Journal of Refractive Surgery

Original Article Supplemental Data

Differences in Posterior Corneal Features Between Normal Corneas and Subclinical Keratoconus

Oren Golan, MD; Eric S. Hwang, BS; Paul Lang, BS; Marcony R. Santhiago, MD, PhD; Adi Abulafia, MD; David Touboul, MD, PhD; Mark Krauthammer, MD; David Smadja, MD

Abstract

PURPOSE:

To compare posterior corneal features and their discriminating power for differentiating normal corneas from subclinical keratoconus using the Placido dual-Scheimpflug analyzer.

METHODS:

Patients were retrospectively included in the study. The preoperative normal right eyes of 79 patients imaged with a Placido dual-Scheimpflug system and with a stable postoperative LASIK follow-up of a minimum of 36 months were included in the normal group and were compared to 39 contralateral topographically normal eyes with clinically evident keratoconus in the fellow eye. The posterior surface variables measured were categorized according to the feature of the corneal shape they were characterizing (curvature, elevation, asymmetry, and eccentricity) and compared between the two groups using the Student's two-sample t test. The discriminating ability of the posterior surface variables was compared by receiver operator characteristics curves.

RESULTS:

Variables that related to asymmetry and elevation of the posterior surface were statistically significantly different between groups (P < .05), whereas eccentricity and curvature–related parameters were not. Receiver operator characteristics curves analysis showed that the maximum posterior elevation over the best-fit toric and aspheric surface reference shape had the highest discriminating ability for distinguishing normal corneas from subclinical keratoconus, with an area under the curve of 0.877, followed by the asphericity asymmetry index, with an area under the curve of 0.871, and posterior inferior-superior value, with an area under the curve of 0.851.

CONCLUSIONS:

Posterior cornea measured with a dual-Scheimpflug analyzer provides useful parameters for differentiating normal corneas from subclinical keratoconus. Of the posterior surface parameters, asymmetry and elevation seem to be the most sensitive shape modifications for differentiating both populations.

[J Refract Surg. 2018;34(10):664–670.]

Abstract

PURPOSE:

To compare posterior corneal features and their discriminating power for differentiating normal corneas from subclinical keratoconus using the Placido dual-Scheimpflug analyzer.

METHODS:

Patients were retrospectively included in the study. The preoperative normal right eyes of 79 patients imaged with a Placido dual-Scheimpflug system and with a stable postoperative LASIK follow-up of a minimum of 36 months were included in the normal group and were compared to 39 contralateral topographically normal eyes with clinically evident keratoconus in the fellow eye. The posterior surface variables measured were categorized according to the feature of the corneal shape they were characterizing (curvature, elevation, asymmetry, and eccentricity) and compared between the two groups using the Student's two-sample t test. The discriminating ability of the posterior surface variables was compared by receiver operator characteristics curves.

RESULTS:

Variables that related to asymmetry and elevation of the posterior surface were statistically significantly different between groups (P < .05), whereas eccentricity and curvature–related parameters were not. Receiver operator characteristics curves analysis showed that the maximum posterior elevation over the best-fit toric and aspheric surface reference shape had the highest discriminating ability for distinguishing normal corneas from subclinical keratoconus, with an area under the curve of 0.877, followed by the asphericity asymmetry index, with an area under the curve of 0.871, and posterior inferior-superior value, with an area under the curve of 0.851.

CONCLUSIONS:

Posterior cornea measured with a dual-Scheimpflug analyzer provides useful parameters for differentiating normal corneas from subclinical keratoconus. Of the posterior surface parameters, asymmetry and elevation seem to be the most sensitive shape modifications for differentiating both populations.

[J Refract Surg. 2018;34(10):664–670.]

Corneal ectasia after excimer laser refractive surgery is one of the major concerns a refractive surgeon should take into account before deciding to perform surgery. An abnormal topography pattern is a major risk factor that can lead to postoperative corneal ectasia if not identified beforehand.1 Screening for subclinical keratoconus remains challenging. Historically, anterior surface parameters have been found to have a stronger predictable power compared to the posterior parameters for the screening of subclinical keratoconus,2–7 but the topic is still debated in the literature because some authors suggest that posterior surface changes may occur in the early stages of keratoconus.8–14 Other studies suggest using the combination of multiple variables such as anterior and posterior surface parameters, wavefront higher order aberrations, spatial corneal thickness, and epithelial thickness maps to reach a higher predictable ability to detect early changes of keratoconus.13,15–23 It is not in the scope of this study to further debate this topic. However, the question of which posterior corneal feature has the highest discriminating power among the posterior surface parameters remains unaddressed.

