The aim of our study was to compare the accuracy of the optical biometry (OB) andconventional ultrasound biometry (UB) in determination of the axial length prior topars plana vitrectomy (PPV), and in silicone oil filled eyes.In Our study, thirty patients undergoing PPV for retinal detachment wereprospectively evaluated. Of the 30 eyes, 16 had PDR (proliferative diabetic retinopathywith tractional rhegmatogenous retinal detachment), 12 had recurrent RDwith PVR and 2 had endophthalmitis. All patients underwent PPV with silicone oilinjection as a primary procedure, followed by pahcoemulsification and silicone oilremoval at least 3 months later. Axial length was determined by optical biometryand conventional ultrasound biometry prior to PPV, After PPV (in the presence ofsilicone oil), and after phacoemulsifaction and silicone oil removal.Our study results showed that prior to PPV the mean axial length was 23.93±2.31mm by UB, and 21.83±2.06 mm by OB (mean difference = 2.1±1.62 mm, p <0.0001). After PPV, and in the presence of silicone oil, the mean axial length was25.21±2.58 mm by UB, and 24.69±2.42 mm by OB (mean difference = 0.52±0.69mm, p = 0.0003). After phacoemulsifaction and silicone oil removal, the mean axiallength was 24.36±2.41 mm by UB, and 24.51±2.37 mm by OB (mean difference = -0.15±0.30 mm, p = 0.01).We can conclude that prior to vitrectomy for retinal detachment; axial lengthdetermination by B guided A-scan is more accurate, as a presence of retinaldetachment causes underestimation of axial length by optical means. In thepresence of silicone oil, both types of biometry yield comparable measurements.