- Email: [email protected]
No effects of high-fiber diets on metabolic control and insulin-sensitivity in type 1 diabetic subjects
Diabetes Research and Clinical Practice, 13 (1991) 15-22
0 1991 Elsevier Science Publishers
B.V. 016%8227/91/$03.50
15
DIABET 00509
No effects of high-fiber diets on metabolic control and insulin-sensitivity in type 1 diabetic subjects D. Bruttomesso,
G. Biolo, S. Inchiostro, C. Fongher, G. Briani, E. Duner, M.C. Marescotti, E. Iori, A. Tiengo and P. Tessari
Cattedra di Malattie de1 Ricambio. Istituto di Medicina Clinica, University of Padova, Padova, Italy
(Received 18 September 1990) (Revision received 13 February 1991) (Accepted 22 February 1991)
Summary The metabolic effects of a three-month treatment with a high-fiber diet (15 grams of guar-gum added to a standard diet) were investigated in seven type 1 diabetic subjects, with a moderately poor metabolic control. HbA,, levels, daily insulin requirement, cholesterol, triglyceride, amino acid and intermediate metabolite concentrations were evaluated before and following the high fiber diet, both in the postabsorptive state at euglycemia and during a euglycemic, hyperinsulinemic, hyperaminoacidemic clamp. Insulinmediated glucose utilization, an index of insulin-sensitivity, was also measured during the clamp. Following the diet, no differences in HbA,, levels (7.6 + 0.7% -+ 7.3 + 0.6%), daily insulin requirement (50 k 5 -_*5 1 k 3 U/d), triglyceride, amino acid and intermediary metabolite concentrations in the basal, euglycemic state, were observed. Only cholesterol concentrations decreased significantly (from 165 + 12 to 142 k 12 mg/dl, P -c 0.01) after the diet. During the clamp, the concentrations of all measured substrates were comparable before and after high fiber treatment. Insulin-mediated glucose disposal was also unchanged by guar-gum treatment. Patients’ body weights were not modified by the diet. In conclusion, our study shows that a high fiber diet, obtained with the addition of 15 grams of guar-gum to a standard diet, is of no benefit to IDDM either as regards the metabolic control or insulin sensitivity. Only cholesterol levels were decreased. Therefore, the costs and benefits of these diets in the treatment of IDDM should be reconsidered. Key words: High-fiber diet; Metabolic
Correspondence
control;
Insulin-sensitivity;
to: D. Bruttomesso, Cattedra di Malattie del Ricambio, Istituto di Medicina Clinica, University of Padova, Padova, Italy.
Diabetes mellitus
Introduction It has been shown by several authors that dietary fibers such as guar and pectins improve glucose metabolism [ l-51 and decrease serum cholesterol [6-91 in patients with diabetes mellitus.
16
Recent studies, however, have failed to support this conclusion, [lo-121 and, at this moment, the reasons for these discrepancies are not apparent. Differences in patient selection, fiber dosage and length of treatment might all explain these differences. The presumed advantage of increasing the fiber content of the diet is that fibers slow the intestinal absorption of carbohydrates and smooth the postprandial rise of plasma glucose and insulin concentrations [ 31. Furthermore, dietary fibers reduce the pool of cholesterol precursors by increasing the fecal loss of bile-salts [ 131. It has been also proposed that fibers could improve insulin sensitivity [ 14-161, but this possibility has not been extensively evaluated, in particular in insulin-dependent diabetes mellitus (IDDM). In this report we have reconsidered the longterm effects of guar-gum, added to a standard diet, on metabolic control, daily insulin requirement, insulin sensitivity, serum lipids, plasma amino acids and intermediary metabolite profiles in patients with type 1 diabetes. Insulin sensitivity was evaluated with the euglycemic hyperinsulinemic glucose clamp. The clamp technique was modified by infusing an amino acid mixture together with insulin and glucose, so as to assess insulin sensitivity under conditions that reproduce the post-prandial hyperaminoacidemia, while by-passing the possible effects of fibers on the intestinal absorption of nutrients.
