- Email: [email protected]
Increased Biliary Lipid Secretion in Celiac Disease
GASTROENTEROLOGY 1985;88:134-42
Increased Biliary Lipid Secretion in Celiac Disease MATTI VUORISTO and T ATU A. MIETTINEN Second Department of Medicine, University of Helsinki, Helsinki, Finland
Direct measurements of biliary lipid outputs, cholesterol absorption, and fecal steroids were carried out in celiac patients before and during a gluten-free diet to show whether an enhanced flux of cholesterol into the gut (found earlier in these patients) is due to increased biliary output or mucosal secretion of cholesterol, or both. The bile flow rate and the secretion of biliary cholesterol, phospholipids, and bile acids were significantly increased in celiac disease and appeared to be normalized by effective gluten-free diet. A significant amount of cholesterol originated from the intestinal mucosa, but the amount was not consistently increased in the celiac patients. Fractional absorption of cholesterol was low, but due to enhanced biliary secretion the amount of cholesterol absorbed was mostly within the normal limits so that fecal neutral steroids of biliary origin and cholesterol synthesis were markedly increased in celiac disease. Despite high biliary bile acid secretion, fractional absorption of bile acids was enhanced. Thus, the effective ileal conservation of bile acids could have contributed to increased bile acid-dependent secretion of biliary cholesterol. The enhanced biliary and fecal output of cholesterol should ultimately be balanced by augmented cholesterol synthesis, but the closer site of the synthesis and regulatory mechanisms between cholesterol and lipoprotein metabolism need further exploration. Celiac disease (CD) is characterized by a gluteninduced villous atrophy of the small intestinal mucosa (1). The damage is more marked in the upper than lower part of the gut (2). Because cholesterol is absorbed mainly in the jejunum (3), mucosal injury of this part of the gut would markedly interfere with cholesterol absorption. In fact, a low absorption Received November 14, 1983. Accepted July 18, 1984. Address requests for reprints to: Professor T. A. Miettinen, M.D., Second Department of Medicine, University of Helsinki, 00290 Helsinki 29, Finland. © 1985 by the American Gastroenterological Association 0016-5085/85/$3.30
percentage and a high elimination of cholesterol as fecal neutral steroids have been found in CD, whereas bile acids, which are mainly absorbed in the ileum, are increased in the feces of some patients only (4). Occasionally, however, the amount offecal neutral steroids is so high that the impaired absorption alone could hardly account for it, suggesting that cholesterol flux into the gut lumen is also increased in CD. Estimation of intestinal cholesterol influx using the absorption percentage of cholesterol from fecal and dietary data indicated that the influx was actually increased in CD, so that the absorption of cholesterol in absolute terms was normal in most patients (5). Because the measurement of the flux was indirect and because it was not possible to determine whether the high input was due to enhanced biliary cholesterol secretion or to mucosal cholesterol loss, or both, cholesterol absorption, fecal steroids, and biliary lipid secretion were measured directly in the present study before and during a gluten-free diet.
Subjects and Methods Subjects The present series consisted of 13 patients with established CD of which 11 were matched for sex, age, and body weight with 7 healthy medical students volunteering as controls (Table 1). The criteria for the diagnosis of CD were villous atrophy of the jejunal mucosa and its improvement on a gluten-free diet. Subjects were informed of the nature and purpose of the investigations and all gave written informed consent for the study which was approved by the Ethic Committee of the hospital.
Studies The subjects were hospitalized and placed on a low-cholesterol (125 mg/2400 kcal) solid-food diet, which contained 100 g of fat per day. The daily energy content of Abbreviations used in this paper: CD. celiac disease; PEG. polyethylene glycol.
January 1985
BILIARY LIPIDS IN CELIAC DISEASE 135
Table 1. Clinical and Laboratory Data of Controls and Patients With Untreated Celiac Disease
Case No.
Fecal steroids (mg/day) Cholesterol Fecal Jejunal Cholesterol Serum lipids (mg/l00ml) villous Sexa Age Body weight Bile Neutral synthesis fat absorption (mg/day) Cholesterol Triglycerides (F/M) (yr) (kg) (relative) height (pm) (g/day) acids sterols (%)
Controls 567 456 498 462 481 525 610
35.8 47.9 26.4 31.0 25.9 38.9 36.2
3.6 3.9 3.9 4.6 2.4 3.4 4.5
384 125 341 284 189 260 260
967 444 684 664 659 604 887
1228 486 925 836 760 752 1047
224 205 209 194 194 163 201
44 52 48 57 51 84 65
514
:!:22
34.6 :!:2.9
3.8 :!:0.3
263 :!:33
701 :!:66
862 :!:90
199 :!:8
58 :!:5
20.5 23.8
6.9 10.8 6.4 4.4 6.1 8.1 10.3 27.9 31.3 12.3 33.6
358 156 480 158 265 174 157 107 262 181 561
841 651 1190 990 1634 1107 1371 1566 1403 1983 1024
1099
71 65 112 82
1193
217 132 186 163 139 263 228 139 112 159
260 :!:44
1251 :!:116
678 387
2022 938
3
M
4 5
M F
6 7
M F
24 25
3/4
25 :!:1
62 :!:4
100 :!:4
25 25 31 20 34 35 41 25 38 29 49
63
48 48 58 68 54 56 46 46 75 54
103 81 94 88 121 100 104 81 82 90 104
203
o o o
41.2 27.7 28.3 12.6 4.3 7.6 23.0 0.1
b
60 :!:60
18.9 :!:4.0
14.4 :!:3.3
o
11.0
41.0 4.8
M
F
Mean :!: SEM
27 32 25 22
92 91 90 100 97 116
80 49 60 62 54 67 65
1
2
22
116
Patients 1
M
2 3 4 5 6 7 8 9
10
M F M F F F F F M
11
F
Mean :!: SEM
7/4
32 :!:3
56 :!:3
95 :!:4
12 13
F M
54 16
49 71
91
a
F, female; M, male.
b
96
o o
287 95
o
55
25
b
185
b
b
724
1582 1048 1787
44
143
54 84 68 60 58 115
1413 :!:112
170 :!:15
74 :!:7
2612 1202
201 151
127 122
1428 1590 1582 2031 1485 b
Statistically Significant difference (p < 0.05) from the respective control values.
the diet (30-35 kcal/kg body wt) was adjusted to maintain a constant body weight during the investigation period. The 3-day fecal collections (6,7) and the measurement of the percentage absorption of cholesterol with the oral double-isotope method were performed as described previously (8,9). For the absorption measurement the subjects received a mixture of 0.7 /LCi [4- 14 Clcholesterol and 1.6 /LCi [22,23- 3 Hlf:l-sitosterol (both from the Radiochemical Centre, Amersham, U.K.) in 4 ml of olive oil with a test meal (5). Simultaneously 800 mg of carmine red was included as a visible marker. Appearance of the dye in feces initiated the 3-day stool collection. The biliary lipid secretions were determined principally as described by Grundy and Metzger (10) using a doublelumen radiopaque tube (type AN 20, H.W. Andersen Products Ltd., U.K.). At the beginning of the study, gallbladder contraction was induced by intravenously administering a bolus (2 U/kg body wt) of cholecystokinin (Pancreozymin, The Boots Company Ltd., U.K.) followed by a constant infusion (0.5 U/kg body wt . h) for up to 8 h. The liquid formula was prepared by sonication of 20 ml of olive oil, 20 g of fat-free milk powder, 4.5 g of glycerol-lmono oleate (Fluka AG, Switzerland), 200 ml of water, and 4 g of unabsorbable polyethylene glycol (PEG) 4000 (Fluka AG, Switzerland) in a final volume of 240 ml. The formula
was infused at a rate of 30 ml/h through a proximal outlet of the tube using an infusion pump (type 871 102, B. Braun, Melsungen AG, West Germany) for up to 8 h; hourly samples were obtained via the other lumen located 10 cm more distally. The samples (5-10 ml) were placed in a water bath at 70°C for 5 min to destroy the lipase activity, and then stored at -20°C until analyzed. A jejunal biopsy specimen was taken with Bult's capsule (Carl Storz, West Germany), which under fluoroscopic control was positioned just past the ligament of Treitz. The mucosal specimen was handled and the villous height was measured as described previously (4). Mucosa with a villous height ",,300 /Lm was regarded as normal.
Chemical and Isotopic Analysis Serum cholesterol was determined according to the method of Huang et al. (11) and serum triglycerides were determined by a modified fluorometric method (12) as described by Kessler and Lederer (13). Fecal neutral steroids and fecal bile acids were quantitated as shown earlier (14,15). Because the fecal recovery of f:l-sitosterol was practically complete, the fecal steroid data were calculated in relation to chromic oxide. Biliary cholesterol
136 VUORISTO AND MIETTINEN
GASTROENTEROLOGY Vol. 88, No.1, Part 1
from the duodenal aspirates was measured using gasliquid chromatography (14), bile acids using an enzymatic method (16), and phospholipids according to Bartlett (17). Polyethylene glycol, used as a nonabsorbable marker, was analyzed according to Hyden (18).
Calculations The bile flow rate was calculated on the basis of PEG as follows: Bile flow rate (mllh) PEG infused (mg/ml) x infusion rate (mllh)
Values for absorbed cholesterol in terms of milligrams per day were obtained by multiplying the fractional absorption by the total flux of cholesterol frow different sources. Bile acid absorption was calculated as the difference between biliary and fecal bile acids, and fractional bile acid absorption was calculated by dividing the amount of absorbed bile acids by the amount of biliary bile acids. Data from the Metropolitan Life Insurance Company (20) were used for the determination of relative body weight. Statistical analysis was carried out using Student's t-test and paired t-test when appropriate. Mean ± SEM values are given in the text.
