Journal of Refractive Surgery

Biomechanics Supplemental Data

Detection of Subclinical Corneal Ectasia Using Corneal Tomographic and Biomechanical Assessments in a Japanese Population

Shizuka Koh, MD, PhD; Renato Ambrósio Jr, MD, PhD; Ryota Inoue, MS; Naoyuki Maeda, MD, PhD; Atsuya Miki, MD, PhD; Kohji Nishida, MD, PhD

Abstract

PURPOSE:

To test the detection of subclinical corneal ectasia using integrated Scheimpflug tomography and biomechanical assessment in a Japanese population.

METHODS:

This prospective, case–control study included 23 patients with very asymmetric ectasia (VAE) and 70 normal controls. Patients with VAE had defined clinical ectasia in one eye and a fellow eye with normal topography (VAE-NT). Objective topography for confirming normal topography in VAE-NT cases included having 0% similarity and 0% severity derived from Placido-disk based topography. Scheimpflug-based corneal tomography and corneal biomechanical assessment were performed. The Belin/Ambrósio Enhanced Ectasia Deviation index (BAD-D), Corvis Biomechanical Index (CBI), and Tomographic Biomechanical Index (TBI) were compared and their discriminating ability for detecting ectasia was assessed.

RESULTS:

For differentiating normal and VAE-NT eyes, the areas under the receiver operating curve for the BAD-D, CBI, and TBI were 0.668, 0.660, and 0.751, respectively. The TBI cut-off of 0.259 provided 52.17% sensitivity and 88.57% specificity. Fourteen VAE-NT cases (60.9%) were abnormal in at least one of the criteria of the BAD-D > 1.60 (39.1%), CBI > 0.5 (26.1%), or TBI > 0.29 (43.5%). Conversely, nine VAE-NT cases (39.1%) exhibited normal values for the BAD-D, CBI, and TBI.

CONCLUSIONS:

In the current study, 40% of VAE-NT eyes were classified as normal by the BAD-D, CBI, and TBI. Although some of these cases may truly represent unilateral ectasia, further advances are needed to enhance ectasia detection and characterize the susceptibility for ectasia progression.

[J Refract Surg. 2019;35(6):383–390.]

Abstract

PURPOSE:

To test the detection of subclinical corneal ectasia using integrated Scheimpflug tomography and biomechanical assessment in a Japanese population.

METHODS:

This prospective, case–control study included 23 patients with very asymmetric ectasia (VAE) and 70 normal controls. Patients with VAE had defined clinical ectasia in one eye and a fellow eye with normal topography (VAE-NT). Objective topography for confirming normal topography in VAE-NT cases included having 0% similarity and 0% severity derived from Placido-disk based topography. Scheimpflug-based corneal tomography and corneal biomechanical assessment were performed. The Belin/Ambrósio Enhanced Ectasia Deviation index (BAD-D), Corvis Biomechanical Index (CBI), and Tomographic Biomechanical Index (TBI) were compared and their discriminating ability for detecting ectasia was assessed.

RESULTS:

For differentiating normal and VAE-NT eyes, the areas under the receiver operating curve for the BAD-D, CBI, and TBI were 0.668, 0.660, and 0.751, respectively. The TBI cut-off of 0.259 provided 52.17% sensitivity and 88.57% specificity. Fourteen VAE-NT cases (60.9%) were abnormal in at least one of the criteria of the BAD-D > 1.60 (39.1%), CBI > 0.5 (26.1%), or TBI > 0.29 (43.5%). Conversely, nine VAE-NT cases (39.1%) exhibited normal values for the BAD-D, CBI, and TBI.

CONCLUSIONS:

In the current study, 40% of VAE-NT eyes were classified as normal by the BAD-D, CBI, and TBI. Although some of these cases may truly represent unilateral ectasia, further advances are needed to enhance ectasia detection and characterize the susceptibility for ectasia progression.

[J Refract Surg. 2019;35(6):383–390.]

Corneal ectatic disorders are characterized by thinning of the central, paracentral, or peripheral cornea. Keratoconus is a major corneal ectatic disorder that manifests as progressive corneal thinning and protrusion.1,2 Visual impairment caused by keratoconus and ectatic corneal diseases is due to irregular corneal astigmatism. Diagnosis of keratoconus and ectatic corneal diseases in the early stages is especially important in preoperative screening for laser refractive surgery to avoid postoperative keratectasia.3,4 Topographic and tomographic analyses have been used in our daily clinical practice to detect alterations in the corneal shape, such as thinning or increased curvature. A conventional Placido ring-based corneal topographer, previously referred to as a video-keratoscope, analyzes contours of the anterior corneal surface.5 Tomographic techniques, such as rotating Scheimpflug imaging, allow the assessment of both anterior and posterior corneal surface configurations and corneal thickness. With recent advances in corneal biomechanical assessment, increased attention has been focused on the biomechanical properties of the cornea, particularly in the context of detection of keratoconus and ectatic corneal disease.

