The degradation of amino acids, proteins, and blood to short-chain fatty acids in colon is prevented by lactulose

GASTROENTEROLOGY

1990:98:353-360

The Degradation of Amino Acids, Proteins, and Blood to Short-Chain Fatty Acids in Colon Is Prevented by Lactulose PER BR0BECH MORTENSEN, KLAVS HOLTUG, HELEN BONNkN, and METTE RYE CLAUSEN Department of Medicine A, Division of Gastroenterology, Rigshospitalet; Department of Medicine B, Division of Gastroenterology and Hepatology, Hvidovre Hospital, University Copenhagen, Copenhagen, Denmark

Short-chain (C,-C,) fatty acids account for 60%-70% of the anions in the colon. Acetate (C,) is nontoxic in contrast to Ct3)(-C, fatty acids (propionate, hutyrate, isobutyrate, valerate, and isovalerate), which induce coma in animals and may be important in the pathogenesis of hepatic coma in humans. An in-vitro fecal incubation system was used to map out shortchain fatty acid production in the presence of lactulose, amino acids, albumin, or blood. Albumin and blood increased production of all C,-C, fatty acids. In contrast, lactulose was converted to acetate only and increased fecal acidity. The degradation of amino acids, albumin, and blood to short-chain fatty acids was completely inhibited by lo-25 mM lactulose. This was caused mainly by the acidifying effect of lactulose. pH-independent inhibition of blood and amino acid degradation to short-chain fatty acids required concentrations of lactulose exceeding 50100 mM. Thus, the effect of lactulose in the treatment of hepatic coma may be related to its rapid fermentation into organic acids at rates exceeding colonic buffering capacity. This probably reduces formation of toxic fatty acids and ammonia from amino acids, polypeptides, and blood in the colon.

n 1964, Ingelfinger suggested that lactulose might be beneficial for patients with portal-systemic encephalopathy by promoting the growth of Lactobacillus species and decreasing the number of putrefactive bacteria (1). Bircher et al. (21 later confirmed the therapeutic efficacy of lactulose in patients with hepatic encephalopathy, which was reconfirmed by other studies. Lactulose has since been used extensively in the treatment of hepatic encephalopathy, although its exact mechanism of action is unknown. It has been

I

of

suggested that by decreasing colonic pH lactulose would suppress the absorption of nonionized ammonia (3,4). Other proposed mechanisms are suppressed bacterial and intestinal ammonia generation (5-71, increased incorporation of ammonia into bacterial protein (7,8), decreased intestinal transit time, reducing the time available for both the production and absorption of toxins (9), and increased fecal nitrogen excretion (10). Lactulose is a disaccharide (p-l-4 galacto-fructose) that is poorly absorbed in the human small intestine because of the lack of lactulase in humans. In the cecum, lactulose is metabolized by colonic bacteria with the formation of short-chain fatty acids (SCFAs), lactate, hydrogen, and, in some subjects, methane. Breakdown of lactulose by fermentation is also followed by a dramatic increase in hydrogen ion concentration. After ingestion of 30-40 g of lactulose per day, pH in the right side of the colon decrease from 6.0 (range 5.5-7.5) to 4.8 (3.5-6.1) (11). Whether or not fecal pH is decreased by lactulose is probably only a matter of dose, and doses that are optimal for the treatment of hepatic coma generally do decrease fecal pH (9). In healthy individuals the main source of SCFAs is bacterial fermentation of dietary polysaccharides not absorbed in the small bowel (12). SCFAs are thus the predominant anions in the colonic lumen (So-120 mM). They are rapidly absorbed by the colonic mucosa (13.14) and are subsequently metabolized in the liver, although acetate may reach peripheral tissues (15). Potentially toxic substances, such as ammonia, mercaptans, and SCFAs (16-30), which arise in the

Abbreviations used in this paper: SCFA, short-chain fatty acid 8 1990 by the American Gastroenterological Association 0016-5065/90/$3.00