PEG (mg/ml) in aspirated fluid
This measures actual intestinal flow (biliary plus intestinal fluid) rate over the mixing segment. The output of biliary cholesterol, bile acids, and phospholipids into the duodenum was calculated according to the following equation: Lipid secretion rate (mg/h) PEG infused (mg/h)
-------'---..:::..~-
PEG (mg/ml) of aspirate
x lipid (mg/ml) of aspirate.
A high secretion rate of biliary lipids into the gut was usually found during the first 1-4 h. Thereafter, constant secretion rates were obtained, indicating that the lipid outputs were attributable to biliary secretion (10). The means of the leveling-off values were used as biliary lipid secretion rates. Cholesterol synthesis was calculated by subtracting the dietary cholesterol from the sum of fecal bile acids and neutral steroids of cholesterol origin. Fractional cholesterol absorption (absorbed fraction of the test meal cholesterol) was calculated according to the following equation: Fractional absorption of cholesterol = 1 -
14Cf3H ratio in feces sample
~____---------':""::~-------,-
14Cf3H ratio in administered test meal
Total flux of cholesterol into the intestinal lumen was calculated by dividing fecal neutral steroids by [1 fractional absorption of cholesterol). In these calculations it is assumed that in both groups of subjects the fractional absorption of cholesterol represents similar mean daily absorption, and that cholesterol entering the gut from diet, bile, and intestinal mucosa has similar fractional absorption. The available data have suggested that exogenous and endogenous cholesterol are absorbed at similar rates (3). Recent studies (19) suggest, however, that absorption of exogenous cholesterol is slightly less than that of endogenous cholesterol. Thus, it is possible that the present figures calculated for the amounts of total flux, mucosal secretion, and absorbed biliary cholesterol may be slightly underestimated but most likely similarly or to the same extent in the two groups. Consequently, fecal neutral steroids of biliary origin were correspondingly slightly overestimated and those of mucosal origin were slightly underestimated.
Results Fecal and Serum Lipids The jejunal villous height, the percentage absorption of cholesterol, and fecal fat indicated that the severity of CD was moderate in the present series (Table 1). Fecal bile acid excretion was normal, but fecal neutral steroids and cholesterol synthesis were significantly increased in the patients as compared with the controls. The serum cholesterol level tended to be low and serum triglycerides were within the control limits. Biliary Lipid Secretion The bile flow rate was quite high in CD patients (Table 2) and was positively correlated with the bile acid output rate both in the controls (r = 0.64; P < 0.001) and in the patients (r = 0.67; P < 0.001).
Secretion rates of biliary cholesterol and phospholipids were significantly higher in the CD group than in the controls (Table 2). In terms of milligrams per kilogram of body weight, significantly increased output rates were found for all three biliary lipids in CD. The molar percentage of cholesterol in the bile was similar in the patients and in the controls (7.4 ± 0.7 vs. 7.2 ± 0.7), whereas that of bile acids was lower (70.0 ± 1.8 vs. 78.2 ± 1.9; P < 0.05) and that of phospholipids was higher (22.6 ± 1.6 vs. 14.6 ± 1.5; P < 0.05) in the patients compared with the controls. The plot of biliary cholesterol against biliary bile acids showed a positive correlation both in the controls and in the patients (Figure 1). At low bile acid output, i.e., <10 J.Lmollkg· h, the cholesterol secretion rate was higher in the patients (1.09 ± 0.07 J.Lmollkg . h) than in the controls (0.68 ± 0.07 J.Lmollkg . h). This indicates that at the low bile acid output the bile may become supersaturated more frequently in CD than normally. Also both in the controls and in the patients bile acid output was significantly correlated with phospholipids (r = 0.66 and 0.86; p < 0.001), which in turn were correlated
January 1985
BILIARY LIPIDS IN CELIAC DISEASE
137
Table 2. Bile Flow and Biliary Lipid Outputs of Controls and Untreated Patients With Celiac Disease flile flow (mllh)
Case No.
Cholesterol
(mg/h)
Bile acids
(mg/kg' h)
(mg/h)
Phospholipids
(mg/kg' h)
(mg/h)
(mg/kg' h)
Controls 1 2 3 4 5 6 7
168 88 63 220 66 127 194
53 19 21 38 14 15 35
0.66 0.39 0.35 0.61 0.26 0.22 0.54
374 279 231 316 154 268 411
4.68 5.69 3.85 5.10 2.85 4.00 6.32
8.2 4.0 4.0 4.7 1.8 2.2 9.1
0.10 0.08 0.07 0.08 0.03 0.03 0.14
Mean ± SEM
132 ± 24
28 ± 5
0.43 ± 0.06
290 ± 33
4.64 ± 0.45
4.9 ± 1.1
0.08 ± 0.02
2 3 4 5 6 7 8 9 10 11
373 312 241 125 99 122 156 223 258 360 279
41 47 78 25 47 32 31 40 47 64 40
0.65 0.97 1.63 0.43 0.69 0.59 0.55 0.87 1.02 0.85 0.74
Mean ± SEM
232 ± 29 0
45 ± 50
0.82 ± 0.10 b
512 ± 92
12 13
211 190
44 53
0.90 0.75
606 366
Patients
i
o
p < 0.05.
b
479 548 1000 166 324 171 317 461 367 1072 731
7.60 11.42 20.83 2.86 4.76 3.17 5.66 10.02 7.98 14.29 13.53 9.28 ± 1.65 0 12.37 5.15
15.2 19.9 29.9 3.5 10.8 5.6 4.5 7.8 13.7 30.1 14.2
0.24 0.41 0.62 0.06 0.16 0.10 0.08 0.17 0.30 0.40 0.26
14.1 ± 2.8 b
0.26 ± 0.05 b
6.5 10.2
0.13 0.14
P < 0.02.
with biliary cholesterol (r
0.73 and 0.77; p
<
0.001).
Cholesterol Influxes, Absorption, and Fecal Excretion
Cholesterol can enter the intestinal lumen from the diet, bile, and intestinal mucosa. Dietary and biliary cholesterol were measured directly in the preseht study. The amount of mucosal output could be estimated only indirectly as the difference betweeh the total influx of cholesterol and the sum of biliary and dietary cholesteroL The data for the fluxes of cholesterol into the gut from the three sources are presented in the upper part of Table 3. Dietary cholesterol is similar in the two groups. The biliary and total influxes of cholesterol are higher in the CD group than in the controls. The difference between the total and biliary plus dietary influxes of cholesterol differed significantly from zero, suggesting that a significant amount of intestinal cholesterol originates from the intestinal mucosa. However, the difference was not significantly higher in the CD group than in the controls, indicating that the high cholesterol flux into the intestinal lumen in CD is mainly of biliary origin. Cholesterol absorption in terms of milligrams per
day (calculated from the influx rates and the fractional absorption) is shown in the middle part of Table 3. Due to the low fractional absorption, the amount of absorbed dietary cholesterol is significantly lowered in CD, whereas the amount of absorbed biliary cholesterol is within the control limits because the biliary output is increased. Absorption of mucosal cholesterol seems to be normal. The total amount of absorbed cholesterol is not significantly lower in CD patients than in the controls. Amounts of fecal steroids from different sources are shown in the lower part of Table 3. The high fecal neutral sterol output in CD is seen to originate mainly from biliary cholesterol, whereas the contributions of dietary and mucosal cholesterol are not significantly increased. Bile Acid Absorption
Increased biliary and normal fecal bile acids indicate that reabsorption of bile acids is increased in CD. In fact, the amount of bile acids reabsorbed (difference between biliary and fecal outputs) was almost twice as high in CD patients (12.0 ± 2.2 g/day) than in the controls (6.7 ± 0.8 g/day). Effectiveness of bile acid absorption is shown also by the finding that the fractional reabsorption (difference
138 VUORISTO AND MIETTINEN
GASTROENTEROLOGY Vol. 88, No.1, Part 1
Correlations r·0.62
5
p
""0
cf70 4
E
o
::J..
...i 0
a:
/
o
3
/
UJ
f0Ul UJ ...J
0 I
/
~~
.r:;
0,
./
o'~
/
/
/
o
• 2
()
cD
10
20
30
40
50
60
B-BILE ACIDS. f,lmol/kg/h
Figure 1. Comparison of biliary secretion rates of cholesterol versus bile acids in the controls (closed circles) and in the patients with celiac disease (open circles). The points consist of the leveling-off values of the subjects. The slopes of the regression lines (y = 0.1654 + 0.0805x and y = 0.8497 + 0.0523x, respectively) were not significantly different.
between biliary and fecal bile acids divided by biliary bile acids) of bile acids was significantly higher in CD patients than in the controls (0.975 ± 0.004 vs. 0.960 ± 0.005; P < 0.05).
Effects of Gluten-Free Diet To show whether the perturbed lipid metabolism of CD patients is corrected by the gluten-free diet, the studies were repeated in 7 subjects after the subjects had consumed the diet for 4-8 mo (mean 7 mol. This caused a significant increase in body weight, villous height, and serum cholesterol and a significant decrease in fecal fat (Table 4). The recovery, however, was not complete so that the villous height, for instance, had reached the control range in only 2 of the 7 patients (Figure 2). Thus, the subjects still had mild CD. The treatment caused a significant decrease in biliary cholesterol secretion, but the control values of Table 2 were not yet reached. The changes in the bile acid, phospholipid, and bile flow rates did not quite reach statistical significarice (Table 4). The diet improved fractional absorption of cholesterol and tended to normalize intestinal cholesterol fluxes (Table 4). In terms of grams per day, bile acid reabsorption only tended to be decreased by the diet (11.3 vs. 10.3 g/day) but no constant change was found in the fractional absorption of bile acids, suggesting that the lower small intestine was still hyperfunctioning.
