Verena C Mulder1, Jan Tode2, Elon Hc van Dijk1,3, Konstantine Purtskhvanidze2, Johann Roider2, Jan C van Meurs1,4, Felix Treumer2. 1. The Rotterdam Eye Hospital/Rotterdam Ophthalmic Institute, Rotterdam, The Netherlands. 2. Department of Ophthalmology, University of Kiel, University Medical Center Schleswig-Holstein, Kiel, Germany. 3. Department of Ophthalmology, Leiden University Medical Center, Leiden, The Netherlands. 4. Department of Ophthalmology, Erasmus University, Rotterdam, The Netherlands.
Abstract
BACKGROUND/AIMS: Patients with rhegmatogenous retinal detachment (RRD) who develop postoperative proliferative vitreoretinopathy (PVR) have been found to have higher preoperative laser flare values than patients with RRD who do not develop this complication. Measurement of laser flare has therefore been proposed as an objective, rapid and non-invasive method for identifying high-risk patients. The purpose of our study was to validate the use of preoperative flare values as a predictor of PVR risk in two additional patient cohorts, and to confirm the sensitivity and specificity of this method for identifying high-risk patients. METHODS: We combined data from two independent prospective studies: centre 1 (120 patients) and centre 2 (194 patients). Preoperative aqueous humour flare was measured with a Kowa FM-500 Laser Flare Meter. PVR was defined as redetachment due to the formation of traction membranes that required reoperation within 6 months of initial surgery. Logistic regression and receiver operating characteristic analysis determined whether higher preoperative flare values were associated with an increased risk of postoperative PVR. RESULTS: PVR redetachment developed in 21/314 patients (6.7%). Median flare values differed significantly between centres, therefore analyses were done separately. Logistic regression showed a small but statistically significant increase in odds with increasing flare only for centre 2 (OR 1.014; p=0.005). Areas under the receiver operating characteristic showed low sensitivity and specificity: centre 1, 0.634 (95% CI 0.440 to 0.829) and centre 2, 0.731 (95% CI 0.598 to 0.865). CONCLUSIONS: Preoperative laser flare measurements are inaccurate in discriminating between those patients with RRD at high and low risk of developing PVR. Published by the BMJ Publishing Group Limited. For permission to use (where not already granted under a licence) please go to http://www.bmj.com/company/products-services/rights-and-licensing/.
BACKGROUND/AIMS: Patients with rhegmatogenous retinal detachment (RRD) who develop postoperative proliferative vitreoretinopathy (PVR) have been found to have higher preoperative laser flare values than patients with RRD who do not develop this complication. Measurement of laser flare has therefore been proposed as an objective, rapid and non-invasive method for identifying high-risk patients. The purpose of our study was to validate the use of preoperative flare values as a predictor of PVR risk in two additional patient cohorts, and to confirm the sensitivity and specificity of this method for identifying high-risk patients. METHODS: We combined data from two independent prospective studies: centre 1 (120 patients) and centre 2 (194 patients). Preoperative aqueous humour flare was measured with a Kowa FM-500 Laser Flare Meter. PVR was defined as redetachment due to the formation of traction membranes that required reoperation within 6 months of initial surgery. Logistic regression and receiver operating characteristic analysis determined whether higher preoperative flare values were associated with an increased risk of postoperative PVR. RESULTS: PVR redetachment developed in 21/314 patients (6.7%). Median flare values differed significantly between centres, therefore analyses were done separately. Logistic regression showed a small but statistically significant increase in odds with increasing flare only for centre 2 (OR 1.014; p=0.005). Areas under the receiver operating characteristic showed low sensitivity and specificity: centre 1, 0.634 (95% CI 0.440 to 0.829) and centre 2, 0.731 (95% CI 0.598 to 0.865). CONCLUSIONS: Preoperative laser flare measurements are inaccurate in discriminating between those patients with RRD at high and low risk of developing PVR. Published by the BMJ Publishing Group Limited. For permission to use (where not already granted under a licence) please go to http://www.bmj.com/company/products-services/rights-and-licensing/.