RESUMO
OBJECTIVE: POSSUM system is widely used and validated for 30-day mortality and morbidity prediction. The aim of this study was to evaluate the performance of five POSSUM's equations (POSSUM, P-POSSUM, V-POSSUM, V-POSSUM physiology and V-POSSUM Cambridge) on predicting 30-day mortality and morbidity in elderly patients undergoing major elective vascular surgery. DESIGN: A retrospective longitudinal cohort study. SETTING: A study conducted at an University Hospital. PARTICIPANTS: 208 elderly patients (≥ 60 years) undergoing major elective vascular surgery. MEASUREMENTS: Data collected from the clinical files included patient's characteristics, diagnosis, surgery, comorbidities, parameters from POSSUM score, 30, 60 and 90-day mortality and 30-day morbidity. POSSUM system's goodness-of-fit for predicting mortality and morbidity was assessed by Hosmer-Lemeshow test (H-L T) and Standardized Mortality/Morbidity Ratio (SMR) and discriminative ability by the area under the ROC curves (ROC-AUC). Patients' average age was 70.8 years, 81% males. INTERVENTIONS: None. MAIN RESULTS: The overall 30-day mortality rate was 2.97% (n=6) and 30-day morbidity was 29.2% (n=59). POSSUM, P-POSSUM, V-POSSUM, V-POSSUM physiology and V-POSSUM Cambridge equation predicted an overall of 29.1, 4.43, 15.3, 21.9 and 13.5 deaths, respectively. POSSUM morbidity equation predicted 105.0 complications. H-LT p-values were 0.001, 0.164, 0.208, 0.011, 0.331 and <0.001, respectively. SMRs and 95% confidence interval (CI) were 0.21[0.04-0.37], 1.35[0.27-2.44], 0.39[0.08-0.71], 0.27[0.06-0.49], 0.44[0.09-0.80] and 0.56[0.42-0.71], respectively. ROC-AUC and 95% CI were 0.72[0.49-0.95], 0.72[0.49-0.95], 0.73[0.51-0.94], 0.69[0.50-0.89], 0.72[0.52-0.92] and 0.71[0.63-0.79], respectively. CONCLUSIONS: P-POSSUM had the best performance predicting 30-day mortality. All the other overestimated 30-day mortality. Prediction of morbidity was inadequate. POSSUM scoring models may not be robust tools for risk prediction in elderly patients undergoing major elective vascular surgery and need further calibration and discrimination.