Hugo van Cleynenbreugel1, Lies Remeijer1, Toine Hillenaar2. 1. Cornea and External Disease Service, The Rotterdam Eye Hospital, Rotterdam, The Netherlands. 2. Cornea and External Disease Service, The Rotterdam Eye Hospital, Rotterdam, The Netherlands; Rotterdam Ophthalmic Institute, Rotterdam, The Netherlands. Electronic address: t.hillenaar@oogziekenhuis.nl.
Abstract
PURPOSE: To ascertain preoperative and intraoperative factors that predict the need for endothelial keratoplasty (EK) in patients with Fuchs' endothelial corneal dystrophy (FECD) undergoing cataract surgery. DESIGN: Prospective, observational cohort study. PARTICIPANTS: Eighty-nine patients (89 eyes) with FECD who require cataract surgery. METHODS: One month before cataract surgery, we assessed best-corrected visual acuity, contrast sensitivity, straylight, keratometry, ultrasonic pachymetry, intraocular pressure, 7 corneal features of FECD and cataract density at slit-lamp examination, and corneal backscatter using in vivo confocal microscopy (IVCM; Confoscan 4, NIDEK Technologies, Padova, Italy). After surgery, measurements were repeated at 1, 2, and 12 months. We used stepwise binary logistic regression analysis to evaluate 30 preoperative and 5 intraoperative parameters for their ability to predict the postoperative need for EK. Receiver operating characteristic (ROC) curves of the predictive factors were used to identify their optimal cutoff points. MAIN OUTCOME MEASURES: Central corneal thickness (CCT) and backscatter at the basal epithelial cell layer (EV). RESULTS: After cataract surgery, 35 (39%) of 89 eyes underwent EK to restore vision. Of all preoperative and intraoperative parameters, only CCT and EV were identified as significant factors, predictive of the need for EK. The area under the ROC curve of EV was significantly higher than that of CCT (P = 0.003), whereas a combination of both factors in a linear discriminant function did not improve the predictive value (P = 0.66). As optimal cutoff points, we chose 1894 scatter units for EV and 630 μm for CCT. Both cutoff points correspond with a specificity of 94% and represent sensitivity of 63% for EV and 40% for CCT. CONCLUSIONS: Backscatter at the basal epithelial cell layer measured by IVCM predicts the need for EK after cataract surgery in patients with FECD. As an indicator for the corneal hydration state, the EV improves patient selection for combined cataract surgery and EK. In deciding whether to perform a triple procedure, CCT remains a less effective, but adequate, alternative. Regardless of the predictive factor used, a tailor-made approach is recommended accounting for individuals' expectations.
PURPOSE: To ascertain preoperative and intraoperative factors that predict the need for endothelial keratoplasty (EK) in patients with Fuchs' endothelial corneal dystrophy (FECD) undergoing cataract surgery. DESIGN: Prospective, observational cohort study. PARTICIPANTS: Eighty-nine patients (89 eyes) with FECD who require cataract surgery. METHODS: One month before cataract surgery, we assessed best-corrected visual acuity, contrast sensitivity, straylight, keratometry, ultrasonic pachymetry, intraocular pressure, 7 corneal features of FECD and cataract density at slit-lamp examination, and corneal backscatter using in vivo confocal microscopy (IVCM; Confoscan 4, NIDEK Technologies, Padova, Italy). After surgery, measurements were repeated at 1, 2, and 12 months. We used stepwise binary logistic regression analysis to evaluate 30 preoperative and 5 intraoperative parameters for their ability to predict the postoperative need for EK. Receiver operating characteristic (ROC) curves of the predictive factors were used to identify their optimal cutoff points. MAIN OUTCOME MEASURES: Central corneal thickness (CCT) and backscatter at the basal epithelial cell layer (EV). RESULTS: After cataract surgery, 35 (39%) of 89 eyes underwent EK to restore vision. Of all preoperative and intraoperative parameters, only CCT and EV were identified as significant factors, predictive of the need for EK. The area under the ROC curve of EV was significantly higher than that of CCT (P = 0.003), whereas a combination of both factors in a linear discriminant function did not improve the predictive value (P = 0.66). As optimal cutoff points, we chose 1894 scatter units for EV and 630 μm for CCT. Both cutoff points correspond with a specificity of 94% and represent sensitivity of 63% for EV and 40% for CCT. CONCLUSIONS: Backscatter at the basal epithelial cell layer measured by IVCM predicts the need for EK after cataract surgery in patients with FECD. As an indicator for the corneal hydration state, the EV improves patient selection for combined cataract surgery and EK. In deciding whether to perform a triple procedure, CCT remains a less effective, but adequate, alternative. Regardless of the predictive factor used, a tailor-made approach is recommended accounting for individuals' expectations.