Role of glycopeptides as part of initial empirical treatment of febrile neutropenic patients: a meta-analysis of randomised controlled trials.

We did a meta-analysis of randomised controlled trials studying glycopeptides as part of the initial empirical treatment of febrile neutropenic patients with a beta-lactam and with or without an aminoglycoside. 14 randomised controlled trials that studied 2413 patients were included in the analysis. A better outcome regarding treatment success, without modification of the initial regimen, was accomplished with the inclusion of a glycopeptide in the empirical therapy; this better outcome applied to the full set of studied patients (OR=1.63, 95% CI 1.17-2.28), as well as in three important subsets of patients--those with microbiologically documented infections (2.03, 1.39-2.97), patients with bacteraemia (1.80, 1.23-2.63), and patients with severe neutropenia, defined as a white blood cell count below 100 cells/microL (2.24, 1.15-4.39). However, mortality was not different in the compared groups (0.67, 0.42-1.05). Overall treatment success was not different if a glycopeptide was added to the antimicrobial regimen in the case of continuation of fever 72 hours or more after the start of treatment (1.02, 0.68-1.52). Also, the inclusion of a glycopeptide in the empirical regimen did not lead to a difference regarding time to defervesence. Adverse effects (4.98, 2.91-8.55), including nephrotoxicity (2.10, 1.12-3.95), were more common in the group receiving a glycopeptide as part of the empirical treatment. In conclusion, our meta-analysis suggests that there are good reasons why glycopeptides should not be routinely used as part of the initial empirical treatment of febrile neutropenic patients.