Abstract
Abstract: :
Purpose: The usefulness of fluorescein pattern analysis for fitting rigid contact lenses to keratoconic corneas is controversial. The purpose of this study is to assess the accuracy and repeatability of contact lens fluorescein pattern interpretation in keratoconus. Methods: Four clinicians evaluated the fluorescein patterns of three keratoconic patients with varying disease severity. For each subject the contact lens base curve that just cleared the corneal apex, the FDACL lens, was determined. At two separate test sessions, the masked clinicians were given six contact lenses, in random order, from the fitting set, including the FDACL lens, three lenses flatter (in 0.1 mm increments) and three lenses steeper (in 0.1 mm increments) than the FDACL lens. They evaluated the fluorescein patterns at the slit lamp biomicroscope and ranked the lenses from flattest to steepest. Accuracy was evaluated by comparing the clinician's ranking to the true rank. Repeatability was determined by comparing the rankings on the two test sessions. Kappa statistics were used as a quantitative measure of agreement. Results: Accuracy of fluorescein pattern analysis was not significantly different between clinicians (p=0.9591), and repetition was not related to accuracy (p=0.45). Data from all clinicians were then combined to calculate weighted kappa statistics: 0.612 (overall), 0.202-0.892 (range for subjects). As with accuracy, repeatability of the ranking was not significantly related to clinician (p=0.60). Data from all clinicians were combined to calculate weighted kappa statistics: 0.481 (overall), 0.007-0.743 (range for subjects). With only one subject in each disease severity group, it is not possible to determine statistically if accuracy or repeatability are related to severity, although this initial data would suggest there are differences among subjects. Conclusion: No difference was found in the accuracy and repeatability of fluorescein pattern analysis among the four clinicians. Agreement between the correct rating and clinician rating of the cornea-to-contact lens base curve relationship was 0.612 overall. Agreement between sessions was 0.481 overall. Variation in individual subject kappa statistics suggests that characteristics of the cornea (e.g., position of the corneal apex) might influence fluorescein pattern interpretation.
Keywords: 450 keratoconus • 367 contact lens • 356 clinical (human) or epidemiologic studies: systems/equipment/techniques