Background/aims: This study was conducted to clarify the impact of pancreaticosplenectomy on the prognosis of patients with gastric carcinoma.
Methodology: Two hundred and seventy-two patients who underwent total gastrectomy with distal pancreatectomy and splenectomy were retrospectively reviewed.
Results: Lymph node metastases at the splenic hilum (#10) and along the splenic artery (#11) occurred in 12.4% and 19.2% of cases, respectively. The 5-year survival rate of those without metastasis at #10 was 62.8%. Once nodal metastasis occurred, the prognosis became very poor; only 18.2% in those with a single positive node and 15.4% of those with two or more positive nodes at this location survived 5 years. Similar trends in survival were observed with respect to nodes at #11. When stratified by nodal status as currently determined by microscopic examination, pancreaticosplenectomy saved 4.5% of patients with positive nodes, but was insufficient in 17.3% of cases and was not necessary in the 78.2% of cases who were node negative at these locations.
Conclusions: The data indicate that pancreaticosplenectomy can save some patients with positive nodes in these regions; however, the small survival benefit does not provide a basis for the general application of this highly morbid procedure. To further evaluate these results in a randomized study, selection of a subset of patients who are likely to have metastasis is the key.