Sexually Transmitted Diseases 0025-7125/90 $0.00 + .

20

Approach to the Patient with Genital

U leer Disease

George P. Schmid, MD*

Disruption of the skin in the genital area is a common presentation of

patients who believe that they have a sexually transmitted disease (STD).
The diagnostic steps taken to evaluate such lesions, generally described as
ulcers, are fairly standard, as is the differential diagnosis. In practice,
however, the definitive etiology of such genital ulcer disease often remains
unknown for a variety of reasons: all or some of the appropriate diagnostic
tests are not available, the appropriate tests are available but not utilized,
diagnostic tests are utilized but the lesions are too advanced for accurate
testing, or the etiology of the lesions is not among the diseases tested for.
As a result of these problems, physicians often must make a clinical
diagnosis, and clinical diagnoses of genital ulcers are consistently incorrect
in 40% of cases. 6, 11, 25 Accurate diagnosis requires consideration of not only
the clinical characteristics of each case, but also the epidemiologic and
demographic features and, most important of all, laboratory investigation
of each case.

WHAT IS A GENITAL ULCER?

Stedman's Medical Dictionary defines an ulcer as "a lesion on the

surface of the skin or a mucous surface, caused by superficial loss of tissue,
usually with inflammation. A wound with superficial loss of tissue from
trauma is not primarily an ulcer . . . . "39 In published reports of patients
with genital ulcers, however, only rarely is a "genital ulcer" defined (Table
1). Consequently, it is difficult to compare the distribution of diagnoses
among reported series of genital ulcers because investigators do not clearly
state their criteria for study entry. Because genital lesions may appear
trivial regardless of etiology, the clinician should consider any "genital
lesion that (is) denuded of the normal epithelium"25 to be an infectious
genital ulcer.

·Chief, Clinical Research Branch, Division of STD/HIV Prevention, Center for Prevention
Services, Centers for Disease Control; and Clinical Associate Professor of Medicine,
Morehouse School of Medicine, Atlanta, Georgia

Medical Clinics of North America-Vo!' 74, No. 6, November 1990 1559

1560 GEORGE P. SCH:\1ID

Table 1. Diseases Characterized by Genital Ulcerations in

Industrialized Countries
Common "Classic" Genital Ulcer Diseases
Syphilis
Genital herpes
Chancroid
Rare "Classic" Genital Ulcer Diseases
Lymphogranuloma venereum
Donovanosis (Granuloma inguinale)
Infrequent Causes of Genital Ulcers
Abrasions and trauma
Fixed drug eruption
Carcinoma
Reiter's syndrome
Infectiou; mononucleosis
Behget's syndrome
Aphthous ulcers
Other microorganisms

REVIEW OF CAUSES OF GENITAL ULCERS

Frequent Causes of Genital Ulcers. Five diseases are "classically"

associated with genital ulcerations (see Table 1) and there are distinct
differences between the developing and industrialized worlds in the fre-
quency of occurrence of these diseases (Table 2). In the developing world,
chancroid is often the most common disease, while in the industrialized
world genital herpes is the most common disease associated with genital
ulcers, and chancroid is frequent only in selected areas. Syphilis is common
in both, but constitutes relatively fewer cases in the developing world.
Lymphogranuloma venereum is not a common disease anywhere, whereas
donovanosis is frequently encountered only in parts of Asia. Important in
both areas is the group, about 5% to 10% of all cases, with coinfection by
more than one pathogen.
No etiology is found in one fourth to one half of all cases of any series.
Most of these are probably cases of syphilis, genital herpes, or chancroid
for which diagnostic tests are falsely negative. These false-negative results
reflect the inadequacy of our current diagnostic tests or the way they are
performed, and sometimes the effect of self-therapy by patients.
Infrequent Causes of Genital Ulcers. Infrequently, other causes of
genital ulcers are found. These include abrasions, fixed drug eruptions,
carcinoma, Reiter's syndrome, infectious mononucleosis, and Beh<;et's syn-
drrome. Unusual organisms have been isolated from genital ulcers, and it
is possible that these and yet-to-be-identified organisms are causes of
ulcers.l7· 40, 43 Numerous organisms causing other genital tract infections
have been recovered from genital ulcers, including Trichorrwnas vaginalis,25
Neisseria gonorrhoeae, 5, 8, 24 Chlamydia trachomatis, II Mycoplasma hom-
inis,5 and Ureaplasma urealyticum. 5, 40 Generally, these same organisms
were isolated elsewhere, and their appearance in the ulcer probably
represents contamination.
 Table 2. Frequency of Diagnosis in Patients with Genital Ulcers, by Percentage, in Indidduals in Developing or Industrialized Countries
DEvELOPING COUI\TRIES

