Outcome comparison of ampicillin/sulbactam or amoxicillin/clavulanate vs. 3rd generation cephalosporin against community-acquired pneumonia

Background: Use of penicillin class (such as ampicillin/sulbactam) is included in a treatment guideline of community-acquired pneumonia (CAP). However, the use of the wide spectrum antibiotics such as 3^rdgeneration cephalosporin is common in the clinical practice. This practice might aggravate the problems of antimicrobial resistance and high medical expenses.

Methods: The case-control study was performed. We did random matching at a ratio of 1:2 (age; ±5 years, sex; same) for ampicillin/sulbactam or amoxicillin/clavulanate: 3^rdgeneration cephalosporin as empiric antibiotics for  CAP. The patients enrolled were between Jan. 2010 and Jun. 2012. The hospitals participated in this study were 4 academic hospitals in Korea.

Results: We enrolled 225 patients (ampicillin/sulbactam: 72, amoxicillin/clavulanate: 3, 3^rd generation cephalosporin: 150). The average age was 60.4 (minimum 19, maximum 90). We compared the selection bias of choice of empiric antibiotics (penicillin group vs. 3^rd generation cephalosporin) that might related with severity of CAP. CURB-65 of penicillin group vs. 3^rd generation cephalosporin were 1.17±1.05, 1.07±0.95 (p=0.487), respectively, and Pneumonia Severity Index (PSI) were 77.5±28.2, 74.7±29.3 (p=0.50), respectively. However, there was statistically significant difference between both groups in the combination of macrolide or quinolone; penicillin group 68.0% (51/75) vs. 3^rd generation cephalosporin 84% (126/150)(p=0.009). The ‘early failure' judged at 72 hours in penicillin group vs. 3^rd generation cephalosporin was 13.3% (20/150) vs. 21.38% (16/75)(p=0.128), respectively. Overall failure judged at 30 days in penicillin group vs. 3^rdgeneration cephalosporin was 2.7% (2/75), 1.3% (2/150) and was not statistically significant. Additionally we conducted multivariate logistic regression analysis with variables (empiric antibiotics, age, Charlson comorbidity score, presence of combination, PSI score, etc.). In this analysis, PSI was significant (p=0.022), and use of penicillins did not show significance of early failure (OR 1.68, 95% CI: 0.79-3.57, p=0.181)

Conclusion: This case-control study showed that ampicillin/sulbactam or amoxicillin/clavulanate might be not inferior to 3^rd generation cephalosporin for the treatment of CAP and could be used with low risk of treatment failure.