ObjectiveTo retrospectively investigate the correlation between tumor immune nutritional indexes and the resectability in patients with pancreatic cancer.MethodsWe selected pancreatic patients with pathological diagnosis who admitted to Xuanwu Hospital of Capital Medical University from January 2015 to December 2018. The clinical data of patients were retrospectively analyzed. Nutritional and inflammatory hematological parameters at one week before operation were carefully collected, the parameters including: the neutrophil count, lymphocyte count, monocyte count, hemoglobin (Hb), platelet count, albumin (Alb), prealbumin (PA), cholesterol, and serum tumor markers (CEA and CA19-9). The ratio of neutrophil count to lymphocyte count (NLR), ratio of platelet count to lymphocyte count (PLR), ratio of lymphocyte count to monocyte count (LMR), prognostic nutrition index (PNI), nutritional risk score (GNIR), and controlled nutritional status score (COUNT) were calculated. The receiver working characteristic curve (ROC curve) was used to evaluate the predictive value of various indexes in radical resection of pancreatic cancer.ResultsOf the 55 patients with pancreatic cancer, 22 received radical surgery and 33 did not. There was no significant difference in gender, BMI, neutrophil count, monocyte count, platelet count, hemoglobin, albumin, prealbumin, cholesterol, and tumor location between the radical operation group and the non-radical operation group (P>0.05), but there were significant differences in age, lymphocyte count, CEA, and CA19-9 between the two groups (P<0.05). There was no significant difference in the area under the curve (AUC) of neutrophil count, lymphocyte count, monocyte count, hemoglobin, platelet count, albumin, prealbumin, cholesterol, NLR, PLR, LMR, PNI, and GNIR to predict the resectability of pancreatic cancer (P>0.05), but there was statistical significance in COUNT score, CEA, and CA19-9 (P<0.05). The AUC values of COUNT, CEA, and CA19-9 were 0.700, 0.705, and 0.739 respectively, the sensitivity corresponding to the best critical point cutoff value were 59.09%, 80.00%, and 100%, as well as the specificity were 87.88%, 66.67%, and 42.42%, respectively. The specificity of COUNT was high, but the sensitivity was poor. The sensitivity of CEA and CA19-9 were high and the specificity were poor.ConclusionsThe COUNT is a simple and useful predictor to predict the resectability of pancreatic cancer. The combination of COUNT and serum tumor markers of CEA and CA19-9 can help to better predict the surgical indications of pancreatic cancer.