Matthew E. Falagas^a,b, MD, MSc, DSc
Konstantinos Z. Vardakas^a, MD
^a Alfa Institute of Biomedical Sciences, Athens, Greece
^b Tufts University School of Medicine, Boston, Massachusetts, USA

In a recently published meta-analysis (CMAJ 2008; 179: 1269-77.) we sought to examine whether respiratory fluoroquinolones were associated with better clinical outcomes in adult patients with pneumonia. We searched PubMed, Current Contents, Scopus, EMBASE, ClinicalTrials.gov and Cochrane for published trials comparing fluoroquinolones (levofloxacin, moxifloxacin and gemifloxacin) with macrolides and/or β-lactams and extracted data regarding effectiveness and toxicity. A meta-analysis including 23 trials was performed. Mortality was not different between the compared antibiotics (odds ratio [OR]: 0.85; 95% confidence interval [CI]: 0.65-1.12). Fluoroquinolones were more effective for the treatment of pneumonia than comparator antibiotics in the intention-to-treat (OR: 1.17; 95% CI: 1.00-1.36), clinically evaluable (OR: 1.26; 95% CI: 1.06-1.50) and microbiologically assessed (OR: 1.67; 95% CI: 1.28-2.20) populations. Fluoroquinolones were also more effective than β-lactam/macrolide combinations (OR: 1.39; 95% CI: 1.02- 1.90), for patients with severe pneumonia (OR: 1.84; 95% CI: 1.02-3.29), for hospitalized patients (OR: 1.30; 95% CI: 1.04-1.61) and for patients requiring intravenous therapy (OR: 1.44; 15% CI: 1.13-1.85). Fluoroquinolones were more effective in open label (OR: 1.35; 95% CI: 1.08-1.69) but not in blinded trials (OR: 1.13; 95% CI: 0.85-1.50). In conclusion, fluoroquinolones were associated with higher treatment success for severe forms of pneumonia, but a benefit in mortality is not evident. A well-designed randomized controlled trial that includes patients with severe pneumonia with or without bacteremia is needed to further clarify this important clinical question.