A newly improved POSSUM scoring system for prediction of morbidity in patients with pancreaticoduodenectomy

BACKGROUND: Postoperative morbidity of pancreaticoduodenectomy (PD) is still prevalent in hospitals around the world all over the world. Chirurgia B and Chen Yingtai et al. compared the existing POSSUM and p-POSSUM scoring system in pancreatoduodenectomy. However, in this retrospective cohort study our aim was to investigate the risk factors associated with post-PD complications, and develop a POSSUM scoring system more specific to PD, which we define as PD-POSSUM. METHODS: Data was gathered from 201 patients who underwent PD procedure at No.4 Hospital of Hebei Medical University between January 2016 and December 2018. All 201 patients were included in the modeling group. The morbidity and associated risk factors were collected and analyzed. The risk factors of complication were investigated by the Pearson χ(2) test, Spearman rho test, multivariable linear regression, univariate/multivariate logistic regression. In addition, the discriminating ability of the PD-POSSUM system to predict morbidity is estimated by the area under the receiver operating characteristic curve (ROC-AUC). RESULTS: Morbidity was 45.5% for the 201 patients. Multivariate logistic regression analysis demonstrated a significant connection between postoperative complications and body mass index (BMI) [odd ratio (OR): 3.700; 95% confidence interval (95% CI): 1.594–4.572], pre-existing respiratory diseases (OR: 3.000; 95% CI: 1.542–5.837), international normalized ratio of prothrombin time (OR: 0.321; 95% CI: 0.099–1.038), alanine aminotransferase (OR: 0.573; 95% CI: 0.375–0.874); total bilirubin level (OR: 1.477; 95% CI: 1.068–2.043); diameter of pancreatic duct (OR: 1.837; 95% CI: 1.221–2.763) and diameter of tumor diameter (OR: 1.837; 95% CI: 1.221–2.763). CONCLUSIONS: [Formula: see text] This risk assessment formula can help estimate and predict postoperative morbidity rate after PD.