54 CATARACT & REFRACTIVE SURGERY TODAY EUROPE SEPTEMBER 2014

COVER STORY

Cassini: Providing True Axis and Magnitude

of AstigmatismMulticolored spot reflection topography produces repeatable measurements.

BY A. JOHN KANELLOPOULOS, MD; AND GEORGE ASIMELLES, PhD

Responsible for approximately two-thirds of the total ocular refractive power, the anterior corneal surface is the most significant contributor to ocular refrac-tion. Because this region of the cornea offers undis-

turbed imaging accessibility, it has been extensively studied. A long-standing technology for anterior surface evalu-

ation is Placido-disc projection topography. These non-contact topography devices enable single-shot capture of the cornea, thus reducing motion artifacts. Placido-disc topography has several limitations,1 however, which include skew ray error,2,3 lower data reliability at the cor-neal center, and susceptibility to error in areas of abrupt corneal elevation changes.4,5

Some newer systems have incorporated color-coded specular reflection and forward ray tracing,6,7 allowing imaging of radial in addition to contour topographic changes in the cornea. Instead of the traditional concen-tric mires pattern used in Placido-ring topographers,8 the prototype VU Topographer (Vrije Universiteit Medical Center) used a color-coded, chess-like pattern6 to reconstruct the anterior corneal surface. The aim of this primitive multicolor-coded corneal topographer was to eliminate source-image mismatch and enable one-to-one correspondence. In principle, it has been shown to be superior in reconstructing features of the anterior cor-neal surface that are not rotationally symmetric.9

LED POINT SOURCESLike the VU Topographer, the commercially available

Cassini system (i-Optics) also uses a multicolor (red, yellow, and green) spot pattern.10 However, instead of projecting a limited number of color-coded squares onto the cornea in a chessboard-like pattern, this topogra-pher applies up to 700 light emitting diode (LED) point sources onto the cornea and evaluates their reflection image pattern as raw data (Figure 1). The Cassini’s

software locates feature points in reflected images and accounts for smearing and deformation in irregular cor-neas.

Cassini provides axial (Figure 2A) and tangential curvature, refractive power (Figure 2B), and elevation maps calculated on an 8.5-mm corneal diameter. The device also calculates steep and flat keratometry (K), axis orientation, and related astigmatism, as well as four topographic indices relating to surface asphericity, two keratoconus indices,11 and four image quality indices.

CLINICAL EVALUATIONDue to its novelty, clinical validation of the Cassini has

yet to be achieved; however, we recently reported its use

Figure 1. The Cassini Multicolored-Spot Reflection

Topographer during acquisition (A). Example of raw data

image: The dotted red line indicates the steep meridian and

the dotted blue line the flat meridian (B).

Figure 2. Topography data is provided by axial curvature (A)

and refractive power (B) maps.

A

A B

B

SEPTEMBER 2014 CATARACT & REFRACTIVE SURGERY TODAY EUROPE 55

COVER STORY

in central corneal dystrophy imaging12 and forme fruste keratoconus diagnosis.13 We have also examined the clini-cal feasibility of multicolored spot reflection corneal topog-raphy in a large number of healthy and precataract eyes by investigating the distribution and repeatability of steep meridian and magnitude of astigmatism measurements.

In our study, three successive acquisitions with Cassini with at least 75% coverage as reported by the quality factor (QF) were obtained in each eye. The magnitude of astig-matism was defined as the difference between the steep meridian K minus the flat meridian K, and the axis of astig-matism was reported by the steep meridian orientation. Measurement repeatability was evaluated by the standard deviation of the three values of each parameter investigated.

Patients were categorized into a control group

(group A; 175 eyes), con-sisting of healthy, noncata-ract patients, and a cata-ract group (group B; 175 eyes), consisting of pre-surgery cataract patients with at least a classification of 1 on the Lens Opacity Classification System III.14,15 These two groups were further stratified according to each eye’s magnitude of astigmatism in order to investigate the relationship between repeatability of axis and magnitude measurement in each eye versus the mean magnitude of astig-matism in the specific eye. The subgroups were as follows: • Subgroup 1: astigma-tism between 0.00 and 0.99 D;• Subgroup 2: astigma-tism between 1.00 and 1.99 D;• Subgroup 3: astigma-tism between 2.00 and 2.99 D; and • Subgroup 4: astigma-tism greater than 3.00 D.

