|| Checking for direct PDF access through Ovid
The aim of this study was to evaluate the prognostic value of nodal involvement in resected adenocarcinoma of the pancreatic head.For the period between 1980 and 2002, 96 patients underwent pancreaticoduodenectomy for pancreatic cancer. Lymph nodes were numbered and classified into groups according to the Japan Pancreatic Society rules. Metastatic lymph nodes were identified based on hematoxylin and eosin staining.Sixty-four (66.7%) patients had positive lymph nodes. The median number of metastatic nodes was 2 (95% confidence interval [CI], 1.0-3.0) and the median ratio of metastatic/resected nodes was 9.7% (95% CI, 7.1%-14.4%). The median survival was 14.2 months (95% CI, 10.7-17.7) and was significantly higher for node-negative than node-positive patients (27.9; 95% CI, 20.9-34.9 vs. 10.6; 95% CI, 8.7-12.5; P < 0.001). The Cox proportional hazards model, including all patients, demonstrated that nodal involvement (hazard ratio [HR], 1.461; 95% CI, 1.177-12.024), moderate or poor tumor differentiation (HR, 2.330; 95% CI, 1.181-6.949), and positive resection margins (HR, 3.838; 95% CI, 1.390-10.597) were independent negative prognostic factors. If the analysis was limited to node-positive patients, lymph node ratio of more than 20% (HR, 1.364; 95% CI, 1.116-2.599), moderate or poor tumor differentiation (HR, 3.393; 95% CI, 1.041-11.061), and positive resection margins (HR, 9.400; 95% CI, 2.235-39.536) significantly correlated with a poorer survival.Lymph node ratio seems to be a new promising prognostic factor in patients with respectable node-positive pancreatic head cancer.