To explore possible reasons and consequences of high biliary cholesterol secretion in CD several correlations were calculated. Serum cholesterol and triglycerides were not associated with biliary cholesterol, whereas the villous height was (Figure 2). Thus, the more severe the disease the higher the biliary cholesterol and other lipid secretion rates. Also, the villous height correlated negatively with the bile acid and phospholipid secretion rates (r = -0.60 and -0.59, respectively; p < 0.01). In addition to the association with biliary bile acids (Figure 1), biliary cholesterol was correlated with the absorbed bile acids both in the controls (r = 0.77; P < 0.05) and in the patients (r = 0.85; P < 0.001) but not quite significantly with bile acid synthesis (r = 0.67 and 0.32, respectively). Furthermore, in CD the fractional absorption of bile acids was correlated positively with the bile acids reabsorbed (r = 0.54; P < 0.05) and the biliary bile acid secretion (r = 0.75; P < 0.001). Thus, effective conservation of bile acids most likely by the hyperplastic ileal mucosa may expand the bile acid pool and consequently increase bile acid secretion and bile acid-dependent biliary secretion of cholesterol. Cholesterol synthesis was also associated with biliary cholesterol secretion in both the controls (r = 0.78; P < 0.05) and the patients (r = 0.47; P < 0.05). Table 3. Sources of Intestinal Cholesterol, Its Absorption, and Fecal Excretion in Control Subjects and Celiac Patients Source
Control subjects a
Celiac patients b
Fractional cholesterol absorption 0.346 ± 0.029 0.189 Cholesterol flux into gut lumen (mg/day) Diet 103 ± 5 98 Bile 669 ± 32 989 Total influx c 1080 ± 99 1573 Mucosa d 308 ± 76 486 Cholesterol absorption (mg/day) Diet 35 ± 3 19 Bile 231 ± 48 187 Mucosa 107 ± 29 114 Total 373 ± 46 315 Fecal output (mg/day) Diet 67 ± 5 79 Bile 436 ± 88 808 Mucosa 198 ± 53 370 Endogenous" 634 ± 63 1178 Total 701 ± 66 1257
p
± 0.040
<0.01
± ± ± ±
5 83 169 153
NS <0.05 <0.05 NS
± ± ± ±
4 38 49 79
<0.01 NS NS NS
± ± ± ± ±
6 77 114 128 129
NS <0.01 NS <0.005 <0.001
NS, not significant. an = 7. b Cholesterol absorption was not measured in patient No.3 (in Table 2); n = 10. C Total fecal output divided by (1 - fractional absorption). d Difference between total influx and the sum of dietary and biliary sources. e Total minus unabsorbed dietary cholesterol or sum of neutral sterols from biliary and mucosal origin.
January 1985
BILIARY LIPIDS IN CELIAC DISEASE
139
Table 4. Effects of Gluten-Free Diet on Cholesterol Metabolism in Celiac Disease Parameter Body weight (kg) Villous height (/-Lm)" Fractional cholesterol absorption b Fecal fat (g/day) Serum lipids (mg/lOO ml) Cholesterol Triglycerides Biliary secretion rates Bile flow (mllh) Cholesterol (mg/h) Cholesterol (mg/kg' h) Bile acids (mg/h) Bile acids (mg/kg . h) Phospholipids (mg/h) Phospholipids (mg/kg' h) Bile acid reabsorption (mg/day) Percentage Fecal steroids Bile acids (mg/day) Bile acids (mg/kg' day) Neutral steroids (mg/day) Neutral steroids (mg/~g . day)
Before GFD 53 38 0.110 22.8
± ± ± ±
During GFD
3 26 0.038 5,3
59 213 0 .237 7.3
159 ± 15 92 ± 11 232 43 0 .83 485 9.45 11 0 .21
± ± ± ± ± ± ±
20 3 0,06 57 1 .24 2 0.04
330 6.3 1282 25.2
± ± ± ±
84 1.6 171 4.0
3 49 0 .028 1.6
<0.01 <0.01 <0,05 <0.01
224 ± 19 81 ± 6
<0.001
± ± ± ± ± ± ±
NS NS
197 37 0 .62 440 7.46 10 0.17
11318 ± 1314 97.2 ± 0.6
± ± ± ±
p
NS
25 3 0 .03 71 1.26 0.5 0 .01
<0,025
NS NS NS NS
10320 ± 1715 97.5 ± 0.5
NS NS
± ± ± ±
NS NS NS NS
233 3.9 832 14.1
23 0.3 122 2.2
Values represent mean ± SEM of 7 patients, GFD. gluten-free diet; NS. not significant. " n = 6; b n = 5 (includes patients 2. 7-9, 11-13 in Table 1).
In the controls the fractional cholesterol absorption is not correlated with the biliary output, indicating that the higher the biliary output the higher the absolute absorption (r = 0.96; P < 0.001) and fecal excretion of cholesterol (r = 0.78; P < 0.05). This would mean that a high cholesterol absorption contributes to biliary cholesterol secretion under normal conditions. In CD, however, the absorption deficiency results in a negative correlation of the biliary cholesterol with the fractional absorption (r = -0.59; P < 0.05). A lack of association between biliary cholesterol and the absorbed cholesterol in CD could indicate a negligible contribution of cholesterol absorption to biliary cholesterol secretion.
Mucosal Cholesterol Loss
The contribution of mucosal cholesterol loss into the intestinal and fecal sterols was quite small but significant in the controls and tended to be
1.6
1,0 .J::
"-
Ol
.:£ "-
Ol
-'
0
Discussion The previous indirect measurements suggested that the amounts of the intraluminal cholesterol (21) and of the endogenous cholesterol flux into the gut are increased in CD (5). The present direct measurements show that the biliary cholesterol secretion is actually enhanced in CD and is the main factor for the expanded pool of intestinal cholesterol in these patients. The absorption findings agree well with our earlier results (5) in that the low fractional absorption of cholesterol in CD associated with the high biliary secretion provides quite a normal absorption of cholesterol in absolute terms.
0 ,8
E
a: w ~ (/)
w
-'
0,6
•
• Patients off GFD o Patients on GFD * Controls
~~
-0
•
0.4
I
,
*
*
0
()
*
* * * *
0,2
[])
°
100
300 VILLOUS
HEIGHT.
500 ~m
Figure 2. Relationship between biliary cholesterol secretion and villous height of jejunal mucosa in the patients off and on the gluten-free diet (GFD) and in the controls (r = -0.65; p < 0.001). Six patients were studied before and during GFD; paired values are connected with continuous lines. '
140
GASTROENTEROLOGY Vol. 88. No.1, Part 1
VUORISTO AND MIETTINEN
increased in the CD patients. The finding that in the patients, but not in the controls, the amount of fecal endogenous neutral steroids was significantly higher than the unabsorbed portion of biliary cholesterol (p < 0.05) may also point to a slightly increased contribution of mucosal cholesterol. An increased epithelial cell turnover of the damaged mucosa (22,23) associated with the high mucosal cholesterol synthesis (24,25) could actually enhance cholesterol loss into the gut lumen, but the direct lipoprotein leakage is not excluded either. In fllct, the perfusion studies (26) have disclosed that the loss of deoxyribonucleic acid (cell loss) from the damaged mucosa in CD was two to three times higher than from the normal mucosa. Furthermore, a positive correlation was found between cell loss and lipid loss, as if mucosal cholesterol secretion had occurred within exfoliated epithelial cells.
Bile Acid Reabsorption In contrast to the negative correlation of the fractional absorption of cholesterol with the biliary cholesterol secretion, that of bile acids was increased proportionately to the increased biliary bile acid secretion in CD. This indicates very effective ileal reabsorption of bile acids probably by a hyperfunctioning ileum. It also suggests that the marked ileal conservation of bile acids significantly contributes to the expanded bile acid pool found by Low-Beer et al. (27) in CD and shown to be associated with a sluggish gallbladder contraction (28). A high vitamin B12 absorption in some patients with CD is also suggestive of a hyperfunctioning terminal ileum (29).
Biliary Lipid Secretion The basic reason for the enhanced biliary cholesterol secretion is not readily apparent in CD. The secretion rate of biliary cholesterol was associated with the villous height, the bile acid absorption and secretion, fecal neutral sterols, and cholesterol synthesis but not serum cholesterol. Villous damage as a primary event causes malabsorption, which in turn apparently results in ileal hyperfunction. Because bile acids strongly determine bile flow and biliary lipid secretion. (30,31), the markedly enhanced bile acid absorption by the hyperfunctioning ileum of clinically manifest CD could subsequently increase bile acid, cholesterol, and phospholipid secretions. It is unlikely that excessive bile acid secretion alone could deplete liver cholesterol to the extent that would, according to the current concept (32), activate hepatic cholesterol synthesis and lower serum cholesterol via increased hepatic low-density
lipoprotein receptors. Bile acid feeding for gallstone dissolution appears to reduce biliary cholesterol secretion (33,34) and inhibit cholesterol synthesis (35-37).
