Fluoroquinolones or macrolides in combination with β-lactams in adult patients hospitalized with community acquired pneumonia: a systematic review and meta-analysis

The best treatment option for hospitalized patients with community-acquired pneumonia (CAP) has not been defined. The effectiveness of β-lactam/fluoroquinolone (BLFQ) versus β-lactam/macrolide (BLM) combinations for the treatment of patients with CAP was evaluated. PubMed, Scopus and the Cochrane Library were searched for observational cohort studies, non-randomized and randomized controlled trials providing data for patients with CAP receiving BLM or BLFQ. Mortality was the primary outcome. A meta-analysis was performed. MINORS and GRADE were used for data quality assessment. Seventeen studies (16 684 patients) were included. Randomized trials were not identified. A variety of β-lactams, fluoroquinolones and macrolides were used within and between the studies. Mortality was reported at different time points. The available body of evidence had very low quality. In the analysis of unadjusted data, mortality with BLFQ was higher than with BLM (risk ratio 1.33, 95% CI 1.15–1.54, I2 28%). BLFQ was associated with higher mortality regardless of the study design, mortality recording time, study period and study BLM group mortality. BLFQ was associated with higher mortality in American but not European studies. No difference was observed in patients with bacteraemia and septic shock. In the meta-analysis of adjusted mortality data, a non-significant difference between the two regimens was observed (eight studies, adjusted risk ratio 1.26, 95% CI 0.95–1.67, I2 43%). In the absence of data from randomized controlled trials recommendations cannot be made for or against either of the studied regimens in this group of hospitalized patients with CAP. Well designed randomized controlled trials comparing the two regimens are warranted.