Lorna Pairman, Yong Tai Beh, Hannah Maher, Sharon J Gardiner, Paul Chin, Jonathan Williman, Stephen T Chambers

Legionnaires' disease (LD) is typically treated with macrolides, including the azalide azithromycin, or quinolones. In 2013, guidelines for empirical treatment of community-acquired pneumonia at Christchurch Hospital, New Zealand were changed to prioritize oral azithromycin over IV clarithromycin. To determine whether the change in antimicrobial guidelines led to altered outcomes for patients subsequently confirmed to have LD. Patients with confirmed LD between 2010 and 2020 were identified from clinical and laboratory data. Hospital records were used to identify mortality, ICU admission, length of hospital stay, time to clinical stability, and time to first anti-Legionella treatment. Mean differences, risk ratios (RRs) and an interrupted time series with propensity adjustment were used to compare patient outcomes before and after the guideline change. There were 323 patients included: 128 before and 195 after the change. Patient outcomes generally improved after the change including: mortality within 30 days (RR 0.4, 95% CI 0.2-0.8); ICU admission (RR 0.6, 95% CI 0.5-0.9); length of stay (difference -2.3 days, 95% CI -4.3 to -0.4); and time to clinical stability (difference -2.4 days, 95% CI -4.3 to -0.5). The interrupted time series analysis suggested improvements in patient outcomes may have occurred regardless of the guideline change. Outcomes for patients with LD were not worsened by the change in antimicrobial guidelines and may have improved. Overall rates of mortality were low. This result was reassuring given the harm that may result from unnecessary use of IV compared with oral antimicrobial agents.