W J Branch

Evidence for the occurrence of microbial breakdown of carbohydrate in the human colon has been sought by measuring short chain fatty acid (SCFA) concentrations in the contents of all regions of the large intestine and in portal, hepatic and peripheral venous blood obtained at autopsy of sudden death victims within four hours of death. Total SCFA concentration (mmol/kg) was low in the terminal ileum at 13 +/- 6 but high in all regions of the colon ranging from 131 +/- 9 in the caecum to 80 +/- 11 in the descending colon. The presence of branched chain fatty acids was also noted. A significant trend from high to low concentrations was found on passing distally from caecum to descending colon. pH also changed with region from 5.6 +/- 0.2 in the caecum to 6.6 +/- 0.1 in the descending colon. pH and SCFA concentrations were inversely related. Total SCFA (mumol/l) in blood was, portal 375 +/- 70, hepatic 148 +/- 42 and peripheral 79 +/- 22. In all samples acetate was the principal anion but molar ratios of the three principal SCFA changed on going from colonic contents to portal blood to hepatic vein indicating greater uptake of butyrate by the colonic epithelium and propionate by the liver. These data indicate that substantial carbohydrate, and possibly protein, fermentation is occurring in the human large intestine, principally in the caecum and ascending colon and that the large bowel may have a greater role to play in digestion than has previously been ascribed to it.

There is now substantial evidence that some dietary polysaccharides, notably dietary fiber, escape absorption in the small bowel and are then broken down in the large intestine of man. The main end products of this colonic digestive process, which is anerobic, are short chain fatty acids (SCFA), and acetic, propionic, and butyric acids. Although these acids are known to be absorbed from the colon, their subsequent fate and significance is unknown. We have measured venous blood SCFA levels in healthy subjects after a 16-h fast, and then following oral doses of either 50 mmol SCFA, 5, 10, or 20 g doses of the fermentable carbohydrate lactulose, or 20 g of pectin. Fasting venous blood acetate was 53.8 +/- 4.4 mumol/liter (SEM) (n = 14). Fasting arterial blood acetate, taken simultaneously with venous blood in six subjects, was higher; 125.6 +/- 13.5 mumol/liter (arterial) vs. 61.1 +/- 6.9 mumol/liter (venous). Significant levels of propionate or butyrate were not detected in any blood samples. Following an oral dose of 50 mmol mixed SCFA, venous blood acetate reached a peak of 194.1 +/- 57.9 mumol/liter at 45 min and returned to fasting levels at 2 h. Blood acetate also rose in response to lactulose, peak levels occurring 2-4 h after the dose: 5 g, 98.6 +/- 23.1 mumol/liter; 10 g, 127.3 +/- 18.2 mumol/liter; and 20 g, 181.3 +/- 23.9 mumol/liter. Pectin fermentation was much slower, with blood acetate levels starting to rise after 6 h and remaining elevated at about twice fasting levels for the subsequent 18 h. However, areas under the blood acetate curves were closely related (r = 0.97; n = 5), whatever the source of acetate. These studies show that the large intestine makes an important contribution to blood acetate levels in man and that fermentation may influence metabolic processes well beyond the wall of this organ.

1. The effect of dietary fibre digestion in the human gut on its ability to alter bowel habit and impair mineral absorption has been investigated using the technique of metabolic balance. 2. Five healthy male students were studied for 9 weeks under controlled dietary conditions and during the last 6 weeks they took 36 g pectin/d. Bowel habit, transit through the gut, faecal fibre excretion, calcium balance and faecal composition were measured. 3. During the control period only 15% of the dietary fibre ingested was excreted in the stools and when pectin was added to the diet there was no increase in fibre excretion. Stool frequency and mean transit time were unchanged by pectin but stool wet weight increased by 33% and faecal excretion increased (%) for fatty acids 80, nitrogen 47, total dry matter 28 and bile acids 35. Ca balance remained unchanged. 4. It may be concluded from these results that dietary fibre is largely metabolized in the human gut and dietary pectin completely so. This could explain its lack of effect on bowel habit and Ca balance. Other changes in the faeces may be related to an increase in bacterial mass.