Fluoroquinolone Use and Risk of Pneumothorax in Adults Hospitalized With Community Acquired Pneumonia: A Retrospective Cohort Study
B. S. Steven Yi, B. S. Saurav Sumughan, M. D. Jessica Cobb, Erika J. YooABSTRACT
Objectives
The objective of this study is to examine the relationship between fluoroquinolone exposure and pneumothorax risk in patients hospitalized with pneumonia.
Methods
This retrospective cohort study used the TriNetX US Collaborative Network. We defined two mutually exclusive cohorts of hospitalized adults based on exposure to antibiotic classes commonly used to treat community acquired pneumonia: fluoroquinolone (ciprofloxacin, moxifloxacin, levofloxacin, and ofloxacin) versus nonfluoroquinolone (amoxicillin, azithromycin, ceftriaxone, and doxycycline). Patients with connective tissue disorders were excluded. The primary outcome was the occurrence of pneumothorax within 60 days. We used Cox proportional hazards regression to estimate hazard ratios (HRs) adjusted for demographics, comorbidities, and markers of illness severity.
Results
Among 1 140 106 eligible patients, 86 814 (7.6%) received fluoroquinolones. Fluoroquinolones as a class were not associated with risk of pneumothorax (HR 1.29, 0.95–1.76, Q = 0.21). Among individual fluoroquinolones, Ciprofloxacin alone was associated with an increased risk of pneumothorax (HR 2.26, 95% CI 1.38–3.69, Q < 0.01). Pooled nonfluoroquinolone antibiotics were associated with a modestly reduced hazard of pneumothorax (HR 0.79, 95% CI 0.69–0.91, Q < 0.01), although this effect was not consistent across individual agents.
Conclusions
Exposure to fluoroquinolones does not appear to impact the risk of pneumothorax in hospitalized adults with pneumonia. Only ciprofloxacin was associated with increased risk of pneumothorax. Future research in both clinical and experimental settings may help elucidate whether the mechanism by which fluoroquinolones affect the pulmonary pleura can vary by acute disease state. Our findings suggest that antibiotic selection for pneumonia should be guided by clinical appropriateness rather than attributable risk for pneumothorax.