354

MORTENSEN ET AL

GASTROENTEROLOGY Vol. 98, No. 2

lumen of the intestinal tract following bacterial degradation of gut contents, are usually removed or detoxified as the portal blood passes through the liver. During portalsystemic encephalopathy such comagenie substances are diverted around or through the liver and into the systemic circulation, through which they are carried to the brain, possibly inducing coma. In recent work (31) we focused on the fatty acid part of colonic polypeptide metabolism, i.e., SCFA formation. It was found that the presence of lactulose detoxified the SCFA profile because protein catabolism was dominated by the breakdown of lactulose to the nontoxic SCFA acetate. In the current study these findings are extended with special reference to the effects of lactulose on bacterial degradation of amino acids to SCFAs. In addition, an attempt was made to elucidate the effect of the acidification induced by

lactulose [compared to the effect of lactulose itself) on these biochemical processes.

Materials and Methods Preparation

of Samples

Fecal samples from 8 healthy persons were collected and within 30 min homogenized with 5 times its weight of sodium chloride (150 mM) for study in an anaerobic fecal incubation system using the method of Vince et al. [S). Several lo-ml aliquots from each fecal sample were incubated simultaneously to allow comparison of different treatments. Incubation time was 24 h at 37’C in all studies, but in some studies additional analyses were performed with 6 or 72 h of incubation [in Table 1 both 6 and 24 h; in Figure 1 both 24 and 72 h). Termination was performed by freezing. Specimens were stored at -18°C until analysis. Results in

Table 1. Effect of Lactulose (300 mM) and Albumin (10 mg/ml] on Production Normal Subjects After 6 and 24 h of Incubation

of Short-Chain

Fatty Acids in Feces From 8

Short-chain fatty acids Incubation time

Total

C*

C,

C,

C,, acetate: C,, propionate; variance.

C,

iC,

PH

WW

(N = 8) Controls oh Mean SD (%) 6h Mean SD (%] 24 h Mean SD (%] Lactulose (300 mM) 6h Mean SD (%] p vs. controls 24 h Mean SD [%] p vs. controls Albumin [IO mg/ml) 6h Mean SD (%) p vs. controls p vs. lactulose 24 h Mean SD (%] p vs. controls p vs. lactulose ANOVA 6 h, p 24h,p

iC,

77.6 28

47.3 25

14.6 43

10.7 55

1.5 32

1.5 40

2.0 32

6.9 5

143.6 18

84.5 21

29.7

19.4

14

35

2.9 28

2.9 20

4.2 32

6.5 4

212.5 29

105.1 36

48.7 17

40.0 50

5.8 19

5.0 19

7.8 27

6.3 5

395.4 46
317.8 53 -co.01

45.6 51 NS

24.3 45 NS

1.8

24h 10.01

3.3 48 NS

2.7 29 10.05

4.6 7
618.5 47
528.4 51
53.8 55 NS

28.1 42 NS

1.8 21
3.9 79 NS

2.6 31
4.0 4 -co.01

217.4 25 <0.05 <0.05

118.4 24 NS co.01

48.8 28 NS NS

32.3 31 NS NS

3.6 30 to.01 to.01

6.6 59 to.01
7.6 41 to.01 to.01

6.6 4 NS
437.9 30
190.2 38 NS
103.5 27
87.7 37 to.01
12.9 35 to.01
16.6 45 co.01 -Lo.01

27.1 31
6.6 4 -co.01 -co.01



NS
NS

<0.05

to.001 to.001

C,, butyrate; iC,, isobutyrate; C,, valerate; iC,, isovalerate; NS, p > 0.05 (l-way ANOVA); ANOVA, analysis of

LACTULOSE AND HEPATIC ENCEPHALOPATHY

February 1990

355

0 c2 0ii(-.!i: $ c4 III 1: iC4

IsI c5 ::

