Background: The only possible curative treatment in gastric carcinoma is surgery, and there is still controversy surrounding the value of extended lymph node dissection.
Study design: A retrospective cohort study was conducted in patients who underwent curative D1 or D2 resection for operable gastric carcinoma. Survival and multivariate prognostic factor analyses were carried out to determine whether dissection type was significant for outcomes, and which subgroup of patients would benefit from D2 dissection.
Results: Three hundred one patients who had potentially curative treatment were eligible to enter the trial. Although mortality rates were 3.1% in the D1 group and 4.3% in the D2 group (p = NS), morbidity rates in the D1 and D2 groups were 6.2% and 27.9%, respectively (p<0.05). Multivariate analysis showed that lymph node dissection type, Borrmann type of tumor, number of metastatic lymph nodes, and depth of wall invasion were the most important independent prognosticators. Five-year disease-free and overall survival rates were 19% and 36% in D1, and 49% and 54% in D2, respectively (p<0.05). After stratifying for pT and pN, the significant survival advantage with D2 was observed in subgroups of pT2, pT3 and pN1, pN2. The subset analysis showed a significant prognostic benefit with D2 dissection in patients in stages II and III-A.
Conclusions: D2 dissections can be carried out with low mortality rates, but they have high morbidity rates and a survival advantage over D1 dissection of only 18%. In principle, a survival benefit with D2 is obtained especially when the tumor invades muscularis propria, penetrates serosa without invasion of adjacent structures, or metastasizes to fewer than fifteen regional lymph nodes. Data in this homogeneous population support the use of extended lymphadenectomy for selected group of patients with gastric carcinoma.