Fever during neutropenia occurs in > 90% and 80% of allogeneic and autologous hematopoietic stem cell transplantation (HSCT) recipients, respectively. Current guidelines recommend the prophylaxis with fluoroquinolones (FQs) in HSCT patients. Although there is evidence that antibiotic prophylaxis improve clinical outcome in patients with chemotherapy-induced neutropenia, prophylactic antibiotic therapy has not been thoroughly evaluated in HSCT recipients. Therefore, we performed a meta-analysis to evaluate the impact of systemic antibiotic prophylaxis in HSCT recipients on mortality, incidence of infection and related adverse events.
We identified reports that were not restricted to those in English and not restricted to published trials through PubMed, the Cochrane Library, and references of identified studies.
We included prospective, randomized studies on systemic antibiotic prophylaxis in HSCT recipients. The outcome measures included the all-cause mortality, infection-related mortality, febrile episodes, incidence of clinically or microbiologically documented infection, bacteremia, or related adverse events. The summarized odds ratios (ORs) were calculated using the Mantel–Haenszel method and the DerSimonian–Laird method.
Seventeen trials with 1453 patients (842 autologous and 407 allogeneic HSCT recipients) were included. The percentage of autologous and allogeneic HSCT recipients was not specified in 2 trials. Systemic antibiotic prophylaxis was compared with placebo or no prophylaxis in 10 trials and with non-absorbable antibiotic in 2 trials, respectively. Systemic antibiotics other than FQs were evaluated in five out of these 12 trials. Four trials evaluated the effect of addition of antibiotics for gram positive bacteria to FQs. Remaining 1 trial compared the two different systemic antibiotic regimens, FQs versus trimethoprim sulfamethoxazole. As a result, systemic antibiotic prophylaxis reduced the incidence of febrile episodes (OR 0.16; 95 percent confidence interval [CI], 0.09-0.30), clinically or microbiologically documented infection (OR 0.41; 95% CI 0.30-0.57) and bacteremia (OR 0.37; 95% CI 0.26-0.53) without the significant effect on all-cause mortality or infection-related mortality (OR 0.89; 95% CI 0.48-1.66, OR 1.37; 95% CI 0.50-3.76, respectively). Impact of prophylaxis with FQs on mortality was inconclusive because of small number of clinical trials evaluated. Adverse events increased in patients with systemic antibiotic prophylaxis compared to controls (OR 3.32; 95% CI 1.45-7.63). In meta-regression, percentage of allogeneic HSCT recipients was not associated with each outcome measure. With regard to the comparison between different prophylactic regimens, addition of antibiotics for gram positive bacteria to FQs decreased the incidence of bacteremia (OR 0.44; 0.24-0.80) without significant effects on all-cause mortality, infection related death and febrile episodes. There was not significant, but consistent decrease in clinically or microbiologically documented infection (OR 0.55; 95% CI 0.30-1.01). There was significant increase of adverse events in patients receiving addition of antibiotics for gram positive bacteria to FQs (OR 6.65; 95% CI 2.15-20.54).
Systemic antibiotic prophylaxis successfully reduced the incidence of infection in HSCT recipients. However, there was no significant impact on mortality. Impact of prophylaxis with FQs on mortality in HSCT recipients was inconclusive because of small number of trials evaluated.
No relevant conflicts of interest to declare.
