The keratometer and Placido disk are reflection-based methods. An object is projected onto a central corneal zone of known diameter and distance from the light source. The cornea acts like a convex mirror, reflecting light, and a virtual upright image is produced. The ratio of the image and object diameters is used to estimate the radius of curvature along a particular meridian (Fig. 1).

With the keratometer, only four data points are generated. Two lie on the steepest axis, and the other two are on an axis 90 degrees away. The keratometer (Von Helmholtz, Javal-Schiotz) reflects the four points over a central zone of 2.88 to 4 mm (depending on the diopteric power of the cornea). The instrument is highly accurate for regular spherocylindrical surfaces such as the central zone of a normal cornea, butit has limited value in irregular corneas. Furthermore, no information is available for zones inside, outside, or between the four reference points. In addition, the keratometer has less accuracy for corneal powers below 36 diopters (D) and above 50 D, although it has a theoretical range of 30 to 60 D.

The Placido disk reflects a series of concentric rings, or mires, off the cornea. Each mire produces an upright virtual image in the anterior chamber of the subject's eye. The distance between each mire to the corneal apex is compared to the image's distance to the corneal apex; the ratio of image distance to mire distance determines the anterior corneal curvature along any particular meridian. Placido devices include photokeratoscopes (Corneoscope) and computerized videokeratoscopes (TMS, EyeSys). Placido disk systems have certain advantages over the keratometer. The photokeratoscope can measure 12 mires over 70% of the corneal surface and over an infinite diopteric range. The computerized videokeratoscope assumes that the corneal surface is spherical and of uniform refractive index; it can measure 15 to 38 mires over 95% of the corneal surface and has a theoretical range of 8 to 110 D¹ (Fig. 2).

Videokeratoscopes display topographical information using different maps. A picture of the cornea is provided as a raw image. This allows the clinician to recognize distortions of the tear film and loss of fixation that may compromise the accuracy of the videokeratoscopy. The data is then presented as a color map with either an absolute or a normalized scale. On color maps, the warmer colors (red, orange, and yellow) represent the steeper areas; the cooler colors (blue and green) represent the flatter regions. The absolute scale is a fixed system in which each color is consistently assigned to a particular curvature or power. This facilitates the comparison of topographies between patients and over time in the same patient. Each videokeratoscope has its own absolute color map with fixed diopteric intervals covering the range of corneal powers. The normalized (relative) scale maximizes the utilization of all possible colors to evaluate a specific cornea. The color map with a normalized scale has narrower intervals between the assigned colors. Although the normalized scale allows the clinician to evaluate a particular topography with greater detail, it makes comparison between examinations difficult because the scale may have changed in range and diopteric interval. Another option available to the clinician is the adjustable scale, which allows the clinician to customize the step interval and diopteric range of the contours.

Placido devices use axial (sagittal, global, average, or standard), tangential (instantaneous, true, or local), or refractive power maps to calculate radius of curvature (Fig. 3). The axial map measures the perpendicular distance from the tangent at a point to the optical axis and therefore has a spherical bias. This gives a global description of shape and may be less accurate in the corneal periphery and with irregular surfaces. On the other hand, a tangential map displays the radius of curvature of a point with respect to its neighboring points along a specific meridian. It has a less spherical bias than an axial map and therefore may be preferred to assess the corneal periphery or in irregular corneas. The refractive map uses Snell's law to measure the focal power of the cornea.

The computer program gives the clinician three options for delineating the major and minor axes for a given cornea. The orthogonal axes mark the major and minor axes as measured at the central 3-mm zone. The zonal axes mark the major and minor axes at the 3-mm as well as the 5-mm and 7-mm zones. This allows the clinician to quickly assess if there is a shift in the major and minor axes with progression toward-s the corneal periphery. The instantaneous axes option draws a continuous line from the center of the cornea to the periphery, showing the true major and minor axes at any given diameter (Fig. 4).

Data can also be manipulated by videokeratoscope computer programs to glean desired information. Cross-sectional maps display the elevation and depression of the anterior corneal surface along any given meridian. Difference maps can compare pre- and post-treatment corneal topographies by subtracting the first map from the second. Statistical indices, which are often used in clinical studies, can also be obtained using computer programs.

Scanning slit topography (e.g., Orbscan) is a projection-based method that uses a series of slit-beam images to generate data regarding anterior surface curvature, posterior surface curvature, and pachymetry. Therefore, the cornea is represented as a three-dimensional (3D) structure. A slit lamp projects a beam at 45 degrees onto the cornea. Twenty slits are projected sequentially on the eye from the left side, and 20 slits from the right, for a total of 40 slits that produce 240 data points per slit. This produces an image that is a height map of the cornea relative to a best-fit sphere. The elevations and depressions are recorded in 5 μm increments. Height maps have a colored reference scale and display elevations in warm colors and depressions in cool colors. The elevation of the anterior and posterior surfaces is determined and the difference between these equals corneal thickness. The instrument averages pachymetry in nine circles of 2-mm diameter located in the central cornea. It also averages pachymetry of eight circles located 3 mm from the visual axis in the mid-periphery² (Fig. 5).

The Orbscan is also capable of calculating corneal power. Mean power maps show the variation of spherical power by calculating the arithmetic average of the principal curvatures for each data point. Astigmatic power maps represent the local astigmatism over the corneal surface, calculating the arithmetic difference of the principal curvatures for each data point. In addition, axial, tangential, and Snell power maps are available, based on the same principles as the Placido disk systems.

