The lymphatic drainage of diseased and normal bowel was studied in 21 patients undergoing surgery for Crohn’s disease. Mesenteric lymphatic obstruction was a consistent feature, identified in areas of small bowel macroscopically affected by Crohn’s disease. This finding was also observed in some areas of apparently unaffected small bowel. Subsequent examination of these areas confirmed the presence of early Crohn’s disease. This method of study may prove to be of value for determining the extent of operative resection in patients with Crohn’s disease. Experimental lymphatic obstruction in animals failed to produce pathological changes of Crohn’s disease, suggesting that this feature is an epiphenomenon and is not of primary importance in the pathogenesis of regional enteritis.

Cellular immunity was studied in three homogeneous groups of patients with cancer to determine whether the pattern of depression of immune competence varied between solid tumours with different patterns of clinical behaviour. Delayed hypersensitivity skin responses were measured in patients with carcinoma of the breast, stomach, and colon and matched controls. Response to 2,4-dinitrochlorobenzene (DNCB) was used as an indication of primary responses and the Mantoux reaction as an index of recall responses. Responses were diminished in all three cancer groups, but there were significant differences between each type of cancer and even between different control groups. Cellular immunity was lost earliest and to the greatest extent in patients with colonic cancers and tended to be retained until a late stage in breast cancer, with gastric cancer occupying an intermediate position. Thus, while there was some degree of correlation between depressed immunity and prognosis our results gave no evidence that general host immune competence could explain the worse prognosis of gastric than colonic cancer. Paradoxical findings in patients with breast cancer suggested a great complexity in the host tumour interaction. Assessments of immune competence in cancer patients must be related to specific types of neoplasms with appropriate control groups if the results are to be meaningful.

No abstract

No abstract