VRE is a well known infectious complication among immunocompromised patients. We previously reported that early-onset VRE bacteremia after aHPCT was associated with a rapidly deteriorating clinical course and death (Bone Marrow Transplant 2005; 35: 497–499). The current analysis was performed to identify risk factors for the development of VRE infection after aHPCT and to determine its prognostic importance for other post transplant outcomes. From 1997–2011, 800 consecutive adult patients with hematologic diseases underwent aHPCT at our institution. These included 448 (56%) males and 723 (90%) Caucasians. The median age was 46 yrs (range, 18–71), 451 (56%) had ≥ 2 chemotherapy regimens before aHPCT, 92 (12%) had prior radiation therapy, 470 (59%) were in remission at transplant, and comorbidity scores included: 280 (36%) low, 229 (30%) intermediate and 268 (34%) high-risk. The most common diagnoses were AML (306), ALL (111), MDS (110), CML (87), and NHL (85). 389 (49%) patients had unrelated donors, 676 (85%) were 8/8 HLA matched, 632 (79%) had myeloablative transplants, and HPC sources included 532 (66%) bone marrow, 213 (27%) peripheral blood progenitor cell, 55 (7%) umbilical cord blood. Eighty-three (10%) pts developed VRE after aHPCT and 8 of these had recurrent episodes. The median time from transplant to the first episode of VRE was 54 days (range, 2–2256); 37 (45%) occurred during the initial transplant hospitalization and 9 (11%) were in outpatient reduced intensity conditioning transplants. Due to the many different underlying diagnoses a recursive partitioning analysis was performed which categorized them into 3 risk groups for VRE: ALL (high risk), AML/MDS/CLL (intermediate risk) and all other diagnoses (low risk). Incidences of VRE by date of aHPCT were 6.1% (1/97-4/01), 10.9% (5/01–8/04), 12.5% (9/04–6/08) and 16.5% (7/08–12/11). Multivariable Cox analysis found 6 risk factors for VRE infection: year of aHPCT (per one year increase) (p=0.047), intermediate + high comorbidity score (p=0.005), active disease at transplant (p=0.014), ALL diagnosis (p=0.003), unrelated donor (p<0.001), and umbilical cord blood cell source (p=0.001). Although African American race, HLA disparity, CD34 + and TNC cell doses were risk factors identified by univariable analysis, they did not retain significance on multivariable analysis. VRE infection was not associated with achievement of donor T-cell chimerism, acute or chronic GVHD, or disease relapse. However, on multivariable analysis it was significantly associated with worse overall survival (HR 4.45, 95% CI 3.39–5.84, p<0.001) and relapse free survival (RFS) (HR 4.28, 95% CI 3.20–5.74, p<0.001). A 1 month landmark analysis for VRE and overall survival is shown in the figure.
Seventy-three (88%) pts with VRE died at a median of 1.2 mos (range, 0–32) after it was detected. However, only 4 (6%) of these deaths were due to VRE while the remaining resulted from other infections (27%), disease relapse (23%), non-pulmonary organ failure (12%), GVHD (22%), pulmonary failure (6%) and graft failure (4%). We conclude that the incidence of VRE after aHPCT has increased over time and is highly associated with mortality although not usually directly from VRE. Strategies to further enhance immune reconstitution post transplant and strict adherence to infection control measures are needed particularly for those with ALL, alternative donors, higher comorbidity scores or not in remission at the time of aHPCT.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.