Rasterphotogrammetry (e.g., PAR Corneal Topography System) is another projection-based system that produces an elevation map of the cornea. Its advantages are that it measures the entire corneal surface, up to 12 mm, and can also measure irregular or de-epithelialized corneas. A projector places a grid onto a corneal surface coated with fluorescein. Fluorescein produces light emissions from the surface of the cornea. Two or more camera systems at a different angle to the cornea from the projector, image the grid. The rays intersect in 3D space, creating the elevation map.³

The latest technology is that of wavefront sensors, which evaluate the sum refractive effects of the cornea, lens, media, and retina. Wavefront sensor technology dates from the 1970s, when Josef Bille, at the University of Heidelberg, applied astronomical principles to the study of the eye. Lightwaves from distant stars traveling through the atmosphere and entering a telescopic lens should be virtually flat. However, interference from the atmosphere produces aberrations in the wavefront. Bille recomposed the imperfect wavefronts into a perfectly round image through the use of different mirrors. Today, wavefront technology is designed to evaluate refractive errors and correct optical aberrations, and it has been applied to laser refractive surgery. Wavefront systems use either the Tscherning aberroscope or the Hartmann-Shack sensor. A laser beam is projected through the cornea, lens, and media onto the fovea; the reflected beam of light produces wavefronts that are captured by a grid of lenses and sensors that lie in a conjugate plane to the cornea. The wavefront's angle of incidence to the lens is recorded by the sensor. The shape of the wave can be reconstructed from the direction that the wave is traveling at each of the grid lenses (Fig. 6). Regions of the wave that are out of phase with a reference point such as the front of the wave can be detected. A computer program develops a color map representing total ametropia at different points. Wavefront maps can determine if the wave emitted from one part of the optical system is out of phase with a wave emitted from another region. The discrepancy in phases can be measured in microns and corresponds to the depth of ablation necessary to produce a flat wavefront. A lens can be formulated to compensate for the aberration, creating a customized ablation pattern. This technology is promising and is currently undergoing clinical trials for use with laser-assisted in-situ keratomileusis (LASIK).^4–9

The normal population consists of a variety of corneal curvatures that are compatible with good vision and within the normal spectrum. Five videokeratoscopic patterns lie on the normal spectrum of corneal curvature: round (23%), oval (21%), symmetric bow tie (18%), asymmetric bow tie (in which the ends of the bow tie are of significantly different sizes) (32%), and irregular(in which the ends are at an angle to one another) (7%) (Fig. 7).¹⁰ In addition, there is diurnal variation due to stromal edema during sleep, menstrual variation, and lifetime variation. Infants usually have spherical corneas. During childhood and adolescence, regular with-the-rule astigmatism develops in about 90% of individuals. This reverses with subsequent aging processes. Diurnal and menstrual variations are generally subclinical.¹¹

Corneal elevation patterns on the Orbscan are measured as compared to a best-fit sphere. Liu et co-workers classified anterior and posterior elevation patterns seen amongst normals into five categories. Anterior corneal elevation patterns occur with the following frequency: the island (72%), incomplete ridge (20%), regular ridge (4%), irregular ridge (2%), and unclassified (2%). Posterior elevation patterns occur with a different order of frequency: the island (33%), regular ridge (30%), incomplete ridge (24%), irregular ridge (13%), and unclassified (0%).¹²

The anterior corneal surface has a power of 49.50 D and the posterior corneal surface has a power of –6.00 D, rendering a total corneal power of 43.50 D. Mean keratometric astigmatism is 0.90 D. Central corneal thickness is 560 nm and peripheral corneal thickness is 1200 nm. Pachymetry patterns were classified by Liu and colleaguesl into four categories: round (41%), oval (48%), decentered round (2%), and decentered oval (9%).¹²

SURFACE MASSES AND PTERYGIA

Irregularities of the corneal surface can result from compression by external agents, abnormalities of the corneal epithelium, or stromal alterations.

Compression of the cornea can occur from eyelid lesions such as chalazia, lid hemangiomas, and dermoids. Compressive lesions have direct as well as indirect effects on corneal topography. There is a flattening of the corneal curvature directly underneath the eyelid lesion and there is a secondary indirect steepening of the corneal curvature adjacent to the area of flattening. Scleral masses will cause scleral flattening immediately beneath the mass and adjacent corneal steepening.

Tear film disturbances and epithelial diseases that affect the tear film canl result in topographic irregularities. The keratometer and Placido disk systems are reflective instruments that rely on an intact air–tear film interface. In such conditions, utilizing a projection-based system to evaluate corneal topography may prove more accurate.

Focal patches of dryness such as in keratoconjunctivitis sicca appear as local areas of flattening. More important, however, corneal topography can often detect subtle changes in epithelial regularity, significant enough to cause patient complaints but not visible by slit-lamp biomicroscopy. This has been seen in recurrent erosion syndrome and in anterior basement membrane dystrophy (Fig. 8). A local depression such as an ulcer or dellen may produce focal flattening. Local elevations, such as Salzmann's nodules or Thygeson's keratitis, produce a focal steepening.¹

