Ultrasound biomicroscopy for examination of the sclerotomy site in eyes with proliferative diabetic retinopathy after vitrectomy

Purpose: We evaluated the capability of ultrasound biomicroscopy (UBM) to p redict fibrovascular proliferation at sclerotomy sites in eyes with postope rative vitreous hemorrhage due to proliferative diabetic retinopathy (PDR). Methods: Ultrasound biomicroscopy was used for examining the sclerotomy sit es in 13 eyes of 11 patients with PDR experiencing postoperative vitreous h emorrhage (PDR group). Thirty-nine sclerotomy sites (all entry sites of eac h eye) were examined before reoperation, and the UBM images were compared w ith findings obtained during revision of the vitrectomy. Thirteen eyes of 1 3 patients undergoing vitrectomy for nondiabetic diseases were used as cont rols and examined after vitrectomy. Results: The UBM images were classified into the following four categories: A, tent; B, spheroid; C, trapezoid; and N, none. The findings were distrib uted as follows in the PDR group: category A, 18%; B, 5%; C, 56%; and N, 21 %; and as follows in the control group: category A, 28%; B, 5%; C, 5%; and N, 62%. In the PDR group, 11 of 12 sclerotomy sites disclosing fibrovascula r proliferation possessed the trapezoidal image. Mean length of trapezoidal base was 2.49 +/- 0.97 mm and 1.51 +/- 0.75 mm in the groups with and with out fibrovascular proliferation, respectively (P < 0.01). The average relat ive reflectivity of the trapezoidal image against the sclera was 0.501 +/- 0.169 in the fibrovascular proliferation group and 0.891 +/- 0.183 in the f ibrous ingrowth group (P < 0.01), Conclusion: Ultrasound biomicroscopy is useful in detecting fibrovascular p roliferation at sclerotomy sites because a large and low-reflecting trapezo idal UBM image is highly correlated to its presence.