The Corvis ST (OCULUS Optikgeräte GmbH, Wetzlar, Germany) monitors corneal dynamic deformation using a constant-pressure air pulse through an ultra-high–speed Scheimpflug camera. Recently, the Corvis Biomechanical Index (CBI)—based on the corneal thickness profile and deformation parameters—has been developed and was shown to demonstrate sensitivity for the discrimination of normal eyes and keratoconic eyes and the detection of subclinical keratoconus.6,7 Furthermore, the recently introduced Tomographic Biomechanical Index (TBI) combines Scheimpflug-based corneal tomography and biomechanics for enhancing ectasia detection.8 Recent studies have demonstrated that the TBI was useful in detecting clinical ectasia and subclinical ectasia in eyes with normal topography in patients with VAE.8–13

The purpose of this study was to investigate the diagnostic ability of combined parameters, based on Scheimpflug-based corneal tomography and biomechanical assessment, to detect subclinical forms of ectasia in a Japanese population. Specifically, we present a case series of patients with subclinical forms of ectasia who demonstrated neither clinical nor topographic signs of ectasia in one eye, but were diagnosed as having keratoconus or a related corneal ectatic thinning disorder in the contralateral eye.

Patients and Methods

This prospective, case–control study was reviewed and approved by the institutional review board of Osaka University Hospital. All participants provided informed consent after the nature and risks/benefits of study participation were explained. All study protocols adhered to the tenets of the Declaration of Helsinki.

Study Patients

The patients with VAE, eyes with normal topography (VAE-NT), and corresponding ectatic eyes examined from June 2017 to May 2018 at Osaka University Hospital were enrolled. Criteria for normality and keratoconus diagnosis were based on a comprehensive ophthalmic examination including Placido-disk corneal topography (TMS-5; Tomey; Aichi, Japan). The diagnosis of keratoconus was based on the presence of at least one slit-lamp finding of corneal stromal thinning, corneal protrusion, Fleischer ring, and Vogt striae. Patients were considered to have VAE when the diagnosis of ectasia was confirmed in one eye and the contralateral eye had a distance-corrected visual acuity of 20/20 or better and a normal topographic map without topographic signs according to both keratoconus screening programs (ie, Klyce/Maeda Keratoconus Index14,15 and Smolek/Klyce Keratoconus Severity Index16). Additionally, the criterion of an inferior-superior asymmetry value (I-S value,17 quantifies the inferior versus superior corneal dioptric asymmetry at 6 mm [3-mm radii]) of less than 1.4 in the topographic map was applied for the eyes with VAE-NT. In accordance with previous research,8 the Belin/Ambrósio Enhanced Ectasia Display was not considered at the diagnosis. Patients with no ocular disorders except refractive errors served as normal controls. From the normal controls, one eye was included in the study.

Corneal Imaging Measurements and Parameters

Scheimpflug-based corneal tomography was performed using the Pentacam HR (OCULUS Optikgeräte) and corneal biomechanical assessment was performed with the Corvis ST (OCULUS Optikgeräte). The principles and procedures involved in the use of these devices have been described elsewhere.8,18 Both measurements were performed by experienced examiners under the same lighting conditions. All patients were examined at least twice to obtain well-focused, properly aligned images of the eye. Belin/Ambrósio Enhanced Ectasia Deviation (BAD-D) values from the Pentacam HR,19 CBI values from the Corvis ST, and TBI values, as a combined parameter, from the Pentacam HR and the Corvis ST were obtained for each eye.

Statistical Analyses

Statistical analysis was performed using Med-Calc Statistical Software (version 16.8.4; MedCalc; Ostend, Belgium; https://www.medcalc.org) and Statistical Package for Social Sciences (SPSS) software (version 25; SPSS, Inc., Chicago, IL). Because the combined parameters were programmed to have their output values range from 0 to 1, BAD-D values were converted to the Belin/Ambrósio Enhanced Ectasia Deviation normalized Index (BAD-DI) to facilitate comparisons.8 Comparisons of age and clinical parameters between normal and VAE-NT groups were performed by the Wilcoxon rank-sum test. The accuracies for discriminating ability in BAD-D, CBI, and TBI were assessed based on receiver operating characteristic (ROC) curves. For each parameter, the area under the ROC curve (AUROC) was calculated and the best cut-off value that yielded the highest accuracy was determined along with the sensitivity and specificity. To compare the performance of diagnostic tests, pairwise comparisons of AUROC among three parameters were performed according to the previously reported approach.20 A P value of less than .05 was considered statistically significant.

