Introduction
Keratoconus, pellucid marginal degeneration, and iatrogenic ectasia all exhibit an irregular form of corneal ‘bulging’ secondary to progressive stromal thinning. These appear to be unrelated to any obvious inflammatory changes. The reported incidence of keratoconus among refractive surgical candidates when refractive surgery was first introduced was as high as 8 to 12 percent.1 This number has dropped to closer to one percent as refractive practices have become more established.2 Importantly, these percentages remain more than 10 times higher than the incidence of keratoconus in the general population (1.3–50 per 100,000 depending on ethnicity).3,4 Most of these patients with asymmetrical corneas experience poor vision quality with spectacle correction or are experiencing a growing intolerance to ill-fitting contact lenses. For those reasons, patients with mild keratoconus tend to seek out other modalities for correction, and excimer laser treatments are novel and attractive alternatives. Thus, it is particularly important for all refractive surgical candidates to undergo careful screening which specifically includes bilateral corneal topography examinations.
As LASIK approaches 1.4 million procedures annually in the United States alone, it has become imperative that the modern day refractive surgeon possess an increasingly expert skill set for the detection of preoperative pathology such as keratoconus and pellucid marginal degeneration as well as the development of post-surgical corneal ectatic changes. Corneal topographers have been the standard of care for preoperative screening of refractive surgical candidates since the early 1990's.5 In this chapter we will explore the basics of topographic pattern recognition essential to the detection of preoperative keratoconus as well as some of the advancing technologies that may complement our current knowledge base.
The mainstay for early detection, diagnosis and tracking of ectasia remains videokeratography, specifically the corneal topographic mapping of axial dioptric power with the color-coded contour map developed in the late 1980's.6 Utilizing Placido-based maps provides the most sensitive and reproducible method for the detection of early ectasia. The amount of information displayed in these maps is determined in part by the topographic scale. However, without the use of a standardized or absolute scale,7,8 certain irrelevant distortions can appear misleading or overemphasized (Figure 1-1).
Although modern corneal mapping systems can display upwards of 22,000 data points on a single topographic map, a 1.5 D scale has sufficient resolution to detect all of the topographic characteristics necessary for diagnosing a variety of topographic abnormalities including: contact lens warpage, early and late keratoconus, penetrating keratoplasty, extracapsular cataract extraction, photorefractive keratotectomy, radial keratotomy, and epikeratophakia.9
Conversely, an ad hoc scale can cause the clinician to overlook early signs of irregular astigmatism or fail to detect progression of ectasia when following a keratoconus suspect. Such scales such as the normalized scale are self-adapting to the power range present in individual topography examinations. This technique can obscure important detail on highly irregular corneas and over emphasize topographic features irrelevant to diagnosis (see Figure 1-1).
The hallmark of corneal ectasia remains an irregular astigmatism which can take several forms. The result is several patterns of irregular astigmatism with which the eye care professional must become familiar. Ectasia from keratoconus and ectasia following refractive surgery can be remarkably similar (Figure 1-2). Keratoconus is most often associated with an inferior localized steepening as shown in Figure 1-2, although the cone can be present centrally or even superiorly.10 However, keratoconus can also present as a central symmetric, but lopsided or ‘lazy eight’ bow tie (Figure 1-3) characterized by the skewed radial axes in the corneal topography recognized by Rabinowitz and Rasheed,11 or as an asymmetric bow tie with or without skewing.3
FIGURE 1-1: A fixed standard scale is essential for proper corneal topography interpretation. A single topography exam from a normal cornea is shown with two scales. The preferred 1.5 D interval scale used on the left panel leaves the correct impression that this cornea is regular with normal peripheral flattening. The right panel casts the same exam with a 0.5 D interval scale. The features that appear at this resolution are not of clinical significance, yet may give the incorrect clinical impression of irregular astigmatism.
FIGURE 1-2: The topographic appearance of keratoconus (left panel) and iatrogenic ectasia (right panel) can be very similar.
