The prognostic impact of nodal involvement in resected pancreatic carcinoma and biliary malignancy has been relatively well established. It has been suggested that lymph node ratio (LNR) may be a more informative way of stratifying patients with node positive disease. Our retrospective review aimed to investigate the significance of such variables and test for independent prognostic factors for survival.
One hundred eighty-three pancreatic and periampullary malignancy cases were registered at the American University of Beirut Medical Center from 1990 to 2004. Of those, 80 had complete data on lymph node status. We analyzed the impact of the number of lymph nodes resected, the number of positive lymph nodes retrieved and LNR using Kaplan-Meier and Cox proportional hazard models. The measured outcome in the KM model was the survival probability at 1, 3, and 5 years while the Cox model was used to measure the hazard ratio (HR) of the previously identified predictors on survival.
For the 80 patients included in this analysis, overall survival rates were 65% (54 to 78), 32% (18 to 47), and 21% (8 to 34) were alive at 1, 3, and 5 years, respectively. The median number of resected lymph nodes was 9. In the node positive patients, those who had >12 nodes examined were found to have a significantly better survival (HR=0.24; P=0.013). On multivariate analysis, our model showed the following factors to be significant: age 60 years or older (HR=5.92; P=0.018), poorly differentiated tumors (HR=21.87; P=0.018), number of lymph nodes examined <12 LN (HR=6.77; P=0.022), 3 or more metastatic LN (HR=7.21; P=0.028), and LNR≥0.2 (HR=7.12; P=0.007).
After pancreaticodudonectomy for adenocarcinoma of the pancreas and biliary malignancies, ratio-based lymph node staging is an independent and powerful prognostic factor.