The objective of our study was to compare the posterior shape features of normal corneas and the mildest stage of keratoconus ectatic disease, defined as subclinical keratoconus. Additionally, to better understand which shape modification is the most relevant, we categorized all parameters investigated according to which shape features they were characterizing: steepening, bulging, asymmetry, or asphericity (cornea becoming more prolate or oblate).

Patients and Methods

This retrospective and comparative study was conducted at the University Hospital of Bordeaux, France, a division of the National Reference Center for Keratoconus, and in the Tel Aviv Sourasky Medical Center in Israel and was approved by the institutional review board of the National Reference Center for Keratoconus. The study was conducted in accordance with the tenets of the Declaration of Helsinki.

Patients

Charts of the patients with keratoconus referred to the National Reference Center for Keratoconus between April 2011 and November 2014 were reviewed, and patients with subclinical keratoconus were selected for the study. The subclinical keratoconus group included contralateral topographically normal eyes with clinically evident keratoconus in the fellow eye, normal topographical aspects with no asymmetry in any direction greater than 0.50 diopter (D), no asymmetric bowtie pattern, no focal or inferior steepening pattern,24 and no suspect keratoconus pattern detected with the Klyce24 and Maeda et al.25 algorithm used in the topographer (TMS 4; Tomey Corporation, Nagoya, Japan). This condition is also known in the literature as “forme fruste keratoconus” because it has already been reported that approximately 50% of clinically normal fellow eyes of patients with unilateral keratoconus progressed to keratoconus within 16 years, with a greater risk during the first 6 years of onset.26 Examples of subclinical keratoconus that compose this sample are shown in Figures AC (available in the online version of this article).

An example of an anterior curvature map showing normal features.

Figure A.

An example of an anterior curvature map showing normal features.

An example of a 4-screen display showing normal features of the anterior curvature map.

Figure B.

An example of a 4-screen display showing normal features of the anterior curvature map.

An example of a 4-screen display showing normal features of the anterior curvature map.

Figure C.

An example of a 4-screen display showing normal features of the anterior curvature map.

Normal corneas, often challenging to define, included corneas that had normal preoperative corneal topography and underwent LASIK surgery with a minimum follow-up of 36 months without developing post-LASIK ectasia. Eyes with previous ocular surgery, ocular pathology, contact lenses worn 1 or 2 weeks before examination (soft or rigid), and low-quality topography maps that did not satisfy the minimum quality required by the system were excluded from the analysis.

Dual-Scheimpflug Analyzer System and Parameters

All measurements were performed with the Galilei dual-Scheimpflug analyzer (version 5.2.1; Ziemer Ophthalmic Systems AG, Port, Switzerland) according to the manufacturer's guidelines. The Galilei is a rotating Scheimpflug tomography–based device combining dual-channel Scheimpflug cameras and a Placido disc. The system acquires between 15 and 60 Scheimpflug images per scan and 2 Placido top view images at 90° apart, as the cameras rotate around the central axis. Placido and Scheimpflug data are acquired simultaneously, and then a motion-correction algorithm is applied to the combined dataset. Height data from the Scheimpflug images and slope data, converted into height data from the Placido device, are merged to provide a surface fitted to the anterior corneal data, whereas posterior corneal surface data are measured using edge detection in images provided by the dual-Scheimpflug system. In addition, the Galilei software has enhanced the posterior edge location and uses ray-tracing with a Snell's law refraction through the anterior surface to locate and help in reliably reconstructing the posterior surface. The following posterior corneal parameters were collected and analyzed and can be briefly described as follows.

Posterior curvature data with the steep, flat, and mean keratometry, and posterior astigmatism are basically the same as for the anterior surface readings, except that keratometry values (steep and flat) at the posterior surface are not simulated because these values are calculated with the real indices of refraction of the cornea (1.376) and the aqueous humor (1.336), which explain the negative sign of posterior keratometry readings.

The posterior inferior–superior (I-S) index has been designed for evaluating the curvature asymmetry of the posterior surface. It is somewhat equivalent to the well-known anterior I-S value proposed by Rabinowitz,27 but at the level of the posterior cornea. It is calculated by subtracting the superior keratometric value from the inferior value. Because keratometry readings are negative at the posterior cornea, the superior and inferior values were calculated by averaging the absolute value of 5 data points, respectively, along the superior or inferior posterior cornea, 3 mm from the center of the cornea at 30° intervals (at 210°, 240°, 270°, 300°, and 330°). Therefore, the higher the value is, the more it reflects an inferior steepening of the posterior surface.

