Eyal Kleinhendler1, Matan J Cohen2, Allon E Moses3, Ora Paltiel4, Jacob Strahilevitz5, Amos Cahan6. 1. Hadassah-Hebrew University Faculty of Medicine, Jerusalem, Israel; Department of Medicine, Sourasky Medical Center, Tel-Aviv, Israel. 2. Hadassah-Hebrew University Faculty of Medicine, Jerusalem, Israel; Department of Clinical Microbiology and Infectious Diseases, Hadassah-Hebrew University Medical Center, Jerusalem, Israel. 3. Department of Clinical Microbiology and Infectious Diseases, Hadassah-Hebrew University Medical Center, Jerusalem, Israel. 4. Department of Hematology, Hadassah-Hebrew University Medical Center, Jerusalem, Israel; School of Public Health and Community Medicine, Hadassah-Hebrew University of Jerusalem, Jerusalem, Israel. 5. Department of Clinical Microbiology and Infectious Diseases, Hadassah-Hebrew University Medical Center, Jerusalem, Israel. Electronic address: jstrahilevitz@hadassah.org.il. 6. Department of Clinical Microbiology and Infectious Diseases, Hadassah-Hebrew University Medical Center, Jerusalem, Israel; IBM T.J Watson Research Center, Yorktown Heights, New York, USA.
Abstract
BACKGROUND: There are several empiric antibiotic treatment options for febrile neutropenia, yet there is no universally-accepted initial protocol. We aimed to assess the performance of a protocol (piperacillin, gentamicin and cefazolin) introduced over 40 years ago and compare its coverage against bacteria isolated from blood of neutropenic patients with that of various commonly used antibiotic treatment protocols. METHODS: Adults with neutropenia admitted between 2003 and 2012 to the hemato-oncologic departments and in whom blood cultures were taken on admission were included. Appropriateness of several common antibiotic protocols was assessed based on the susceptibility of the blood isolates. Crude mortality rates were computed by the susceptibility of bacteria isolated from patients' blood to the actual treatment given. RESULTS: In total, 180 admissions of neutropenic patients (95 in patients who had fever above 38 °C) with positive blood cultures were analyzed. The actual antibiotic regimen prescribed was deemed appropriate in 82% of bacteremia episodes. The recommended institutional protocol was used in 62% of bacteremia episodes in neutropenic patients. This protocol would have been appropriate in 85% of all neutropenic bacteremia episodes and 89% of episodes in febrile neutropenia patients compared with piperacillin/tazobactam (79%, P = 0.13 and 76%, P = 0.002, respectively) and imipenem (93%, P = 0.004 and 92%, P = 0.74, respectively). Isolation of bacteria resistant to the actual antibiotic treatment given was associated with higher mortality at one week and at 30 days. CONCLUSION: Common current antibiotic regimens provide similar coverage among febrile neutropenic patients, whereas broad spectrum antibiotic combinations maximize coverage among neutropenic patients.
BACKGROUND: There are several empiric antibiotic treatment options for febrile neutropenia, yet there is no universally-accepted initial protocol. We aimed to assess the performance of a protocol (piperacillin, gentamicin and cefazolin) introduced over 40 years ago and compare its coverage against bacteria isolated from blood of neutropenicpatients with that of various commonly used antibiotic treatment protocols. METHODS: Adults with neutropenia admitted between 2003 and 2012 to the hemato-oncologic departments and in whom blood cultures were taken on admission were included. Appropriateness of several common antibiotic protocols was assessed based on the susceptibility of the blood isolates. Crude mortality rates were computed by the susceptibility of bacteria isolated from patients' blood to the actual treatment given. RESULTS: In total, 180 admissions of neutropenicpatients (95 in patients who had fever above 38 °C) with positive blood cultures were analyzed. The actual antibiotic regimen prescribed was deemed appropriate in 82% of bacteremia episodes. The recommended institutional protocol was used in 62% of bacteremia episodes in neutropenicpatients. This protocol would have been appropriate in 85% of all neutropenic bacteremia episodes and 89% of episodes in febrile neutropeniapatients compared with piperacillin/tazobactam (79%, P = 0.13 and 76%, P = 0.002, respectively) and imipenem (93%, P = 0.004 and 92%, P = 0.74, respectively). Isolation of bacteria resistant to the actual antibiotic treatment given was associated with higher mortality at one week and at 30 days. CONCLUSION: Common current antibiotic regimens provide similar coverage among febrile neutropenicpatients, whereas broad spectrum antibiotic combinations maximize coverage among neutropenicpatients.