D H Roukos1, P Paraschou, M Lorenz. 1. Department of Surgery, University Hospital of Frankfurt, Frankfurt am Main, Germany. roukos@hol.gr
Abstract
BACKGROUND: Curative resection (R0) is the treatment of choice for distal gastric cancer, but it is unclear whether this operation should include a total gastrectomy (TG) with splenectomy and extended (D2) lymph node dissection. A new concept was developed based on the fact that residual metastatic lymph nodes after a limited (D1) subtotal gastrectomy (SG) may be the source of fatal relapse. We conducted a prospective study on patients who had undergone a D2 TG to evaluate whether certain stations left behind after a D1 SG contain metastasis. METHODS: We studied 1207 nodes obtained from 35 eligible patients who underwent a TG within 2 years. Of these patients, 29 fulfilled the criterion for a D2 dissection with curative potential. Numbers of retrieved and tumor-containing nodes by each station according to the Japanese Research Society for Gastric Cancer were documented prospectively in a standardized protocol. All lymph nodes were studied in sections smaller than 2 mm, but emphasis was given to the study of nodes from stations 1 and 2 (paracardial right and left), station 10 (splenic hilum), and stations 7 through 12 (around celiac axis, and in hepatoduodenal ligament) that can be dissected with a TG, splenectomy, and D2 dissection, respectively. For quality control of D2 dissection, the numbers of nodes retrieved by each compartment II nodal station (7-12) documented by a pathologist were used and compared with proposed reference values. Long-term survival and cumulative risk of relapse were calculated in terms of lymph node status and presence of metastasis in compartment II nodes. RESULTS: A mean total node yield of 37.4 from stations 1-12 and 11.4 from compartment II (stations 7-12) was obtained from 29 patients who had a D2 TG with curative intent. A substantial variation in node yields was found, and sometimes several stations contained no lymph nodes, which suggested an important cause of noncompliance (no yield of lymph nodes detected by the pathologist from that indicated for dissection stations) and difficulties for quality control. No positive node was detected in stations 1, 2, and 10 among patients who had a curative TG with splenectomy. However, substantially high was the incidence of metastasis in compartment II nodes, which was detected in one third of patients with node-positive disease. After 10 years of follow-up, overall survival and relapse rates among R0 D2 patients with negative compartment II nodes (pN0/pN1 disease) were 47% and 44%, respectively. CONCLUSIONS: Our results suggest the necessity of D2 dissection, but not of TG with splenectomy, to achieve an R0 resection for patients with distal gastric carcinoma. A large prospective study based on our protocol and findings may clarify whether a D2 R0 resection would result in a survival benefit.
BACKGROUND: Curative resection (R0) is the treatment of choice for distal gastric cancer, but it is unclear whether this operation should include a total gastrectomy (TG) with splenectomy and extended (D2) lymph node dissection. A new concept was developed based on the fact that residual metastatic lymph nodes after a limited (D1) subtotal gastrectomy (SG) may be the source of fatal relapse. We conducted a prospective study on patients who had undergone a D2 TG to evaluate whether certain stations left behind after a D1 SG contain metastasis. METHODS: We studied 1207 nodes obtained from 35 eligible patients who underwent a TG within 2 years. Of these patients, 29 fulfilled the criterion for a D2 dissection with curative potential. Numbers of retrieved and tumor-containing nodes by each station according to the Japanese Research Society for Gastric Cancer were documented prospectively in a standardized protocol. All lymph nodes were studied in sections smaller than 2 mm, but emphasis was given to the study of nodes from stations 1 and 2 (paracardial right and left), station 10 (splenic hilum), and stations 7 through 12 (around celiac axis, and in hepatoduodenal ligament) that can be dissected with a TG, splenectomy, and D2 dissection, respectively. For quality control of D2 dissection, the numbers of nodes retrieved by each compartment II nodal station (7-12) documented by a pathologist were used and compared with proposed reference values. Long-term survival and cumulative risk of relapse were calculated in terms of lymph node status and presence of metastasis in compartment II nodes. RESULTS: A mean total node yield of 37.4 from stations 1-12 and 11.4 from compartment II (stations 7-12) was obtained from 29 patients who had a D2 TG with curative intent. A substantial variation in node yields was found, and sometimes several stations contained no lymph nodes, which suggested an important cause of noncompliance (no yield of lymph nodes detected by the pathologist from that indicated for dissection stations) and difficulties for quality control. No positive node was detected in stations 1, 2, and 10 among patients who had a curative TG with splenectomy. However, substantially high was the incidence of metastasis in compartment II nodes, which was detected in one third of patients with node-positive disease. After 10 years of follow-up, overall survival and relapse rates among R0 D2 patients with negative compartment II nodes (pN0/pN1 disease) were 47% and 44%, respectively. CONCLUSIONS: Our results suggest the necessity of D2 dissection, but not of TG with splenectomy, to achieve an R0 resection for patients with distal gastric carcinoma. A large prospective study based on our protocol and findings may clarify whether a D2 R0 resection would result in a survival benefit.