Durban' Nairobi24 Nairobi" Bangkokt 41

DIAGI\OSIS (n = 100 Men) (n = 100 Men) (n = 100 Women) (n = 120 Men)

Syphilis 35 7 12* 1
Genital herpes 6 6 2 10
Chancroid 30 45 48 36
Lymphogranuloma venereum 1
Oonovanosis
More than one 11 6 1 2
Trauma PNS PNS 1 PNS
Other 1 2 2
Unknown 16 36 33 50

INDUSTRIAL COUNTRIES

Amsterdam40 WinnipeglO Edmonton" Detroit'

(n = 46; 39 Men) (n = 28; 25 Men) (n = 73 Men; 41 Women) (n = 100 Men)
Syphilis 26 32 36 20 17
Genital herpes 15 21:j: 38 76 34
Chancroid 22 4 2
Lymphogranuloma venereum
Oonovanosis 4
More than one 7
Trauma 9 (?) 4 P:'-JS PNS 7
Other 1
Unknown 22 36 26 5 40
Abbreviation: PNS, probably not sought.
*Eight were ulcerated condylomata lata.
tSixty percent of all patients had recently taken antimicrobials .
...... :j:Patients with vesicles were excluded.
Vl
~
......
1562 GEORGE P. SCH'\1JD

EPIDEMIOLOGY

In addition to knowing the relative probabilities of individual diseases

occurring in a given community, patient demographics can be diagnostically
helpful. Genital ulcers are diseases of men. One reason for this is the
greater visibility of the male anatomy, such that men are far more likely
than women to notice painless lesions (i.e., women are underdiagnosed).
In 1989, the male:female ratio for primary syphilis in the United States was
3.9:1, yet the ratio for secondary syphilis, in which men and women would
have an equal chance of noting a rash, was 1:1.3. 4 This ratio reversal
indicates that many women do not notice the painless ulcers of primary
syphilis. In contrast, the male:female ratio for genital herpes, which is
characterized by painful ulcerations, is about 1:1 (see Table 2).3 Another
factor contributing to male predominance among patients with genital
ulcers is that prostitution for money or drugs is associated in some
communities with syphilis 28 and chancroid 18 such that a small number of
women infect many men.
Racial characteristics, whether representing differences in socioeco-
nomic status, sexual activity, or genetic predisposition to selected diseases,
also may be helpful diagnostically. In the United States, syphilis and
chancroid are strongly associated with minority populations. In 1988, 83%
of cases of primary and secondary syphilis occurred in blacks or hispanics,4
as do most cases of chancroid (for which no racial data are compiled). In
contrast, genital herpetic ulcer disease appears to be relatively more
common among whites, 9 despite the finding that antibody to herpes simplex
virus type 2 (HSV-2), which causes 70% to 90% of genital infections in the
United States, occurs in 65% of blacks aged 60 to 74 years compared with
20% of whites. 14 This disparity between rates of infection and those of
disease may be a consequence of higher rates of cross-protective herpes
simplex virus type 1 (HSV-l) antibody among blacks.
A sexual practice history is often helpful. Despite general adoption of
safer sexual practices among homosexual men, 13% of cases of infectious
syphilis studied in several areas in 1987 occurred in homosexual men. 2
Syphilis and especially chancroid are associated with prostitution, whether
for money or drugs. IS. 28
Syphilis. Of American men called to service during World War 11,
375,000 (5%) required treatment for syphilis. 21 With the advent of penicillin,
the numbers of cases of syphilis declined markedly. A nadir of 69,721 cases
(of all stages) was reached in 1985,4 a decline partly attributable to fear of
AIDS within the homosexual community, in which many cases were
occurring. Since then, however, rates of syphilis have increased. In 1989,
19,455 cases of primary syphilis occurred, 15,474 (80%) of which were in
men. 4
Genital Herpes. As many as 700,000 new cases of genital herpes, and
20,000,000 episodes of recurrent genital herpes, are estimated to occur
annually in the United States. 44 Initial visits to a physician's office for genital
herpes occur equally for men and women, but women are more likely to
return for subsequent attacks. 3
Chancroid. During the last decade, chancroid has again become a
ApPROACH TO THE PATIENT WITH GENITAL ULCER DISEASE 1563

disease commonly encountered in the United States. 35 In 1988, 5001 cases,

the most cases since 1949, were reported. 4 The disease is localized to
selected communities, in particular New York City, from where about three
fifths of cases are reported. The majority of cases occur in men.
Lymphogranuloma Venereum. Lymphogranuloma venereum, caused
by infection with the L serovars of C. trachomatis, is a rare disease in the
United States. Fewer than 100 cases are reported annually.4 Many of the
diagnoses of these cases are based solely on serologic criteria (a chlamydial
complement fixation titer of ~1:64) without clinical features characteristic
of lymphogranuloma venereum.
Donovanosis (Granuloma Inguinale). Donovanosis is caused by Calym-
matobacterium granulomatis, an elusive bacterium that has rarely been
cultured; no isolate exists in the American Type Culture Collection.
Donovanosis is a disease found almost exclusively in the Far East (princi-
pally, New Guinea and southern India), and is rarely diagnosed in the
United States. In 1988, 11 cases were reported, and the criteria upon
which these diagnoses were based are not known. 4