Try

Phe

Ala

Asp

Arg

His

Leu

LYS

Hypro

Pro I iC5

GIY

TYr

Val

Met

Ser

Glu

lieu

CySSCy CySH

Blood

Figure 1.Production of SCFAs (expressed as percentage of concentrations in control assays) from 19 different amino acids (100 mM) and blood (100ml/L) in 24-h fecal incubations. Seventy-two-hour incubations resulted in similar or higher SCFA production. Columns marked with dotted lines represent 72-h incubations that exceeded the 24-h incubations by more than 50% (of control values). SCFAs: C,, acetate; C,, propionate; C,, butyrate; iC,, isobutyrate; C,, valerate; iC,, isovalerate. Amino acids: Try, tryptophan; Phe, phenylalanine; Ala, alanine; Asp, aspartate; Arg, arginine; His, histidine; Leu, leucine; Lys, lysine; Hypro, hydroxyproline; Pro, proline; Gly, glycine; Tyr, tyrosine; Val, valine; Met, methionine; Ser, serine; Glu, glutamate; Ileu, isoleucine; CySSCy, cystine; CySH, cystein. The upper 99% confidence interval for control incubations is indicated by the horizontal line.

Table 1 represent mean values found in feces from 8 persons, whereas results in Table 2 and in Figures 1,2, and 3 from 1 subject only, were based on fecal suspensions although from different batches and persons between experiments. Lactulose was added at 0 h to give concentrations of 10, 25, 50, 100, and 300 mM (in Table 1 only 300 mM), whereas blood (1 ml/10 ml], albumin (100 mg/lO ml], and amino acids (100 mM) were always added in constant amounts at 0 h. Investigations on the effect of pH in assays [Figure 2) were performed in suspensions titrated to initial pH values 2, 3, 4, 5, 6, 7, and 8 by addition of hydrochloric acid or sodium hydroxide before incubation with blood or amino acids. Sodium hydroxide was added to incubations titrated to constant pH to bring them to a pH of 7-8 during incubation (Figure 3). In addition, they were homogenized with sodium bicarbonate (150 mM) instead of sodium chloride to increase buffering capacity. Because addition of substrate or alkali increased osmolality, control experiments were performed to examine the effect of increased osmolality on SCFA formation: fecal suspensions homogenized with water, 150 and 300 mM sodium chloride, did not differ significantly in production rates of SCFAs when incubated without substrate addition or with 100 mM lactulose, respectively.

Analysis Short-chain fatty acid concentrations were determined by steam distillation followed by gas-liquid chromatography, essentially as described by Zijlstra et al. (321, with modifications by Hdverstad et al. (33). Glass column-packing material, SP-1200/l% H,PO, on 80/100 Chromosorb W AW, was obtained from Supelco, Inc., Bellefonte, Pa. Short-chain fatty acids, 2-ethylbutyric acid (internal standard], lactulose, D-glucose, D-fructose, albumin, hemoglobin, and amino acids were obtained from Merck, FRG, and Sigma Chemical Co., St. Louis, MO. Concentrations were calculated from areas of gas-liquid chromatography peaks automatically computerized by a Hewlett-Packard 5830A gas chromatograph [Hewlett-Packard, Palo Alto, Calif.) equipped with an FID detector. The pH of fecal suspensions was measured by a pH meter (Radiometer, Copenhagen).

Statistics Statistical analysis is l-way analysis of variance with the difference between individual groups using the least significant difference.

356

MORTENSEN

ET AL.

GASTROENTEROLOGY

Table 2. Effect of Lactulose (O-300 mM] on Production Fecal Incubation Svstem

of Short-Chain

Vol. 98. No. 2

Fatty Acids From Blood and 7 Amino Acids in a 24-h Short-chain fatty acids

C*

C5

Controls” OhfN=41

h&an 24;ef;i

PH

(mM1

’ 8)

SD (%]

46.1 6.6

16.5 6.1

14.1 5.9

1.9 4.7

124.3 12.0

62.1 7.4

54.4 8.3

7.7 10.7

:::

:::

6.9 0.9

88::

11.7 9.5

6.6 0.9

6.8 6.4

(% of 24-h control incubations)

24-h control limits

142 58

+99% -99%

126 74

129 71

138 63

131 69

133 67

100

100

100

100

Lactuloee added (O-300 mM) LactB1ose Only

100

a; 50 100 300 Bloom (100 ml/L) 10

Ei:

100

300 SerinOe(100 mM) 10 25 50 100 300

Vali?

