OBJECTIVE: To evaluate changes in epidemiological characteristics and outcome of patients with pneumococcal bacteremia during a 15-year period in a Swiss university hospital. PATIENTS AND METHODS: We reviewed the medical records of all hospitalized adults at the University Hospital Basel, Basel, Switzerland, whose blood culture yielded Streptococcus pneumoniae from January 1, 1986, through December 31, 2000. RESULTS: We analyzed 405 episodes of pneumococcal bacteremia in 394 patients. The mean annual incidence of 1.78 episodes per 1000 hospital admissions was inversely related to the mean atmospheric temperature of the area. During the study period, penicillin nonsusceptibility increased from 0% to 17%. The overall case-fatality rate was 25%, which decreased from 33% to 17% between the first and the second half of the study period (P<.001). The proportion of women with pneumococcal bacteremia increased from 37% to 52%. Independent risk factors for fatal outcome were coronary artery disease (P<.001; relative risk [RR], 4.3; 95% confidence interval [CI], 3.4-5.1), neutropenia (P=.001; RR, 3.2; 95% CI, 1.9-4.8), and age 65 years or older (P=.001; RR, 2.9; 95% CI, 1.8-4.2), whereas prior respiratory tract infection (P=.03; RR, 0.3; 95% CI, 0.1-0.5) and the occurrence of pneumococcal bacteremia in the second half of the study period (P=.01; RR, 0.4; 95% CI, 0.2-0.6) were independent predictors of survival. The case-fatality rate in human immunodeficiency virus (HIV)-infected patients was significantly lower than in patients not infected with HIV or in those with unknown HIV status (9% vs 27%; P=.006), which correlated with the younger mean +/- SD age of HIV-infected patients (33.2+/-6.6 years) compared with patients not infected with HIV (63.1+/-18.1 years) (P<.001). CONCLUSIONS: The case-fatality rate of patients with pneumococcal bacteremia decreased significantly between the first and second half of the study period, despite the increased prevalence of penicillin-nonsusceptible isolates. Independent risk factors for fatal outcome were coronary artery disease, neutropenia, and age 65 years or older, whereas prior respiratory tract infection and the occurrence of pneumococcal bacteremia in the second half of the study period were independent predictors of survival. HIV infection was a predisposing factor for pneumococcal bacteremia but was not a risk factor for fatal outcome.
OBJECTIVE: To evaluate changes in epidemiological characteristics and outcome of patients with pneumococcal bacteremia during a 15-year period in a Swiss university hospital. PATIENTS AND METHODS: We reviewed the medical records of all hospitalized adults at the University Hospital Basel, Basel, Switzerland, whose blood culture yielded Streptococcus pneumoniae from January 1, 1986, through December 31, 2000. RESULTS: We analyzed 405 episodes of pneumococcal bacteremia in 394 patients. The mean annual incidence of 1.78 episodes per 1000 hospital admissions was inversely related to the mean atmospheric temperature of the area. During the study period, penicillin nonsusceptibility increased from 0% to 17%. The overall case-fatality rate was 25%, which decreased from 33% to 17% between the first and the second half of the study period (P<.001). The proportion of women with pneumococcal bacteremia increased from 37% to 52%. Independent risk factors for fatal outcome were coronary artery disease (P<.001; relative risk [RR], 4.3; 95% confidence interval [CI], 3.4-5.1), neutropenia (P=.001; RR, 3.2; 95% CI, 1.9-4.8), and age 65 years or older (P=.001; RR, 2.9; 95% CI, 1.8-4.2), whereas prior respiratory tract infection (P=.03; RR, 0.3; 95% CI, 0.1-0.5) and the occurrence of pneumococcal bacteremia in the second half of the study period (P=.01; RR, 0.4; 95% CI, 0.2-0.6) were independent predictors of survival. The case-fatality rate in human immunodeficiency virus (HIV)-infectedpatients was significantly lower than in patients not infected with HIV or in those with unknown HIV status (9% vs 27%; P=.006), which correlated with the younger mean +/- SD age of HIV-infectedpatients (33.2+/-6.6 years) compared with patients not infected with HIV (63.1+/-18.1 years) (P<.001). CONCLUSIONS: The case-fatality rate of patients with pneumococcal bacteremia decreased significantly between the first and second half of the study period, despite the increased prevalence of penicillin-nonsusceptible isolates. Independent risk factors for fatal outcome were coronary artery disease, neutropenia, and age 65 years or older, whereas prior respiratory tract infection and the occurrence of pneumococcal bacteremia in the second half of the study period were independent predictors of survival. HIV infection was a predisposing factor for pneumococcal bacteremia but was not a risk factor for fatal outcome.