Ten RCTs were identified (3,940 patients, range 33 to 712). Mean Jadad score was 3.1, range 2 to 4, with eight high quality studies and two low quality studies. Results were generally given for an intention-to-treat analysis and mainly for a fixed-effect analysis, unless stated otherwise.
Treatment success
Significantly higher treatment success for gemifloxacin versus other antibiotics (OR 1.39, 95% CI 1.15 to 1.68; Ι²=11%; eight studies). Clinical success was also significantly higher for gemifloxacin versus other quinolones (OR 1.69, 95% CI 1.28 to 2.23; Ι²=0%; five studies) but not for gemifloxacin versus β-lactams and/or macrolides (OR 1.16, 95% CI 0.89 to 1.50; Ι²=43%; three studies). Results did not change when they were calculated for clinically evaluable populations. An intention-to-treat analysis found treatment success was higher for gemifloxacin versus other antibiotics for the two subsets of diseases (OR 1.37, 95% CI 1.03 to 1.83; Ι²=18%; four studies for community-acquired pneumonia; and OR 1.40, 95% CI 1.09 to 1.80; Ι²=28%; five studies for acute exacerbation of chronic bronchitis) but the results were not significant when they were calculated for clinically evaluable populations.
All-cause mortality
There were no significant differences in all-cause mortality for gemifloxacin versus other antibiotics, versus other quinolones, or versus β-lactams and/or macrolides.
Adverse events
For gemifloxacin versus other quinolones, there were no significant differences in total adverse events (OR 0.89, 95% CI 0.56 to 1.41; Ι²=63%; four studies) or nausea for gemifloxacin versus other quinolones. However, patients experienced significantly more diarrhoea (OR 2.01, 95% CI 1.00 to 4.06; Ι²=0%; three studies) and rash (OR 2.66, 95% CI 1.03 to 6.87; Ι²=0%; two studies) with gemifloxacin.
For gemifloxacin versus β-lactams and/or macrolides, there were significantly lower total adverse events (OR 0.71, 95% CI 0.57 to 0.89; Ι²=41%; five studies) for gemifloxacin versus β-lactams and/or macrolides, including diarrhoea (OR 0.47, 95% CI 0.32 to 0.69; Ι²=51%; five studies). There were higher rates of nausea and rash (Ι²=0%; three studies) for gemifloxacin but the effects were not significant.
There were no significant differences between gemifloxacin and comparators for microbiological outcomes. Results of sensitivity analyses were reported. A funnel plot indicated that there was no significant publication bias.