Although our understanding of corneal power has come a long way since the first keratometers, many of the main principles remain the same. The original keratometer grossly estimated corneal curvature by aligning the images reflected by the tear film interface using multiple assumptions. With recent advancements in technology, we can more accurately characterize corneal curvature and calculate true corneal power. This becomes increasingly important when trying to detect subclinical corneal ectasia in refractive surgery candidates and for improving cataract postsurgical outcomes.
Corneal curvature and refractive power can be measured with both topography (a 3-dimensional representation of the anterior corneal surface) and tomography (a 3-dimensional reconstruction of the entire anterior segment using cross-sectional images). In our clinic, we rely on both topography and tomography. At a time when patient expectations are sky high, updates to the devices utilizing these techniques help us to better understand true corneal power.
Placido disc technology, designed to measure corneal reflection, has significantly improved over the past 200 years. Placido disc topography utilizes the first and second purkinje images that arise from the reflections of the anterior and posterior cornea, respectively. Concentric rings project from a light source onto the corneal surface. A computer detects the rings and quantifies them into keratometric values. The Atlas 9000 corneal topography system (Carl Zeiss, Meditec), uses 22 concentric rings, each of which contains 180 data points that reflect back to the computer.1 The central (0.0-3.0 mm), mid-peripheral (3.0-6.0 mm) and peripheral (6.0-9.0 mm) corneal zones are calculated with an angular resolution of 2 degrees. Other devices that use placido ring topography to measure the anterior corneal curvature include the TMS-5 (Tomey GmbH), OPD-Scan III (Nidek, Inc.) and Easygraph (Oculus, Inc.). This type of technology relies heavily on the quality of the tear film and only measures the anterior corneal surface.
Previously, many surgeons relied solely on anterior corneal measurements when evaluating patients for refractive surgery. Mathematical algorithms were employed to rule out corneal ectasias. Specifically, topographers reported the mean simulated keratometry (sim K), which calculated the apical keratometric power. The sim K was utilized numerically as a cut-off for early keratoconus.2 But relying on the sim K alone, with its low sensitivity and specificity, has obvious setbacks. Other estimations, like the maximum anterior sagittal curvature (K max), measures only a small, central steep area of the cornea and ignores the posterior corneal contribution to keratectasias.2 Other helpful calculations that compared the inferior to superior cornea included the I-S value and the skewed radial axis (SRAX). These calculations have helped diagnose keratoconus, but today, fall short of other algorithms that can measure the posterior cornea and potentially catch forme fruste.
Placido disc technology is still universally used today but recent developments in technology have allowed calculation of both anterior and posterior corneal astigmatism. Including the posterior cornea’s contribution to total corneal power is important for detecting early-onset irregular astigmatism, such as forme fruste keratoconus, and for measuring more accurate total astigmatism; the assumption being that distinction of posterior corneal power from just anterior power alone results in a poor astigmatism estimate, as posterior corneal astigmatism can differ from its anterior counterpart by > 0.5 diopters or >10 degrees in meridian.3 As technology develops, we continue to create new systems to measure total corneal power and toricity.
The Cassini color light-emitting diode (LED) tomographer (i-Optics Corp.) is a corneal tomographer that utilizes the same principle of reflection, but with a different spin. The Cassini uses point-source color LED technology to calculate total corneal astigmatism by point-to-point ray tracing.4 It uses roughly 700 red, yellow and green LED lights. The central 3-mm corneal zone can then be calculated from the first and second purkinje images, thereby establishing an estimate of posterior corneal curvature. The color-LED tomography system was found to be highly repeatable, though slightly less repeatable than placido disc topography.4
The Cassini can also calculate keratoconus indices. The keratoconus form factor index characterizes the gradual change in corneal curvature from the center to periphery.5 The surface irregularity index (SRI) compares the fluctuations in corneal surface in the central 3-mm corneal zone with normal corneal values <1.55.5 In addition to the SRI, the surface asymmetry index (SAI) calculates corneal symmetry by summing the differences in corneal power between points 180 degrees from each other.
More complex imaging systems can create cross sections of the cornea. Scheimpflug tomography has rotating cameras; these create a cross section portraying posterior corneal curvature and elevation data. The Pentacam HR rotating Scheimpflug camera (Oculus, Inc.) uses a single, 180-degree rotating Scheimpflug camera to calculate corneal measurements. The camera has a 470 nm slit LED light source that communicates with the opposing static camera to create 50 cross section images equaling 138,000 data points of the anterior segment.1 After cross sections are obtained, the Pentacam recreates a three-dimensional view of the anterior segment. The Galilei G4 Dual-Scheimpflug/Placido Analyzer (Ziemer Ophthalmic Systems AG) uses dual Scheimpflug cameras in addition to placido disc imaging. The Galilei produces simulated keratometric values in the 1.0 to 4.0 mm central corneal zone by using 20 rings in combination with two rotating Scheimpflug cameras situated 180 degrees apart. The cross section image, in addition to the placido rings, combines roughly 122,000 data points into keratometric values and anterior segment structures.1
The Scheimpflug devices allow for a more accurate measurement of the posterior cornea and can detect keratectasia at an earlier stage. Specifically, the Pentacam uses a Belin-Ambrosio enhanced ectasia display (BAD) that compares both anterior and posterior maps of curvature and pachymetry to screen for ectasia.2 Both Scheimpflug devices allow for a best-fit sphere (BFS) in the central 8.0 mm corneal zone to be compared to the rest of the cornea. They also use an enhanced reference surface to represent a normal peripheral cornea thereby allowing for central irregularities to manifest.2 The Galilei has a function that can compare elevation differences between the best-fit sphere in order to determine forme fruste keratoconus.2
Anterior segment optical coherence tomography (AS-OCT), though similar to Scheimpflug technology in its 3D imaging, differs in how it acquires anterior segment cross sections. The RTVue 100 Fourier domain OCT (Optovue, Inc.) uses 26,000 axial scans per second, which produce high resolution (5-10 um) images of the anterior segment. The machine aligns centrally with the apex of the cornea or the center of the pupil, after which a 6 mm line scans through eight corresponding meridians.1 This scanning capability allows for both mapping of the power and pachymetry of the cornea. Though the OCT has excellent resolution of most anterior segment structures, it has poor resolution of the anterior chamber angle.
