Safety of a paraaortic node dissection for selected advanced gastric cancer patients.

Background/aims: Comprehensive evaluations of a superextended (D3) dissection concerning its benefits as well as potential disadvantages have been scanty in the English literature.

Methodology: The patient selection criteria for a D3 dissection were pre- and intraoperatively > or = T3 and/or > or = N2 diseases, no paraaortic node involvement, no distant or peritoneal metastases, and a negative peritoneal lavage cytology. A D3 dissection involved a node clearance up to the third tier including the middle paraaortic region. Twenty-seven patients were entered into our D3 dissection strategy in a prospective manner between 1997 and 2000, and were divided into 2 groups: pathologically meeting the D3 criteria (D3 criteria fit; 12 patients) and pathologically proved to have less advanced disease (less advanced/D3; 15 patients). Data from age-matched 53 patients who underwent a D2 dissection between 1991 and 1996 were collected and also divided according to the current D3 criteria. Surgical invasiveness, morbidity and mortality, nutritional parameters, and survival were compared between D2 and D3 patients. Additionally, positive paraaortic node patients with a D3 dissection (4 patients) were used for preliminary survival comparison.

Results: A D3 dissection resulted in longer surgery and more blood loss, but morbidity and mortality, and nutritional impairment were similar to those of a D2 dissection. A more promising survival rate by a D3 dissection over a D2 dissection was observed only in the D3 fit patients but not in the less advanced disease patients. On the contrary, survivals of the positive paraaortic node patients were pessimistic despite the R0 resection.

Conclusions: A D3 dissection is an invasive procedure but can be performed as safely as a D2 dissection. Our results provide a starting point for a D3 challenge; however, paraaortic node positive patients should be excluded from a D3 dissection.