In group A, the mean age was 35.7 ±12.3 years (range, 10–65 years) and the mean preoperative

astigmatism was 1.60 ±1.45 D (range, 0.04–9.94 D). We characterized the astigmatism in this group as with-the-rule (WTR) because the steep meridian was between 80° and 110° in 73% of eyes. The average steep axis orientation was 94.78 ±28.01° (range, 4º–179.78°; Figure 3A).

The mean age in group B was 73.6 ±6.4 years (range, 61–91 years), and the mean preoperative astigmatism was 1.28 ±0.98 D (range, 0.13–6.45 D). In this group, the astigmatism was characterized as against-the-rule (ATR), as the steep axis in 69% of the eyes was between either 0º and 20° or 150º and 180°. The average steep axis orien-tation was 137.4 ±55.8° (range, 2.00º–178.8°; Figure 3B).

RESULTSResults of our study are outlined in Tables 1 and 2 and

Figure 3. Vector plots illustrate the distribution of steep meridian axis in groups A (A) and B (B).

Figure 4. Boxplot illustration of steep magnitude (A) and axis measurement (B) repeatability in

the astigmatism range in group A.

A

A

B

B

Figure 5. Boxplot illustration of steep magnitude (A) and axis measurement (B) repeatability in

the astigmatism range in group B.

A B

56 CATARACT & REFRACTIVE SURGERY TODAY EUROPE SEPTEMBER 2014

COVER STORY

in Figures 4 and 5. Astigmatism magnitude repeatability was not statistically significantly different among the paired subgroups. In group A, astigmatism repeatability ranged from 0.25 in subgroup 1 to 0.33, 0.64, and 0.59 D in sub-groups 2, 3, and 4, respectively. In group B, astigmatism repeatability ranged from 0.35 in subgroup 1 to 0.59, 0.65, and 0.61 D in subgroups 2, 3, and 4, respectively.

Our work confirms previous studies investigating cor-neal astigmatism distribution16 and the increasing preva-lence of ATR astigmatism with age.17 Specifically, in our study, the younger population had predominantly WTR astigmatism and the older population predominantly ATR.

Additionally, our study results were in line with previ-ous studies regarding the distribution of astigmatism. In a recent evaluation of 1,230 eyes in which mean patient age

was 75.54 ±10.71 years, 79.5% of eyes had 1.50 D or less of corneal astigmatism, 9.69% had more than 2.08 D, 4.61% had more than 2.50 D, 1.93% had 3.00 D or more, and 0.96% had more than 3.50 D.18 Others have reported that more than 40% of eyes had more than 1.00 D of astigma-tism19 and 25% had more than 1.50 D.20

When we used Cassini to calculate the distribution of astigmatism in our study population, about 60% of group A and 50% of group B had 1.00 D or more of astigmatism. Additionally, the repeatability of astigmatism measurement in both groups was less than 0.65 D, a parameter that slightly increased as the magnitude of astigmatism increased. By contrast, the repeatability of axis measurement improved as the magnitude of astigmatism increased (Tables 1 and 2 and Figure 4). Specifically, for example, in group B, repeat-ability was 1.25° in subgroup 2 and 0.62° in subgroup 4; in the

TABLE 1. REPEATABILITY OF ASTIGMATISM MAGNITUDE AND AXIS MEASUREMENTS

VERSUS THE AMOUNT OF ASTIGMATISM IN IN GROUP A

Repeatability Astigmatism (D)

Repeatability Axis (°)

AverageAstigmatism (D)

Subgroup 1 (n=70)

mean 0.25 2.49 0.66

SD 0.28 1.47 0.22

min 0.01 0.23 0.17

max 1.51 5.67 0.99

Subgroup 2 (n=56)

mean 0.33 2.05 1.43

SD 0.35 1.26 0.31

min 0.01 0.21 1.01

max 1.54 4.16 1.99

Subgroup 3 (n=26)

mean 0.64 1.44 2.43

SD 0.90 0.95 0.33

min 0.02 0.05 2.00

max 3.51 4.16 2.99

Subgroup 4 (n=23)

mean 0.59 1.14 4.80

SD 0.49 0.68 1.69

min 0.10 0.14 3.14

max 1.91 2.65 9.22

SD = standard deviation; min = minimum; max = maximum

TABLE 2. REPEATABILITY OF ASTIGMATISM MAGNITUDE AND AXIS MEASUREMENTS VERSUS THE AMOUNT OF ASTIGMATISM

IN GROUP B

Repeatability Astigmatism (D)

