Statin Use and Reduced Risk of Pneumonia in Patients with Melioidosis: A Lung-Specific Statin Association

Abstract

Rationale: 3-Hydroxy-3-methylglutaryl coenzyme A reductase inhibitor (statin) use is associated with a lower risk of incident pneumonia and, less robustly, with nonpulmonary infections. Whether statin use is associated with a lower risk of pneumonia than other clinical presentations of infection with the same pathogen is unknown. Objectives: To assess whether preadmission statin use is associated with a lower risk of pneumonia than nonpneumonia presentations among patients hospitalized with Burkholderia pseudomallei infection (melioidosis). Methods: We performed a secondary analysis of a prospective multicenter cohort study of patients hospitalized with culture-confirmed B. pseudomallei infection (melioidosis). We used Poisson regression with robust standard errors to test for an association between statin use and pneumonia. We then performed several sensitivity analyses that addressed healthy user effect and indication bias. Results: Of 1,372 patients with melioidosis enrolled in the parent cohort, 1,121 were analyzed. Nine hundred eighty (87%) of 1,121 were statin nonusers, and 141 (13%) of 1,121 were statin users. Forty-six (33%) of 141 statin users presented with pneumonia compared with 432 (44%) of 980 statin nonusers. Statin use was associated with a lower risk of pneumonia in unadjusted analysis (relative risk, 0.74; 95% confidence interval, 0.58-0.95; P = 0.02) and, after adjustment for demographic variables, comorbidities, environmental exposures, and symptom duration (relative risk, 0.73; 95% confidence interval, 0.57-0.94; P = 0.02). The results of sensitivity analyses, including active comparator analysis and inverse probability of treatment weighting, were consistent with the primary analysis. Conclusions: In hospitalized patients with melioidosis, preadmission statin use was associated with a lower risk of pneumonia than other clinical presentations of melioidosis, suggesting a lung-specific protective effect of statins.

Keywords: Burkholderia pseudomallei; hydroxymethylglutaryl–coenzyme A reductase inhibitors; low- and middle-income countries; pneumonia; statin.

Figure 1.

Study flow diagram. BMI = body mass index.

Figure 2.

Primary adjusted analysis and sensitivity analyses of pneumonia risk among statin users compared with nonusers. For inverse probability of treatment weighting, the average treatment effect is reported rather than a relative risk and is not shown on the figure. The average treatment effect was −0.15 (95% CI, −0.26 to −0.04; P = 0.01). CI = confidence interval. See Table E1 for tabular results.