Results

Twenty-three patients with VAE and 70 control subjects were enrolled in this study. The demographic and clinical data (topographic, tomographic, and biomechanical parameters) of both groups are summarized in Table 1. All studied parameters were significantly different between VAE-NT eyes and normal eyes (P < .05), except for minimum central corneal thickness. The distributions of BAD-D, BAD-DI, CBI, and TBI for both groups are shown as dot plot graphs (Figure 1).

Demographic and Clinical Characteristics

Table 1:

Demographic and Clinical Characteristics

The box and dot plots showing distribution of the BAD-D (top left), BAD-DI (top right), CBI (bottom left), and TBI (bottom right) for the normal group and the subclinical ectasia group (fellow eye from patients with very asymmetric ectasia with normal topography). BAD-D = Belin/Ambrósio Enhanced Ectasia Deviation; BAD-DI = Belin/Ambrósio Enhanced Ectasia Deviation normalized index; CBI = Corvis Biomechanical Index; TBI = Tomographic Biomechanical Index; VAE-NT = fellow eye from patients with very asymmetric ectasia with normal topography

Figure 1.

The box and dot plots showing distribution of the BAD-D (top left), BAD-DI (top right), CBI (bottom left), and TBI (bottom right) for the normal group and the subclinical ectasia group (fellow eye from patients with very asymmetric ectasia with normal topography). BAD-D = Belin/Ambrósio Enhanced Ectasia Deviation; BAD-DI = Belin/Ambrósio Enhanced Ectasia Deviation normalized index; CBI = Corvis Biomechanical Index; TBI = Tomographic Biomechanical Index; VAE-NT = fellow eye from patients with very asymmetric ectasia with normal topography

ROC curves were established to distinguish between the normal and VAE-NT groups (Figure 2). Table 2 demonstrates the optimum cut-off values with the highest sensitivity and specificity for BAD-D, CBI, and TBI. For comparison between normal eyes and VAENT eyes, the AUROC for BAD-D, CBI, and TBI values were 0.668, 0.660, and 0.751, respectively. There were no significant differences among the three parameters (P > .05). The 95% confidence interval (CI) for each parameter was remarkably large. The TBI cut-off of 0.259 provided the highest predictive accuracy with 52.17% sensitivity and 88.57% specificity.

Receiver operating characteristic curves indicating the ability of each index to distinguish between the normal group and the subclinical ectasia group (VAE-NT: fellow eye from patients with very asymmetric ectasia with normal topography). BAD-D = Belin/Ambrósio Enhanced Ectasia Deviation; CBI = Corvis Biomechanical Index; TBI = Tomographic Biomechanical Index

Figure 2.

Receiver operating characteristic curves indicating the ability of each index to distinguish between the normal group and the subclinical ectasia group (VAE-NT: fellow eye from patients with very asymmetric ectasia with normal topography). BAD-D = Belin/Ambrósio Enhanced Ectasia Deviation; CBI = Corvis Biomechanical Index; TBI = Tomographic Biomechanical Index

ROC Analysis With AUROC, Sensitivity, and Specificity Between Normal Eyes and VAE-NT Eyes

Table 2:

ROC Analysis With AUROC, Sensitivity, and Specificity Between Normal Eyes and VAE-NT Eyes

