only target the SCCmec
cassette is mandatory.

139

Rene te Witt bw.indd 139 24-11-11 10:35

 Chapter 8

Table 4. Comparison of performance on like samples in the QCMD 2006, 2007, 2008 and 2009 MRSA EQA Programmes
Sample content Sample concentration Percentage correct results
CFU/ml 2006 2007 2008 2009
MRSA a 1x10E9 96.6 96.3 93.2 NIP
MRSA a 5x10E6 82.8 96.3 91.9 100
MRSA a 5x10E5 NIP NIP NIP 100
MRSA a 5x10E5 NIP NIP NIP 98.8
MRSA a 2.5x10E4 NIP NIP NIP 97.6
MRSA a 2.5x10E3 51.7 72.2 54.1 46.3
MRSA a 1x10E3 51.7 66.7 55.4 NIP
MRSA a 2.5x10E2 12.1 37.0 20.3 15.9
MSSA b 1x10E9 87.9 92.6 89.2 NIP
MSSA b 5x10E6 NIP NIP NIP 87.8
MSSA b 1x10E3 NIP 94.4 95.9 NIP
MRCoNS c 1x10E7 82.8 88.9 94.6 80.5
MSSA b + MRCoNS c 1x10E3 + 1x10E5 94.8 83.3 86.5 NIP
MSSA b + MRCoNS c 1x10E3 + 1x10E4 96.6 77.8 83.8 NIP
MSSA d 5x10E6 NIP NIP NIP 36.6
MSSA e 1x10E7 NIP NIP NIP 39.0
E. coli f 5x10E6 93.1 92.6 97.3 89.0
S. aureus Neg medium - NIP 92.6 98.6 NIP
NIP: not in panel.
MRSA a: Methicillin-resistant Staphylococcus aureus N315.
MSSA b: Methicillin-sensitive Staphyloccoous aureus ATCC 29213.
MRCoNS c: Methicillin-resistant Coagulase-negative Staphylocci 634.
MSSA d: Methicillin-sensitive Staphyloccoous aureus (mecA negative 92-1866 [SCCmec III]).
MSSA e: Methicillin-sensitive Staphyloccoous aureus (mecA negative 93-2886 [SCCmec I]).
E. coli f: Eschericia coli ATCC 35218.

In conclusion, the quality of molecular diagnostic tests still needs improvement and proper
and regular quality control and international standardisation for MRSA diagnostics should be
mandatory for the years to come.
We present the first QCMD EQA program for the genetic characterisation of MRSA strains.
Clear differences in resolution were observed between the datasets. Some PFGE protocols,
which were implemented by most of the participating laboratories, proved to be suboptimal
as a low level of discrimination in the high and/or low molecular weight fragments was ob-
served in the majority of the results reported. This suggests the need for optimisation of the
PFGE program. This lack of resolution was most evident within the group of closely related
MRSA strains in the panel (MRSATP09-08, MRSATP09-09 and MRSA09-10). These strains were
incorrectly reported in 27% of datasets. Participants reported using a range of criteria for
determining genotype and subtype. The guidance according to Tenover et al was the most
prominent method (12).
To improve the performance and quality of molecular diagnostics, both laboratories
and manufacturers should be encouraged to participate in EQAs. The availability of EQA

140

Rene te Witt bw.indd 140 24-11-11 10:35

 External Quality Assessment of molecular diagnostics and genotyping of methicillin-resistant Staphylococcus aureus

panels for detection and typing should also be developed for other important (nosocomial)
infectious agents including vancomycin resistant enterococci (VRE) and extended spectrum
β-lactamase (ESBL) producing Enterobacteriaceae.

Acknowledgements

We acknowledge Deborah Kreft for her technical assistance in the preparation of the panels.

Chapter 8

141

Rene te Witt bw.indd 141 24-11-11 10:35

 Chapter 8

References
1. Brown, D. F., et al. (2005). Guidelines for the laboratory diagnosis and susceptibility testing of methicillin-resistant
Staphylococcus aureus (MRSA). J Antimicrob Chemother 56(6): 1000-18.
2. Weller, T. M. (2000). Methicillin-resistant Staphylococcus aureus typing methods: which should be the international
standard? J Hosp Infect 44(3): 160-72.
3. Ibrahem, S., et al. (2009). Carriage of methicillin-resistant Staphylococci and their SCCmec types in a long-term-
care facility. J Clin Microbiol 47(1): 32-7.
4. Bartels, M. D., et al. (2009). A common variant of staphylococcal cassette chromosome mec type IVa in isolates
from Copenhagen, Denmark, is not detected by the BD GeneOhm methicillin-resistant Staphylococcus aureus
assay. J Clin Microbiol 47(5): 1524-7.
5. Jansen, W. T., et al. (2006). Novel mobile variants of staphylococcal cassette chromosome mec in Staphylococcus
aureus. Antimicrob Agents Chemother 50(6): 2072-8.
6. Ender, M., Berger-Bachi, B. and McCallum, N. (2007). Variability in SCCmecN1 spreading among injection drug users
in Zurich, Switzerland. BMC Microbiol 7: 62.
7. Ichiyama, S., et al. (1991). Genomic DNA fingerprinting by pulsed-field gel electrophoresis as an epidemiological
marker for study of nosocomial infections caused by methicillin-resistant Staphylococcus aureus. J Clin Microbiol
29(12): 2690-5.
8. van Belkum, A., et al. (1998). Assessment of resolution and intercenter reproducibility of results of genotyping
Staphylococcus aureus by pulsed-field gel electrophoresis of SmaI macrorestriction fragments: a multicenter study.
J Clin Microbiol 36(6): 1653-9.
9. Kerremans, J. J., et al. (2008). Detection of methicillin-resistant Staphylococcus aureus in a low-prevalence setting
by polymerase chain reaction with a selective enrichment broth. Diagn Microbiol Infect Dis 61(4): 396-401.
10. Huletsky, A., et al. (2004). New real-time PCR assay for rapid detection of methicillin-resistant Staphylococcus
aureus directly from specimens containing a mixture of staphylococci. J Clin Microbiol 42(5): 1875-84.
11. van Belkum, A., et al. (2007). Quality control of direct molecular diagnostics for methicillin-resistant Staphylococcus
aureus. J Clin Microbiol 45(8): 2698-700.
12. Tenover, F. C., et al. (1995). Interpreting chromosomal DNA restriction patterns produced by pulsed-field gel
electrophoresis: criteria for bacterial strain typing. J Clin Microbiol 33(9): 2233-9.

142

Rene te Witt bw.indd 142 24-11-11 10:35

 Discussion and future perspectives

Rene te Witt bw.indd 143 24-11-11 10:35

Rene te Witt bw.indd 144 24-11-11 10:35
 Chapter 9
New clinical microbiological diagnostics

René te Witt
Alex van Belkum
Willem B. van Leeuwen

Submitted to Nederlands Tijdschrift voor Medische Microbiologie.

Rene te Witt bw.indd 145 24-11-11 10:35

 Chapter 9

Abstract

Current diagnostics of infectious diseases is based on conventional microscopy and culture-

methods. Essentially, laboratories can perform simple tests on a large number of clinical
samples on a daily basis, using inexpensive culture media and simple techniques. This strat-
egy also provides bacterial isolates, which can be further characterized e.g. more detailed
identification of the species, analysis of their antibiotic susceptibility profiles and, in special
cases, epidemiological typing for outbreak analysis can be obtained. Many new diagnostic
techniques have been developed during the past two decades, all with the objective to save
time and to improve accuracy. Next to nucleic acid amplification techniques, there is a trend
towards biophysical technologies such as matrix-assisted laser desorption/ionisation time-
of-flight mass spectrometry (MALDI-TOF MS), the “electronic nose” and Raman spectroscopy.
These will be discussed in this chapter.
Multiplex screening offers significant advantages with respect to time, costs, sample
requirements and the amount of data that can be generated. Many different strategies are
(highly) suited for multiplex detection and candidates for implementation in clinical diagnos-
tic laboratories are discussed in this chapter as well.

146

Rene te Witt bw.indd 146 24-11-11 10:35

 New clinical microbiological diagnostics

Molecular diagnostics

Introduction

Pathogens can be detected in different ways in the daily routine of a clinical microbiological
laboratory. Most prominently used methods are culture, microscopy and (indirect) antigen-
and antibody detection. Molecular diagnostics can also be of great value to clinical microbio-
logical diagnostics. By using molecular techniques, pathogens can be detected by species-
unique nucleic acid targets. Within the molecular diagnostics laboratory, Polymerase Chain
Reaction (PCR) is the most widely used method (1). Real-time PCR can be considered as the
most important recent innovation within microbiological diagnostics (2-4). This technique
combines target amplification with detection of the product by using a fluorescent probe in a
closed system. Amplification and detection take 1-2 hours, which is faster than conventional
PCR and corresponding post-amplification process. The risk of contamination is marginal and
performing real-time PCR demands less hands-on-time and expertise.
In contrast, there are drawbacks for PCR. There is a chance of false-positive (low specificity,
contamination) or false-negative (low sensitivity, variability in the target, inhibition) results.

Bacterial typing

After detection and identification of (drug resistant) bacteria, typing is necessary to assess
the occurrence of bacterial transmission and whether infection-prevention measures are
mandatory. In addition, typing is essential for elucidation of local or (inter)national dissemi-
nation of clones. Currently, many different genotyping methods are in use in the diagnostic
laboratory. Examples are pulsed-field gel electrophoresis (PFGE) of macro restriction frag-
ments of genomic DNA, multi-locus variable number of tandem repeat analysis (MLVA) and

Chapter 9
direct nucleotide sequence-based approaches, such as spa-sequence typing and multi-locus
sequence typing (MLST) (Table 1). Due to the complex, expensive and technical protocols
of these techniques, typing is mainly performed retrospectively and only when there is a
sufficient number of cases (e.g. ≥5 patients).
Multiple new diagnostic techniques have been developed during the last decade and
there is an increasing interest in biophysical techniques. These analytical platforms are more
and more adapted for identification and typing of microorganisms. A significant example is
matrix-assisted laser desorption/ionisation time-of-flight mass spectrometry (MALDI-TOF MS).
This technique identifies microorganisms by their proteomes. Another example with potency
in microbiology is the “electronic nose”; biosensors that recognize specific bacterial odours by
which bacteria can be identified. For bacterial typing, Raman spectroscopy will be discussed
in this chapter. With this technique it is possible to determine clonal relationships (also called
typing) between bacteria, but also between isolates of specific yeast or fungus species.

147

Rene te Witt bw.indd 147 24-11-11 10:35

 Chapter 9

Finally, different candidates for implementation of rapid multiplex screening in clinical

diagnostic laboratories will be discussed.

Table 1. Comparison of the main currently available molecular typing methods.

Method Principle/target Strengths Weaknesses Example(s) References
Pulsed-field gel Restriction - High - Technically demanding MRSA, Pseudomonas (5-8)
electrophoresis polymorphism of the discriminatory - Slow aeruginosa
(PFGE) whole genome power - Limited portability
- Multiple nomenclatures
Multilocus sequence Sequence - Phylogenetic - Limited discriminatory Candida albicans, (5-7)
typing determination of structure of core power MRSA
(MLST) allelic variants of genome - Low throughput
housekeeping genes - Portability - Expensive
- Standard
nomenclature
spa-sequence Polymorphism - Rapid - Moderate S. aureus (MRSA) (8-9)
typing of number and - High throughput discriminatory power
(for S. aureus typing sequence of repeat - Portability - Misclassification of STs*
only) elements of the - Standard due to recombination
hypervariable gene nomenclature
rep-PCR typing Polymorphism - Rapid - Limited discriminatory MRSA, multi-drug (10-13)
(Diversilab) in chromosomal - High throughput power for gram positive resistant bacteria, P.
inter-repeat element - Portability cocci aeruginosa
spacers - No validated
interpretation criteria
- No standard
nomenclature
Multilocus VNTR Polymorphism - Rapid - Limited discriminatory Mycobacterium (14-18)
analysis in number of - High throughput power tuberculosis, Bacillus
(MLVA) chromosomal VNTR** - Standard anthracis, MRSA
elements nomenclature
* ST: Sequence Type
** VNTR: Variable Number of Tandem Repeat.

148

Rene te Witt bw.indd 148 24-11-11 10:35

 New clinical microbiological diagnostics

MALDI-TOF MS

Introduction

MALDI-TOF MS is a relatively new technology within microbiology (5-6). The most common
bacterial pathogens, yeasts and fungi that are currently isolated from clinical samples in a
clinical microbiology laboratory can be identified using MALDI-TOF MS (e.g. Enterobacteria-
ceae, staphylococci, streptococci and Nocardia spp) (7-8). The practical procedure of MALDI-
TOF MS is simple and robust. The sample is transferred onto a sample slide and the cells are
embedded into a matrix. The principle of the procedure is illustrated in Figure 1.

detection ion detector

+ + fi ld f
field-free
separation drift range
+ +
grid electrode
acceleration
acceleration
+ zone
ionization
+
desorption
matrix/analyte
crystals
y

template

Figure 1. Principle of MALDI-TOF MS (Adapted with permission from bioMerieux, Marcy l’Etoile, France).

Chapter 9
The matrix is essential for two reasons. Firstly, the matrix must absorb the energy of the laser
and, secondly, the matrix stimulates ionisation of the microbial fragments after laser excita-
tion. The ions that form are accelerated in an electromagnetic field between the sample and
a detector. After passing two electrodes, the ions are transferred into a vacuum tube, which
contains a detector at the end. The “time-of-flight” of the ions is inversely proportional to the
mass-charge ration of the ions. A complete cycle consists of desorption, ionisation, accelera-
tion, separation and detection. Depending on the microorganism, the sample is processed
40-100 times to accumulate a stable, average mass spectrum. The actual data acquisition
with MALDI-TOF MS is performed in an automated way. The laser scans the sample and ac-
cumulates a number of mass spectra by launching a defined number of laser pulse cycles.

149

Rene te Witt bw.indd 149 24-11-11 10:35

 Chapter 9

In Figure 2, the different spectra that can be observed between different species are illus-
trated.

Certain bacterial species and yeasts are more difficult to identify with MALDI-TOF MS than via
direct transfer of the colony. Special protocols have been developed to improve the quality
of the spectra. These protocols use chemical agents to break the cell walls to expose intracel-
lular proteins, which greatly improves the ability of MALDI-TOF MS to identify yeasts (8-9).
The essential step for species identification is comparing the unknown mass spectrum
of the sample to a database containing reference mass spectra. Obviously, the quality and
completeness of such a database are crucial for an optimal performance. This requires refer-
ence mass fingerprints of all species of interest and mass fingerprints of multiple strains per
species (10).

Figure 2. Examples of MALDI-TOF MS generated spectra. Distinct differences between different, related species can be observed (Adapted with
permission from Ferroni et al).

During the last five years, two commercial systems with extended and secure databases
have been developed for routine diagnostics: MALDI Biotyper from Bruker (Bremen, Ger-
many) and Vitek MS from bioMérieux (Marcy l’Etoile, France) (29-30). With both systems,
batches of 100 samples can be analyzed simultaneously in less than one hour; the analysis
of 1.000 samples per day with a single instrument is realistic. Throughput may be enhanced
even further by efficient automation of the sample preparation process.

