A B S T R A C T Using a triple-lumen constant perfusion system, the following observations were made in normal subjects. First, chloride, bicarbonate, and sodium were found to exhibit net movement across ileal mucosa against electrochemical gradients. Second, during perfusion with a balanced electrolyte solution simulating plasma, the ileum generally absorbed, but sometimes secreted fluid. A reciprocal net movement of chloride and bicarbonate was noted when sodium movement was zero. Increasing rates of sodium absorption were associated with decreasing bicarbonate secretion rates and finally bicarbonate absorption. Even when bicarbonate was absorbed ileal contents were alkalinized (by contraction of luminal volume). Third, net chloride movement was found to be sensitive to bicarbonate concentration in ileal fluid. For instance, chloride was absorbed from solutions containing 14 or 44 mEq/liter of bicarbonate, but was secreted when ileal fluid contained 87 mEq/liter of bicarbonate. Fourth, when chloridefree (sulfate) solutions were infused, the ileum absorbed sodium bicarbonate and the ileal contents were acidified. Fifth, when plasma-like solutions were infused, the potential difference (PD) between skin and ileal lumen was near zero and did not change when chloride was replaced by sulfate in the perfusion solution.These results suggest that ileal electrolyte transport occurs via a simultaneous double exchange, Cl/HCOs and Na/H. In this model neither the anion nor the cation exchange causes net ion movement; net movement results from the chemical reaction between hydrogen and bicarbonate. No other unitary model explains all of the following observations: (a) human ileal transport in vivo is essentially nonelectrogenic even though Na, Cl, and HCO3 are transported against electrochemical Dr. Turnberg's present address is Manchester Royal Infirmary, Oxford Road, Manchester 13, England.

Using a constant perfusion technique, sodium and bicarbonate absorption was studied in human subjects. The following observations were made on sodium absorption from saline solution: (a) the rate of sodium absorption is markedly influenced by bulk water flow, (b) when net water flow is zero, sodium absorption is zero if there are no concentration gradients between plasma and lumen that favor net NaCl diffusion; and (c) the PD between abraded skin and jejunal lumen is near zero when saline is perfused and does not change with partial substitution of sulfate or bicarbonate for chloride. Based on these observations, we conclude that sodium absorption from saline is entirely passive in the human jejunum. On the other hand, in the presence of bicarbonate sodium is absorbed actively against electrochemical gradients. The mechanism of the link between bicarbonate and sodium absorption was studied in normal subjects and in 11 patients with pernicious anemia; the latter were chosen because they do not secrete gastric acid which can react with bicarbonate in the jejunal lumen. We observed that bicarbonate absorption (a) occurs against steep electrochemical gradients, (b) does not generate a potential difference between abraded skin and jejunal lumen, (c) is inhibited by acetazolamide, and (d) generates a high CO2 tension in jejunal fluid. These observations suggest that bicarbonate absorption is mediated by active hydrogen secretion, rather than by bicarbonate ion transport per se, and that the link between sodium and bicarbonate transport is best explained by a sodium-hydrogen exchange process.

SUMMARY Oral formulations of 5-aminosalicylic acid (mesalazine) appear less toxic than sulphasalazine. We have therefore compared sulphasalazine, low dose mesalazine and high dose mesalazine in the treatment of mild to moderate relapse of ulcerative colitis. Sixty one patients (32 men, aged 20-78 years) were randomly allocated to sulphasalazine 2 g daily, mesalazine 800 mg daily, or mesalazine 2.4 g daily in a double blind, double dummy, four week trial. Groups were comparable for age, sex, extent of disease, and pretrial sulphasalazine intake. Four patients were unable to complete the study because of treatment failure (two taking sulphasalazine and two high dose mesalazine). A further two patients taking sulphasalazine developed side effects necessitating withdrawal. Within treatment comparisons revealed significant improvement of: sigmoidoscopic grade in the sulphasalazine group; rectal bleeding, sigmoidoscopic and histological grade in the low dose mesalazine group; stool frequency, rectal bleeding and sigmoidoscopic grade in the high dose mesalazine group. Greater improvement in rectal bleeding (p<005) and sigmoidoscopic appearances (p<005) occurred in patients taking high dose mesalazine than in those taking sulphasalazine. In two patients taking high dose mesalazine minor rises of plasma creatinine concentrations occurred, suggesting the need to monitor renal function.Although the main value of sulphasalazine (SSZ) is in the maintenance of ulcerative colitis remission'-3 it is also of benefit in active disease.' Many patients, however, are unable to take the drug as side effects are common, particularly at high doses.' As most of these side effects are similar to those seen with sulphonamide drugs and as they correlate well with plasma sulphapyridine concentrations it seems likely that the sulphapyridine component of SSZ is responsible for much of the drug's toxicity.' Therapeutic activity, on the other hand, seems to reside in the 5-aminosalicylic acid (5-ASA) component of SSZ.`'4Unfortunately, 5-ASA is unstable in gastric acid and is rapidly absorbed from the small intestine.`As

No abstract

No abstract

No abstract

No abstract

SUMMARY We investigated the possibility that loperamide might influence absorption and secretion in the human jejunum in vivo. Using a triple lumen tube perfusion technique in healthy normal volunteers we showed that loperamide did not affect net absorption of water or electrolytes under basal condition. When secretion was induced by prostaglandin E2, however, loperamide significantly reduced that secretion and in three out of six subjects secretion was abolished. Loperamide was effective when it was given either before or after secretion had been initiated. The results lend support to the suggestion that the antidiarrhoeal activities of loperamide may include an antisecretory effect. screening. The mixing segment was 15 cm long, the test segment 30 cm and the infusion rate 10 ml/min. In basal studies the equilibration period was 40 minutes and the test period one hour. In experiments with prostaglandin E2 (PGE2) the equilibration period was 30 minutes and the test period 40 minutes. During test periods jejunal aspirate from the proximal end of the test segment was collected by suction with a hand held syringe at a rate of 1.5 ml/min. Aspirates from the distal end of the segment were collected by continuous low grade suction. At 15 minute intervals small samples of aspirate were withdrawn and used to measure pH and pCO2 for the calculation of bicarbonate concentration.Two perfusion solutions were used, one contained bicarbonate (Na, 135; K,5; Cl, 105; HCO3, 35; mmolI1) and the other was bicarbonate-free (Na, 135; K,5; Cl, 140; mmol/l). Both perfusates contained polyethylene glycol 4000 (PEG) 2 gm/l and 14C PEG 0 5 ,uCi/l as a non-absorbable marker.Loperamide solution was administered intraluminally as a bolus of 4 mg or 8 mg followed half an hour later by perfusion with a solution containing loperamide (3 mg/l or 6 mg/l). In control periods the same volumes of a placebo solution were given. PGE2 was administered intraluminally by adding it to the perfusate in a final concentration of 5x10-6M.Transit time was assessed by noting the time taken for a 1-5 ml bolus of bromsulphthalein dye (BSP) to pass along the 30 cm segment as previously described. The infusion rate was measured at the beginning and end of the experiment.