19 Trends in the Incidence and Spectrum of Organisms Responsible for Bacteremia after Allogeneic Stem Cell Transplantation Using Two Different Prophylactic Strategies: A 12-Year Experience at Memorial Sloan-Kettering Cancer Center

Track: BMT Tandem "Scientific" Meeting
Thursday, February 14, 2013, 4:45 PM-6:15 PM
Ballroom A-D (Salt Palace Convention Center)
Ubonvan Jongwutiwes, MD , Department of Medicine, Infectious Disease Service, Memorial Sloan-Kettering Cancer Center, New York, NY
Kun Xiao, MD MPH , Department of Medicine, Infectious Disease Service, Memorial Sloan-Kettering Cancer Center, New York, NY
Dick Chung , Department of Medicine, Infectious Disease Service, Memorial Sloan-Kettering Cancer Center, New York, NY
Anindita Chakraborty, MD , Memorial Sloan-Kettering Cancer Center, New York, NY
George I. Getrajdman, MD , Department of Radiology, Interventional Radiology Service, Memorial Sloan-Kettering Cancer Center, New York, NY
Sergio A. Giralt, MD , Department of Medicine, Adult Bone Marrow Transplant Service, Memorial Sloan-Kettering Cancer Center, New York, NY
Ann A. Jakubowski, PhD MD , Department of Medicine, Adult Bone Marrow Transplant Service, Memorial Sloan-Kettering Cancer Center, New York, NY
Susan K. Seo, MD , Department of Medicine, Infectious Disease Service, Memorial Sloan-Kettering Cancer Center, New York, NY
Genovefa Papanicolaou, MD , Department of Medicine, Infectious Disease Service, Memorial Sloan-Kettering Cancer Center, New York, NY

The present study aimed to analyze changes in the incidence of and pathogens associated with bacteremias and the development of microbial resistance during two different prophylactic strategies in adult recipients of allogeneic hematopoietic stem cell transplantation (HSCT) at Memorial Sloan-Kettering Cancer Center.

Methods: This was an observational cohort study of 1,138 consecutive adult HSCT from 2000-2011. Bacteremias occurring from day -7 to +100 post HSCT were entered into a computerized database. Beginning November 2005, all patients except those receiving nonmyeloablative-conditioning regimens were given vancomycin prophylaxis from day -2 through day +7. Fluoroquinolone (FQ) prophylaxis was initiated in November 2006. We compared the overall incidence of bacteremia, changing spectrum of the organisms, and the resistance pattern between 2000-2005 (period A) and 2006-2011 (period B).

Results:  During the study period, 312 (28%) patients had 363 bacteremic episodes with 392 pathogens isolated. The median onset of first episode was 7 days post transplant, and 62% of the episodes occurred before neutrophil recovery. The incidence of gram-positive bacteria (GPB) and gram-negative rod (GNR) bacteremia was 19% and 15%, respectively. Enterococcus faecium was the most commonly isolated GPB (46%), followed by viridans group streptococci (17%), while Escherichia coli was the most common GNR (26%). Vancomycin-resistant enterococci comprised the majority of enterococcal isolates (81% in period A and 90% in period B, P=0.2011).  Eighty-nine percent of the Viridans streptococcal bacteremias (VSB) and 60% of the GNR bacteremias occurred before neutrophil recovery. The overall incidence of bacteremia decreased significantly from 32% in period A to 25% in period B (P=0.005). Viridans group streptococci comprised 31 of 168 (18%) organisms in period A compared to 6 of 224 (3%) in period B (P <0.0001). The incidence of enterococcal bacteremia remained stable (11% versus 9%, p=0.47). There was no change in the incidence of GNR bacteremia during the study period (17% in 2000 – 2006 and 14% in 2007 – 2011, p=0.27).  Among GNR, the rate of FQ resistance was 39% and was unchanged between the two periods.

Conclusions: During a 12-year period, the overall incidence of bacteremia decreased. The decrease was mainly due to the dramatic decrease of VSB after institution of vancomycin for prophylaxis. The incidence of GNR and VRE bacteremia remained stable in the periods pre and post FQ prophylaxis.  The rate of FQ resistance among GNR was also unchanged in the two periods. Continued surveillance is necessary to identify changes in local epidemiology, to reevaluate prophylactic regimens and to guide new empiric-therapy schemes.