Possible drawbacks

Naturally, MALDI-TOF MS-based identification of microorganisms has limits. Some limits are
of a technical nature and include the amount of biomass required for a reliable result (104-105

150

Rene te Witt bw.indd 150 24-11-11 10:35

 New clinical microbiological diagnostics

cells). Consequently, culturing of bacteria prior to identification is needed, which results in

delay in diagnosis. Another technical drawback is variability between duplicate measure-
ments (11). Other limits are of a biological nature and concern the taxonomic resolution
that can be achieved. Examples are the failure to discriminate between Escherichia coli and
Shigella spp. or difficulties with discriminating species of the Streptococcus mitis complex (7,
12). Furthermore, “coagulase negative staphylococci” and “Gram-negative non-fermenting
bacteria” can be identified difficulty by using MALDI-TOF MS. On the one hand, the analyzed
ribosomal proteins provide little discriminatory information; on the other hand this may be
caused by an incomplete database.
However, in the near future, these shortcomings can and will be solved. Moreover, the ap-
plicability of MS for clinical microbiological diagnostics will be expanded further. A number
of studies has evaluated the possibility to differentiate drug resistant strains from sensitive
strains of the same species, such as methicillin resistant S. aureus (MRSA) and methicillin
sensitive S. aureus (MSSA) (13-14). Also, an approach to detect enzymes that are responsible
for antibiotic resistance has been published (15).
Multiple studies have evaluated applications of MALDI-TOF MS to analyze mixed bacterial
species (16-17). Despite the fact that two or three bacterial species can be separated and
identified simultaneously, the discriminatory power of MALDI-TOF MS is restricted when one
single species predominates in the mixture. The background caused by commensal bacteria
hinders the detection of potential pathogens in clinical samples. For urinary tract infections
(UTI), where usually one single species predominates, reliable identification of UTI causing
pathogens has been described (18). Identification of pathogens directly from patient blood is
difficult due to the low titre, but recently pathogenic bacteria have been efficiently detected
by MS directly from positive blood cultures (19).
Currently, possibilities for typing with MALDI-TOF MS are limited as a consequence of the
limited resolution. Optimization of sample preparation and new approaches in data analysis

Chapter 9
or MS technology will possibly allow the discriminatory power of MALDI-TOF MS to increase.

Conclusion

MALDI-TOF MS is currently settling in clinical microbiological diagnostics. During the next 5

years, currently used biochemical identification techniques will be replaced by a combination
of MALDI-TOF MS and automated antibiotic susceptibility testing, such as Vitek (bioMérieux)
or Phoenix (BD Diagnostics, Breda, The Netherlands). The addition of automated sample
preparation will make MALDI-TOF MS an indispensible identification platform in clinical
microbiology.

151

Rene te Witt bw.indd 151 24-11-11 10:35

 Chapter 9

Electronic nose

Introduction

One of the aspects of conventional bacteriological diagnostics is the specific odour of some
bacterial species, such as Clostridium difficile (smells like “stable”), Haemophilus influenzae
(“burned caramel”) and Pseudomonas aeruginosa (“blossom”). Based on the simplicity and
ease-of-use of odour-based diagnostics, this may be an attractive alternative for identifica-
tion of bacteria.
The concept of an electronic nose was developed during the seventies of the 20th century.
The availability of computers made it possible to recognize patterns, analogous to human
smell perception, when measuring complex mixtures of compounds with biosensors. Hence
the name electronic nose.
The metabolism of microorganisms proceeds via different routes, depending on the (ge-
netic) possibilities of a microorganism and on the available nutrients in the environment. As a
consequence, microorganisms produce a wide variety of volatile organic compounds (VOCs)
during growth. By measuring these VOCs, analysis of the metabolic activity can lead to the
development of a fast identification system.

E-nose

In all papers describing the classification of bacteria with electronic noses the analysis models
used a single sensing device (20-21). A company called C-it (Zutphen, The Netherlands) has
developed an electronic nose, called the MonoNose, where complex bacterial VOC signals
are measured using multiple metal-oxide sensors (22). The MonoNose is a broadly applicable,
inexpensive system, which uses real-time VOC pattern recognition and matching of this pat-
tern with a previously identified reference database (Figure 3).
In the already mentioned study, eleven different, clinically relevant bacterial species were
measured. A total of fifty-two clinically relevant strains were tested, showing diagnostic
specificity to be 100% for Clostridium difficile to 67% for Enterobacter cloacae, with an overall
average of 87%. Thanks to the continuous measuring process, reliable identification can be
achieved in 4-8 hours. This research was performed on pure cultures and not directly on clini-
cal samples, so an extra culturing step is necessary to identify the potential causative agents.
Multiple clinical samples, such as urine, and blood, need to be tested to determine the influ-
ence of the sample matrix and the possibilities and drawbacks of direct measurements.

152

Rene te Witt bw.indd 152 24-11-11 10:35

 New clinical microbiological diagnostics

Figure 3. Composition of the MonoNose (Adapted with permission of C-it, Zutphen, The Netherlands).

Possible drawbacks

VOC-based identification of microorganisms has limitations. The available nutrients in the

medium will be depleted faster when applying a high inoculum. In this way, there is a need
to switch metabolism route, which will change the analysis (time) of the E-nose. Obviously,
a higher inoculum will lead to faster identification. The relation between incubation tem-
perature and replication speed is known and will lead to an altered VOC pattern. This pattern
depends on the presence and concentration of nutrients in the medium. Furthermore, these

Chapter 9
influences are organism dependent, which leads to more challenges in creating a reference
database.

Conclusion

Above-mentioned factors are crucial for the future of electronic noses for (direct) identifica-
tion of microorganisms. Direct real-time identification from clinical samples may be a real
point-of-care test. An example of the applicability of the electronic nose is a bench-top blood
culturing machine; blood cultures can be incubated and analysed next to the patient. Another
example can be a surgical bandage with an incorporated nose that gives a signal far before
clinical symptoms appear and can be observed by medical staff. A final example may be the
direct analysis of expiration gasses from a patient to diagnose pneumonia or tuberculosis.

153

Rene te Witt bw.indd 153 24-11-11 10:35

 Chapter 9

Raman spectroscopy

Introduction

The Centre for Optical Diagnostics and Therapy (CODT) of the Erasmus MC in Rotterdam
develops analytical and therapeutic applications based on Raman spectroscopy. The depart-
ment cooperates with pathology and urology (healthy and sick tissue differ in molecular
composition), dermatology (diagnosis and classification of skin tumours and multiple skin
diseases) and medical microbiology (molecular composition of microbial samples, species
identification and typing of bacteria) (23-25).
Raman spectroscopy is named after its discoverer, the Indian physicist Chandrasekhara
Venkata Raman. In 1928, Raman was the first to find experimental proof for in-elastic scat-
tering of light through matter. For this discovery “of extraordinary great importance for our
knowledge of the structure of molecules”, Raman received the Nobel Prize for physics in 1930.

Principle of Raman spectroscopy

When matter in general or molecules in particular, are radiated with (monochromatic) light,
most photons will be scattered with the same, identical wavelength. This process is called
elastic scattering. However, a very small fraction of the light will be scattered in-elastically
and will, thanks to loss of energy, differ from the incident light. This process is called Raman
scattering (Figure 4).

Figure 4. Raman scattering. The incident light λ is being scattered via the molecule, which results in light with the same wavelength (λ), and, in
a very small fraction, in a shift in wavelength (λ+Δ).

Atoms in a molecule vibrate around their equilibrium state. When Raman scattering occurs,
an incident photon will transfer a small fraction of its energy onto the molecule. This results
in elevated vibration(s) and in a photon with a lower energy. This lower energy results in a
shift in wavelength, which can be measured by the detector of the Raman spectrometer. The
amount of energy needed to ensure molecular vibration of the molecule depends on the
mass of the atoms, their chemical bonds, the molecular structure, the micro-environment
(such as pH) and the interactions of the molecule with the environment. This all explains the
molecular specificity of Raman spectra.

154

Rene te Witt bw.indd 154 24-11-11 10:35

 New clinical microbiological diagnostics

To exactly determine the shift(s) in wavelength between incident and scattered light,
monochromatic light is used to create a Raman spectrum of a sample. All molecules of a
sample (e.g. a bacterial cell) contribute to the total Raman spectrum, where the intensity of
the Raman signal depends on the concentration.

Instrument

Nowadays, a Raman spectroscope is relatively simple and consists of 4 base components:

a laser (monochromatic light), a sample place (radiation of the sample), a spectrometer
(detection of scattered light) and a computer (analysis of the collected spectra). Since Ra-
man spectroscopy does not use labels or dyes, a highly pure sample is essential. Any kind of
contamination contributes to the Raman spectrum and will interfere with (the result of ) the
analysis.
A high reproducibility is important to maintain the variance within one group smaller than
the variance between two groups. Only then two (closely) related samples/bacteria can be
distinguished from each other.
Since samples are (almost) always measured versus a background, the composition of this
background needs to be reproducible with minimal contribution to the Raman spectrum,
since this background will interfere with (the result of ) the analysis.

Bacterial typing

Raman research within the CODT has led to the introduction of a commercial typing system,
the SpectraCellRA analyzer (SCRA) (River Diagnostics, Rotterdam) in 2009. This system analy-
ses clonal relationship between bacterial strains. The workflow is easy, with little hands-on-
time. Shortly: isolates are cultured for 20h on Trypticase Soy Agar (TSA). Biomass is obtained

Chapter 9
from the agar and suspended in water. This suspension is transferred onto a Microslide. The
slide is air-dried and Raman spectra can be measured (Figure 5).

Figure 5. Workflow for the SpectraCellRA analyzer.

155

Rene te Witt bw.indd 155 24-11-11 10:35

 Chapter 9

When all measurements are done, the Raman spectra are compared in silico. Subsequently,
a similarity matrix is calculated where the relationships between the different samples are
displayed.

Possible drawbacks

Despite the enormous potential, there are some limitations to Raman spectroscopy. Raman
spectroscopy is a phenotypical method, where minor differences in culture medium, incuba-
tion temperature or – time, in theory, can cause differences in Raman spectra. These prob-
lems can only be intercepted by using culture media of constant quality and strict incubation
times. Extended validation studies are needed to show the influence of the abovementioned
factors by using Raman spectroscopy within clinical microbiology. Next to this, there is a lack
of standard nomenclature, which gives suboptimal portability; Raman type 1 can be different
everywhere.

Conclusion

Raman spectroscopy certainly is highly useful within clinical microbiology thanks to the
simplicity, high reproducibility and the high discriminatory power. With a small amount of
biomass, it can be easily determined if two (or more) individuals are infected or colonized
with the same bacterial strain. Furthermore, due to the possibility of building a database,
it can be implemented to generate transmission alerts when a previous found SCRA type
is detected again. In theory, this can lead to accurate real-time and continuous, instead of
retrospective, monitoring of isolates to prevent health-care associated infections.

156

Rene te Witt bw.indd 156 24-11-11 10:35

 New clinical microbiological diagnostics

Multiplex detection

Multiplex analysis offers significant advantages regarding time, costs, sample requirements
and the amount of data that can be generated. Many different techniques are (highly) suit-
able for multiplex detection. The most important ones are reviewed in the next paragraphs.

Luminex

Different applications of Luminex for the simultaneous detection of multiple respiratory

viruses or bacterial pathogens have been reported (26-30). The technique has also been
used for other diseases like meningococcal infection and fungal infections (50-51). Luminex
technology incorporates flow cytometric methods using custom-designed microbeads that
are coloured with slightly different shades of the red spectrum (31). Each coloured bead is
labelled with a different fluorescent probe that is specific to each particular target. After mul-
tiplex PCR amplification of targets, the PCR products can be incubated with the beads. The
bead mixture is sorted by flow cytometry based on the internal bead colours and hybridized
samples produce a fluorescent signal (32).
The rapid detection of antibodies against S. aureus with Luminex technology showed that
antibody levels were associated with the presence of specific genes in infectious S. aureus
isolates (33). Another study showed that persistent carriage of S. aureus is influenced by and
associated with genetic variation in host inflammatory response genes (34). Both approaches
may be useful in fast screening for (susceptibility) to (methicillin-resistant) S. aureus carriage
or infection.

Microarray

Chapter 9
Microarrays are massively parallel detection platforms that were first used for gene expres-
sion studies, but have also been successfully applied for microbial detection and identifica-
tion (35). The technology has enabled researchers to gain insight into the microbial diversity
of human samples and has enabled studies for multi-pathogen infections. In contrast to gene
expression studies, the concentrations of targets in analyzed samples for microbial detection
are usually much lower, which requires the use of nucleic acid amplification.
The adaptation of microarray technology for diagnostics allows full differentiation of a
pathogen from a non-pathogen. This could allow the nucleic acid from a sample to be ex-
tracted and be hybridized to a glass slide containing DNA fragments complementary to the
genes of interest, unique to the tested pathogen. Rather than individual tests for each patho-
gen, sample type screening would be able. For example, a microarray slide could be created
containing a number of unique gene segments for all pathogens associated with respiratory

157

Rene te Witt bw.indd 157 24-11-11 10:35

 Chapter 9

tract infections. This would then allow the laboratory to process a single respiratory sample
for identification of all pathogens present in that sample in one quick assay.
Another great benefit is that such a gene approach could also employ antibiotic resistance
genes, so that antibiotic resistance present in a sample could be detected, which is still a ma-
jor challenge in adopting molecular diagnostics over culture-based isolation and character-
ization. Antimicrobial resistance is one of the most important and urgent problems in clinical
bacteriology and the use of microarrays has already demonstrated its suitability in general
(36-37). For complex resistance problems, such as detection of different Extended Spectrum
Beta Lactamase (ESBL) variants or MRSA, DNA microarrays were published (38-39). All these
assays, although they need further optimization regarding implementation in routine clini-
cal laboratories, are characterized by a short time-to-result (4-5 hours). Besides the fact that
resistance-related information is crucial for the adequate treatment of life-threatening infec-
tions, the obtained information can also be used for epidemiological surveillance.
Of course, there are a large number of obstacles before such technology can be employed
in a clinical setting. These include the costs and the level of user skill required for the technol-
ogy. Also the amount of pathogen DNA required to label and then hybridize to the slide
to produce a detectable signal needs to be sufficient. This can be overcome by PCR-based
amplification of the DNA in the sample, but this is extremely difficult if attempting to amplify
hundreds of gene segments for the type of array previously mentioned (40). Most problem-
atic is the fact that still one only finds what one is searching for: arrays always need prior
sequence information and are incapable of identifying novel sequence variants. Most likely,
the use of microarrays in clinical diagnostics will be overpowered by the introduction of
(whole-genome) next generation sequencing.