Pterygia affect the corneal epithelium as well as the underlying stroma. Visual disturbances can result from tear-film irregularity or from growth onto the cornea, resulting in astigmatism. Both regular and irregular astigmatism can be induced. There is progressive flattening as the corneal periphery is approached. As the size of pterygium increases, the amount of astigmatism and associated topographic abnormalities increase.^13–15 O'Brart and associates determined that the mean keratometric astigmatism induced is about 4 D with-the-rule but the mean refractive astigmatism is about 2 D.¹⁶ This does not affect refraction significantly until the pterygium head encroaches on the visual axis (Fig. 9). There are two hypotheses regarding the mechanism of induced astigmatism. One theory proposed by Hochbaum and associates suggests that horizontal steepening is induced in the semi-meridian aligned with the pterygium due to traction from subepithelial fibrosis.¹⁷ The second theory proposes that the mechanism is due to the tear film; the tear meniscus in front of the head of the pterygium pools in the angle between the pterygium and the paracentral cornea, leading to the observation of topographic flattening. Pterygium removal results in decreased astigmatism because of steepening of the flat meridian and flattening of the steep perpendicular meridian. In eyes with large pterygia that encroach within 2 mm of the line of sight, astigmatism may not resolve completely. Interestingly, studies have shown that postoperatively, there is an overall increase in mean central corneal power and a reduction of corneal asphericity, suggesting that the ratio of the astigmatic change in the treated meridian to that 90 degrees away was not one-to-one. The amount of surgically induced change in astigmatism increases with the size of the pterygium.^18,19 Refractive changes stabilize 1 month postoperatively; therefore, Tomidokoro et colleagues recommend that cataract or other refractive surgery be postponed until this time.^13,14

STROMAL ECTASIAS

Crneal ectasias are non-inflammatory diseases that are characterized by thinning and protrusion of the corneal stroma resulting in shape changes. This family of diseases consists of keratoconus, keratoglobus, and pellucid marginal degeneration. Normally, the stromal collagen lamellae run circumferentially in the corneal periphery, producing a round shape. However, progressive thinning of the stroma leads to flattening of the corneal curvature along that meridian. This induces the peripheral ring of collagen lamellae to assume a more oval shape and transmits a compressive force to the lamellae that are 90 degrees away, resulting in corneal steepening in that meridian. This mechanism is known as biomechanical coupling. Furthermore, intraocular pressure at the site of weakness causes protrusion of the cornea.

Keratoconus

Keratoconus is a bilateral but typically asymmetric ectasia that has onset in the late teens and usually progresses slowly over many years. Patients have a history of progressive myopia, oblique astigmatism, and reduction of spectacle-corrected visual acuity. Prior to the introduction of methods to assess corneal topography, the diagnosis was based on history and the presence of clinical signs. In mild disease, however, these clinical signs are subtle or altogether absent. The advent of refractive surgery has made the detection of subclinical keratoconus of increasing importance in order to prevent the surgical treatment of these eyes. Keratoconus has an incidence of 1 in 2000 inf the general population but is being detected in 5% of myopes who present for refractive surgery evaluation.²⁰

Keratoconus can be categorized according to the severity of power (mild, moderate, severe); location of cone (superior, central, inferior); and shape of cone (oval, globus, nipple) (Fig. 10). Corneal thinning most commonly occurs in the inferocentral cornea, and protrusion also occurs in this region. The point of maximal protrusion is referred to as the apex of the cone. The steepest corneal slope lies just peripheral to the apex (usually inferior in central cones). The region of smallest radius of curvature (therefore the greatest corneal power) lies between the cone's apex and its steepest slope. The mechanism of biomechanical coupling causes the flattest meridian to be approximately horizontal and the steepest meridian to lie close to the vertical meridian. Placido disk-based videokeratographs mirror this distortion by producing mires that are typically oval. The distance between rings is smallest at the steepest corneal slope and farthest apart superiorly where the cornea is flattest. Tangential curvature maps of projection-based systems provide additional information. On these maps the steepest slope is easily located as being inferior to the apex, producing an asymmetric bow tie. This corresponds to the exaggerated prolate shape of the keratoconic eye. Projection-based systems can be used to locate the apex of the cone on elevation maps as the highest point. The apex is surrounded by concentric zones of decreasing elevation. A comparison of anterior and posterior elevation maps reveals that there is a greater change in height from periphery to central cornea posteriorly than anteriorly. On tangential curvature maps, the apex of a cone has a slope of zero. Protrusions such as proud nebulae also have a slope of zero.

Several authors have recommended topographical indices for detecting early keratoconus and suspected keratoconus. The diagnosis, however, still requires the presence of Vogt's striae, a Fleischer ring, or corneal thinning. The surface asymmetry index (SAI) measures the irregularity of the cornea in the central 4.5-mm zone. It measures the difference in corneal power between points on the same ring 180 degrees apart. The inferior-superior (I-S) value measures the average power at five superior points 3 mm from the center at 30-degree intervals and compares this to five inferior points 3 mm from the center at 30-degree intervals. K is the central K-reading; when used alone, a value greater than 47.2 is suggestive of keratoconus. The KCI%, KPI% and KISA% are values derived by a combination of other indices. For example, the keratoconus predictability index (KPI) combines the SAI with seven other indices in an algorithm to predict the presence of keratoconus with 68% sensitivity and 99% specificity. Auffarth and co-workers. used the Orbscan to evaluate a series of keratoconus patients and noticed that the apex and thinnest point were located separately but with no consistent distance or pattern.²⁰ The thinnest point was less than 0.500 mm. Furthermore, there is a high degree of nonsuperimposable mirror-image symmetry in the location of the cones between the right and left eyes of the same patient. The nonsuperimposability is due to the variation in radii between the apex and the thinnest point.^20–23