CLINICAL PRESENTATIONS

Syphilis. Mter a mean incubation period of 3 weeks, most patients

first note a papule or macule at the site where an ulcer will develop.7
Although ulcers caused by Treponema pallidum are commonly thought to
be solitary, as many as one half of cases may have more than one. 7 The
ulcers are surprisingly painless, although mild tenderness can be elicited
by firm pressure such as would occur when taking a specimen for darkfield
examination. Characteristically, the ulcers have smooth margins with firm,
palpable (indurated) borders, and a clean, indurated base that has a serous,
but not purulent, exudate (Fig. 1). Atypical ulcers are, however, common.
In men, one half of ulcers are located on the foreskin (inner or outer
surface) or coronal sulcus, with the glans and shaft being the next most
common sites. 7. 19 In homosexual men, one third of cases have ulcers in the
anorectal area that may require careful inspection to detect.19 In women,
the external genitalia is the most common site for ulcers. Ulcers may occur
in the vagina, and, in one series, 21% of women with primary syphilis had
ulcers in the cervix. 19 As many as 80% of patients have inguinal lymphad-
enopathy, which is bilateral in most cases. The adenopathy is small, rubbery
in consistency, and not painful, although one third of cases have tenderness. 7
Genital Herpes. Genital herpes is characterized by a prodrome of
paresthesias 12 to 48 hours prior to the appearance of blisters. Grouped
vesicles on an erythematous base are the hallmark of genital herpes. These
rupture, leaving multiple shallow and painful ulcerations (see Fig. 1). The
severity of first episodes of genital herpes varies, depending considerably
upon whether the individual has previously been exposed to herpes simplex
virus (HSV). Primary first-episode disease (the initial infection in someone
without preexisting antibody to HSV) is severe; painful vesicles remain for
1564 GEORGE P. SCHMID

Figure 1. Showing the appearance of the ulcers of chancroid, syphilis, and herpes. The
ulcer of chancroid has irregular margins and is deep with undermined edges (like an
Erlenmeyer flask). The syphilis ulcer has a smooth, indurated border and a smooth base. The
genital herpes ulcer is superficial and inflamed. (Schmid GP, Schalla WO, DeWitt WE:
Chancroid. In Morse SA, Moreland AA, Thompson SE (eds): Atlas of Sexually Transmitted
Diseases. Philadelphia, JB Lippincott, 1990; with permission.)

