Anterior Segment Optical Coherence Tomography Indices and Their Value in Diagnosing Corneal Ectasia

Purpose: To determine the diagnostic value of the anterior segment optical coherence tomography (AS-OCT) indices in differentiating normal from ectatic corneas. Material and Methods: Two groups of patients—with corneal ectasia and normal controls were compared. Each group consists of 80 eyes of 43 age and sex-matched patients. All of them underwent corneal topography with OCULUS Keratograph 5M and corneal pachymetry with AS-OCT with RTVue-100. The indices generated by the AS-OCT pachymetric scans were analyzed. Results: There was a statistically significant difference for all the examined indices between the two groups with p values <0.001 and a confidence interval of 95%. The minimal corneal thickness (Min) was the best performing index according to the ROC analysis with an area under the curve of 0.976 and a combination of sensitivity and specificity of 0.925 and 0.911 respectively, and a “cut-off” value of 484 microns, followed by the indices of focal thinning—Min-Med and Min-Max with an area under the curve of 0.973 and 0.971 and sensitivity/specificity of 0.938/0.962 and 0.938/0.937 respectively. The rest of the examined parameters had an area under the curve in the range between 0.950 for the central corneal thickness and 0.814 for the outer superior segment. Conclusion: The anterior segment OCT indices showed excellent capability in differentiating ectatic from normal corneas.


Introduction
The keratoconus [1] [2] and the marginal pellucid degeneration [3] are the two How to cite this paper: Atanassov The diagnostics of the corneal ectasias is based on several clinical features-patient's refractive error, the presence or absence of Vogt striae, Fleisher ring, Munson sign as well as changes in the topographical [4] [5] and the biomechanical properties of the cornea [6] [7] [8] [9].
The main features of keratoconus according to the participants in the latest consensus [10] on the topic are posterior cornea ectasia, abnormal corneal thickness profile and noninflammatory thinning. In contrast to the corneal topography, which provides information about the changes in the anterior surface of the cornea, the corneal tomography and the anterior segment optical coherence tomography are excellent tools for observation of the changes in the corneal properties.
The purpose of the study is to determine the diagnostic capability of the anterior segment OCT indices in patients with keratoconus and pellucid marginal degeneration.

Patients and Methods
The current study includes two groups of patients-healthy controls and patients with corneal ectasia, that were age and sex-matched. Each group consists of 80 eyes of 43 caucasian individuals, 23 male and 20 female. Each participant underwent a standard ophthalmological examination followed by corneal topography with OCULUS Keratograph 5M (Oculus, Germany) and anterior-segment ОСТ with OptovueRTVue-100 (OptoVue, USA). Patients wearing contacts were instructed to discontinue them for at least 3 days before the examination.
Inclusion criteria for the ectatic group are: • corneal astigmatism greater than 3 diopters.
• history of keratitis.
We took into consideration the following parameters generated by the anterior-segment OCT ( Figure 1): Mean corneal thickness in 17 sectors, Min (minimal corneal thickness in the 6 mm zone), Min-Max index (the maximum corneal thickness subtracted from the minimal corneal thickness), Min-Median index (the mean corneal thickness subtracted from the minimal corneal thickness), SN-IT index (the mean corneal thickness of the interior temporal sector subtracted from the mean corneal thickness of the superior nasal sector), S-I index (the mean corneal thickness of the inferior corneal sector subtracted from the mean corneal thickness of the superior sector).
The statistical analysis was performed using SPSS version 15 and the following tests: Mann-Whitney U test, Independent-Samples T-test and ROC (Receiver

Results
The average age in the control group was 34,4 years and 34,6 years in the group with ectasia (Table 1).
After determining the distribution of the data, we used the independent-sample T-test to compare the following indices: central corneal thickness (CCT), superior nasal corneal thickness, inferior nasal corneal thickness, temporal corneal thickness, superior temporal corneal thickness, outer superior nasal corneal thickness, outer nasal corneal thickness, outer inferior nasal corneal thickness, outer inferior temporal corneal thickness, outer temporal corneal thickness, SN-IT and S-N ( Table 2).
For the rest of the parameters (superior corneal thickness, nasal corneal thickness, inferior corneal thickness, inferior temporal corneal thickness, outer superior nasal corneal thickness, outer superior temporal corneal thickness, minimal corneal thickness (Min), Min-Max, Min-Med) we used the Mann-Whitney U test (Table 3). The null hypothesis was rejected for all of the examined indices On the other hand, thickness asymmetry indices SN-IT and S-I had a lot worse performance (Figure 2).   Table 4.
The area under the curve for the remainder of the indices is in the range from 0.914 to 0.814.

Discussion
In the current study, several indices show excellent diagnostic capabilities. The central corneal thickness (ССТ) and the minimum corneal thickness (Min) are the two best performing parameters with an area under the curve of 0.950 and 0.976. Before the introduction of the Shleimflug devices and their new indices-ART (Ambrosio relational thickness), PPI (pachymetric progression index), PTI (percentage thickness increase) and others, the CCT [22] [23] [24] and the Min [22] [23] [24] [25] [26] were the most used commonly used pachymetric indices.
There is plenty of publications on the topic, but most of them are based on a tomographic [22] [23] [24] [25] examination and only a few are with ultrasound [27] or OCT devices [26] [27] [28]. Some of the authors haven't taken into consideration the central corneal thickness and for those that have, it is performing worse than MIN [22] [23] [24]. There is a good reason for that-for most of the cases, the zone with the thinning is situated slightly inferior and temporal [27]  and not centrally, so the Min and the CCT value don't match.
Most of the authors also have lower AUC for both CCT and Min ( Table 5).
The mean corneal thickness in several sectors also shows promising results. The area under the curve for the following zones: outer inferior, inferior temporal, outer inferior, and temporal, is in the range between 0.956 and 0.945.
Unfortunately, there are only a few publications with the OptoVueRTVue 100 device, and none of them examined the mean corneal thickness. As already stated, the thinnest zone in patients with ectasia is displaced inferiorly and temporally (in keratoconus) or inferiorly (in PMD), so it makes sense for those sectors to be with the best diagnostic value.
In the current study, the indices of focal thinning (Min-med and Min-max) outperform the thickness asymmetry ones (SN-IT and S-N). The area under the curve for Min-Med is 0.973 and for Min-Maх 0.971. In comparison, SN-IT and S-I have a lower area under the curve of 0.861 and 0.843 respectively. Most of the publications on the topic also confirm this finding.
Li et al. [26] were one of the first to use anterior segment OCT in the diagnostics of corneal ectasia. With the Zeiss OCT device, they found AUC for Min-Med of 0.929. The I-S and IT-SN indices, which are analogous to the ones we used, also have worse performance when compared to Min-Med with values of 0.868 and 0.894 respectively.
With the RTvue OCT, Qin et al. [28] have reported an AUC of 0.925 for Min-Med (they haven't tested Min-Max). The results for SN-IT and S-I are like the ones we have reported-0.896 and 0.861.

Conclusion
The anterior segment OCT and its pachymetric indices and maps offer important information about the corneal thickness profile in patients with corneal ectasias. The focal thinning indices alongside the mean corneal thickness in the inferior and the temporal parts of the cornea can be used to distinguish normal from ectatic corneas.

Conflicts of Interest
The authors declare no conflicts of interest regarding the publication of this paper.