Su1542 DEVELOPMENT AND VALIDATION OF A SCORING SYSTEM FOR MORTALITY PREDICTION AFTER ENDOBILIARY STENT PLACEMENT IN PATIENTS WITH UNRESECTABLE MALIGNANT BILIARY OBSTRUCTION

Abstract

The prognosis of patients with malignant biliary obstruction (MBO) is generally poor due to advanced stage at the time of clinical presentation. Palliative biliary decompression with endobiliary stent placement is the mainstay treatment for these patients. Despite the improvement of biliary obstruction, the survival benefit after stent placement is arguable. This study aimed to identify factors predicting mortality, develop and validate mortality predictive score for unresectable MBO patients. Aretrospective study of ERCP database from 2007 to 2018. Five-hundred and eleven patients with unresectable MBO undergoing first-time ERCP-guided endobiliary stent placement were enrolled. We randomly divided patients in 3:1 fashion into a derivation cohort (n=383) and validation cohort (n=128)using computer generation. Variables that might influence 90-day mortality after ERCP-guided endobiliary stent were identified and included in the model development. The model was validated using data from the validation cohort. The baseline characteristics between the two groups were comparable. In the derivation cohort, the overall 90-day mortality rate was 22.2%. After multivariate analysis using Cox’s regression model, there were five predictors associated with 90-day mortality including liver metastasis (HR 3.10; 95% CI 1.66-5.80, p<0.001), pre-stenting serum alkaline phosphatase (ALP) of ≥3 times of upper normal limit (HR 3.02; 95% CI 1.42-6.40, p=0.004), pre-stenting serum albumin of < 3 g/dl (HR 2.32; 95% CI 1.26-4.31, p=0.008),peritoneal carcinomatosis (HR 2.23, 95% CI 1.06-4.71, p=0.035) andbilirubin resolution of ≥50% at 2 weeks after stent placement (HR 0.43; 95% CI 0.22-0.82, p=0.010) (Table 1). The scoring system was developed to identify patients at risk of death after endobiliary stenting within 90 days. Patients were categorized into three groups using the following cut-off point: risk score <1, risk score 1-2.5 (HR 3.55; 95% CI 1.66-7.62; probability of death 19%), and risk score ≥2.5 (HR 20.31; 95% CI 7.73-53.36; probability of death 66%). In the validation cohort, the risk score > 2.5 provides HR of 12.23 (95% CI 1.09-137.99) underlining the usability of this scoring system (Table 2). This studyproposes a useful and validated scoring system derived from pre-stenting serum albumin, ALP, bilirubin regression after stenting, liver metastasis and peritoneal carcinomatosis to predict 90-day mortality after ERCP with endobiliary stenting in patients with unresectable MBO. This score permits physicians to stratify the risk of mortality in these patients and may be helpful in endobiliary stent selection and providing proper palliative management.View Large Image Figure ViewerDownload Hi-res image Download (PPT)