Posterior eccentricity (p∊2) is one of the four parameters by which the shape of a conic section can be described; Q (asphericity), p (P value), and E (corneal shape factor) are the others. These terms are mathematically related by the following equation: ∊2 = E = 1 − p = −Q. It is calculated within a central diameter of 8 mm averaged over all meridians of the anterior corneal surface. A positive value refers to a prolate shape of the corneal surface, whereas a negative value refers to an oblate shape.

Posterior elevation data were measured with two different reference bodies over an 8-mm calculation zone: best-fit sphere (BFS) in float mode and best-fit toric and aspheric body (BFTA). Values were recorded over three locations by manually guiding the cursor over the posterior elevation maps: maximum posterior elevation value within the 8-mm diameter zone (MPE), posterior elevation value at the thinnest point, and elevation value at the maximum keratometry location.

The posterior asphericity asymmetry index (AAI) has been designed for quantifying the asymmetry of asphericity of the corneal surface. It has been recently shown to be highly discriminant for identifying sub-clinical keratoconus.13 It is calculated over the BFTA map as the absolute value of the difference between the maximum negative elevation value (maximum depression) and MPE value within the central 6-mm diameter data zone as illustrated in Figure 1.

How to calculate the posterior asphericity asymmetry index (AAI). BFTA = best-fit toric and aspheric body

Figure 1.

How to calculate the posterior asphericity asymmetry index (AAI). BFTA = best-fit toric and aspheric body

Data and Statistical Analyses

In an attempt to distinguish what features of the corneal shape were the most relevant to track down for differentiating these two populations, all variables analyzed were categorized according to four potential features and the actions that they were characterizing, as summarized in Table A (available in the online version of this article) and compared between the two groups: (1) steepening of the cornea (curvature-derived parameters); (2) bulging of the cornea (elevation-derived parameters); (3) asymmetry of the cornea; and (4) eccentricity/asphericity metrics.

Categorization of the Posterior Corneal Parameters Analyzed According to the Corneal Shape Characteristic They Represent

Table A:

Categorization of the Posterior Corneal Parameters Analyzed According to the Corneal Shape Characteristic They Represent

Statistical analyses were performed using JMP software (version 8.0; SAS Institute, Inc., Cary, NC). Normal distribution of quantitative data was investigated and confirmed by the Kolmogorov–Smirnov test. Therefore, differences between data were evaluated using the Student's two-sample t test. Data were expressed as mean ± standard deviation (SD). The level of significance for each parameter was set at a P value less than .05. Receiver operating characteristics (ROC) curves were used as described4 to assess the discriminating ability and to determine the optimal cut-off values. Comparisons of area under the ROC (AUROC) values were made as previously described to test differences between corneal shape features. Statistical analyses were performed using SPSS software (version 24; SPSS, Inc., Chicago, IL).

Results

This study included 118 eyes of 118 patients divided into two groups (normal and subclinical keratoconus), including 39 eyes of 39 patients with subclinical keratoconus and 79 eyes of 79 normal patients. Baseline demographic characteristics of the patients by groups are summarized in Table 1.

Baseline Clinical and Demographic Characteristics by Group

Table 1:

Baseline Clinical and Demographic Characteristics by Group

Comparison of Mean Posterior Corneal Values

None of the posterior curvature and eccentricity–derived parameters was statistically significantly different between the groups, as summarized in Table B (available in the online version of this article). However, except for the MPE analyzed over the BFS reference surface, all elevation and asymmetry–derived parameters were statistically significantly different between the two groups (P < .001). The mean values and statistical comparison are summarized in Tables 23.

Comparison of the Posterior Corneal Variables Representing Steepening and Eccentricity Between Normal and Subclinical Keratoconus

Table B:

Comparison of the Posterior Corneal Variables Representing Steepening and Eccentricity Between Normal and Subclinical Keratoconus

Comparison of Posterior Corneal Variables Representing Bulging Between Normal and Subclinical Keratoconus

Table 2:

Comparison of Posterior Corneal Variables Representing Bulging Between Normal and Subclinical Keratoconus

Comparison of Posterior Corneal Variables Representing Asymmetry and Eccentricity Between Normal and Subclinical Keratoconus Eyes

Table 3:

Comparison of Posterior Corneal Variables Representing Asymmetry and Eccentricity Between Normal and Subclinical Keratoconus Eyes

Discriminating Ability and ROC Curves

Of the seven parameters found to be statistically significant, the three most discriminant were MPE (BFTA), AAI, and posterior I-S ratio. The MPE (BFTA) had an AUROC of 0.877 and sensitivity and specificity of 79.5% and 84.8% for a cut-off point of 11.5 μm, respectively. The AAI had an AUROC of 0.871, and sensitivity and specificity of 84.6% and 81%, respectively, for a cut-off value set at 21.5 μm. The posterior I-S ratio had an AUROC of 0.851 with sensitivity and specificity of 79.5% and 63.2%, respectively, for the cut-off point of −0.2. All results are summarized in Table 4 and illustrated in Figure 2. When comparing the discriminating abilities between these strongest variables, although the MPE (BFTA) had the highest AUROC, there was no statistically significant difference with the posterior AAI and the posterior I-S ratio, as summarized in Table C (available in the online version of this article).

