Thirty elderly long-stay patients were randomly allocated to receive either placebo or dietary supplementation with vitamins A, C and E for 28 days. Nutritional status and cell-mediated immune function were assessed before and after the period of supplementation. Following vitamin supplementation, cell-mediated immune function improved as indicated by a significant increase in the absolute number of T cells (p less than 0.05), T4 subsets (p less than 0.05), T4 to T8 ratio (p less than 0.01) and the proliferation of lymphocytes in response to phytohaemagglutinin (p less than 0.01). In contrast, no significant changes were noted in the immune function of the placebo group. We conclude that supplementation with the dietary antioxidants vitamins A, C and E can improve aspects of cell-mediated immune function in elderly long-stay patients.

I . The effect of pectin on the structure and function of the rat small intestine was compared with that of a standard pellet diet and of a fibre-free basal diet.2. The length and wet weight of the small bowel was significantly greater inpect in-fed rats than in either pellet-or basal-diet-fed rats.3. Histological measurements of longitudinal sections from the small bowel showed a significantly greater crypt depth and muscle layer thickness in the mid-jejunum and ileum of the pectin fed rats. Villous height showed less variation. 4.The specific activity of alkaline phosphatase (EC 3 . I . 3 . I ) and leucyl-P-naphthylamidase (EC 3.4. I I . I ) in mucosal scrapings was significantly lower in the upper jejunum of pectin-fed rats compared with either of the other dietary groups. The differences were not so marked in mid-jejunum or ileum.5 . Glucose absorption measured in vivo from jejunal and ileal loops was similar in all three dietary groups.6. With two minor exceptions there were no significant differences in any of these measurements between the pellet-and basal-diet-fed rats.7. These findings could be explained by increased epithelial cell turnover caused by pectin. The possible mechanisms of this are discussed.8. The effect of pectin on the human small bowel requires study before it can be widely prescribed in man.There is increasing interest in both the physiological and pharmacological effects of dietary fibre. Pectin, a partially methoxylated polymer of galacturonic acid is one form of dietary fibre attracting particular attention, as it seems to have the most consistent hypocholesterolaemic effect of all forms of fibre so far tested in man (Truswell, 1977). How pectin acts as a hypocholesterolaemic agent is unclear, but it may be by interfering with cholesterol absorption (Leveille & Sauberlich, 1966). Subjects taking pectin show an increase in their faecal neutral steroid excretion as well as increased bile acid and faecal fat excretion (Kay & Truswell, 1977). If pectin, or other similar plant materials, are to be used pharmacologically, then more information is required about their effects both on the bowel itself and on the absorption of nutrients and drugs.Pectin-fed rats excrete an oral dose of paracetamol more rapidly than rats on a fibrefree diet, the difference being due to more rapid absorption of paracetamol from the small bowel (Brown et al. 1979). The present paper describes studies of the small intestine of rats fed on a pectin diet compared with rats given either a fibre-free diet or a standard pelleted diet. Changes in intestinal anatomy have been quantitated, functional changes have been assessed by measuring in vivo glucose absorption from small bowel loops and biochemical changes have been assessed by measuring mucosal protein and DNA, and the activity of two brush border enzymes, alkaline phosphatase (EC 3 . I . 3 . I) and leucyl-P-naphthylamidase (EC 3.4. I I . I). MATERIALS A N D METHODS AnimaZsFemale Wistar rats, initial weight 175-220 g were used.

A tubeless pancreatic function test (BTP test) using N-benzoyl-l-tyrosyl-p-aminobenzoic acid was used to assess exocrine function from urinary recovery of p-aminobenzoic acid produced by hydrolysis of the peptide by chymotrypsin. Patients with acute pancreatitis were studied at various time intervals after the acute attack and compared with controls with abdominal pain that was not pancreatic in origin. The initial BTP test carried out in the convalescent period was abnormal in all of 30 patients with acute pancreatitis but normal in 10 patients with non-pancreatic abdominal pain and also in 8 patients who had recovered from an attack of acute pancreatitis 2-6 years previously. Results were improved or normal in 12 of 15 patients re-tested 1 year after the attack of pancreatitis, but 4 out of 6 still had abnormal results 2-6 months after the attack. The data suggest that an attack of acute pancreatitis may impair exocrine pancreatic function for several months.

In 100 patients with cystic fibrosis the severity of steatorrhoea was assessed by three separate methods. Using chemical faecal fat assay as the gold standard, two other rapid and inexpensive methods were compared with it. The steatocrit method proved unreliable in our hands and gave little indication of the presence or severity of steatorrhoea. The more simple microscopy method was highly sensitive (97%) and only three of 80 patients with steatorrhoea were missed. All patients with severe steatorrhoea (>60 mmol fat/day) were clearly demonstrated. The method is applicable to spot faecal samples and can readily be carried out on an outpatient basis. In centres where faecal fat assays are not available, the simple and cheap microscopic examination will give some indication of the response to enzyme treatment and may also help to identify non-compliant individuals.