Sequencing

Another example of multiplex analysis is the use of molecular techniques to determine the
nucleotide sequence of 16S ribosomal RNA (rRNA), which was already described in 1980 (41).
Where phenotypic identification is subject to variations in morphology and metabolic status,
genotypic classification is based on relatively stable and uniform molecular targets. The use
of 16S rRNA gene sequences for identification of bacteria has been most successful for a
number of reasons. These reasons include the presence of 16S rRNA in all bacteria, that the
function of 16S rRNA gene over time has not changed (hence its relatively strong sequence
conservation) and that the 16S rRNA gene is large enough (1500 bp) for discrimination
between different species (42). Furthermore, 16S rRNA provides the possibility to detect
nonviable bacteria and bacteria with special culture requirements.
Naturally, there are challenges associated with 16S rRNA sequencing. The conserved na-
ture of ribosomal genome sequences across the different genera of bacteria and the high
sensitivity of PCR easily allows for the generation of false-positive results due to target con-

158

Rene te Witt bw.indd 158 24-11-11 10:35

 New clinical microbiological diagnostics

tamination. There are different conserved regions within the 16S rRNA gene. Therefore there
are different primer sequences that can be used for 16S rRNA PCR (43). Despite being broad
range primers, it is unlikely that any primer set will amplify all bacteria. Because of difficulties
in the interpretation of DNA electropherograms resulting from direct sequencing of clinical
polybacterial samples, the use of 16S rRNA sequencing has initially been limited to clinical
samples of infections that are predominantly monobacterial. However, Kommedal et al have
developed an algorithm for analysis of mixed chromatograms, which makes it possible to
apply 16S rRNA sequencing as a diagnostic tool (44-45).
Sequencing is a powerful approach for decoding a number of human infectious diseases.
The introduction of next-generation sequencing (NGS) technologies has dramatically re-
duced sequencing costs and significantly increased the output and throughput, making
whole-genome sequencing possible. Although a relatively new technology and only used in
research laboratories, whole genome sequencing has the potential to provide an enormous
load of diagnostic capabilities as the technology improves and costs becomes reasonable.
Examples are platforms developed by companies such as 454 Genome Sequencer FLX Sys-
tem (GS FLX; Roche Life Sciences, Basel, Switzerland), Genome Analyzer II (GA; Illumina, CA,
USA), Sequencing by Oligonucleotide Ligation and Detection (SOLiD; Life Technologies, CA,
USA), HeliScope Single Molecule Sequencer (Helicos Biosciences, MA, USA) and Ion Torrent
(Life Technologies, CA, USA).
NGS technologies have reduced both cost-per-base and time required to decode an entire
genome, making DNA sequencing a cost-effective option for many experimental approaches.
Although differing in sequencing chemistries and technical details, all NGS platforms utilize
a similar strategy, miniaturization and multiplexing of individual sequencing reactions to
overcome the limited power of traditional Sanger sequencing (46-47).
Traditional Sanger sequencing involves creating different sizes of fluorescently labelled
PCR products in the same reaction tube. The templates used in this reaction are purified plas-

Chapter 9
mids or amplified DNA fragments. There are four different coloured labels that are specific to
one of the four DNA bases. As a result, numerous different-sized products are created which
are separated on a capillary electrophoresis instrument by size and detected by fluorescence.
The sequence can be compared with database entries for identification of the species of a
microorganism or genes indicative for drug resistance. For example, one study describes
the use of sequencing for identification of different Mycobacterial species by using several
different gene targets (48-49).
The Roche 454 GS FLX System is based on emulsion PCR and pyrophosphatase detection
techniques (50-52). Compared with other NGS platforms, the strength of the 454 GS FLX
system are its longer sequence reads. The Roche 454 GS FLX can generate more than 1 mil-
lion individual sequence reads with lengths over 400 bases during a 10 h run. Although its
per-base cost is much higher than that of other NGS platforms, the Roche 454 system is best

159

Rene te Witt bw.indd 159 24-11-11 10:35

 Chapter 9

suited for certain applications, such as de novo sequencing of genomes for which long read
lengths are essential (50, 53).
The Illumina GA system is the first short-read sequencing platform (46). It uses an array
technique to achieve solid-phase DNA amplification. Solid-phase amplification can produce
100-200 million spatially separated template clusters, providing free ends to which a universal
sequencing primer can be hybridized to initiate the NGS reaction. This repetitive sequencing
process takes approximately 2-3 days to generate 50 million reads, with a read-length of 36
bases. The tedious preparatory steps, however, are being automated as we speak.
The SOLiD system is based on the techniques described by Shendure et al and McKernan
et al (54-55). Library construction for the SOLiD system is similar to Roche 454 technology.
However, unlike the Roche 454 sequencing approach, the sequencing-by-synthesis in the
SOLiD system is based on ligation chemistry.
The HeliScope Genetic Analysis System is the first commercialized single-molecule DNA
sequencer. The technology is based on work by Braslavsky et al and relies on the cyclic inter-
rogation of an array of sequencing features (56). Each sequencing cycle consists of the suc-
cessive addition of polymerase and a different type of dye-labelled nucleotide. The HeliScope
instrument is capable of imaging billions of single molecules per run.
Last in line is Ion Torrent, which uses a sequencing strategy similar to the 454, except
that hydrogen ions (H+) are detected (instead of a pyrophosphatase cascade) and standard
microchips are used as sequencing chips. Detection of H+ means that no lasers, cameras or
fluorescent dyes are needed. Using common microchip design standards means that low-
cost manufacturing can be used.
There are important differences among the abovementioned NGS technologies. For
example, the Roche 454 sequencer and the Ion Torrent may be preferable for de novo se-
quencing because of its longer read length (53). The Illumina and Life Technologies platforms
are particularly well suited for variant discovery by resequencing genomes because gigantic
volumes of high-quality base sequences are produced per run (57). The Helicos platform
is well suited for applications that demand quantitative information in RNA-seq or direct
RNA sequencing (58-60). Table 2 provides an overview of the main features of these NGS
technologies.

Once the fragment sequences have been determined, they are aligned to a known reference
sequence or assembled de novo. These technologies can provide gigabytes of sequence
information about the whole organism, which has the potential to change the treatment
for specific infectious diseases. However, there are a number of challenges that need to be
overcome before sequencing can become the predominant diagnostic and epidemiological
technology in clinical microbiology.
Cost, ease-of-use and need for technical skilled staff limit this method to be introduced in
routine laboratories. There is an urgent need for studies examining the mode and tempo of

160

Rene te Witt bw.indd 160 24-11-11 10:35

 Table 2. Overview of next-generation sequencing technologies.

Rene te Witt bw.indd 161

Platform NGS chemistry Read length (bases) Run time (days) Yield Mb/run Advantages Disadvantages Applications References
Roche 454 GS FLX Pyrosequencing ~ 400 0.5-1 50 - Longer reads - Expensive - Bacterial genome de (64, 75)
System - Short run times novo assemblies
- High throughput - Identification of
multibacterial samples
Illumina GA system Sequencing by synthesis 35-150 1.5-8 1.000-96.000 - Currently the most - Slightly higher costs - Variant discovery (SNP (76-77)
with reversible terminators widely used system detection)
- Evolution
Life Technologies Sequencing by ligation ~ 50 3-12 71.000-200.000 - Cheap - Long run times - Variant discovery (68-69)
SOLiD chemistry - High accuracy - Relatively short reads
- Ability to save failed
sequencing cycles
Helicos Bioscience True single molecule 25-55 8 28.000 - Large numbers of - High error rate - RNA-seq (79-81, 85)
HeliScope sequencing by synthesis reads directly from - RNA sequencing
single molecules
Ion Torrent Synthesis with H+ detection ~ 100 0.25-0.5 100->1.000 - Cheap - Sample preparation time - Bacterial genome de (78)
- Longer reads novo assemblies
SNP: Single Nucleotide Polymorphism.

161
New clinical microbiological diagnostics

Chapter 9

24-11-11 10:35
 Chapter 9

genomic evolution in different bacterial lineages and the performance of these technologies
when applied in real-life. There is a need for improved user-friendly informatics platforms
that allow the average technician to analyze and exploit high-throughput sequencing data.
Furthermore, approaches for Single Nucleotide Polymorphism (SNP) detection and valida-
tion need to be improved. More informative epidemiological markers need to be identified.
The focus should be on improved read lengths to capture longer sequences and, finally, there
is a need for better analytical tools and databases.
Towards the end of this decade it is likely that NGS technologies have become a com-
mon clinical diagnostic tool and that medical laboratories will directly sequence DNA or
RNA extracted from clinical samples. To achieve this, we will need improved approaches to
sample preparation. Finally, it still remains to be seen if even the uncovery of full genetic and
transcriptional potential can be used to define phenotypes that usually depend on genome-
environment interactions as well.

Concluding remarks

We can state that conventional diagnostics still is and will remain present in clinical microbio-
logical laboratories for the coming years. Molecular diagnostics can be implemented as an
additional diagnostic tool, but with restrictions.
MALDI-TOF MS is highly suited for high-throughput identification of “common” clinically
relevant microorganisms. The electronic nose has the potential to be developed for direct
detection and identification of microorganisms, as a real bed-site (point-of-care) test for
infectious disease diagnostics. In the field of typing, Raman spectroscopy is a candidate for
fast and reliable typing of bacterial strains, in both large and small laboratories.
High-speed multiplex analysis will be the ultimate diagnostic strategy for future diagnos-
tics of infectious diseases, which will ensure more efficient patient management.
Furthermore, the technology push is not over yet! Techniques such as bacterial life-dead
assays and various applications of spectroscopy (e.g. infrared, single cell, electrospray ioniza-
tion) are being explored for their use in clinical diagnostics microbiology as we speak.

162

Rene te Witt bw.indd 162 24-11-11 10:35

 New clinical microbiological diagnostics

References
1. Mullis, K. B. and Faloona, F. A. (1987). Specific synthesis of DNA in vitro via a polymerase-catalyzed chain reaction.
Methods Enzymol 155: 335-50.
2. Gibson, U. E., Heid, C. A. and Williams, P. M. (1996). A novel method for real time quantitative RT-PCR. Genome Res
6(10): 995-1001.
3. Heid, C. A., et al. (1996). Real time quantitative PCR. Genome Res 6(10): 986-94.
4. Holland, P. M., et al. (1991). Detection of specific polymerase chain reaction product by utilizing the 5’----3’ exo-
nuclease activity of Thermus aquaticus DNA polymerase. Proc Natl Acad Sci U S A 88(16): 7276-80.
5. Claydon, M. A., et al. (1996). The rapid identification of intact microorganisms using mass spectrometry. Nat
Biotechnol 14(11): 1584-6.
6. Pieles, U., et al. (1993). Matrix-assisted laser desorption ionization time-of-flight mass spectrometry: a powerful
tool for the mass and sequence analysis of natural and modified oligonucleotides. Nucleic Acids Res 21(14): 3191-6.
7. Bizzini, A., et al. (2010). Performance of matrix-assisted laser desorption ionization-time of flight mass spectrom-
etry for identification of bacterial strains routinely isolated in a clinical microbiology laboratory. J Clin Microbiol
48(5): 1549-54.
8. van Veen, S. Q., Claas, E. C. and Kuijper, E. J. (2010). High-throughput identification of bacteria and yeast by matrix-
assisted laser desorption ionization-time of flight mass spectrometry in conventional medical microbiology
laboratories. J Clin Microbiol 48(3): 900-7.
9. Marklein, G., et al. (2009). Matrix-assisted laser desorption ionization-time of flight mass spectrometry for fast and
reliable identification of clinical yeast isolates. J Clin Microbiol 47(9): 2912-7.
10. Lartigue, M. F., et al. (2009). Identification of Streptococcus agalactiae isolates from various phylogenetic lineages
by matrix-assisted laser desorption ionization-time of flight mass spectrometry. J Clin Microbiol 47(7): 2284-7.
11. Cherkaoui, A., et al. (2010). Comparison of two matrix-assisted laser desorption ionization-time of flight mass
spectrometry methods with conventional phenotypic identification for routine identification of bacteria to the
species level. J Clin Microbiol 48(4): 1169-75.
12. Welker, M. and Moore, E. R. (2011). Applications of whole-cell matrix-assisted laser-desorption/ionization time-of-
flight mass spectrometry in systematic microbiology. Syst Appl Microbiol 34(1): 2-11.
13. Du, Z., et al. (2002). Identification of Staphylococcus aureus and determination of its methicillin resistance by

Chapter 9
matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. Anal Chem 74(21): 5487-91.
14. Edwards-Jones, V., et al. (2000). Rapid discrimination between methicillin-sensitive and methicillin-resistant
Staphylococcus aureus by intact cell mass spectrometry. J Med Microbiol 49(3): 295-300.
15. Camara, J. E. and Hays, F. A. (2007). Discrimination between wild-type and ampicillin-resistant Escherichia coli by
matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. Anal Bioanal Chem 389(5): 1633-8.
16. Wahl, K. L., et al. (2002). Analysis of microbial mixtures by matrix-assisted laser desorption/ionization time-of-flight
mass spectrometry. Anal Chem 74(24): 6191-9.
17. Warscheid, B. and Fenselau, C. (2004). A targeted proteomics approach to the rapid identification of bacterial cell
mixtures by matrix-assisted laser desorption/ionization mass spectrometry. Proteomics 4(10): 2877-92.
18. Ferreira, L., et al. (2010). Direct identification of urinary tract pathogens from urine samples by matrix-assisted laser
desorption ionization-time of flight mass spectrometry. J Clin Microbiol 48(6): 2110-5.
19. Drancourt, M. (2010). Detection of microorganisms in blood specimens using matrix-assisted laser desorption
ionization time-of-flight mass spectrometry: a review. Clin Microbiol Infect 16(11): 1620-5.

163

Rene te Witt bw.indd 163 24-11-11 10:35

 Chapter 9

20. Guernion, N., et al. (2001). Identifying bacteria in human urine: current practice and the potential for rapid, near-
patient diagnosis by sensing volatile organic compounds. Clin Chem Lab Med 39(10): 893-906.
21. Dutta, R., et al. (2002). Bacteria classification using Cyranose 320 electronic nose. Biomed Eng Online 1: 4.
22. Bruins, M., et al. (2009). Device-independent, real-time identification of bacterial pathogens with a metal oxide-
based olfactory sensor. Eur J Clin Microbiol Infect Dis 28(7): 775-80.
23. Maquelin, K., et al. (2002). Identification of medically relevant microorganisms by vibrational spectroscopy. J
Microbiol Methods 51(3): 255-71.
24. Willemse-Erix, D. F., et al. (2009). Optical fingerprinting in bacterial epidemiology: Raman spectroscopy as a real-
time typing method. J Clin Microbiol 47(3): 652-9.
25. Willemse-Erix, D. F., et al. (2010). Towards Raman-based epidemiological typing of Pseudomonas aeruginosa. J
Biophotonics 3(8-9): 506-11.
26. Li, H., et al. (2007). Simultaneous detection and high-throughput identification of a panel of RNA viruses causing
respiratory tract infections. J Clin Microbiol 45(7): 2105-9.
27. Benson, R., et al. (2008). Development and evaluation of a novel multiplex PCR technology for molecular dif-
ferential detection of bacterial respiratory disease pathogens. J Clin Microbiol 46(6): 2074-7.
28. Mahony, J., et al. (2007). Development of a respiratory virus panel test for detection of twenty human respiratory
viruses by use of multiplex PCR and a fluid microbead-based assay. J Clin Microbiol 45(9): 2965-70.
29. Pabbaraju, K., et al. (2008). Comparison of the Luminex xTAG respiratory viral panel with in-house nucleic acid
amplification tests for diagnosis of respiratory virus infections. J Clin Microbiol 46(9): 3056-62.
30. Krunic, N., et al. (2007). xTAG RVP assay: analytical and clinical performance. J Clin Virol 40 Suppl 1: S39-46.
31. Dunbar, S. A. (2006). Applications of Luminex xMAP technology for rapid, high-throughput multiplexed nucleic
acid detection. Clin Chim Acta 363(1-2): 71-82.
32. Dunbar, S. A., et al. (2003). Quantitative, multiplexed detection of bacterial pathogens: DNA and protein applica-
tions of the Luminex LabMAP system. J Microbiol Methods 53(2): 245-52.
33. Verkaik, N. J., et al. Immunogenicity of toxins during Staphylococcus aureus infection. Clin Infect Dis 50(1): 61-8.
34. Emonts, M., et al. (2008). Host polymorphisms in interleukin 4, complement factor H, and C-reactive protein associ-
ated with nasal carriage of Staphylococcus aureus and occurrence of boils. J Infect Dis 197(9): 1244-53.
35. Smoot, L. M., et al. (2005). DNA microarrays as salivary diagnostic tools for characterizing the oral cavity’s microbial
community. Adv Dent Res 18(1): 6-11.
36. Westin, L., et al. (2001). Antimicrobial resistance and bacterial identification utilizing a microelectronic chip array.
J Clin Microbiol 39(3): 1097-104.
37. Monecke, S., et al. (2008). DNA microarray-based genotyping of methicillin-resistant Staphylococcus aureus strains
from Eastern Saxony. Clin Microbiol Infect 14(6): 534-45.
38. Grimm, V., et al. (2004). Use of DNA microarrays for rapid genotyping of TEM beta-lactamases that confer resis-
tance. J Clin Microbiol 42(8): 3766-74.
39. Strommenger, B., et al. (2007). DNA microarray for the detection of therapeutically relevant antibiotic resistance
determinants in clinical isolates of Staphylococcus aureus. Mol Cell Probes 21(3): 161-70.
40. Call, D. R. (2005). Challenges and opportunities for pathogen detection using DNA microarrays. Crit Rev Microbiol
31(2): 91-9.
41. Fox, G. E., et al. (1980). The phylogeny of prokaryotes. Science 209(4455): 457-63.
42. Patel, J. B. (2001). 16S rRNA gene sequencing for bacterial pathogen identification in the clinical laboratory. Mol
Diagn 6(4): 313-21.