Optimized Cut-off Values, Corresponding Pairs of Sensitivity/Specificity Values, and AUROC Values/Discriminate Ability Between Subclinical Keratoconus and Normal Corneas

Table 4:

Optimized Cut-off Values, Corresponding Pairs of Sensitivity/Specificity Values, and AUROC Values/Discriminate Ability Between Subclinical Keratoconus and Normal Corneas

Area under the curve (AUROC) for each posterior wall significant parameter. max PE = maximum posterior elevation; BFTA = best-fit toric and aspheric body; AAI = asphericity asymmetry index; I-S = inferior-superior; TP = thinnest point; Ks = steepest keratometry; BFS = best-fit sphere

Figure 2.

Area under the curve (AUROC) for each posterior wall significant parameter. max PE = maximum posterior elevation; BFTA = best-fit toric and aspheric body; AAI = asphericity asymmetry index; I-S = inferior-superior; TP = thinnest point; Ks = steepest keratometry; BFS = best-fit sphere

Comparison of Discriminating Ability Between the Strongest Parameters for Distinguishing Normal and Subclinical Keratoconus

Table C:

Comparison of Discriminating Ability Between the Strongest Parameters for Distinguishing Normal and Subclinical Keratoconus

Discussion

Screening performance for keratoconus has reached higher sensitivity with major advancements in imaging technology. Among the technological progress is the ability to image and measure parameters from the posterior cornea. However, the question of which posterior corneal feature has the highest discriminating power for distinguishing between normal and subclinical keratoconus remained unaddressed. In our study, we intended to address this issue by categorizing posterior corneal variables according to which corneal shape they were representing: steepening, bulging, asymmetry, or asphericity. We were able to demonstrate that, of the posterior surface metrics, asymmetry and bulging were the first detectable manifestations in earlier stages of keratoconus disease. We found that posterior steepening (curvature-related parameters) and asphericity (the cornea becoming more prolate or oblate) were not relevant for screening subclinical keratoconus at this early stage of the ectatic process.

As opposed to our findings, while studying the performance of posterior elevation in discriminating between normal and subclinical keratoconus using a Scheimpflug device (Pentacam; Oculus Optikgeräte, Wetzlar, Germany), de Sanctis et al.4 reported that the MPE calculated over the BFS was highly discriminant, with an AUROC of 0.93 and sensitivity and specificity of 68% and 90.8%, respectively, although they recommended not to use it alone but rather to combine it with anterior variables for improving screening performance. Similar findings were also reported by Uçakhan et al.28 using the same imaging technology, with a lower AUROC of 0.79 for discriminating between normal corneas and subclinical keratoconus.

In our study, we found that MPE calculated over a BFS reference surface was not statistically significantly different between subclinical keratoconus and normal corneas. We included a milder stage of the disease in our study because we defined subclinical keratoconus as normal-appearing anterior Placido features (Figures AC), whereas in both aforementioned studies, the authors used more advanced forms of keratoconus that already had “inferior–superior asymmetry and/or bowtie pattern and skewed radial axes, as detected on tangential Placido disc-based videokeratographs” to define their subclinical keratoconus. This difference is likely to explain the disparity between our findings because the more the disease progresses, the more bulging at the posterior surface you would expect to observe, making all posterior elevation parameters more relevant. Moreover, they limited their analyses on posterior elevation with the BFS, which did not allow them to investigate the relevance of other posterior corneal features, including other reference surface shape for calculating elevation. However, studies published by Smadja et al.14 and Kovács et al.29 have highlighted the relevance of using a BFTA reference surface shape over the BFS for improving the screening sensitivity of subclinical keratoconus. These findings were actually confirmed in our study because we found a better discriminating ability of the posterior elevation values over the BFTA compared to the BFS.

