Outcome Of Escherichia Coli (E.coli) Blood Stream Infections (BSI) In Haematology-Oncology Patients- Hematopoetic Stem Cell Transplantation (HSCT) Is Not Associated With Increased Mortality

E.coli are one of the commonest organisms causing blood stream infections (BSI) in oncology and hematology patients. With the growing incidence of extended spectrum beta lactam resistance (ESBL) this single centre retrospective analysis was undertaken to identify the predictors of outcome of E.coli septicaemia in hemato-oncology patients. From 2002 to 2012, 1089 patients with haematological malignancies had 3528 episodes of BSI and of them 257 in 177 patients (7.3%) were with E.coli.

177 (M: 114, F: 63; median age: 52yr., range: 16-89) patients with haematological malignancies treated from April 2002 to April 2012 were included in analysis. Underlying diagnosis was AML (n=47), ALL (n=23), Lymphoma (n=73), MDS (n=6), myeloma (n=19) or other conditions (n=9). There were 257 episodes of E.coli septicaemia in 177 patients and each episode was analysed as an individual event. Of the 257 episodes, 34 (13%) developed following stem cell transplant (Allo=18, Auto=16), 85 (33%) following intensive chemotherapy, 55 (21%) after non-intensive therapy, 47 (18%) after various other forms of chemotherapy and 36 (14%) did not have any therapy prior to BSI. E.coli were isolated from peripheral blood (PB, n=112, 44%), central venous access (Line, n=130, 51%) or unspecified source (NK, n=15, 6%). At the time of developing E.coli BSI, ANC was below 0.5 in 136 cases (53%), 167 (65%) had platelet count below 100, 78 (30%) had BUN above 10, 69(27%) had creatinine above 120, 74(29%) had serum bilirubin above 34 and 129 (50%) had trans-aminitis. E.coli BSI increased significantly from year 2007 onwards (110/2073, 5% vs. 137/1455, 9%; p=0.0001). Mortality within 30 days of BSI with E.coli was used as a marker to estimate the impact of the event. 28 of 257 episodes (10.9%) resulted in death within 30 days. In univariate analysis, mortality was significantly higher with age above 45yr. (26/185 vs. 2/72, p=0.01), GGT more than 50 (25/180 vs. 3/77, p=0.02), BUN above 10 (17/78 vs. 11/179, p=0.0002), creatinine above 120 (12/22 vs. 16/188, p=0.035) and PB as source of BSI (18/112 vs. 10/145, p=0.02). In Cox multivariate analysis, PB as source (HR: 2.2, 95% CI: 1.0-4.9, p=0.048), GGT above 50 (HR: 4.4, 95% CI: 1.2-15.5, p=0.022), age above 45 yr. (HR: 4.5, 95% CI: 1.1-19.6, p=0.04) and creatinine above 120 (HR: 3.1, 95% CI: 1.1-8.7, p=0.03) were independently associated with mortality within 30 days of E.coli BSI. Mortality increased significantly with the number of risk factors [0-1: 2/68 (3%), 2-3: 22/140 (16%), 4: 8/25 (32%), HR: 3.9, 95% CI: 2.0-7.4, p=0.0001]. 16% of E.coli were resistant to beta lactam antibiotics but the number was too small to assess the impact on mortality. HSCT was not associated with increased mortality.

E.coli BSI are associated with significant mortality that is not associated with underlying diagnosis but with the age and associated organ dysfunction. Continued surveillance is needed to assess the impact of ESBL incidence, status and its clinical significance.

Cavet:Celgene: Consultancy, Honoraria; Eli Lilly: Honoraria.