Colonic tumour is a general designation of the colonic lesion featured by its protrusion into the bowel lumen, regardless of its benign or malignant nature ; most of the colonic tumours are the so-called benign polyp, which is said to have a close relation to carcinoma on one hand, and is sometimes most difficult to tell from carcinoma, on the other. The purpose of this paper is to find reliable criteria for differentiation between benign polyp and polypoid carcinoma of the colon.1. Analysis of endoscopic observations The materials are 73 cases with 106 lesions, in total, of endoscopically benign, colonic tumours without ulceration, adherence of blood clot or necrotic material detected by colonic fibrescopy (with CF-SB, CF-MB and CF-LB, Olympus) at the First Department of Internal Medicine, University of Tokyo and at the Department of Medicine, Tokyo Medical College, until May 1972 and at the Medical Department of Luisen Hospital, Aachen, at the Department of Gastroenterology, Leiden University, and at the Department of Medicine, Laennec Hospital Paris from May to August 1970; those casess of the latter 3 institutes are the personal experiences of Niwa, one of the authers. Of the above mentioned lesions, 75 lesions in 61 cases, in which sufficient histological specimens were taken by direct vision biopsy, were selected for the study of relation between size and shape of protrusion, on one hand, and histological diagnosis on the other. The shape of protrusion was classified as sessile, semi-pedunculated and pedunculated: The size was assessed by comparison with the opened biopsy forceps, measuring ca. 5 mm in length, and graded as about 0.5 cm, 1 cm and 2 cm in diametre. The incidence of carcinoma was significantly higher (p<0.01) in the semi-pedunculated tumours (4/31) than in the sessile ones (0/36) ; the differences between pedunculated (1/8) and sessile, and between pedunculated and semi-pedunculated, were insignificant. Comparison between size and incidence of carcinoma revealed a higher incidence in 2.0 cm group (2/8) than in 0.5 cm group (1/44) with some statistical significance (p=0.058), but the differences in the other combinations were insignificant. With the 46 lesions, which were well photographed, surface appearance and colour were studied by the photographs with reference to incidence of malignancy. The incidence of carcinoma was significantly higher (p<0.01), in uneven surface (4/10) than in even surface (1/36) ; the colour of the surface was divided into three categories, id., red, normal and pale, but the difference in carcinoma incidence was insignificant among them.2. Study of surgical materials Hundred and eighty-six colonic tumours seen in the colectomy specimens at the University of Tokyo Hospital from January 1967 to December 1971, and at the National Cancer Centre until November 1971 were studied for the relation of shape and size to histological malignancy. Concerning the shape, incidence of malignancy was significantly higher (p<0.01) in the semi-pedunculated tumours (9/23) than in sessile (11/81) or pedunculated (11/82). Size was closely related to malignancy (p<0.001) . The larger the tumours, the higher the incidence : carcinoma was seen in 1/97 tumours less than 1 cm in diametre, 10/56. 1 to 1.9 cm, and 20/33 larger than 2cm. Expected occurrance rate of carcinoma in each group (with the reliability coefficient of 0.90) was estimated as 0.05 to 4.8% in the tumours less 1 cm in diametre, 10 to 29% in the tumours 1 to 1.9 cm, and 44 to 75% in the tumours larger than 2cm. Summary and Conclusion Ulceration, bleeding, and adherence of necrotic materials over the tumour are generally accepted as suggestive evidences of malignancy. The present study revealed the presence of malignancy even without those findings. Other findings were also found by the present study to suggest malignancy : semi-pedunculated tumour, rough surface and large size were all associated with malignancy. Even with that
抄録全体を表示