1320. Initial Antibiotic Selection and Hospital Length of Stay in Adults Hospitalized with Community-acquired Pneumonia (CAP) in the CDC Etiology of Pneumonia in the Community (EPIC) Study
Session: Oral Abstract Session: Clinical Management of Infectious Disease
Saturday, October 20, 2012: 11:00 AM
Room: SDCC 26 AB
Background: Whether CAP outcomes are affected by selection of different guideline-concordant empiric antibiotic regimens remains uncertain. We compared the two most common antibiotic treatments—fluoroquinolone (FQN) monotherapy vs. ceftriaxone plus azithromycin (CFT/AZITHRO)—on hospital length of stay (LOS) for adults hospitalized with CAP in the CDC EPIC study.   Methods: Adults ≥ 18 years old with clinical and radiographic CAP were enrolled at 5 hospitals in Chicago and Nashville. This study was restricted to adults enrolled from Jan. 2010 through Oct. 2011, admitted to non-intensive care unit beds and who received treatment with either FQN monotherapy or CFT/AZITHRO only during the first two days of hospitalization. Antibiotics were selected by treating clinicians.  We compared LOS between the antibiotic regimen groups using a multivariate proportional hazards regression model with adjustment for potential confounders. In a secondary analysis, we applied propensity score matching to reduce concerns about confounding by indication.   Results: Of 1542 adults enrolled in EPIC during the study period, 740 (48%) were admitted to a non-ICU bed and received FQN monotherapy (n=423) or CFT/AZITHRO (n=317). The median LOS (interquartile range) was 58 (41, 94) hours for the FQN group and 71 (48,122) hours for the CFT/AZITHRO group.  There were 4 in-hospital deaths in the FQN group and 0 in the CFT/AZITHRO group. Complete data were available for 641 subjects for the multivariate proportional hazards model, which demonstrated shorter LOS for the FQN group (adjusted hazard ratio [HR]: 0.78; 95% confidence interval (CI): 0.66, 0.94). The propensity score matching strategy retained 177 patients in each antibiotic exposure group; results from this analysis were consistent with the multivariate model (HR: 0.76; 95% CI: 0.61, 0.93) (Figure).   Conclusion: Preliminary findings from this study of adults hospitalized with CAP outside the ICU suggest that treatment with FQN monotherapy may be associated with shorter hospital LOS than CFT/AZITHRO. Continued enrollment in this study may increase the precision of our estimates and allow for evaluation of secondary clinical outcomes.
Figure:    
Wesley H. Self, MD, MPH1, Richard G. Wunderink, MD2, Kathryn Edwards, MD, FIDSA1, Derek J. Williams, MD, MPH1, Sherene Fakhran, MD, MPH3, Evan J. Anderson, MD4, Robert Balk, MD5, Christopher Trabue, MD6, Lauri A. Hicks, DO7, Yuwei Zhu, MD, MS8, James D. Chappell, MD, PhD1, Carrie Reed, DSc, MPH7, Lyn Finelli, DrPH, MS7, Anna M. Bramley, MPH9, Seema Jain, MD7 and Carlos G. Grijalva, MD, MPH1, (1)Vanderbilt University School of Medicine, Nashville, TN, (2)Pulmonary and Critical Care, Northwestern University Feinberg School of Medicine, Chicago, IL, (3)John H. Stroger, Jr. Hospital of Cook County, Chicago, IL, (4)Children's Memorial Hospital and Northwestern Memorial Hospital, Chicago, IL, (5)Rush University Medical Center, Chicago, IL, (6)University of Tennessee Health Sciences Center, Baptist Hospital, Nashville, TN, (7)Centers for Disease Control and Prevention, Atlanta, GA, (8)Biostatistics, Vanderbilt University School of Medicine, Nashville, TN, (9)Influenza Division, Centers for Disease Control and Prevention, Atlanta, GA

Disclosures:

W. H. Self, None

R. G. Wunderink, None

K. Edwards, None

D. J. Williams, None

S. Fakhran, None

E. J. Anderson, None

R. Balk, bioMerieux: Investigator, Research grant

C. Trabue, None

L. A. Hicks, None

Y. Zhu, None

J. D. Chappell, None

C. Reed, None

L. Finelli, None

A. M. Bramley, None

S. Jain, None

C. G. Grijalva, None

Findings in the abstracts are embargoed until 12:01 a.m. PST, Oct. 17th with the exception of research findings presented at the IDWeek press conferences.