10 to 12 days before rupturing and forming ulcers, which last another 1 to

2 weeks. Nonprimary first-episode genital herpes (the first episode of
genital herpes in an individual who has preexisting antibody to HSV) is
milder than primary disease, and attacks of recurrent genital herpes are
milder still with a total duration of about 10 days. In first-episode genital
herpes, lesions are bilateral or in the midline; with recurrent disease,
lesions tend to be unilateral.
Genital herpes, unlike chancroid or syphilis, is sometimes accompanied
by diagnostically useful manifestations aside from the ulcer. Systemic
symptoms of fever, myalgia, and headache are most common with first-
episode disease. Herpes simplex virus can be cultured from the cervix of
88% of women with primary disease, 65% with nonprimary first-episode
disease, and 12% to 20% with recurrent disease; the cervix is visibly
abnormal, with erythema and erosions in 90% of instances when HSV is
recovered, although changes may be subtle. 44 Dysuria and urethritis may
occur, and HSV is frequently recovered from these sites in individuals with
these symptoms and genital herpes.
Unfortunately, the clinician is often confronted with atypical cases of
genital herpes. The patient may have no history of a prodrome or vesicles,
only one or several shallow ulcers, extensive shallow ulceration due to
coalescence of smaller ulcers, or only one or several deep ulcers (perhaps
due to coalescence of smaller ulcers). Salzman32 described ulcers in 33
patients thought to be chancroid and recovered HSV from 16 (48%). Diaz-
MitomalO studied 28 individuals with genital ulcers selected for the absence
ApPROACH TO THE PATIENT WITH GENITAL ULCER DISEASE 1565
of a vesicular phase to their illness and recovered HSV from the ulcers of
6 (21%).
Chancroid. Chancroid is characterized by one or a few painful ulcers
occurring after an incubation period of 4 to 7 days. The most distinguishing
feature of chancroid is the pain associated with the ulcer, and patients are
reluctant to have the ulcer manipulated. The pain is probably the result of
the "rawness" of the ulcers, which tend to be deep, "beefy," purulent, and
with an irregular undermined border (see Fig. 1).25. 41 The ulcers can
become large, reaching the size of a quarter. Despite this characteristic
appearance, the ulcers of chancroid may occasionally be shallow, similar to
those of genital herpes. The ulcers are characteristically located in the
coronal sulcus in circumcised men and on the prepuce of uncircumcised
men. In women, although the large male:female ratio of cases has prompted
diligent searches for asymptomatic women with ulcers on the vaginal walls
or cervix, such ulcers were found in only 7% of women in one series. 25
Instead, ulcers occur externally and are painful.
About one half of men with chancroid develop painful inguinal lymph-
adenopathy, which can be unilateral or bilateral. Pain and tenderness are
characteristic of the lymphadenopathy accompanying chancroid, and, when
such adenopathy occurs in conjunction with a solitary genital ulcer, it is
virtually pathognomonic of chancroid. Without treatment, or occasionally
in spite of appropriate treatment, lymph nodes rupture, with subsequent
drainage of pus. Lymphadenitis is less common in women than in men,
presumably because lymphatic drainage from the posterior two thirds of
the vagina is to sacral nodes. Nevertheless, lymphadenitis occurred in 35%
of Kenyan women with chancroid. 25
Lymphogranuloma Venereum. Lymphogranuloma venereum is often
included in the differential diagnosis of genital ulcers, yet an ulcer is a
minor feature of this disease. At most, 30% of men, and fewer women,
have or remember an ulcer, and one is rarely present when the patient is
seen by the clinician. 42 The ulcer of lymphogranuloma venereum is a
transient, shallow, painless ulcer which precedes by 7 to 30 days the
appearance of inguinal lymphadenopathy, which is the prominent feature
of lymphogranuloma venereum. Inguinal lymphadenopathy occurs as a
result of multiple enlarged, matted, and tender nodes, which may coalesce
with suppuration and bubo formation. Because of the prominence of
adenopathy in chancroid, it may be misdiagnosed as lymphogranuloma
venereum by inexperienced clinicians.
Donovanosis (Granuloma Inguinale). Calymmatobacterium granulom-
atis causes a slowly progressive disease of the genital area that is character-
ized as much by heaped-up granulomatous tissue as it is by genital
ulceration. The lesions are painless and slowly destructive, but do not
typically present as an acute ulcerative process. 27 Nevertheless, other
diseases (in particular, chancroid or carcinoma) may mimic donovano-
sis. 16• 45 Lymphadenopathy in donovanosis is unusual, although a "pseudo-
bubo" may be formed 45 by subcutaneous extension of the granulomatous
process into the inguinal area.
1566 GEORGE P. SClI\llD

DIAGNOSIS

Syphilis. Because T. pallidum can be recovered from lesion material

only by inoculation of rabbit testes, definitive diagnosis depends upon
demonstrating T. pallidum in ulcer material by darkfield or fluorescence
microscopy. Most microscopes can be fitted with a darkfield condensor,
and microscopy takes only minutes to perform. Fluorescence microscopy is
potentially available at many hospitals but requires fluorescein-tagged
monoclonal antibodies, which currently are available only in research
settings,'30 Immunofluorescence microscopy will probably best be used in
situations when specimens cannot be immediately examined by darkfield
microscopy.
Although darkfield microscopy provides a definitive diagnosis, serology
provides only presumptive evidence of infection because a positive titer
cannot distinguish present from past infection. A nontreponemal test,
generally either the rapid plasma reagin (RPR) or Venereal Disease Research
Laboratory (VDRL), should be done in every case of genital ulcer disease.
These so-called nontreponemal tests do not measure specific treponemal
antibody but rather antibody directed against cardiolipin, which makes up
about 10% of the cell lipids of T. pallidum but is also found in mammalian
tissueY These tests are positive in about 80% of cases of primary syphilis
(the RPR is slightly more sensitive than the VDRL). All positive nontrepo-
nemal tests should be confirmed by a specific treponemal test (e.g., the
fluorescent treponemal antibody-absorbed [FTA-ABS] or microhemagglu-
tination-Treponema pallidum [MHA-TP]), because biologic false-positive
tests may occur, particularly with intravenous drug use (6% to 9%, compared
to 1% in the general population 15), acute viral infections (including genital
herpes), some vaccinations, and autoimmune diseases.
Genital Herpes. Although genital herpes is sometimes characteristic
enough to be diagnosed without a laboratory test,9 I think it is reasonable
to confirm clinical suspicions with a laboratory test whenever possible, and
always if there is any diagnostic uncertainty. For many persons, the
diagnosis of genital herpes is traumatic, alters self-image, and affects the
perceived ability to enter new sexual relationships and bear children. Thus,
laboratory tests should be used to ensure that the correct diagnosis is made.
All tests for HSV are most sensitive during the vesicle stage. Culture is the
most sensitive and specific test, with sensitivity approaching 100% in the
vesicle stage and 89% in the pustular stage, and dropping to as low as 33%
in patients presenting with ulcers.22, 38 A variety of nonculture antigen-
detection or nucleic acid hybridization tests have become available. Overall,
these nonculture tests are about 80% as sensitive as culture,44 although
some tests are more sensitive. For any test performed during the ulcer
stage, a negative result cannot be relied upon to exclude the diagnosis of
genital herpes.
An often overlooked but easily performed cytologic test is the Tzanck
smear. The Tzanck smear detects multinucleated giant cells that are formed
as a result of infection with herpes viruses. The test is done by scraping
the base of the vesicle or ulcer with a swab or scalpel, placing the material
on a slide, staining the slide (often with Wright's stain, used in hematology
ApPHOACH TO TilE PATIENT WITIl GENITAL ULCEH DISEASE 1567