Protocol Seven male insulin-dependent diabetic patients, of a mean age of 30 k 2 (m k SE) years, were studied. Their body mass index was normal (Table 1). In all, the plasma level of C-peptide stimulated by i.v. glucagon was undetectable (< 0.3 ng/ml). In four of them the metabolic control was fairly good (HbA,. 4-7.7% ; normal range : 4-6.5 %), in three it was poor (HbA,, > 9.0%). No patients had any evidence of autonomic neuropathy. They were admitted to our metabolic unit to obtain a baseline metabolic profile some days before the study, while following their usual diet (about 2000 Kcal, containing - 50% carbohydrates, - 30% lipids, - 20% proteins and -30 g of natural fibers). Blood samples were obtained for the assay of fasting plasma cholesterol, triglycerides, HbA,, and amino acid levels. On the evening before the clamp, the patients received at dinner one half of their usual dose of regular insulin, while longacting and intermediate insulins were withheld. During the night, euglycemia was maintained with the intravenous infusion of regular insulin at the rate of l-2 U/h. In the morning of the study, four basal samples for the determination of substrate and hormone concentrations were taken over 30 min with the patients kept at euglycemic conditions. Thereafter, a euglycemic, hyperinsulinemic, hyper-
TABLE 1 Clinical characteristics Subjects
of the diabetic patients Age (yrs)
BMI @g/m* )
Duration of diabetes
HbA,, (%I
Ws) M.A. F.C. T.V. C.L. C.M. C.S. M.A.
20 25 28 32 28 35 40
24.5 24.2 27.1 23.9 21.8 22.6 27.7
9 2 10 9 1.5 7 7
M f SE
29.7 k 2.4
24.5 t 0.8
6.5 + 1.3
Insulin requirement w/24
6.8 4.6 6.1 10.2 9.0 7.7 9.0 7.6 + 0.7
63 50 63 63 53 35 32 50+ 5
h)
17
aminoacidemic clamp was started. Insulin (Actrapid, Novo, Bagswaert, Denmark), was infused intravenously at the rate of 0.05 U/m?&’ for 3 h (O-180 min) by means of a calibrated Harvard pump (Harvard Apparatus, South Nartick, MA, U.S.A.). Insulin was dissolved in 0.9% saline solution along with 2 ml of the patient’s blood, to prevent absorption of the hormone to the plastic tubing. Blood glucose was kept at -90 mg/dl by means of a Biostator GCIIS (Miles Laboratories, Elkhart, IN). During the clamp, an amino acid solution was also infused intravenously, at a rate corresponding to a total nitrogen infusion rate of 0.23 mg/kg * min by means of a Harvard pump. The composition of the amino acid mixture is reported on Table 2. This amino acid infusion rate was chosen in order to increase the concentration of most amino acids by SO-150%, i.e., similar to post-prandial values. Blood samples were drawn every 30 min for assay of glucose, amino acids, intermediary metabolites and free insulin. Four samples were also taken over 30 min at the end of the clamp. These values were averaged and used for the comparison to the basal data. At the end of the study day, the patients were dismissed with their usual diet. In addition, they were asked to take 5 g of guar-gum, as suspension in water, 3 times a day immediately before breakfast, lunch and dinner. Therefore, the total fiber content of the diet was increased from 30 g of natural fibers per day to a total of 45 g (30 g/natural and 15 g/extractive). The patients were contacted by telephone by one of the investigators about compliance to diet TABLE 2 Amino acid concentrations L-alanine L-arginine L-phenylalanine Glycine L-isoleucine L-histidine L-leucine
84.3 34.5 6.1 120.0 68.7 15.5 83.4
in the infused solution (mmol/l) L-lysine L-methionine L-proline L-serine L-threonine L-tryptophan L-valine
41.6 6.1 69.6 41.6 37.8 3.1 71.8
and drug prescription at least once a month. The patients were also recommended not to change their usual insulin dosage, except for major derangement in the metabolic control. The whole study, i.e., the metabolic profile and the euglycemic, hyperaminoacidemic, hyperinsulinemic clamp was repeated after 3 months of treatment during a short admission to our metabolic ward. Methods