Topography maps were obtained using Placido-disk corneal topography. Pentacam/Corvis ST ARV (Ambrósio, Roberts, and Vinciguerra) TBI display was obtained for the 23 VAE-NT cases. Figure 3 shows 12 of the 23 cases (data for all 23 cases are available from the authors). In all eyes of the VAE-NT group, Placido-disk corneal topography demonstrated a Keratoconus Index of 0% (absence of keratoconus),14,15 a Keratoconus Severity Index of 0%, 18 and an I-S value lower than 1.40 D. Table A (available in the online version of this article) summarizes topographic, tomographic, and biomechanical parameters in 23 patients with VAE. The questionnaire results on the presence of known keratoconus-related factors such as ocular allergy, atopy, and eye rubbing are also shown. Based on currently published criteria,6,8,21 among the 23 eyes with VAE-NT, 9 eyes had a BAD-D score higher than 1.60 (39.1%), 6 had a CBI score higher than 0.5 (26.1%), and 10 eyes had a TBI score higher than 0.29 (43.5%). Seven eyes, which had abnormal values on either the BAD-D or CBI, showed abnormal TBI values. Three eyes (patients 5, 13, and 21) showed abnormal TBI values, but had normal values on both the BAD-D and CBI. In total, 14 of 23 VAE-NT cases (60.9%) were diagnosed as abnormal based on the BAD-D, CBI, and TBI values. In contrast, 9 VAE-NT cases (patients 1, 3, 4, 8, 11, 12, 14, 15, and 16) had normal values on the BADD, CBI, and TBI. Twelve of the 23 patients with VAE (52.2%) had an ocular allergy and 11 of the 23 patients (47.8%) had atopy. Sixteen of the 23 patients (69.6%) reported current or previous habitual eye rubbing.

Topography maps were obtained by using Placido-disk corneal topography. Pentacam/Corvis ST ARV (OCULUS Optikgeräte, Wetzlar, Germany) (Ambrósio, Roberts, and Vinciguerra) TBI display was obtained for the 23 normal control eyes of the very asymmetric ectasia group. Topography map with keratoconus screening indices (left) and Pentacam/Corvis ST ARV TBI display (right) from each patient are shown for 12 of the 23 cases.

Figure 3.

Topography maps were obtained by using Placido-disk corneal topography. Pentacam/Corvis ST ARV (OCULUS Optikgeräte, Wetzlar, Germany) (Ambrósio, Roberts, and Vinciguerra) TBI display was obtained for the 23 normal control eyes of the very asymmetric ectasia group. Topography map with keratoconus screening indices (left) and Pentacam/Corvis ST ARV TBI display (right) from each patient are shown for 12 of the 23 cases.

Topographic, Tomographic, and Biomehanical Parameters in VAE Cases

Table A:

Topographic, Tomographic, and Biomehanical Parameters in VAE Cases

Discussion

The current study demonstrated the detection capability of subclinical corneal ectasia by the current integrated Scheimpflug tomography and biomechanical assessments in a Japanese population. Consistent with previous studies,8,9,11–13 our results demonstrated that the TBI had high diagnostic capability equivalent to that of the BAD-D and CBI. However, the accuracy of the current approach for the detection of subclinical VAE-NT cases was less than optimal. Although the TBI presented the highest discriminating capability of VAE-NT in this study, it was lower than those of other studies.8,10–13 Because racial differences in the dimension of anterior eye segments have been reported,22 validation of these parameters in different racial populations is important.

Comparison of the AUROC and cut-off values of the TBI with previous studies8–13 are shown in Table B (available in the online version of this article). To date, six studies have reported the discriminating capability of VAE-NT from normal eyes. Regarding the AUROC, our results demonstrated bulky curves with large 95% CIs, which made it difficult to determine “optimal” cut-off value. Although the small sample size of VAENT in this study should be taken into consideration, we speculate that the reasons for a lower detection capability of the TBI for the VAE-NT group in this study would be partly attributed to racial distribution. In addition, the criterion of “VAE-NT” was different among the studies. Notably, we only enrolled the patients with VAE-NT who had a distance-corrected visual acuity of 20/20 or better and a normal Placido-disk topographic map (Klyce/Maeda Keratoconus Index14,15 and Smolek/Klyce Keratoconus Severity Index16), based on the definition of forme fruste keratoconus.23,24 Moreover, there was no description about the criteria values of the Keratoconus Index and Keratoconus Severity Index for Placido-disk corneal topography in previous studies.8,10–13 In the current study, 13 of 23 eyes with VAE-NT (56.5%) showed normal TBI values (less than 0.29). Nine (patients 1, 3, 4, 8, 11, 12, 14, 15, and 16) of these 13 cases had completely normal values in all parameters, including the Keratoconus Index for Placido-disk corneal topography, BAD-D, CBI, and TBI. These cases can be considered as the group of fellow eyes from patients with VAE-NT and tomography cases, based on Steinberg et al.13 The ratio of the patients with VAE-NT with completely normal BAD-D, CBI, and TBI values in our population (39.1%) might be higher than in other studies,8,10–13 resulting in the large 95% CIs in the AUROC results. According to the statement by the “Global Consensus on Keratoconus and Ectatic Corneal Diseases,”25 true unilateral keratoconus does not exist. However, considering the asymmetry apparent in these 9 patients, the TBI may indicate the existence of true unilateral ectasia, allowing the separation of true unilateral ectasia from true fruste cases. This result also suggests that forme fruste keratoconus may need to be redefined.

