8 Clinical Instrumentation in

Contact Lens Practice

JAMES S. WOLFFSOHN and FRANK EPERJESI

CHAPTER CONTENTS Corneal Topography, 158 Anterior Eye Photography (Slit-Lamp

Photokeratoscopy or Cameras), 167
Videokeratoscopy, 160 Other Instruments for Assessing the
Types of Corneal Topographers, 160 Anterior Segment, 169
Slit-Lamp Biomicroscropy (Slit-Lamp), 164

Contact lens practitioners have a range of instrumentation The eye is constantly moving, even during apparently
available to them with which they can assess the anterior steady fixation. It is therefore difficult to measure directly
eye for its suitability to wear or continue to wear lenses, and the size of an image reflected by the cornea. However, if the
with which they determine the most appropriate lenses to image is doubled by passing it through a prism or a doubly-
be fitted. This chapter includes most of the commonly used refractive crystal, then when the base of one resultant image
instruments, although there are others such as the lipid layer is aligned with the top of the other, the displacement will
interferometer (see Chapter 5) which can further aid contact equal the exact height of the object (Fig. 8.3). This principle
lens fitting and aftercare. can be used in the form of either:
■ fixed doubling, where a predetermined amount of doubling
is incorporated and the mire moved until the image pro-
Corneal Topography duced is of the predetermined height (e.g. Javal-Schiotz- or
Zeiss-type mires)
Determination of corneal curvature is of prime importance ■ variable doubling, where the object is set to a predeter-
especially with rigid contact lens fitting. Comparing corneal mined size while the doubling system is varied until the
astigmatism with spectacle astigmatism can assist contact image is displaced through its exact height (e.g. Bausch
lens selection, such as whether rigid or soft lenses are likely & Lomb type mires, Fig. 8.2).
to give a better visual result or whether a back, front or
bitoric lens might be necessary (see Chapter 11). Corneal The mires reflected from the corneal surface vary in appear-
curvature measurements use the reflection of light off the ance among manufacturers (Figs 8.2, 8.4 & 8.5).
convex tear film coated cornea, which alters depending on As the distance of the eye and change in image size
the topography of the surface. Early methods were purely resulting from its reflection from the cornea is now known,
observational, with the first clinically applicable technique the radius of curvature can be calculated. It is read on an
being the keratometer. internal or external scale in millimetres or in dioptres – the
latter making the assumption that the refractive index of the
cornea is on average 1.3375, including a compensation for
KERATOMETRY
the back surface of the cornea having a power of –10% of
The keratometer has been partially superseded by the corneal the power of the front surface (see Chapter 7).
topographer, but it is still useful in providing a measure- The instrument must be focused before use. As the cornea
ment of the radius of curvature and continues to be used to usually has two principal meridians at 90° to each other, the
select the first contact lens. Keratometers are also available instrument is first rotated until the horizontal limbs of the
combined with an autorefractor or as a hand-held version mires are coincident. If the keratometer is a one-position
that is useful for measuring central corneal curvature in instrument (e.g. Bausch & Lomb–type mires), the image
small children who require contact lenses (see Chapter 24). is doubled in two directions at right angles to each other,
The keratometer is a poor guide to overall corneal shape which provides a means of measuring the two meridians
as it assesses only the central 3.0–3.5 mm of the cornea. simultaneously. The instrument is then adjusted until the
At a fixed viewing distance, an object of known size will two parts of the mire are superimposed. The reflected mires
be imaged and the image size will depend on the radius of show any corneal distortion, lens flexure and tear film stabil-
curvature of the reflecting surface (Fig. 8.1). The working ity. Two-position instruments (e.g. those incorporating Zeiss
distance of the keratometer is usually monitored through or Javal-Schiotz mires) only assess the corneal curvature in
a Scheiner disc or similar system. This produces a doubled one meridian and therefore need to be rotated by 90° to
image of the object in the eyepiece unless the instrument is measure the second principal meridian. The two meridians
used at the exact working distance required by the instru- may not be at 90° to each other if the astigmatism is irregu-
ment design (Fig. 8.2). lar. Javal-Schiotz mires are usually of two different colours
158
 8 • Clinical Instrumentation in Contact Lens Practice 159

B'
h
h'
C A
Q'

h' r/2
= But d x
h x
r/2
h' r/2 2 dh'
∴ ∴r
h d h
r

x Fig. 8.1 The optical principle of

d keratometry.

Fig. 8.2 The Bausch & Lomb type keratometer

mires. When not set at the correct working
distance, a Scheiner disc creates a doubled
Out of focus Focused and aligned image of the mires in the eyepiece.

Aligned

Not along principal meridian Along principal meridian

Fig. 8.3 The principle of doubling used to measure image height. Fig. 8.4 The Zeiss type mire.
160 SECTION 3 • Instrumentation and Lens Design

so that any overlapping of the mires produces a change in cornea, and the clarity and regularity of the reflection is
mire colour, aiding precise alignment. With the mires aligned evaluated.
in the steeper meridian, rotation of the instrument head Corneal topography can be particularly useful for:
through 90° will result in one mire-step overlap for each ■ assessing irregular corneas (e.g. trauma, displaced apices)
dioptre of astigmatism (Fig. 8.5). ■ advanced lens fitting such an orthokeratology (see
Errors in the use of a keratometer involve: Chapter 19)
■ features of the instrument design ■ keratoconus (see Chapter 20)
■ inaccuracies in paraxial ray theory ■ showing corneal shape changes, especially in advancing
■ the assumption that the peripheral areas from which keratoconus (see Chapter 20)
the mires are reflected have the same curvature as the ■ following postrefractive surgery (see Chapter 23).
corneal pole
■ operator-induced errors
■ inaccurate alignment
■ focusing errors Types of Corneal Topographers
■ proximal accommodation
■ orientation of the instrument TOPOGRAPHERS USING A PLACIDO DISC
■ patient induced
■ poor fixation
A corneal topographer (or photo/video keratoscope) is an
■ corneal distortion
automated version of the Placido disc. It uses a bowl or cone
to act as the illumination source to reflect off the tear film
coated cornea. Instruments using this technology include
the Oculus Keratograph (bowl technology) and the Medmont
Photokeratoscopy or E300 (cone) (Fig. 8.6a and b). A camera attached to an
Videokeratoscopy internal or external computer images the rings as they are
reflected off the central to mid-peripheral 10 mm of the tear
Corneal shape is more complex than the central curvature film coated cornea. Image capture can be manually triggered
measurements of keratometry. It is usually represented by a when the image is centred and in focus (often highlighted
prolate ellipse (one that flattens in the periphery), but there by indicator scales on the screen) or activated automatically,
is wide variation among individuals (Durr et al. 2015). to plot the contours of the corneal surface.
A Placido disc is a hand-held circular disc with concentric, Image processing detects the separation of the rings in
alternate black-and-white rings which are reflected off the multiple meridians, which can be interpreted as curvature
at that position of the anterior corneal surface. The data
are displayed in the form of contour maps and simulated
keratometry readings in the principal axes (see Fig. 8.9). The
latter is generated from the innermost rings, the diameters
of which most nearly equate to a conventional keratometer.
The average asphericity of the cornea can also be calculated.
Measures of asphericity include:
■ eccentricity = e
■ shape factor = p
■ asphericity parameter = Q
where e2 = 1 – p = –Q
The average asphericity of the human eye (Q) is about
–0.2 to –0.3 (e = 0.45–0.55) and varies with meridian. It
Fig. 8.5 The Javal-Schiötz type mire. The steps on the second mire is similar across ethnicity (Ying et al. 2012, Maseedupally
represent 1-dioptre intervals of astigmatism. et al. 2015).

