Infections and bacterial colonization during cytotoxic therapy in patients with acute leukemia

Abstract

Infectious complications are a remarkable cause of mortality, morbidity and costs among patients with hematological malignancies. In Finland, 33% of nosocomial bloodstream infections (BSI) occur among hematological patients [1]. During the past 2 decades, the microbial etiology of BSI in patients with febrile neutropenia has shifted from gramnegative to -positive organisms in Western countries [2]. In recent studies, the predominant pathogens recovered from blood cultures of patients with hematological malignancies and cancer have been coagulase-negative staphylococci, viridans-group streptococci, enterococci, Escherichia coli, Klebsiella spp., Pseudomonas aeruginosa and Staphylococcus aureus in various orders [3–5]. Infections due to otherwise extremely rare bacteria such as glucose nonfermenting gram-negative bacilli (GNF-GNB) other than P. aeruginosa and non-tuberculous mycobacteria have also been reported increasingly [4, 6–8]. These emerging inducers of infection are known to colonize hospital water distribution systems and to have a tendency to contaminate fluids and equipment. The significance of environment-associated bacteria as a source of infection in hematological patients was evaluated in an 800-bed referral tertiary hospital where an unusual cluster of bacteremic infections caused by GNF-GNB was previously traced to the hospital’s water distribution systems [9]. In the present study, the microbial data obtained from the blood culture results of patients in the hospital’s 18-bed adult hematology unit during the years 1992–2003 and stored in the Department of Clinical Microbiology database were analyzed retrospectively. In addition, infection and colonization in 11 consecutive adult patients with newly diagnosed acute leukemia (i.e., five patients with acute myeloid leukemia, five with acute lymphoid leukemia and one with myelodysplastic syndrome) were followed during the active phase of their disease in a 1-year prospective surveillance study. Approval for the study was obtained from the hospital’s ethical committee, and the patients gave written informed consent before being enrolled in the study. Throat swabs and fecal samples were collected from each patient at admission and weekly during hospitalization; they were then cultivated for aerobic bacteria, legionellae and mycobacteria. Throat samples were cultured on 5% sheep blood agar with Columbia agar base and on cysteine–lactose-deficient-agar (BBL, Becton Dickinson, Sparks, Maryland, USA), both of which were incubated at 30°C for 4 days to increase the recovery of possible environmental colonizers, and on non-selective charcoalbuffered yeast extract agar plates (BCYEα) (Oxoid, Basingstoke, UK), which were incubated in 5% CO2 at 35°C for 14 days. Fecal samples were diluted stepwise in NaCl from 1:100 to 1:10, and 0.1 ml of each dilution was spread onto cysteine–lactose-deficient agar plates. For culture of legionellae, each sample was pre-treated in HCl– KCl, plated onto non-selective and selective buffered charcoal yeast extract agar plates (BCYEα and BMPA, Oxoid) and incubated as described above. Throat and fecal samples for mycobacterial culture were respectively decontaminated with 2% NaOH–NALC and consecutively with 4% H2SO4 and 2% NaOH–NALC; both sample types were cultured using a Bactec MGIT 960 (Becton Dickinson) automated mycobacteria detection system. Cubital venous blood samples were obtained for blood culture whenever fever or other signs suggestive of bloodstream infection (e.g., chills, hypotension) were present. Two to three consecutive samples were collected into a set of aerobic and O. Perola (*) . A. Laatikainen . M. L. Katila Department of Clinical Microbiology, Kuopio University Hospital, P.O. Box 1777, 70211 Kuopio, Finland e-mail: Outi.Perola@kuh.fi Tel.: +358-17-173231 Fax: +358-17-173202

Topics

0 Figures and Tables

    Download Full PDF Version (Non-Commercial Use)