A comparison of double β-lactam combinations with netilmicin/ureidopenicillin regimens in the empirical therapy of febrile neutropenic patients

In a randomized trial ceftazidime plus piperacillin or azlocillin, and netilmicin plus piperacillin or azlocillin were used as initial empirical therapy in 202 febrile neutropenic episodes. Netilmicin plus azlocillin was the most effective combination with a clinical response rate of 81% in clinically and microbiologically documented infections compared with 63% for ceftazidime plus piperacillin. All of the episodes of Gram-negative bacteraemia treated with azlocillin responded compared with 43% of those treated with piperacillin. Gram-positive organisms accounted for 52% of all bacteriologically documented infections and 40% of the febrile episodes were treated with vancomycin for presumptive or documented Gram-positive infection. Patients treated with netilmicin had significantly more nephrotoxicity than those given the double β-lactam combinations (14.8% vs. 3.5%; P < 0.05). However, this difference was not shown in those patients who did not receive concurrent vancomycin or amphotericin. The double β-lactam combinations were associated with more hypokalaemia (58.2% vs. 37.7%; P < 0.05) and more colonization with yeasts (24% vs. 10.4%; P < 0.05) but there was no evidence that their use was associated with prolongation of neutropenia. These results indicate that ceftazidime plus a ureidopenicillin would be adequate empirical therapy in situations where the concomitant use of nephrotoxic agents precludes the use of aminoglycoside containing combinations.