A final characteristic of keratoconus topography is that its progression is usually uneven between the two eyes of a patient, and a small number of patients will appear to have unilateral keratoconus.124
However, recognizing that keratoconus is a genetic disease, whose expression is variable between eyes, it is clear that if one eye of a patient exhibits keratoconus, the other eye will have the defect as well and be of similar risk for ectasia should it be treated with refractive surgery.
Another more rare form of naturally occurring progressive corneal ectasia and equally a risk factor for refractive surgical candidates is pellucid marginal degeneration.13 Pellucid marginal degeneration involves an arcuate peri-limbal thinning which differentiates it from keratoconus in its more advanced stages. However, before stromal thinning can be measured, topographic signs emerge as the only detectable measure of a potential underlying pathology. The topographical pattern of pellucid marginal degeneration is typically a ‘claw’, or ‘C’ shape as seen in Figure 1-4. Note, however, if these topographic patterns occur without pachymetry, then the interpretation must be topographic pellucid marginal degeneration. This is because keratoconus can also exhibit the pellucid topographic pattern in some cases.
Other topographic signs of suspect keratoconus in addition to inferior steepening have been proposed. Levy, et al demonstrated that a ‘J’ pattern and an ‘Inverted J’ pattern in corneal topography were over represented in relatives of patients with clearly established Familial Keratoconus.
FIGURE 1-4: Pellucid marginal degeneration has a characteristic ‘C’ shape or claw shape on topography. Note the against the rule astigmatism frequently exhibited and the tear drop inferior depression. However, diagnosis of this cornea was only confirmed by slit lamp examination showing a peri-limbal inferior arcuate band of thinning, unique to this disorder.
These minimal images of the classic asymmetric bow tie with inferior steepening were not seen with statistical significance in their study's control population.14 Other workers claim that the presence of a ‘vertical D’ pattern in corneal topography was a sign of keratoconus15 and a risk factor for refractive surgery. Detected retrospectively in two patients who developed ectasia after LASIK, such abnormalities in preoperative corneal topography serve as a warning sign and potential exclusion criteria for refractive surgical candidates.
Differentiating these irregular patterns from dry eye, contact lens warpage or simply normal variants of regular astigmatism remains one of the more difficult and critical diagnostic dilemmas facing today's eye care professional. To better elucidate early ectatic or forme fruste corneal changes from more stable states, Maeda, et al demonstrated the benefits of incorporating an expert system that classified corneal maps based on discriminant analysis and a classification tree that considers eight different topographic indexes.16 This method was shown to be more sensitive and specific to keratoconus detection than looking for elevated average Simulated Keratometry (Sim K) readings or previously utilized methods that relied on central corneal (K) power and Infero-Superior asymmetry (I-S) values.
Today this statistical approach has been extended through the use of neural networks to detect and interpret map patterns.17 The Magellan Mapper from Nidek (Gamagori, Japan) features software that includes a neural network application capable of predicting various corneal diseases and possible post-surgical outcomes. This application of artificial intelligence utilizes a previously trained set of logic rules ‘learned’ from sets of abnormal and normal patient topographies. Unique to this system is the ability to differentiate between astigmatism, keratoconus suspects, true keratoconus, and pellucid marginal degeneration.18 Furthermore, the Magellan is able to assign a percentage of probability, or grade, to these disease states. Figure 1-5 illustrates a typical Magellan printout with a traditional axial map, a grouping of indices, and an easy to interpret bar graph that includes a percentage of each classification's probability. Other categories of potential classification by the neural network include: previous penetrating keratoplasty (PKP), myopic refractive surgery (MRS), hyperopic refractive surgery (HRS) and ‘other’ (OTH). This last designation is significant because the system has the ability to recognize and differentiate irregular patterns produced by early or forme fruste keratoconus suspect (KCS) from those asymmetries seen with dry eye or contact lens warpage.
However, it is important to note that like other clinical tests corneal topographer interpreters may not be 100% accurate. In some cases, contact lens warpage can produce inferior steepening that mimics keratoconus so closely that a false report of keratoconus can be given. The particulars related to contact lens wear should always be noted in the chart for patients being screened for refractive surgery. If there is inferior steepening, contact lens wear should be discontinued and a repeat topography taken after 2–3 weeks have passed to differentiate keratoconus and contact lens warpage. With keratoconus, the inferior steepening will tend to become more pronounced as contact lenses tend to press on the cone. If the inferior is due to contact lens warpage, the asymmetry should diminish.