Comparison of AUROC and Cut-off Values of the TBI in VAE-NT Cases With Previous Studies

Table B:

Comparison of AUROC and Cut-off Values of the TBI in VAE-NT Cases With Previous Studies

In the presented case series of 23 normal eyes of the VAE-NT group, 10 eyes (43.5%) showed abnormal TBI values. Among these 10 patients, 7 eyes had abnormal values on either the BAD-D or CBI. Particularly, 3 eyes (patients 5, 13, and 21) were diagnosed as normal by objective indices derived from topographic, tomographic, and biomechanical assessment alone; however, only the TBI showed abnormal values. In contrast, 2 eyes (patients 17 and 18) showed abnormal values only on the BAD-D and the other 2 eyes (patients 2 and 9) showed abnormal values only on the CBI while having normal values in other parameters. These results indicate the need for careful assessment of subclinical ectasia using different techniques/devices in combination. With increased use of tomographic and biomechanical assessment in corneal ectatic disease, it is of interest to clarify the potential capability of the TBI in detecting subclinical ectasia and to interpret TBI data in the context of pathophysiological changes of corneal ectasia.

To date, few studies have described the long-term observation of unilateral keratoconus using corneal topography or corneal tomography.26–29 According to Li et al.,27 using clinical evaluation and videokeratography, 30 of 85 (35.3%) clinically normal fellow eyes developed keratoconus. Among these 30 patients, 83% developed keratoconus within the first 6 years after the initial keratoconus diagnosis. In comparison, Imbornoni et al.28 presented 5 cases of unilateral keratoconus in which changes indicative of clinical keratoconus did not develop in the fellow eye, based on tomographic evaluation over a period of 5 years. Long-term follow-up observations of eyes with VAE-NT are essential.29 It is of great interest to observe whether 9 VAE cases with normal values in all parameters (patients 1, 3, 4, 8, 11, 12, 14, 15, and 16) remain stable over an extended period despite high susceptibility to ectasia progression. A longitudinal TBI assessment study addressing this issue is underway in our institute.

Corneal ectasia occurs due to mechanical forces, such as eye rubbing or atopy, and the inherent ectasia susceptibility of the patient. In total, 70% of the 23 patients reported current or previous habitual eye rubbing and all except 2 of these patients had an ocular allergy or atopy. Unilateral eye rubbing is considered one of the environmental factors contributing to uni-lateral clinical presentation in keratoconus.25 In this study, unilateral eye rubbing was not identified in the face-to-face questionnaire as a factor contributing to unilateral keratoconus, because it seemed to be difficult for the patients themselves to recognize a habit of unilateral eye rubbing. High prevalence of eye rubbing observed in patients with VAE highlights the importance of education and treatment of patients with corneal ectasia, such as through verbal guidance regarding not rubbing one's eyes and use of topical anti-allergic medication in patients with allergies.

Limitations of the current study include the small sample size of the VAE group. Further investigations regarding the sensitivity and specificity of the TBI for distinguishing VAE-NT from normal eyes with a larger number of VAE-NT cases are warranted. As noted in the previous study,8 the criteria for inclusion in the normal group may be considered. In fact, some cases in the normal group showed higher values on the BAD-D, CBI, and TBI, suggesting the relatively rare but potential susceptibility to corneal ectasia. Recently, epithelial thickness mapping with the use of anterior segment optical coherence tomography has been useful in detecting subclinical keratoconus.30 The potential of other variables such as those related to demographics, epithelial characteristics, or a more sophisticated representation of corneal surface characteristics by Zernike polynomials should be considered in artificial intelligence algorithms. The current study revealed the necessity of further technical advances and data pooling to obtain a robust metric that can be applied to a broad range of eyes. A recent study reported an enhanced tomographic assessment to detect corneal ectasia based on artificial intelligence methods.31 Accuracy in the detection of subclinical corneal ectasia is still a challenge, and there are still subthreshold differences that cannot be detected with current technology. Longitudinal and transversal studies to analyze data from “completely normal” eyes would be helpful in confirming true unilateral ectasia.

The integration of corneal tomographic and biomechanical data, such as that observed in the TBI, enhances the ability to detect abnormalities in most eyes with normal topography in VAE cases and in rare normal cases in a Japanese population. Although keratoconus is defined as a bilateral disease, some VAE-NT cases may represent true unilateral ectasia, possibly related to external environmental factors, such as eye rubbing or atopy. Larger population studies, including longitudinal evaluations, are warranted in the future.

