The effect of povidone iodine flush versus drops on conjunctival colonization before intravitreal injections

To determine the most effective method of applying povidone iodine 5% to decrease conjunctival colonization before intravitreal injections.

Twenty-eight patients from two tertiary care centers undergoing intravitreal injection for diffuse diabetic macular edema, exudative age-related macular degeneration, venous occlusive disease, or refractory pseudophakic cystoid macular edema were prospectively randomized to two study arms. One arm received 2-3 drops of 5% povidone iodine (drops group) and the second received a 10 ml flush of the same solution (flush group). The inferior conjunctival fornix was cultured before and after antiseptic technique was performed in all patients. Three culture media, thioglycollate broth, chocolate agar and blood agar, were used for each sample.

Each study group had 14 patients. Prior to antisepsis, 22 of the 28 (78.6%) subjects had positive conjunctival cultures. 16 and 14 bacterial organisms were isolated in the first and second groups, respectively. After using 2-3 drops of 5% povidone iodine in the first study arm of patients, three of 16 (18.7% reduction) bacterial organisms were no longer isolated in thioglycollate broth media. With flush irrigation of 10 ml of 5% povidone iodine, seven of 14 (50% reduction) bacterial organisms were no longer isolated (P-value 0.07) in broth media. No difference in reduction of bacterial colonization was found on plated media (chocolate agar and blood agar).

Irrigating the conjunctival fornix with 5% povidone iodine results in greater reduction of bacterial colonization compared with drop application of the same solution. Flush irrigation may provide better protection against the risk of endophthalmitis with intravitreal injections.