Bloodstream infections in neutropenic patients with haematological malignancies

Background: Patients with haematological malignancies have higher risk of acquiring bloodstream infection (BSI). Neutropenia resulting from cytotoxic chemotherapy is the most common risk factor. Infections can progress rapidly with poor outcomes. Understanding the epidemiology may enable prevention and effective management. We investigated and compared the incidence of BSI amongst patients with haematological malignancies and neutropenia and examined the changing spectrum of organisms, and their antimicrobial profiles. Methods: BSI data between July 1st 2009 and June 30th 2015 was reviewed. Results: Three hundred and fifty five BSI were identified in 255 neutropenic patients. Acute myeloid leukaemia (AML) accounted for 40%, Non-Hodgkin's lymphoma for 22% and Acute lymphocytic leukaemia (ALL) for 11.8%. A neutrophil count of <500 cells/mL was present in 93.2%. The overall incidence was 5.40 BSI per 1000 Haematology Occupied Bed days (OBD). Viridans streptococci and Enterococcus species were the most predominant Gram-positives. Vancomycin resistant Enterococcus faecium (VRE) emerged as the predominant Enterococcus species during the study period . Escherichia coli was the most predominant Gram-negative and Extended-spectrum beta-lactamases (ESBL) were detected in 7.1% of isolates. Amongst the Enterobacteriaceae and Pseudomonas aeruginosa dual resistance to Piperacillin-tazobactam and Gentamicin was detected in 5.4%. Conclusion: Our incidence of BSI was 5.40 per 1000 OBD, however variability in reporting of rates in neutropenic patients with haematological malignancies makes comparison between studies difficult, highlighting the need for rate reporting standardization. The epidemiology of organisms causing BSI has changed over time. There is a trend towards an increasing incidence of VRE and multidrug resistant Gram-negatives. (C) 2019 Australasian College for Infection Prevention and Control. Published by Elsevier B.V. All rights reserved.