Abstract
Purpose :
There is a lack of agreement on the definition of progression of Keratoconus (KCN) after treatment. Here we compare two preoperative definitions of progression after corneal cross-linking (CXL) for KCN and compare the percentage of eyes identified as progressed.
Methods :
This was a retrospective case series analysis of patients affected by KCN that underwent CXL. Maximum keratometry (Kmax), thinnest corneal thickness (ThCT), corrected distance visual acuity (CDVA) and ABCD progression criteria were recorded before the treatment and at the last follow-up. Progression of keratoconus after CXL treatment was defined according to either the widely adopted standard criteria for progression (i.e., Kmax increase > 1 Diopter (D) or ThCT reduction > 20 mm) or the ABCD progression criteria. Furthermore, we simulated the percentage of eyes incorrectly defined as progressed according to different cut-offs (1, 1.25 and 1.5 D) of the difference between two Kmax readings. To assess whether repeatability might improve the classification, standard deviation values of the difference of single (0.81 D) or mean-of-three (0.46 D) Kmax readings were derived from the literature and used for comparison.
Results :
Thirty-seven KCN eyes of 21 patients underwent CXL. Mean (SD) age was 27.4 ± 7 and mean (SD) follow-up was 519 ± 128 days; 13 patients had a history of atopy. Kmax showed a significant reduction from baseline values after CXL (55.2 ± 7.8 D vs 54.3 ± 7.5 D, p < 0.01), while no difference in ThCT (465.5 ± 40 mm vs 463.5 ± 40.8 mm, p = 0.12) and in CDVA (0.22 ± 0.26 vs 0.18 ± 0.30 LogMAR, p = 0.1) was found.
Applying standard criteria for progression 18.9% (7/37) of eyes were classified as progressed and 43.2% (16/37) according to the ABCD method. In a simulation of 20,000 keratoconic eyes with no progression, wrong classification occurred in 2170, 1228 and 641 eyes (1, 1.25 and 1.5 D cut-offs, respectively) based on the difference of single Kmax readings between baseline and last follow-up, dropping to 297, 66 and 12 when based on the difference of mean-of-three Kmax readings (Figure 1).
Conclusions :
The definition of disease stability after CXL in KCN varies greatly according to the definition of progression used. Furthermore, low repeatability of measurements increases the number of eyes incorrectly identified as progressed. We encourage to use the mean of three repeated measurements to provide better definition of corneal changes after CXL.
This abstract was presented at the 2023 ARVO Annual Meeting, held in New Orleans, LA, April 23-27, 2023.