Pellucid Marginal Degeneration

Pellucid marginal degeneration (PMD) is another bilateral corneal-thinning disorder. There may be asymmetry in the severity of disease between the two eyes. Furthermore, there are reports of PMD in one eye and keratoconus in the fellow eye. Typically in PMD, a 2-mm wide band of thinned stroma occurs 1 to 2 mm from the inferior limbus and spans the central four clock hours. Unlike in keratoconus, the areas of thinning and protrusion are not the same; corneal protrusion occurs in the cornea that lies above the area of thinning. Cases of superior PMD, nasal PMD, and circumferential extension of inferior PMD have been recognized. Classically, in inferior PMD topography maps the lowest corneal power to a narrow corridor of central cornea that is close to the vertical meridian, producing an against-the-rule astigmatism. The power increases markedly toward the inferior periphery within this narrow corridor of central cornea. The mires show elongation along the vertical axis with compression inferiorly. The area of highest power extends along the inferior cornea and then turns toward the central cornea along the inferior oblique meridians. The term applied to this ring of high cylinder power is the loop cylinder. Progression to the superior corneal periphery does not show an increase in power (Fig. 11). Conversely, in superior PMD the area of highest corneal power is in the superior periphery with extension toward the central cornea from the superior nasal and superior temporal oblique semimeridians producing a superior loop cylinder. If the thinning progresses toward the horizontal, there is a shift in the meridians of highest and lowest corneal powers. Extension of thinning nasally will cause the lowest power to shift from the vertical toward the temporal and the highest power to shift toward a more nasal meridian.^24,25

Fig. 11. Pellucid marginal degeneration (PMD). A. Slit topography showing inferior corneal thinning (arrow) 1 to 2 mm from the limbus, extending from the 5- to 8-o'clock positions in both eyes. B. Videokeratoscopic image shows a typical pear-shaped image with compression of the inferior rings. C. Corneal topographic maps (absolute scale) showing against-the-rule astigmatism of 10.6 diopters (D) in the right eye. The left eye shows enantiomorphic symmetry (mirror image) to the right eye with 11.9 D of against-the-rule astigmatism. In early PMD the power of the cornea is least at a vertical axis very close to 90 degrees. The area of greater power is presented in a bow-tie configuration of two semimeridians inferior and oblique to the horizontal axis. (Karabatsas CH, Cook SD: Topographic analysis in pellucid marginal degeneration and keratoglobus. Eye 10:451–455, 1996.)

Keratoglobus

Keratoglobus is characterized by a diffuse globoid protrusion of cornea. The stroma is thinned diffusely, including the limbus. The condition is very rare and there were few reports in the literature. Keratoglobus may present bilaterally but has also been reported with the presence of other ectasias in the fellow eye. Karabatasas documented the topographic picture of keratoglobus in an individual who had classic PMD in the fellow eye.²⁵ The keratoglobic eye had resolving hydrops in the inferotemporal quadrant and a band of circumferential peripheral thinning similar to that seen in PMD, suggesting that the condition may have arisen out of advanced PMD. The topography demonstrated a very asymmetric bow-tie pattern with a shift of 35 degrees from the vertical in the axis of lowest corneal power (Fig. 12).

PERIPHERAL THINNING DISORDERS

Terrien's Marginal Degeneration

Terrien's marginal degeneration is a slowly progressive, painless thinning of the corneal stroma of unknown etiology. It is typically bilateral and is predominantly seen in men. Initially there is deposition of refractile, yellow-white lipid deposits in the superior perilimbal anterior stroma, with radial superficial vessels. Subsequently a gutter 1- to 2-mm wide with intact epithelium develops. This gutter runs parallel to, but does not involve, the limbus. Over time there is deeper involvement of the stroma as well as progression circumferentially. Complications include astigmatism and perforation.²⁶

Mooren's Ulcer

Mooren's ulcer is a rare form of peripheral ulcerative keratitis. In 30% of cases it is bilateral. Although the pathogenesis is unknown, there is a strong evidence that it is an autoimmune genetic disease, which is supported by the findings of HLA class II DR17 and DQ2 in 83% of cases; an inflammatory immune response; antibodies to corneal antigens; and resolution with immunosuppressive therapy. It is seen more commonly in Africa and India and is characterized by a previous history of trauma, surgery or infection. The patient presents with an acute, painful ulceration of the peripheral cornea that progresses in a circumferential or transverse fashion. The borders of the ulcer are overhanging; vessels extended from the limbus into the ulcer bed without adjacent scleral melt.About 20% of the cases may progress to perforation. Treatment includes local immunosuppression, systemic immunosuppression, and/or removal of local stimulatory antigens such as in lamellar conjunctivosclerokeratectomy.^27,28

In these peripheral thinning disorders, flattening of the cornea occurs in the affected meridian and steepening in the perpendicular meridian due to biomechanical coupling. Characteristically there is high against-the-rule astigmatism in a bow-tie pattern if the process is confined to the superior and/or inferior periphery. As the thinning progresses circumferentially to encompass the horizontal meridians, the flat area extends and the perpendicular steep meridians merge closer together, resulting in an arching bow-tie pattern. In some cases, the central cornea remains relatively spherical, such as with 360-degree peripheral thinning or, conversely, with a very limited area of thinning (Fig. 13). These topography changes may not be specific but may be helpful in the diagnosis and in the differentiation from other diseases.^26–28

TYPES OF CORNEAL SURGERY

Post-penetrating Keratoplasty

The use of topography in the management of the cornea-transplanted patient has facilitated visual rehabilitation. Topography can aid in:

