Clinical Impact of Colonization By Multi-Drug Resistant Organisms in Stem Cell Transplantation

Hematologic patients (pts) are at high risk for colonization by multi-drug resistant organisms (MDROs), primarily due to their immunosuppressed status and inpatient treatments. Colonization status (CS) is a known risk factor for MDRO infections. Detecting CS is essential to guide empiric antibiotic therapy, as MDRO infections have higher mortality rates and inadequate antibiotic coverage compromises their prognosis.

We conducted a retrospective, single-center study to analyze the impact of MDRO CS in Stem-Cell Transplantation (SCT). We included 272 pts, tested 255 pts (93.8%) admitted for SCT from June 2021 to June 2024 (38.8% allogeneic SCT and 61.2% autologous SCT). MDRO colonization screening was performed including rectal swabs in 254 pts (99.6%), pharyngeal swabs in 249 pts (97.6%) and nasal swabs in 150 pts (58.8%). The target MDROs for identification included Enterobacteriaceae, Acinetobacter, Enterococcus, Pseudomonas aeruginosa, and Staphylococcus aureus. None of the pts received antibacterial prophylaxis either before or during hospitalization. Our empiric antibiotic therapy is piperacillin-tazobactam plus amikacin based on our resistance patterns. This therapy is adjusted based on MDRO CS as necessary. Statistical analysis was performed using the Chi-square and Fisher’s exact tests.

Of the tested 255 pts, 36 (14.1%) pts tested positive for MDRO colonization with 6 isolated species. Of the 36 pts, 34 (94.4%) presented with a single isolated MDRO: Escherichia coli (36.1%), Klebsiella pneumoniae (47.2%), Staphylococcus aureus (5.6%) and Enterococcus faecium (5.6%). The other 2 pts (5.6%) had two different isolated MDRO: Pseudomonas aeruginosa/ Citrobacter freundii and Escherichia coli/ Enterococcus faecium. Regarding multiresistance mechanisms of gram-negative bacteria (GNB), extended-spectrum β-lactamases (ESBLs) were 69.4%, carbapenemasas were 8.3%, both ESBLs and carbapenemasas were 5.6% and class D β-lactamase (OXA-48 oxacilinase) were 2.8%. We compared variables including sex, hospital of origin, type of stem cell transplantation (autologous vs. allogeneic) and risk of colonization based on baseline disease (low-risk disease with outpatient treatments vs. high-risk disease with inpatient treatments). No significant differences were found in baseline characteristics between carriers (CRs) and non-carriers (non-CRs) pts. Of the 255 pts, 247 (96.9%) presented with fever during admission. Microbiological cultures were positive in 76 (30.8%) pts, of whom 9/76 (11.8%) were positive for MDROs. Our empiric antibiotic therapy was adequate in 54/76 (71.1%) pts. In the group of MDRO colonized pts, 35/36 (97.2%) presented with fever, leading to modification of empiric antibiotic therapy based on MDRO colonization in 29/35 (82.9%) pts. Microbiological cultures were positive in 9/35 (25.7%) pts, of whom 2/9 (22.2%) were positive for MDROs. The empiric antibiotic therapy based on CS was adequate in 8 of these pts. The only MDRO-related death occurred in a non-tested patient at admission, resulting in a 3-day delay in appropriate antibiotic therapy until carbapenem-resistant Klebsiella pneumoniae was isolated, which was also detected in rectal swabs at UCI admission. In the entire cohort, the median hospital stay was 25 days (range: 10-137). No significant differences were found between CRs and non-CRs pts regarding clinical outcomes at admission, including rates of ICU hospitalization, infection-related mortality, and overall mortality.

Given that infectious complications are significant causes of mortality in SCT, testing for MDRO CS may be a useful strategy for guiding empiric antibiotic therapy. In our cohort, despite a relatively low rate of MDRO colonization compared to other registries, our findings suggest that testing for MDRO CS at admission for SCT could be beneficial. Notably, our only MDRO-related death occurred in a patient who was not tested, resulting in a delay in the administration of the appropriate antibiotic. Adjusting antibiotics based on MDRO CS in our study allowed us to cover more than 70% of the microbial isolations with the initial antibiotic regimen.