The Value Of Orthopedic POSSUM And P-POSSUM Scoring System In Predicting Operation Risk In Aged Patients With Hip Fracture

BackgroundOur countries with ageing populations are seeing an increase in the number of older people who suffer hip fracture. It is a consensus that early intervention and rehabilitation can improve the outcome. It is of great significance that how to objectively and accurately predict mortality and morbidity for patients and their family members as well as the operation, which is also a topic for doctors in recent years. POSSUM and P-POSSUM score systems, one of the most reliable and wide scoring systems, have been proved that they can predict morbidity and mortality very well in general surgery, But there is few reports about the value in aged patient with hip fracture in our country through validation methods including discrimination (C-index statistic), observed:expected (0:E) ratio, calibration with the Hosmer-Lemeshow test. ObjectiveTo evaluate the value of orthopedic POSSUM and P-POSSUM scoring system in predicting inhospital mortality and morbidity in aged patients undergoing hip fracture operation. Materials And MethodsPatient data were collected retrospectively in Jinshan Hospital Fudan University. Overall 164 aged patients (54 man and 110 woman), at least 60 years of age (mean age 76.53 years) (range 60-94 years), suffered hip fracture over six years (from January 2007 to December 2012), were studied with orthopedic POSSUM and P-POSSUM scoring system. The predicted postoperative mortality and morbidity were compared with the observed values. The ability was evaluated through three characteristics: discrimination (C-index statistic), observed:expected (0:E) ratio, calibration with the Hosmer-Lemeshow test. ResultsOrthopedic POSSUM performed well in predicting morbidity, the overall 0:E ratio was 0.82 without significant difference(X2=2.09, P>0.05), and discrimination was good (C-index= 0.87), and calibration was good (Hosmer-Lemeshow test H 2=3.66, df=8, P>0.05). P-POSSUM could perform well in predicting mortality too, the overall 0:E ratio was 0.80 without significant difference (X2=0.21, P>0.05), and discrimination was good (C-index= 0.93) and calibration was good (Hosmer-Lemeshow test H2 =2.59, df=4, P>0.05). While POSSUM showed overestimation the inhospital mortality, the overall 0:E ratio was 0.27 with significant difference (X2=31.0,P>0.05),and discrimination was good (C-index= 0.93) and calibration was good (Hosmer-Lemeshow test H2=8.45, df=4, P<0.05). ConclusionsGood discrimination, nonsignificant overall 0:E values, as well as good calibration makes Orthopedic POSSUM and P-POSSUM valuable tools when they are used to predict inhospital morbidity and mortality in aged people undergoing hip fracture operation respectively, but Orthopedic POSSUM overestimate mortality...