A randomized prospective study of ceftazidime and ciprofloxacin with or without teicoplanin as an empiric antibiotic regimen for febrile neutropenic patients.

Febrile neutropenic patients were randomized to receive ceftazidime and ciprofloxacin with the addition of teicoplanin in cases with clinical suspicion of a Hickman line-associated infection. At 48 h both clinical and bacteriological analyses were made. Patients were categorized as: success, in cases of improvement or stability in the clinical condition; failure, if there was a deterioration of the clinical condition; and non-evaluable, in cases of protocol violation or death due to other causes. A total of 86 patients have completed the study so far; 43 patients were randomized to the ceftazidime and 43 to the ciprofloxacin group. Teicoplanin was added to the monotherapy in 12 patients in the ceftazidime group and 15 in the ciprofloxacin group. The diagnoses were: acute myelogenous leukaemia (25), non-Hodgkin's lymphoma (34), Hodgkin's disease (17); acute lymphoblastic leukaemia (9) and chronic granulocytic leukaemia (1). Forty-eight hour clinical assessment showed a response rate in the ceftazidime group of: success 18/31 (58%), failure 13/31 (42%); in the ceftazidime and teicoplanin group: success 8/12 (67%), failure 4/12 (33%); in the ciprofloxacin group: success 23/28 (82%), failure 5/28 (18%); and in the ciprofloxacin and teicoplanin group: success 11/15 (73%), failure 4/15 (27%). Blood cultures were positive in 48/86 (56%) cases, with more than one organism isolated in seven of these 48 cases. Organisms isolated were: coagulase-negative staphylococci (20), Escherichia coli (13), Pseudomonas aeruginosa (4), Staphylococcus aureus (3), diphtheroids (3), Klebsiella sp. (3), Enterobacter cloacae (2), Streptococcus faecalis (2), Ent. adenocarboxylata (1), Clostridium septicum (1). Strept. B (1), alpha streptococcus (1) and P. fluorescens (1), In-vitro testing showed that all Gram-positive organisms were sensitive to teicoplanin, and Gram-negative organisms to ciprofloxacin or ceftazidime. There were seven cases of superimposed infections with eight organisms isolated, all cases occurring in patients receiving ciprofloxacin alone: coagulase-negative staphylococci (4), Strep. sangius (1), Strep. viridans (1), diphtheroids (1) and P. maltophilia (1). Both ciprofloxacin and ceftazidime are comparable in efficacy as empirical monotherapy for febrile neutropenic patients. There is, however, a significant increase in the incidence of superimposed infection in patients receiving only ciprofloxacin. In view of the very high incidence of Gram-positive septicaemia, we suggest that teicoplanin should be added as part of the initial empiric antibiotic regimen in all febrile neutropenic patients.