Identifying Clinical Predictors of Proliferative Sickle Cell Retinopathy

Abstract

Purpose To identify systemic and/or ophthalmologic predictors of proliferative sickle retinopathy. Methods Cross-sectional study comparing clinical, laboratory, and structural choriorretinal aspects between sickle cell disease patients with and without proliferative retinopathy. Patients underwent complete systemic and ophthalmologic evaluation. Enhanced depth spectral domain optical coherence tomography with choroidal binarization and optic coherence tomography angiography were performed and choriorretinal vascular components were compared. Results Forty-five eyes from 45 sickle cell patients were included. Ninety-one percent of patients were diagnosed with sickle cell retinopathy, 29% with proliferative retinopathy. Mean corpuscular volume, lactate dehydrogenase, and percentage of fetal hemoglobin were reduced in the subgroup of patients with proliferative retinopathy when compared with patients without proliferative retinopathy (p ≤ 0.001; p = 0.04; p ≤ 0.001, respectively). The best predictor of proliferative retinopathy was mean corpuscular volume (AUC = 0.842; p = 0.001), followed by the percentage of fetal hemoglobin (AUC = 0.763, p = 0.009) and lactate dehydrogenase (AUC curve = 0.706; p = 0.039). No differences were found between groups in the quantitative analysis of retinal vascularization using OCTA and choroidal vascularization using OCT (p ≥ 0.05). Conclusion Fetal hemoglobin and mean corpuscular volume may be good predictors of proliferative sickle retinopathy. The association between proliferative retinopathy and reduced levels of lactate dehydrogenase and mean corpuscular volume points to hypoxia and not hemolysis as a possible driving force in its pathophysiology.