Modern topographers can aid in the screening for early ectatic diseases and detect subtle irregularities such as those produced from over wear of contact lenses. For those patients going on to wavefront-guided excimer laser ablations, these findings may allow the surgeon to collect a more accurate portrait of patient wavefront information and thus produce better postoperative results.6
FIGURE 1-5: Screen shot of the NIDEK Corneal Navigator automatically interpreting keratoconus (KC) with a 94.5% likelihood that the topography matches keratoconus. The severity (Keratoconus Severity Index, KSI) is given as 6.6%.
The above discussion centers on the use of the axial power map for displaying corneal topography. It is noted that there are three other common topography displays on most corneal topographers. These include the tangential or instantaneous power map, the refractive power map, and a height or elevation map. The former two are in units of diopters while the latter is in units of mm or microns. The elevation maps are presented as the difference between a best fit sphere and the measured shape of a corneal surface. There are also scanning slit-based instruments (see below) that measure the position of the two corneal surfaces to yield pachymetry maps which are an important adjunct to corneal topography in the screening of patients for refractive surgery as well as differentiating keratoconus from pellucid marginal degeneration by the pattern of thinning.
Tangential power maps can be useful to show details of refractive surgery, in particular the characteristics of the transition zone after the correction of myopia. Abrupt transition between the optical zone and the peripheral corneal can produce symptoms of haloes and monocular polyopia. Refractive power maps present the true refractive power of the corneal surface, but are not generally 7useful for clinical diagnosis. In fact, refractive power maps can obscure mild inferior keratoconus or pellucid marginal degeneration, since with the refractive power display, the algorithms used to calculate refractive power display a steepened peripheral cornea. The steep periphery in the refractive power map can mask the peripheral steepening characteristic of mild keratoconus.
Slit Scanning and Scheimpflug Imaging
By providing anterior and posterior corneal surface information, along with whole corneal pachymetry, the Orbscan (Bausch & Lomb, Rochester, NY), the Pentacam (Oculus, Inc., Lynnwood, Washington), and the Galilei™ Dual Scheimpflug Analyzer (Ziemer Ophthalmic Systems AG, Port, Switzerland) imaging systems can further aid in the detection and progression of corneal ectasia.
The Orbscan uses two scanning slit lamps to project a series of 40 slit beams on the anterior cornea, posterior cornea, anterior iris and anterior lens. This system is combined with a calibrated video and Placido disk imaging. Figure 1-6 shows a typical Orbscan quad map. This provides elevation displays created as the difference between corneal surface elevation and the best fit sphere (the so-called anterior and posterior “float” displays).
FIGURE 1-6: Keratoconus examination with the Orbscan II. Anterior (top left) and posterior (top right) floats provide the difference in elevation between the corneal surfaces and best fit spheres. Axial topography (bottom left) and pachymetry (bottom right) are shown. The thinnest point on this cornea is 597 microns illustrating the fact that keratoconus is not always associated with an abnormally thin cornea.
The Orbscan also displays the more clinically useful traditional Placido disk keratometric topography and full corneal thickness measurements over a broad area of the cornea.