164

Rene te Witt bw.indd 164 24-11-11 10:35

 New clinical microbiological diagnostics

43. Isenbarger, T. A., et al. (2008). The most conserved genome segments for life detection on Earth and other planets.
Orig Life Evol Biosph 38(6): 517-33.
44. Kommedal, O., Karlsen, B. and Saebo, O. (2008). Analysis of mixed sequencing chromatograms and its application
in direct 16S rRNA gene sequencing of polymicrobial samples. J Clin Microbiol 46(11): 3766-71.
45. Kommedal, O., et al. (2009). Direct 16S rRNA gene sequencing from clinical specimens, with special focus on
polybacterial samples and interpretation of mixed DNA chromatograms. J Clin Microbiol 47(11): 3562-8.
46. Metzker, M. L. (2010). Sequencing technologies - the next generation. Nat Rev Genet 11(1): 31-46.
47. Sanger, F., Nicklen, S. and Coulson, A. R. (1977). DNA sequencing with chain-terminating inhibitors. Proc Natl Acad
Sci U S A 74(12): 5463-7.
48. Adekambi, T. and Drancourt, M. (2004). Dissection of phylogenetic relationships among 19 rapidly growing My-
cobacterium species by 16S rRNA, hsp65, sodA, recA and rpoB gene sequencing. Int J Syst Evol Microbiol 54(Pt 6):
2095-105.
49. Bao, J. R., et al. (2010). Identification of acid-fast bacilli using pyrosequencing analysis. Diagn Microbiol Infect Dis
67(3): 234-8.
50. Droege, M. and Hill, B. (2008). The Genome Sequencer FLX System--longer reads, more applications, straight
forward bioinformatics and more complete data sets. J Biotechnol 136(1-2): 3-10.
51. Nyren, P., Pettersson, B. and Uhlen, M. (1993). Solid phase DNA minisequencing by an enzymatic luminometric
inorganic pyrophosphate detection assay. Anal Biochem 208(1): 171-5.
52. Tawfik, D. S. and Griffiths, A. D. (1998). Man-made cell-like compartments for molecular evolution. Nat Biotechnol
16(7): 652-6.
53. Su, Z., et al. (2011). Next-generation sequencing and its applications in molecular diagnostics. Expert Rev Mol
Diagn 11(3): 333-43.
54. Shendure, J., et al. (2005). Accurate multiplex polony sequencing of an evolved bacterial genome. Science
309(5741): 1728-32.
55. McKernan, K. J., et al. (2009). Sequence and structural variation in a human genome uncovered by short-read,
massively parallel ligation sequencing using two-base encoding. Genome Res 19(9): 1527-41.
56. Braslavsky, I., et al. (2003). Sequence information can be obtained from single DNA molecules. Proc Natl Acad Sci U
S A 100(7): 3960-4.
57. Harismendy, O., et al. (2009). Evaluation of next generation sequencing platforms for population targeted se-

Chapter 9
quencing studies. Genome Biol 10(3): R32.
58. Ozsolak, F., et al. (2009). Direct RNA sequencing. Nature 461(7265): 814-8.
59. Lipson, D., et al. (2009). Quantification of the yeast transcriptome by single-molecule sequencing. Nat Biotechnol
27(7): 652-8.
60. Wang, Z., Gerstein, M. and Snyder, M. (2009). RNA-Seq: a revolutionary tool for transcriptomics. Nat Rev Genet 10(1):
57-63.

165

Rene te Witt bw.indd 165 24-11-11 10:35

Rene te Witt bw.indd 166 24-11-11 10:35
 Chapter 10
Summarizing discussion and future
perspectives

Rene te Witt bw.indd 167 24-11-11 10:35

Rene te Witt bw.indd 168 24-11-11 10:35
 Summarizing discussion and future perspectives

In this thesis, new diagnostic tests for infectious diseases were explored in addition to ap-
plications where conventional culture- and microscopy-based methods have long been and
may still be considered as the “gold standard”. The following chapter will discuss the previous
chapters and a perspective on future developments within the field of diagnostic clinical
microbiology will be given.

Introduction

Chapter 2 outlined the microbiological tests that are currently available for the detection
of Mycoplasma pneumoniae, Chlamydophila pneumoniae and Legionella spp. from clinical re-
spiratory tract samples. Historically, atypical etiological agents of respiratory tract infections
(RTI) were detected on the basis of microbiological culture or serology. Without exception,
sensitivity and specificity of such cultures did not meet criteria of excellence. Therefore, al-
ternatives were sought and a variety of antigen- or antibody mediated tests was developed.
Unfortunately, none of these individual tests met the performance criteria to reliably identify
these atypical causative infectious agents of RTI in all cases. Moreover, the more recently
developed wide variety of nucleic acid amplification tests did not (yet) solve these diagnostic
problems as well. In today’s clinical microbiology laboratories, no widely accepted gold
standard for the detection of M. pneumoniae, C. pneumoniae and Legionella spp. is available.
Technicians and physicians have to rely on a combination of test results that, in combination
with clinical patient data, may lead to a presumptive identification of these microorganisms
at best. Although significant improvements of the diagnostic platforms have been made over
the past twenty years, no reliable tool is as yet available. Future developments for identifi-
cation in clinical diagnostics are still required before the ultimate test for the diagnosis of
atypical RTI agents will become available.

Biomarkers

The first part of Chapter 3 discussed the performance of procalcitonin (PCT)-, neopterin- and
Chapter 10

soluble triggering receptor expressed on myeloid cells-1 (sTREM-1) levels as potentially suit-
able biomarkers to discriminate severe Plasmodium falciparum disease from non-severe P.
falciparum disease and non-P. falciparum diseae at the initial clinical assessment of travellers
with imported malaria. In developed countries such as The Netherlands, fatal cases of malaria
are only reported occasionally (1-2), which is in contrast to the situation in regions of high
malaria endemicity. Therefore, most clinicians in developed countries encounter problems
making a proper diagnosis of malaria.
The quantification of sTREM-1 levels in admission samples of the patient did not result
in proper discrimination of severe P. falciparum malaria and uncomplicated P. falciparum
malaria or non-P. falciparum malaria. Significantly higher levels of PCT and neopterin were

169

Rene te Witt bw.indd 169 24-11-11 10:35

 Chapter 10

observed in travellers with severe P. falciparum malaria upon hospital admission as com-
pared to travellers with uncomplicated P. falciparum malaria or non-P. falciparum malaria,
respectively. These findings are in agreement with those of several other studies performed
in semi-immune malaria patients originating from malaria-endemic regions (3-4). When the
ROC curve characteristics of neopterin and procalcitonin were compared to those of plasma
lactate, the AUROC of neopterin appeared superior whereas the AUROC of PCT appeared
inferior to lactate. This suggests that neopterin may provide the highest accuracy in its diag-
nostic performance to identify severe P. falciparum malaria in this cohort of travellers.
Unfortunately, the applicability of both neopterin and PCT in the initial clinical assessment
of patients with severe P. falciparum malaria will probably be limited by the poor positive
predictive value of both tests, indicating that neither test can serve as a useful tool for the
diagnosis of severe P. falciparum malaria. On the other hand, the high negative predictive
value of both neopterin and procalcitonin suggests that these tests can still be of value by
providing an exclusion criterion of severe disease. With either a PCT level of less than 0.9 ng/
ml or a neopterin level of less than 7.9 ng/ml in serum on admission as a cut-off point for
severe P. falciparum malaria, no patient with severe disease would have been denied access
to high-level monitoring and intensive treatment.
In conclusion, although neither neopterin nor PCT testing can probably serve as a useful
single diagnostic tool for severe P. falciparum malaria, the high negative predictive value of
both neopterin and PCT may be helpful for a rapid exclusion of severe P. falciparum malaria
upon admission. This may be valuable – particularly if available as a cheap and rapid diagnos-
tic test - for physicians only occasionally dealing with ill-returned travellers and who need to
decide on subsequent oral anti-malarial treatment or a timely referral to a specialized centre
for high-level monitoring and intensified parenteral treatment.

The second part of Chapter 3 focused on the diagnostic application of PCT- and neopterin
levels as biomarkers to discriminate between a bacterial and a viral cause of infection. The
value of PCT and neopterin in clinical decision making was shown in several studies (5-10),
however not in all (11). In our study we found a disappointingly poor diagnostic accuracy for
both PCT and neopterin for bacterial and viral causes of fever in ill-returning travellers. In ad-
dition, neopterin even had a more accurate performance for bacterial than for viral disease.
These findings may be explained by the observation that S. Typhi and Paratyphi and Rickettsia
species are intracellular bacteria and that other intracellular bacteria such as tuberculosis and
melioidosis also lead to increased neopterin levels (10).
Of the more traditional leukocyte differential count-based parameters, only presence of
leukopenia, lymphocytosis and presence of atypical lymphocytes were suggestive of viral
disease. Its use in clinical practice is hampered by its considerable overlap with the findings
in the bacterial infections. Interestingly, when empirical antibiotic treatment was withheld on
admission on the basis of combined presence of lymphocytosis and/or CRP ≤ 10 mg/l, only 5

170

Rene te Witt bw.indd 170 24-11-11 10:35

 Summarizing discussion and future perspectives

of 33 patients with Dengue would receive unnecessary antibiotic treatment but all patients
with enteric fever and rickettsial diseases would receive antibiotic treatment; an almost 85%
reduction of unnecessary empirical antibiotic treatment. This finding alone would make the
use of this simple combination of diagnostics a mandatory procedure for clinical microbiol-
ogy laboratories.
Since the present findings were based on a relatively small number of selected patients with
possibly inherent study limitations such as selection bias and lack of power, the suggested
clinical decision rule for empirical antibiotic treatment in returning febrile travellers without
leukocytosis on admission needs validation in properly designed prospective studies.

Culture

Chapter 4 described the in vitro evaluation of the performance of Granada tube broth (GT)
(bioMérieux, Marcy l’Etoile, France) for detection of Group B Streptococcus (GBS) in swabs,
both direct and after transport. We compared the performance of GT with the currently
recommended Amies transport medium (AT).
Direct incubation of GT at 37°C resulted in the highest yield of GBS. No significant differ-
ences in growth of GBS were observed between transport of GT at RT (73%) or at 4°C (55%).
When GT were subsequently incubated at 37°C, the sensitivity increased to almost 100%.
The limitation of our study was its in vitro character. The application of purified isolates
of GBS in GT and onto AT may not reflect GBS survival on swabs containing vaginal and/or
rectal flora. Nonetheless, we can conclude that GT is a highly sensitive transport and culture
medium to detect GBS as the survival rate of GBS was significantly higher in GT as compared
to AT.
GT may be especially suited for the transport of swabs from pregnant women, attending
general practitioners or midwifery practices, to the laboratory, which may take 2-3 days.
Specifically, if a woman has low-density GBS colonization, extended transport times of swab
specimens at RT or higher could reduce the culture sensitivity for AT but possibly not for GT.
GT may not be suitable for the direct detection of GBS from clinical samples, as the absence
Chapter 10

of orange pigment does not conclude in the absence of GBS. Therefore, GT should always
be sub-cultured for optimal use. Furthermore, GT should always be incubated at 37°C to
improve its sensitivity.
However, the question remains if culture is the most optimal method for detection of GBS
since PCR–based assays have fared better in the detection of GBS. One multicenter study
demonstrated that a molecular-based assay for GBS colonization during labour was highly
sensitive and specific (12). More rapid molecular tests allow for point-of-care testing. Gen-
eXpert GBS (Cepheid Diagnostics, Sunnyvale, CA) is such a rapid, real-time PCR-based assay
for the intra-partum detection of GBS colonization, with high sensitivity and specificity (13).
The system fully automates DNA extraction and PCR within the assay cartridges and results

171

Rene te Witt bw.indd 171 24-11-11 10:35

 Chapter 10

are presented in less than 75 minutes. The use of this system could result in a reduction of
early-onset GBS disease in neonates, better than culture.

Nucleic acid amplification

Chapter 5 provided an update on molecular diagnostics and typing of methicillin-resistant

Staphylococcus aureus (MRSA). MRSA is responsible for a large and still growing number of
both health-care and community-associated infections, resulting in increased morbidity and
excessive healthcare costs. Screening of individuals combined with an aggressive infection
control program has become the standard for management of these infections. Rapid screen-
ing methods that allow reliable detection of MRSA within hours are now available. The short
time-to-result is a clear advantage that has provided a tool for successful infection control
strategies. However, every assay should be evaluated against the local MRSA diversity before
being introduced in the diagnostic microbiological laboratory. Continuous evolution of SC-
Cmec, constrains continuously monitoring of the assay performance and positive results of
direct MRSA testing should always be confirmed by a culture method or a second molecular
test. For laboratories with high false positivity rates or in regions with low prevalence of
MRSA, confirmation is essential.
The quality of molecular diagnostic tests and typing techniques is still under discussion.
Adequate internal and external quality control and international standardization for MRSA
diagnostics should be improved immediately. To achieve this, both laboratories and manu-
facturers should be encouraged to participate in EQAs.

Bacterial typing

In Chapter 6, we have analyzed a representative selection of an international collection of

epidemic MRSA strains, the Harmony collection, originating from 11 different European
countries using the DiversiLab System (bioMérieux), pulsed-field gel electrophoresis (PFGE)
with SmaI macro-restriction analysis and multi-locus sequence typing (MLST). The HARMONY
isolates represent all major hospital-acquired MRSA types found in Europe.
The primary aim of this study was to assess the performance and convenience of the
DiversiLab system for discrimination of MRSA isolates in a microbiological diagnostic routine
setting and to compare evaluation criteria obtained with PFGE and MLST.
DiversiLab results were fully concordant with the CC classification as defined with MLST.
The discriminatory power of the DiversiLab system was comparable to that of MLST, while
PFGE was the most discriminatory technique.
The reproducibility of the DiversiLab system was excellent (>99%). The feasibility of the
DiversiLab system with respect to labour intensity and time-to-result was good. The most
labour-intensive step is the extraction, which requires almost half a working day hands-on

172

Rene te Witt bw.indd 172 24-11-11 10:35

 Summarizing discussion and future perspectives

time, depending on the number of isolates being tested. Total turn-around-time is one
working day with a maximum of 36 samples, which is less than PFGE (3 to 5 days) and MLST
(approximately 3 days). Costs of PFGE, MLST and DiversiLab are comparable and are approxi-
mately €50 per sample, respectively.
In summary, DiversiLab is a rapid and non labour-intensive technique, but it lacks resolu-
tion to differentiate genetically and epidemiologically unique MRSA strains, as is needed for
detailed outbreak analysis. Therefore, it may be useful as a library typing system for long-term
epidemiological studies.

The primary aim of the study described in Chapter 7 was to assess the performance of the
SpectraCellRA system (SCRA; River Diagnostics, Rotterdam, The Netherlands) for typing of an
epidemiologically perfectly defined set of MRSA isolates.
Results of SCRA analysis at the isolate level were concordant with the gold standard of epi-
demiology for 95.6% (108/113). However, when our results were analyzed at the household
cluster level, results were concordant for 90.7% (49/54). In other words, when using SCRA
in practice, SCRA gives discrepant results in 9% of the small clusters. Analysis of discrepant
results using PFGE showed equal results as epidemiological data. When the discrepant
households were analyzed for possibilities of exo-household acquisition of MRSA, no pos-
sible transmission of another known source or contact with a person with known risk factors
were found. This indicates that the discriminatory power of the SCRA system might be too
high for adequate outbreak analysis of small clusters.
During transmission and acquisition of MRSA, micro-evolution may take place (14-15).
PFGE may not be able to detect this micro-heterogeneity, where Raman spectroscopy does.
However, since a household can be considered as a close community, we can assume that
transmission and detection have occurred in a short time period. Therefore, it is very likely
that the discrepancy between SCRA and the gold standard is not due to micro-evolution of
the organism.
Multiple independent measurements of 3 reference isolates and the duplicate measure-
ments of 26 MRSA strains resulted in a reproducibility of 100%. This indicates that SCRA is
Chapter 10

stable over a longer period of time, which has been published before (16).
Our findings are similar to those found in another study, where the observed concordance
between SCRA and the gold standard of epidemiological data and PFGE was 97% (17). Repro-
ducibility in our study was better (100% vs. 95%). In both studies isolates were tested as full
biological replicates at different points in the study.
The feasibility of the SCRA system with respect to hands-on-time (~ 3 h) and time-to-result
(36-48 h) was good and better than PFGE. Time-to-result may be improved by applying 1
subculture instead of 2 subcultures. In this way, results will be available the next day, what is
enormous valuable for prevention control.