::

100

300 Glutamine (100 mM) 0

10

25 50

100

(100 mM)

(100 mM]

10 25 50

100

300 Leu$~e (100 mM) 10 25 50 100 300 Isole~e

154 208 265 328 515 473

160 207 250 180 148 116

(100 mM)

10

25 50 100 300

1:; 127

58

89 166 236 343 557 458 235 163 258 467 554 498 196 272 207 405 548 540 -

95 143 192 138 155 126 :: 144 143 137 122 1:: 139 152 153 142

94 117 152 50 60 49 199 103 68 :: 48 181 205 133 60 61 54 -

138 220 219 390 552 582

105 149 162 156 148 154

64 86 140

108 148 191 320 467 434

104 132 163 141 139 126

98 113 136

96 126 181

102 118 161 157 148 148

292

443 484

56

:: 343 162 138

EJ 60

i! 22 a:

506 133 88

234 167 127 88 72 53

10

10 25 50 100 300 Prolp

113 151 188 150 145

(100 mM)

300 Hyd;xyproline

100

140 203 351 574 547

229 181 n 94 60 70 61

u

z 6:s 6.1 4.0 4.0

421 122 51

6.3 5.9 5.0 4.5 4.2 4.0

s9;

63 88

49

23

559 37

20 23 19

6.8 6.0 5.1 4.4 4.2 4.1

90 89

:“7

;: 36 29

23 15 18 16

319 ::

---I

6.5 5.7 5.1 4.3 4.0 4.0 5.0 4.8 4.5

60 45 50 56 56 51

E 4.1

92 83 38 52 60 80

88 85 22 19 21 20

6.0 5.5 5.0 4.3 3.9 4.0

101 88 52 55

93

6.4 5.6

x: 21 20 23

::: 3.9 3.9

z: 51

62 64 48 49 53 51

92 78 50 31 29 28

6.4 5.7 5.0 4.4 4.0 4.0

99 98 94 55

55 46 44 49

99 76 56 35

::

Zf

::

8.5 5.7 5.1 4.5 4.0 4.0

C,. acetate; C,, propionate: C,, butyrate; iC,. isobutyrate; C,, valerate; iC,. isovalerate. ‘Lactulose not added. Boxes indicate SCFAs specifically formed from amino

acids or blood.

February 1990

LACTULOSE AND HEPATIC ENCEPHALOPATHY

357

ACETATE p”

I

I

I

‘1

0

I

I

1

PRIOPIONATE

I

BUTYRATE

60 40 Figure 2. Concentrations of SCFAs (mM) after 24-h incuhation at different pHs in control assays ( q) and in assays to which blood (o), vallme (VI, proline ( 0) or leucine (A) were added.

20 0

Results Conversion of Lactulose and Albumin to Short-Chain Fatty Acids Fecal incubations from 8 normal individuals showed a spontaneous production of all C,-C, SCFAs (acetate, propionate, butyrate, isobutyrate, valerate, and isovalerate), resulting in an average increase in concentration from 77.6 to 143.6 and 212.5 mM after 6and 24-h incubations, respectively (Table 1). Whereas the interindividual variation both before and after 24-h incubation was approximately 30% (Table l), the coefficient of variation on determining SCFAs within each fecal batch was naturally lower and usually less than 10% (Table 2). Addition of lactulose (300 mM) to incubations increased acetate production considerably but did not increase the production of other SCFAs significantly. In addition a marked decrease in pH of the incubate was observed. In contrast, incubation with albumin (10 mg/ml] resulted in a significant increase in all C,-C, SCFAs, whereas pH did not differ from values seen in control incubations (Table 1).The SCFA profile produced by albumin metabo-