When screening refractive surgery candidates, several ‘red flags’ or indices have been associated with possible signs of early ectasia.19 With the Orbscan, these include: pachymetry readings with a thinnest point less than a certain threshold (470-500 microns), a minimum peripheral corneal thickness that is not at least 20 microns greater than the central cornea, posterior float greater than 50 microns,20 high irregularity indices at the 3 mm and 5 mm zones, and the overall correlation of the highest/thinnest point coinciding on the anterior, posterior and pachymetry maps. Additional risk factors have been identified by Randleman and colleagues.13 The strongest correlate was found to be abnormal topography, but other risk factors included high myopia, reduced preoperative corneal thickness, reduced residual stromal bed after refractive surgery, and age. Note that these indications are regarded as warning signs that have been correlated with keratoconus; the signs of keratoconus seen in the axial power map from the Placido topography remains the most sensitive and reliable clue to the presence of keratoconus. There are many reports of large series of refractive surgical patients who have had one or more of these warning signs (but with normal topography) and who have not developed iatrogenic ectasia. Again, apart from axial topography, none of these warning signs provide a strong indicator of forme fruste keratoconus by themselves. In addition, topographic irregularities at both the 3 mm and 5 mm zones may simply point to the presence of increased higher order aberrations without specific pathology. Further, the repeatability of peripheral slit-based pachymetry measurements is somewhat controversial compared to the more consistent central corneal thickness values.21
Note also, that corneas with scarring or moderate dry eye will frequently produce inaccurate pachymetry maps as the scarred or dry surface may be interpreted as the posterior corneal surface.
With the Pentacam, by orienting the camera lens and lens plane at intersecting angles, the Scheimpflug camera is able to record the corneal surfaces directly. Direct central recording is unavailable with the traditional Placido devices which places the viewing lens in the center of the Placido mires. Similar to the Orbscan, Pentacam slit images are gathered that image both surfaces of the cornea as well as the surface of the crystalline lens. Although slit-based corneal front surface data is not as high in resolution as Placido imagery, in cases of advanced ectasia, Placido mires become obscured while the slit images can be used to present topography (Figure 1-7). This is a major advantage of slit-based topography.
For screening purposes, it is claimed that Pentacam anterior elevation values between +12 to +15 microns (above a reference sphere) are suspicious for ectasia and should prompt further investigation, whereas, a central deviation of the cornea's anterior elevation of more than +15 microns is indicative of keratoconus.22 However, these guidelines have not received rigorous experimental proof.
The Ziemer Galilei is a dual Scheimpflug instrument that, like the Orbscan, has also a Placido function. This instrument combines slit and Placido data in displaying corneal topography.
Wavefront Sensing
It is becoming more apparent that wavefront data can further enhance our topographic diagnostic abilities. Maeda, et al showed that wavefront aberrometers may provide additional clues for the detection of early corneal ectasia.23 The authors compared the total eye wavefront aberrations in normals to those in keratoconic eyes. An increase in the total higher order aberrations was noted in keratoconus and attributed to the corneal shape. Coma-like aberrations were dominant and increased in the keratoconus eyes.9
FIGURES 1-7A AND B: Comparison of the Humphrey map (A) with the Pentacam map (B) in the same eye of a keratoconus patient. With very steep corneas, the slit-based topography is able to track the surface, while Placido mires can merge with the loss of data. The data is compared with the same scales using VolPro software (Sarver and Associates, Carbondale, IL) (Topography exams courtesy of Renato Ambrósio, MD, PhD).
Moreover, subsets of corneal ectasia have been shown to produce unique wavefront profiles. Pepose and Applegate demonstrated that patients with pellucid marginal degeneration could be differentiated from keratoconus based on wavefront data.24 The patients with pellucid marginal degeneration were noted to possess higher amounts of peripheral aberrations (especially trefoil), whereas the keratoconus patients tended to show higher degrees of coma as in the Maeda study. An example of wavefront analysis of a keratoconus cornea is shown in Figure 1-8.
Corneal Hysteresis
To date, the bulk of our biomechanical corneal knowledge arises from the measurement of its geometrical aspects such as topography and pachymetry. When attempting to diagnose and treat a poorly understood corneal progressive thinning disease like keratoconus, any information regarding the biomechanical properties of the cornea would be welcome. The Reichert Ocular Response Analyzer (ORA; Reichert, Buffalo, NY) provides measurement of Corneal Hysteresis (CH) and the Corneal Resistance Factor (CRF) which are the result of viscous damping in the corneal tissue. The ORA utilizes a rapid pulse of air, and an advanced electro-optical system to record the displacement or deflection of the corneal surface before, during, and after the perturbation. The signal obtained during this process is shown in Figure 1-9. The response measured by the ORA is a complex combination of corneal and possibly other ocular biomechanical properties.