Fig. 8.6 (a) Medmont E300. (b) Close-up of the

Medmont cone-shaped illumination source. A B
 8 • Clinical Instrumentation in Contact Lens Practice 161

OTHER TYPES OF CORNEAL TOPOGRAPHERS and the Pentacam can provide further information about
the anterior segment (Fig. 8.8).
Other systems are not simply based on a Placido disc (Yeung THE EYE SURFACE PROFILER (EAGLET EYE) is a
& Kit 2015); they use different approaches to produce corneal corneal and scleral topographer that can measure the cur-
contour maps: vature and sagittal height of up to 20 mm diameter of the
anterior surface of the eye. The instrument captures the
■ Orbscan uses slit scanning.
image by projecting two Moiré fringe patterns onto the eye
■ Pentacam (Oculus) uses Scheimpflug imaging (Fig. 8.7).
after instilling fluorescein (see Fig. 14.28).
■ Sirius (CSO Italia) combines a Scheimpflug camera with
Placido technology.
ORBSCAN II (BAUSCH & LOMB) scans a series of slit
sections of the cornea in addition to the traditional Placido
disc rings to produce a topographical map. The camera plane
is at 45° to the light slit to improve the depth of field of the
optic section (like swinging the illumination system on a
slit-lamp to better view the corneal layers).
PENTACAM (OCULUS) (see Fig. 8.7) uses a Scheimp-
flug camera set in a rotating wheel to directly measure the
corneal topography and analyse the cornea while a second
static camera within the fixation target (a monochromatic
slit) monitors fixation (Jain & Grewal 2009). In addition,
the iris camera lens measures the horizontal visible iris
diameter.
CASSINI TOTAL CORNEAL ASTIGMATISM (CASSINI)
uses multicoloured (red, green and yellow) light emitting
diodes (LED) spaced throughout the bowl and uses ray tracing
to measure the relative position of each point in order to
produce a map. The manufacturers suggest that this has
fewer errors than from ring overlap in irregular corneas or
tear film distortion. The machine also utilises the second
Purkinje reflections of each LED to quantify posterior corneal
curvature.
These topographers are able to give corneal curvature data
Fig. 8.7 Pentacam. (Courtesy of Lynne Speedwell.)
for the anterior and posterior cornea, and the Orbscan II

Fig. 8.8 Pentacam scan showing the Overview presentation, which gives an array of useful anterior segment information including corneal thickness
(pachymetry) and correction factor for intraocular pressure, anterior chamber depth, anterior chamber angle and pupil and corneal diameter. (Courtesy
of Lynne Speedwell.)
162 SECTION 3 • Instrumentation and Lens Design

A B

Fig. 8.9 Contour maps: (a) axial cur-

vature (b) tangential curvature (c)
corneal height; (d) refractive power. C D
(Reproduced with permission from
Medmont International.)

mathematical derivation of the radius of curvature with

CONTOUR MAPS, PLOTS OR SCANS
radii centres not restricted to a single axis.
These are generated from the point contour values, with ■ Advantages – gives a more accurate representation of

similar values connected to form zones of equal curvature. the position of the apex and other corneal structures
The zones are coloured in spectral order, with the red end and a better corneal shape for comparing the plot to an
(warm colours) corresponding to steeper (shorter) corneal observed contact lens fluorescein pattern.
radii and the blue end (cooler colours) corresponding to flatter ■ Disadvantages – difficult to verify and has a lower repeat-

(longer) corneal curvatures. ability than the sagittal plot.

Relative scales grade the image presentation to cover the ■ CORNEAL HEIGHT OR ELEVATION MAPS (Fig. 8.9c)
entire difference in curvature across the image, highlight- (or X, Y, Z coordinates or Z values) – based on the difference
ing any differences occurring regardless of their magnitude. in height from a reference sphere (the reference sphere
Absolute scales are set by the user and attribute each scale can vary among instruments).
increment to a set radius or power change. Careful note must ■ Advantages – they are the most direct measure of corneal

be made of the type of scale used and the magnitude of the shape and can predict areas of corneal touch of a contact
increments in order to correctly interpret contour maps (see lens on an irregular cornea.
Chapter 19). ■ Disadvantages – they have the lowest repeatability.

There are four main ways in which contour maps are ■ REFRACTIVE POWER MAPS (Fig. 8.9d) convert the
presented. detected curvature at any point into presumed refractive
power based on assumptions of the refractive index of the
■ SAGITTAL (OR AXIAL) MAPS (Fig. 8.9a) determine
cornea. Clinicians tend to think in terms of power rather
the radius of curvature of the cornea at each measured
than radii, but the true power of the cornea is based on
point. This is based on a single refracting surface formula
more than front surface curvature.
(paraxial ray theory). It assumes rotational symmetry of ■ Advantage – can infer the quality of vision from the
the surface and predicts that all rays will be focused on
corneal surface especially after corneal surgery.
the axis of symmetry (Fig. 8.9a). ■ Disadvantage – makes presumption that corneal has
■ Advantages – easy to verify, has the highest repeatability
standard refractive index in all individuals.
and is the most widely used; possible to correlate the
anterior surface shape with the refractive status, e.g. Most instruments have software to simulate the expected
determines the type and shape of astigmatism. fluorescein patterns of specific lenses (custom-made by the
■ Disadvantages – distorts the position of the apex and manufacturer or the practitioner’s own design), allowing
features such as ablation areas. improved empirical fitting accuracy (Fig. 8.10; and further
■ TANGENTIAL MAPS (or instantaneous representation) material available at: [Link]
(Fig. 8.9b) calculate the actual radius of curvature mea- Comparative data are also available that show, for example,
sured at a tangent (90°) to its surface. This is based on a how keratoconus changes over time (Fig. 8.11).
 8 • Clinical Instrumentation in Contact Lens Practice 163

Fig. 8.10 Simulated fluorescein

pattern produced by the contact
lens fitting program of the Medmont
Corneal topographer. (Reproduced
with permission from Medmont
International.)

