We aim to describe mortality in orthopaedic patients with Clostridium difficile associated diarrhoea (CDAD), to identify prognostic factors for 30 day mortality, and to modify a CDAD risk score to fit to orthopaedic patients.

This was a two centre, retrospective, observational study including consecutive patients with a first episode of CDAD between 2005–2007. 79 patients were identified, comprising 11 elective patients (14%) and 68 emergency patients (86%). 73 patients (92%) underwent surgery and all but two patients received broad spectrum antibiotics prior to CDAD. The overall 30 day mortality was 29% (n=26). The predominant diagnosis was a fractured femoral neck (66%, n=52).

The most significant multivariable model in predicting 30 day mortality comprised increasing white cell count (WCC, OR 1.20 [for 10% variable increase]; 95% CI 1.06–1.36 p=0.003) and decreasing albumin (OR 0.86 [for single unit decrease]; 0.86–0.95, p=0.003), with adjustment for age ł80 years (OR 6.39, 1.15–35.52, p=0.04). CRP was found to be not significant. Based on this, modification of the previously described Clostridium difficile prognostic index leads to a point awarded for WCC ł20, albumin Ł20, age ł80, urea ł15 or clinically severe disease (peritonitis, sepsis, ł10 episodes of diarrhoea per day). This produces low (0–1 points), medium (2–3 points) and high (4–5 points) risk of death groups, with mortalities of 15%, 47% and 75% respectively for all orthopaedic patients, and 14%, 41% and 67% respectively in only the validation cohort.

CDAD in orthopaedic patients mainly affects emergency patients, in particular those with fractured femoral neck. Inpatient mortality is high, and a high white count and low albumin are significant predictors of mortality. Modification of an easily remembered scoring system based on this can help identify orthopaedic patients likely to die from an episode of CDAD, allowing early aggressive therapy and early objective referral to gastrointestinal teams.