Origin of Excessive Biliary Cholesterol Excess of biliary and fecal cholesterol is most likely balanced by increased cholesterol synthesis in CD, because cholesterol absorption is within normal limits and long-term mobilization of tissue cholesterol is impossible (4). Two sources of synthesis, the intestinal mucosa and the liver, should be considered. Intestinal origin. In CD, intestinal mucosa effectively produces cholesterol (24,25)' which is not accumulated in mucosal cells (25). Thus, some a'f it may be lost to the intestine but its lymphatic transport to the liver would be increased (normal quantity of absorbed cholesterol plus newly synthesized mucosal cholesterol) so that hepatic cholesterol would not be depleted. A cholesterol load to the liver caused by a high-cholesterol diet can enhance biliary reexcretion of cholesterol (38). In CD, excessive bile acid flux may potentiate reexcretion but proteincalorie malnutrition (39) may reduce the release of hepatic cholesterol to the circulation as very lowdensity lipoprotein, allowing more to be secreted in the bile. In fact, the production rate of very lowdensity lipoprotein triglycerides and the synthesis and receptor-mediated catabolism of low-density lipoprotein apoprotein are low (40). Accordingly, excessive mucosal synthesis and transport of intestinal cholesterol to the liver, impaired hepatic lipoprotein production, and excessive enterohepatic circulation of bile acids could enhance biliary cholesterol secretion in CD. Hepatic origin. Alternatively, the transport of intestinal cholesterol may compensate insufficiently for the enhanced cholesterol loss in CD, resulting in depletion of hepatocyte cholesterol and subsequently in enhanced hepatic cholesterol synthesis. This type of event when caused by bile acid malabsorption is associated with an increased very low-density lipoprotein production (41) and enhanced receptormediated catabolism of low-density lipoprotein (42,43). However, the opposite findings were actually observed in CD (40). Thus, hepatic cholesterol synthesis may not be increased in CD or the synthesis is not coordinated by the hepatic low-density lipoprotein receptors. In fact, although serum concentrations of cholesterol precursors are frequently low in CD (44), a gluten-free diet lowers,their contents within lipoproteins (unpublished observation) as if hepatic cholesterol synthesis had been enhllnced (37,44) under basal conditions.
January 1985
References 1. Paulley JV. Observations on the aetiology of idiopathic steatorrhea. Jejunal and lymph-node biopsies. Br Med J 1954;4: 1318-21. 2. McDonald WC, Brandborg LL, Flick AL, Trier JS, Rubin CEo Studies of celiac sprue. IV . The response of the whole length of the small bowel to a gluten-free diet. Gastroenterology 1964;47:57 3-89. 3. Borgstrom B. Studies on intestinal cholesterol absorption in human. J Clin Invest 1960;39:809-15. 4. Vuoristo M, Tarpila S, Miettinen TA. Serum lipids and fecal steroids in patients with celiac disease: effects of gluten-free diet and cholestyramine. Gastroenterology 1980;78:1518-25 . 5. Vuoristo M, Miettinen TA. Cholesterol absorption, elimination and synthesis in coeliac disease. Eur J Clin Invest 1982;12:285-91. 6. Grundy SM, Ahrens EH Jr, Salen G. Dietary f3-sitosterol as an internal standard to correct for cholesterol losses in sterol balance studies. J Lipid Res 1968;9:374-87. 7. Davignon J, Simmonds WJ, Ahrens EH Jr. Usefulness of chromic oxide as an internal standard for balance studies in formula-fed patients and assessment of colonic function. J Clin Invest 1968;47:127-38. 8. Borgstrom B. Quantification of cholesterol absorption in man by fecal analysis after the feeding of a single isotope-labeled meal. J Lipid Res 1969;10:331-7. 9. Sodhi HS, Horlick L, Nazir OJ, Kudchodhar BJ. A simple method for calculating absorption of dietary cholesterol in man. Proc Soc Exp Bioi Med 1971;137:277-9. 10. Grundy SM, Metzger AL. A phYSiological method for estimation of hepatic secretion of biliary lipids in man. Gastroenterology 1972;62:1200-17. 11. Huang TC, Chen CP, Weller V, Raftery A. A stable reagent for the Liebermann-Burchard reaction. Application to rapid serum cholesterol determination. Anal Chern 1961;33:1405-7. 12 . Noble RO, Campbell FM. Improved accuracy in automated Iluorometric determination of plasma triglycerides. Clin Chern 1970;16:166-70. 13. Kessler G, Lederer H. Fluorometric measurement of triglycerides. In: Skeggs LT, ed . Automation in analytical chemistry. New York: Medical Inc., 1966:341-4. 14. Miettinen TA, Ahrens EH IrT Grundy SM. Quantitative isolation and gas-liquid chromatographic analysis of total dietary and fecal neutral steroids. J Lipid Res 1965;6:411-24 . 15. Grundy SM, Ahrens EH Jr, Miettinen TA. Quantitative isolation and gas-liquid chromatographic analysis of total fecal bile acids. J Lipid Res 1965 ;6:397-410. 16. Fausa O. Quantitative determination of serum bile acids using a purified 3-f3-hydroxysteroid dehydrogenase. Scand J Gastroenterol 1975;10:747-52. 17. Bartlett GR. Phosphorus assay in column chromatography. J BioI Chem 1959;234:466-8. 18. Hyden S. A turbidometric method for the determination of higher polyethene glycols in biological materials. Kungl Lantbrukshogskolans Annaler 1955 ;2 2:139-45 . 19. Samuel P, McNamara OJ. Differential absorption of exogenous and endogenous cholesterol in man. JLipid Res 1983;24:26576. 20. Metropolitan Life Insurance Company. New weight standards for men and women. Statist Bull 1959;40(Nov-Dec):1. 21. Miettinen TA, Siurala M. Micellar solubilization of intestinal lipids and sterols in gluten enteropathy and liver cirrhosis. Scand J GastroenteroI1971 ;6:527-35. 22. Padykula HA, Strauss EW, Landman AJ, Gardner FH. A
BILIARY LIPIDS IN CELIAC DISEASE
141
morphologic and histochemical analysis of the human jejunal epithelium in nontropic sprue. Gastroenterology 1961 ;40: 735-65. 23. Wright N, Watson A, Morley A, Appleton 0, Marks J. Cell kinetics in flat (avillous) mucosa of the human small intestine. Gut 1973;14:701-10. 24. Dietschy JM, Gamel WG. Cholesterol synthesis on the intestine of man: regional differences and control mechanisms. J Clin Invest 1971;50:872-80. 25 . Vuoristo M, Miettinen TA. Synthesis of squalene, methyl sterols and cholesterol in jej unal biopsies of patients with coeliac disease. In: Garsi, ed. Abstract book of the VI World Congress of Gastroenterology. Madrid, 1978:49. 26. Croft ON, Cotton PB. Gastrointestinal cell loss in man. Digestion 1973;8:144-60. 27. Low-Beer TS, Heaton KW, Po mare EN, Read AE. The effect of coeliac disease bile salts. Gut 1973;14:204-8. 28. Low-Beer TS, Heaton KW, Heaton ST, Read AE. Gallbladder inertia and sluggish enterohepatic circulation of bile salts in coeliac disease. Lancet 1971 ;i:991-4. 29. MacKinnon AM, Short MD, Elias E, Dowling RH. Adaptive changes in vitamin Bt2 absorption in celiac disease and after proximal small-bowel resection in man. Am J Dig Dis 1975; 20:835-40. 30. Preisig R, Cooper HL, Wheeler HO . The relationship between taurocholate secretion rate and bile production in the unanesthetized dog during cholinergic blockade and during secretin administration. J Clin Invest 1962;41:1152-62. 31. Schersten T, Nilsson S, Cahlin E, Filipson M, Brodin-Persson G. Relationship between th e biliary excretion of bile acids and the excretion of water, lecithin, and cholesterol in man. Eur J Clin Invest 1971 ;1 :2 42-7 . 32. Brown MS , Goldstein JL. Lipoprotein receptors in the liver; control signals for plasma cholesterol traffic. J Clin Invest 1983;72:743-7. 33. Adler RD , Bennion LJ, Duane WC, Grundy SM. Effect of low doses of chenodeoxycholic acid feeding on biliary lipid metabolism. Gastroenterology 1975;68:326-34. 34. Northfield TC, LaRusso NF, Hofmann AF. Biliary lipid output during three meals and an overnight fast. II. Effect of chenodeoxycholic acid treatment in gallstone subjects. Gut 1975 ;16: 12-7. 35. Coyne MI. Bonorris GG, Goldstein LI, Schoenfield LJ. Effect of chenodeoxycholic acid and phenobarbital on the rate-limiting enzymes of hepatic cholesterol and bile acid synthesis in patients with gallstones. I Lab Clin Med 1976;87:281-91. 36. Maton PN, Murphy GM, Dowling RH. Ursodeoxycholic acid treatment of gallstones. Dose-response study and possible mechanism of action. Lancet 1977;ii:1297-301. 37. Miettinen T A. Effects of bile acid feeding and depletion on plasma and biliary squalene, methyl sterols and lathosterol. In: Paumgartner G, Stiehl A, Gerok W, eds. Bile acids and lipids. Lancaster: MTP Press, 1981:255-62. 38. Quintao E, Grundy SM, Ahrens EH Jr. Effect of dietary cholesterol on the regulation of total body cholesterol in man. J Lipid Res 1971;12:233-47. 39. Green PA, Wollaeger EE. The clinical behavior of sprue in th e United States. Gastroenterology 1960;38:399-418. 40. Kesaniemi YA, Vuoristo M, Miettinen TA. Metabolism of very low density and low density lipoproteins in patients with coeliac disease. In: Carlsson LA, Olsson AG, eds. Proceedings of the 41st European Atherosclerosis Group Meeting, Stockholm. New York: Raven , 1983:49-53. 41. Angelin B, Einarsson K, Hellstrom K. Leijd B. Effects of cholestyramine and chenodeoxycholic acid on the metabolism of endogenous triglyceride in hyperlipoproteinemia. J Lipid Res 1978;19:1017-24.
142
VUORISTO AND MIETTINEN
42 . Spengel FA, Jadhav A, Duffield RGM , Wood CB , Thomson GR. Superiority of partial ileal bypass over cholestyramine in reducing cholesterol in familial hypercholesterolaemia. Lancet 1981 ;ii:768-70. 43 . Shepherd J, Packard CJ, Bicker S, Lawrie TDV, Morgan HG.
GASTROENTEROLOGY Vol. 88, No.1, Part 1
Cholestyramine promotes receptor-mediated low-density lipoprotein catabolism. N Engl J Med 1980;302 :1219-22. 44. Miettinen T A. Detection of changes in human cholesterol metabolism. Ann C1in Res 1970;2:300-20.