laboratories to stain blood smears), and looking for the characteristic cells
by light microscopy. Solomon et al 38 found the sensitivity of the Tzanck
smear (compared to culture) to be 67% in the vesicle stage, 75% in the
pustule stage, and 50% in the ulcer stage. Despite the relatively low
sensitivity, the specificity of the test in this study was 94%. Others,
however, cite a lower specificity;'6 which may relate to the fact that
experience is important in reading slides.
Chancroid. Because of the uncertainty of accuracy and unavailability,
respectively, of Gram stain of ulcer secretions and culture, most cases are
diagnosed on clinical grounds. Although the Gram stain is said to be
insensitive and nonspecific, 8. 40 others have reported a sensitivity of 62%
and a specificity of 99% for slides with gram-negative coccobacilli in parallel
rows or in a clustered, "school of fish" appearanceY Culture is definitive,
and many enriched media have been devised. 20 More than one type should
be used, as some isolates will grow on one medium and not the other (we
use a rabbit blood agar and a chocolated horse blood agar). Incubation at
33°C, in a moist environment with added carbon dioxide, is preferable to
35°C. 34 Unfortunately, culture is difficult and insensitive, accounting for
the fact that most clinical microbiology laboratories cannot isolate the
organism at all. Even in experienced research laboratories, Haemophilus
ducreyi is recovered from at most 80% of clinically suspected cases. 3 !
Haemophilus ducreyi can be cultured from lymph nodes, but recovery
is less successful than from ulcers for reasons that are unclear. Gram-
stained smears of pus from lymph nodes show white blood cells, but
organisms are infrequently seen, suggesting that organisms are fewer in
number in nodes than in ulcers.
Immunofluorescence of ulcer secretions, using a fluorescein-tagged
monoclonal antibody against H. ducreyi, has been little studied. 33 Recent
serologic advances, currently research tools, appear promising.
Lymphogranuloma Venereum. The diagnosis of lymphogranuloma ve-
nereum is usually based on serologic criteria (a chlamydial complement
fixation titer of 2: 1:64). Although such titers support the diagnosis of
lymphogranuloma venereum when present in a patient with clinical features
compatible with lymphogranuloma venereum, titers of this magnitude
occasionally occur as a result of infection by the far more commonly
occurring non-L serovars of C. trachomatis; thus, titers of 1:64 or greater
are not specific for lymphogranuloma venereum. Specific diagnosis of
lymphogranuloma venereum requires culture of C. trachomatis with sub-
sequent immunotyping of the isolate, or the use of microimmunofluores-
cence serology, which is able to distinguish type-specific serologic responses
to infection with C. trachomatis. Microimmunofluorescence is available in
only several research laboratories in the United States.
Donovanosis (Granuloma Inguinale). Because C. granulomatis cannot
be successfully cultured, diagnosis depends upon visualizing it within biopsy
specimens. This can be accomplished simply by use of a "crush preparation."
A fragment of tissue is removed, placed between two slides, and crushed.
The slide is then stained (Wright's stain is satisfactory) and examined for
typical "Donovan bodies," bacillary organisms within histiocytes, which are
1568 GEORGE P. SCHMlD

diagnostic of granuloma inguinale. Alternatively, biopsy specimens can be

examined. No serology is available.

