Abstract
Purpose:
Preoperative aqueous laser flare measurements are thought to have a predictive value in the development of proliferative vitreoretinopathy (PVR) in patients undergoing surgery for a retinal detachment. In this study we tested if preoperative aqueous flare is a useful screening tool to decide whether a patient should receive a possible preventive therapy.
Methods:
Two hundred consecutive patients with a primary retinal detachment were asked to participate and gave written informed consent. Exclusion criteria were active vasculitis, active uveitis, proliferative diabetic retinopathy, diabetic macular edema, exudative age-related macular degeneration, and retinal vein occlusion. In both eyes aqueous flare was measured preoperatively using a Kowa FM-500 Laser Flare Meter (Kowa Company Ltd, Tokyo,Japan) 15 minutes after instillation of 0.5% tropicamide. From each eye an average of five measurements expressed in photoncounts per millisecond (pc/ms) was taken. Extent of retinal detachment, number of horseshoe tears, presence of curled edges during surgery, type of surgery and medication history were recorded. All patients were evaluated at least six months after surgery to determine if they underwent additional surgery due to PVR redetachment.<br /> Median flare values were compared using the Mann-Whitney U test. Sensitivity and specificity at different cut-off values were determined with a ROC analysis.
Results:
The median preoperative flare value of patients who developed PVR (mdn = 18.4 pc/ms, n=10) and patients that did not (mdn = 9.35 pc/ms, n=92) differed significantly, p =0.007. The area under the ROC curve was 0.76 (95% CI 0.62; 0.89). A cut-off value of 10pc/ms leads to acceptable sensitivity (80%) of this measurement. However, it yields a very low specificity (50%). If the cut-off value is raised to 15pc/ms (specificity 80%) the sensitivity decreases to only 60%.
Conclusions:
Although median flare values seem to differ between groups, preoperative flare has low sensitivity and specificity as predictor for PVR in patients with a primary rhegmatogenous retinal detachment. Therefore it has limited clinical usefulness as a screening tool for preventive therapy.