Increased Biliary Lipid Secretion in Celiac Disease MATTI VUORISTO and T ATU A. MIETTINEN Second Department of Medicine, University of Helsinki, Helsinki, Finland
Direct measurements of biliary lipid outputs, cholesterol absorption, and fecal steroids were carried out in celiac patients before and during a gluten-free diet to show whether an enhanced flux of cholesterol into the gut (found earlier in these patients) is due to increased biliary output or mucosal secretion of cholesterol, or both. The bile flow rate and the secretion of biliary cholesterol, phospholipids, and bile acids were significantly increased in celiac disease and appeared to be normalized by effective gluten-free diet. A significant amount of cholesterol originated from the intestinal mucosa, but the amount was not consistently increased in the celiac patients. Fractional absorption of cholesterol was low, but due to enhanced biliary secretion the amount of cholesterol absorbed was mostly within the normal limits so that fecal neutral steroids of biliary origin and cholesterol synthesis were markedly increased in celiac disease. Despite high biliary bile acid secretion, fractional absorption of bile acids was enhanced. Thus, the effective ileal conservation of bile acids could have contributed to increased bile acid-dependent secretion of biliary cholesterol. The enhanced biliary and fecal output of cholesterol should ultimately be balanced by augmented cholesterol synthesis, but the closer site of the synthesis and regulatory mechanisms between cholesterol and lipoprotein metabolism need further exploration. Celiac disease (CD) is characterized by a gluteninduced villous atrophy of the small intestinal mucosa (1). The damage is more marked in the upper than lower part of the gut (2). Because cholesterol is absorbed mainly in the jejunum (3), mucosal injury of this part of the gut would markedly interfere with cholesterol absorption. In fact, a low absorption Received November 14, 1983. Accepted July 18, 1984. Address requests for reprints to: Professor T. A. Miettinen, M.D., Second Department of Medicine, University of Helsinki, 00290 Helsinki 29, Finland. © 1985 by the American Gastroenterological Association 0016-5085/85/$3.30
percentage and a high elimination of cholesterol as fecal neutral steroids have been found in CD, whereas bile acids, which are mainly absorbed in the ileum, are increased in the feces of some patients only (4). Occasionally, however, the amount offecal neutral steroids is so high that the impaired absorption alone could hardly account for it, suggesting that cholesterol flux into the gut lumen is also increased in CD. Estimation of intestinal cholesterol influx using the absorption percentage of cholesterol from fecal and dietary data indicated that the influx was actually increased in CD, so that the absorption of cholesterol in absolute terms was normal in most patients (5). Because the measurement of the flux was indirect and because it was not possible to determine whether the high input was due to enhanced biliary cholesterol secretion or to mucosal cholesterol loss, or both, cholesterol absorption, fecal steroids, and biliary lipid secretion were measured directly in the present study before and during a gluten-free diet.
Subjects and Methods Subjects The present series consisted of 13 patients with established CD of which 11 were matched for sex, age, and body weight with 7 healthy medical students volunteering as controls (Table 1). The criteria for the diagnosis of CD were villous atrophy of the jejunal mucosa and its improvement on a gluten-free diet. Subjects were informed of the nature and purpose of the investigations and all gave written informed consent for the study which was approved by the Ethic Committee of the hospital.
Studies The subjects were hospitalized and placed on a low-cholesterol (125 mg/2400 kcal) solid-food diet, which contained 100 g of fat per day. The daily energy content of Abbreviations used in this paper: CD. celiac disease; PEG. polyethylene glycol.
January 1985
BILIARY LIPIDS IN CELIAC DISEASE 135
Table 1. Clinical and Laboratory Data of Controls and Patients With Untreated Celiac Disease
Case No.
Fecal steroids (mg/day) Cholesterol Fecal Jejunal Cholesterol Serum lipids (mg/l00ml) villous Sexa Age Body weight Bile Neutral synthesis fat absorption (mg/day) Cholesterol Triglycerides (F/M) (yr) (kg) (relative) height (pm) (g/day) acids sterols (%)
Controls 567 456 498 462 481 525 610
35.8 47.9 26.4 31.0 25.9 38.9 36.2
3.6 3.9 3.9 4.6 2.4 3.4 4.5
384 125 341 284 189 260 260
967 444 684 664 659 604 887
1228 486 925 836 760 752 1047
224 205 209 194 194 163 201
44 52 48 57 51 84 65
514
:!:22
34.6 :!:2.9
3.8 :!:0.3
263 :!:33
701 :!:66
862 :!:90
199 :!:8
58 :!:5
20.5 23.8
6.9 10.8 6.4 4.4 6.1 8.1 10.3 27.9 31.3 12.3 33.6
358 156 480 158 265 174 157 107 262 181 561
841 651 1190 990 1634 1107 1371 1566 1403 1983 1024
1099
71 65 112 82
1193
217 132 186 163 139 263 228 139 112 159
260 :!:44
1251 :!:116
678 387
2022 938
3
M
4 5
M F
6 7
M F
24 25
3/4
25 :!:1
62 :!:4
100 :!:4
25 25 31 20 34 35 41 25 38 29 49
63
48 48 58 68 54 56 46 46 75 54
103 81 94 88 121 100 104 81 82 90 104
203
o o o
41.2 27.7 28.3 12.6 4.3 7.6 23.0 0.1
b
60 :!:60
18.9 :!:4.0
14.4 :!:3.3
o
11.0
41.0 4.8
M
F
Mean :!: SEM
27 32 25 22
92 91 90 100 97 116
80 49 60 62 54 67 65
1
2
22
116
Patients 1
M
2 3 4 5 6 7 8 9
10
M F M F F F F F M
11
F
Mean :!: SEM
7/4
32 :!:3
56 :!:3
95 :!:4
12 13
F M
54 16
49 71
91
a
F, female; M, male.
b
96
o o
287 95
o
55
25
b
185
b
b
724
1582 1048 1787
44
143
54 84 68 60 58 115
1413 :!:112
170 :!:15
74 :!:7
2612 1202
201 151
127 122
1428 1590 1582 2031 1485 b
Statistically Significant difference (p < 0.05) from the respective control values.
the diet (30-35 kcal/kg body wt) was adjusted to maintain a constant body weight during the investigation period. The 3-day fecal collections (6,7) and the measurement of the percentage absorption of cholesterol with the oral double-isotope method were performed as described previously (8,9). For the absorption measurement the subjects received a mixture of 0.7 /LCi [4- 14 Clcholesterol and 1.6 /LCi [22,23- 3 Hlf:l-sitosterol (both from the Radiochemical Centre, Amersham, U.K.) in 4 ml of olive oil with a test meal (5). Simultaneously 800 mg of carmine red was included as a visible marker. Appearance of the dye in feces initiated the 3-day stool collection. The biliary lipid secretions were determined principally as described by Grundy and Metzger (10) using a doublelumen radiopaque tube (type AN 20, H.W. Andersen Products Ltd., U.K.). At the beginning of the study, gallbladder contraction was induced by intravenously administering a bolus (2 U/kg body wt) of cholecystokinin (Pancreozymin, The Boots Company Ltd., U.K.) followed by a constant infusion (0.5 U/kg body wt . h) for up to 8 h. The liquid formula was prepared by sonication of 20 ml of olive oil, 20 g of fat-free milk powder, 4.5 g of glycerol-lmono oleate (Fluka AG, Switzerland), 200 ml of water, and 4 g of unabsorbable polyethylene glycol (PEG) 4000 (Fluka AG, Switzerland) in a final volume of 240 ml. The formula
was infused at a rate of 30 ml/h through a proximal outlet of the tube using an infusion pump (type 871 102, B. Braun, Melsungen AG, West Germany) for up to 8 h; hourly samples were obtained via the other lumen located 10 cm more distally. The samples (5-10 ml) were placed in a water bath at 70°C for 5 min to destroy the lipase activity, and then stored at -20°C until analyzed. A jejunal biopsy specimen was taken with Bult's capsule (Carl Storz, West Germany), which under fluoroscopic control was positioned just past the ligament of Treitz. The mucosal specimen was handled and the villous height was measured as described previously (4). Mucosa with a villous height ",,300 /Lm was regarded as normal.
Chemical and Isotopic Analysis Serum cholesterol was determined according to the method of Huang et al. (11) and serum triglycerides were determined by a modified fluorometric method (12) as described by Kessler and Lederer (13). Fecal neutral steroids and fecal bile acids were quantitated as shown earlier (14,15). Because the fecal recovery of f:l-sitosterol was practically complete, the fecal steroid data were calculated in relation to chromic oxide. Biliary cholesterol
136 VUORISTO AND MIETTINEN
GASTROENTEROLOGY Vol. 88, No.1, Part 1
from the duodenal aspirates was measured using gasliquid chromatography (14), bile acids using an enzymatic method (16), and phospholipids according to Bartlett (17). Polyethylene glycol, used as a nonabsorbable marker, was analyzed according to Hyden (18).
Calculations The bile flow rate was calculated on the basis of PEG as follows: Bile flow rate (mllh) PEG infused (mg/ml) x infusion rate (mllh)
Values for absorbed cholesterol in terms of milligrams per day were obtained by multiplying the fractional absorption by the total flux of cholesterol frow different sources. Bile acid absorption was calculated as the difference between biliary and fecal bile acids, and fractional bile acid absorption was calculated by dividing the amount of absorbed bile acids by the amount of biliary bile acids. Data from the Metropolitan Life Insurance Company (20) were used for the determination of relative body weight. Statistical analysis was carried out using Student's t-test and paired t-test when appropriate. Mean ± SEM values are given in the text.
