Previous studies suggest that smoking is independently associated with decreased mortality in patients with pneumonia. We hypothesized that this is a result of acquiring differential pneumococcal serotypes (ie, smokers with pneumococcal pneumonia are more likely to experience bacteremia, with low case fatality rate (CFR) serotypes). We tested this hypothesis in a population-based cohort of patients with bacteremic pneumococcal pneumonia (BPP).
Our prospective population-based clinical registry included 1,636 adults (≥ 18 years) with BPP who were hospitalized between 2000 and 2010 in northern Alberta, Canada. Using multivariable logistic regression, we determined the adjusted risk of all-cause in-hospital mortality according to smoking status (current vs not current) and conducted stratified analyses by serotypes (low CFR vs all other CFRs) according to smoking status.
The average patient age was 54 years, 57% were men, 49% were current smokers, and 41% had low-CFR serotypes. Overall, 62 of 809 current smokers died in the hospital vs 164 of 827 nonsmokers (8% vs 20%; adjusted OR, 0.52; 95% CI, 0.36-0.77; P = .001). Current smokers were more likely to have low-CFR-serotype isolates than were nonsmokers (53% vs 29%; adjusted OR, 1.67; 95% CI, 1.31-2.12; P < .001) and in models adjusted for low-CFR serotype, smoking remained independently associated with reduced mortality (P = .001).
Compared with nonsmokers, current smokers with BPP had a decreased risk of in-hospital mortality and were more likely to experience bacteremia with low CFR serotypes. These findings, at least in part, may explain why previous studies showed that smoking was associated with lower mortality in patients with pneumonia.