Background and Aims: Gallbladder carcinoma (GBC) is a malignant tumor with a high mortality rate. The lack in quick and effective methods of preoperative diagnosis and prognostic assessment has increased the difficulty of operation and follow-up treatment for GBC. In recent years, the relations of the peripheral blood inflammation-related indicators with tumor prognosis have been extensively studied, but the predictive values are often limited due to the instability of the peripheral blood parameters. The purpose of this study was to investigate the role of the new peripheral blood parameter models preoperative plus postoperative neutrophil-lymphocyte ratio (PP-NLR) and Glasgow prognosis score (GPS) in prognostic assessment of GBC. 
Methods: The clinical data of 140 patients with GBC treated from January 2005 to December 2015 were retrospectively analyzed. The cut-off values of NLR before and after operation were determined by ROC curve, based which, the NLR value was assigned as 1 if it was increased, or as 0 if not, and the PP-NLR value was defined as the sum of the two assigned values and was 0, 1 and 2, respectively; the preoperative serum albumin <35 g/L and preoperative CRP >10 g/L were assigned as 1 respectively and were assigned as 0 if not, and GPS value was defined as the sum of the two assigned values and was 0, 1 and 2, respectively. The relations of PP-NLR and GPS with the prognosis and clinicopathologic factors of the patients were analyzed by Kaplan-Meier method, Log-rank test and univariate analysis, respectively. The correlations of PP-NLR and GPS with the clinicopathologic factors were determined by Spearman correlation analysis. The independent prognostic factors were determined by multivariate Cox hazard model. 
Results: The cut-off values of NLR before and after operation determined by ROC curve were 2.51 (sensitivity: 0.961, specificity: 0.788) and 2.38 (sensitivity: 0.745, specificity: 0.712). Survival analysis showed that the survival rates were significantly different among patients with different PP-NLR and GPS levels (all P<0.05), which presented a successive decrease in PP-NLR=1, PP-NLR=2 and PP-NLR=3 group, and was significantly higher in GPS=0 group than those in GPS=1 group or GPS=2 group (both P<0.05), but had no significant difference between GPS=1 group and GPS=2 group (P>0.05). Univariate analysis suggested that both PP-NLR and GPS were related to the radical resection rate, tumor invasion, lymph node or distal metastases, TNM classification and degree of differentiation as well as the inflammatory indexes and tumor markers (all P<0.05). The correlation analysis indicated that PP-NLR and GPS were significantly correlated to the radical resection rate, tumor invasion, lymph node or distal metastases, TNM classification and degree of differentiation (all P<0.05). Univariate analysis demonstrated that both the increased PP-NLR and GPS were significantly associated with low survival rate (both P<0.05), and multivariate analysis revealed that PP-NLR was an independent risk factor affecting the prognosis of the patients (PP-NLR=1: HR=0.357, 95% CI=0.221–0.575, P<0.05; PP-NLR=2: HR=0.357, 95% CI=0.221–0.575, P<0.05). 
Conclusion: Both PP-NLR and GPS are related to the prognosis of GBC patients, and PP-NLR is an independent prognostic factor, suggesting that peripheral blood parameters PP-NLR and GPS can easily, quickly and effectively assess the prognosis of the patient. In addition, PP-NLR integrates the pre- and postoperative systemic inflammation and immune status, so it is more comprehensive and reliable that that GPS and preoperative or postoperative NLR alone for prediction, and can provide a theoretical basis for follow-up treatment.