PEG (mg/ml) in aspirated fluid
This measures actual intestinal flow (biliary plus intestinal fluid) rate over the mixing segment. The output of biliary cholesterol, bile acids, and phospholipids into the duodenum was calculated according to the following equation: Lipid secretion rate (mg/h) PEG infused (mg/h)
-------'---..:::..~-
PEG (mg/ml) of aspirate
x lipid (mg/ml) of aspirate.
A high secretion rate of biliary lipids into the gut was usually found during the first 1-4 h. Thereafter, constant secretion rates were obtained, indicating that the lipid outputs were attributable to biliary secretion (10). The means of the leveling-off values were used as biliary lipid secretion rates. Cholesterol synthesis was calculated by subtracting the dietary cholesterol from the sum of fecal bile acids and neutral steroids of cholesterol origin. Fractional cholesterol absorption (absorbed fraction of the test meal cholesterol) was calculated according to the following equation: Fractional absorption of cholesterol = 1 -
14Cf3H ratio in feces sample
~____---------':""::~-------,-
14Cf3H ratio in administered test meal
Total flux of cholesterol into the intestinal lumen was calculated by dividing fecal neutral steroids by [1 fractional absorption of cholesterol). In these calculations it is assumed that in both groups of subjects the fractional absorption of cholesterol represents similar mean daily absorption, and that cholesterol entering the gut from diet, bile, and intestinal mucosa has similar fractional absorption. The available data have suggested that exogenous and endogenous cholesterol are absorbed at similar rates (3). Recent studies (19) suggest, however, that absorption of exogenous cholesterol is slightly less than that of endogenous cholesterol. Thus, it is possible that the present figures calculated for the amounts of total flux, mucosal secretion, and absorbed biliary cholesterol may be slightly underestimated but most likely similarly or to the same extent in the two groups. Consequently, fecal neutral steroids of biliary origin were correspondingly slightly overestimated and those of mucosal origin were slightly underestimated.
Results Fecal and Serum Lipids The jejunal villous height, the percentage absorption of cholesterol, and fecal fat indicated that the severity of CD was moderate in the present series (Table 1). Fecal bile acid excretion was normal, but fecal neutral steroids and cholesterol synthesis were significantly increased in the patients as compared with the controls. The serum cholesterol level tended to be low and serum triglycerides were within the control limits. Biliary Lipid Secretion The bile flow rate was quite high in CD patients (Table 2) and was positively correlated with the bile acid output rate both in the controls (r = 0.64; P < 0.001) and in the patients (r = 0.67; P < 0.001).
Secretion rates of biliary cholesterol and phospholipids were significantly higher in the CD group than in the controls (Table 2). In terms of milligrams per kilogram of body weight, significantly increased output rates were found for all three biliary lipids in CD. The molar percentage of cholesterol in the bile was similar in the patients and in the controls (7.4 ± 0.7 vs. 7.2 ± 0.7), whereas that of bile acids was lower (70.0 ± 1.8 vs. 78.2 ± 1.9; P < 0.05) and that of phospholipids was higher (22.6 ± 1.6 vs. 14.6 ± 1.5; P < 0.05) in the patients compared with the controls. The plot of biliary cholesterol against biliary bile acids showed a positive correlation both in the controls and in the patients (Figure 1). At low bile acid output, i.e., <10 J.Lmollkg· h, the cholesterol secretion rate was higher in the patients (1.09 ± 0.07 J.Lmollkg . h) than in the controls (0.68 ± 0.07 J.Lmollkg . h). This indicates that at the low bile acid output the bile may become supersaturated more frequently in CD than normally. Also both in the controls and in the patients bile acid output was significantly correlated with phospholipids (r = 0.66 and 0.86; p < 0.001), which in turn were correlated
January 1985
BILIARY LIPIDS IN CELIAC DISEASE
137
Table 2. Bile Flow and Biliary Lipid Outputs of Controls and Untreated Patients With Celiac Disease flile flow (mllh)
Case No.
Cholesterol
(mg/h)
Bile acids
(mg/kg' h)
(mg/h)
Phospholipids
(mg/kg' h)
(mg/h)
(mg/kg' h)
Controls 1 2 3 4 5 6 7
168 88 63 220 66 127 194
53 19 21 38 14 15 35
0.66 0.39 0.35 0.61 0.26 0.22 0.54
374 279 231 316 154 268 411
4.68 5.69 3.85 5.10 2.85 4.00 6.32
8.2 4.0 4.0 4.7 1.8 2.2 9.1
0.10 0.08 0.07 0.08 0.03 0.03 0.14
Mean ± SEM
132 ± 24
28 ± 5
0.43 ± 0.06
290 ± 33
4.64 ± 0.45
4.9 ± 1.1
0.08 ± 0.02
2 3 4 5 6 7 8 9 10 11
373 312 241 125 99 122 156 223 258 360 279
41 47 78 25 47 32 31 40 47 64 40
0.65 0.97 1.63 0.43 0.69 0.59 0.55 0.87 1.02 0.85 0.74
Mean ± SEM
232 ± 29 0
45 ± 50
0.82 ± 0.10 b
512 ± 92
12 13
211 190
44 53
0.90 0.75
606 366
Patients
i
o
p < 0.05.
b
479 548 1000 166 324 171 317 461 367 1072 731
7.60 11.42 20.83 2.86 4.76 3.17 5.66 10.02 7.98 14.29 13.53 9.28 ± 1.65 0 12.37 5.15
15.2 19.9 29.9 3.5 10.8 5.6 4.5 7.8 13.7 30.1 14.2
0.24 0.41 0.62 0.06 0.16 0.10 0.08 0.17 0.30 0.40 0.26
14.1 ± 2.8 b
0.26 ± 0.05 b
6.5 10.2
0.13 0.14
P < 0.02.
with biliary cholesterol (r
0.73 and 0.77; p
<
0.001).
Cholesterol Influxes, Absorption, and Fecal Excretion
Cholesterol can enter the intestinal lumen from the diet, bile, and intestinal mucosa. Dietary and biliary cholesterol were measured directly in the preseht study. The amount of mucosal output could be estimated only indirectly as the difference betweeh the total influx of cholesterol and the sum of biliary and dietary cholesteroL The data for the fluxes of cholesterol into the gut from the three sources are presented in the upper part of Table 3. Dietary cholesterol is similar in the two groups. The biliary and total influxes of cholesterol are higher in the CD group than in the controls. The difference between the total and biliary plus dietary influxes of cholesterol differed significantly from zero, suggesting that a significant amount of intestinal cholesterol originates from the intestinal mucosa. However, the difference was not significantly higher in the CD group than in the controls, indicating that the high cholesterol flux into the intestinal lumen in CD is mainly of biliary origin. Cholesterol absorption in terms of milligrams per
day (calculated from the influx rates and the fractional absorption) is shown in the middle part of Table 3. Due to the low fractional absorption, the amount of absorbed dietary cholesterol is significantly lowered in CD, whereas the amount of absorbed biliary cholesterol is within the control limits because the biliary output is increased. Absorption of mucosal cholesterol seems to be normal. The total amount of absorbed cholesterol is not significantly lower in CD patients than in the controls. Amounts of fecal steroids from different sources are shown in the lower part of Table 3. The high fecal neutral sterol output in CD is seen to originate mainly from biliary cholesterol, whereas the contributions of dietary and mucosal cholesterol are not significantly increased. Bile Acid Absorption
Increased biliary and normal fecal bile acids indicate that reabsorption of bile acids is increased in CD. In fact, the amount of bile acids reabsorbed (difference between biliary and fecal outputs) was almost twice as high in CD patients (12.0 ± 2.2 g/day) than in the controls (6.7 ± 0.8 g/day). Effectiveness of bile acid absorption is shown also by the finding that the fractional reabsorption (difference
138 VUORISTO AND MIETTINEN
GASTROENTEROLOGY Vol. 88, No.1, Part 1
Correlations r·0.62
5
p
""0
cf70 4
E
o
::J..
...i 0
a:
/
o
3
/
UJ
f0Ul UJ ...J
0 I
/
~~
.r:;
0,
./
o'~
/
/
/
o
• 2
()
cD
10
20
30
40
50
60
B-BILE ACIDS. f,lmol/kg/h
Figure 1. Comparison of biliary secretion rates of cholesterol versus bile acids in the controls (closed circles) and in the patients with celiac disease (open circles). The points consist of the leveling-off values of the subjects. The slopes of the regression lines (y = 0.1654 + 0.0805x and y = 0.8497 + 0.0523x, respectively) were not significantly different.
between biliary and fecal bile acids divided by biliary bile acids) of bile acids was significantly higher in CD patients than in the controls (0.975 ± 0.004 vs. 0.960 ± 0.005; P < 0.05).
Effects of Gluten-Free Diet To show whether the perturbed lipid metabolism of CD patients is corrected by the gluten-free diet, the studies were repeated in 7 subjects after the subjects had consumed the diet for 4-8 mo (mean 7 mol. This caused a significant increase in body weight, villous height, and serum cholesterol and a significant decrease in fecal fat (Table 4). The recovery, however, was not complete so that the villous height, for instance, had reached the control range in only 2 of the 7 patients (Figure 2). Thus, the subjects still had mild CD. The treatment caused a significant decrease in biliary cholesterol secretion, but the control values of Table 2 were not yet reached. The changes in the bile acid, phospholipid, and bile flow rates did not quite reach statistical significarice (Table 4). The diet improved fractional absorption of cholesterol and tended to normalize intestinal cholesterol fluxes (Table 4). In terms of grams per day, bile acid reabsorption only tended to be decreased by the diet (11.3 vs. 10.3 g/day) but no constant change was found in the fractional absorption of bile acids, suggesting that the lower small intestine was still hyperfunctioning.