173

Rene te Witt bw.indd 173 24-11-11 10:35

 Chapter 10

Mistyping of strains by using the SCRA system for outbreak analysis is a risk. Isolates
epidemiologically assigned to one cluster could now be considered as unique isolates and
as a single event each, which has serious consequences for infection prevention measures.
Prevention measures to a single event, or 2 separate single events, are different than the
response on an outbreak (2 or more MRSA cases belonging to one cluster). For 90-95% of
cases a considerable amount of time for implementing prevention measures is gained. For
the remaining 5-10%, further analysis will be necessary.
Many studies on typing of MRSA focus on a small set of MRSA isolates and the relevant
epidemiological data of the isolates are often (partially) unknown. We had the unique op-
portunity to use the epidemiological relationships within households as the gold standard,
together with PFGE.
Although the high discriminatory power of SCRA may lead to discrepancies, and accompa-
nying consequences, in 5-10% of epidemiologically related MRSA clusters, we conclude that
the SpectraCellRA system is a highly reproducible, easy-to-use and fast typing platform that
can compete with the currently used typing techniques.

Quality control

Chapter 8 focused on the External Quality Assessment (EQA) of molecular diagnostics and
genotyping of MRSA. The primary aim of our EQA programmes was to assess the proficiency
of laboratories in the molecular detection and characterisation of MRSA strains. We concluded
that the molecular detection of MRSA in samples with high CFU counts is reliable, which
can and has been implemented in various laboratory settings with confidence. All tests per-
formed equally well. However, we also had to conclude that several of the currently applied
tests do not meet the requested quality criteria. The sensitivity of many tests is relatively low
and the specificity needs to be improved.
Pre-enrichment of clinical samples leads to concentrations of MRSA exceeding 109 CFU/ml,
which is higher than the concentrations of MRSA likely to be found in a patient sample, and
those in this EQA panel. In the panel, only inactivated cells were present. As a consequence,
pre-enrichment was not possible. This may have influenced the results of laboratories that
have designed their MRSA assay to be optimal after enrichment.
The incorrect results found in this study underscore the need for improved specificity of
molecular MRSA tests. Therefore, positive results should always be confirmed by a culture
method or a second molecular test. For laboratories with high false positivity rates or in
regions with low prevalence of MRSA such as The Netherlands, confirmation is necessary.
Of course, this will cause a delay in results. A correct balance between reliability and speed
needs to be found.

174

Rene te Witt bw.indd 174 24-11-11 10:35

 Summarizing discussion and future perspectives

We presented the first EQA for typing of MRSA strains. Clear differences in resolution were
observed between the datasets, which underscores the need for improved international
standardization.
In conclusion, the quality of molecular diagnostic tests still needs improvement and regular
quality control and international standardization for MRSA diagnostics and typing should be
mandatory for the years to come. To improve the performance and quality of overall molecu-
lar diagnostics, both laboratories and manufacturers should be encouraged to participate
in EQAs. EQA panels for detection and typing should also be developed for other important
(health-care associated) infectious agents, such as vancomycin resistant enterococci (VRE)
and extended spectrum β-lactamase (ESBL) producing Enterobacteriaceae.

Discussion and future perspectives

In Chapter 9, an overview was given on new diagnostic tools (matrix-assisted laser desorp-
tion/ionisation time-of-flight mass spectrometry [MALDI-TOF], electronic nose and Raman
spectroscopy) that are currently being developed and implemented for detection and typing
of bacteria in clinical laboratories. Next to this, the most promising candidates for multiplex
analysis were discussed.

Chapter 10

175

Rene te Witt bw.indd 175 24-11-11 10:35

 Chapter 10

Future perspectives

Current diagnostics of infectious diseases is based on conventional microscopy and culture.

Essentially, laboratories can perform simple tests on many clinical samples on daily basis,
using inexpensive culture media and simple techniques. This strategy also provides bacterial
isolates, which can be further characterized e.g. more detailed identification of the species
and analysis of their antibiotic susceptibility profiles and/or epidemiological typing for out-
break analysis. Anti-microbial susceptibility testing (AST) is still performed in a manual or, at
most, semi-automated way. These conventional techniques are time-consuming, delaying
proper treatment of the patient.
Molecular diagnostics (e.g. real-time PCR) can be of great value for clinical microbiological
diagnostics. The combination of excellent sensitivity and specificity, the low contamination
risk, the ease-of-use and short time-to-result, has lead to the prominent position of real-time
PCR in current diagnostics of infectious diseases. In contrast, there are drawbacks for PCR.
There is a chance of false-positive (low specificity, contamination) or false-negative (low
sensitivity, variability in the target, inhibition) results.

Point-of-care diagnostics

The last years we have seen an exponential rise in the amount of research and financial
support for the development of new diagnostic techniques, resulting in rapid tests for di-
agnostics of infectious diseases, so called point-of-care (POC) tests. Probably the greatest
success for rapid diagnostics to date is the Cepheid GeneXpert system (Cepheid Diagnostics).
The GeneXpert is a fully automated, closed system that automates sample preparation, DNA
amplification and detection. There are currently 11 commercial GeneXpert tests available, of
which the MRSA assay is the most notable and successful one (18). Total assay time for MRSA
detection on the GeneXpert is 75 minutes, with a total hands-on-time of approximately 2
minutes. However, despite the encouraging findings of clinical studies, the overall perfor-
mance of the GeneXpert MRSA assay might be improved. The manufacturer’s interpretation
is based on the application of a Ct cut-off value of 36 cycles, regardless of the presence or
absence of evidence of amplification. This may introduce false-negative results even when
tests demonstrate clear evidence of MRSA amplification, but Ct values exceed 36 cycles.
Conversely, false-positive results may also occur, where Ct values of <36 are obtained but
amplification curves fail to demonstrate an exponential rise. The inclusion of an equivocal
result for tests where the amplification data of the MRSA target DNA fail to support the Ct
cut-off result may be needed.

What feature and functions should the ideal POC test possess? The primary driver for the
development of such a test is shortened time-to-result. The duration of a test should be

176

Rene te Witt bw.indd 176 24-11-11 10:35

 Summarizing discussion and future perspectives

maximally one hour, including hands-on-time and analytical time. The device should be por-
table and the cost per test should be low (± €25 for developed countries, however far less for
developing countries!). The device should use a disposable cartridge and no pre-processing
of the specimen or user intervention once the assay has been started should be required. The
results generated should be clear and easy to understand. The performance characteristics
of a POC assay should be comparable with those of the laboratory-performed version. One
important consideration is the input volume of the specimen, which can affect the sensitivity.
This is especially important for infectious disease testing, where the numbers of pathogen
cells may be small. Accuracy and precision within and among devices is another factor that
requires consideration. Internal controls that are tested along with the target should be
included in every unit and will ensure that the test is performing correctly. The inclusion of
external controls should help ensure comparable performance among devices.
The technology should also be available to populations that do not have access to tradi-
tional laboratory testing, such as those in developing countries in e.g. Africa.

Automation

In order for rapid diagnostics to be fully accepted in clinical and POC settings, perhaps the
major challenge is automation. This will save hands-on-time and allow multiple samples to
be analyzed at once. It may reduce contamination of samples, thus reducing false-positive
results. Automation will permit more extensive standardization of tests and systems that
will open the diagnostic process for operation by less-skilled staff, increasing the number of
people capable to run diagnostic analyses.
In addition, automation of laboratory data processing is an important issue as well, since
optimal patient management depends on transfer of the result to the clinician. With the
introduction of multiple new technologies and equipments in the clinical microbiology
laboratory, laboratory information systems (LIS) need to be adapted and expanded. A good
LIS is essential to guarantee optimal clinical servicing.
Chapter 10

Conclusion

In line with the contrasting advantages and challenges listed in this thesis, the development
of POC diagnostics is an area that is a major focus of research to date. Yet, at the same time,
very few platforms are currently commercially available. In the next 5-10 years, we will observe
a shift in the diagnosis of infectious diseases towards rapid POC diagnostics. Tests developed
for research-use-only are currently being reformatted for integration onto novel diagnostic
platforms. High-throughput, full automation and short time-to-result will be combined with
little sample preparation and a large panel of tests. This will allow diagnosis of infectious

177

Rene te Witt bw.indd 177 24-11-11 10:35

 Chapter 10

diseases to occur faster with a lesser burden on currently overloaded clinical laboratories,
resulting in efficient treatment.
Furthermore, research needs to be expanded to simplify and speed up AST testing, which
is still one of the biggest challenges in the clinical microbiological laboratory.

178

Rene te Witt bw.indd 178 24-11-11 10:35

 Summarizing discussion and future perspectives

References
1. van Genderen, P. J., Hesselink, D. A. and Bezemer, J. M. (2008). Imported malaria is falling in Netherlands and
Europe. BMJ 337: a1026.
2. van Rijckevorsel, G. G., et al. (2010). Declining incidence of imported malaria in the Netherlands, 2000-2007. Malar
J 9: 300.
3. Brown, A. E., et al. (1992). Urinary neopterin in volunteers experimentally infected with Plasmodium falciparum.
Trans R Soc Trop Med Hyg 86(2): 134-6.
4. Biemba, G., et al. (2000). Markers of inflammation in children with severe malarial anaemia. Trop Med Int Health
5(4): 256-62.
5. Simon, L., et al. (2004). Serum procalcitonin and C-reactive protein levels as markers of bacterial infection: a
systematic review and meta-analysis. Clin Infect Dis 39(2): 206-17.
6. Chirouze, C., et al. (2002). Low serum procalcitonin level accurately predicts the absence of bacteremia in adult
patients with acute fever. Clin Infect Dis 35(2): 156-61.
7. Hoen, B. (2009). [Differentiating bacterial from viral meningitis: contribution of nonmicrobiological laboratory
tests] Diagnostic differentiel entre meningite bacterienne et meningite virale : apport des examens non microbi-
ologiques. Med Mal Infect 39(7-8): 468-72.
8. Pfafflin, A. and Schleicher, E. (2009). Inflammation markers in point-of-care testing (POCT). Anal Bioanal Chem
393(5): 1473-80.
9. Ip, M., et al. (2007). Value of serum procalcitonin, neopterin, and C-reactive protein in differentiating bacterial from
viral etiologies in patients presenting with lower respiratory tract infections. Diagn Microbiol Infect Dis 59(2): 131-6.
10. Fuchs, D., et al. (1984). Neopterin as an index of immune response in patients with tuberculosis. Lung 162(6):
337-46.
11. Hesselink, D. A., et al. (2009). Procalcitonin as a Biomarker for a Bacterial Infection on Hospital Admission: A Criti-
cal Appraisal in a Cohort of Travellers with Fever after a Stay in (Sub)tropics. Interdiscip Perspect Infect Dis 2009:
137609.
12. Davies, H. D., et al. (2004). Multicenter study of a rapid molecular-based assay for the diagnosis of group B Strep-
tococcus colonization in pregnant women. Clin Infect Dis 39(8): 1129-35.
13. El Helali, N., et al. (2009). Diagnostic accuracy of a rapid real-time polymerase chain reaction assay for universal
intrapartum group B streptococcus screening. Clin Infect Dis 49(3): 417-23.
14. Feil, E. J. (2004). Small change: keeping pace with microevolution. Nat Rev Microbiol 2(6): 483-95.
15. Aziz, R. K. and Nizet, V. (2010). Pathogen microevolution in high resolution. Sci Transl Med 2(16): 16ps4.
Chapter 10
16. Willemse-Erix, D. F., et al. (2009). Optical fingerprinting in bacterial epidemiology: Raman spectroscopy as a real-
time typing method. J Clin Microbiol 47(3): 652-9.
17. Wulf, M. W., et al. (2011). The use of Raman spectroscopy in the epidemiology of methicillin-resistant Staphylococ-
cus aureus of human- and animal-related clonal lineages. Clin Microbiol Infect.
18. Rossney, A. S., et al. (2008). Evaluation of the Xpert methicillin-resistant Staphylococcus aureus (MRSA) assay using
the GeneXpert real-time PCR platform for rapid detection of MRSA from screening specimens. J Clin Microbiol
46(10): 3285-90.

179

Rene te Witt bw.indd 179 24-11-11 10:35

Rene te Witt bw.indd 180 24-11-11 10:35
 Chapter 11
Nederlandse samenvatting

René te Witt

Rene te Witt bw.indd 181 24-11-11 10:35

Rene te Witt bw.indd 182 24-11-11 10:35
 Nederlandse samenvatting

Geschiedenis

Het leven is veranderd sinds de Nederlandse botanicus Anthonie van Leeuwenhoek (1632-
1723) de diversiteit en het bestaan van de microbiële wereld aantoonde door middel van de
ontdekking van de microscoop. Microscopisch bewijs leverde in de 19e eeuw ondersteuning
voor de “microbe-theorie”. In de jaren 80 van de 19e eeuw stelde Robert Koch zijn postulaten
op om te bepalen of een micro-organisme wel of niet de veroorzaker is van een ziekte. Sinds-
dien is het spectrum aan bacteriën, schimmels, virussen en parasieten exponentieel gegroeid
dankzij verbeterde kweektechnieken en de ontwikkeling van geavanceerde microscopische
technieken. Echter, de grootste stap voorwaarts sinds Van Leeuwenhoek is de ontdekking
van nucleïnezuren in 1871 door Miescher, die heeft geleid tot de ontdekking van DNA als
bron van genetische informatie en als basis voor de karakterisatie van een micro-organisme
in 1953 door Watson, Crick en Wilkins (1-2).

Diagnostiek van infectieziekten

Vandaag de dag zijn infectieziekten nog steeds een wereldwijd probleem. Infectieziekten
kunnen alomtegenwoordig zijn (tuberculose, cholera, malaria) of opkomen als jaarlijkse
epidemieën (norovirus, influenza, seizoensverkoudheid). Ze kunnen zich presenteren als in-
cidenteel (vogelgriep, SARS) of worden veroorzaakt door antibioticumresistente pathogenen
(methicilline-resistente Staphylococcus aureus (MRSA), Extended Spectrum Beta-Lactamase
(ESBL) producerende bacteriën, carbapenem-resistente bacteriën). Tot slot kunnen infectie-
ziekten ook voorkomen als pandemieën (AIDS, de recente H1N1 uitbraak).
Huidige diagnostiek van de meeste bacteriële infecties is nog steeds gebaseerd op con-
ventionele microscopie en kweek. Voordelen hiervan zijn de lage kosten, de grote capaciteit
en het gebruiksgemak. Op deze manier kunnen laboratoria honderden tot duizend monster
per dag verwerken, gebruikmakend van goedkope kweekmedia en simpele technieken.
Bovendien levert deze strategie ook bacteriële isolaten op die, indien nodig, verder kunnen
worden gekarakteriseerd. Denk hierbij aan antibioticum gevoeligheidsbepalingen (AST) en
typering voor uitbraakanalyse.
Echter, er zijn ook nadelen. Kweek is ongevoelig en het duurt lang voor er een definitieve
uitslag kan worden gegeven. Versnelling en verbetering van diagnostiek van infectieziekten
Chapter 11

leidt tot betere en gerichtere behandeling van patiënten. Aan versnelde identificatie wordt
momenteel veel onderzoek gedaan, maar het bepalen van antibioticum gevoeligheid gaat
nog steeds langzaam.
In dit proefschrift zijn nieuwe mogelijkheden voor diagnostiek onderzocht.