2345678

pH in assay

lism was thus completely by lactulose.

different from that produced

Conversion of Amino Acids and Proteins to Short-Chain Fatty Acids Figure 1 shows the metabolism of 19 amino acids and blood into acetate, propionate, butyrate, isobutyrate, valerate, and isovalerate in fecal suspensions incubated for 24 h. Results are given as a percentage of simultaneously incubated suspensions from the same batch (controls) without external addition of amino acids or blood. The individual amino acids were selectively metabolized to 1 or more specific SCFAs. Acetate was formed from histidine, hydroxyproline, serine, and glutamine; propionate from aspartate; butyrate from hydroxyproline, serine, and glutamine; isobutyrate from valine; valerate from hydroxyproline and proline; and isovalerate from leutine and isoleucine. These and some further amino acids were converted to specific SCFAs after very long incubation times of 72 h [Figure 1;only 72-h incubations producing more SCFA than 24-h incubations

358

MORTENSEN

ET AL.

GASTROENTEROLOGY

Vol. 98, No. 2

ISOBUTYRATE

I

I

I

I

1

I

PROPIONATE

0

1

1

T

B&RAT; 60 40 20 0.

1

I

0

10

I

25

I 50

I

100

300

Lactulose

0

added

( mmol II)

+5O% of control values are shown: columns marked by dotted lines). Only few amino acids reached the 99% significance level of SCFA production after 6 h of incubation (not shown). The qualitative SCFA response remained specific for each amino acid when studies were repeated on fecal suspensions from 4 individuals, whereas the quantities produced varied up to 100% between the individuals. It was thus foreseeable that the addition of albumin (Table l), hemoglobin (not shown), and blood (Figure 1) resulted in a production of all the C,-C, SCFAs. Effect of Lactulose on the Metabolism Amino Acids and Blood

of

Table 2 shows that increasing concentrations of lactulose increase the acetate production approximately 5-fold and reduce pH to approximately 4. Addition above 100 mM did not result in further production of acetate. Even moderate concentrations (10-50 mM) of lactulose altered the typical metabolic pattern of blood into a predominant production of acetate, signifying that lactulose inhibits the metabolism of blood proteins.

10

25

50

100

300

Figure 3. Concentrations of SCFAs (mM) in control assays (0) and in assays to which blood (0), valine (V), proline (01, or leucine (A) and lactulose were added (O-NWmMk 34-h incubation in unbuffered @H 7+4; open symbols) and buffered assays (pH constant 7-8; closed symbols).

To determine whether this effect was caused by an inhibition of proteolysis or an inhibition of amino acid metabolism, amino acids were incubated with increasing concentrations of lactulose. Table 2 shows that the amino acid specific production of SCFAs also was completely inhibited by 25-50 mM lactulose. Effect of pH on the Metabolism Acids and Blood

of Amino

Figure 2 shows the effect of pH in assays titrated with hydrochloric acid to initial values between 2-8. pH 2-4 inhibited the formation of all C2-C, SCFAs in fecal suspensions incubated for 24 h with or without blood. Similarly, the production of isobutyrate, valerate, and isovalerate from valine, proline, and leucine, respectively, was almost completely inhibited at low pH levels. The effect was especially noticeable when pH was reduced from 5 to 4 and was much more moderate in the range of pH between 6-8. Titration to alkaline pH above 10 also reduced SCFA production to a minimum (not shown). Hence, acidification inhibited bacterial protein and amino acid metabolism in the absence of lactulose.

