Article Text


PWE-103 A risk stratification score for predicting 30-day mortality in Clostridium difficile associated diarrhoea
  1. R Sood1,
  2. L L Wong1,
  3. R Swallow2,
  4. S J Smale1
  1. 1Department of Gastroenterology, York Teaching Hospital, York, UK
  2. 2Pharmacy Department, York Teaching Hospital, York, UK


Introduction Clostridium difficile associated diarrhoea (CDAD) causes significant morbidity and mortality in susceptible patients. There are no validated risk stratification scores to identify those patients at the greatest risk of death.

Methods Data were collected on 125 sequential patients diagnosed with CDAD in our institution between August 2008 and October 2010. Data on age, co-morbidities, number of antibiotics prescribed and course length as well as other relevant medications such as proton pump inhibitor (PPI) were recorded. The length of and timing of any admissions in the preceding 13-month and physiological variables such as the patient's temperature, blood pressure and pulse at diagnosis of CDAD were recorded as were any haematological and biochemical investigations taken within 24 h of diagnosis.

Results Hospital admissions between 14 and 30 days (p=0.031), increasing co-morbidities (p=0.05), systolic blood pressure <100 mm Hg (p=0.048) and heart rate greater than 100 beats per minute (p=0.043) were significantly associated with increased 30-day mortality. There was a clear trend of increased mortality for increasing length of stay (LOS) on the current admission before development of diarrhoea (If LOS greater than 7 days before onset, p=0.07), temperature <35°C (p=0.07), recent discharge from hospital within 30 days (p=0.106) and to a lesser degree, low albumin (albumin <30 g/l p=0.221) at the time of diagnosis, age >85 years (p=0.947) and current PPI use (p=0.224). From the data, a scoring system was derived whereby 1 point was given to the non-statistically significant trends and 2 points to the statistically significant trends. If there were 1 to 3 co-morbidities, this was scored as 1 and greater than 3 co-morbidities scored as 2. There was a maximum score of 14. On applying this retrospectively to the existing database, 39 patients were identified where all the parameters in the scoring system were available. Of these, a score of 4 or less was associated with low risk of 30-day mortality (0% mortality in 13 patients). A score of 5 to 7 was associated with moderate risk of 30-day mortality (33.3% mortality in 15 patients) and a score of 8 or more to be associated with a high risk of 30-day mortality (72.7% mortality in 11 patients).

Conclusion A simple scoring system holds promise for defining those patients at greatest risk of 30-day mortality from CDAD in our population. However a note of caution should be exercised. Applying a scoring system retrospectively to a dataset from which information has been used to derive the score does not validate it and a prospective study is planned to validate the York scoring system.

Competing interests None declared.

Statistics from

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.