183

Rene te Witt bw.indd 183 24-11-11 10:35

 Chapter 11

Biomarkers

Koorts bij een patiënt kan zowel een ziekte betekenen die snel verergert en zelfs dodelijk
kan zijn (sepsis, meningitis), maar het kan ook een uiting zijn van een triviale infectie (Sal-
monellosis, griep) of zelfs van een steriele ontsteking (allergie). De arts moet binnen korte
tijd beslissen of de infectie (waarschijnlijk) wordt veroorzaakt door een bacterie, virus of
parasiet en of de patiënt voor behandeling moet worden opgenomen. Resultaten van
routine (microbiologische) laboratoriumtesten geven een eerste indicatie; bijvoorbeeld
een verhoogd aantal witte bloedcelen is een aanwijzing voor een bacteriële infectie en een
verlaagd aantal witte bloedcellen voor een virale infectie. Helaas kan de arts niet volledig op
dit soort bepalingen afgaan, omdat er ook bacteriële infecties zijn die geassocieerd worden
met een normaal tot laag aantal witte bloedcellen. Het gebruik van biomarkers zou snelle en
betrouwbare detectie van een echte infectie wellicht mogelijk maken.
Diverse studies hebben de relevantie van biomarkers om bacteriële en virale infecties bij
opname van een patiënt van elkaar te onderscheiden al beschreven. Procalcitonine zou een
goed voorspellende marker zijn voor bacteriële infecties (sepsis, longontsteking) (3-6), waar
neopterine voorspellend zou zijn voor virale infecties (7).
In het eerste gedeelte van hoofdstuk 3 hebben we de mogelijke bruikbaarheid van
procalcitonine en neopterine getest om bij opname van de patiënt ernstige malaria te on-
derscheiden van niet-ernstige malaria. Er zijn significant hogere waarden procalcitonine en
neopterine gevonden bij reizigers met ernstige malaria dan bij reizigers met niet-ernstige
malaria. Helaas zal de toepasbaarheid van beide testen beperkt zijn doordat beide testen
een slechte positief voorspellende waarde hebben en geen van beide testen in staat is om
met grote zekerheid de diagnose ernstige malaria te geven. Aan de andere kant suggereert
de hoge negatief voorspellende waarde van zowel procalcitonine als neopterine dat beide
testen van waarde kunnen zijn bij het uitsluiten van ernstige malaria, zeker wanneer deze
test beschikbaar is als een goedkope en snelle diagnostische test. Dit kan artsen die slechts
sporadisch hebben te maken met zieke reizigers helpen in het maken van de juiste beslissing
om mensen te behandelen met malariapillen of door te verwijzen naar een gespecialiseerd
ziekenhuis.
Het tweede gedeelte van hoofdstuk 3 behandelt de diagnostische relevantie van procal-
citonine - en neopterine waardes als biomarkers voor bacteriële of virale infectie bij reizigers
die ziek terugkomen uit de (sub)tropen. In deze studie vonden we een teleurstellend lage
diagnostische toepasbaarheid voor zowel procalcitonine als neopterine.

Microscopie

Detectie van humane pathogenen door middel van direct microscopisch onderzoek van het
klinisch monster is de oudste, maar nog steeds uiterst relevante, diagnostische methode

184

Rene te Witt bw.indd 184 24-11-11 10:35

 Nederlandse samenvatting

binnen de medische microbiologie. Veel verschillende, celwand-specifieke kleurmethodes

zijn beschikbaar (Gramkleuring, Ziehl-Neelsen etc.), waardoor samen met de vorm en lig-
ging van de bacterie een eerste indicatie kan worden gegeven van de aanwezigheid van
een mogelijke pathogeen. Echter, bijna altijd zal de definitieve identificatie worden gegeven
op basis van fenotypische (biochemische) eigenschappen van de bacterie na kweek op een
geschikt medium.

Kweek

Een conventionele en veelgebruikte diagnostische techniek is kweek op en in diverse vaste

en vloeibare media. Deze media kunnen universeel zijn (groei is mogelijk voor de meest
voorkomende klinisch relevante bacteriën: b.v. bloedagar, chocolade agar, Müller Hinton
bouillon), specifiek zijn (b.v. agar of bouillon speciaal voor Gram-positieve kokken of anaero-
be bacteriën) of zelfs speciespecifiek zijn (b.v. MRSA ophopingsmedium, chromogene agars)
door middel van het toevoegen van specifieke substraten, groeiremmers en/of bepaalde
antibiotica.
Kweekmethoden eisen tijd voordat de bacterie is gegroeid, over het algemeen 24-72 uur.
Dit resulteert in een vertraging van gerichtere behandeling van de patiënt. Voor veeleisende
of traaggroeiende bacteriën, zoals Mycobacterium tuberculosis, wordt deze tijd nog verder
verlengd, terwijl bacteriën die helemaal niet groeien op conventionele media en onder con-
ventionele condities, vanzelfsprekend, niet worden gevonden. Bovendien zullen logistieke
factoren de bacteriële groei verder beïnvloeden. Monsters moeten naar het laboratorium
worden vervoerd en gedurende dit transport kunnen monsters van samenstelling verande-
ren door bijvoorbeeld veranderingen in temperatuur. Dit kan een dramatisch effect hebben
op de bacteriële kwaliteit en kwantiteit.
Een voorbeeld van een bacterie die voornamelijk wordt gedetecteerd door middel van
kweek is de Groep B Streptokok (GBS). Gedurende de zwangerschap kan GBS verschillende
ernstige ziektes veroorzaken, bij zowel de moeder als het kind (8-9). In de meeste gevallen
krijgen neonaten van gekoloniseerde moeders (~35%) GBS tijdens de bevalling (10-12).
Om dragerschap van en infectie met GBS vast te stellen zijn er diverse kweektechnieken
ontwikkeld. Slechts weinig aandacht is echter besteed aan de mogelijke gevolgen van trans-
portcondities aangaande tijd, temperatuur en medium. Richtlijnen stellen dat GBS isolaten
Chapter 11

enkele dagen bij kamertemperatuur in leven blijven in Amies transport medium. Wel wordt
gedurende 1-4 dagen een vermindering in de opbrengst van GBS gezien van bijna 30%,
zeker bij verhoogde temperaturen (13-14). Zelfs wanneer geschikte transport media worden
gebruikt is de sensitiviteit van GBS kweek het meest optimaal wanneer het monster voor
kweek wordt bewaard bij 4°C en de kweek wordt ingezet binnen 24 uur na monsterafname
(8, 13-17).

185

Rene te Witt bw.indd 185 24-11-11 10:35

 Chapter 11

Een nieuw transport- en ophopingsmedium, de Granada Tube (GT) bouillon (bioMérieux)

is recentelijk geïntroduceerd. In dit medium wordt een oranje pigment gevormd in de aan-
wezigheid van GBS (18-19). In hoofdstuk 4 wordt de in vitro evaluatie van de performance
van deze bouillon voor de detectie van GBS beschreven. We hebben deze performance
vergeleken met die van het huidige aanbevolen transport medium, Amies transport medium
(AT). De gevoeligheid van GT was significant hoger dan die van AT, in het bijzonder voor lage
concentraties en verlengde transport/incubatietijden. GT is mogelijk niet bruikbaar voor de
directe detectie van GBS van klinische monsters, omdat bij afwezigheid van oranje pigment
niet direct geconcludeerd mag worden dat GBS afwezig is. Daarom zal elke GT altijd moeten
worden afgeënt voor een optimaal resultaat. Bovendien zullen GT altijd bij 37°C moeten
worden geïncubeerd om de gevoeligheid te verhogen.
De vraag is echter of kweek de toekomst heeft bij het zoeken naar GBS. Op PCR gebaseerde
methodes zijn gevoeliger voor de detectie van GBS (20). GeneXpert GBS (Cepheid) is een
snelle, op real-time PCR gebaseerde assay voor de detectie van GBS kolonisatie, met een
hoge sensitiviteit en specificiteit (21). DNA extractie en PCR zijn volledig geautomatiseerd en
resultaten zijn binnen 75 minuten beschikbaar.

Serologie

In diagnostische laboratoria worden microscopie en kweek vaak aangevuld met serologische

technieken voor (in)directe diagnostiek van infecties door het detecteren van antilichamen
of antigenen. Er zijn echter diverse tekortkomingen. Serologieresultaten kunnen onbetrouw-
baar zijn als gevolg van kruisreactiviteit. Bovendien, als gevolg van de noodzaak tot het
testen van een nieuw monster na 1-4 weken, zijn resultaten vaak te laat beschikbaar om
direct invloed te kunnen hebben op de behandeling van de patiënt.

Identificatie en antimicrobiële gevoeligheidsbepalingen

Na kweek van een micro-organisme, is identificatie de volgende stap. De meeste klinische

laboratoria gebruiken biochemische en immunologische identificatie testen. Echter, de resul-
taten hiervan kunnen worden beïnvloed door de testcondities, zoals kweektijd, -temperatuur,
- omstandigheden (b.v. met of zonder O2) en/of samenstelling van het kweekmedium. Geau-
tomatiseerde identificatie - en AST systemen worden tegenwoordig veelgebruikt binnen de
klinische bacteriologie. Deze systemen zijn gebaseerd op het gebruik van enzymatische en
biochemische eigenschappen van een bacterie voor de identificatie en verdunning voor AST.

186

Rene te Witt bw.indd 186 24-11-11 10:35

 Nederlandse samenvatting

Antibioticum resistentie

Penicilline resistentie in S. aureus verscheen zeer snel na het klinisch gebruik van penicilline in
1943 en het resistentiemechanisme was de productie van β-lactamase (22). Het eerste geval
van MRSA werd gerapporteerd in 1962, slechts 2 jaar na de introductie van methicilline ter
behandeling van penicilline resistente S. aureus. De originele MRSA stammen circuleerden
binnen gezondheidszorginstellingen en konden op deze manier resistent worden tegen een
breed scala aan antibiotica. Echter, sinds de jaren 90 van de vorige eeuw is er een stijging
waar te nemen in MRSA infecties in voorheen gezonde individuen zonder contact met de
gezondheidszorg. Deze MRSA stammen zijn niet ontsnapt uit het ziekenhuis, maar ontstaan
in de samenleving (23-25).
Echter, de grootste bedreiging in antimicrobiële resistentie wordt momenteel gevormd
door de snelle en globale verspreiding van multiresistente Gramnegatieve bacteriën. Er
worden regelmatig nieuwe β-lactamases en andere nieuwe resistentiemechanismen ge-
vonden die gemakkelijk van bacterie naar bacterie kunnen worden overgedragen. Labora-
toriumdetectie is moeilijk omdat meerdere resistentiemechanismen aanwezig kunnen zijn
en resistentie “verstopt” kan zijn. Het behandelen van ESBL producerende bacteriën met
carbapenem heeft geleid tot toenemende resistentie tegen carbapenem en de ontdek-
king van diverse carbapenemases. De in 2010 ontdekte New Delhi metallo-β-lactamase 1
(NDM-1) en de globale verspreiding hiervan is van grote zorg aangezien er zeer weinig effec-
tieve antibiotica zijn tegen deze zeer resistente bacteriën (26). Bovendien is de detectie van
carbapenemases problematisch doordat er verschillende methodes zijn, die verschillende
resultaten opleveren. Nucleïnezuur amplificatietechnieken zijn dan ook essentieel om deze
resistentiemechanismen te ontdekken.

Nucleïnezuur amplificatie (moleculaire diagnostiek)

Elk species bezit een unieke nucleïnezuur (DNA) vingerafdruk, die kan worden gebruikt als
voor de detectie en identificatie van micro-organismen in klinische monsters. Bovendien kan
op deze manier ook aanvullende informatie van dat micro-organisme, zoals resistentiegenen
en virulentiefactoren, worden gevonden. Duidelijke voordelen ten opzichte van kweek zijn
een sneller resultaat en verhoogde sensitiviteit en specificiteit.
Chapter 11

Echter, er bestaat een risico op foutpositieve (contaminatie of kruisreactiviteit) en fout-

negatieve (ongevoelige test, variabiliteit van het target of remming van de test) uitslagen.
Tevens vindt men alleen waarop men test.
Bij het invoeren van dit soort assays dient het hele proces (monsterafname, DNA extractie,
amplificatie en interpretatie van de resultaten) uitvoerig gevalideerd te worden. Bovendien
dienen positieve - en negatieve controles en remmingcontroles te worden getest. Daarnaast

187

Rene te Witt bw.indd 187 24-11-11 10:35

 Chapter 11

moet het monster worden voorbehandeld om een goede DNA opbrengst te kunnen garan-
deren en moeten alle reagentia regelmatig worden getest.
In hoofdstuk 5 wordt een overzicht gegeven van de laatste stand van zaken ten aanzien
van de moleculaire diagnostiek en typeren van MRSA.

Bacterieel typeren

Na detectie en identificatie van (antibioticum resistente) bacteriën, kan het noodzakelijk zijn
om deze te identificeren op stamniveau (ook bekend als typeren) om potentiële transmissie
vast te stellen. Bovendien is typeren geschikt om (inter)nationale verspreiding van bacteriën
aan te tonen. Tegenwoordig zijn er veel verschillende typeertechnieken beschikbaar in het
diagnostisch laboratorium.
In hoofdstuk 6 hebben we een gedeelte van een internationale collectie MRSA stammen
afkomstig uit 11 Europese landen (27) geanalyseerd met behulp van het DiversiLab systeem
(bioMérieux), pulsed-field gel electroforese (PFGE) en multi-locus sequence typing (MLST).
Het doel van deze studie was het bepalen van de performance en het gebruiksgemak van het
Diversilab systeem ten opzichte van de huidige gebruikte systemen.
We hebben geconcludeerd dat het Diversilab system een snelle typeertechniek is. De
resolutie, nodig voor uitbraakanalyse, is echter onvoldoende om genetisch en epidemiolo-
gisch unieke MRSA stammen te onderscheiden. Het Diversilab systeem is waarschijnlijk meer
bruikbaar voor lange-termijn studies.

Het doel van de studie in Hoofdstuk 7 was het bepalen van de performance en het gebruiks-
gemak van een andere typeertechniek, het SpectraCell systeem (SCRA; River Diagnostics),
voor het typeren van MRSA.
Op stamniveau kwamen de gevonden resultaten voor 95.7% overeen met de gouden stan-
daard. Echter, wanneer we onze resultaten analyseerden op het niveau van de huishoudens,
waren deze concordant voor 90.7%. Een indicatie dat het discriminerend vermogen van
SCRA wellicht te hoog is voor adequate uitbraak analyse.
De reproduceerbaarheid was 100%. Dit wijst erop dat het SCRA systeem over langere tijd
stabiel is. Onze resultaten komen overeen met eerdere publicaties (28-29).
De toepasbaarheid van het SCRA systeem met betrekking tot hands-on-time en de door-
looptijd voor het produceren van een uitslag is goed. Zowel de hands-on-time als de door-
looptijd kunnen nog worden verbeterd door te gaan werken met 1 overnachtkweek in plaats
van 2 overnachtkweken. Op deze manier zijn resultaten de volgende dag al beschikbaar. Dit
levert een enorme tijdswinst op ten opzichte van de huidige typeertechnieken.
Ondanks dat het hoge discriminerende vermogen van SCRA kan leiden tot discrepante
uitslagen in 5-10% van epidemiologisch gerelateerde clusters, met bijbehorende conse-
quenties op het gebied van infectiepreventie, concluderen we dat het SpectraCellRA system

188

Rene te Witt bw.indd 188 24-11-11 10:35

 Nederlandse samenvatting

een uiterst reproduceerbare, gemakkelijk te gebruiken en snelle typeertechniek is, die kan
concurreren met de huidige gebruikte technieken.