Fig. 8.11 Pentacam comparative axial (sagittal) scans. The difference plot on the right, plotted on the left image, indicate how much the cornea has
steepened and thinned over 10 months.

Sources of Error ■ The number of ‘points’ analysed on each ring is usually

the same, resulting in less sampling per unit area in the
In addition to presentational distortions in accuracy, there periphery than in the centre.
are other sources of error. ■ Narrower, more closely spaced rings allow greater sam-
■ Difficulties caused by the virtual image not being accurately pling, but they are more difficult to detect as being distinct
detected or being broken-up due to from one another in distorted corneas.
■ a poor tear film ■ Peripheral rings in bowl-based topographers are often
■ inaccurate instrument focusing limited by shadows from anatomical features such as
■ corneal disease or epithelial disorders. nose or eyelashes. Cone-shaped topographers avoid this
164 SECTION 3 • Instrumentation and Lens Design

problem, allowing larger areas of the cornea to be anal- ■ red-free (enhances contrast between blood vessels and the
ysed (this is more relevant in orthokeratology). Orienta- cornea or sclera to render vascularisation and hyperaemia
tion of the patient’s head to the opposite side from the eye more visible)
being measured assists in minimising shadows in bowl ■ blue (typically cobalt, but a peak of 495 nm is optimal)
topographers and allows cone-shaped topographers to be used in conjunction with fluorescein to assess corneal
positioned sufficiently close to the eye. damage or the fitting of rigid lenses. The blue light will
■ Ocular accommodation and vergence. cause fluorescein dye to fluoresce to ~515 nm, and the
■ The two principal meridians of a toric cornea are not addition of a yellow barrier filter with a sharp cutoff at
imaged in the same plane. 500 nm (typically a Wratten 12) in the observation system
■ Alignment errors, often due to a patient’s high prescription will enhance the contrast and maximise the visibility of
and/or poor visual acuity, can cause inaccuracy, although the fluorescence (Peterson et al. 2006).
the magnitude may be limited if the working distance is
sufficiently long (Nieves & Applegate 1992). OBSERVATION SYSTEM
■ More consistent and significant errors occur because the
image is centred on the visual axis, which may not coin- This comprises a microscope with converging or parallel
cide with the geometric axis of the cornea. This results in eyepieces. A turret of an objective lens or optical zoom, which
nasal displacement of the mire image and measurement allows a greater range of magnification, can be connected
error that affects nasal readings more than temporal ones. to the latter without the need to change eyepieces.
It will also induce an error in calculating the peripheral
curvature of aspheric surfaces (McCarey et al. 1992).
SLIT-LAMP TECHNIQUES
■ Accuracy reduces with the irregularity of the cornea and
varies among different instruments (Dave et al. 1998, Hil- To ensure the slit image is produced in the same plane as
mantel et al. 1999, McMahon et al. 2001, Cairns et al. the focus of the microscope, each eyepiece must be focused
2002, Cho et al. 2002). individually.
There are a number of different methods of illumination,
which are discussed below (Stockwell & Stone 1988, Morris
Slit-Lamp Biomicroscropy & Stone 1992). However, in routine slit-lamp examination,
(Slit-Lamp) the field of view of the observation system is always larger
than the area illuminated by the slit, such that several types
Careful observation of the eye and monitoring of the ocular of illumination are evident at any one time. Fig. 8.12 high-
response to contact lens wear are prerequisites to successful lights different types of illumination that can be seen within
contact lens fitting. The slit-lamp is the main tool used in the field of view. Scanning of the structures by moving both
this process, providing a magnified, illumination-controlled, illumination and observation system across the surfaces under
binocular view of the ocular structures. examination will allow objects of interest to pass from one
The principal components of the slit-lamp are an illumi- method of illumination to another. It is often the change in
nation system that provides a focused slit image of light, appearance of the objects as they are illuminated in different
and a microscope with high resolution providing magnifica- ways that makes them readily visible.
tion typically between 6 and 40 times. Both systems can be The conventional examination technique is to scan the
moved around a common centre of rotation so that there is whole cornea in three sweeps with the illumination system
a common point of focus, which is constant as the system moved laterally so that it is always on the same side of the
is moved across the curved surfaces of the eye. This can be midline as the part of the cornea under examination (Stock-
uncoupled in order to employ special methods of illumina- well & Stone 1988).
tion (see below). Diffuse illumination – a diffuser placed in front of the
illumination system when a focused slit is not needed, for
Illumination System general examination of the external eye and adnexa under
To assess changes in the optically transparent media, a light low magnification (6-10x) and the assessment of a contact
source with clearly defined edges is needed. This ensures lens fit under white light (Fig. 8.13a and b). This is more
that light is not diffused away from the point of focus comfortable for the patient and does not flood the eye with
unless there is irregularity in the media. An optical system light, with the risk of reflex tearing.
projects an image of a mechanical slit aperture, which Direct illumination – focusing and observing the slit of light
can be varied in height and width, onto the surface being directly on a structure. The slit is narrowed to produce an
examined. optic section and passed perpendicularly through the cornea,
requiring the illumination arm to be swung as the cornea
Filters is traversed. This is used to determine depth, such as of
The slit illumination can be augmented by using filters, for fluorescein-stained corneal damage. A parallelepiped is a
example: wider optic section allowing observation of corneal features
such as nerves and striae. This can be reduced in height to
■ diffusing form a conical beam which can be shone through the pupil
■ heat reducing (dichroic filters that reduce infrared radia- at various angles to the visual axis in a dark room to detect
tion transmission) cells in the anterior chamber, seen as flare. In Fig. 8.12 the
■ neutral density (less light so more comfortable for the cornea is seen in direct illumination, whereas in Fig. 8.14
patient) the iris can be seen in direct illumination.
 8 • Clinical Instrumentation in Contact Lens Practice 165

Direct/indirect illumination

Y = Direct illumination
Z = Indirect illumination
Y Z Parallelepiped

Endothelium

Epithelium
A B C
Fig. 8.12 Illumination: (a) the beam on the left is direct illumination, the darker area to the right of the direct beam is indirect illumination, while the
light reflected off the iris will retro-illuminate the cornea; (b) schematic diagram showing illumination system on the left and microscope in the centre,
(c) a cross section of the cornea as seen in direct illumination. (Reproduced with permission from Vistakon, Synoptik and Tom Løfstrøm, Denmark.)