APPROACH TO THE PATIENT WITH GENITAL ULCER DISEASE

Diagnostic efforts are traditionally directed at excluding syphilis be-

cause of the consequences of inappropriate therapy (the development of
tertiary syphilis or, in pregnant women, congenital syphilis). Nevertheless,
it is also important in women to document a history of genital herpes
because pregnant women should inform their obstetricians of a past history
of genital herpes, as such women are at small but increased risk of delivering
an infant with neonatal herpes. Last, all three common causes of genital
ulcers (syphilis, genital herpes, chancroid) have been associated with
increased or high seroprevalence rates of human immunodeficiency virus
(HIV).12, 13.37 Proper diagnosis is important in selecting appropriate therapy,
with the possible consequence of decreased likelihood of acquiring or
transmitting HIV infection.
Frequent Cases of Genital Ulcers. Despite the imprecision of the
clinical diagnosis of genital ulcer disease, several presentations are highly
suggestive of specific diagnoses:
1. A nonpainful and minimally tender ulcer, not accompanied by
inguinal adenopathy (or accompanied by small, rubbery nodes), is likely to
be syphilis, especially if the ulcer is indurated.
2. Grouped vesicles mixed with small ulcers, particularly with a
preceding history of such lesions, are almost pathognomonic of genital
herpes.
3. One to three extremely painful ulcers, accompanied by tender
inguinal lymphadenopathy, are unlikely to be anything except chancroid; if
the adenopathy is fluctuant, the diagnosis is secure.
4. An inguinal bubo accompanied by one or several ulcers is most
likely to be chancroid. If there is no ulcer, the most likely diagnosis is
lymphogranuloma venereum.
Infrequent Causes of Genital Ulcers. Trauma is commonly thought by
patients to be a cause of ulcers, especially among men ("she was tight," "I
noticed it right after intercourse," or "I caught it in my zipper "), but is an
infrequent diagnosis that should be made only with a compatible history
(e.g., history of a bite) and after investigation for other causes. Fixed drug
eruptions often are manifest as genital ulcers; tetracycline and phenophthal-
ein are common causes. Diagnosis is based upon a history of medication
consumption, exclusion of other causes, and ideally a history of a previous
similar result of medication administration. A carcinoma will be a subacute
lesion and will not respond to empiric antimicrobial therapy. Reiter's
syndrome, which affects mostly young men, involves the skin (frequently
in the genital area) in about one half of cases. If it occurs on moist areas
(e.g., in uncircumcised men), shallow, painless ulcers occur on the glans
and may encircle the glans ("circinate balanitis"). If disease occurs on dry,
keratinized surfaces, however, macules develop. Genital ulcers have been
reported in infectious mononucleosis, and Epstein-Barr virus has been
ApPROACH TO THE PATIE:-iT WITH GE~ITAL ULCEI\ DISEASE 1569
recovered from them 26 ; despite the ubiquity of infectious mononucleosis,
this event appears to be unusual. Beh<;et's syndrome and aphthous ulcers
may be characterized by oral or genital ulcers.
Approach to Diagnosis. Because not all medical facilities have a full
complement of diagnostic tests, and because few tests yield an immediate
result, the clinician must use clinical judgment in making a diagnosis. Such
judgment is based on patient risk factors (e. g., syphilis and chancroid are
associated with prostitution), demographics, medication history (e. g., fixed
drug eruptions following laxative use), self-medication history (which may
lead to false-negative tests), presence or absence within a community of
diseases that are not universally common, and the clinical appearance of
the ulcer.
All individuals should have a serologic test for syphilis, and, ideally, a
darkfield examination and a test for herpes simplex virus performed (Fig.
2). In areas where chancroid is common, a Gram stain of ulcer exudate
may suggest the diagnosis and the lesions(s) should be cultured for H.
ducreyi. If initial darkfield and syphilis test results are negative or unreliable
(perhaps due to self~medication, either oral or topical), and the diagnosis
of syphilis is thought likely, reliable patients can be asked to return the
following day for re-examination.
If buboes are present, a diagnostic and therapeutic aspiration per-
formed through intact skin (and not through inflamed skin, for fear of fistula
formation) should be performed. With the exudate, a Gram stain and
culture for H. ducreyi can be made, and a culture for C. trachomatis can

-Positive Treat for syphilis

and perform STS

- - - - <.... Positive

No

1·3 painless Multiple shallow, Possibly serafast,

ulcers with painful ulcers. but treat for
nontender often with preceding syphilis and
nodes blisters; nodes may consider other
be tender
,I
causes of ulcer

Figure 2. Algorithm for clinical approach to the diagnosis of genital ulcers, concentrating
on common causes. In all cases, a careful history should be taken and additional diagnostic
tests used when possible, e. g., Tzanck, culture, or antigen- or nucleic acid-detection test for
suspected genital herpes. STS = serologic test for syphilis; LGV = lymphogranuloma vener-
eum.
1570 GEORGE P. SCH\1ID

be performed if lymphogranuloma venereum is suspected. Even the in-

guinal lymph nodes of syphilis may yield T. pallidum if aspirated (a small
amount of nonbacteriostatic saline can be injected and a drop of resultant
fluid examined by darkfield microscopy), although this is rarely done.