Kwaliteitscontrole

Voor het uitvoeren van kwalitatief hoogstaande moleculaire diagnostiek is kwaliteitscontrole

essentieel. Verschillende laboratoria kunnen verschillende assays gebruiken, met elk hun
eigen sensitiviteit en specificiteit. Een manier om hiermee om te gaan is door het verzorgen
van externe kwaliteitscontrole programma’s (EQA). Deze programma’s dienen goed gedefini-
eerd te zijn en te worden georganiseerd door onafhankelijke instituten.
In hoofdstuk 8 hebben we de resultaten beschreven van een EQA voor moleculaire diag-
nostiek van MRSA. We concluderen dat de moleculaire detectie van MRSA in monsters met
een hoge concentratie betrouwbaar is. Tevens concluderen we dat de detectie van MRSA
in monsters met een lage concentratie onvoldoende is en dat de specificiteit moet worden
verhoogd. Ophoping van klinische monsters leidt tot MRSA concentraties boven de 109
kolonievormende eenheden per ml, een concentratie hoger dan verwacht mag worden bij
een patiënt en veel hoger dan concentraties van dit EQA panel. In het rondgestuurde panel
zaten alleen geïnactiveerde stammen, waardoor ophoping niet mogelijk was. Dit zou een
mogelijke verklaring kunnen zijn voor de gevonden lage gevoeligheid. De gevonden foute
resultaten onderstrepen de noodzaak voor verbetering van de specificiteit van moleculaire
MRSA testen.
Om de kwaliteit van moleculaire diagnostiek te verbeteren zullen zowel laboratoria als
fabrikanten aangemoedigd moeten worden om deel te nemen aan EQA en zal internationale
standaardisatie ingesteld moeten worden. Daarnaast zullen EQA panels ook beschikbaar
moeten komen voor detectie van andere belangrijke (resistente) bacteriën, zoals de al eerder
genoemde ESBL producerende bacteriën.

Afhankelijk van de technologische en financiële mogelijkheden, zal een klinisch microbiolo-

gisch laboratorium de meest optimale diagnostische strategie voor de detectie van micro-
organismen moeten zoeken door conventionele technieken te combineren met moleculaire
diagnostiek. Hierbij zal een balans moeten worden gevonden tussen performance (sensi-
tiviteit, specificiteit) and gebruikersgemak (hands-on-time, kosten, etc.). Uiteindelijk zal dit
Chapter 11

leiden tot de meest optimale patiëntenzorg.

189

Rene te Witt bw.indd 189 24-11-11 10:35

 Chapter 11

Discussie en toekomstperspectief

In hoofdstuk 9 is een overzicht gegeven van nieuwe diagnostische technieken voor de de-
tectie en het typeren van bacteriën in klinische laboratoria (MALDI-TOF, elektronische neus
en Raman spectroscopie) die momenteel worden ontwikkeld.
Bovendien zijn in hoofdstuk 9 de belangrijkste kandidaten voor multiplex analyse (Lu-
minex, microarray en sequencing) bediscussieerd. Snelle multiplex analyse is de ultieme
diagnostische strategie. Met betrekking tot Luminex en microarray is het grootste probleem
dat beide technieken niet in staat zijn om nieuwe (varianten van) micro-organismen te
ontdekken. Het gebruik van beide systemen zal dan ook waarschijnlijk worden verdrongen
door de introductie van (totaal genoom) next generation sequencing. Richting het einde van
dit decennium is het waarschijnlijk dat next generation sequencing een veelvoorkomend
klinisch diagnostische techniek zal zijn, die door laboratoria direct op klinisch materiaal zal
worden ingezet.

Toekomstperspectief

Point-of-care diagnostiek

De laatste jaren hebben we een exponentiële stijging gezien in de hoeveelheid onderzoek

en financiële ondersteuning voor de ontwikkeling van nieuwe snelle diagnostische testen,
de zogeheten point-of-care (POC) testen.
Welke eigenschappen zou een ideale POC test moeten bezitten? De primaire drijfveer voor
de ontwikkeling van een dergelijke test is een snel resultaat. De totale tijd totdat het resultaat
bekend is, zal, ideaal gezien, maximaal 1 uur zijn. Het apparaat zal draagbaar moeten zijn en
de kosten per test moeten laag zijn (± €25 voor de “westerse wereld”, maar veel minder voor
de “3e wereld”). De uitslagen zullen duidelijk en gemakkelijk te begrijpen moeten zijn. De
performance karakteristieken van een POC test zullen vergelijkbaar moeten zijn met die van
de laboratoriumversie. Een belangrijke overweging zal het input volume van het monster
zijn, wat mogelijk gevolgen heeft voor de gevoeligheid. Nauwkeurigheid en precisie binnen
1 apparaat en tussen meerdere apparaten is een andere overweging. Het gelijktijdig met
het monster testen van interne controles zal moeten worden ingevoerd om te zorgen dat
elke test correct wordt uitgevoerd. Externe controles zullen moeten worden meegenomen
om de performance tussen apparaten te waarborgen. De technologie zal tevens beschikbaar
moeten zijn voor bevolkingsgroepen die geen of minder toegang hebben tot traditionele
laboratoriumtesten, zoals in ontwikkelingslanden in b.v. Afrika.

190

Rene te Witt bw.indd 190 24-11-11 10:35

 Nederlandse samenvatting

Automatisering

Om snellere diagnostiek als POC tests mogelijk te maken is automatisering een van de
grootste uitdagingen. Automatisering zal hands-on-time verlagen en maakt het mogelijk om
meerdere monsters tegelijk te testen. Bovendien zal automatisering leiden tot uitgebreidere
standaardisering, wat ervoor zorgt dat ook minder opgeleid personeel deze POC testen zal
kunnen uitvoeren. Op deze manier wordt het aantal mensen dat diagnostische testen kan
uitvoeren vergroot.
Automatisering van de resultaatverwerking is ook een belangrijke factor. Hoe sneller een
arts een resultaat heeft, hoe adequater en gerichter de patiënt behandeld kan worden. Met
de introductie van meerdere nieuwe technieken en apparaten, zullen laboratorium informa-
tie systemen (LIS) hiervoor geschikt moeten worden gemaakt. Een goed LIS is essentieel om
optimale klinische zorg te kunnen garanderen.

Conclusie

In overeenstemming met de contrasterende voordelen en nadelen zoals beschreven in dit

proefschrift, is de ontwikkeling van POC diagnostiek een speerpunt van huidig onderzoek.
Tegelijkertijd zijn er momenteel nog weinig testen commercieel verkrijgbaar. In de komende
5-10 jaar zullen we in de diagnostiek van infectieziekten een verschuiving zien richting snelle
POC diagnostiek. Dit zal het mogelijk maken om diagnostiek te versnellen, met efficiëntere
behandeling van patiënten als resultaat. Bovendien zal er veel onderzoek nodig zijn om AST
te versnellen en vereenvoudigen, wat nog steeds een van de grootste uitdagingen is binnen
de klinische microbiologie.

Chapter 11

191

Rene te Witt bw.indd 191 24-11-11 10:35

 Chapter 11

References.
1. Watson, J. D. and Crick, F. H. (1953). Genetical implications of the structure of deoxyribonucleic acid. Nature
171(4361): 964-7.
2. Watson, J. D. and Crick, F. H. (1953). Molecular structure of nucleic acids; a structure for deoxyribose nucleic acid.
Nature 171(4356): 737-8.
3. Chirouze, C., et al. (2002). Low serum procalcitonin level accurately predicts the absence of bacteremia in adult
patients with acute fever. Clin Infect Dis 35(2): 156-61.
4. Hoen, B. (2009). [Differentiating bacterial from viral meningitis: contribution of nonmicrobiological laboratory
tests] Diagnostic differentiel entre meningite bacterienne et meningite virale : apport des examens non microbi-
ologiques. Med Mal Infect 39(7-8): 468-72.
5. Pfafflin, A. and Schleicher, E. (2009). Inflammation markers in point-of-care testing (POCT). Anal Bioanal Chem
393(5): 1473-80.
6. Simon, L., et al. (2004). Serum procalcitonin and C-reactive protein levels as markers of bacterial infection: a
systematic review and meta-analysis. Clin Infect Dis 39(2): 206-17.
7. Ip, M., et al. (2007). Value of serum procalcitonin, neopterin, and C-reactive protein in differentiating bacterial from
viral etiologies in patients presenting with lower respiratory tract infections. Diagn Microbiol Infect Dis 59(2): 131-6.
8. Verani, J. R., et al. (2010). Prevention of perinatal group B streptococcal disease--revised guidelines from CDC,
2010. MMWR Recomm Rep 59(RR-10): 1-36.
9. Muller, A. E., et al. (2006). Morbidity related to maternal group B streptococcal infections. Acta Obstet Gynecol
Scand 85(9): 1027-37.
10. Bergseng, H., et al. (2007). Real-time PCR targeting the sip gene for detection of group B Streptococcus coloniza-
tion in pregnant women at delivery. J Med Microbiol 56(Pt 2): 223-8.
11. Valkenburg-van den Berg, A. W., et al. (2006). Prevalence of colonisation with group B Streptococci in pregnant
women of a multi-ethnic population in The Netherlands. Eur J Obstet Gynecol Reprod Biol 124(2): 178-83.
12. Campbell, J. R., et al. (2000). Group B streptococcal colonization and serotype-specific immunity in pregnant
women at delivery. Obstet Gynecol 96(4): 498-503.
13. Rosa-Fraile, M., et al. (2005). Specimen storage in transport medium and detection of group B streptococci by
culture. J Clin Microbiol 43(2): 928-30.
14. Stoner, K. A., Rabe, L. K. and Hillier, S. L. (2004). Effect of transport time, temperature, and concentration on the
survival of group B streptococci in amies transport medium. J Clin Microbiol 42(11): 5385-7.
15. Schrag, S., et al. (2002). Prevention of perinatal group B streptococcal disease. Revised guidelines from CDC.
MMWR Recomm Rep 51(RR-11): 1-22.
16. Ostroff, R. M. and Steaffens, J. W. (1995). Effect of specimen storage, antibiotics, and feminine hygiene products on
the detection of group B Streptococcus by culture and the STREP B OIA test. Diagn Microbiol Infect Dis 22(3): 253-9.
17. Hakansson, S., et al. (2008). Group B streptococcal carriage in Sweden: a national study on risk factors for mother
and infant colonisation. Acta Obstet Gynecol Scand 87(1): 50-8.
18. Martinho, F., et al. (2008). Evaluation of liquid biphasic Granada medium and instant liquid biphasic Granada
medium for group B streptococcus detection. Enferm Infecc Microbiol Clin 26(2): 69-71.
19. Heelan, J. S., et al. (2005). Evaluation of a new selective enrichment broth for detection of group B streptococci in
pregnant women. J Clin Microbiol 43(2): 896-7.

192

Rene te Witt bw.indd 192 24-11-11 10:35

 Nederlandse samenvatting

20. Davies, H. D., et al. (2004). Multicenter study of a rapid molecular-based assay for the diagnosis of group B Strep-
tococcus colonization in pregnant women. Clin Infect Dis 39(8): 1129-35.
21. El Helali, N., et al. (2009). Diagnostic accuracy of a rapid real-time polymerase chain reaction assay for universal
intrapartum group B streptococcus screening. Clin Infect Dis 49(3): 417-23.
22. Kirby, W. M. (1944). Extraction of a Highly Potent Penicillin Inactivator from Penicillin Resistant Staphylococci.
Science 99(2579): 452-3.
23. McDougal, L. K., et al. (2003). Pulsed-field gel electrophoresis typing of oxacillin-resistant Staphylococcus aureus
isolates from the United States: establishing a national database. J Clin Microbiol 41(11): 5113-20.
24. Tenover, F. C. and Goering, R. V. (2009). Methicillin-resistant Staphylococcus aureus strain USA300: origin and
epidemiology. J Antimicrob Chemother 64(3): 441-6.
25. Tenover, F. C., et al. (2008). Characterization of Staphylococcus aureus isolates from nasal cultures collected from
individuals in the United States in 2001 to 2004. J Clin Microbiol 46(9): 2837-41.
26. Kumarasamy, K. K., et al. (2010). Emergence of a new antibiotic resistance mechanism in India, Pakistan, and the
UK: a molecular, biological, and epidemiological study. Lancet Infect Dis 10(9): 597-602.
27. Cookson, B. D., et al. (2007). Evaluation of molecular typing methods in characterizing a European collection of
epidemic methicillin-resistant Staphylococcus aureus strains: the HARMONY collection. J Clin Microbiol 45(6): 1830-
7.
28. Willemse-Erix, D. F., et al. (2009). Optical fingerprinting in bacterial epidemiology: Raman spectroscopy as a real-
time typing method. J Clin Microbiol 47(3): 652-9.
29. Wulf, M. W., et al. (2011). The use of Raman spectroscopy in the epidemiology of methicillin-resistant Staphylococ-
cus aureus of human- and animal-related clonal lineages. Clin Microbiol Infect.

Chapter 11

193

Rene te Witt bw.indd 193 24-11-11 10:35

Rene te Witt bw.indd 194 24-11-11 10:35
 Appendices

Rene te Witt bw.indd 195 24-11-11 10:35

Rene te Witt bw.indd 196 24-11-11 10:35
 Dankwoord

Rene te Witt bw.indd 197 24-11-11 10:35

Rene te Witt bw.indd 198 24-11-11 10:35
 Dankwoord

Zo, dan kan ik nu iets gaan schrijven wat iedereen als eerste (en vaak ook als enige) leest bij
ontvangst van een proefschrift.

De afgelopen 4 jaar heb ik hulp gehad van veel mensen en uiteraard wil ik iedereen bedanken
die op enige wijze heeft bijgedragen aan het tot stand komen van dit boekje. Vanzelfspre-
kend heeft de een wat meer bijgedragen dan de ander en de volgende personen wil ik in het
bijzonder noemen.

Mocht u vinden dat uw naam ontbreekt, dan kunt u dit simpel oplossen door uw naam
hieronder in te vullen op de stippellijn:

Beste …………………………………………………………………………………., bedankt!

199

Rene te Witt bw.indd 199 24-11-11 10:35

 Dankwoord

Prof.dr.dr. van Belkum, beste Alex. Bedankt voor de kans die je mij hebt gegeven. De op-
dracht was simpel: schrijf in 4 jaar tijd 5 artikelen, klop er een nietje door en klaar! Zoals vaak
verschillen theorie en praktijk van elkaar, maar het eindresultaat is er! En ja, het staat in elk
dankwoord, maar je snelle en humoristische manier van overleg en beoordelen van artikelen
is echt ongeëvenaard en, net als je muzieksmaak, zeer gewaardeerd. Zelfs nu je niet meer
bij ons werkt en voor een commercieel bedrijf de hele wereld rondreist, is je betrokkenheid
nog net zo hoog, bewonderenswaardig! Ik mis de ochtendverwelkoming met harde klanken
van Rammstein, Muse of Madrugada, net als de legendarische “JA!” na een klop op en “Mot
je’n bakkie?” na openen van je deur! Ik gok en hoop dat we in de toekomst nog meer zaken
kunnen doen.

Dr. van Leeuwen, beste Willem. Ik heb veel van je geleerd, op wetenschappelijk gebied, maar
ook op het sociale vlak. Vanaf het begin klikte het uitstekend en ik denk dat ik voor mijn
persoon geen betere co-promotor had kunnen hebben. Je hebt me de ruimte en vrijheid
gegeven om rustig te werken aan dit boekje, zonder ook maar een moment het onderzoek
(en mij) uit het oog te verliezen. Het is duidelijk dat je bent “opgeleid” door Alex, want ook jij
reageert zeer snel op e-mails en verzoeken om artikelen te beoordelen! Wel jammer dat je
qua muzieksmaak “wat” bent achtergebleven. Bedankt voor alle wijze lessen!

Prof.dr. Verbrugh. Bedankt voor het plaatsnemen als secretaris in de kleine commissie en het
beoordelen van mijn proefschrift. Ik ben blij dat ik de opleiding tot MMM kan volgen om
zo verder te kunnen werken aan innovatieve (moleculaire) diagnostiek binnen de afdeling
MMIZ.

Prof.dr. Hooijkaas. Bedankt voor het plaatsnemen in de kleine commissie en voor het snel
beoordelen van mijn proefschrift.
Prof.dr. Savelkoul, beste Paul. Bedankt voor het plaatsnemen in de kleine commissie en voor
de snelle beoordeling van mijn proefschrift.