February

1990

Effect of Lactulose on the Metabolism of Amino Acids and Blood at Constant pH Titration with sodium hydroxide to maintain the assay pH between 7 and 8 increased the metabolism of lactulose to acetate, especially at lactulose concentrations above 100 mM [Figure 3). Lactulose in nonbuffered incubations inhibited its own metabolism, presumably because of the induced drop in pH. Moreover, when assays were buffered to pH between 7 and 8, the inhibitory effect of lactulose on the metabolism of blood, valine, proline, and leucine to C,-C, SCFAs was considerably reduced and required lactulose concentrations above 50-100 mM compared to 10-25 mM in nonbuffered assays. Discussion Since lactulose was introduced in the treatment of hepatic coma in 1966, the mechanism behind this disaccharide’s clinical effect has been the subject of speculation. It was at first suggested that its effect was secondary to an acidification of the colonic contents, causing an increased ionization of ammonia and hence a reduced absorption of ammonia (3,4). However, fecal excretions of ammonia were not increased during treatment with lactulose in clinical studies (9). Vince et al. (8) found that ammonia production was reduced in a fecal incubation system when lactulose was added, and that lactulose increased ammonia assimilation for bacterial protein synthesis. The present study is carried out in a similar fecal incubation system, but with focus on organic acid formation rather than nitrogen metabolism. In a previously published report, Mortensen et al. (31) found that lactulose was largely metabolized to acetate, whereas blood and proteins resulted in a production of all C,-C, SCFAs and that protein and blood metabolism were inhibited by lactulose. The present work shows that polypeptides are metabolized to all C,-C, SCFAs because a number of amino acids after deamination are converted to the corresponding organic acid, which is further metabolized to 1 or more specific SCFAs. In contrast, fructose and galactose, the constituents of lactulose, are primarily metabolized to acetate (341. Hence, SCFA production reflects whether the biochemical processes are dominated by the metabolism of lactulose or polypeptides. The present results suggest that lactulose decreases protein degradation in colonic contents by an inhibition of the amino acid metabolism. Apparently this is not primarily caused by substrate competition, which was initially believed most probable. In terms of SCFA production, the main effect of lactulose on the bacterial metabolism is secondary to its acidifying properties. Protein and amino acid metabolism was

LACTULOSE AND HEPATIC ENCEPHALOPATHY

359

already inhibited at lactulose concentrations of lo-25 mM, corresponding to a pH of approximately 5. Even the fermentation of lactulose to acetate ceases at high concentrations (100 mM), which reduce pH to minimal values of 4. However, concentrations of lactulose necessary to inhibit SCFA production from proteins or blood in buffered assays were close to the upper levels that would occur in a clinical situation. The concentration of lactulose after a single dose of 20 g is 50-100 mM in the cecum (35) with a corresponding decrease in pH to values in the range 3.5-6.1 in the proximal colon (11,35). It is believed that products from the bacterial metabolism in colon are coma inducing in patients with liver insufficiency. Not only ammonia but also mercaptans, SCFAs, and medium-chain fatty acids (16-30) are suspected of being comagenic substances. None of these substances alone, however, seems to be responsible. In this connection it is noteworthy that in contrast to SCFAs with more carbon atoms, acetate does not induce coma in animals (25-30). Thus inhibition of protein degradation by lactulose presumably hinders both the formation of ammonia and toxic amino acid-derived C,-C, SCFAs. It might even be speculated that the lactulose induced acidification may reduce a bacterial production of toxic substances other than those investigated in this study. The present results suggest that lactulose should be given in doses that do not only acidify the cecum but also overcome buffering capacity in the left half of the colon. It is possible that a lactulose enema could be a useful supplement if fecal pH is not reduced sufficiently. Moreover, lactulose administered as enemas could theoretically be replaced by the less expensive glucose, since malabsorption in the small intestine is not needed. Studies that clinically show that the antiencephalopathic effect of lactulose is not specific but seems to be unspecifically related to the supply of carbohydrate to the colon are in agreement with the present results. The malabsorbed disaccharide lactilol is thus clinically equivalent to lactulose (36,371, just as lactose is used in areas of the world where the population is lactase deficient (38). Dietary fiber, i.e.. nonstarch polysaccharides undergoing fermentation in colon, also seem to improve chronic hepatic encephalopathy (391. References 1. Ingelfinger