References

  1. Krachmer JH, Feder RS, Belin MW. Keratoconus and related noninflammatory corneal thinning disorders. Surv Ophthalmol. 1984;28:293–322. doi:10.1016/0039-6257(84)90094-8 [CrossRef]
  2. Rabinowitz YS. Keratoconus. Surv Ophthalmol. 1998;42:297–319. doi:10.1016/S0039-6257(97)00119-7 [CrossRef]
  3. Binder PS, Lindstrom RL, Stulting RD, et al. Keratoconus and corneal ectasia after LASIK. J Refract Surg. 2005;21:749–752.
  4. Ambrósio R Jr, Randleman JB. Screening for ectasia risk: what are we screening for and how should we screen for it?J Refract Surg. 2013;29:230–232. doi:10.3928/1081597X-20130318-01 [CrossRef]
  5. Klyce SD. Computer-assisted corneal topography: high-resolution graphic presentation and analysis of keratoscopy. Invest Ophthalmol Vis Sci. 1984;25:1426–1435.
  6. Vinciguerra R, Ambrósio R Jr, Elsheikh A, et al. Detection of keratoconus with a new biomechanical index. J Refract Surg. 2016;32:803–810. doi:10.3928/1081597X-20160629-01 [CrossRef]
  7. Vinciguerra R, Ambrósio R Jr, Roberts CJ, Azzolini C, Vinciguerra P. Biomechanical characterization of subclinical keratoconus without topographic or tomographic abnormalities. J Refract Surg. 2017;33:399–407. doi:10.3928/1081597X-20170213-01 [CrossRef]
  8. Ambrósio R Jr, Lopes BT, Faria-Correia F, et al. Integration of Scheimpflug-based corneal tomography and biomechanical assessments for enhancing ectasia detection. J Refract Surg. 2017;33:434–443. doi:10.3928/1081597X-20170426-02 [CrossRef]
  9. Sedaghat MR, Momeni-Moghaddam H, Ambrósio R Jr, et al. Diagnostic ability of corneal shape and biomechanical parameters for detecting frank keratoconus. Cornea. 2018;37:1025–1034. doi:10.1097/ICO.0000000000001639 [CrossRef]
  10. Chan TCY, Wang YM, Yu M, Jhanji V. Comparison of corneal tomography and a new combined tomographic biomechanical index in subclinical keratoconus. J Refract Surg. 2018;34:616–621. doi:10.3928/1081597X-20180705-02 [CrossRef]
  11. Ferreira-Mendes J, Lopes BT, Faria-Correia F, Salomão MQ, Rodrigues-Barros S, Ambrósio R Jr, . Enhanced ectasia detection using corneal tomography and biomechanics. Am J Ophthalmol. 2019;197:7–16. doi:10.1016/j.ajo.2018.08.054 [CrossRef]
  12. Kataria P, Padmanabhan P, Gopalakrishnan A, Padmanaban V, Mahadik S, Ambrósio R Jr, . Accuracy of Scheimpflug-derived corneal biomechanical and tomographic indices for detecting subclinical and mild keratectasia in a South Asian population. J Cataract Refract Surg. 2019;45:328–336. doi:10.1016/j.jcrs.2018.10.030 [CrossRef]
  13. Steinberg J, Siebert M, Katz T, et al. Tomographic and biomechanical Scheimpflug imaging for keratoconus characterization: a validation of current indices. J Refract Surg. 2018;34:840–847. doi:10.3928/1081597X-20181012-01 [CrossRef]
  14. Maeda N, Klyce SD, Smolek MK, Thompson HW. Automated keratoconus screening with corneal topography analysis. Invest Ophthalmol Vis Sci. 1994;35:2749–2757.
  15. Maeda N, Klyce SD, Smolek MK. Comparison of methods for detecting keratoconus using videokeratography. Arch Ophthalmol. 1995;113:870–874. doi:10.1001/archopht.1995.01100070044023 [CrossRef]
  16. Smolek MK, Klyce SD. Current keratoconus detection methods compared with a neural network approach. Invest Ophthalmol Vis Sci. 1997;38:2290–2299.
  17. 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. Erratum in: J Cataract Refract Surg. 2000;26:480. doi:10.1016/S0886-3350(99)00195-9 [CrossRef]
  18. Chan TC, Wang YM, Yu M, Jhanji V. Comparison of corneal dynamic parameters and tomographic measurements using Scheimpflug imaging in keratoconus. Br J Ophthalmol. 2018;102:42–47. doi:10.1136/bjophthalmol-2017-310355 [CrossRef]
  19. Ambrósio R Jr, Dawson DG, Salomão M, Guerra FP, Caiado AL, Belin MW. Corneal ectasia after LASIK despite low preoperative risk: tomographic and biomechanical findings in the unoperated, stable, fellow eye. J Refract Surg. 2010;26:906–911. doi:10.3928/1081597X-20100428-02 [CrossRef]
  20. DeLong ER, DeLong DM, Clarke-Pearson DL. Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics. 1988;44:837–845. doi:10.2307/2531595 [CrossRef]
  21. Ambrosio R Jr, Belin MW. Enhanced screening for ectasia risk prior to laser vision correction. Int J Kerat Ect Cor Dis. 2017;6:23–33.
  22. Qin B, Tang M, Li Y, Zhang X, Chu R, Huang D. Anterior segment dimensions in Asian and Caucasian eyes measured by optical coherence tomography. Ophthalmic Surg Lasers Imaging. 2012;43:135–142. doi:10.3928/15428877-20120102-03 [CrossRef]
  23. Amsler M. The “forme fruste” of keratoconus [article in German]. Wien Klin Wochenschr. 1961;73:842–843.
  24. Klyce SD. Chasing the suspect: keratoconus. Br J Ophthalmol. 2009;93:845–847. doi:10.1136/bjo.2008.147371 [CrossRef]
  25. Gomes JAP, Tan D, Rapuano CJ, et al. Global consensus on keratoconus and ectatic diseases. Cornea. 2015;34:359–369. doi:10.1097/ICO.0000000000000408 [CrossRef]
  26. Holland DR, Maeda N, Hannush SB, et al. Unilateral keratoconus: incidence and quantitative topographic analysis. Ophthalmology. 1997;104:1409–1413. doi:10.1016/S0161-6420(97)30123-7 [CrossRef]
  27. 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]
  28. Imbornoni LM, Padmanabhan P, Belin MW, Deepa M. Long-term tomographic evaluation of unilateral keratoconus. Cornea. 2017;36:1316–1324. doi:10.1097/ICO.0000000000001295 [CrossRef]
  29. Koh S, Inoue R, Maeda N, et al. Long-term chronological changes in very asymmetric keratoconus. Cornea. 2019;38:605–611. doi:10.1097/ICO.0000000000001890 [CrossRef]
  30. Li Y, Chamberlain W, Tan O, Brass R, Weiss JL, Huang D. Sub-clinical 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]
  31. Lopes BT, Ramos IC, Salomão MQ, et al. Enhanced tomographic assessment to detect corneal ectasia based on artificial intelligence. Am J Ophthalmol. 2018;195:223–232. doi:10.1016/j.ajo.2018.08.005 [CrossRef]