Prof.dr. Ieven, Prof.dr. Hermans en Prof.dr. Endtz. Dank voor het plaatsnemen in de grote
commissie.
Dr. Claas, beste Eric. Bedankt voor het plaatsnemen in de grote commissie. Ik kijk uit naar
onze samenwerking en de gevulde koeken in het kader van mijn MMM-opleiding.

Heren Reid, Geleijnse en van Tol. Geweldig dat jullie “Fokke & Sukke” belangeloos ter beschik-
king hebben gesteld. Bedankt!

200

Rene te Witt bw.indd 200 24-11-11 10:35

 Dankwoord

Alle co-auteurs, dank voor de samenwerking en jullie bijdrage aan dit boekje.
Anouk, bedankt voor je bijdrage aan en de lessen over groep B streptokokken.
Perry, bedankt voor de prettige samenwerking. Ik heb veel van je geleerd over het schrijven
van een artikel. Hopelijk kunnen we de samenwerking voortzetten.
Pieter, bedankt voor je kijk op microbiologische diagnostiek en je hulp bij de biomarkers.
William, thank you for the very pleasant cooperation. I hope that we can continue this in the
future.
Greet, bedankt voor je kritische blik en je hulp bij het tot stand komen van het laatste artikel.
Sybren, nieuwe kamergenoot. Dank voor je hulp bij het laatste artikel. Enne, een kamerinde-
ling op basis van muzieksmaak is zo gek nog niet!

Kees, kleine man. Jouw enthousiasme voor Raman heeft mij overtuigd. Ik heb veel van je
geleerd over het doen van onderzoek, Raman spectroscopie en bioinformatica, waarvoor
dank. Ik ga ervan uit dat we blijven samenwerken.
Diana. Had ik je toch bijna ingehaald! Bedankt voor de prettige samenwerking die we als
vanzelfsprekend zullen voortzetten.
Femke, bedankt voor je hulp op het lab.
Jan-willem, intussen heb je een andere richting gekozen, maar het Raman werk heb je niet
kunnen loslaten. Jij hebt mij leren werken met Raman, waarvoor dank, ook voor de bakkies
pleur.

Peter, Hans, Frans en Theo, bedankt voor de gedegen microbiologische basis die jullie mij
hebben geleerd.

Een promovendus kan niet zonder een kudde stageschapen, zeker niet als ze naast pipette-
ren ook zelf kunnen denken. Bouchra, Ersilia, Fenny, Koen, Marloes, Mitchell, Sutha en Vikash,
bedankt voor jullie enorme inzet en fraaie resultaten!

Guus en Martin, Sjef, Leendert, Marcel, Gerwin en Bas, bedankt voor de prettige samenwer-
king.

Anne, we zijn ongeveer tegelijk gestart met ons promotie-onderzoek. Leuk dat we elkaar nog
regelmatig zien bij de WMDI en congressen. Veel succes met je eigen promotie!

“Nieuwe” collega’s:
Norbert, van af en toe eens overleggen tot een van de belangrijkste collega’s. Fijn om te
merken dat we op 1 lijn zitten. Bedankt voor je hulp. Ik verwacht nog veel van je te kunnen
leren!
Jaap, bedankt voor de prettige samenwerking en het delen van je parasitologische kennis.

201

Rene te Witt bw.indd 201 24-11-11 10:35

 Dankwoord

Liesbeth en Willemien, intussen zijn we goed op elkaar ingespeeld. Ik hoop dat ik nog lang
met jullie mag werken om samen het niveau van onze moleculaire diagnostiek naar een
hoger peil te tillen. Bedankt voor jullie inzet!
Angelique, Esther, Nathalie, Renske, Ilona, Marlize, bedankt voor jullie inzet en de fijne sa-
menwerking.
Ad, bedankt voor je open deur en de prettige samenwerking.
Liset, dank voor je geduld als ik weer eens langskom met een LIMS vraag of kom zeuren over
stammen.
Marieke, jij bent vreemd genoeg de enige die mijn zangkwaliteiten waardeert tijdens de
jaarlijkse karaoke; wanneer doen we weer een duet?

“Oude” collega’s:
Corné, onze master of Luminex. Bedankt voor het bierhalen. Snel maar weer een game+pizza-
avond organiseren?
Gerjo, eindelijk ook promovendus. Bedankt voor de magazijnsamenwerking en succes met
je onderzoek!
Suzanne, ik ken maar weinig mensen die zo sociaal zijn als jij. Bedankt voor het omzetten
hiervan in een enorme hoeveelheid vrijdagmiddagborrels.
Patricia, bedankt voor de espresso’s. Heel veel succes met je eigen laatste loodjes.
Jasper, jouw humor is een welkome aanwinst voor de afdeling gebleken. Wat heet!
Irma, bedankt voor je hulp tijdens mijn overstap van de 2e naar de 3e.
Astrid, bedankt voor de fietstochtjes naar huis. Succes met je eigen promotie.
Bibi, Jeannine, Jurriaan, Marcel, Martijn, Rogier, Sanne, Saskia, Wendy Ka, Wendy Kl, veel suc-
ces verder met jullie promotie-onderzoek.
Deborah, bedankt voor je labhulp en labkennis.
Denise, Hélène, John, Kimberley, Marian tK, Mehri, Nicole, Susan, Wendy vdS, Willem vW,
bedankt voor de goede tijd.
Carla, zonder jou zou de unit R&D hopeloos verloren zijn. Bedankt voor al je hulp en je Rot-
terdamse nuchterheid.
Ger, bedankt voor je “keukenhulp”.
Marian H, bedankt voor je hulp op het administratieve vlak.

Beste Adrie, Kees en Martin, bedankt voor de hulp met onze 3 kinderen en voor de interesse
in mijn onderzoek de afgelopen jaren.

Naat, Huis, Sas en Chris, we zien elkaar te weinig, maar echte vrienden blijven vrienden! Huis
en Chris, ook voor jullie geldt dat we snel weer een game-avond moeten regelen.

202

Rene te Witt bw.indd 202 24-11-11 10:35

 Dankwoord

Michiel, drie jaar geleden kwam je hier werken en vanaf het begin klikte het uitstekend;
zal wel iets met humor, wetenschap en veel nerd-punten zijn. Je algemene kennis in het
algemeen en je kennis van South Park in het bijzonder is ongeëvenaard!

Martijn, bedankt voor 25 jaar vriendschap. Op naar de volgende 25! Ik zou me geen prettiger
iemand naast me kunnen voorstellen.

Lieve Xandra, Martijn, Marjo, Micha, Krista, Gijs en Dorien. Ik hoef jullie niet te vertellen hoe
belangrijk jullie zijn in mijn leven. Hopelijk is er na deze drukke maanden weer meer tijd
voor onze beroemde etentjes, weekenden-weg en kaasfondue-, wijn- en discussieavonden.
Bedankt voor jullie steun en liefde.

Lieve Dirk-Jan, ondanks dat je dit niet leest, wil ik je toch noemen. Zonde dat een promotie
jou nooit is gegund. Bedankt voor je inspirerende vriendschap. Ik denk nog vaak aan je.

Lieve papa en mama, ondanks dat jullie waarschijnlijk maar een klein beetje doorhebben
waar ik mee bezig ben, hebben jullie me altijd gesteund. Dank, dank, dank! Jullie hebben
het heel goed gedaan (en doen het nog steeds uitstekend) met 3 van zulke kinderen en 3
kleinkinderen!

Lieve Monique en Saskia, fijn om 2 van zulke zusjes te hebben. Jullie verblijfplaats vertoont
momenteel een omgekeerd evenredig verband. Mo, leuk dat je nu nog dichter bij woont.
Kom maar vaak langs! Sas, in Singapore samen met Michel. Veel succes daar en we komen
snel langs!

Lieve Lanne, Ymke en Jinte. Jullie zijn misschien wel het mooiste dat deze periode heeft
voortgebracht. Bedankt voor jullie (pogingen tot) geduld als papa weer zat te schrijven.
Vanaf nu heb ik weer echt tijd om met jullie te spelen.

Lieve Hedwig. Het is af! Bedankt voor je laatste duwtje om aan dit promotietraject te begin-
nen. Bedankt voor je steun en begrip, zeker deze laatste maanden. Bedankt voor onze eigen
K3. Luv you!

203

Rene te Witt bw.indd 203 24-11-11 10:35

 Curriculum vitae

Curriculum vitae

René te Witt was born in Rotterdam, on December 16, 1975. He passed his secondary school
(VWO) exam in 1994 at the Scholengemeenschap Blaise Pascal in Spijkenisse. From 1994 to
1998 he studied Medical Microbiology at the Hogeschool Rotterdam & Omstreken where he
graduated for his bachelor degree (Bsc/Ing) on a project studying Enterococcus faecalis and
Enterococcus faecium. In 1998 he started working at the Department of Medical Microbiology
and Infection prevention at the Sint Franciscus Gasthuis in Rotterdam. During his nearly 9
years at the SFG he participated in routine microbiological diagnostics and specialized in
molecular diagnostics, serology, parasitology and Mycobacterium tuberculosis.
In 2007 he moved to the Department of Medical Microbiology and Infectious Diseases,
Unit Research & Development at the Erasmus MC in Rotterdam, where he started his PhD
project on “Clinical Microbiological Diagnostics 2.0” under the supervision of Prof.dr.dr. A. van
Belkum and Dr. W. van Leeuwen.
After obtaining his registration for Medical Microbiological Researcher (Medisch Micro-
biologisch Onderzoeker; SMBWO), René will start his training for the Medical Molecular
Microbiologist (Medisch Moleculair Microbioloog) registration in March 2012.

Curriculum vitae

204

Rene te Witt bw.indd 204 24-11-11 10:35

 List of publications

List of publications

te Witt, R., P.M. Oostvogel, R. Yahiaoui, Y. Wu, A. van Belkum, and A.E. Muller. 2011. In vitro
evaluation of the performance of Granada selective enrichment broth for the detection of
group B streptococcal colonization. Eur J Clin Microbiol Infect Dis.

te Witt, R., A. van Belkum, W.G. MacKay, P.S. Wallace, and W.B. van Leeuwen. 2010. External
quality assessment of the molecular diagnostics and genotyping of meticillin-resistant
Staphylococcus aureus. Eur J Clin Microbiol Infect Dis 29:295-300.

te Witt, R., M.E. van Wolfswinkel, P.L. Petit, J.J. van Hellemond, R. Koelewijn, A. van Belkum,
and P.J. van Genderen. 2010. Neopterin and procalcitonin are suitable biomarkers for exclu-
sion of severe Plasmodium falciparum disease at the initial clinical assessment of travellers
with imported malaria. Malar J 9:255-62.

te Witt, R., A. van Belkum, and W.B. van Leeuwen. 2010. Molecular diagnostics and genotyp-
ing of MRSA: an update. Expert Rev Mol Diagn 10:375-80.

te Witt, R, W.B. van Leeuwen, and A. van Belkum. 2010. Specific diagnostic tests for atypical
respiratory tract pathogens. Infect Dis Clin N Am 24:229-48.

te Witt, R., V. Kanhai, and W.B. van Leeuwen. 2009. Comparison of the DiversiLab system,
Pulsed-Field Gel Electrophoresis and Multi-Locus Sequence Typing for the characterization of
epidemic reference MRSA strains. J Microbiol Methods 77:130-3.

Goessens, W.H., P. de Man, J.G. Koeleman, A. Luijendijk, R. te Witt, H.P. Endtz, and A. van
Belkum. 2005. Comparison of the COBAS AMPLICOR MTB and BDProbeTec ET assays for
detection of Mycobacterium tuberculosis in respiratory specimens. J Clin Microbiol 43:2563-6.

van Den Braak, N., A. van Belkum, D. Kreft, R. te Witt, H.A. Verbrugh, and H.P. Endtz. 1999.
The prevalence and clonal expansion of high-level gentamicin-resistant enterococci isolated
from blood cultures in a Dutch university hospital. J Antimicrob Chemother 44:795-8.

van den Braak, N., A. van Belkum, R. te Witt, H.A. Verbrugh, and H.P. Endtz. 1998. Glycopeptide
resistance amongst Staphylococcus spp. J Antimicrob Chemother 42:673-5.

List of publications

205

Rene te Witt bw.indd 205 24-11-11 10:35

 PhD Portfolio

PhD Portfolio

Name PhD student: René te Witt

Erasmus MC Department: Medical Microbiology and Infectious Diseases
PhD period: September 2007 – December 2011
Promotor: Prof.dr.dr. Alex van Belkum
Co-promotor: Dr. Willem B. van Leeuwen

PhD training Year ECTS

In-depth courses
• Quality and troubleshooting with real-time PCR 2008 1.1
• Molecular Diagnostics 2008 0.6
• Molecular Microbiology of Infectious Diseases 2008 0.6
• Biomedical Research Techniques 2008 0.1
• Epidemiology 2009 0.1
• Primer and probe design 2010 0.6
• Molecular Typing Methods 2011 0.8
• “Succesvol gebruik van moleculaire diagnostiek” 2011 1.4

Seminars and workshops

• Department Journal Clubs 2007-2011 4
• Department Research Meetings 2007-2011 4
• PhD Day Erasmus MC 2008, 2011 1
• Research Day Dept. MMID 2008-2010 1

National and international conferences

• EMMD, Scheveningen 2007 0.6
• ECCMID, Barcelona, Spain 2008 1
• ASM, Philadelphia, USA 2009 1
• EMMD, Scheveningen 2009 0.6
• ECCMID, Vienna, Austria 2010 1
• Scientific Spring Meeting NVMM, Papendal 2010 1
• IMMEM, Wernigerode, Germany 2010 1
• EMMD, Scheveningen 2011 1

Presentations and lecturing

• Poster presentation, ECCMID, Barcelona 2008 0.5
• Poster presentation, ASM, Philadelphia 2009 0.5
• Poster presentation (2x), EMMD, Scheveningen 2009 1

206

Rene te Witt bw.indd 206 24-11-11 10:35

 PhD Portfolio

• Poster presentation, Mol Med Day, Rotterdam 2009 0.5

• Lecture, medical students, Rotterdam 2010, 2011 0.5
• Oral presentation (2x), ECCMID, Vienna 2010 1
• Poster presentation, ECCMID, Vienna 2010 0.5
• Oral presentation, bioMérieux symposium, Kaatsheuvel 2010 0.5
• Oral presentation, IMMEM, Wernigerode 2010 0.5
• Lecture (3x), Hogeschool Leiden, Leiden 2010 0.5
• Oral presentation, STAR laboratory symposium, Rotterdam 2010 0.5
• Oral presentation, HET instrument, Amsterdam 2010 0.5
• Oral presentation, Mol Med Day, Rotterdam 2011 0.5
• Oral presentation, NVML symposium, Ede 2011 0.5
• Poster presentation, EMMD, Scheveningen 2011 0.5
• Lecture, Hogeschool Leiden, Leiden 2011 0.5
• Oral presentations, Dpt. MMID, Rotterdam 2008-2011 2

Scientific meetings
• WMDI meeting, Utrecht 2007-2011 3
• Mol Med Day, Rotterdam 2009 0.3
• Roche symposium, Utrecht 2009 0.3
• Microbial typing symposium, Utrecht 2009 0.3
• Mol Med Day, Rotterdam 2011 0.3
• “2e Nederlandse Infectieziektendag”, Zeist 2011 0.3
• Cepheid symposium, Leiden 2011 0.3
• BD symposium, Eindhoven 2011 0.3

Supervision of students
• Supervision of Bachelor of Science students 2007-2011 34
• Supervision of medical students Dermatology 2009, 2011 4
• Supervision of medical students “Vaardigheidsonderwijs” 2008, 2010, 2011 3

Other activities
• 3 months “Internship” molecular diagnostics, MMID 2009 12

207

Rene te Witt bw.indd 207 24-11-11 10:35

Rene te Witt bw.indd 208 24-11-11 10:35