The incidence of lymph node metastasis in early gastric cancer according to the expanded criteria in comparison with the absolute criteria of the Japanese Gastric Cancer Association: a systematic review of the literature and meta-analysis.

BACKGROUND AND AIMS: Japanese criteria for curative endoscopic resection of early gastric cancer initially included nonulcerated, well-differentiated mucosal lesions ≤2 cm in diameter, known as the absolute criteria. Subsequently, these indications were expanded to include larger, ulcerated, and undifferentiated mucosal lesions as well as differentiated lesions with slight submucosal invasion. Whether patients meeting the expanded criteria can be managed safely without gastrectomy and lymph node dissection has been controversial. The risk of lymph node metastasis (LNM) in patients who met the expanded criteria is a critical factor in determining the best course of management for these patients.

METHODS: We comprehensively searched main reference databases for studies that included patients who underwent gastrectomy and lymph node dissection for early gastric cancer. A meta-analysis was conducted by using the random effects model. Relative risk reduction was used to compare the incidence of LNM in patients meeting the absolute criteria as compared with those meeting the expanded criteria.

RESULTS: Twelve studies met the inclusion criteria, providing a total of 9798 patients. The incidence of LNM was 0.2% for patients who met the absolute criteria as compared with 0.7% for patients who met the expanded criteria. Analysis of the various components of the expanded criteria was conducted, revealing that the incidence of LNM for differentiated mucosal lesions ≤3 cm with ulceration and for differentiated mucosal lesions without ulceration, irrespective of size, was 16 of 2814 (0.57%), reference range (RR) 3.01; P = .02 and 8 of 3004 (0.27%), RR 1.69; P = .37, respectively, only marginally higher than the risk of LNM associated with the absolute criteria. In contrast, undifferentiated mucosal lesions ≤2 cm and differentiated lesions <3 cm with slight submucosal invasion had a significantly higher incidence of LNM in comparison with the absolute criteria (25/972 [2.6%], RR 6.79; P = .0004 and 8/315 [2.5%], RR 6.30; P = .004, respectively).

CONCLUSION: Overall, expanding the indication for endoscopic resection to include mucosal nonulcerated differentiated lesions irrespective of size and differentiated mucosal ulcerated lesions <3 cm is justified with minimal increased risk in comparison to the absolute criteria. However, expanding the indication for undifferentiated lesions ≤2 cm and differentiated lesions with slight submucosal invasion (T1b) should be balanced with the risks of surgery, given the increased risk of LNM in these patients.