Gafter-Gvili, A., Fraser, A., Paul, M., Vidal, L., Lawrie, T.A., van de Wetering, M.D., . . . Leibovici, L. (2012) Antibiotic prophylaxis for bacterial infections in afebrile neutropenic patients following chemotherapy. Cochrane Database of Systematic Reviews, 1, CD004386.

The purpose of this meta analysis and sytematic review was to evaluate the effect of antibiotic prophylaxis on mortality and infection in neutropenic patients. In addition, subgroups of patients who may benefit the most were identified, and whether or not the effectiveness of different antibiotic regimens were similar was evaluated, as were the adverse effects of different regimens and the emergence of quinolone-resistant bacteria.

• Databases searched from 1996–2011 included MEDLINE, EMBASE, Cochrane databases of controlled trials and cancer network registry of trials, multiple relevant conference proceedings.
• Regarding keywords, appendices provided extensive detail of exact search terms used per database.
• Inclusion criteria included patients with cancer and neutropenia from chemotherapy or after bone marrow transplantation. Control groups received placebo, no intervention or an alternate intervention, quasi-randomized controlled trials, or randomized clinical trials (RCTs) for initial review through 2005. An update to 2011 included only RCTs. Exclusion criteria were not stated,.

119 total references were retrieved. Cochrane Handbook for Systematic Reviews methods were used to evaluate and commend on the literature used.

• 109 total studies were evaluated.
• 13,579 cases were reviewed.
• Key characteristics: patients with cancer—sites not specified—but most involved patients with leukemia. Some studies sampled febrile episodes rather than cases.

Active antitumor treatment

Antibiotic prophylaxis resulted in significant reduction in risk of mortality across 46 trials analyzed (RR = 0.66, 95% confidence interval [CI] [0.55, 0.79], p < 0.00001). The greatest effect was with quinolones, although differences between regimens was not statistically significant. The effect was larger for trials in which prophylaxis was begun at the onset of neutropenia. An advantage was seen for all quinolones except for norfloxacin. Antibiotic prophylaxis significantly reduced infection-related mortality (RR = 0.61, 95% CI [0.48, 0.77], p = 0.04), decreased occurrence of fever, documented infection, and occurrence of bacteremia. Quinolones and TMP-SMZ were both associated with side effects that were mostly diarrhea and nausea. TMP-SMZ was associated with drug resistant bacteria cultures (RR = 2.42, 95% CI [1.35, 4.36]).  With quinolones, no significant differences were noted between study groups compared to placebo or other interventions. Addition of gram-positive coverage did not show any apparent benefits in terms of mortality.

Findings support use of quinolones as prophylaxis of choice since they reduced risk of death compared to placebo or not intervention and were generally associated with fewer side effects and less resistant bacterial cultures in treated patients. Levofloxacin or ciprofloxacin are recommended.

• All cause mortality was only available for 47 of the studies. 
• Length of follow-up in studies may have been too short to fully detect emergence of resistant bacteria. 
• Most studies were limited to patients with hematologic cancers. 
• Applicability to patients with solid tumor types requires further study.

Prophylactic quinolone antibiotic therapy is recommended for patients with hematologic cancers and those who are likely to develop neutropenia. Additional research is needed to better define patients with solid tumors that may benefit from antibiotic prophylaxis. In most studies, prophylaxis was begun when chemotherapy was initiated, rather than when neutropenia occurred.  Prophylaxis should be accompanied by surveillance to monitor quinolone-resistant gram-negative bacteria and other resistant organisms.