Fluoroquinolone prophylaxis in patients with neutropenia: a meta-analysis of randomized placebo-controlled trials

A recent meta-analysis, which included non-placebo open-labeled trials, showed that fluoroquinolone prophylaxis reduces mortality in neutropenic patients, whereas two recent large trials failed to show a similar benefit. Therefore, we performed a meta-analysis of randomized, blinded, placebo-controlled trials of fluoroquinolone prophylaxis in neutropenic patients. We searched several databases for relevant trials in any language. We used random effects models for pooling dichotomous data and assessed the between-study inconsistency with I (2). Two investigators independently assessed the eligibility and quality of the included trials. A total of 2,721 patients were randomized in eight eligible trials. Compared to the placebo, there was a statistically non-significant but consistent decrease in mortality with fluoroquinolone prophylaxis (4.5% vs. 3.9%, relative risk (RR) 0.76, 95% confidence interval (CI) 0.54, 1.08, p = 0.13, I (2) = 0%). Significant inconsistency, however, accompanied the pooled analysis of febrile episode (39% vs. 31%, RR 0.76, 95% CI 0.55, 1.03, p = 0.08, I (2) = 96.5%). To an extent, this inconsistency was explained in the subgroup analyses by the type of patient population studied and the type of fluoroquinolone used (p for interaction