The POSSUM And P-POSSUM Scoring System For Risks Assessment In Pancreatoduodenectomy

Objective:Pancreatoduodenectomy (PD), as a classical surgical operation,is widely applied in the diseases treatment on duodenum, pancreatic head,middle and lower segment of common bile duct and ampulla of Vater.However, PD operation which involved in large area excision need to be takenmuch time and easily infected with patients, who have some diseases beforesuch as severe obstructive jaundice and malnutrition, thereby leading to theoccurrence of postoperative complications, and even more threatening thepatients' life. From1935Whipple1firstly adopted PD to the1970s, themortality was up to25%after PD, while in recent years medical standardsimprovement, experience accumulation on curing postoperative complications,and application to new operation equipment have decreased mortalitysignificant to5%~10%2. Nevertheless, PD as one of the most complicatedabdominal surgical operation still had a high incidence of postoperativecomplications which reached up to30%~50%3. Under medical technology,PD remains a higher operation risk. Consequently, an objective, accuratescoring method is urgently established to predict PD postoperative risk andfurther evaluate patients' condition accurately thereby strengthening treatmentin period of peri operation, improving the postoperative complications andfactors incur death, and adjusting patients' body function state to prevent orreduce postoperative complications and mortality, so that patients can receivethe greatest treatment effect. At present, several scoring method have beenapplied to predict the risk of surgical operation, such as acute physiology andchronic health scoring Ⅱ(acute physiology and chronic health evaluation,APACHE II, mainly used to assess the death risk of ICU patients), Goldmancardiac risk index (CRI, mainly used for the prediction of cardiaccomplication after non-cardiac operation), Charlson complication index (mainly applied to forecaste the incidence of pulmonary complications),mortality enumeration and physiological morbidity and the operation severityscoring (physlological and operativescore for the enumeration of mortalityand morbidity, POSSUM) and so on. POSSUM scoring system explored byCopeland4in1991was based on the patients' preoperative physiological statusand operative severity to predict postoperative mortality and the incidence ofcomplications, and further assess the risks of operation. POSSUM scoringsystem is comprised of12physiological indices (PS) and6operation indices(OS), and each index is classified into four grades in terms of the degree ofseverity, and was respectively marked at the score of1,2,4,8, and thendifferent regression equation is adopted to predict the patients' postoperativecomplications and mortality. In1996, Whiteley5observed clinical trial in thePortsmouth region and learned that expected mortality in operation calculatedby POSSUM scoring system was higher than the observed mortality.Moreover, POSSUM scoring system overestimated the postoperative deathrisks significantly when it was used to predict lower risk operation inparticular. As a result, Whiteley modified POSSUM regression equation intoPortsmouth-POSSUM only to predict postoperative mortality, and verifiedthat expected mortality value got by P-POSSUM scoring system was moreclose to the observed mortality. Later Whiteley and Prytherch6revisedP-POSSUM regression equation more perfect when they researched theregression to the larger sample. At this stage, POSSUM and P-POSSUMscoring systems are widely used in Europe and the United States, especially inthe fields of rectum7,8, stomach, orthopedic9and so on; while in ChinaPOSSUM scoring system is used narrowly and fell behind Europe and theUnited States obviously, therefore, the reports on PD risk prediction is stillfewer. In this paper, POSSUM and P-POSSUM scoring system is applied topredict postoperative morbidity and mortality of pancreatic duodenalcompared with the actual incidence and judge the clinical predictive ability,besides, provide available method for the reasonable assessment to PDoperation risks. Methods:165consecutive PDs were performed in Hebei MedicalUniversity fourth hospital hepatobiliary surgery from January2010toDecember2011. POSSUM and P-POSSUM scored the clinical data ofpreoperative physiological and operative severity and calculated thepostoperative expected mortality and morbidity for each case, whilst theobserved values were recorded and compared with expected values.Statistical methods: The statistical analyses were subjected to test formeasurement data and Chisquare test for enumeration data using IBM SPSS19.0software. The least significant difference (LSD) test was used to separatethe means when the difference was significant between groups (P <0.05).Results:67cases in165consecutive PDs had complications (40.6%),such as abdominal infection (n=25, morbidity=15.2%), pancreatic fistula (n=19, morbidity=11.5%), gastrointestinal bleeding (n=17, morbidity=10.3%); and in operation2cases died (mortality=1.2%), depending on thefailure to control infection caused by septic shock and multiple organsdisorder. POSSUM scoring system predicted that86cases have got thepostoperative complications (expected morbidity=52.1%), between whichand observed morbidity had significant differences (Df=1, Χ2=4.40, P <0.05). Stratified analyses revealed O: E values were relative higher (≥0.5),and predictive value was the highest when POSSUM scores ranged from0.40to0.60(O: E≈1, P>0.05); POSSUM predicted22patients died, which had asignificant difference with comparison to observed death cases (P <0.01, Df=1, Χ2=17.97). Additionally, for survival group and death group, they had thesignificant differences (P <0.05) between in physiology and POSSUM scores.Whereas P-POSSUM exhibited7patients for mortality, compared withobserved mortality, which had no significant differences (P>0.05, Df=1,Χ2=1.83).Conclusion: POSSUM scoring system overestimated PD postoperativecomplications and mortality; while P-POSSUM scoring system has high valuefor predict postoperative mortality of patients.