Repeatability Axis (°)

AverageAstigmatism (D)

Subgroup 1 (n=81)

mean 0.35 3.03 0.65

SD 0.37 1.48 0.19

min 0.06 0.53 0.17

max 1.42 5.86 0.99

Subgroup 2 (n=49)

mean 0.59 2.70 1.34

SD 0.53 1.30 0.22

min 0.04 0.97 1.04

max 1.90 4.88 1.73

Subgroup 3 (n=26)

mean 0.65 1.25 2.61

SD 0.89 0.83 0.28

min 0.00 0.35 2.22

max 2.46 2.42 2.93

Subgroup 4 (n=10)

mean 0.61 0.62 3.51

SD 0.37 0.62 0.39

min 0.18 0.13 3.09

max 0.82 1.32 3.88

SD = standard deviation; min = minimum; max = maximum

58 CATARACT & REFRACTIVE SURGERY TODAY EUROPE SEPTEMBER 2014

COVER STORY

same subgroups in group A, repeatability was 1.44° and 1.14°, respectively. These results may be explained by the fact that, in the older age group B, astigmatism was largely ATR, with the steep axis oriented largely horizontally. This may bear clinical value, as the Cassini device may be used to determine the correct axis for toric IOL placement.

SURGICAL PLANNINGAccurate representation of the anterior corneal surface

on topography has become an important component of cataract surgery planning, and continual developments in this area have led to innovative proposals for more accu-rate imaging.21,22 In addition to accounting for spherical error, achieving emmetropia often requires astigmatism correction with toric IOLs and/or astigmatic keratotomy. Therefore, proper identification of preoperative corneal astigmatism, both in magnitude and axis, is crucial.

Astigmatism shifts from WTR to ATR occur with age. In our study, repeatability of astigmatism axis was less than 3º and repeatability of astigmatism magnitude measurements was less than 0.60 D when taken with the multicolored spot reflection topography of the Cassini. These may be seen as impressive sensitivity benchmarks. Even better repeatability on axis measurement was noted in eyes with higher astigmatism.

AN INTEGRAL TOOLOver the past few years, we have become accustomed

to the use of extensive corneal imaging for preoperative planning in cataract and refractive surgeries, for postopera-tive assessments in these surgeries, and also when dealing with corneal irregularities such as scars and keratoconus. The traditional modalities of Placido-disc topography and Scheimpflug tomography have been illustrative; however,

the Cassini system with its brilliant precision and repeat-ability has further helped in cases in which the first two technologies were in disagreement or central irregularities had to be studied.

For the past year, the Cassini system has been an inte-gral component of our routine diagnostics protocol for IOL calculation, corneal screening for refractive surgery, post-LASIK assessments, and understanding irregular cor-neas and their topography-guided normalization. n

George Asimellis, PhD, is in the clinical research department of the LaserVision.gr Eye Institute in Athens, Greece. He states that he has no financial interest in the products or companies mentioned.

A. John Kanellopoulos, MD, is the Director of the LaserVision.gr Eye Institute in Athens, Greece, and is a Clinical Professor of Ophthalmology at New York University School of Medicine. He is an Associate Chief Medical Editor of CRST Europe. Dr. Kanellopoulos states that he is a consultant to Alcon, Avedro, i-Optics, and Oculus. He may be reached at tel: +30 21 07 47 27 77; e-mail: ajkmd@mac.com.

