Infusion of Microbially Contaminated Hematopoietic Cell Products Has No Adverse Effect on Clinical Outcome.

Microbial contamination of hematopoietic progenitor cell (HPC) products has been described in variable frequencies, with typically no adverse outcome reported for the recipients.

Methods: We retrospectively reviewed 3291 consecutive autologous and allogeneic infusions of HPCs and donor lymphocyte infusions (DLI) administered at MD Anderson between 2001 and 2005. Products were bone marrow (BM), mobilized peripheral blood (PB) or DLIs. Sterility testing was performed at collection, thawing for cryopreserved products and infusion. Appropriate antibiotic therapy was started prior to infusion for patients who received products known to have a positive culture; or as soon as identification of a positive culture for the ones who received fresh products. Following IRB approval, clinical data were extracted from the medical records and BMT Dept Database.

Results: Among the 3291 infusions, 37 patients (1.1%) received contaminated BM (n=15) or PB (n=22) products. No contaminated DLI were infused. Median age was 49 years (range, 20–68); 23 were males and 14 females. Diagnoses were NHL (n=10), Hodgkin s Disease (n=7), Multiple Myeloma (n=6) AML (n=3), ALL (n=1), Aplastic Anemia (n=2), Systemic Mastocytosis (n=1) and solid tumors (n=3). Products were autologous (n= 21), related allogeneic (n=6) or unrelated allogeneic (n=10); products were manipulated (n=11; 29.7%), fresh (n=16; 43.2%) or frozen (n=21; 56.8%). Contamination at harvest or apheresis was detected in 29 (78.4%) products, 16 freshly infused and 13 that proceeded to cryopreservation. At thawing and infusion, 10 products had positive cultures: 2 of them had a positive culture of apheresis collection and 8 had been negative. Isolated agents were Coagulase-negative Staphylococcus (CNS) in 27/29 positive products at collection and 6/10 at infusion; gram-positive cocci, unspeciated, were detected in 2 products at collection and 1 at infusion; Bacillus species, Enterococcus faecalis and Micrococcus species were detected in 1 product each at infusion (not detected until 1–5 days post-infusion). Two patients had complications at infusion which were unrelated to the postive culture: 1 patient with a cardiomyopathy developed transient pulmonary edema and 1 patient experienced chest pain; both resolved without evidence of infection. Seven patients had positive blood cultures after infusion between days +2 and +8; 6/7 were with different organisms than those identified in their product; 1/7 had CNS which responded completely to vancomycin and had no complication due to this infection. All autologous patients had neutrophil recovery in a median of 10 days (range, 8–14); all allogeneic patients but one engrafted in a median of 13 days (range, 5–21). Admissions lasted a median of 21 days (8–34) for autologous and 28 days (20–65) for allogeneic transplant patients. At a median follow-up of 568 days (49–1971), 8 of the 37 patients died. Cause of death was recurrent disease (n=2); secondary AML after lymphoma treatment (n=1); acute GVHD (n=1); regimen-related toxicity after allogeneic transplant (n=2); brain abscess due to nocardia (n=1) and sepsis 75 days post transplant with an organism not previously identified in the product (n=1). None of the 37 patients experienced complications or had clinical sequelae due to contaminated infusions.

Conclusion: Infusion of microbially contaminated products in this series did not adversely affect the outcome of the transplant patient.