Comparative Effectiveness of Ceftriaxone Alone Versus in Combination with Azithromycin For Community-Acquired Pneumonia Among Children Enrolled in the CDC Etiology of Pneumonia in the Community (EPIC) Study

Background:   The combination of macrolide and beta-lactam therapy, compared with beta-lactam monotherapy, has been associated with reduced mortality for adults with bacteremic pneumococcal pneumonia. The potential benefit of macrolide combination therapy for moderate to severe pediatric community-acquired pneumonia (CAP)in children is unclear.  To  address this question we studied children hospitalized with CAP and enrolled in the CDC EPIC study.

Methods:   We compared the effectiveness of ceftriaxone monotherapy versus ceftriaxone plus azithromycin among children aged <18 years hospitalized with clinical and radiographic evidence of CAP at three children's hospitals in Tennessee and Utah.  The analyses were restricted to children receiving initial therapy with either ceftriaxone (reference) or ceftriaxone plus azithromycin during the first two days of hospitalization.  Those receiving other antibiotics were excluded.  Antibiotic selection was based on clinical judgment.  The primary outcome was hospital length of stay (LOS), and was analyzed using proportional hazards regression, adjusting for sociodemographics, comorbidities, clinical symptoms, and physiologic measurements collected at admission.  We also examined LOS using a propensity score matching strategy to reduce concerns about confounding by indication.

Results:   Of the 678 children included in the analyses (46% of enrolled children), 522 (77%) received ceftriaxone alone (median age 18 months) and 156 (23%) received ceftriaxone plus azithromycin (median age 57 months).  Median LOS was similar between groups, 62 hours (ceftriaxone) vs. 58 hours (ceftriaxone plus azithromycin), unadjusted hazard ratio [HR] 1.05 (0.87-1.25); adjusted HR 0.86 (95% CI, 0.70-1.05). Propensity score matching retained 132 children in each exposure group (median ages 46 and 45 months, respectively). Results from this analysis were similar (adjusted hazard ratio 0.94, 95% CI 0.74-1.20) (Figure). 

Conclusion:   The addition of azithromycin to ceftriaxone therapy for children hospitalized with CAP was not associated with differences in LOS in our study; however, analyses to evaluate important subgroups and to examine additional clinical outcomes are ongoing.