Impact of atypical coverage for patients with community-acquired pneumonia managed on the medical ward: Results from the United States community-acquired pneumonia project

Study Objective. As current guidelines for treatment of community-acquired pneumonia (CAP) recommend empiric antimicrobial coverage for atypical pathogens, we evaluated the need for atypical coverage by examining length of hospital stay (LOS) and mortality in patients with CAP who were managed on the medical ward. Methods. Medical records of patients with CAP admitted from January 1, 1997-December 31, 2001, from 176 United States nonteaching community hospitals were reviewed. Patients were divided into one of three mutually exclusive groups on the basis of intravenous antimicrobials received on days 1 or 2 of hospital stay: ceftriaxone monotherapy, ceftriaxone plus a macrolide, or levofloxacin. Variables evaluated for their ability to predict outcome were patient age, year of hospital admission, geographic region, preadmission setting, preadmission antimicrobial treatment, timing of antimicrobial administration, comorbid disease, and duration of intravenous antimicrobial treatment. The impact of initial antimicrobial regimen on LOS and mortality was evaluated in regression models while controlling for significant predictors of outcome. Results. Of 8975 patients evaluated, 2453 met the inclusion criteria. Significant differences were noted among patients who received ceftriaxone (932 patients), ceftriaxone plus a macrolide (872), and levofloxacin (649) with respect to mean ± SD age (72 ± 16, 67 ± 18, and 70 ± 17 yrs, respectively; p<0.0001), admission from a nursing home (21%, 11%, and 15%, respectively; p<0.0001), and duration of intravenous antimicrobial treatment (4.4 ± 2.7, 4.0 ± 2.6, and 3.6 ± 2.5 days, respectively; p<0.0001). The LOS predictors were age, geographic region, coexisting heart failure, and duration of intravenous antimicrobial therapy. Mortality predictors were age, admission from a nursing home, coexisting heart failure, and coexisting cancer. After controlling for these predictors of outcome, no significant differences were noted among the three groups for LOS (5.5 ± 3.5, 4.8 ± 2.9, and 4.8 ± 2.9 days, respectively; p=0.2791) or mortality (3.1%, 2.0%, and 2.6%, respectively; p=0.8461). Conclusion. Initial coverage for atypical pathogens does not affect LOS or mortality among patients with CAP managed on the medical ward.