【Objective】 To evaluate the effect of POSSUM

P-POSSUM

Cr-POSSUM

and APGBI scoring systems in predicting risk of colorectal cancer resection. 【Methods】 Clinical data of 320 cases of patients accepting surgery on colorectal cancer from July 2005 to March 2010 in Gastrointestinal Department of the Third Affiliated Hospital of Sun Yet-Sen University were collected both pre- and intra-operatively. POSSUM

P-POSSUM

Cr-POSSUM and APGBI scoring systems were used to calculate scores and predict incidence of postoperative complications and mortality. Afterwards

ROC curves of predicted results were drawn. 【Results】 The predicted incidence of postoperative complications and mortality were 39.4% and 10% respectively estimated with POSSUM scoring system

which were higher than the actual ones (25.6% and 2.8% respectively

P < 0.01). While in the elderly group of patients

there was no difference in the incidence of postoperative complications between predicted result and the actual one. The area under curve was 0.767

and the most critical value was 45%. There was no difference in mortality among predicted results estimated with P-POSSUM (3.8%)

Cr-POSSUM (4.1%)

and APGBI (4.4%) scoring systems and the actual one. Respectively

the areas under curve were 0.872

0.906

and 0.936

and the most critical values were 7%

5% and 4%. 【Conclusions】 POSSUM scoring system was effective in estimating incidence of postoperative complications of high risk group of patients while it over estimated mortality. P-POSSUM

Cr-POSSUM

and APGBI scoring systems were all effective in estimating mortality.