RESUMEN
PURPOSE: To evaluate the efficacy and safety of early pars plana vitrectomy (PPV) for the treatment of acute infective endophthalmitis, and identify prognostic factors for better visual outcome. DESIGN: Retrospective cohort study. METHODS: Consecutive patients who underwent early PPV within 72 hours of presentation for the treatment of acute infective bacterial endophthalmitis and presented to a large tertiary referral center in New South Wales, Australia, between January 2009 and December 2013 were included. Changes in best-corrected visual acuity (VA) from baseline to 1 year were examined. RESULTS: A total of 64 patients were included. The inciting events were cataract surgery (53%), intravitreal injection (36%), trabeculectomy (3%), and endogenous (3%). The mean VA improved from 3.1 logMAR (hand motion) at baseline to 1.02 (approximately 20/200) at 1 year, with 42% achieving final VA equal to or better than 0.477 logMAR (20/60) following early PPV. Positive prognostic factors were negative microbial cultures (P < 0.01) and etiology of post-cataract surgery (P < 0.01). In multivariable analyses adjusting for age and prognostic factors, patients with baseline VA of light perception and hand motion achieved greater visual gains than those with counting fingers, with gains of logMAR of -2.68, -2.09, and -0.85, respectively (P < 0.0001). CONCLUSIONS: Most patients who undergo early PPV experience substantial VA improvement. Negative microbial cultures and endophthalmitis after cataract surgery were associated with better final visual outcome. Patients with presenting VA of light perception or hand motion achieved higher visual gains than those with counting fingers, suggesting the possibility that early PPV may be beneficial in both groups.