Demographic and Clinical Characteristics

ParameterNormal (n = 70)VAE-NT (n = 23)P
Age (y).034
  Mean ± SD42.8 ± 13.847.2 ± 10.2
  Median (range)40.0 (22.0 to 80.0)46.0 (24.0 to 66.0)
Min CCT (mm).063
  Mean ± SD550.46 ± 33.33538.55 ± 31.12
  Median (range)550.68 (474.12 to 645.06)538.74 (493.23 to 633.51).063
Kmax (D).398
  Mean ± SD44.40 ± 1.4044.59 ± 1.38
  Median (range)44.10 (41.29 to 48.46)44.57 (41.70 to 46.58)
I-S value.005
  Mean ± SD0.18 ± 0.530.54 ± 0.50
  Median (range)0.21 (−1.09 to 1.44)0.53 (−0.41 to 1.21)
KISA (%).003
  Mean ± SD8.21 ± 10.2417.81 ±18.97
  Median (range)4.02 (0.33 to 37.42)9.47 (0.39 to 71.20)
BAD-D.016
  Mean ± SD0.96 ± 0.561.48 ± 1.14
  Median (range)1.00 (−0.52 to 2.12)1.47 (−0.11 to 5.65)
BAD-DI.002
  Mean ± SD0.04 ± 0.130.19 ± 0.34
  Median (range)0.00 (0.00 to 0.61)0.03 (0.00 to 1.00)
CBI.021
  Mean ± SD0.07 ± 0.120.27 ± 0.37
  Median (range)0.01 (0.00 to 0.58)0.04 (0.00 to 1.00)
TBI< .001
  Mean ± SD0.12 ± 0.160.33 ± 0.30
  Median (range)0.05 (0.00 to 0.87)0.26 (0.00 to 1.00)