THERAPY

If immediate clinical and laboratory evaluation is unrewarding, patients

should be treated for the most likely diagnosis and seen again in 7 days,
unless symptomatology worsens. At that time, the patient should be
reevaluated. If initial therapy did not include benzathine penicillin and
syphilis was thought to be a possibility, a repeat syphilis test should be
performed. In all patients not receiving adequate therapy for syphilis, but
particularly in those without a prompt response to therapy, a syphilis test
2 to 3 months following therapy is prudent. Some clinicians, but particularly
those practicing in areas with high rates of syphilis, prefer to treat all cases
of genital ulcers of unclear etiology with benzathine penicillin, in addition
to therapy for other possible diseases. Because genital ulcers appear to
facilitate HIV transmission (or may be markers of high-risk activity for its
acquisition), testing of patients with genital ulcers for HIV infection,
particularly those with syphilis or chancroid, should be encouraged. If
initial testing is negative, a repeat test might be performed in 3 months.
Sexual partners of patients with genital ulcer disease should be examined
promptly, tested, and treated appropriately. 1

REFERENCES

1. Centers for Disease Control: Sexually transmitted diseases treatment guidelines. MMWR
38(S-8): 1989
2. Centers for Disease Control: Continuing increase in infectious syphilis-United States.
MMWR 37:35-38, 1988
3. Centers for Disease Control: Genital herpes infection-United States, 1966-1984. MMWR
35:402-404, 1986
4. Centers for Disease Control: Unpublished statistics
5. Chapel T, Brown WJ, Jeffries C, et al: The microbiological flora of penile ulcerations. J
Infect Dis 137:50-56, 1978
6. Chapel TA, Brown WJ, Jeffries C, et al: How reliable is the morphological diagnosis of
penile ulcerations? Sex Transm Dis 4:150-152, 1977
7. Chapel TA: The variability of syphilitic chancres. Sex Transm Dis 5:68-70, 1978
8. Coovadia YM, Kharsany A, Hoosen A: The microbial aetiology of genital ulcers in black
men in Durban, South Africa. Genitourin Med 61:266-269, 1985
9. Corey L: Genital herpes. In Holmes KK, Mardh PA, Sparling PF, et al (eds): Sexually
Transmitted Diseases. New York, McGraw-Hill, 1990, pp 391-413
10. Diaz-Mitoma F, Benningen G, Slutchuk M, et al: Etiology of nonvesicular genital ulcers
in Winnipeg. Sex Transm Dis 14:33-36, 1987
11. Fast MV, D'Costa LJ, Nsanze H, et al: The clinical diagnosis of genital ulcer disease in
men in the tropics. Sex Transm Dis 11:72-76, 1984
12. Greenblatt RM, Lukehart SA, Plummer FA, et al: Genital ulceration as a risk factor for
human immunodeficiencv virus infection. AIDS 2:47-50, 1988
13. Holmberg SO, Stewart JA: Gerber AR, et al: Prior herpes simplex virus type 2 infection
as a risk factor for HIV infection. JAM A 259:1048-1050, 1988
ApPROACH TO TIlE PATIENT WITH GENITAL ULCER DISEASE 1571
14. Johnson RE, "Iahmias AJ, Magder LS, et al: A seroepidemiologic survey of the prevalence
of herpes simplex virus type 2 infection in the United States. N Engl J Med 321:7-12,
1989
15. Kaufman RE, Weiss S, Moore JD, et al: Biological false positive serological tests for
syphilis among drug addicts. Br J Vener Dis 50:350-353, 1974
16. Kraus SJ, Werman BS, Biddle JW, et al: Pseudogranuloma inguinale caused by Haemo-
philus ducreyi. Arch Dermatol 118:494-497, 1982
17. Leibovitz A: An outbreak of pyogenic penile ulcers associated with a microaerophilic
streptococcus resembling Haemophilis ducreyi. Am J Syph Neurol 38:203-215, 1954
18. McLaughlin M, Wilkes M, Blum S, et al: Risks associated with acquiring chancroid genital
ulcerative disease and HIV infection: A case control study [abstract]. In Programs and
Abstracts of the Fifth International Conference on AIDS, Montreal, June 1989
19. Mindel A, Tovey SJ, Timmins DJ, et al: Primary and secondary syphilis, 20 years'
experience. 2. Clinical features. Genitourin Med 65:1-3, 1989
20. Morse SA: Chancroid and Haemophilus ducreyi. Clin Microbiol Rev 2:137-157, 1989