1. Making decisions about trephination and graft size
   
2. Identifying thin areas to be avoided in the graft-host junction
   
3. Choosing a suturing technique
   
4. Managing selective suture removal or adjustment
   
5. Deciding on the need for a relaxing incision or a wedge resection in astigmatism greater than 8 D
   
6. Correcting refractive errors by a excimer laser procedureedit
   
7. Postsurgical fitting of a contact lens²⁹
   

Postoperative astigmatism is the major limiting factor for good vision in patients with clear grafts. It has been reported that after penetrating keratoplasty, 10% of the eyes have at least 5.00 D of keratometric astigmatism. Studies have shown that the refractive power map and axial power map correlate better with the manifest refractive cylinder than tangential maps or keratometry. However, the axis of the astigmatism is more accurate when measured by keratometry.³⁰ Almost any topographical pattern may be seen post-keratoplasty. The asymmetric bow tie was most commonly seen by many investigators, but others have found the irregular pattern to be the most frequent.^30–32 Several factors can affect corneal topography following surgery, including the following:

1. Graft size (most surgeons oversize the donor button by 0.25 mm because a difference of 0.5 mm may produce steepening, whereas a difference of 0 mm may flatten the cornea)
   
2. Depth of incision (lamellar vs. penetrating keratoplasty)
   
3. Centration of trephination
   
4. Preexisting astigmatism in the host and donor corneas
   
5. Suture technique (interrupted vs. continuous, symmetry, and radiality; long and deep bites may cause more compression and steepening, but loose and superficial sutures produce wound gape and flattening
   

Corneal steepening is usually the result of a tight suture. Tight sutures result in local flattening of the host-graft interface but a secondary steepening of the central cornea. As a result of biomechanical coupling, there is also flattening of the perpendicular meridian. Another cause may be a vertical wound misalignment in which the central edge underrides the peripheral edge. Tissue contraction due to cauterization or edema of the wound edges may also lead to steepening. Corneal flattening is often due to a loose suture. It may also result from vertical wound misalignment in which the central edge overrides the peripheral edge, too superficial sutures that produces a posterior wound gape, or delay in wound healing. Corneal irregular astigmatism may result when two nonperpendicular and nonadjacent sutures are tight, resulting in steepness in two meridians. Non-radial suture bites may also result in torsion and irregular astigmatism.

Different suturing techniques have been used with the objective of decreasing postoperative astigmatism. A single continuous running suture has the advantage over interrupted sutures in that it may be adjusted intraoperatively or postoperatively to redistribute corneal tension. The suture is rotated from flat meridians toward steep meridians. Intraoperative adjustment has been shown to be superior to postsurgical adjustment. Interrupted sutures alone, or in combination with a continuous suture, facilitate postoperative selective suture removal.

Suture removal can decrease the refractive cylinder and improve visual acuity. There is no universal protocol that establishes how many sutures should be removed at one time, the interval between suture removals or how many diopters of astigmatism would be improved. However, most clinicians agree that if refractive astigmatism is less than 3 D and visual acuity is acceptable, no sutures are removed. If this is not the case then interrupted sutures can be removed as early as 3 weeks postoperatively as long as there is a continuous suture in place. Interrupted sutures are removed no earlier than 12 weeks if there is no continuous suture in place. Usually one interrupted suture is removed in the steep meridian (Fig. 14). The patient is then re-evaluated with topography and refraction at 3 week intervals.

REFRACTIVE CORNEAL SURGERY

Topography is relied on in refractive corneal surgery for screening diseases (contact lens–induced corneal warpage, keratoconus), planning surgery (incision location, length, and depth) and in wavefront-guided LASIK. Postoperatively, corneal topography changes can be quantified by difference maps, tangential (local) maps, and indices such as the surface regularity index (SRI) and theSAI. These tools allow the recognition of decentered treatment zones, flap irregularities, and stromal ectasias. Topography can also assist in calculation of intraocular lens power in postrefractive surgery patients who are undergoing cataract surgery. Projection-based systems are preferred because they do not require the anterior corneal surface to be reflective and therefore can be used immediately postoperatively.

Cornea shape alteration can affect just the anterior surface or all the layers of the cornea. In the former case, the superficial tissue is removed (excimer laser, keratectomy), added (epikeratoplasty), or structurally altered (laser thermokeratoplasty) without alterations in the stroma or posterior cornea.³⁴ In the latter case, the corneal stroma is changed either by incisions (radial keratotomy, astigmatic keratotomy) or by applying persistent mechanical forces (intrastromal corneal rings).

Radial Keratotomy

Radial keratotomy (RK) induces central flattening by deep radial incisions. The central 3-mm optical zone remains untouched. The corneal profiles seen after RK (Table 1) are oblate (79%) and a mixed pattern of oblate/prolate (18%). The radial incisions produce a rapid change in slope called the paracentral knee (inflection zone) that lies between the zone of peripheral steepening and central cornea flattening. Topographic patterns seen after RK are the same spectrum as for normal corneas with the addition of the polygonal pattern. The polygonal pattern is seen in 59% of cases and is a concentric pattern with two or more angles (135 degrees) and three or more nearly straight lines that correspond to the central ends of the radial incisions. In all RK cases the corneal asphericity is increased. Initially, maximal flattening is present at the proximal end of the incisions in the paracentral zone, whereas the central cornea remains steeper. With time, the central cornea becomes flatter. The cornea can mimic a multifocal lens as a result of the increased range of diopters within the pupillary aperture³⁵ (Fig. 15).

