ABSTRACT
BACKGROUND: To perform a microbiological contamination analysis of the vitreous during office-based micro-incision vitrectomy surgery (MIVS) assessing whether the bacteria detected correlated with patient's ocular conjunctival flora. METHODS: This is a prospective, interventional, nonrandomized case series of patients undergoing office-based MIVS, anti-VEGF, and dexamethasone intravitreal injections (triple therapy) for the treatment of wet age-related macular degeneration (AMD) and diabetic macular edema (DME).All patients were operated at a small procedure room in an ambulatory clinic of the Department of Ophthalmology, University of Montreal, Quebec, Canada. Conjunctival samples were done before placing the sclerotomies. The MIVS was done with a 23-gauge retractable vitrector, a 27-gauge infusion line, and a 29-gauge chandelier. Undiluted and diluted vitreous were collected for aerobic, anaerobic and fungal cultures. Outcomes measured were bacterial species identification within samples collected from the conjunctiva and the vitreous. RESULTS: Thirty-seven patients (37 eyes) were recruited and completed over 17 months of follow-up. Twenty-eight had wet AMD and nine had DME. There were 13 men and 24 women, with a mean age of 78 years. Eighteen patients (46%) had culture positive conjunctival flora. Twenty-six bacterial colonies were tabulated in total from the conjunctival swabs. All bacteria detected were gram-positive bacteria (100%), most commonly: Staphylococcus epidermitis in 11 (42%) and Corynebacterium sp. in 6 (23%). Only 1/18 patients had more than 3 species isolated, 6/18 patients had 2 species and 11/18 patients had 1 species identified on the conjunctival swab. Only 1 of the 37 undiluted midvitreous samples was culture positive, equating to a contamination rate of 2.7%. None of the diluted vitreous samples were culture positive. All cultures were negative for fungus. No serious postoperative complications occurred, including bacterial endophthalmitis, choroidal detachment, and retinal detachment. CONCLUSION: This preliminary study of office-based MIVS gives us insights on the ocular surface microbial profile and vitreous contamination rate of performing such procedures outside the OR-controlled environment. Our initial results seem to indicate that there is little risk of bacterial translocation and contamination from the conjunctiva into the vitreous. Therefore, if endophthalmitis occurs post-operatively, the source may likely arise after the procedure. Larger studies are needed to confirm our data.