Importantly, the ORA provides a repeatable, Goldmann tonometer-correlated intraocular pressure (IOP) measurement. The additional parameters of Corneal Hysteresis and the Corneal Resistance Factor are being studied to determine whether they can be utilized for the detection of keratoconus. Early studies suggest a relationship between CH and the presence of keratoconus, although there is a great deal of overlap between normal corneas and those with keratoconus.25–27 It is hoped that CH and CRF might be useful as diagnostic tools for determining who might be at risk for developing post-refractive ectasia.10
FIGURE 1-8: Wavefront analysis of keratoconus corneal topography using the Nidek Magellan topographer. With this cornea, there is a significant amount of spherical aberration, coma (usually a dominant aberration in keratoconus), and higher order aberrations. The residuals map shows the corneal wavefront that is not fit by the 6th order Zernike series.
It should be noted that interpretation of CH and CRF is confounded by the fact that the entire globe and lamina cribosa are involved in the response to an air pulse delivered to the corneal surface. Hence, the characteristics of the response may not reflect only corneal biomechanics, but include those of other ocular components as well. Further, to date, no study has produced evidence that the early changes in the cornea associated with the forme fruste or mild stages of keratoconus can be detected with this approach.
In conjunction with CH and CRF, the other signal waveform characteristics may also yield clues to the presence of altered biomechanics as occurs in keratoconus. Signal analysis of a normal eye shows great symmetry and height between peak 1 and peak 2 and a smoother, less erratic waveform (Figure 1-9). The signal waveform peaks of a keratoconic eye appears altered and CH and CRF can be lower than normal (Figure 1-10). Current research is focusing on further analysis of the waveform characteristics to see if additional biomechanical properties can be gleaned, particularly from the earliest onset of pathology. Success may yield another test that can be used to assess risk factors associated with refractive surgery.
FIGURES 1-10A AND B: Magellan corneal topography (A) of a keratoconus patient and the scan from the ORA (B). Note the reduction in amplitude of the second peak, and additional ‘noise’ on the raw signal.
Conclusion
In order to detect early ectatic disease, all refractive surgery candidates should have corneal topography examinations prior to any elective procedure. The Placido corneal topographer with its axial power map remains the most reliable method to screen for forme fruste keratoconus or forme fruste pellucid marginal degeneration. Contact lens history must be considered prior to conclusive diagnosis. Always use zonal pachymetry whether from multiple ultrasound readings (central, nasal, temporal, superior, and inferior) or from scanning slit or Scheimpflug map data. Although no one single diagnostic tool may yet provide the screening “crystal ball” desired to predict future disease in all cases, following the topographic guidelines outlined above will greatly enhance the ability to avoid potential postoperative complications12
References
- Kalin NS, Maeda N, Klyce SD, Hargrave S, Wilson SE. Automated topographic screening for keratoconus in refractive surgery candidates. CLAO J 1996;22:164–7.
- Ambrósio R, Klyce SD, Wilson SE. Corneal topographic and pachymetric screening of refractive surgery patients. J Refractive Surg 2003;19:24–9.
- Nielsen K, Hjortdal J, Aagaard Nohr E, Ehlers N. Incidence and prevalence of keratoconus in Denmark. Acta Ophthalmol Scand. 2007;85:890–2.
- Assiri AA, Yousuf BI, Quantock AJ, Murphy PJ. Incidence and severity of keratoconus in Asir province, Saudi Arabia. Br J Ophthalmol 2005;89:1403–6.
- Wilson SE, Klyce SD. Advances in the analysis of corneal topography. Surv Ophthalmol 1991;35:269–77.
- Maguire LJ, Singer DE, Klyce SD. Graphic presentation of computer – analyzed keratoscope photographs. Arch Ophthalmol 1987;105:223–30.
- Wilson SE, Klyce SD, Husseini ZM. Standardized color-coded maps for corneal topography. Ophthalmology 1993;100:1723–7.