1. Rand RH, Howland HC, Applegate RA. Mathematical model of a Placido disk keratometer and its implications for recovery of corneal topography. Optom Vis Sci. 1997;74(11):926-930.2. Klein SA. Axial curvature and the skew ray error in corneal topography. Optom Vis Sci. 1997;74:931-944.3. Iskander DR, Davis BA, Collins MJ. The skew ray ambiguity in the analysis of videokeratoscopic data. Optom Vis Sci. 2007;84:435-442.4. Klein SA. Corneal topography reconstruction algorithm that avoids the skew ray ambiguity and the skew ray error. Optom Vis Sci. 1997;74:945-962. 5. Kanellopoulos AJ, Asimellis G. Comparison of Placido disc and Scheimpflug image-derived topography-guided excimer laser surface normalization combined with higher fluence CXL: the Athens Protocol, in progressive keratoconus. Clin Ophthalmol. 2013;7:1385-1396.6. Vos FM, van der Heijde RGL, Spoelder HJW, et al. A new instrument to measure the shape of the cornea based on pseudorandom color coding. IEEE Trans Instrum Meas. 1997;46:794-797.7. Snellenburg JJ, Braaf B, Hermans EA, et al. Forward ray tracing for image projection prediction and surface reconstruction in the evaluation of corneal topography systems. Opt Express. 2010;18:19324-19338.8. Sicam VA, Snellenburg JJ, van der Heijde RG, et al. Pseudo forward ray-tracing: a new method for surface valida-tion in cornea topography. Optom Vis Sci. 2007;84(9):915-923.9. Sicam VA, van der Heijde RG. Topographer reconstruction of the nonrotation-symmetric anterior corneal surface features. Optom Vis Sci. 2006;83(12):910-918.10. Weikert MP, Koch DD, Wang D. Evaluation of corneal topography based on color LED technology. Paper presented at: the ASCRS annual symposium; April 20, 2013; San Francisco.11. Smolek MK, Klyce SD. Current keratoconus detection methods compared with a neural network approach. Invest Ophthalmol Vis Sci. 1997;38:2290-2299.12. Kanellopoulos AJ, Asimellis G. Novel multicolor spot reflection topography. Initial clinical findings. Digital J Ophthalmol. 2014; [under review].13. Kanellopoulos AJ, Asimellis G. Forme fruste keratoconus imaging and validation via point-source reflection topography. Case Rep Ophthalmol. 2013;4(3):199-209.14. Chylack LT Jr, Wolfe JK, Singer DM, et al; for the Longitudinal Study of Cataract Study Group. The lens opacities classification system III. Arch Ophthalmol. 1993;111:831-836.15. Gupta M, Ram J, Jain A, et al. Correlation of nuclear density using the Lens Opacity Classification System III versus Scheimpflug imaging with phacoemulsification parameters. J Cataract Refract Surg. 2013;39(12):1818-1823.16. Hayashi K, Hayashi H, Hayashi F. Topographic analysis of the changes in corneal shape due to aging. Cornea. 1995;14(5):527-532.17. Asano K, Nomura H, Iwano M, et al. Relationship between astigmatism and aging in middle-aged and elderly Japanese. Jpn J Ophthalmol. 2005;49(2):127-133.18. Khan MI, Muhtaseb M. Prevalence of corneal astigmatism in patients having routine cataract surgery at a teach-ing hospital in the United Kingdom. J Cataract Refract Surg. 2011;37(10):1751-1755.19. Ferrer-Blasco T, Montés-Micó R, Peixoto-de-Matos SC, et al. Prevalence of corneal astigmatism before cataract surgery. J Cataract Refract Surg. 2009;35:70-75.20. Hoffer KJ. Biometry of 7,500 cataractous eyes. Am J Ophthalmol. 1980;90:360-368.21. Maguire LJ, Singer DE, Klyce SD. Graphic presentation of computer-analyzed keratoscope photographs. Arch Ophthalmol. 1987;105(2):223-230.22. Koch DD, Foulks GN, Moran CT, et al. The Corneal EyeSys System: Accuracy analysis and reproducibility of first-generation prototype. Refract Corneal Surg. 1989;5(6):424-429.

• Inprinciple,amulticolor-codedcornealtopographercanreconstructfeaturesoftheanteriorcornealsurfacethatarenotrotationallysymmetric.

• TheCassinisystemusesared,yellow,andgreenspotpatternandappliesupto700LEDpointsourcesontothecornea.Thesoftwareevaluatesthereflectionimagepatternasrawdataandlocatesfeaturepointsinreflectedimagesandaccountsforsmearinganddeformationinirregularcorneas.

• Cassiniprovidesaxialandtangentialcurvature,refractivepower,andelevationmapscalculatedonan8.5-mmcornealdiameterandcalculatessteepandflatK,axisorientation,andrelatedastigmatismaswellasfourtopographicindicesrelatingtosurfaceasphericity,twokeratoconusindices,andfourimagequalityindices.

TAKE-HOME MESSAGE