Diffuser

A B
Fig. 8.13 Diffuse illumination can be used to obtain a general overview of the adnexa and anterior structures of the eye: (a) image; (b) schematic
diagram. (Reproduced with permission from Vistakon, Synoptik and Tom Løfstrøm, Denmark.)

KEY POINT

A central corneal section can be better examined if the micro- Purkinje Images
scope is swung away from the illumination system so that ■ Purkinje I is from the front surface of the tear film, and,

the profile of the section can be viewed. following a blink, the tears can be seen moving just to the
side of the bright reflection. This can be used to assess
Specular reflection occurs when the angle of incidence tear film stability.
(illumination) and the angle of reflection (observation) are ■ Purkinje II is from the corneal endothelium/aqueous inter-

equal and opposite in relation to the normal to the surface. face, and the endothelial mosaic of hexagonal cells can
This creates a Purkinje image wherever there is a change in be seen under high (~40×) magnification in the dimmer
refractive index. reflection (see ‘Specular Microscopy’, page 170).
166 SECTION 3 • Instrumentation and Lens Design

A A

i u

X
i=u

B
Fig. 8.14 Specular reflection of the anterior surface of the crystalline
lens: (a) image of specular reflection from the lens surface; (b) schematic Offset of illumination
diagram of specular reflection of the cornea, where i is the incident
angle and u is the angle of reflection. (Reproduced with permission from
Vistakon, Synoptik and Tom Løfstrøm, Denmark.) B
Fig. 8.15 Sclerotic scatter using total internal reflection of the cornea:
(a) image; (b) schematic diagram; X = coupled; Y = uncoupled. (Repro-
duced with permission from Vistakon, Synoptik and Tom Løfstrøm, Denmark.)
■ Purkinje III is from the front surface of the lens, and the
dimpled appearance can be seen with quite a narrow angle
between the observation and illumination system (see Fig.
8.14a and b). the opposite side to the retro-illuminated surface), whereas
this is not the case for the empty space within vacuoles (see
KEY POINT
Chapter 12 and Figs 12.5 and 12.9).
The slit-lamp needs to be ‘uncoupled’ to accurately view Sclerotic scatter requires the light to be displaced to the
structures in indirect or retro-illumination or sclerotic scatter. limbus while the cornea is viewed (Fig. 8.15a and b). Light
This is achieved by physical rotation of the illumination incident on the limbus travels through the cornea by total
system to the side while maintaining the position of the internal reflection and will only be visible within the cornea if
observation system (microscope). there is an irregularity or opacity that causes light to scatter
outwards, such as rigid lens–induced oedema.
The slit-lamp is uncoupled for the following: The magnification system can be ignored and the result
Indirect illumination – light is shone to the side of the object viewed with the naked eye, but to view the central cornea
to be viewed to highlight features such as neovascularisation with magnification requires uncoupling.
that would otherwise be obscured by direct bright light (the Tangential illumination can be used to inspect the iris for
area of cornea to the right of the corneal section is seen in raised naevi. The illumination system is set parallel to the
indirect illumination in Fig. 8.12). iris and the iris observed perpendicular to the visual axis.
Retro-illumination – the illumination is reflected off the iris
or retina to view corneal or lens opacities such as vacuoles, APPLICATIONS
microcysts and cataract (see Fig. 8.12). Microcysts contain
fluid with a higher refractive index than the surrounding Table 8.1 indicates some typical conditions and the preferred
stroma, so they show reverse illumination (illumination on methods of illumination to render them most visible.
 8 • Clinical Instrumentation in Contact Lens Practice 167

Table 8.1 Application of Slit Lamp Techniques to Ocular Examination

Area/abnormality to be
Observed Magnification Illumination Method Slit Width Filter Accessory
Lids and general view of Low-medium Diffuse Wide Diffuser
external eye
Lashes Medium Direct focal Medium to wide None
Localised oedema Low Sclerotic scatter Medium Uncoupled system
Corneal opacities/defects Medium-high Direct focal/ Indirect focal Medium to narrow None
Depth of opacity/defect Medium-High Direct focal Narrow None
within cornea
Microcysts High Retro Narrow None
Striae High Indirect Medium to narrow None
Neovascular changes/ Medium-high Indirect retro Medium to narrow None
ghost vessels Green filter may aid if blood flow
Endothelium High Specular reflection Medium None
Dystrophies Low Direct retro Wide None or Neutral density
Fluorescein staining Low-medium Direct focal or diffuse Wide Blue in illumination system and yellow
(500 nm cut-off in front of observation)

