ObjectiveThe Physiological and Operative Severity Score for the enUmeration of Mortality and morbidity (POSSUM) model and its Portsmouth (P-POSSUM) modification are used extensively to predict post-operative mortality and morbidity in general surgery. The aim was to analysis the predictive value of these models in patients undergoing hepatobiliary surgery. MethodsEligible articles were identified by searching such electronic databases as PubMed, The Cochrane Library (Issue 10, 2013), Science Citation Index, CNKI, WanFang Data and CBM from 1991 to October 2013. Each study was assessed according to the inclusion and exclusion criteria. Then data were extracted, pooled, and analyzed using Comprehensive Meta Analysis Version 2. ResultsTen studies were included. The morbidity analysis included five studies and 683 patients on POSSUM with a weighted O/E ratio 0.71 (95%CI 0.60 to 0.81). The mortality analysis included seven studies with 1 291 patients on POSSUM and six studies with 1 793 patients on P-POSSUM. Weighted O/E ratios for mortality were 0.42 (95%CI 0.27 to 0.57) for POSSUM and 0.74 (95%CI 0.53 to 0.95) for P-POSSUM. ConclusionPOSSUM significantly overestimates postoperative morbidity in patients undergoing hepatobiliary surgery. Compared with the original POSSUM, P-POSSUM is more accurate for predicting post-operative mortality. Modifications to POSSUM and P-POSSUM are needed for audit in hepatobiliary surgery.
Abstract ： Objective To evaluate the clinical validity of Physiological and Operative Severity Score for theenUmeration of Mortality and Morbidity (POSSUM) in primary lung cancer patients undergoing surgery in order to get clinical treatment reference. Methods A total of 179 patients，with 124 males and 55 females，with primary lung cancer surgery between January 2007 and October 2010 were included in the First Affiliated Hospital of Xinjiang Medical University. Their age was 59.2±11.4 years．Before the surgery，POSSUM was used to each patient to rate the results and substituted the results into Copeland equation to calculate the predicted postoperative mortality and morbidity. The actual number of complications and death were calculated after surgery and the patients were divided into one group with postoperative complications and another group without postoperative complications. The physiological score and the operative risk score were compared between the two groups. Actual number of complications and death were compared with thenumber predicted by POSSUM respectively. The clinical factors related to the actual number of complications and death were analyzed. Results Among 179 patients， there were postoperative complications in 78 patients. The physiological score and the operative severity score were significantly higher in the group in whose complications occurred compared with those without complications (16.11±2.53 points versus 14.88±1.86 points for physiological score，P=0.000 ； 13.47±2.83 points versus 12.88±2.57 points for operative severity score，P=0.000). There was no statistical difference in complication between the predicted and actual number (65/179 versus 78/179，χ2=1.968，P=0.161). There was statistical difference in death between the predicted and actual number(12/179 versus 3/179，χ2=5.636，P=0.018).Univariable analysis revealed that 5 single factors were related to the complications， only hemoglobin was related to the death. Conclusion The POSSUM gives satisfactory prediction in morbidity rate but overrates the mortality rate in primary lung cancer patients undergoing surgery， and 5 single clinical factors show a better clinical value.