Causes of hospital-acquired septicaemia--a case control study.

Hospital-acquired septicaemia is common and an important cause of morbidity and expense. Few studies have explored its cause by comparing features of people who do and do not develop septicaemia--the aim of this study. All blood cultures were monitored over one year in a 400-bed tertiary care hospital where special care is taken with intravenous cannulae to prevent septicaemia. Cases had at least one positive blood culture taken at least 48 h after admission. The nearest two patients in the same ward were controls, and information was abstracted on all three from the hospital records by a research nurse. There were 84 cases during the 12 months (3.8 per 1000 admissions) and 167 controls. Univariate odds ratios (95% CI) were 5.57 (2.06-15.95) with the presence of a central venous line, 3.40 (1.16-16.40) with total parenteral nutrition, 2.41 (1.30-4.32) with blood transfusion, 2.10 (1.16-4.56) with immunosuppressive disease, 2.06 (1.02-4.44) with the presence of a urinary catheter and 0.32 (0.17-0.69) with recent surgery. Conditional logistic regression identified a central venous line and blood transfusion to be independent risk factors for septicaemia, odds ratios (and 95% CI) being 5.14 (1.74-15.23) and 2.74 (1.28-5.88), respectively, while recent surgery and antibiotic prophylaxis were independently 'protective' at 0.31 (0.12-0.79) and 0.38 (0.16-0.90), respectively. Thus, in a hospital where great care has been taken to avoid septicaemia, especially by attention to intravenous cannulae, there were only two independent risk factors for the development of hospital-acquired septicaemia--a central venous line and blood transfusion.(ABSTRACT TRUNCATED AT 250 WORDS)