To explore possible reasons and consequences of high biliary cholesterol secretion in CD several correlations were calculated. Serum cholesterol and triglycerides were not associated with biliary cholesterol, whereas the villous height was (Figure 2). Thus, the more severe the disease the higher the biliary cholesterol and other lipid secretion rates. Also, the villous height correlated negatively with the bile acid and phospholipid secretion rates (r = -0.60 and -0.59, respectively; p < 0.01). In addition to the association with biliary bile acids (Figure 1), biliary cholesterol was correlated with the absorbed bile acids both in the controls (r = 0.77; P < 0.05) and in the patients (r = 0.85; P < 0.001) but not quite significantly with bile acid synthesis (r = 0.67 and 0.32, respectively). Furthermore, in CD the fractional absorption of bile acids was correlated positively with the bile acids reabsorbed (r = 0.54; P < 0.05) and the biliary bile acid secretion (r = 0.75; P < 0.001). Thus, effective conservation of bile acids most likely by the hyperplastic ileal mucosa may expand the bile acid pool and consequently increase bile acid secretion and bile acid-dependent biliary secretion of cholesterol. Cholesterol synthesis was also associated with biliary cholesterol secretion in both the controls (r = 0.78; P < 0.05) and the patients (r = 0.47; P < 0.05). Table 3. Sources of Intestinal Cholesterol, Its Absorption, and Fecal Excretion in Control Subjects and Celiac Patients Source
Control subjects a
Celiac patients b
Fractional cholesterol absorption 0.346 ± 0.029 0.189 Cholesterol flux into gut lumen (mg/day) Diet 103 ± 5 98 Bile 669 ± 32 989 Total influx c 1080 ± 99 1573 Mucosa d 308 ± 76 486 Cholesterol absorption (mg/day) Diet 35 ± 3 19 Bile 231 ± 48 187 Mucosa 107 ± 29 114 Total 373 ± 46 315 Fecal output (mg/day) Diet 67 ± 5 79 Bile 436 ± 88 808 Mucosa 198 ± 53 370 Endogenous" 634 ± 63 1178 Total 701 ± 66 1257
p
± 0.040
<0.01
± ± ± ±
5 83 169 153
NS <0.05 <0.05 NS
± ± ± ±
4 38 49 79
<0.01 NS NS NS
± ± ± ± ±
6 77 114 128 129
NS <0.01 NS <0.005 <0.001
NS, not significant. an = 7. b Cholesterol absorption was not measured in patient No.3 (in Table 2); n = 10. C Total fecal output divided by (1 - fractional absorption). d Difference between total influx and the sum of dietary and biliary sources. e Total minus unabsorbed dietary cholesterol or sum of neutral sterols from biliary and mucosal origin.
January 1985
BILIARY LIPIDS IN CELIAC DISEASE
139
Table 4. Effects of Gluten-Free Diet on Cholesterol Metabolism in Celiac Disease Parameter Body weight (kg) Villous height (/-Lm)" Fractional cholesterol absorption b Fecal fat (g/day) Serum lipids (mg/lOO ml) Cholesterol Triglycerides Biliary secretion rates Bile flow (mllh) Cholesterol (mg/h) Cholesterol (mg/kg' h) Bile acids (mg/h) Bile acids (mg/kg . h) Phospholipids (mg/h) Phospholipids (mg/kg' h) Bile acid reabsorption (mg/day) Percentage Fecal steroids Bile acids (mg/day) Bile acids (mg/kg' day) Neutral steroids (mg/day) Neutral steroids (mg/~g . day)
Before GFD 53 38 0.110 22.8
± ± ± ±
During GFD
3 26 0.038 5,3
59 213 0 .237 7.3
159 ± 15 92 ± 11 232 43 0 .83 485 9.45 11 0 .21
± ± ± ± ± ± ±
20 3 0,06 57 1 .24 2 0.04
330 6.3 1282 25.2
± ± ± ±
84 1.6 171 4.0
3 49 0 .028 1.6
<0.01 <0.01 <0,05 <0.01
224 ± 19 81 ± 6
<0.001
± ± ± ± ± ± ±
NS NS
197 37 0 .62 440 7.46 10 0.17
11318 ± 1314 97.2 ± 0.6
± ± ± ±
p
NS
25 3 0 .03 71 1.26 0.5 0 .01
<0,025
NS NS NS NS
10320 ± 1715 97.5 ± 0.5
NS NS
± ± ± ±
NS NS NS NS
233 3.9 832 14.1
23 0.3 122 2.2
Values represent mean ± SEM of 7 patients, GFD. gluten-free diet; NS. not significant. " n = 6; b n = 5 (includes patients 2. 7-9, 11-13 in Table 1).
In the controls the fractional cholesterol absorption is not correlated with the biliary output, indicating that the higher the biliary output the higher the absolute absorption (r = 0.96; P < 0.001) and fecal excretion of cholesterol (r = 0.78; P < 0.05). This would mean that a high cholesterol absorption contributes to biliary cholesterol secretion under normal conditions. In CD, however, the absorption deficiency results in a negative correlation of the biliary cholesterol with the fractional absorption (r = -0.59; P < 0.05). A lack of association between biliary cholesterol and the absorbed cholesterol in CD could indicate a negligible contribution of cholesterol absorption to biliary cholesterol secretion.
Mucosal Cholesterol Loss
The contribution of mucosal cholesterol loss into the intestinal and fecal sterols was quite small but significant in the controls and tended to be
1.6
1,0 .J::
"-
Ol
.:£ "-
Ol
-'
0
Discussion The previous indirect measurements suggested that the amounts of the intraluminal cholesterol (21) and of the endogenous cholesterol flux into the gut are increased in CD (5). The present direct measurements show that the biliary cholesterol secretion is actually enhanced in CD and is the main factor for the expanded pool of intestinal cholesterol in these patients. The absorption findings agree well with our earlier results (5) in that the low fractional absorption of cholesterol in CD associated with the high biliary secretion provides quite a normal absorption of cholesterol in absolute terms.
0 ,8
E
a: w ~ (/)
w
-'
0,6
•
• Patients off GFD o Patients on GFD * Controls
~~
-0
•
0.4
I
,
*
*
0
()
*
* * * *
0,2
[])
°
100
300 VILLOUS
HEIGHT.
500 ~m
Figure 2. Relationship between biliary cholesterol secretion and villous height of jejunal mucosa in the patients off and on the gluten-free diet (GFD) and in the controls (r = -0.65; p < 0.001). Six patients were studied before and during GFD; paired values are connected with continuous lines. '
140
GASTROENTEROLOGY Vol. 88. No.1, Part 1
VUORISTO AND MIETTINEN
increased in the CD patients. The finding that in the patients, but not in the controls, the amount of fecal endogenous neutral steroids was significantly higher than the unabsorbed portion of biliary cholesterol (p < 0.05) may also point to a slightly increased contribution of mucosal cholesterol. An increased epithelial cell turnover of the damaged mucosa (22,23) associated with the high mucosal cholesterol synthesis (24,25) could actually enhance cholesterol loss into the gut lumen, but the direct lipoprotein leakage is not excluded either. In fllct, the perfusion studies (26) have disclosed that the loss of deoxyribonucleic acid (cell loss) from the damaged mucosa in CD was two to three times higher than from the normal mucosa. Furthermore, a positive correlation was found between cell loss and lipid loss, as if mucosal cholesterol secretion had occurred within exfoliated epithelial cells.
Bile Acid Reabsorption In contrast to the negative correlation of the fractional absorption of cholesterol with the biliary cholesterol secretion, that of bile acids was increased proportionately to the increased biliary bile acid secretion in CD. This indicates very effective ileal reabsorption of bile acids probably by a hyperfunctioning ileum. It also suggests that the marked ileal conservation of bile acids significantly contributes to the expanded bile acid pool found by Low-Beer et al. (27) in CD and shown to be associated with a sluggish gallbladder contraction (28). A high vitamin B12 absorption in some patients with CD is also suggestive of a hyperfunctioning terminal ileum (29).
Biliary Lipid Secretion The basic reason for the enhanced biliary cholesterol secretion is not readily apparent in CD. The secretion rate of biliary cholesterol was associated with the villous height, the bile acid absorption and secretion, fecal neutral sterols, and cholesterol synthesis but not serum cholesterol. Villous damage as a primary event causes malabsorption, which in turn apparently results in ileal hyperfunction. Because bile acids strongly determine bile flow and biliary lipid secretion. (30,31), the markedly enhanced bile acid absorption by the hyperfunctioning ileum of clinically manifest CD could subsequently increase bile acid, cholesterol, and phospholipid secretions. It is unlikely that excessive bile acid secretion alone could deplete liver cholesterol to the extent that would, according to the current concept (32), activate hepatic cholesterol synthesis and lower serum cholesterol via increased hepatic low-density
lipoprotein receptors. Bile acid feeding for gallstone dissolution appears to reduce biliary cholesterol secretion (33,34) and inhibit cholesterol synthesis (35-37).
