Prospective multicentric study of the etiology of 1051 bacteremic episodes in 782 cancer patients
- 1 January 1993
- journal article
- research article
- Published by Springer Nature in Supportive Care in Cancer
- Vol. 1 (1) , 34-46
- https://doi.org/10.1007/bf00326637
Abstract
A total of 1051 bacteremic episodes (782 patients) were prospectively recorded in 10 cancer centers (9 French, 1 Belgian), with: patient's age (mean 53, range 1–89 years), underlying cancer, neutropenia (E. coli (10.7%), Klebsiella-Enterobacter-Serratia (6.1%), other enterobacteriaceae (2.2%), Pseudomonas aeruginosa (4.8%), other nonfermenters (4.7%), coagulase-negative staphylococci (CNS, 40.8%), Staphylococcus aureus (9.9%), streptococci (5.4%), enterococci (2.2%), anaerobes (3.4%), yeasts (3.5%), and other bacteria (6.9%). The CDC (Centers for Disease Control) criteria (1988) were used to assess significance: group 1: pathogenic species (616 episodes; 59%); group 2: clinical signs and isolation of a “contaminant” species (47; 4.5%); group 3: as in group 2 with an i.v. line and empiric antibiotic treatment (181 episodes including 176 CNS; 17%); group 4: non-significant (207 episodes including 203 CNS; 20%). Groups 1–3, in which the episodes were considered to be significant (844 episodes; 80%) were compared with non-significant episodes (Fisher). Significant differences (P≤0.05) were seen in time to growth (median growth within 24 h vs 48 h), fever (86% vs 54%), chills (40% vs 3%), hypotension (10% vs 2%), septic shock (9% vs 1%), polymicrobial etiology (10% vs 0.5%), and initiation of empiric antibiotic treatment (71% vs 4%). Bacteremic episodes of CDC groups 1,3 and 4 were further studied in episodes with a single isolate as a doubtful clinical significance (482 episodes) and episodes with ≥2 bottles positive of probable clinical significance (569 episodes; 54%). In group 1 (218 doubtful, 398 probably significant episodes) significant differences were seen in chills (36% vs 52%), shock (7% vs 13%), polymicrobial (8% vs 17%), initiation of empiric antibiotic treatment (60% vs 72%); in group 3 (87 doubtful, 94 probably significant) in time to growth delay; in group 4 (177 doubtful, 30 probably significant) in proportion with implanted catheter (26% vs 52%), fever (62% vs 10%), and time to growth. This study confirms the predominant role of Gram-positive cocci in bacteremia occurring in cancer patients.Keywords
This publication has 35 references indexed in Scilit:
- Streptococcal and Enterococcal Bacteremia in Patients with CancerClinical Infectious Diseases, 1992
- Vancomycin Added to Empirical Combination Antibiotic Therapy for Fever in Granulocytopenic Cancer PatientsThe Journal of Infectious Diseases, 1991
- Gram-positive bacteraemia in granulocytopenic cancer patientsEuropean Journal of Cancer and Clinical Oncology, 1990
- A Three-Year Study of Positive Blood Cultures, with Emphasis on PrognosisClinical Infectious Diseases, 1990
- CDC definitions for nosocomial infections, 1988American Journal of Infection Control, 1988
- Nosocomial Infection Rates at an Oncology CenterInfection Control & Hospital Epidemiology, 1988
- Clinical importance of polymicrobial bacteremiaDiagnostic Microbiology and Infectious Disease, 1986
- Infection in cancer patients: A continuing associationThe American Journal of Medicine, 1986
- Polymicrobial Septicemia in the Cancer PatientMedicine, 1986
- The Clinical Significance of Positive Blood Cultures: A Comprehensive Analysis of 500 Episodes of Bacteremia and Fungemia in Adults. I. Laboratory and Epidemiologic ObservationsClinical Infectious Diseases, 1983