Thirty-day mortality of Clostridium difficile infection in a UK National Health Service Foundation Trust between 2002 and 2008

Few standardised data are available on mortality rates in patients with Clostridium difficile infection (CDI). The literature often reports 'attributable' mortality or cannot be universally applied. We aimed to investigate the pattern and trends in all-cause mortality in a large unselected cohort of patients affected by CDI. This was done by means of a retrospective cohort study between 2002 and 2008 of all patients with positive stool toxin tests indicating CDI in one National Health Service (NHS) Trust, comprising three general hospitals and seven community hospitals. Vital status of the patients was determined from two sources. In total, 2571 patients with a first episode of CDI were identified (1638 females; median age 82.1 years). Cumulative mortality at 7 days, 14 days, 30 days and 1 year was 13.4%, 20.8%, 32.5% and 58.7%, respectively. There was no significant difference in mortality between sex, year of diagnosis or hospital site. Mortality at 30 days increased incrementally from 3.4% in those aged <40 years to 41% in those >90 years. Mortality rates were significantly higher than reported by previous studies but were remarkably consistent over the time period and between different hospitals within the Trust. Prognosis falls with increasing age, and the age of this cohort may explain the high 30-day absolute mortality. CDI infection is associated with high early mortality. To reduce mortality, new interventions need to be introduced soon after diagnosis. There is a need for standardised outcome data for CDI. (C) 2010 The Hospital Infection Society. Published by Elsevier Ltd. All rights reserved.