For measurements such as blood vessel encroachment, contact (see Section 8, History, available at: [Link]
lens movement and opacity dimensions, the slit width or length [Link]/), a tear analyser (e.g. EasyTearView+™; see
scale on the slit-lamp can be used (although it is important to Chapter 5) and various camera attachments (see later).
check that it is calibrated). The slit can be turned through up
to 90° to take measurements at the appropriate angle. Alterna-
tively, a reticule can be incorporated into one of the eyepieces. Anterior Eye Photography
Tear Film Assessment
(Slit-Lamp Cameras)
For tear film assessment that is carried out using a slit-lamp, With modern smartphones, which have an inbuilt camera and
see Chapter 5. See also further material available at: https:// an inexpensive macro lens, it is simple to photograph and com-
[Link]/. municate any pathology of the anterior eye. This can be a good
option when a more purpose-designed system is not available, but
the illumination and camera position are not easily manipulated
GRADING
to photograph the eye and it is difficult to include patient and date
In order to make decisions based on slit-lamp findings over a details. Many new ophthalmic instruments are multifunctional
period of time or among clinicians, it is important to be able and allow diffuse or blue light anterior eye imaging in addition
to make valid comparisons. Grading scales, such as the BHVI to corneal topography, meibomian gland imaging with infrared
grading scale (see further information available at: https:// light (meibography; see Chapter 5) or retinal imaging.
[Link]/ and also Appendix B) or the Efron There are two main options for anterior segment imaging
scale, allow a given ocular feature to be gauged relative to with a slit-lamp:
predetermined images. These are chosen to represent different
degrees of the condition on an ordinal scale. Grading scales 1. A camera system can be attached to the existing eyepieces
vary in the number of images shown for each condition, but (typically a C-mount screw fitting). The main advantage
are usually descriptive, artistically rendered, photographic, is the relatively low cost although a computer image
computer morphed or combine several of these features. database storage program and often an image board are
Typically, the absence of a sign is given a grade zero on a still required. The eyepieces have optics designed for the
five-point scale. Expansion of the grading scale beyond five 60 D cornea/lens assembly, which have to be adapted
levels (such as by grading to one-tenth of a unit) increases using the camera’s optics to allow in-focus imaging by a
discrimination (Wolffsohn et al. 2015). camera. Therefore the optical path is different from that
Even using a pictorial grading scale as a reference, there of a purpose-dedicated photographic slit-lamp. Light loss
are marked differences in grades allocated by different clini- occurs at the eyepiece lens assembly but an internal beam
cians, although interobserver variability appears to improve splitter is unnecessary. The field of view of the image is
with practice (see also Section 9, Addendum, available at: also generally reduced and the camera obscures at least
[Link] one eyepiece, so the advantages of a binocular system
are lost in aligning and focusing an object of interest.
Slit-Lamp Mounted Devices 2. The more common slit-lamp camera involves a beam splitter
There are various attachments available to use with the slit- being inserted into the optical path of the slit-lamp when
lamp. Probably the most commonly used is the Goldmann photography is required. The use of a beam splitter still
tonometer but others are also available including a pachymeter allows binocular viewing through the eyepieces, and hence
168 SECTION 3 • Instrumentation and Lens Design

storage. The key aspects to be aware of with a slit-lamp digital

camera system are:
■ The digital chip:
■ Type – CMOS (complementary metal-oxide semiconduc-

tor) chips* are similar to charged couple devices (CCD),

but both photon-to-electron and electron-to-voltage con-
versions are conducted within the pixel together with
digitisation of the signal, leaving less room for the light-
sensitive part of the sensor. Normally a microlens is used
to capture more light within the pixel area and bend
it towards the light-sensitive part (the fill factor) of the
pixel. The CMOS has the advantage of being cheaper and
less power hungry than the CCD due to having fewer
components, making it more reliable. The development
of smartphones has greatly improved this technology.
■ Resolution – the number of pixels in the horizontal mul-

tiplied by the vertical meridian. The higher the number,

the more detailed the image (if the pixel size is main-
tained) but the larger the image file. However, viewing
an image of higher resolution than the monitor can
Fig. 8.16 Topcon slit-lamp with a beam splitter attached to a Nikon
resolve will reduce image quality (Peterson et al. 2005).
■ Size – the larger the chip (normally 1 to 1 inch), the
camera. 4
larger the light-sensitive area and hence the image
contrast for a certain exposure time. Each pixel receptor
will be larger on a larger chip of the same resolution
as a smaller chip. The larger the pixel receptor target,
the more chance the photon has of hitting it.
■ Colour rendering – digital chips convert light into elec-

trons so colour can be extracted by coating each pixel in

a different colour, spatially arranged in a mosaic pattern
(providing twice as many green as red or blue pixels)
interpolating colour data from the surrounding pixels
(100% spatial, but only 90–95% spectral accuracy).
Alternatively, the light can be split with prisms onto three
chips, each with a different-coloured filter (red, green
or blue) allowing 100% spatial and spectral efficiency.
These cameras are more expensive, delicate, heavy and
bulky than single-chip cameras and due to the light
loss from the two beam splitters, require a higher light
output from the slit-lamp for equivalent performance.
■ Connection interface to the computer – generally Firewire
(transfer speeds of ≥50 (Megabits per second) MBps) or
USB (up to 60 MBps with USB2 and 640 MBps with USB3)
that provide autoconfiguration and plug-and-play technol-
ogy. The speed of transfer will affect the temporal resolution
Fig. 8.17 CSO-integrated digital image slit-lamp camera with the shutter (refresh rate) of the display, which is especially important
button on the joystick. with high-resolution chips.
■ File capture format – although disk space is becoming less of
a limiting factor, storing large image files can slow the speed
the camera only receives about 50% of the available light of recall from patient management systems. Photographs
(depending on the reflectance of the beam splitter) (Fig. are stored as bitmaps, with the image divided into pixels
8.16). Some systems use a commercial camera features and the colour of each recorded. Each camera manufacturer
such as autofocusing disabled (usually by selecting infinity captures the information in a proprietary format (RAW form)
(‘mountain’) viewing), while others integrate the camera but converts this to standardised forms such as:
into the beam splitter module (Fig. 8.17) generally allow- ■ TIF (tagged information file) – a lossless format, storing
ing better control through the associated software. Apart all the data from the camera once its internal processing
from being more compact, the integrated camera has the (such as colour interpolation) has taken place. However,
shutter button on the slit-lamp joystick so that the focus- the stored images are still relatively large.
sing and image capture can be done with one hand.

