Comparison Of 2 Scoring Systems For Detection Of Early Forms Of Keratoconus

The aim of the study is to investigate the efficacy of the new objective K SCORE  method from ANTERION (Heidelberg Engineering GmbH) in comparison to the well established Belin Ambrosio enhanced ectasia display (BAD-D) score from Pentacam AXL Wave (OCULUS Optikgerate GmbH) for the early detection of ectasia susceptible eyes

This is a prospective study in which 200 eyes of 200 patients undergoing refractive surgery were included (one eye was randomly chosen). 

After taking informed consent, all eyes underwent Pentacam AXL Wave (rotating scheimpfkug technology) and ANTERION (AS-OCT) scanning to derive the respective keratoconus score. The agreement between the parameters used by the two classification softwares were evaluated using interclass correlation coefficient (ICC) and paired simple t-test. Efficacy of the scoring system was assessed by reciever operating characteristics (ROC) curve analysis. All tests were performed using Medcalc 20.115 (MedCalc Software Ltd, Belgium.

Anterior surface keratometry (K) parameters K1, K2 and astigmatism had mean difference and confidence interval of -0.2 [-0.25, -0.16]D, -0.21 [-0.25, -0.16]D and -0.01 [-0.05, 0.03]D, respectively. Similarly, the same for the posterior surface were -0.17 [-0.18, -0.16]D, -0.14 [-0.15, -0.13]D and -0.03 [-0.03, -0.02]D, respectively. The above parameters for both surfaces were significantly different between the devices in the t-test (p<0.001). However, the difference was not clinically relevant (<0.25 D). ICC for the above parameterswas >0.92. ROC analysis of BAD-D had a 94.74% sensitivity and 99.18% specificity. ROC analysis of K SCORE had a 92.86% sensitivity and 91.79% specificity.

Overall the ANTERION and Pentacam AXL Wave parameters showed good agreement. The sensitivity and specificity of the K SCORE were comparable to the BAD-D.