Keratoconus has always been known as an asymmetric disease, and many relevant indices calculated over the anterior cornea have been reported to be highly sensitive for improving the detection of the ectatic disease, such as the I-S value,27 opposite sector index,25 surface asymmetry index,30 and coma-like aberration.31,32 Despite the surge of interest in posterior surface measurements for improving the screening performance at earlier stages of the ectatic disease, few studies have reported the relevance of posterior asymmetry in diagnosing the subclinical forms of keratoconus. This interesting finding may be due to the absence of asymmetry-related parameters in most of the commercial devices that usually are limited to measurements of elevation, curvature, and asphericity.

However, in a recent study published by Smadja et al.,13 the posterior AAI was the most discriminant parameter among 56 other corneal variables by an artificial intelligence software that received the task of building an automated decision tree with discriminating rule for distinguishing subclinical keratoconus from normal corneas. Similar findings were also reported by Arbelaez et al.23 using different artificial intelligence software, where posterior surface features including posterior AAI markedly improved the sensitivity in diagnosing subclinical keratoconus. Bae et al.2 also compared corneal asymmetry between normal corneas and subclinical keratoconus by manual calculation, adding the absolute value of the maximum depression and maximum elevation at both the anterior and posterior cornea, over the BFS elevation maps. Both anterior and posterior asymmetries were significantly different between the two groups, with AUROCs of 0.734 and 0.735, respectively. However, although not significant, posterior corneal asymmetry had slightly higher sensitivity and specificity values of 57.14% and 88.24%, respectively, compared to the anterior corneal asymmetry with 50% and 85.29%, respectively. Using similar manual calculations and multivariable regression analysis, Nilforoushan et al.33 reported posterior asymmetry to be the strongest predictor of suspect corneas.

Although our study used the posterior AAI that is calculated over the BFTA reference surface and not the BFS, we found that it belongs to the group of the posterior surface metrics with the highest predictive ability to distinguish between normal corneas and subclinical keratoconus, with an AUROC of 0.871 and a sensitivity and specificity of 84.6% and 81%, respectively, for the cut-off set at 21.5 μm. Second, using BFTA to calculate the asymmetry index seems to improve the sensitivity of detection,14,29 as we could observe when comparing the difference in AUROC values between the current study (0.871) and the two aforementioned studies that used the BFS for calculating posterior asymmetry 0.735 and 0.80, respectively, in Bae et al.'s2 and Nilforoushan et al.'s33 studies.

In light of our findings, although their discriminating abilities were not high enough to be used alone in identifying the earliest stage of the ectatic disease, when looking at posterior surface parameters as part of the screening process, special attention should be paid to the asymmetry and elevation–related parameters because they provide the highest discriminating ability among the posterior surface metrics to distinguish between normal corneas and subclinical keratoconus.

