EFFECTS OF ASPIRIN ON VITREOUS PRERETINAL HEMORRHAGE IN PATIENTS WITHDIABETES-MELLITUS - EARLY TREATMENT DIABETIC-RETINOPATHY STUDY REPORTNO-20

Objective: To assess whether the use of aspirin exacerbates the severi ty or duration of vitreous/preretinal hemorrhages in patients with dia betic retinopathy. Design: The Early Treatment Diabetic Retinopathy St udy (ETDRS), a multicenter randomized clinical trial, was designed to assess the effect of photocoagulation and aspirin on 3711 patients wit h mild to severe nonproliferative or early proliferative diabetic reti nopathy. Intervention: Patients were randomly assigned to either an as pirin (650 mg/d) or a placebo group. One eye of each patient was rando mly assigned to early photocoagulation and the other to deferral of ph otocoagulation. Main Outcome Measures: The severity and duration of th e vitreous/preretinal hemorrhages were determined from gradings of the annual, seven standard stereoscopic field, fundus photographs. Clinic al examinations scheduled every 4 months also provided information on the presence and duration of hemorrhages. Results: Annual fundus photo graphs of eyes assigned to deferral of photocoagulation revealed vitre ous/preretinal hemorrhages at some time during follow-up in 564 patien ts (30%) assigned to the placebo group and 585 patients (32%) assigned to the aspirin group (P=.48). Based on gradings of fundus photographs , there were no statistical differences in the severity of vitreous/pr eretinal hemorrhages (P=.11) or their rate of resolution (P=.86) betwe en the groups. Clinical examination of eyes assigned to deferral of ph otocoagulation revealed 721 eyes (39%) assigned to the aspirin group a nd 689 (37%) assigned to the placebo group that had vitreous/preretina l hemorrhages during the course of the study (P=.30). Again, no statis tically significant difference was found between the rates of resoluti on, as assessed clinically, between the two treatment groups (P=.43). Conclusions: As previously reported, the use of aspirin did not increa se the occurrence of vitreous/preretinal hemorrhages in patients enrol led in the ETDRS. The data presented in this report demonstrate that t he severity and duration of these hemorrhages were not significantly a ffected by the use of aspirin and that there were no ocular contraindi cations to its use (650 mg/d) in persons with diabetes who require it for treatment of cardiovascular disease or for other medical indicatio ns.