Reply to Baiocchi et al.

We are grateful for the comments and discussion provided by Baiocchi et al. with regard to the treatment of type 1 gastric carcinoid tumors. Rappel et al. [1] and Soga [2] have reported metastasis in a few patients with gastric carcinoids associated with type A gastritis. Soga [2] also showed that the rate of metastasis is directly related to the size and depth of the tumors. On the basis of these reports, we believe that endoscopic mucosal resection (EMR) can be used to treat type 1 carcinoids associated with hypergastrinemia in patients who do not have more than three to five tumors measuring 1 cm or less at their widest diameter. After EMR, we carry out a histopathological assessment of the resected specimens and recommend additional surgery in patients who have invasion of the muscularis propria, vascular invasion, or residual tumor.

The case described by Dr. Baiocchi was a single tumor measuring about 2 cm in diameter. In our guidelines, surgery would be indicated for this tumor. Polypectomy was carried out, but effective submucosal resection is not possible with this method. Since most gastric carcinoids develop primarily in the submucosa, EMR would have been a more suitable treatment, as it allows the submucosal tissue to be resected. At our hospital, we carry out surgery in patients who have carcinoids measuring 1 cm or more in diameter or who have more than three to five carcinoids.

Hosokawa et al. [3] carried out a follow-up study of type 1 carcinoids (all 1 cm or less in size) that were left untreated. The patients were regularly examined using endoscopy, measurement of serum gastrin levels, and abdominal computed tomography. The patients had no evidence of tumor progression, and there were some in whom the carcinoid tumors disappeared. These findings suggest that some patients may not require prompt treatment. As stated above, however, since the risk of metastasis is directly proportional to tumor size and depth, EMR performed while the tumors are still small is useful both for diagnosis and treatment.

Our series included only five patients, in all of whom the lesions were asymptomatic; the carcinoid tumors were detected during screening examinations. The tumors were 1 cm or less in diameter in four patients. In patient 2, whose tumor measured 10 × 12 mm, surgery was carried out because residual tumor was suspected on evaluation after EMR.

Some cases of gastric carcinoid cannot be clearly classified into the three types proposed by Rindi. In our series, patients 1, 4, and 5 had lesions that were distinctly type 1. Patients 2 and 3 also had lesions approximating to type 1, because of their association with hypergastrinemia. Retrospectively, we noted that treatment was carried out in accordance with the guidelines proposed by Gilligan et al. [4].

We conclude that EMR is indicated for the treatment of patients with three to five type 1 carcinoid tumors measuring 1 cm or less at their longest diameter. Histopathological assessment of the resected specimens should be carried out and the patients should receive close follow-up observation. Patients who do not satisfy the criteria for EMR are candidates for surgical therapy.

References

• 1 Rappel S, Altendorf-Hofmann A, Stole M. et al . Prognosis of gastric carcinoid tumors.  Digestion. 1995;  56 455-462
• 2 Soga J. Gastric carcinoids: a statistical evaluation of 1094 cases collected from the literature.  Surg Today Jpn J Surg. 1997;  27 892-901
• 3 Hosokawa O, Kaizaki Y, Watanabe K. et al . Endoscopic surveillance of gastric carcinoid tumors associated with gastritis type A.  Stomach and Intestine. 2000;  35 1395-1404
• 4 Gilligan C J, Lawton G P, Tang L H. et al . Gastric carcinoid tumors: the biology and therapy of an enigmatic and controversial lesion.  Am J Gastroenterol. 1995;  90 338-352