References

  1. Randleman JB, Woodward M, Lynn MJ, Stulting RD. Risk assessment for ectasia after corneal refractive surgery. Ophthalmology. 2008;115:37–50. doi:10.1016/j.ophtha.2007.03.073 [CrossRef]
  2. Bae GH, Kim JR, Kim CH, Lim DH, Chung ES, Chung TY. Corneal topographic and tomographic analysis of fellow eyes in unilateral keratoconus patients using Pentacam. Am J Ophthalmol. 2014;157:103–109. doi:10.1016/j.ajo.2013.08.014 [CrossRef]
  3. de Sanctis U, Aragno V, Dalmasso P, Brusasco L, Grignolo F. Diagnosis of subclinical keratoconus using posterior elevation measured with 2 different methods. Cornea. 2013;32:911–915. doi:10.1097/ICO.0b013e3182854774 [CrossRef]
  4. de Sanctis U, Loiacono C, Richiardi L, Turco D, Mutani B, Grignolo FM. Sensitivity and specificity of posterior corneal elevation measured by Pentacam in discriminating keratoconus/subclinical keratoconus. Ophthalmology. 2008;115:1534–1539. doi:10.1016/j.ophtha.2008.02.020 [CrossRef]
  5. Fam HB, Lim KL. Corneal elevation indices in normal and keratoconic eyes. J Cataract Refract Surg. 2006;32:1281–1287. doi:10.1016/j.jcrs.2006.02.060 [CrossRef]
  6. Rabinowitz YS, Rasheed K. KISA% index: a quantitative videokeratography algorithm embodying minimal topographic criteria for diagnosing keratoconus. J Cataract Refract Surg. 1999;25:1327–1335. doi:10.1016/S0886-3350(99)00195-9 [CrossRef]
  7. Smolek MK, Klyce SD. Current keratoconus detection methods compared with a neural network approach. Invest Ophthalmol Vis Sci. 1997;38:2290–2299.
  8. Saad A, Gatinel D. Topographic and tomographic properties of forme fruste keratoconus corneas. Invest Ophthalmol Vis Sci. 2010;51:5546–5555. doi:10.1167/iovs.10-5369 [CrossRef]
  9. Pinero DP, Alió JL, Aleson A, Escaf Vergara M, Miranda M. Corneal volume, pachymetry, and correlation of anterior and posterior corneal shape in subclinical and different stages of clinical keratoconus. J Cataract Refract Surg. 2010;36:814–825. doi:10.1016/j.jcrs.2009.11.012 [CrossRef]
  10. Tomidokoro A, Oshika T, Amano S, Higaki S, Maeda N, Miyata K. Changes in anterior and posterior corneal curvatures in keratoconus. Ophthalmology. 2000;107:1328–1332. doi:10.1016/S0161-6420(00)00159-7 [CrossRef]
  11. Schlegel Z, Hoang-Xuan T, Gatinel D. Comparison of and correlation between anterior and posterior corneal elevation maps in normal eyes and keratoconus-suspect eyes. J Cataract Refract Surg. 2008;34:789–795. doi:10.1016/j.jcrs.2007.12.036 [CrossRef]
  12. Lim L, Wei RH, Chan WK, Tan DT. Evaluation of keratoconus in Asians: role of Orbscan II and Tomey TMS-2 corneal topography. Am J Ophthalmol. 2007;143:390–400. doi:10.1016/j.ajo.2006.11.030 [CrossRef]
  13. Smadja D, Touboul D, Cohen A, et al. Detection of subclinical keratoconus using an automated decision tree classification. Am J Ophthalmol. 2013;156:237–246. doi:10.1016/j.ajo.2013.03.034 [CrossRef]
  14. Smadja D, Santhiago MR, Mello GR, Krueger RR, Colin J, Touboul D. Influence of the reference surface shape for discriminating between normal corneas, subclinical keratoconus, and keratoconus. J Refract Surg. 2013;29:274–281. doi:10.3928/1081597X-20130318-07 [CrossRef]
  15. Saad A, Gatinel D. Combining Placido and corneal wavefront data for the detection of forme fruste keratoconus. J Refract Surg. 2016;32:510–516. doi:10.3928/1081597X-20160523-01 [CrossRef]
  16. Li Y, Chamberlain W, Tan O, Brass R, Weiss JL, Huang D. Subclinical keratoconus detection by pattern analysis of corneal and epithelial thickness maps with optical coherence tomography. J Cataract Refract Surg. 2016;42:284–295. doi:10.1016/j.jcrs.2015.09.021 [CrossRef]
  17. Kovács I, Miháltz K, Kránitz K, et al. Accuracy of machine learning classifiers using bilateral data from a Scheimpflug camera for identifying eyes with preclinical signs of keratoconus. J Cataract Refract Surg. 2016;42:275–283. doi:10.1016/j.jcrs.2015.09.020 [CrossRef]
  18. Muftuoglu O, Ayar O, Hurmeric V, Orucoglu F, Kilic I. Comparison of multimetric D index with keratometric, pachymetric, and posterior elevation parameters in diagnosing subclinical keratoconus in fellow eyes of asymmetric keratoconus patients. J Cataract Refract Surg. 2015;41:557–565. doi:10.1016/j.jcrs.2014.05.052 [CrossRef]
  19. Ruiseñor Vázquez PR, Galletti JD, Minguez N, et al. Pentacam Scheimpflug tomography findings in topographically normal patients and subclinical keratoconus cases. Am J Ophthalmol. 2014;158:32–40. doi:10.1016/j.ajo.2014.03.018 [CrossRef]
  20. Bühren J, Schäffeler T, Kohnen T. Validation of metrics for the detection of subclinical keratoconus in a new patient collective. J Cataract Refract Surg. 2014;40:259–268. doi:10.1016/j.jcrs.2013.07.044 [CrossRef]
  21. Jafarinasab MR, Feizi S, Karimian F, Hasanpour H. Evaluation of corneal elevation in eyes with subclinical keratoconus and keratoconus using Galilei double Scheimpflug analyzer. Eur J Ophthalmol. 2013;23:377–384. doi:10.5301/ejo.5000226 [CrossRef]
  22. Saad A, Gatinel D. Evaluation of total and corneal wavefront high order aberrations for the detection of forme fruste keratoconus. Invest Ophthalmol Vis Sci. 2012;53:2978–2992. doi:10.1167/iovs.11-8803 [CrossRef]
  23. Arbelaez MC, Versaci F, Vestri G, Barboni P, Savini G. Use of a support vector machine for keratoconus and subclinical keratoconus detection by topographic and tomographic data. Ophthalmology. 2012;119:2231–2238. doi:10.1016/j.ophtha.2012.06.005 [CrossRef]
  24. Klyce SD. Chasing the suspect: keratoconus. Br J Ophthalmol. 2009;93:845–847. doi:10.1136/bjo.2008.147371 [CrossRef]
  25. Maeda N, Klyce SD, Smolek MK, Thompson HW. Automated keratoconus screening with corneal topography analysis. Invest Ophthalmol Vis Sci. 1994;35:2749–2757.
  26. Li X, Rabinowitz YS, Rasheed K, Yang H. Longitudinal study of the normal eyes in unilateral keratoconus patients. Ophthalmology. 2004;111:440–446. doi:10.1016/j.ophtha.2003.06.020 [CrossRef]
  27. Rabinowitz YS. Videokeratographic indices to aid in screening for keratoconus. J Refract Surg. 1995;11:371–379.
  28. Uçakhan ÖÖ, Cetinkor V, Özkan M, Kanpolat A. Evaluation of Scheimpflug imaging parameters in subclinical keratoconus, keratoconus, and normal eyes. J Cataract Refract Surg. 2011;37:1116–1124. doi:10.1016/j.jcrs.2010.12.049 [CrossRef]
  29. Kovács I, Miháltz K, Ecsedy M, Németh J, Nagy ZZ. The role of reference body selection in calculating posterior corneal elevation and prediction of keratoconus using rotating Scheimpflug camera. Acta Ophthalmol. 2011;89:e251–e256. doi:10.1111/j.1755-3768.2010.02053.x [CrossRef]
  30. Kanellopoulos AJ, Asimellis G. Revisiting keratoconus diagnosis and progression classification based on evaluation of corneal asymmetry indices, derived from Scheimpflug imaging in keratoconic and suspect cases. Clin Ophthalmol. 2013;7:1539–1548. doi:10.2147/OPTH.S44741 [CrossRef]
  31. Bühren J, Kook D, Yoon G, Kohnen T. Detection of subclinical keratoconus by using corneal anterior and posterior surface aberrations and thickness spatial profiles. Invest Ophthalmol Vis Sci. 2010;51:3424–3432. doi:10.1167/iovs.09-4960 [CrossRef]
  32. Alió JL, Shabayek MH. Corneal higher order aberrations: a method to grade keratoconus. J Refract Surg. 2006;22:539–545.
  33. Nilforoushan MR, Speaker M, Marmor M, et al. Comparative evaluation of refractive surgery candidates with Placido topography, Orbscan II, Pentacam, and wavefront analysis. J Cataract Refract Surg. 2008;34:623–631. doi:10.1016/j.jcrs.2007.11.054 [CrossRef]

