Abstract Background The SHEA/IDSA/APIC Strategies to Prevent MRSA Transmission and Infection Practice Recommendations advised that facilities consider decolonizing patients on hemodialysis. We implemented a nasal decolonization intervention in which patients self-administered povidone-iodine (PVI) at each dialysis session. We aimed to assess intervention safety and effectiveness. Methods We performed a stepped wedge cluster randomized trial at 16 outpatient hemodialysis units affiliated with 5 academic medical centers between 2020-2023. Adverse events were self-reported at 1 and 6 months. While the analysis was at the hemodialysis unit level, patients were required to give verbal informed consent for PVI use. Outcomes included National Healthcare Safety Network reportable dialysis events aggregated at hemodialysis unit level, including bloodstream infections (BSI), access-related BSI, and central venous catheter (CVC) BSI for all pathogens and for Staphylococcus aureus (SA). The primary outcome was SA BSI. A generalized linear mixed model with a negative binomial distribution, log link function, and an offset for person-months with a random intercept for each hemodialysis unit was performed. Results Overall, 1,351 patients received hemodialysis at these centers each month. Of those, 362 patients verbally consented to use PVI. Among these, 3.9% reported side effects: nasal drip, congestion or burning/stinging, unpleasant smell, headache, or minor nose bleed. A reported side effect ‘yellow tears’ was assessed via chart review and resolved by discontinuing PVI. There were no statistically significant associations between unit level randomization to the PVI intervention and infections. However, there was a non-statistically significant trend toward a protective association between unit-level randomization to PVI and SA infections, particularly SA CVC BSI.(Table) Conclusion Long-term nasal decolonization with PVI was safe with few adverse events. Unit level randomization to the PVI intervention did not significantly decrease unit-level infections. Given low patient enrollment and added infection prevention interventions due to COVID-19, the study could not determine if PVI decolonization could decrease BSI rates in the hemodialysis setting. Disclosures Marin Schweizer, PhD, 3M: Grant/Research Support Anitha Vijayan, MD, Baxter: Honoraria|NxStage: Advisor/Consultant|Qanta: Honoraria David A. Pegues, MD, DaVita/Total Renal Care: Advisor/Consultant Daniel Diekema, MD, Affinity Biosensors: Grant/Research Support|bioMerieux, Inc: Grant/Research Support Loreen Herwaldt, MD, 3M: Grant/Research Support|PDI: Grant/Research Support
Read full abstract