Corneal topographers are gaining popularity as a valuable addition to the consulting room. This article considers the clinical applications of the instruments, how to interpret the data they capture, and their use in the timely detection of keratoconus.
The cornea is the first major refractive surface of the eye and is responsible for over two thirds of its overall dioptric power. Therefore, relatively small changes in corneal shape – secondary to surgery or disease – can have a substantial impact on clarity of vision. Mapping of the corneal shape through the use of corneal topographers enables eye care practitioners to accurately analyse this important structure. Despite their availability, topographers have not been commonplace in High Street practices and have historically been associated with hospital optometry, refractive surgery clinics, and those specialising in advanced contact lens fitting. Now, with exposure to the benefits of topography in tear film/dry eye analysis, myopia intervention, and its use as a diagnostic tool in relation to corneal irregularity and potential visual quality, we are seeing a rise in the number of topographers in optometric practice.
Techniques used to measure corneal topographyCurrently, most eye care practitioners use keratometers to measure corneal curvature. However, this technique only measures 6-7% of the corneal area and only gives an average curve over the central 3mm, with no information on peripheral curvature.1 Therefore, conditions which exhibit peripheral corneal changes cannot typically be detected and monitored with keratometry.
Gross corneal topography was first assessed by Placido in 1880 by projection of a simple concentric ring target onto the cornea.1 Many modern day topography systems continue to use the technique of Placido disc projection to measure corneal curvature, with information captured using specialised video systems. With the use of computer-aided complex software and the capability to translate the information captured by the camera into useful information on corneal shape, these systems are also referred to as videokeratography.
Placido-based topographers comprise the majority of units in optometric practice today. The Topcon CA800, Medmont E300 and Oculus Keratograph 5M are just three examples of currently available Placido-based topographers; these systems use the tear film as a convex mirror to reflect a series of concentric rings (see Figure 1: Placido-based topographers project concentric rings onto the cornea. The tear film must be good quality to ensure complete, smooth rings are formed (left). Tear thinning and bear-up can be seen when the rings become distorted (right)).2 Corneal shape is assessed by analysis of the regularity and separation of the reflected rings to give curvature and power information. Placido disc-based systems cannot obtain true corneal height information. This type of topographer requires a good quality tear film to enable accurate measurement; therefore, it is advisable to ensure the patient has a good blink immediately before capture. Alternatively, artificial tears can be instilled if required. Placido-based topography systems have been shown to provide repeatable measurement of corneal topography in healthy and keratoconic eyes.3
Scheimpflug topography (a type of projection topography) is an alternative method of measuring corneal curvature, a technique, which is utilised by the Oculus Pentacam. With this technique, the cornea is illuminated with a slit of light, causing backscatter of light which is captured by a camera, oriented according to the Scheimpflug principle, thus creating a perfectly sharp image (see Figure 2: A slit projection image of the cornea captured by a Scheimpflug camera on the Oculus Pentacam. Anterior and posterior corneal surface information can be obtained with this technique)4,5 A series of radial images are captured around the eye, then combined to create a three-dimensional model of the entire anterior portion of the eye from the anterior lens to the anterior corneal surface. The rotating measurement principle used in Scheimpflug imaging avoids measurement errors that would result from horizontal scanning. Captured images are mathematically analysed to generate data on elevation, curvature and pachymetry.4 While Scheimpflug topography is considered the gold standard for the measurement of corneal shape, the high cost limits its availability in optometric practice.
Interpretation of corneal topography maps
When analysing corneal topography maps, a clear systematic approach is advised to avoid reading errors. When assessing any computer-generated patient results, the first thing to consider is the quality of the data obtained; with Placido-based topography systems, this is achieved by viewing the raw data and assessing the regularity and completeness of the concentric rings (see Figure 1). Care should be taken while capturing images using Placido-based topographers, as often the nose can block the projection of the nasal rings onto the cornea, and in these instances the patient’s head should be twisted away from the eye being imaged to enable more complete image capture.
Once image quality is checked, the scale used should be noted as this can vary and change the appearance of the topographic map. While many scales exist, most fall into two categories: ‘absolute’ and ‘normalised’ (see Figure 3: (left to right) Curvature maps showing a regular astigmatic cornea on an axial map and tangential map with absolute scale, and an axial map with normalised scale. In all curvature maps, hotter colours indicate steeper areas of the cornea, while cooler colours indicate flatter areas).6 The absolute scale uses large, fixed intervals to cover the whole scale of possible curvature values; the same scale and colours are used for all eyes. While the absolute scale can mask fine details, it should always be used to facilitate comparisons over time. The normalised scale is not fixed and varies for each eye and image. The range for the normalised scale is determined by the flattest and steepest values of the cornea it is examining. While this scale reveals fine corneal detail, care should be taken when using it as small details can appear to be magnified by an inappropriately narrow scale.6
A radius of curvature map is a common way of displaying corneal topographic data, particularly when using a Placido-based topographer. Corneas with a steep surface slope have a small radius of curvature and high dioptric power, while corneas with a flat surface slope have a large radius of curvature and low dioptric power. In curvature maps, hotter colours (yellow and red) represent steeper areas of the cornea, whereas cooler colours (green and blue) represent flatter areas of cornea. Radius of curvature can be expressed in mm or can be converted to power (in dioptres) by using the standard keratometric index (SKI), 1.3375.
