Mortality and risk stratification in patients with Clostridium difficile-associated diarrhoea

Aim This study aimed to describe the mortality in hospital patients with a first documented episode of Clostridium difficile-associated diarrhoea (CDAD) and to identify prognostic risk factors. Method A cohort study of 158 patients was carried out with CDAD diagnosed over a 8-month period in a large acute UK teaching hospital. Logistic multivariable regression aided construction of a scoring system to stratify risk of death. The main outcome measure was the 30-day inpatient mortality. Results Most affected patients were medical (n = 101, 64%), with general surgical (n = 30, 19%) and orthopaedic patients (n = 27, 17%) forming the rest. General surgical patients (mean age 78 years) were significantly younger than medical (82 years) or orthopaedic patients (85 years, P = 0.008). Overall 30-day mortality was 38%. 30-day mortality was higher in medical (46%) and orthopaedic patients (37%) compared with general surgical patients (13%, P = 0.006). Most surgical patients were those admitted as an emergency. A scoring system was devised and used within the first 72 h. A point was awarded for each of the following: age >= 80 years, clinically severe disease (sepsis, peritonitis, >= 10 episodes of diarrhoea in 24 h), WCC >= 20 or CRP >= 150, urea >= 15, albumin < 20. Point counts of 0-1, 2-3 and 4-5 were associated with mortality rates of 22%, 55% and 89% respectively. Conclusion Inpatient mortality from CDAD is high. Variability exists between different specialities. Patients at high risk of death can potentially be identified earlier using clinical and biochemical risk factors.