When utilizing AS-OCT for detection of keratectasia, pachymetry proves to be the most reliable measure. In fact, AS-OCT has been shown to obtain more repeatable pachymetry measurements than Scheimpflug imaging in eyes with keratoconus.2 Accurate pachymetry mapping has become increasingly important for measuring the epithelial changes in early keratoconus. Multiple studies have confirmed that forme fruste keratoconus can manifest first with epithelial thinning in comparison to normal eyes.6
Optical low coherence reflectometry (OLCR) deserves its own mention as yet another way to assess total corneal power. It acquires corneal measurements by using A-scans with partial coherence superposition of light with a wavelength of 820 um.1 The Lenstar LS 900 optical low-coherence reflectometer/autokeratometer (Haag-Streit AG) takes 32 points in two concentric rings to give six axial length readings, four keratometry readings and one anterior chamber depth reading per measurement.1 A swept-source optical biometer (SS-OCT) like the IOL Master 700 (Carl Zeiss Meditec), uses optical B scans, rather than A scans, to measure the cornea. The IOL Master uses the B scan to penetrate the entire eye in a cross section through six meridians. As with the Lenstar, the IOL Master reads multiple components, including 18 full-length and 15 keratometry measurements per scan.
Given the many types of corneal imaging technologies, it is most important to discuss studies that compare these devices. In particular, can one device be equally compared to another or are similar devices interchangeable? Many studies compare and contrast some corneal imaging technologies, but none compare them all. The mean corneal powers calculated by the Cassini, the Atlas 9000 and the Galilei G2 were not statistically different per one study in normal and postrefractive surgery eyes.7 The Cassini was compared to the Atlas 9000 and the Galilei G2; the difference in astigmatism magnitude between the Cassini and the Atlas system increased as astigmatism increased.7
When comparing the Pentacam to RTVue 100 Fourier Domain OCT, results showed that there was repeatability and reproducibility of the corneal powers obtained by both machines, with RTVue achieving slightly steeper measures by 0.6 D.1 One study found that the Lenstar OLCR measures slightly thinner central corneal thickness, lens thickness and axial length in comparison to the Pentacam, but measured greater anterior chamber depth.8
More technically advanced machines capture more data points, which in turn create more accurate measurements for corneal power. There is much work to be done however, as no gold standard exists for measuring total corneal power and astigmatism.
We are lucky, though, to have many topographers and tomographers on hand to help us understand the cornea’s shape. This helps us in our quest to provide excellent refractive outcomes for our refractive and cataract patients. OM
- Guimaraes de Souza R, Montes de Oca I, Esquenazi I, Al-Mohtaseb Z, Weikert MP. Updates in Biometry. Int Ophthalmol Clin. 2017; 57: 115-124.
- Duncan JK, Esquenazi I, Weikert MP. New Diagnostics in Corneal Ectatic Diseases. Int Ophthalmol Clin. 2017; 57: 63-74.
- Ho JD, Tsai CY, Liou SW. Accuracy of Corneal Astigmatism Estimation by Neglecting the Posterior Corneal Surface Measurement. Am J Ophthalmol. 2009; 147: 788-795.
- Ventura BV, Al-Mohtaseb Z, Wang L, Koch DD, Weikert MP. Repeatability and Comparability of Corneal Power and Corneal Astigmatism Obtained from a Point Source Color Light-emitting Diode Topographer, a Placido-based Corneal Topographer, and a Low-coherence Reflectometer. J Cataract Refract Surg. 2015; 41: 2242-2250.
- Pinero DP, Nieto JC, Lopez-Miguel A. Characterization of Corneal Structure in Keratoconus. J Cataract Refract Surg. 2012; 38: 2167-2183.
- Li Y, Meisler DM, Tang M, et al. Keratoconus Diangosis with Optical Coherence Tomography pachymetry Mapping. Ophthalmology. 2008; 115: 2159-2166.
- Ventura BV, Wang L, Ali SF, Koch DD, Weikert MP. Comparison of Corneal Power, Astigmatism, and Wavefront Aberration Measurements Obtained by a Point-source Color Light-emmitting Diode-based Topography, a Placido-disk Topographer, and a Combined Placido and Dual Scheimpflug Device. J Cataract Refract Surg. 2015; 41: 1658-1671.
- Cinar Y, Cingu AK, Sahin M, et al. Compaison of Optical versus Ultrasonic Biometry in Keratoconic Eyes. J Ophthalmol. 2013; 2013: 1-6.