21. Morton RS: Evolution ofvenerealogy as a specialty. In Holmes KK, Mardh PA, Sparling
PF, et al (eds): Sexually Transmitted Diseases. New York, McGraw-Hill, 1984, pp 30-
38
22. Mosley RC, Corey L, Benjamin D, et al: Comparison of viral isolation, direct immuno-
fluorescence, and indirect immunoperoxidase techniques for detection of genital herpes
simplex virus infection. J Clin Microbiol 13:913-918, 1981
23. Musher D: Biology of Treponema pallidum. In Holmes KK, Mardh PA, Sparling PF, et
al (eds): Sexually Transmitted Diseases. New York, McGraw-Hill, 1990, pp 205-219
24. Nsanze H, Fast MV, D'Costa LJ, et al: Genital ulcers in Kenya: Clinical and laboratory
study. Br J Vener Dis 57:378-381, 1981
25. Plummer FA, D'Costa JL, Nsanze H, et al: Clinical and microbiologic studies of genital
ulcers in Kenyan women. Sex Transm Dis 12:193-197, 1985
26. Portnoy J, Ahronheim GA, Ghibu F, et al: Recovery of Epstein-Barr virus from genital
ulcers. N Engl J Med 311:966--968, 1984
27. Richens J: Donovanosis-a review. Papua New Guinea Med J 28:67-74, 1985
28. Rolfs RT, Goldberg M, Sharrar RG: Risk factors for syphilis: Cocaine use and prostitution.
Am J Public Health, in press
29. Romanowski B, Blackwelder M, Draker JM: Etiology of genital ulceration [abstract]. In
Programs and Abstracts of the Thirty-second General Assembly of The International
Union Against Venereal Diseases and Treponematoses, Montreal, June 17-21, 1984
30. Romanowski B, Forsey E, Prasad E, et al: Detection of Treponema pallidum by a
fluorescent monoclonal antibody test. Sex Transm Dis 14(3):156-159, 1987
31. Ronald AR, Plummer FA: Chancroid and Haemophilus ducreyi [editorial]. Ann Intern
Med 102:705-707, 1985
32. Salzman RS, Kraus SJ, Miller RG, et al: Chancroidal ulcers that are not chancroid: Cause
and epidemiology. Arch Dermatol 120:636--639, 1984
33. Schalla WO, Sanders LL, Schmid GP, et al: Use of dot-immunobinding and immunofluo-
rescence assays to investigate clinically suspected cases of chancroid. J Infect Dis
153:879-887, 1986
34. Schmid GP, Faur YC, Valu JA, et al: Effect of temperature on the recovery of Haemophilus
ducreyi from clinical specimens [abstract], In Programs and Abstracts of the Twenty-
Ninth Interscience Conference on Antimicrobial Agents and Chemotherapy, Houston,
September, 1989
35. Schmid GP, Sanders LL, Blount JH, et al: Chancroid in the United States: Reestablish-
ment of an old disease. JAMA 258:3265-3268, 1987
36. Sever JL, Nerurkar LS: Diagnosis of herpes simplex. In Straus SE (moderator): Herpes
simplex virus infection: Biology, treatment, and prevention. Ann Intern Med 103:404-
419, 1985
37. Simonsen IN, Cameron W, Gakinya MN, et al: Human immunodeficiency virus infection
among men with sexually transmitted diseases: Experience from a cent er in Africa. N
Engl J Med 319:274-278, 1988
38. Solomon AR, Rasmussen JE, Varani J, et al: The Tzanck smear in the diagnosis of
cutaneous herpes simplex. JAMA 251:633-636, 1984
39. Stedman's Medical Dictionary. Baltimore, Williams & Wilkins, 1976
1572 GEORCE P. SCIIMID

40. Sturm AW, Stolting GJ, Cormane HH, et al: Clinical and microhiological evaluation of 46
episodes of genital ulceration. Genitourin Med 63:98-101, 1987
41. Taylor DN, Duangmani C, Suvongse C, et al: The role of Haemophilus ducreyi in penile
ulcers in Bangkok, Thailand. Sex Transm Dis 11:148-151, 1984
42. U.S. Department of Health, Education, and Welfare: Chancroid, donovanosis, lympho-
granuloma venereum. DHEW Publication No. 75-8302, 1964
43. Ursi JP, Van Dyck E, Ballard RC: Characterisation of an unusual bacterium isolated from
genital ulcers. J Med Microbiol 15:97-103, 1982
44. Webb DH, Fife KH: Genital herpes simplex virus infection. Infect Dis Clin North Am
1:97-122, 1987
45. Werman BS, Herskowitz LJ, Olansky S, et al: A clinical variant of chancroid resembling
granuloma inguinale. Arch Dermatol 119:890--894, 1983

Address reprint requests to:

George P. Schmid, MD
Technical Information Services (E06)
Centers for Disease Control
Atlanta, GA 30333