Mortality and associated risk factors in consecutive patients admitted to a UK NHS trust with community acquired bacteraemia

Purpose Within the UK, there is lack of contemporary data on clinical outcomes in patients admitted to hospital with severe community acquired infection. The purpose of this study was to determine outcomes and risk factors associated with mortality in consecutive patients admitted to a UK NHS trust with community acquired infections that cause bacteraemia. Methods From September 2007 to August 2008, demographic, clinical and microbiological data were collected on patients with laboratory confirmed bacteraemia. Multivariate logistic regression was used to determine the association between predicted variables and likelihood of death. Results 686 bacteraemic episodes occurred in 681 patients. The most common sites of infection were non-catheter associated urinary tract infections (140, 20.4%) and biliary tract infections (62, 9.1%). The most common organisms were Escherichia coli (238, 34.7%), Staphylococcus aureus (84, 12.2%) and Streptococcus pneumoniae (40, 5.8%). Of the E coli infections, extended spectrum beta-lactamase (ESBL) producers accounted for 21/238 (8.8%), and of the S aureus infections, methicillin resistant S aureus (MRSA) accounted for 14/84 (16.7%). 124 (18.2%, 95% CI 15.3% to 21.1%) people died within 7 days and 170 (25.0%, 95% CI 21.7% to 28.2%) within 30 days. Age (OR 2.17, 95% CI 1.54 to 3.06), Charlson comorbidity index (OR 1.21, 95% CI 1.10 to 1.34), and Pitt score (OR 1.49, 95% CI 1.32 to 1.67) were highly significantly associated with 30 day mortality (p<0.001). Delay in appropriate antibiotic treatment (OR 1.35, 95% CI 1.05 to 1.75) and an undefined site of infection (OR 2.05, 95% CI 1.19 to 3.53) were less significantly associated with 30 day mortality (p<0.05). Conclusion The 30 day mortality rate in consecutive patients with community acquired bacteraemic infection was 25.0%. These figures could be used as performance indicators to compare outcomes in different UK NHS trusts. With the exception of delay in appropriate antibiotic treatment, predictors of mortality at 30 days were non-modifiable.