Whole blood pyruvate, lactate, 3/$hydroxybutyrate @-OH), acetoacetate (AcAc) and glycerol were analyzed using an enzymatic fluorometric method [ 17,181. Plasma free fatty acid concentrations were measured according to Shimuzu [ 191. Plasma glucose was determined using the glucose oxidase method of the Beckman Glucose Analyzer II [20]. The Bio-Rad column test [21] was used for determining HbA,. Plasma C-peptide concentration was evaluated according to Kuzuya et al. [22]. Free insulin was determined according to Nakagawa et al. [24] and triglycerides [ 251 were determined using enzymatic methods. Plasma amino acids were determined liquid chromatography by high-performance (HPLC) [26]. Insulin-mediated glucose disposal (M) was determined by averaging the exogenous glucose infusion rates (mg/kg . min), at 10 min intervals, in the last hour of the clamp. At the plasma insulin our achieved in study concentrations (N 60 pU/ml), endogenous glucose production is virtually suppressed [ 271. Therefore, the M value represents total glucose utilization. The statistical analysis was performed using the two-tailed Student c-test for paired data in comparison between the two groups before and after treatment. A P value equal or less than 0.05 was considered as statistically significant. Results Basal euglycemic state
The patients
showed a good tolerance
to guar-
18
TABLE 3
unchanged
Haemoglobin A,, and insulin mediated glucose (mean k SE) before and after high fiber diet HbA,, Subject
disposal
M (mg/kg min)
(%)
Diet Standard
High fiber
Standard
High fiber
M.A. F.C. T.V. C.L. C.M. C.S. M.A.
6.84 4.6 6.12 10.2 9 9
8.13 5.2 1.1 5.81 1.8 8 8.2
6.5 1.5 5.1 6.1 3.1 9.6 1.3
5.3 I 8 1.9 10.1 9.2 6.1
Mean SE
1.6 0.1
1.3 0.5
6.5 0.1
1.1 0.6
1.1
(50 + 5 before,
51 f 3 after guar-
gum). In the postabsorptive state, at comparable insulin infusion rates, plasma glucose (102 + 6 mg/dl before, 102 k 10 mg/dl after guar-gum) and free insulin concentrations (Table 4) were identical before and after the high fiber diet as well as plasma amino acid and whole-blood intermediary metabolite concentrations (Tables 4,5). The only change observed after treatment was a significant decrease in the fasting levels of cholesterol, from 165 ? 12 to 142 k 12 mg/dl (P < 0.005, Fig. 1). In contrast, serum triglycerides remained unchanged (Fig. 1). Clamp studies
gum. In general, they complained of flatulence and meteorism only during the first few days of treatment. Only one patient had to reduce the guar-gum to 10 g/day because of abdominal discomfort. No patient had to stop guar-gum intake. Patients’ body weights did not change significantly during the study (75.5 kg before, to 76.8 kg, after guar-gum, NS). The addition of guar-gum failed to improve the metabolic control of diabetes, as judged by HbA,, levels (Table 3). Insulin dosage was also
During the clamp, insulin was raised to comparable levels in the two studies (Table 4). Plasma glucose was also clamped at comparable values before (88 + 3 mg/dl, coefficient of variation : 7.5 k 2% and 83 + 3% mg/dl, coefficient of variation 10 + 2%) after guar-gum treatment. Plasma amino acid and lactate concentrations (Tables 4, 5) increased (P < 0.05) to values commonly found after a meal, while acetoacetate, FFA and glycerol decreased. P-OH and pyruvate concentrations did not change (Table 4). Fiber treatment failed to alter the concentrations of these metabolites following the euglycemic, hyperinsulinemic, hyperaminoacidemic clamp
TABLE 4 Free-insulin and metabolite concentrations postabsorptive state and during the clamp
(means k SE) in diabetic subjects
Normal fiber Basal Free-insulin (uU/ml) Lactate (mM/l) Pyruvate (mM/l) BGH (mM/l) AcAc (mM/l) Glycerol (mM/l) FFA (mM/l) * P < 0.05 vs basal.
