Topical antiseptics in minimizing ocular surface bacterial load before ophthalmic surgery: a randomized controlled trial

Purpose: To investigate the reduction of the ocular surface bacterial load induced by two commercially available ophthalmic antiseptic formulations, povidone-iodine (PVI) 0.6% and chlorexidine (CLX) 0.02%, before ocular surgery. Design: Randomized controlled trial. Methods: Seventy adult patients, undergoing intraocular surgery (phacoemulsification), were randomized to receive in the index eye PVI (group A) QID for 3 days or CLX (group B)" QID for 3 days before surgery. The untreated eye was used as control. A conjunctival swab was taken in both eyes before (T0) and after (T1) therapy. Microbial DNA was quantified with real-time PCR analysis. The Mick algorithm was used to compare the abundance of each genus/genera against the distribution of abundances from the reference. At T1, patients filled a questionnaire to evaluate therapy-induced symptoms. Primary outcome was the reduction of bacterial DNA at T1 (microbial load), vs. control arm, expressed as mean number of real-time PCR cycle times (CT). Secondary outcomes were taxonomic composition, differential abundance, and therapy-induced ocular symptoms. Results: The difference T0-T1 in CT was significant in group B, but not in group A (mean [95% CI], 0.99[0.33] vs. 0.26[0.15], p<0.001, and 0.65[0.3] vs. 0.45[0.41], p=0.09, respectively). The taxonomic composition, alpha and beta diversity remained consistent at all time-points in both groups. The rate of patients reporting therapy-induced ocular symptoms and the mean discomfort grade were greater in group A than in group B (97% vs 26% and 4.97±2.48 vs 0.66±1.53, respectibely). Conclusions: Compared to PVI 0.6%, CLX 0.02% induced a greater reduction of ocular surface bacterial load, with no significant alterations of the taxonomic composition. Moreover, CLX was better tolerated than PVI.