Related Experiment Video
Updated: Jun 21, 2026
Competing-Risk Nomogram for Predicting Cancer-Specific Survival in Multiple Primary Colorectal Cancer Patients after Surgery
Published on: September 27, 2024
Predicting post-operative mortality in patients undergoing colorectal surgery using P-POSSUM and CR-POSSUM scores: a
Edmund Leung1, Ali M Ferjani, Nigel Stellard
1Department of Surgery, University Hospitals Coventry and Warwickshire, Warwick, UK. ed.leung@doctors.org.uk
Introduction:
POSSUM and its variants Portsmouth POSSUM (P-POSSUM) and Colorectal POSSUM (CR-POSSUM) equations were derived from a heterogeneous general surgical population, which have been used successfully to provide risk-adjusted operative mortality rates. CR-POSSUM utilises fewer parameters, allowing ease of use. The aim of this study was to predict the mortality outcome in colorectal surgery using these scoring systems compared to the observed mortality and to devise a new scoring system with improved accuracy.
Methods:
The study was conducted prospectively on all consecutive patients requiring elective and emergency colorectal surgery between April 2002 and May 2005. The outcome parameter was defined as 30-day mortality. The observed mortality was compared with predicted mortality by the scoring systems. Hosmer and Lemeshow test was used to assess statistical accuracy of POSSUM.
Results:
Eight hundred ninety-nine patients underwent colorectal surgery during the study period. There were 619 elective and 281 emergency patients. Observed 30-day mortality rate was 9%, compared with predicted mortality rate of 13.5% with POSSUM, 5% with P-POSSUM and 9.5% with CR-POSSUM.
Conclusion:
POSSUM's mortality rate was overestimated, while P-POSSUM's mortality rate was underestimated. CR-POSSUM, the simplest system of all three, most accurately predicted mortality in our unit.