ROC Analysis With AUROC, Sensitivity, and Specificity Between Normal Eyes and VAE-NT Eyes

ParameterBAD-DCBITBI
AUROC0.670.660.75
  95% CI0.56 to 0.760.56 to 0.760.65 to 0.84
  Cut-off> 1.430> 0.515> 0.259
Sensitivity (%)60.8730.4352.17
  95% CI (%)38.5 to 80.313.2 to 52.930.6 to 73.2
Specificity (%)85.7098.5788.57
  95% CI (%)75.3 to 92.992.3 to 10078.7 to 94.9

Topographic, Tomographic, and Biomehanical Parameters in VAE Cases

PtAge (y)SexVAE-NTVAE-EEye RubbingOcular AllergyAtopy


EyeBAD-DCBITBIEyeBAD-DCBITBI
144ML0.2800.06R17.100.010.95
240MR1.440.790.23L27.7911
340MR0.7000L31.5610.93++
442ML0.4300R11.4210.98++
545MR1.030.200.31L10.9511+++
624MR2.190.980.49L6.6711+++
746MR1.990.010.35L24.86NANA
866ML0.9600.12R40.55NANA
959ML1.510.750.17R12.68NANA++
1052ML2.770.781R8.6511++
1150MR−0.060.100.06L11.3811++
1264ML0.730.020.07Roperated+++
1340FR1.1800.38L12.6310.97+
1446MR1.470.020.14Loperated++
1556ML1.50.170.1Roperated+++
1653FR0.600L7.9611+++
1736MR1.6400.12L6.5811+
1859FL1.620.040.26R7.330.981++
1935FR1.780.570.38L6.7711++
2059MR2.490.010.86L22.5411+
2140ML1.440.220.49R6.9111
2248FL1.610.020.38R1311+++
2341FR5.6511L6.40.991++

Comparison of AUROC and Cut-off Values of the TBI in VAE-NT Cases With Previous Studies

StudyComparisonAUROCCut-offSensitivity (%)Specificity (%)
Ambrósio et al.8480 normal vs 94 VAE-NT0.985> 0.2990.496.0
Chan et al.1037 normal vs 23 VAE-NT0.925> 0.1684.482.4
Ferreira-Mendes et al.11312 normal vs 57 VAE-NT0.960> 0.29589.591.0
Kataria et al.12100 normal vs 100 VAE-NT0.901> 0.1684.086.0
Steinberg et al.13105 normal vs 32 VAE-NT0.825> 0.1172.071.0
Current study70 normal vs 23 VAE-NT0.750> 0.25952.1788.57
Authors

From the Department of Innovative Visual Science, Osaka University Graduate School of Medicine, Osaka, Japan (SK, RI); the Department of Ophthalmology, Osaka University Graduate School of Medicine, Osaka, Japan (NM, AM, KN); Instituto de Olhos Renato Ambrósio and Visare Personal Laser, and the Department of Ophthalmology, Federal University of the State of Rio de Janeiro (UNIRIO), Rio de Janeiro, Brazil (RA); and SEED Co., Ltd., Tokyo, Japan (RI).

Drs. Koh and Maeda received lecture fees from, and Dr. Ambrósio is a consultant for OCULUS Optikgeräte GmbH. Mr. Inoue is an employee of SEED Co., LTD. The remaining authors have no financial or proprietary interest in the materials presented herein.

The authors thank Dr. Sven Reisdorf and Ms. Stefanie Berger (OCULUS Optikgeräte GmbH) for helpful discussions and calculations.

AUTHOR CONTRIBUTIONS

Study concept and design (SK, RA, NM); data collection (SK, RI, NM, AM); analysis and interpretation of data (SK, RA, RI, NM, AM, KN); writing the manuscript (SK, RA); critical revision of the manuscript (RI, NM, AM, KN); statistical expertise (SK, RA); administrative, technical, or material support (KN); supervision (KN)

Correspondence: Shizuka Koh, MD, PhD, Department of Innovative Visual Science, Osaka University Graduate School of Medicine, Room E7, 2-2 Yamadaoka, Suita Osaka 565-0871, Japan. E-mail: cizciz@gmail.com

Received: August 31, 2018
Accepted: April 15, 2019

10.3928/1081597X-20190417-01

Sign up to receive

Journal E-contents