Assessment of corneal epithelial thickness mapping by spectral-domain optical coherence tomography

Abstract

Background: To assess corneal epithelial-thickness (ET) mapping resulting from spectral-domain-optical-coherence-tomography (SD-OCT) by analysing its repeatability and reproducibility and its utility for screening corneal-refractive-surgery (CRS) candidates.

Methods: ET was measured in 25-sectors by two-operators. Intra-subject-standard-deviation, coefficient-of-repeatability (CoR) and coefficient-of-variability (CoV) were calculated to evaluate repeatability. Reproducibility was evaluated using a Bland-Altman analysis. Scheimpflug-tomography, refraction, visual acuity, and patient history were used to make a decision on eligibility for CRS. After this decision, the surgeon was shown the patient's ET map and was asked to reconsider his analysis. The percentage of screenings that changed after evaluating the ET maps was determined.

Results: Forty-three eyes with normal corneas (CRS-group) and 21 eyes not suitable for CRS (non-CRS-group) were studied. For the CRS-group, CoR ranged from 2.03 (central) to 19.73 μm (outer-inferonasal), with the central-sector showing the highest repeatability (CoV: 1.53-1.80%). For the non-CRS-group, CoR ranged from 3.82 (central-middle-superonasal) to 13.42 μm (middle-inferotemporal), with the inner-superonasal-sector showing the highest repeatability (CoV: 2.86-4.46%). There was no statistically significant difference between operators (p > 0.01). In the CRS-group, the outcomes showed a narrow 95% limits-of-agreement (LoA) for the central-and inner-nasal-sectors (about 4 μm), and wider for the inner-superior, outer-superotemporal and outer-inferonasal (about 10-14 μm). In the non-CRS-group, they were for the outer superonasal (about 4 μm), and for the middle-inferotemporal and outer-temporal (about 10 μm), respectively. Candidacy for CRS changed in 7.82% of patients after evaluation of the ET maps, with all of them screened-out.

Conclusion: The SD-OCT provided repeatable and reproducible corneal ET measurements and may alter candidacy for CRS.

Clinical trial registration: German Clinical Trials Register: https://drks.de/search/en/trial/DRKS00032797, identifier: DRKS00032797.

Keywords: cornea; epithelial thickness; optical coherence tomography; repeatability; reproducibility; spectral domain.

Figure 1

Pachymetry (top) and epithelial thickness (bottom) maps (microns) obtained using the CIRRUS 5000 HD-OCT device showing the mean values for the 25 sectors analysed in a healthy patient for right and left eyes (S, superior; N, nasal; I, inferior; T, temporal).

Figure 2

Bland–Altman plots showing agreement in measurements for different sectors for the two operators in the refractive surgery group: central (A), inner nasal (B), inner superonasal (C), inner superior (D), inner superotemporal (E), inner temporal (F), inner inferotemporal (G), inner inferior (H), inner inferonasal (I), middle nasal (J), middle superonasal (K), middle superior (L), middle superotemporal (M), middle temporal (N), middle inferotemporal (O), middle inferior (P), middle inferonasal (Q), outer nasal (R), outer superonasal (S), outer superior (T), outer superotemporal (U), outer temporal (V), outer inferotemporal (W), outer inferior (X) and outer inferonasal (Y). The mean (continuous line), lower and upper limits of agreement [±1.96 SD (standard deviation), peripheral dotted lines], and the lower and upper confidence intervals (95%) are depicted.

Figure 3

Bland–Altman plots showing agreement in measurements for different sectors for the two operators in the non-refractive surgery group: central (A), inner nasal (B), inner superonasal (C), inner superior (D), inner superotemporal (E), inner temporal (F), inner inferotemporal (G), inner inferior (H), inner inferonasal (I), middle nasal (J), middle superonasal (K), middle superior (L), middle superotemporal (M), middle temporal (N), middle inferotemporal (O), middle inferior (P), middle inferonasal (Q), outer nasal (R), outer superonasal (S), outer superior (T), outer superotemporal (U), outer temporal (V), outer inferotemporal (W), outer inferior (X) and outer inferonasal (Y). The mean (continuous line), lower and upper limits of agreement [±1.96 SD (standard deviation), peripheral dotted lines], and the lower and upper confidence intervals (95%) are depicted.