F. Editorial comments. In: Beeson PB, ed. Year Book of Medicine. Chicago: Year Book Medical, 1964-1965:591-Z. 2. Bircher J, Miiller J, Guggenheim P, Haemmerli UP. Treatment of chronic portalsystemic encephalopathy with lactulose. Lancet 1966;1:890-3. 3. Caste11 DO, Moore EW. Ammonia absorption from the human

360 MORTENSEN

ET AL.

colon. The role of non-ionic diffusion. Gastroenterology 1971;60: 33-42. 4. Bircher J, Haemmerli UP, Trabert E. The mechanism of action of lactulose in portal-systemic encephalopathy. Non-ionic diffusion of ammonia in the canine colon. Rev Eur Etud Clin Biol 1971;16:352-7. 5. Vince A, Dawson AM, Park N, O’Grady F. Ammonia production by intestinal bacteria. Gut 1973;14:171-7. 6. Van Leeuwen PAM. Lactulose in acute portal-systemic encephalopathy. In: Van Leeuwen PAM. Ammonia generation in the gut and the influence of lactulose and neomycin: review of the literature and experimental studies in the rat. Doctoral thesis, University of Maastricht, The Netherlands, 1985:39-40. 7. Vince AJ. Burridge SM. Ammonia production by intestinal bacteria: the effects of lactose, lactulose and glucose. J Med Microbial 1980;13:177-91. 8. Vince A, Killingley M, Wrong OM. Effect of lactulose on ammonia production in a fecal incubation system. Gastroenterology 1978;74:544-9. 9. Agostini L. Down PF, Murison J, Wrong OM. Faecal ammonia and pH during lactulose administration in man: comparison with other cathartics. Gut 1972;13:859-66. 10. Weber FL Jr. The effect of lactulose on urea metabolism and nitrogen excretion in cirrhotic patients. Gastroenterology 1979; 77:518-23. 11. Bown RL, Gibson JA, Sladen GE, Hicks B, Dawson AM. Effects of lactulose and other laxatives on ileal and colonic pH as measured by a radiotelemetry device. Gut 1974;15:999-1004. 12. Cummings JH. Fermentation in the human large intestine: evidence and implication for health. Lancet 1983;1:1206-9. 13. McNeil NI, Cummings JH, James WPT. Short chain fatty acid absorption by the human large intestine. Gut 1978;19:819-22. 14. Ruppin H, Bar-Meir S, Soergel KH, Wood CM, Schmitt MG. Absorption of short chain fatty acids by the colon. Gastroenterology 1980;78:1500-7. 15. Pomare EW, Branch WJ, Cummings JH. Carbohydrate fermentation in the human colon and its relation to acetate concentrations in venous blood. J Clin Invest 1985;75:1448-54. 16. Zieve FJ, Zieve L, Doizaki WM. Gilsdorf RB. Synergism between ammonia and fatty acids in the production of coma: implications for hepatic coma. J Pharmacol Exp Ther 1974;191: 10-6. 17. Zieve L. Nicoloff DM. Pathogenesis of hepatic coma. Ann Rev Med 1975;26:143-57. 18. Zieve L. Nicoloff DM. Alterations in volatile free fatty acids of blood after hepatectomy. Surgery 1976;80:554-7. 19. Zieve L. The mechanism of hepatic coma. Hepatology 1981;1:3605. 20.Chen S. Mahadevan V, Zieve L. Volatile fatty acids in the breath of patients with cirrhosis of the liver. J Lab Clin Med 1970;75:622-7. 21. Takahashi Y. Serum lipids in liver disease. Liver disease and the relationship of serum lipids and hepatic coma. Jap J Gastroenterol1963;60:571-9. 22. Muto Y, Takahashi Y. Short-chain fatty acids and the liver (abstr). Postgrad Med 1965:37A:A158. 23.Dankert J, Zijlstra JB, Wolthers BG. Volatile fatty acids in human peripheral and portal blood: quantitative determination by vacuum distillation and gas chromatography. Clin Chim Acta 1981;110:301-7. 24.Lai JCK, Silk DBA, Williams R. Plasma short-chain fatty acids in fulminant hepatic failure. Clin Chim Acta 1977;78:305-10. 25. Samson FE, Dahl N Jr. Dahl DR. A study on the narcotic action of the short chain fatty acids. J Clin Invest 1956;35:1291-8.

