Letters to the Editor

To the Editor:

We have read with great interest the paper written by Siewert et al ¹ concerning the prognostic factors in gastric cancer. The study prospectively shows that in a large patient population, D2 extended lymphadenectomy can improve survival in gastric cancer patients without affecting the incidence of postoperative complications. Moreover the so-called phenomenon of “stage migration” has been satisfactorily assessed. However, two methodologic aspects of the study should be further stressed.

First, the authors state that “the technique of lymph node dissection was performed according to the recommendations of the JRSGC ² … en bloc resection of the stomach with lymph node dissection of compartments I and II was recommended as the procedure of choice. Compartment I comprises all lymph nodes along the major and minor curvature of the stomach (i.e., lymph node stations 1–6 in those undergoing a total gastrectomy and lymph node stations 3–6 in those undergoing a subtotal gastrectomy). Compartment II comprises lymph node stations 7 to 12 in the Japanese classification.” According to the Japanese rules, for tumor located at proximal and middle third of the stomach, lymph nodes along the splenic artery (n.11) and at the splenic hilus (n.10) belong to the compartment II and therefore must be dissected to obtain a complete D2 lymphadenectomy. It is well known that to achieve such a dissection, it is necessary to perform a concomitant splenectomy. ^3–5 However, if we look at the data, the overall splenectomy rate reported in the paper was 492 cases, whereas the number of tumors located at proximal, middle or entire stomach undergoing extended lymph node dissection was 836. Therefore, even we assume that all the splenectomized patients belong to the extended lymph node dissection group, at least 41% of patients did not undergo a complete D2 dissection, thus leading to a possible misunderstanding of the results.

Second, the cutoff of 25 nodes could not be really effective in differentiating the extent of lymphadenectomy; in actual fact, the mean number of dissected perigastric (n.3, 4), pericardial (n.1, 2), and supra- and infrapiloric nodes (n.5, 6) in patients undergoing total gastrectomy at our Institute during the last year is 36.3 ± 12.8. ⁶ Therefore, even a D1 lymphadenectomy can lead to a quite large number of dissected nodes. D2 lymphadenectomy can be defined as complete only if all the stations of the second compartment, according to tumor location, have been dissected; the absolute number of dissected nodes is the result and not the limit of dissection.

Despite these two methodologic criticisms, however, the authors should be congratulated because they have definitively demonstrated that extended lymphadenectomy in gastric cancer can be safely and efficaciously performed even by Western surgeons in Western patients.

December 10, 1998

F. Pacelli MD
G. B. Doglietto MD
P. Caprino MD