Baseline Clinical and Demographic Characteristics by Group

CharacteristicNormalSubclinical KeratoconusPa
No. of patients7939
Mean ± SD age (y)29 ± 9.8528 ± 8.8.66
Male/female ratio38/4029/10
Anterior SimK steep (D)44.12 ± 2.1144.32 ± 2.53.65
Anterior SimK flat (D)43.08 ± 1.6143.07 ± 1.84.97
Keratometric astigmatism (D)1.04 ± 0.961.25 ± 1.40.34
Average corneal thickness (μm)543.83 ± 34.65516.08 ± 33.58.0000726

Comparison of Posterior Corneal Variables Representing Bulging Between Normal and Subclinical Keratoconus

ParameterMean ± SD (Range)Pa,b

NormalSubclinical Keratoconus
MPE-TP (BTFA)−0.67 ± 2.19 (−7 to 5)2.23 ± 2.91 (−9 to 9)< .001
MPE-Ks (BFTA)4.4 ± 3.77 (−5 to 17)8.18 ± 6.6 (−3 to 24)< .001
Max PE (BFTA)8.73 ± 2.88 (4 to 21)14.95 ± 5.05 (4 to 30)< .001
MPE-TP (BFS)2.27 ± 2.26 (−2 to 9)6.28 ± 4.98 (−3 to 22)< .001
MPE-Ks (BFS)1.73 ± 3.57 (−8 to 10)4.38 ± 5.47 (−11 to 21)< .001
Max PE (BFS)12.03 ± 4.41 (3 to 27)13 ± 6.01 (3 to 34).33

Comparison of Posterior Corneal Variables Representing Asymmetry and Eccentricity Between Normal and Subclinical Keratoconus Eyes

VariableMean ± SD (Range)Pa,b

NormalSubclinical Keratoconus
AAI posterior17.35 ± 5.13 (0 to 34)29.49 ± 9.85 (10 to 55)< .001
Posterior I-S ratio−0.40 ± 0.48 (−1.4 to 0.7)0.98 ± 1.2 (−1.4 to 4.6)< .001