Table 1. Topographic Patterns Seen after RK Radial Keratotomy (RK) (From Bogan et al.)

Astigmatic Keratotomy

Astigmatic keratotomy corrects astigmatism by flattening the steep meridian (with paired relaxing corneal incisions) or steepening the flat meridian (with wedge resection and/or compressive sutures). The most common topographic patterns seen after astigmatic keratotomy are symmetric bow tie and asymmetric bow tie.¹

Intrastromal Corneal Rings

Intrastromal corneal rings induce a steep peripheral ring and central corneal flattening. The natural aspheric prolate shape is preserved¹ (Fig. 16).

Photorefractive Keratectomy

Photorefractive keratectomy (PRK) and LASIK correct refractive error by reshaping the superficial corneal tissues. In LASIK, a corneal flap of 120 to 190 μm is created using a microkeratome; the stromal bed is then ablated using an excimer laser. At least 30% or 50 μm of the stromal bed is left untouched in order to prevent iatrogenic ectasia. Correction of myopia involves ablation of the central cornea; topography after a successful procedure for myopia shows a well-centered central corneal region of uniformly reduced corneal power resembling an oblate configuration. There is a smooth power change gradient with progressively decreasing power change proceeding from the center to the periphery of the treatment zone. Topography after successful treatment of hyperopia shows a larger-diameter treatment zone in which the peripheral cornea has been concentrically ablated, producing an exaggerated prolate pattern.

Immediately after PRK, the topographic map shows the delineation of the treatment zone. The most effective way to see the change produced by PRK is by the use of a difference map. The difference between the preoperative and postoperative map shows the profile of the ablation and the uniformity of the laser beam. Preoperative corneas with a round or oval (no astigmatism) pattern maintain this pattern after hyperopic or myopic correction. Hyperopic astigmatics who have only sphere correction do not have a change in their bow-tie pattern. However, in myopic astigmatics who have only sphere correction, the red (steeper) bow-tie pattern becomes a blue (flatter) bow-tie pattern in the perpendicular meridian. In astigmatic correction, the preoperative red bow tie (where the oval ablation zone was performed) becomes a blue bow tie in the same meridian.

In general, the topography of post-LASIK eyes appears to change very little after the first postoperative visit, unlike post-PRK eyes (Fig. 17). Following the boom in refractive surgery procedures, a variety of complications have been recognized. A visual acuity of 20/20 does not necessarily correlate with a successful procedure. Subtle irregularities in the cornea can result in complaints of glare, halos, decreased contrast sensitivity, and polyopia. Topography assists in the recognition of these irregularities. Several investigators have proposed various classification schemes for the topography patterns seen following LASIK with various excimer lasers.^36–38 The topographical patterns seen after PRK are shown in Table 2.^39–42

Table 2. Topographic Patterns After Photorefractive Keratotomy.

The two most common complications are central islands and decentration. Regardless of classification schemes, most authors define a central island as a central area of relatively less flattening. It measures at least 1.0 mm in diameter and has at least 1.00 D greater power than the surrounding cornea. It is surrounded 360 degrees by an area of greater flattening and therefore does not extend into the periphery (Fig. 18). Central islands may occur as a result of reduced central ablation or irregular healing. Reduced ablation may occur from uneven hydration during ablation, with accumulation of fluid centrally, masking the stroma. Delayed clearance of ablation debris may also mask the central stroma. Cooler laser beams centrally may also cause a central island. Central islands are associated with undercorrection, loss of best corrected visual acuity, glare, monocular diplopia and halos. The highest incidence is at 1 week postoperatively; at 1 year postoperatively, the incidence is less than 2%. The resolution of central islands is due to epithelial-subepithelial hyperplasia and corneal haze.^43–48

Decentration occurs most commonly as a result of patient movement secondary to loss of fixation. Decentration is measured in relation to the pupil center, rather than to the center of the cornea. Higher-order refractive corrections have a greater incidence of decentration due, presumably, to the longer treatment time. Decentration is considered significant if it is greater than 0.5 mm, but its effect is modified by the size of the pupil and the diameter of the ablation zone. Decentration shows a similar pattern to that of asymmetric astigmatism (show topography). This phenomenon produces visual effects (loss of visual acuity, contrast sensitivity, glare, halos and polyopia) that are greater in patients with large pupils and smaller treated optical zones. Two types of eye-tracking systems have been developed to prevent decentration during laser procedures. Postsurgical centration can be demonstrated by using software that measures the distance from the center of the pupil to the center of the treatment zone and compares this with the corneal vertex, which is the center of the cornea topography.

Other complications that may be seen post-LASIK are debris or epithelial cells under the flap, ectasia, striae, and flap defects resulting in irregular astigmatism. Ectasia occurs as a result of changes in the posterior corneal curvature (PCC). Iatrogenic keratoectasia has been reported when the stromal bed is less than 250 μm after LASIK. The average amount of ectasia induced under these conditions is 13 μm. It is important to keep in mind that the error of the ORBSCAN system in pachymetry measurement is ± 20 μm.³⁴

CONTACT LENSES

Contact lenses have been used to compensate for distortions of the anterior corneal surface, leading to improved visual acuity. Computerized videokeratoscopy systems contain software to assist the clinician in selecting a lens that is most likely to fit a patient. These systems can provide the clinician with initial lens parameters (posterior lens curvature, optic zone diameter, overall lens diameter, and edge lift) and a fluorescein-pattern simulation. The advantage is that the patient has to try on fewer trial lenses before an appropriate fit is found. In most cases, the clinician determines the final power of the lens after trial lens insertion and over-refraction. Some software programs will also provide the final power of the lens as well.

