You can understand the logic, at least on the surface. Primary care physicians lean toward broad-spectrum antibiotics in community-acquired pneumonia (CAP) when their patients are very sick; the expensive, broader-spectrum agents are perceived as being somehow “stronger.” But they are not—they simply wipe out a broader swath of pathogens (and unfortunately, non-pathogens). What’s more, severity of disease has little to do with the likely organism causing a patient’s pneumonia; age, host defense, interaction with comorbidities, nutrition level—these are the determinants of severity.
Kate Kupiec, PharmD, of Wake-Forest Baptist Health in Winston-Salem, presented evidence at IDWeek 2013 favoring narrower-spectrum agents for CAP, rather than broad-spectrum agents that are more expensive, associated with greater risk, and whose use is not supported by literature. The Infectious Diseases Society of America guidelines provide a nice framework for choosing antimicrobials in CAP. Note that the current IDSA guideline for CAP on which the case-matched control group is based dates from 2007.
In this retrospective, case-control study, 62 patients were enrolled; 31 cases and 31 controls, matched for age and APACHE II severity score. Thirty-day mortality was 16.1% among cases (5 deaths) and 6.4% among controls (2 deaths), for a P value of .43, which is not statistically significant (in this case, there is nearly a 50% likelihood that the observed 3-death difference is due to chance). Patients were identified using criteria from the CMS Pneumonia Core Measure–Initial Antibiotic Selection for CAP in Immunocompetent Patients–ICU (so these were sick ICU patients). Exclusion criterion: lack of radiographic evidence of pneumonia. Cases received broad-spectrum therapy (eg, piperacillin/tazobactam) and controls received IDSA guideline-recommended therapy (ceftriaxone-based regimens). The primary outcome was 30-day mortality. Secondary outcomes included days of therapy (DOT), hospital length of stay (LOS), readmission within 30 days, Clostridium difficile infection, acute kidney injury (RIFLE criteria), and Candida superinfection (Table).
|Outcome||Broad-spectrum||Controls (IDSA-guideline)||P value|
|Intravenous DOT, median (range)||9 (2-52)||5 (1-22)||0.01|
|Total DOT, median (range)||9 (3-52)||7 (2-27)||0.02|
|Ventilated days, median (range)||7 (1-59)||5 (1-18)||0.26|
|ICU LOS, median (range)||4.5 (1-69)||3 (1-26)||0.03|
|Total LOS, median (range)||9.5 (2-69)||6.5 (2-29)||0.05|
|Readmission within 30 days, n (%)||6 (23.1)||4 (13.3)||0.49|
|C difficile infection, n||2||0||1.00|
|Candida superinfection, n||5||1||0.20|
|Acute kidney injury, n (%)||8 (25.8)||4 (12.9)||0.34|
Broad-spectrum antibiotics were associated with statistically significant longer duration of therapy, days on ventilator, ICU days, and length of stay. Statistically insignificant but suggestive associations were seen for 30-day readmission rate, C difficile infection, Candida superinfection, and acute kidney injury. Broad-spectrum therapy conferred no measurable clinical benefit in patients with severe CAP. The study is small, and retrospective, but the attempt at severity matching is fairly impressive—there’s no evidence here that sicker pneumonia patients are the ones that were given broad-spectrum agents. Dr Kupiec recommended that patients with CAP should be treated with IDSA guideline-recommended antibiotics, regardless of disease severity.
Kupiec K, Johnson J, Ohl C, et al. Broad-spectrum antibiotics offer no advantage over guideline-recommended antibiotics for patients with severe community-acquired pneumonia (CAP). Abstract presented at: IDWeek 2013; October 4, 2013; San Francisco. View all abstracts at IDWeek 2013 Web site.