Origin of Excessive Biliary Cholesterol Excess of biliary and fecal cholesterol is most likely balanced by increased cholesterol synthesis in CD, because cholesterol absorption is within normal limits and long-term mobilization of tissue cholesterol is impossible (4). Two sources of synthesis, the intestinal mucosa and the liver, should be considered. Intestinal origin. In CD, intestinal mucosa effectively produces cholesterol (24,25)' which is not accumulated in mucosal cells (25). Thus, some a'f it may be lost to the intestine but its lymphatic transport to the liver would be increased (normal quantity of absorbed cholesterol plus newly synthesized mucosal cholesterol) so that hepatic cholesterol would not be depleted. A cholesterol load to the liver caused by a high-cholesterol diet can enhance biliary reexcretion of cholesterol (38). In CD, excessive bile acid flux may potentiate reexcretion but proteincalorie malnutrition (39) may reduce the release of hepatic cholesterol to the circulation as very lowdensity lipoprotein, allowing more to be secreted in the bile. In fact, the production rate of very lowdensity lipoprotein triglycerides and the synthesis and receptor-mediated catabolism of low-density lipoprotein apoprotein are low (40). Accordingly, excessive mucosal synthesis and transport of intestinal cholesterol to the liver, impaired hepatic lipoprotein production, and excessive enterohepatic circulation of bile acids could enhance biliary cholesterol secretion in CD. Hepatic origin. Alternatively, the transport of intestinal cholesterol may compensate insufficiently for the enhanced cholesterol loss in CD, resulting in depletion of hepatocyte cholesterol and subsequently in enhanced hepatic cholesterol synthesis. This type of event when caused by bile acid malabsorption is associated with an increased very low-density lipoprotein production (41) and enhanced receptormediated catabolism of low-density lipoprotein (42,43). However, the opposite findings were actually observed in CD (40). Thus, hepatic cholesterol synthesis may not be increased in CD or the synthesis is not coordinated by the hepatic low-density lipoprotein receptors. In fact, although serum concentrations of cholesterol precursors are frequently low in CD (44), a gluten-free diet lowers,their contents within lipoproteins (unpublished observation) as if hepatic cholesterol synthesis had been enhllnced (37,44) under basal conditions.
January 1985
References 1. Paulley JV. Observations on the aetiology of idiopathic steatorrhea. Jejunal and lymph-node biopsies. Br Med J 1954;4: 1318-21. 2. McDonald WC, Brandborg LL, Flick AL, Trier JS, Rubin CEo Studies of celiac sprue. IV . The response of the whole length of the small bowel to a gluten-free diet. Gastroenterology 1964;47:57 3-89. 3. Borgstrom B. Studies on intestinal cholesterol absorption in human. J Clin Invest 1960;39:809-15. 4. Vuoristo M, Tarpila S, Miettinen TA. Serum lipids and fecal steroids in patients with celiac disease: effects of gluten-free diet and cholestyramine. Gastroenterology 1980;78:1518-25 . 5. Vuoristo M, Miettinen TA. Cholesterol absorption, elimination and synthesis in coeliac disease. Eur J Clin Invest 1982;12:285-91. 6. Grundy SM, Ahrens EH Jr, Salen G. Dietary f3-sitosterol as an internal standard to correct for cholesterol losses in sterol balance studies. J Lipid Res 1968;9:374-87. 7. Davignon J, Simmonds WJ, Ahrens EH Jr. Usefulness of chromic oxide as an internal standard for balance studies in formula-fed patients and assessment of colonic function. J Clin Invest 1968;47:127-38. 8. Borgstrom B. Quantification of cholesterol absorption in man by fecal analysis after the feeding of a single isotope-labeled meal. J Lipid Res 1969;10:331-7. 9. Sodhi HS, Horlick L, Nazir OJ, Kudchodhar BJ. A simple method for calculating absorption of dietary cholesterol in man. Proc Soc Exp Bioi Med 1971;137:277-9. 10. Grundy SM, Metzger AL. A phYSiological method for estimation of hepatic secretion of biliary lipids in man. Gastroenterology 1972;62:1200-17. 11. Huang TC, Chen CP, Weller V, Raftery A. A stable reagent for the Liebermann-Burchard reaction. Application to rapid serum cholesterol determination. Anal Chern 1961;33:1405-7. 12 . Noble RO, Campbell FM. Improved accuracy in automated Iluorometric determination of plasma triglycerides. Clin Chern 1970;16:166-70. 13. Kessler G, Lederer H. Fluorometric measurement of triglycerides. In: Skeggs LT, ed . Automation in analytical chemistry. New York: Medical Inc., 1966:341-4. 14. Miettinen TA, Ahrens EH IrT Grundy SM. Quantitative isolation and gas-liquid chromatographic analysis of total dietary and fecal neutral steroids. J Lipid Res 1965;6:411-24 . 15. Grundy SM, Ahrens EH Jr, Miettinen TA. Quantitative isolation and gas-liquid chromatographic analysis of total fecal bile acids. J Lipid Res 1965 ;6:397-410. 16. Fausa O. Quantitative determination of serum bile acids using a purified 3-f3-hydroxysteroid dehydrogenase. Scand J Gastroenterol 1975;10:747-52. 17. Bartlett GR. Phosphorus assay in column chromatography. J BioI Chem 1959;234:466-8. 18. Hyden S. A turbidometric method for the determination of higher polyethene glycols in biological materials. Kungl Lantbrukshogskolans Annaler 1955 ;2 2:139-45 . 19. Samuel P, McNamara OJ. Differential absorption of exogenous and endogenous cholesterol in man. JLipid Res 1983;24:26576. 20. Metropolitan Life Insurance Company. New weight standards for men and women. Statist Bull 1959;40(Nov-Dec):1. 21. Miettinen TA, Siurala M. Micellar solubilization of intestinal lipids and sterols in gluten enteropathy and liver cirrhosis. Scand J GastroenteroI1971 ;6:527-35. 22. Padykula HA, Strauss EW, Landman AJ, Gardner FH. A
BILIARY LIPIDS IN CELIAC DISEASE
141
morphologic and histochemical analysis of the human jejunal epithelium in nontropic sprue. Gastroenterology 1961 ;40: 735-65. 23. Wright N, Watson A, Morley A, Appleton 0, Marks J. Cell kinetics in flat (avillous) mucosa of the human small intestine. Gut 1973;14:701-10. 24. Dietschy JM, Gamel WG. Cholesterol synthesis on the intestine of man: regional differences and control mechanisms. J Clin Invest 1971;50:872-80. 25 . Vuoristo M, Miettinen TA. Synthesis of squalene, methyl sterols and cholesterol in jej unal biopsies of patients with coeliac disease. In: Garsi, ed. Abstract book of the VI World Congress of Gastroenterology. Madrid, 1978:49. 26. Croft ON, Cotton PB. Gastrointestinal cell loss in man. Digestion 1973;8:144-60. 27. Low-Beer TS, Heaton KW, Po mare EN, Read AE. The effect of coeliac disease bile salts. Gut 1973;14:204-8. 28. Low-Beer TS, Heaton KW, Heaton ST, Read AE. Gallbladder inertia and sluggish enterohepatic circulation of bile salts in coeliac disease. Lancet 1971 ;i:991-4. 29. MacKinnon AM, Short MD, Elias E, Dowling RH. Adaptive changes in vitamin Bt2 absorption in celiac disease and after proximal small-bowel resection in man. Am J Dig Dis 1975; 20:835-40. 30. Preisig R, Cooper HL, Wheeler HO . The relationship between taurocholate secretion rate and bile production in the unanesthetized dog during cholinergic blockade and during secretin administration. J Clin Invest 1962;41:1152-62. 31. Schersten T, Nilsson S, Cahlin E, Filipson M, Brodin-Persson G. Relationship between th e biliary excretion of bile acids and the excretion of water, lecithin, and cholesterol in man. Eur J Clin Invest 1971 ;1 :2 42-7 . 32. Brown MS , Goldstein JL. Lipoprotein receptors in the liver; control signals for plasma cholesterol traffic. J Clin Invest 1983;72:743-7. 33. Adler RD , Bennion LJ, Duane WC, Grundy SM. Effect of low doses of chenodeoxycholic acid feeding on biliary lipid metabolism. Gastroenterology 1975;68:326-34. 34. Northfield TC, LaRusso NF, Hofmann AF. Biliary lipid output during three meals and an overnight fast. II. Effect of chenodeoxycholic acid treatment in gallstone subjects. Gut 1975 ;16: 12-7. 35. Coyne MI. Bonorris GG, Goldstein LI, Schoenfield LJ. Effect of chenodeoxycholic acid and phenobarbital on the rate-limiting enzymes of hepatic cholesterol and bile acid synthesis in patients with gallstones. I Lab Clin Med 1976;87:281-91. 36. Maton PN, Murphy GM, Dowling RH. Ursodeoxycholic acid treatment of gallstones. Dose-response study and possible mechanism of action. Lancet 1977;ii:1297-301. 37. Miettinen T A. Effects of bile acid feeding and depletion on plasma and biliary squalene, methyl sterols and lathosterol. In: Paumgartner G, Stiehl A, Gerok W, eds. Bile acids and lipids. Lancaster: MTP Press, 1981:255-62. 38. Quintao E, Grundy SM, Ahrens EH Jr. Effect of dietary cholesterol on the regulation of total body cholesterol in man. J Lipid Res 1971;12:233-47. 39. Green PA, Wollaeger EE. The clinical behavior of sprue in th e United States. Gastroenterology 1960;38:399-418. 40. Kesaniemi YA, Vuoristo M, Miettinen TA. Metabolism of very low density and low density lipoproteins in patients with coeliac disease. In: Carlsson LA, Olsson AG, eds. Proceedings of the 41st European Atherosclerosis Group Meeting, Stockholm. New York: Raven , 1983:49-53. 41. Angelin B, Einarsson K, Hellstrom K. Leijd B. Effects of cholestyramine and chenodeoxycholic acid on the metabolism of endogenous triglyceride in hyperlipoproteinemia. J Lipid Res 1978;19:1017-24.
142
VUORISTO AND MIETTINEN
42 . Spengel FA, Jadhav A, Duffield RGM , Wood CB , Thomson GR. Superiority of partial ileal bypass over cholestyramine in reducing cholesterol in familial hypercholesterolaemia. Lancet 1981 ;ii:768-70. 43 . Shepherd J, Packard CJ, Bicker S, Lawrie TDV, Morgan HG.
GASTROENTEROLOGY Vol. 88, No.1, Part 1
Cholestyramine promotes receptor-mediated low-density lipoprotein catabolism. N Engl J Med 1980;302 :1219-22. 44. Miettinen T A. Detection of changes in human cholesterol metabolism. Ann C1in Res 1970;2:300-20.