It is advisable to try different slit-lamp cameras before *An etched pixelated metal oxide semiconductor made from silicone, sensi-
deciding which to buy and also take into account image tive in the visible and near-infrared spectrum.
 8 • Clinical Instrumentation in Contact Lens Practice 169

■ JPEG (joint photographic experts group) – a compressed

format, which attempts to eliminate redundant or unnec-
essary information (lossy compression). Some systems
offer a modified JPEG file type known as an exchangeable
image file (EXIF) which store ‘tags’ onto the header of
the image file containing technical data such as time,
exposure settings and camera make. This feature allows
documented proof of when an image was captured,
should it be needed for litigation protection.
■ BMP – Microsoft Windows native bitmap format. It is

most commonly used as an uncompressed format, so

file sizes can be large.
■ Image capture – the physical action necessary to capture an
image differs among systems, with more advanced systems
overriding the software or camera control in favour of a
foot-pedal or joystick button.
■ Optics – the quality of any camera image can only be
as good as the lens system which captures the light and
focuses it on the light receptor. This is even more critical Fig. 8.18 Burton or UV lamp. (Courtesy of Lynne Speedwell.)
with smaller-sized light-receptor chips.
■ Illumination – digital cameras have the ability to ‘turn
on’ the light receptors for a set period of time (electronic
shutter) with short exposures reducing blur from motion
at the expense of the amount of light captured. Additional
lighting (often from a fibre optic connected directly to
the slit-lamp illumination source) is essential for opto-
metric imaging due to the loss of light from intervening
beam splitters and lenses, incomplete fill factor of the
sensor pixels and a reduced light sensitivity of the chip
substrate compared with the human eye. CCD and CMOS
photoreceptors are more responsive to the red end of the Fig. 8.19 Pachmate hand-held ultrasonic pachymeter.
spectrum. Therefore they often have an infrared filter and
compensate for the low blue sensitivity by amplifying blue
CORNEAL PACHYMETRY (OR PACHOMETRY)
signals within the image processing. The blue illumination
channel is likely to exhibit more noise than the red or Corneal pachymetry (pachometry) is the measurement of
green channels, and fluorescein viewing with blue light corneal thickness. It is useful both as a clinical and research
can be a good way to examine the quality of a digital tool and is used especially to assess corneal thickness before
camera (Wolffsohn, 2008). and after refractive surgery and to monitor keratoconus, assess
■ Software – purpose-designed systems usually allow not only corneal oedema and as a correction factor in the calculation
the image to be captured, but also the image to be enhanced of intraocular pressure.
and annotated. Most allow connection to an anterior eye There are various pachymeters (pachometers) available
image-capture device as well as a fundus camera as part including:
of a paperless patient management system. ■ optical pachymeter – an attachment for the slit-lamp (see
Section 8, History, available at: [Link]
[Link]/)
Other Instruments for Assessing ■ ultrasound (Fig. 8.19) – e.g. Pachmate® hand-held ultrasound
pachymeter. At each major corneal interface, an echo is evident
the Anterior Segment on an ultrasound trace. Once calibrated, the distance between
the epithelial and endothelial echoes will provide a measure
BURTON LAMP of corneal thickness that is converted to a digital readout.
■ optical coherence tomography (OCT) (see below)
A Burton lamp or hand-held UV lamp (Fig. 8.18) consists of ■ Scheimpflug imaging – incorporated into some corneal
a wide magnifying lens (typically 2×) with white light and topographers such as the Pentacam (see Fig. 8.7) and
near ultraviolet radiation (UV) light sources. It is used to Oculis Corvis (Fig. 8.24)
evaluate the fit of RGP lenses or to measure pupil size under ■ raster scan topography (swept across the scan, line by line)
dim illumination (as the pupils are relatively insensitive to ■ specular microscopy (see below)
UV and hence do not constrict). It is particularly useful for ■ confocal microscopy where the physical movement of the
viewing both eyes simultaneously such as when comparing probe required to image from the front of the epithelium to
the position of the lens fit in both eyes. When using a Burton the endothelium is recorded (see ‘Confocal Microscopy’).
lamp, the head position is a much more natural one than
on a slit-lamp, so the position and movement of the contact The techniques all use the refractive index of each layer
lens is likely to better emulate real life. to determine how much compensation is needed to slow
170 SECTION 3 • Instrumentation and Lens Design

down the light waves (or sound in the case of ultrasound) The Konan CellCheck is a noncontact endothelial camera
as they pass through the corneal layers. Some light or sound that has an imaging field of 0.1 mm2. It allows repeatability of
is reflected back to the sensor at each interface as the refrac- the specular images, which helps relocate a small area of cells
tive index changes. in a cornea. It also incorporates a pachymeter. Autoanalysis
provides information about cell count and cell morphology
(see Chapter 3).
SPECULAR MICROSCOPY – See Chapter 3
Corneal sensitivity is occasionally tested in contact lens prac- CONFOCAL MICROSCOPY
tice, in particular in patients who have had corneal disease
or fifth nerve palsy. If it is found to be reduced, contact lens Confocal microscopy was patented by Minsky in 1957. A
fitting should proceed with caution. Corneal sensitivity varies modern confocal microscope allows all layers of the cornea
depending on the location in the eye (see Chapter 3). to be viewed down to cellular level and is useful when exam-
Sensitivity is traditionally measured with a Cochet-Bonnet ining for corneal disease and, corneal dystrophies as well
aesthesiometer, which consists of a thin nylon thread held as monitoring contact lens–induced changes. The confocal
perpendicular to the cornea with the patient reporting when microscope views endothelial cells at higher resolution than
they are aware of discomfort. The shorter the thread length the specular microscope as the area examined is smaller.
used, the higher the stimulus intensity. An alternative is the Huang et al. (2017) found that automated cell counting with
noncontact air-jet aesthesiometer, which uses a controlled confocal microscopy was superior to a specular microscope
jet of air to elicit ocular surface sensitivity. The stimulated in cases of Fuchs endothelial dystrophy but they suggested
receptors can be investigated in more detail by changing the that automated counting overestimated cell counts compared
temperature and gas used (Murphy et al. 1998). with manual counting.
In standard microscopy, only a single layer of an object
can be viewed in the focal plane. However, layers above and
SPECULAR MICROSCOPY
below the focal plane still contribute to the image albeit out
David Maurice was the first to develop a specular microscope of focus. In a confocal microscope, a pinhole or slit aperture
in 1968. Specular microscopy utilises the difference in refrac- is inserted between the object being viewed (in this case, the
tive index at the corneal endothelium/aqueous interface to cornea) and the eyepiece. This illuminates a single point of
view and photograph the endothelial cells (see above ‘Purkinje tissue which is then viewed by a camera in the same plane
II’ and ‘Specular reflection’ page 165). Individual corneal as the light source (i.e. confocal) (Tavakoli et al. 2008).
endothelial cells can be seen because there is a difference To increase the area seen in laser scanning confocal
between the reflectivity of the cell bodies and the aqueous microscopy, a pair of oscillating mirrors raster scan (line
which is not present at the cell borders (Böhnke & Masters by line) the laser light across the object being viewed via
1999). Light is projected onto the cornea and the reflected the objective lens.
image captured from the endothelial/aqueous interface. Ini- Fluorescence emitted by the object (cornea) passes back
tially an applanating cone was used to stabilise the image, but through the mirror systems to a beam splitter which blocks
noncontact microscopy is now available which will analyse out-of-focus images, and the remaining fluorescence then
the cells in the area covered. passes through the pinhole or slit to generate the image
Being able to view the endothelium in detail enables the (Fig. 8.20a).
clinician to monitor endothelial dystrophies and to make deci- There are several confocal microscopes available. Fig. 8.20b
sions about the most appropriate contact lens modality and shows the Nidek Confoscan 4, which has a scanning mirror
material. For example, if the endothelial cell count is low over- confocal system and uses slits to illuminate the tissue and
night lens wear is not advisable as oedema is likely to develop. to filter unwanted reflected light. It can scan each corneal