The radius of curvature of a cornea can be calculated, and thus displayed in two ways: axial and tangential (see Figure 3). Axial (global/sagittal) curvature measures the curvature of each section of the cornea in relation to the optical axis, resulting in measurements that have a spherical bias.7 Localised changes in curvature and peripheral data are poorly represented, making axial maps suboptimal for the assessment of irregular corneas.6,8,9 As axial curvature maps produce large, diffuse and smooth patterns with little noise, they tend to be better for visualising regular corneal astigmatism, for contact lens fitting and for estimating the general corneal curvature. Axial curvature maps of normal corneal topography can be classified into five groups: round, oval, symmetrical bowtie, asymmetric bowtie and irregular (see Figure 4: Patterns of corneal topography as described by Bogan et al. From left to right: round (spherical cornea), oval, symmetric bowtie (regular astigmatism), asymmetric bowtie (irregular astigmatism), irregular).10 In regular astigmatism, if the bowtie is vertical, the eye has with-the-rule astigmatism; if it is horizontal it has against-the-rule astigmatism.9
Tangential (local/instantaneous) curvature measures the curvature of each point on the cornea with respect to its neighbouring points. Therefore, tangential mapping is more accurate in assessing local irregularities and for mapping the peripheral cornea, but produces maps that look a lot noisier.6,9,11 Tangential maps should always be used over axial maps for the detection and monitoring of keratoconus as they allow more accurate assessment of the cone location and power.
Curvature maps (combined with an absolute scale) can be used to produce difference maps, which allow for monitoring of a disease process over time,or for assessing the effect of surgical intervention (see Figure 5: Difference maps in an eye that has previously undergone refractive surgery. The top right map is pre-surgery, top left post-surgery. The differential map shows how the cornea has changed, with cool areas showing flattening centrally and hot areas showing peripheral steepening).
True height (elevation) maps can only be created using a projection-based topography system; height maps generated from Placido-based topographers require approximations based on presumptions about corneal shape.6 Height maps are not typically viewed in their raw form, but in a comparison to a reference sphere, showing how the corneal elevation deviates from a known shape.5 This technique magnifies the differences, giving a qualitative map that highlights clinically significant areas. In elevation maps, hot colours represent elevation above the reference sphere (that is flatter curvature), while cool colours represent areas lower than the reference sphere (that is steeper curvature). An elevation map of a normal eye has an hourglass-like pattern (see Figure 6: Elevation maps. A diagrammatic depiction of elevation maps (A – top), decribed in relation to the best-fit sphere (pale blue solid circle). The steep meridian (red line) is below the best-fit sphere, and the flatter meridian (blue line) falls above best-fit sphere. In the elevation map (A – bottom) the flatter meridian is seen as elevated above the best-fit sphere (warm colours), while the steeper meridians is seen as below the best-fit sphere (cool colours) An elevation map of a normal cornea typically shows an hourglass pattern (B)).
Height maps are particularly useful for assessment prior to laser surgery where the refractive effect is dependent upon the depth of tissue removed. They can also be useful for predicting the fit of a rigid contact lens. When the reference sphere is set to the BOZR of the contact lens, warm colours demonstrate where it would be expected for fluorescein to be displaced, while cool colours represent where the fluorescein would pool.
Detecting keratoconus with corneal topography
The most significant increase in the demand for topographers in optometric practices relates to the seismic shift in keratoconus management since the NICE approval of corneal cross-linking (CXL) for keratoconus in 201312 together with the realisation that there are certain populations within the UK where the prevalence of keratoconus is much higher and onset much younger than the average for the population as a whole. Recent studies have reported the worldwide incidence of keratoconus to be one in 1,650, compared to one in 15,000 30 years ago.13
It is likely that increasing use of topography has contributed to the rise in keratoconus prevalence due to greater detection rates compared to those obtained with more traditional examination techniques including high contrast VA, retinoscopy and ophthalmolscopy, all of which only tend to detect keratoconus when it is more advanced. The usefulness of contrast sensitivity measurements is limited by the variability of measurements. Retinoscopy in experienced hands when used on all patients, is relatively sensitive to early keratoconus but provides no information about changes over time, other than refractive change, in a group of patients who tend to show myopia progression associated with their age. It is no longer acceptable to pick up keratoconus only when it becomes a significant visual problem and referring when complex lenses or surgery are required. With CXL now proven and available, our duty of care is to detect keratoconus as early as possible, capture a base line corneal map and refer for CXL if local progression criteria are met.