GASTROENTEROLOGY

Vol. 98, No. 2

26. Walker CO, MC Candless DW, MC Garry JD. Schenker S. Cerebral energy metabolism in short-chain fatty acid-induced coma. J Lab Clin Med 1970;76:569-83. 27. White RP, Samson FE Jr. Effects of fatty acid anions on the electroencephalogram of unanesthetized rabbits. Am J Physiol 1956;186:271-4. 28. Muto Y, Takahashi Y, Kawamura H. Effects of short chain fatty acids anions on the electrical activity of neo-, paleo-, and archicortical systems. Brain Nerve 1964;16:608-14. 29. Teychenne PF, Walters I, Claveria LE. Calne DB, Price F, Macgillivaty BB, Gompertz D. The encephalopathic action of five-carbon-atom fatty acids in the rabbit. Clin Sci Mol Med 1976;50:463-72. 30.Dahl DR. Short chain fatty acid inhibition of rat brain Na-K adenosine triphosphatase. J Neurochem 1968;15:815-20. 31. Mortensen PB, Rasmussen HS, Holtug K. Lactulose detoxifies in vitro short-chain fatty acid production in colonic contents induced by blood: implications for hepatic coma. Gastroenterology 1988;94:750-4. 32.Zijlstra JB, Beukema J, Wolthers BG, Byrne BM, Groen A, Donkert L. Pretreatment methods prior to gas chromatographic analysis of volatile fatty acids from faecal samples. Clin Chim Acta 1977;78:243-50. 33. Hgverstad T, Fausa 0, Bjflrneklett A, Bglhmer T. Short-chain fatty acids in the normal human feces. Stand J Gastroenterol 1984;19:375-81. 34. Mortensen PB, Holtug K, Rasmussen HS. Short-chain fatty acid production from mono- and disaccharides in a fecal incubation system: implications for colonic fermentation of dietary fiber in humans. J Nutr 1988;118:321-5. 35. Florent C, Flourie B, Leblond A, Rautureau M. Bernier J, Rambaud J. Influence of chronic lactulose ingestion on the colonic metabolism of lactulose in man (an in vivo study). J Clin Invest 1985;75:608-13. 36. Morgan MY, Hawley KE, Stambuk D. Lactilol versus lactulose in the treatment of chronic hepatic encephalopathy. A doubleblind, randomized, cross-over study. J Hepatology 1987;4:23644. 37. Patil DH, Westaby D, Mahida YR, Palmer KR, Rees R, Clark ML, Dawson AM, Silk DBA. Comparative modes of action of lactilol and lactulose in the treatment of hepatic encephalopathy. Gut 1987;28:255-9. 38. Uribe M, Marquez MA, Garcia-Ramos G, Escobedo V. Murillo H. Guevara L, Lisker R. Treatment of chronic portal-systemic encephalopathy with lactose in lactase-deficient patients. Dig Dis Sci 1980;25:924-8. 39. Uribe M, Dibildox M, Malpica S, Guillermo E. Villallobos A, Nieto L, Vargas F, Ramos GG. Beneficial effect of vegetable protein diet supplemented with psyllium plantago in patients with hepatic encephalopathy and diabetes mellitus. Gastroenterology 1985;88:901-7.

Received March 13,1989. Accepted August 1.1989. Address requests for reprints to: Per Brdbech Mortensen, M.D., Department of Medicine A. Division of Gastroenterology, Rigshospitalet, DK-2100 Copenhagen, Denmark. This work was supported financially by the Danish Medical Research Council, the Foundation of Max and Anna Friedmann, the Foundation of Else and Svend Madsen, the NOVO Foundation, and the Danish Hospital Foundation for Medical Research, region of Copenhagen, the Faroe Islands, and Greenland. The skillful technical assistance of Kirsten Lund Larsen is highly appreciated.