cornea
scanning mirror

camera

front lens

fixed slits
fixation lights

light source
A B
Fig. 8.20 (a) Diagrammatic representation of scanning mirror confocal optical system. (b) Nidek confocal microscope. (Reproduced with permission
from Nidek.)
 8 • Clinical Instrumentation in Contact Lens Practice 171

■ to assess tear volume through imaging the tear meniscus

(Del Aguilla-Carrasco et al. 2015; see Chapter 5)
■ to evaluate anterior chamber angles in glaucoma.

OCT technology can be divided into two distinct subar-

eas: time domain OCT (TD-OCT) and Fourier domain OCT
(FD-OCT). FD-OCT can be further split into: spectral domain
OCT (SD-OCT) and swept source OCT (SS-OCT).
Time domain OCT (so called because the interference pat-
terns were produced as a function of time) was the original
OCT system. It has generally been superseded by spectral
domain and swept source which scan at much faster speeds
and produce higher resolution images. Time domain OCT
relies on the principle of interference of broad-spectrum,
low-coherence (white light) sources, and it physically moves
a mirror in order to scan. For further details, see Section 9,
Addendum, available at: [Link]
Fourier Domain OCT
Fig. 8.21 The Zeiss Cirrus HD-OCT machine. (Courtesy of Lynne Speedwell.) Spectral domain OCT – The first SD-OCT system was demon-
strated in 1995 (Fercher et al. 1995) using a broad-spectrum
source similar to TD-OCT systems (see Section 9, Addendum,
available at: [Link] The major
difference is that the reference arm (reference mirror) does not
move when scanning, giving the system a far faster operating
capability, and the balance detector is replaced by a system
of diffraction gratings and a line scan camera.
Swept source OCT is a further advancement of the FD-OCT
concept. The system shares similar components to SD-OCT
but swaps the diffraction gratings and line scan camera
for a frequency swept light source and a photodiode detec-
tion system (Fig. 8.23). The timing between pulsed laser
and data capture is controlled via a precision timing line
giving interference data for a wide spread of frequencies.
Fig. 8.22 An OCT image of a cornea showing inserted Kerarings for Postprocessing (or quality-improvement digital image
keratoconus (see Chapter 20). The image is taken using an Optovue processing) uses Fourier transforms to construct depth and
OCT (Haag-Streit). (Courtesy of Tony Phillips.) reflection information as with FD-OCT. Improved technol-
ogy provides A scan frequencies of 370 kHz (Huber & Adler
2006) and light sources with 170-nm spread (Chong et al.
layer separately, both centrally and pericentrally, and measure 2008).
corneal thickness (pachymetry). (For further information, see Section 9, Addendum, avail-
able at: [Link]
OPTICAL COHERENCE TOMOGRAPHY (OCT)
Optical Coherence Tomography Resolution
(See the Section 8, History, and also Section 9, Addendum, All biological tissue, such as the ocular optical components,
available at: [Link] has different properties of scattering, absorbing and reflect-
An optical coherence topographer (Fig. 8.21) uses a laser ing light at differing wavelengths. Generally, the lower the
scanner to produce high-resolution, three-dimensional tomo- frequency, the less absorption is experienced, thus the better
graphical images. As well as its better-known use in retinal the depth penetration achieved. For ophthalmic imaging
imaging, it can also be used for imaging of the anterior of the anterior chamber, typically ≥3 mm past the corneal
segment of the eye. In particular: apex needs to be visualised, requiring a coherence length
■ to examine the cornea (Fig. 8.22), its shape and its effect of ≥6 µm.
on lens fit (Shen et al. 2011, Wolffsohn et al. 2013)
■ to aid rigid/scleral lens fitting (Luo & Jacobs 2012; see MULTIFUNCTION INSTRUMENTS
Chapter 14)
■ to measure and monitor corneal thickness and curvatures As mentioned previously, some instruments such as the
in: Pentacam (see Fig. 8.7) and the Orbscan are able to carry
■ keratoconus out several functions. The Oculus Corvis® ST (Fig. 8.24)
■ before and after refractive surgery measures intraocular pressure and, using Scheimpflug tech-
■ before and after corneal transplants nology, simultaneously determines the corneal thickness
■ incorporated within surgical microscopes to aid incision (pachymetry) and records the deformation or applanation
making and intraocular lens placement (El-Haddad & of the cornea to a defined pulse of air. The software is com-
Tao 2015) patible with the Pentacam topographer, and the combined
172 SECTION 3 • Instrumentation and Lens Design

Swept Source
Frequency-Domain OCT
Reference mirror

ω (1)
ω (2)
Collimating
ω (3) lens

G
Sc ani
al

an c m
v
High Coherence

ni irr
Beam

ng o
Swept Source
splitter
Narrow Spectrum

r
A-scan
Condensing
Voltage

lens

Distance Photo
diode
FFT
Eye
Light intensity

DAQ

Intensity profile
Fig. 8.23 Diagrammatic representation of a Swept source Frequency-Domain OCT.

data provide biomechanical information about the cornea,

in particular, keratoconus detection and risk of developing
ectasia after corneal refractive surgery.

