Abstract
Abstract: :
Purpose:Systematic review of the incidence of endophthalmitis after intravitreal injection. Methods:English language reports published from 1972–2002 that evaluated intravitreal injections of various substances as a sole modality of therapy in human eyes were analyzed for the following: indication for intravitreal injection, substance injected, preinjection preparation technique, number of injections, duration of follow–up, patient AIDS status and the frequency & type of endophthalmitis. Inclusion criteria included: (a) human primary studies of intravitreal injection; (b) >10 intravitreal injections performed in the study, and (c) a mean followup of –>2 months. Exclusion criteria included reports where: (a) type and frequency of complications were not mentioned, (b) other ocular procedures or surgeries (e.g., vitrectomy) were performed concurrently with intravitreal injection, and (c) intravitreal injection in eyes with previously diagnosed endophthalmitis. Results:10,563 intravitreal injections in 1,326 eyes from 42 separate published reports were analyzed. The overall reported incidence of endophthalmitis as a direct complication of intravitreal injection was 0.17% (18 cases). Causative organisms of endophthalmitis were identified in 0.07% (8 cases) and included S. epidermidis (3), coagulase–negative staphylococci (2), S. aureus (1), S. pneumoniae (1), and S. viridans (1). Intravitreal injections in eyes of reported AIDS patients were not significantly more likely to have endophthalmitis as a complication of intravitreal injection than those eyes in patients without reported AIDS (0.17% vs. 0.15%, p = 0.889). Intravitreal injections reportedly performed in the operating room were not significantly less likely to develop endophthalmitis as a complication than injections reportedly performed in the outpatient clinic (0.11% [4/3720] vs 0.17% [5/2965], p = 0.498). The use of topical betadine prior to injection resulted in a lower incidence of endophthalmitis compared to not using betadine (0.14% [9/6314] vs. 0.69% [6/869], p = 0.0009). Conclusions:This comprehensive systematic review confirms that endophthalmitis is a serious, but infrequent complication of intravitreal injection. The analyses suggest that patients may not be more likely to develop endophthalmitis after intravitreal injection if they have AIDS or if their injection is performed outside of operating room facilities. However, the use of topical betadine prior to injection may substantially decrease the risk of endophthalmitis after intravitreal injection.
Keywords: endophthalmitis • injection • vitreous