BACKGROUND: This study evaluated the predictive ability of preoperative carbohydrate antigen 19-9 (CA19-9) level in assessing tumour resectability in patients with gallbladder carcinoma (GBC). METHODS: We retrospectively analysed preoperative serum levels of CA19-9 in 292 patients with potentially resectable GBC between January 2000 and March 2016 in our institution. All final diagnoses were confirmed by pathological examination. The optimal cut-off point of the CA19-9 for predicting resectability was determined by the receiver operating characteristic curve. The univariate analysis and multivariate Cox regression model were applied to assess the relationship between the parameters to resectability. RESULTS: A total of 292 patients with surgical treatment were included, of whom 195 were identified as curative resection (R0 resection), 69 were classified as R1/2 resection and the remaining 28 patients were operated on with palliative surgery. The receiver operating characteristic curve analysis calculated the best CA19-9 cut-off point of 98.91 U/mL in the prediction of resectability. Meanwhile, the sensitivity, specificity, positive predictive value and negative predictive value were 76.3%, 70.8%, 85.7% and 56.5%, respectively. In the multivariate logistic regression analysis, CA19-9 >98.91 U/mL (odds ratio (OR) 6.339, 95% confidence interval (CI) 3.562-11.284, P < 0.001), tumour located on hepatic side (OR 1.787, 95% CI 1.022-3.123, P = 0.042) and advanced American Joint Committee on Cancer stage (OR 2.156, 95% CI 1.180-3.940, P = 0.013) were independent determinants of resectability in patients diagnosed as GBC. CONCLUSION: Preoperative CA19-9 predicts resectability in patients with radiological resectable GBC. Increased preoperative CA19-9 is related to poor resectability rate.
BACKGROUND: This study evaluated the predictive ability of preoperative carbohydrate antigen 19-9 (CA19-9) level in assessing tumour resectability in patients with gallbladder carcinoma (GBC). METHODS: We retrospectively analysed preoperative serum levels of CA19-9 in 292 patients with potentially resectable GBC between January 2000 and March 2016 in our institution. All final diagnoses were confirmed by pathological examination. The optimal cut-off point of the CA19-9 for predicting resectability was determined by the receiver operating characteristic curve. The univariate analysis and multivariate Cox regression model were applied to assess the relationship between the parameters to resectability. RESULTS: A total of 292 patients with surgical treatment were included, of whom 195 were identified as curative resection (R0 resection), 69 were classified as R1/2 resection and the remaining 28 patients were operated on with palliative surgery. The receiver operating characteristic curve analysis calculated the best CA19-9 cut-off point of 98.91 U/mL in the prediction of resectability. Meanwhile, the sensitivity, specificity, positive predictive value and negative predictive value were 76.3%, 70.8%, 85.7% and 56.5%, respectively. In the multivariate logistic regression analysis, CA19-9 >98.91 U/mL (odds ratio (OR) 6.339, 95% confidence interval (CI) 3.562-11.284, P < 0.001), tumour located on hepatic side (OR 1.787, 95% CI 1.022-3.123, P = 0.042) and advanced American Joint Committee on Cancer stage (OR 2.156, 95% CI 1.180-3.940, P = 0.013) were independent determinants of resectability in patients diagnosed as GBC. CONCLUSION: Preoperative CA19-9 predicts resectability in patients with radiological resectable GBC. Increased preoperative CA19-9 is related to poor resectability rate.