Corneal Warpage

Contact lenses have been known to induce topographic changes in the cornea. This phenomenon is known as corneal warpage and is generally reversible with discontinuation of contact lens use. While warpage has been documented with both rigid and soft lenses, polymethylmethacrylate (PMMA) and rigid gas-permeable (RGP) lenses have a greater mechanical effect on the cornea than soft lenses. Signs of corneal warpage are keratometry mire distortion, significant changes in keratometric measurements, and decreased spectacle visual acuity in the absence of clinically observable corneal edema. Generally, RGP lenses induce relative flattening of the cornea under the resting position of the lens and may induce steepening peripherally beyond the lens edge, producing an asymmetric oblate bow-tie pattern.^49,50 Corneal warpage may simulate keratoconus with loss of central radial symmetry and the presence of irregular astigmatism due to superior flattening and inferior steepening. Lebow and Grohe compared topography indices of keratoconics and corneal warpage subjects; they determined that corneal shape factor, corneal irregularity measure, and corneal toricity were greater in true keratoconics subjects.⁵¹ Refractive surgery candidates are required to discontinue use of contact lenses for a period of time before ocular measurements are made in order to resolve any changes induced by lenses. Corneal topography can be used to diagnose warpage and to follow the resolution (Fig. 19).

Orthokeratology

Orthokeratology intentionally manipulates the potential of RGP lenses to modify anterior topography as a nonsurgical and variably reversible means of improving vision in disorders such as myopia or keratoconus. The technique involves fitting progressively flatter reverse-geometry rigid contact lenses (lenses in which the secondary curvature steepens) until the anterior corneal curvature has been altered to the desired level of myopia reduction. Myopia reduction is modest, 1 to 2 D, and is associated with central epithelial thinning and midperipheral stromal thickening.^52,53

1. Corbett MC, Rosen ES, O'Brart DPS: Corneal Topography: Principles and Applications. London, BMJ Books, 1999.

2. Orbscan Operator's Manual for the Slit Scan Corneal Tomography/Pachymetry System: Software Version 2.10. ORBTEK, Inc., Utah.

3. Gills JP: Corneal Topography: The State of the Art. Slack, Thorofare, NJ, 1995.

4. Naseri A, McLeod SD, Lietman T: Evaluating the human optical system: Corneal topography and wavefront analysis. Ophthalmol Clin North Am 14:269–273, 2001.

5. Haman H: A quick method for analyzing Hartmann-Shack patterns: Application to refractive surgery. J Refract Surg 16: S636–S642, 2000.

6. Mrochen M, Kaemmerer M, Seiler T: Wavefront-guided laser in in-situ keratomileusis: Early results in three eyes. J Refract Surg 16:116–120, 2000.

7. Wilson S, Ambrosio R: Computerized corneal topography and its importance to wavefront technology. Cornea 20:441–454, 2001.

8. Pallikaris IG, Panagopoulou SI, Molebny VV: Clinical experience with the Tracey technology wavefront device. J Refract Surg 16:S588–S591, 2000.

9. Hong X, Thibos LN: Longitudinal evaluation of optical aberrations following laser in in-situ keratomileusis surgery. J Refract Surg 16:S647–S650, 2000.

10. Bogan SJ, Waring GO, Ibrahim O, et al: Classification of normal corneal topography based on computer-assisted videokeratography. Arch Ophthalmol 108:945–949, 1990.

11. Marin-Amat M: The physiological variations of the corneal curvature during life: Their significance in ocular refraction. Bull Soc Belge Ophtalmol 136:263, 1957.

12. Liu Z, Huang A, Pflugfelder SC: Evaluation of corneal thickness and topography in normal eyes using the Orbscan corneal topography system. Br J Ophthalmol 83:774–778, 1999.

13. Tomidokoro A, Oshika T, Amano S, et al: Quantitative analysis of regular and irregular astigmatism induced by pterygium. Cornea 18:412–415, 1999.

14. Tomidokoro A, Miyata K, Sakaguchi Y, et al: Effects of pterygium on corneal spherical power and astigmatism. Ophthalmology 107:1568–1571, 2000.

15. Avisar R, Loya N, Yassur Y, et al: Pterygium-induced corneal astigmatism. Isr Medl Assoc Jl: Imag 2:14–15, 2000.

16. O'Brart DPS, Corbett MC, Rosen ES: The topography of corneal disease. Eur J Implant Refract Surg 7:173–183, 1995.

17. Hochbaum DR, Moskowitz SE, Wirtschafter JD: A quantitative analysis of astigmatism induced by pterygium. J Biomech 10:735–746, 1977.

18. Cinal A, Yasar T, Demirok A, et al: The effect of pterygium surgery on corneal topography. Ophthalmic Surg Lasers. 32:35–40, 2001.

19. Budak K, Khater TT, Friedman NJ, et al: Corneal topographic changes induced by excicision of perilimbal lesions. Ophthalmic Surg Lasers 30:458–464, 1999.

20. Auffarth GU, Wang L, Völcker HE: Keratoconus evaluation using the Orbscan Topography System. J Cataract Refract Surg 26:222–228, 2000.

21. Chastang PJ, Borderie VM, Carvajal-Gonzalez S, et al: Automated keratoconus detection using the EyeSys videokeratoscope. J Cataract Refract Surg 26:675–683, 2000.