- Smolek MK, Klyce SD, Hovis JK. The universal standard scale: Proposed improvements to the ANSI standard corneal topography map. Ophthalmology 2002;109:361–9.
- Oshika T, Klyce SD. Corneal topography: basic concepts. In FS Brightbill (Ed): Corneal Surgery: Theory, Technique, & Tissue. Part II. Testing and Measuring Corneal Function prior to Surgery (3rd edn). Mosby, Inc., St. Louis, 1999, 91–97, Color Plates 4–13.
- Wilson SE, Lin DT, Klyce SD. The corneal topography of keratoconus. Cornea 1991;10:2–8.
- Rabinowitz YS, Rasheed K. KISA% index: a quantitative videokeratography algorithm embodying minimal topographic criteria for diagnosing keratoconus. J Cataract Refract Surg 1999;25:1327–35.
- Holland DR, Maeda N, Hannush SB, Riveroll LH, Green MT, Klyce SD, Wilson SE. Unilateral keratoconus. Incidence and quantitative topographic analysis. Ophthalmology 1997;104:1409–13.
- Randleman JB. Ectatic disorders associated with a claw-shaped pattern on corneal topography. Am J Ophthalmol 2007;144:977–8.
- Levy D, Hutchings H, Rouland JF, Guell J, Burillon C, Arné JL, Colin J, Laroche L, Montard M, Delbosc B, Aptel I, Ginisty H, Grandjean H, Malecaze F. Videokeratographic anomalies in familial keratoconus. Ophthalmology 2004;111:867–74.
- Abad JC, Rubinfeld RS, Del Valle M, Belin MW, Kurstin JM. Vertical D: A novel topographic pattern in some keratoconus suspects. Ophthalmology 2007;114:1020–6.
- Maeda N, Klyce SD, Smolek MK, Thompson HW. Automated keratoconus screening with corneal topography analysis. Invest Ophthalmol Vis Sci 1994;35:2749–57.
- Maeda N, Klyce SD, Smolek MK. Application of neural networks to the classification of corneal topography: preliminary demonstration. Invest Ophthalmol Vis Sci 1995;36:1327–35.
- Klyce SD, Karon MD, Smolek MK. Screening patients with the corneal navigator. J Refractive Surg 2005;21:S617–22.
- Randleman JB, Trattler WB, Stulting RD. Validation of the ectasia risk score system for preoperative laser in situ keratomileusis screening. Am J Ophthalmol 2008;145:813–8.
- Klein SR, Epstein RJ, Randleman JB, Stulting RD. Corneal ectasia after laser in situ keratomileusis in patients without apparent preoperative risk factors. Cornea 2006;25:388–403.
- Cho P, Cheung SW. Repeatability of corneal thickness measurements made by a scanning slit topography system. Ophthalmic Physiol Opt 2002;22:505–10.
- Maus M, Krober S, Swartz T, Belin MW, Michaelson M, Sutphin J, Wang M. Pentacam. In Wang M (Ed): Corneal Topography in the Wavefront Era, Chapter 24, Slack, Inc, Thorofare, NJ, 2006;281–93.
- Maeda N, Fujikado T, Kuroda T, Mihashi T, Hirohara Y, Nishida K, Watanabe H, Tano Y. Wavefront aberrations measured with Hartmann-Shack sensor in patients with keratoconus. Ophthalmology 2002;109:1996–2003.
- Pepose JS, Applegate RA. Making sense out of wavefront sensing. Am J Ophthalmol. 2005;139:335–43.
- Luce DA. Determining in vivo biomechanical properties of the cornea with an ocular response analyzer. J Cataract Refract Surg 2005;31:156–62.
- Kirwan C, O’Keefe M. Measurement of Intraocular Pressure in LASIK and LASEK patients using the Reichert Ocular Response Analyzer and Goldmann applanation tonometry. J Refract Surg 2008; 24: 366–70.
- Ortiz D, Piñero D, Shabayek MH, Arnalich-Montiel F, Alió JL. Corneal biomechanical properties in normal, post-laser in situ keratomileusis, and keratoconic eyes. J Cataract Refract Surg 2007;33:1371–5.