Optimized Cut-off Values, Corresponding Pairs of Sensitivity/Specificity Values, and AUROC Values/Discriminate Ability Between Subclinical Keratoconus and Normal Corneas

ParameterSensitivity (%)Specificity (%)Cut-off ValueAUROCa
MPE-TP (BTFA)71.874.60.5 μm0.782
MPE-Ks (BFTA)5974.76.5 μm0.668
MPE (BFTA)79.584.811.5 μm0.877
MPE-TP (BFS)64.1824.5 μm0.778
MPE-Ks (BFS)66.773.43.5 μm0.693
AAI84.68121.50.871
Posterior I-S ratio79.563.2−0.20 D0.851

Categorization of the Posterior Corneal Parameters Analyzed According to the Corneal Shape Characteristic They Represent

Steepening (D)Bulging (μm)AsymmetryEccentricity
Axial posterior K steepMPE-TP (BFTA)Posterior AAI (μm)Posterior eccentricity
Axial posterior K flatMPE-Ks (BFTA)Posterior I-S
Axial posterior cylinderMax PE (BFTA)
Central posterior power (1–4 mm)MPE-TP (BFS)
Paracentral posterior power (4–7 mm)MPE-Ks (BFS)
Peripheral posterior power (7–8 mm)Max PE (BFS)

Comparison of the Posterior Corneal Variables Representing Steepening and Eccentricity Between Normal and Subclinical Keratoconus

ParameterMean ± SD (Range) (D)Pa

NormalSubclinical Keratoconus
Posterior curvature
  Axial kertametry steepest−6.39 ± 0.37 (−8.50 to −5.70)−6.36 ± 0.38 (−7.34 to −5.63).63
  Axial keratometry flattest−6.09 ± 0.22 (−7.40 to −5.60)−6.04 ± 0.27 (−6.64 to −5.46).34
  Cylinder−0.29 ± 0.14 (−1.20 to −0.04)−0.32 ± 0.22 (−1.24 to −0.07).57
  Axial average, central−6.24 ± 0.34 (−8.00 to −5.65)−6.20 ± 0.32 (−7.01 to −5.55).49
  Instantaneous average, mid−5.8 ± 0.35 (−6.60 to −4.16)−5.76 ± 0.31 (−6.34 to −4.95).51
  Instantaneous average, outer−5.22 ± 0.36 (−5.80 to −3.46)−5.17 ± 0.32 (−5.74 to −4.34).42
Posterior eccentricity0.25 ± 0.50 (−0.20 to 3.50)0.24 ± 0.29 (−0.23 to 0.99).89

Comparison of Discriminating Ability Between the Strongest Parameters for Distinguishing Normal and Subclinical Keratoconus

ParameterAUROC95% CIAAIMax PEPosterior I-S Ratio
AAI0.8710.793 to 0.950
MAX PE (BFTA)0.8770.803 to 0.952P > .05P > .05
Posterior I-S ratio0.8510.763 to 0.939P > .05
Authors

From the Department of Ophthalmology, Tel Aviv Sourasky Medical Center, Tel Aviv University, Tel Aviv, Israel (OG, MK); the Department of Ophthalmology, Keck School of Medicine, University of South California, Los Angeles, California (ESH, PL); the Department of Ophthalmology, Federal University of Rio de Janeiro, Rio de Janeiro, Brazil (MRS); the Department of Ophthalmology, Shaare Zedek Medical Center, Jerusalem, Israel (AA, DS); and University Center Hospital of Bordeaux, Anterior Segment and Refractive Surgery Unit, Bordeaux, France (DT, DS).

Drs. Smadja and Santhiago are consultants for Ziemer Ophthalmic Systems AG. Dr. Abulafia is a consultant for Hoya, Physiol, Zeiss, and Haag-Streit. The remaining authors have no financial or proprietary interest in the materials presented herein.

AUTHOR CONTRIBUTIONS

Study concept and design (OG, DS); data collection (OG, DT, MK, DS); analysis and interpretation of data (OG, ESH, PL, MRS, AA, DS); writing the manuscript (OG, ESH, PL, DS); critical revision of the manuscript (OG, MRS, AA, DT, MK, DS); statistical expertise (DS); administrative, technical, or material support (DS); supervision (DS)

Correspondence: Oren Golan, MD, Department of Ophthalmology, Tel Aviv Sourasky Medical Center, Tel Aviv, Israel. E-mail: golanoren@hotmail.com

Received: December 18, 2017
Accepted: August 22, 2018

10.3928/1081597X-20180823-02

Sign up to receive

Journal E-contents