More information on topography is available at: https://

[Link]

ACKNOWLEDGMENT
With thanks to Lynne Speedwell for her help in updating
this chapter.
References
Böhnke, M., Masters, B.R., 1999. Confocal microscopy of the cornea. Prog.
Retin. Eye Res. 18 (5), 553–628.
Cairns, G., McGhee, C.N., Collins, M.J., et al., 2002. Accuracy of Orbscan
II slit-scanning elevation topography. J. Cataract Refract. Surg. 28 (12),
2181–2187.
Cho, P., Lam, A.K., Mountford, J., et al., 2002. The performance of four dif-
ferent corneal topographers on normal human corneas and its impact on
orthokeratology lens fitting. Optom. Vis. Sci. 79 (3), 175–183. Erratum
in: Optom. Vis. Sci. 2002;79 (7), 462.
Chong, C., Morosawa, A., Sakai, T., 2008. High-speed wavelength-swept
laser source with high-linearity sweep for optical coherence tomography.
IEEE J. Sel. Top. Quantum Electron. 14, 235–242.
Dave, T., 1998. Current developments in measurement of corneal topog-
raphy. Cont. Lens Anterior Eye 21 (Suppl. 1), S13–S30.
Durr, G.M., Auvinet, E., Ong, J., et al., 2015. Corneal shape, volume,
Fig. 8.24 Oculis Corvis®. (Courtesy of Lynne Speedwell.) and interocular symmetry: parameters to optimize the design of
 8 • Clinical Instrumentation in Contact Lens Practice 172.e1

For more information on topography, please refer to https:// This website has a wealth of information on topography and
[Link]/user/MedmontPtyLtd/videos novices to topography should first watch ‘Basic Functions of
the E300 Topographer’. (permission kindly supplied to the
editors by Randy Kojima, Clinical Advisor, Medmont and
Jason Anderson, New Development Manager, Medmont
International.)
 8 • Clinical Instrumentation in Contact Lens Practice 173

biosynthetic corneal substitutes. Invest. Ophthalmol. Vis. Sci. 56, 4275– Minsky, M., 1961. Microscopy apparatus. US patent 3013467 A, filed
4282. November 1957.
El-Haddad, M.T., Tao, Y.K., 2015. Automated stereo vision instrument track- Morris, J., Stone, J., 1992. The slit-lamp biomicroscope in optometric practice.
ing for intraoperative OCT guided anterior segment ophthalmic surgical Optometry Today 7 September, 26–28; 5 October, 16–19; 2 November,
manoeuvres. Biomed. Opt. Express 6, 3014–3031. 28–30.
Fercher, A., Hitzenberger, C., Kamp, G., et al., 1995. Measurement of intra- Murphy, P.J., Lawrenson, J.G., Patel, S., et al., 1998. Reliability of the non-
ocular distances by backscattering spectral interferometry. Opt. Commun. contact corneal aesthesiometer and its comparison with the Cochet-Bonnet
117, 43–48. aesthesiometer. Ophthalmic Physiol. Opt. 18 (6), 532–539.
Hilmantel, G., Blunt, R.J., Garrett, B.P., et al., 1999. Accuracy of the tomey Nieves, J.E., Applegate, R.A., 1992. Alignment errors and working distance
topographic modeling system in measuring surface elevations of asym- directly influence the accuracy of corneal topography measurements.
metric objects. Optom. Vis. Sci. 76 (2), 108–114. Invest. Ophthalmol. Vis. Sci. 33, 993.
Huang, J., Maram, J., Tepelus, T., et al., 2017. Comparison of noncontact Peterson, R.C., Wolffsohn, J.S., 2005. The effect of digital image resolution
specular and confocal microscopy for the evaluation of corneal endothe- and compression on anterior eye imaging. Br. J. Ophthalmol. 89, 828–830.
lium. Eye Contact Lens. pii: S1888-4296(17)30049-3. doi:10.1016/j. Peterson, R.C., Wolffsohn, J.S., Fowler, C.W., 2006. Optimization of anterior
optom.2017.06.001. [Epub ahead of print]. eye fluorescein viewing. Am. J. Ophthalmol. 142, 572–575.
Huber, R., Adler, D., 2006. Buffered Fourier domain mode locking: unidi- Shen, M., Cui, L., Riley, C., 2011. Characterization of soft contact lens
rectional swept laser sources for optical coherence tomography imaging edge fitting using ultra-high resolution and ultra-long scan depth optical
at 370,000 lines/s. Opt. Lett. 31, 2975–2977. coherence tomography. Invest. Ophthalmol. Vis. Sci. 52, 4091–4097.
Jain, R., Grewal, S.P.S., 2009. Pentacam: Principle and clinical applications. Stockwell, H., Stone, J., 1988. Anterior eye examination. In: Edwards, K.H.
J. Curr. Glaucoma Pract. 3 (2), 20–32. (Ed.), Optometry. Butterworth-Heinemann, Oxford.
Luo, Z.K., Jacobs, D.S., 2012. Current and potential applications of anterior Wolffsohn, J.S., 2008. Doshi, S., Harvey, W. (Eds.), Eye Essentials: Ophthalmic
segment optical coherence tomography in contact lens fitting. Semin. Imaging, first ed. Butterworth Heinemann. eBook ISBN: 9780702039201.
Ophthalmol. 27, 138–142. Wolffsohn, J.S., Drew, T., Dhallu, S., et al., 2013. Impact of soft contact lens
Maseedupally, V., Gifford, P., Swarbrick, H., 2015. Variation in normal edge design and midperipheral lens shape on the epithelium and its inden-
corneal shape and the influence of eyelid morphometry. Optom. Vis. tation with lens mobility. Invest. Ophthalmol. Vis. Sci. 54, 1690–1696.
Sci. 92, 286–300. Wolffsohn, J.S., Naroo, S.A., Christie, C., et al., 2015. Anterior eye health
Maurice, D., 1968. Cellular membrane activity in the corneal endothelium recording. Cont. Lens Anterior Eye 38, 266–271.
of the intact eye. Experientia 24 (11), 1094–1095. Tavakoli, M., Hossain, P., Malik, R.A., 2008. Clinical applications of corneal
McCarey, B.E., Zurawski, C.A., O’Shea, D.S., 1992. Practical aspects of a confocal microscopy. Clin. Ophthalmol. 2 (2), 435–445.
corneal topography system. CLAO J. 18, 248–254. Yeung, K.K., Kit, B., 2015. Diagnostic imaging technology for corneal
McMahon, T.T., Anderson, R.J., Joslin, C.E., et al., 2001. Precision of three analysis. Rev. Corn. Contact Lens June 2015.
topography instruments in keratoconus subjects. Optom. Vis. Sci. 78 Ying, J., Wang, B., Shi, M., 2012. Anterior corneal asphericity calculated
(8), 599–604. by the tangential radius of curvature. J. Biomed. Opt. 17 (7), 075005.