Curvature topography maps of keratoconic eyes typically show inferior corneal steepening, with cone localisation best detected with a tangential map and normalised scale (see Figure 7: Curvature maps of a keratoconic left eye. While the axial map with absolute scale shows significant infero-temporal steepening (left), when viewed as a tangential map with a normalised scale, the apex of the cone can be easily identified unmasking the comparatively flatter periphery). Occasionally, early keratoconus can present on the posterior cornea. It is usually associated with anterior corneal changes but in the rare cases where it is not, it will only be identified by a slit-projection topographer or wavefront analyser. Most modern topographers now provide a probability value indicating the likeliness of keratoconus, known as the keratoconus probability index (KPI); this probability is calculated using an algorithm which incorporates several indices based upon geometrical and refractive properties of the cornea.14 Incorporation of this calculation into topography systems is particularly beneficial for practitioners who are less confident when analysing corneal topography maps. Keratoconic indices also allow quantification of changes over time – a useful tool when deciding when to refer, as well as in the monitoring of eyes post-surgery.
Corneal topography is an essential tool for the assessment of corneal curvature in complex contact lens fitting and for the detection and monitoring of keratoconus and other ectatic disorders. With the NICE approval of collagen cross-linking having profound implications for keratoconic outcomes, practitioners must have the tools and skills available to detect the earliest signs of this ectatic disease. The minimum requirement is to capture a baseline measurement, or refer to a colleague for this to be undertaken. Clinicians in primary eye care are ideally placed to monitor the progression of keratoconus, and help minimise corneal clinic waiting times by referring in accordance with local referral guidelines.
About the authors
Dr Rachel Hiscox PhD, MCOptom, is the education and clinical affairs manager for Topcon GB and an optometrist with experience in both hospital and High Street optometry. Dr Hiscox gained a PhD from Cardiff University in 2013 investigating the retina in cystic fibrosis. Professor Catharine Chisholm is an optometrist who has spent much of her career lecturing and researching, firstly at Anglia Ruskin, then City University London and more recently the University of Bradford. She left academia in 2014 to become clinical affairs manager for Topcon GB and now leads global education for the Topcon Eyecare Company.
- Dave T. Current developments in measurement of corneal topography. Contact Lens and Anterior Eye. 1998 21:S13-S30
- Mejia-Barbosa Y, Malacara-Hernandez D. A review of methods for measuring corneal topography. Optometry and vision science : official publication of the American Academy of Optometry. 2001;78(4):240-53
- Ortiz-Toquero S, Rodriguez G, de Juan V, Martin R. Repeatability of placido-based corneal topography in keratoconus. Optometry and vision science : official publication of the American Academy of Optometry. 2014;91(12):1467-73
- Belin MW, Khachikian SS. An introduction to understanding elevation-based topography: how elevation data are displayed - a review. Clinical & experimental ophthalmology. 2009;37(1):14-29
- Konstantopoulos A, Hossain P, Anderson DF. Recent advances in ophthalmic anterior segment imaging: a new era for ophthalmic diagnosis? The British journal of ophthalmology. 2007;91(4):551-7
- Corneal topography. American Academy of Ophthalmology. Ophthalmology. 1999;106(8):1628-38
- Wilson SE, Klyce SD. Advances in the analysis of corneal topography. Survey of ophthalmology. 1991;35(4):269-77
- Sinjab M. Skillful interpretation of corneal imaging. Cataract and Refractive Surgery Today. 2014;April:68-72
- Sinjab M. A guide to interpreting corneal topography. Cataract and Refractive Surgery Today. 2012;Jan:20-4
- Bogan SJ, Waring GO, 3rd, Ibrahim O, Drews C, Curtis L. Classification of normal corneal topography based on computer-assisted videokeratography. Archives of ophthalmology (Chicago, Ill : 1960). 1990;108(7):945-9
- Roberts C. Corneal topogprahy: a review of terms and concepts. Journal of Cataract and Refractive Surgery. 1996;22:624-9
- NICE. Photochemical corneal collagen cross‑linkage using riboflavin and ultraviolet A for keratoconus and keratectasia [21 August 2015]. Available from NICE
- Gokhale NS. Epidemiology of keratoconus. Indian Journal of Ophthalmology. 2013;61(8):382-3
- Rabinowitz YS, Rasheed K. KISA% index: a quantitative videokeratography algorithm embodying minimal topographic criteria for diagnosing keratoconus. Journal of Cataract & Refractive Surgery.25(10):1327-35