22. Rabinowitz YS, Rasheed K: KISA % index: A quantitative videokeratography algorithm embodying minimal topographic criteria for diagnosing keratoconus. J Cataract Refract Surg 25:1327–1335, 1999.

23. Maeda N, Klyce SD, Smokek MK: Comparison of methods for detecting keratoconus using videokeratology. Arch Ophthalmol 113:870–874, 1995.

24. Rao SK, Fogla R, Padmanabhan P, et al: Corneal topography in atypical pellucid marginal degeneration. Cornea 18:265–272, 1999.

25. Karabatasa CH, Cook SD: Topographic analysis in pellucid marginal degeneration and keratoglobus. Eye 10:451–455, 1996.

26. Wilson S, Lin D, Klyce S, et al: Terrien's marginal degeneration: Corneal topography. Refract Corneal Surg 6:15–20, 1990.

27. Zegans ME, Srinivasan M, McHugh T, et al: Mooren ulcer in South India: Serology and clinical risk factors. Am J Ophthalmol 128:205–210, 1999.

28. Zegans ME, Srinivasan M: Mooren's ulcer. Int Ophthalmol Clin 38:81–88, 1998.

29. Riddle HK Jr , Parker DA, Price FW Jr : Management of postkeratoplasty astigmatism. Curr Opin Ophthalmol 9I:15–28, 1998.

30. Borderie VM, Touzeau O, Laroche L: Videokeratography, keratometry, and refraction after penetrating keratoplasty. J Refract Surg 15:32–37, 1999.

31. Shimazaki J, Shimmura S, Tsubota K: Intraoperative versus postoperative suture adjustment after penetrating keratoplasty. Cornea 17:590–594, 1998.

32. Karabatzas CH, Cook SD, Sparrow JM: Proposed classification for topographic patterns seen after penetrating keratoplasty. Br J Ophthalmol 83:403–409, 1999.

33. Touzeau O, Borderie VM, Allouch C, et al. Effects of penetrating keratoplasty suture removal on corneal topography and refraction. Cornea 18:638–644, 1999.

34. Hernandez-Quintela E, Smapunphong S, Khan B: Posterior corneal surface changes after refractive surgery. Ophthalmology 108:1415–1422, 2000.

35. Waring GO 3rd, Lynn MJ, McDonnell PJ:. Results of the Prospective Evaluation of Radial Keratotomy (PERK) study ten years after surgery. Arch Ophthalmol 112:1298–1308, 1994.

36. Barker NH, Couper TA, Taylor HR: Changes in corneal topography after laser in situ keratomileusis for myopia. J Refract Surg 15:46–52, 1999.

37. Sano Y, Carr JD, Takei K, et al: Videokeratography after excimer laser in situ keratomileusis for myopia. Ophthalmology 107:674–684, 2000.

38. Hersh PS: A standardized classification of corneal topography after laser refractive surgery. J Refract Sur 13:571–578, 1997.

39. Lin DT, Sutton HF, Berman A: Corneal topography following excimer photorefractive keratectomy for myopia. J Cataract Refract Surg 19:149–154, 1993.

40. Lin DTC: Corneal topographic analysis after excimer laser photorefractive keratectomy. Ophthalmology 101:1432–1439, 1994.

41. Hersh PS, Schwartz-Goldstein BH: The Summit Photorefractive Keratectomy Topography Study Group: Corneal topography of phase III excimer laser photorefractive keratectomy: characterization and clinical effects. Ophthalmology 102:963–978, 1995.

42. Hersh PS, Shah SI, Holladay JT: Summit Photorefractive Keratectomy Topography Study Group: Corneal asphericity following excimer laser photorefractive keratectomy. Ophthalmic Surg Lasers 27:S421–S428, 1996.

43. Krueger RR, Saedy NF, McDonnell PJ: Clinical analysis of steep central islands after excimer laser photorefractive keratectomy. Arch Ophthalmol 114:377–381, 1996.

44. McGhee CN, Bryce IG: Natural history of central topographic islands following excimer laser photorefractive keratectomy. J Cataract Refract Surg 22:1151–1158, 1996.

45. Levin S, Carson CA, Garrett SK, et al: Prevalence of central islands after excimer laser refractive surgery. J Cataract Refract Surg 21:21–26, 1995.

46. Klyce SD: Using topography to manage LASIK complications. Rev Refract Surg 20–26. 2001.

47. Oshika T, Klyce SD: Corneal topography in LASIK. Semin Ophthalmol 13:64–70, 1998.

48. Abbas UL, Hersh PS, et al: Late natural history of corneal topography after excimer laser photorefractive keratectomy. Ophthalmology 108:953–959, 2001.

49. Novo AG, Pavlopoulos G, Feldman ST: Corneal topographic changes after refitting polymethylmethacrylate contact lens wearers into rigid gas permeable materials. CLAO J 21:47–51, 1995.

50. Calossi A, Verzella F, Zanella SG: Corneal warpage resolution after refitting an RGP contact lens wearer into hydrophilic high water content material. CLAO J 22:242–244, 1996.

51. Lebow KS, Grohe RM: Differentiating contact lens induced warpage from true keratoconus using corneal topography. CLAO J 25:114–122, 1999.

52. Swarbrick HS, Wong G, O'Leary DJ: Corneal response to orthokeratology. Optom Vis Sci 75:791–799, 1998.

53. Polse KA, Brand RJ, Vastine DW, et al: Corneal change accompanying orthokeratology: Plastic or elastic? Results of a randomized controlled clinical trial. Arch Ophthalmol 101:1873–1878, 1983.