Difficulties in the differential diagnosis of gastric carcinoma and gastritis; surgical considerations.

Abstract

My contribution to this symposium will appear to some pessimistic and negativistic. My thesis is brief. I say that the radiologist, yes, and even the gastroscopist as he popularly uses his instrument today, are unable to differentiate with sufficient accuracy between certain cases of gastritis and gastric neoplasms to warrant withholding surgical biopsy (excision) from patients with stomach disease falling into this differential classification. After critical analysis, many among us have found it impossible by study of the gross morphology of some gastric lesions (by either roentgen or gastroscopic methods, or even when handling and viewing the surgically removed specimen) to differentiate certain inflammatory diseases of the stomach from infiltrating types of cancer. Not being eminently qualified as an authority on this problem, I must draw upon the literature and statements of those men recognized in radiology and gastroscopy to reinforce my categorical assertion. B. R. Kirklin (23, 17) has said that there are cases in which he finds it impossible to distinguish between gastritis and carcinoma. H. W. Schmidt (22) has informed me that, in spite of what is being written in the literature today about the value of gastroscopy, it is often impossible for a gastroscopist to make a differentiation between severe gastritis and carcinoma until such a time as he can safely obtain tissue from the stomach. Moersch (1) reported an error of about 10 per cent in the gastroscopic differential diagnosis between certain types of gastritis and gastric cancer. Likewise, Moersch and Kirklin (2) found an experienced roentgenologist about 10 per cent in error when distinguishing certain gastric neoplasms from gastritis. In 6 to 10 per cent of later proved cases of gastritis and carcinoma, even the surgeon at exploration could not tell whether the gastric disease was benign or malignant, the final diagnosis depending upon microscopic study of excised tissue. Palmer et al. (3), presenting 6 cases of giant hypertrophic gastritis subjected to surgery, say that from the point of view of clinical diagnosis differentiation from carcinoma was not satisfactory either by gastroscopy or roentgenology. They also note that, although areas of mucosal polypoid hypertrophy are not common in diffuse giant hypertrophic gastritis, the gross separation of the latter from polypoid adenomata is difficult or impossible. And since some 20 per cent of gastric polyadenomata en nappe will have malignant changes in one or several of the polyps (3), a diagnosis of either would merit gastric resection. Many authors, as Vaughan (4), Arendt (5), Harris (6), Bank et al. (7), Rennie (8), Lyall and Leider (10), report one or more cases showing certain types of gastritis to be indistinguishable clinically from carcinoma.