14.8 0.63 0.03 0.14 0.13 0.05 0.340
High fiber Clamp
f 2.5 f 0.1 5 0.006 * 0.05 f 0.02 & 0.006 f 0.10
before and after high fiber diet, in the basal
59.2 0.91 0.025 0.03 0.01 0.04 0.051
Basal f 6.4* + 0.06* k 0.005 + 0.003 + 0.01* & 0.002* It: 0.01*
13.18 0.53 0.04 0.08 0.11 0.06 0.225
Clamp k f f f f + k
2.4 0.05 0.002 0.03 0.01 0.009 0.01
5.1* 0.01* 0.01 0.006 + 0.008* 0.01 0.04 + 0.009* 0.043 + 0.01*
51.6 1.09 0.06 0.03
+ * + f
19 TABLE 5 Plasma amino acid concentrations (means + SE) in the diabetic subjects before and after high fiber diet, in the basal post-absorptive state and during the clamp Standard
Valine @M/l) Leucine Isoleucine Phenyl-Alanine Tryptophan Lysine Proline Alanine Glycine
diet
Basal
Clamp
Basal
Clamp
301 + 20 145+ 11 65 + 7 40 f 4 52+ 7 121 f 13 238 + 27 17Ok 26 178k13
497 f 12 218_+ 7 196 + 12 36+ 7 62+ 7 170 f 24 476 f 40 29Ok21 322k25
270 k 17 120_+ 6 53 f 3 32+ 3 45+ 8 945 7 195 + 27 158 f 20 158+13
483 k 35 202_+12 191 f 10 30* 2 53+ 9 155& 17 409 k 63 296 + 24 322k31
< ’ ’ 0.005
200
High fiber diet
T
m
STANDARD
DIET
HIGH
DIET
FIBER
150 n
100
z) 50
CHOLESTEROL
TRIGLYCERIDES
Fig. 1. Postabsorptive plasma cholesterol and triglyceride concentrations before and after high fiber diet.
(Tables 4, 5). Therefore also insulin sensitivity was measured at comparable circulating substrate levels. Insulin-mediated glucose disposal, an index of peripheral insulin sensitivity, did not significantly change after treatment (Table 3). Discussion This study indicates that a long-term treatment with guar-gum does not improve glucose metabolism in type 1 diabetes, as reflected by both metabolic control and peripheral insulin sensitivity. This finding is in agreement with the results of
some studies performed both in IDDM and NIDDM [ 10-121, although other authors have found favorable effects following the supplementation of the diet with extractive fibers [ 1,2,4]. The discrepancy among the various studies could be due to differences in patient selection, in the length of treatment or in the amount of fiber administered. The patients selected in our study had, on average, a moderately poor metabolic control (HbA,, = 7.6 + 0.7x, normal range 46.5%). In contrast, most studies that reported a beneficial effect of fibers included patients with much worse HbA,, levels [5]. Also two out of three of our patients with the highest basal level of HbA,, (10.2% and 9%) showed a marked improvement of both HbA,, (5.8% and 7.8%, respectively) and insulin sensitivity (6.1-+ 7.9 and 3.7 --* 10.1 mg/kg emin). Therefore, it is possible that high fiber diets may be useful only in poorlycontrolled diabetic patients. As regards the duration of treatment, it appears also from previous studies [28,29] that a possible initial beneficial effect of high dietary fibers is lost with a more prolonged treatment. As concerns the total amount of dietary fibers, some of the authors that reported favorable results have raised dietary fiber content to a higher level than that achieved in our study [ 301. In fact, we have previously observed in type 1 diabetes a significant decrease of the plasma levels of HbA,, after doubling, for one month, the content of natural fibers of the diet, from 30 to 70 g per day [31]. However, since large amounts of dietary fiber are not tolerated in more prolonged studies because of increased side effects, like flatulence and meteorism, we have used in the present longterm study lower amounts of extractive fibers (15 g/day). According to our patients’ reports, this is probably the maximum amount of fiber supplement that they would accept on a long term basis and is the likely limit of feasibility of such studies. The only positive effect observed in our patients was the lowering of serum cholesterol. It is interesting to note that both in this and in other studies, this beneficial effect has been obtained
20 with relatively low doses of fibers and is apparently persistent for a long period [ 10,301. Therefore, a small increase in dietary fiber may be particularly beneficial as regards cholesterol also in type 1 diabetes, a disease at increased risk for macrovascular complications and at increased prevalence of risk factors for macroangiopathy. In conclusion, the addition of a relatively low amount of extractive fiber to the usual diet of type 1 diabetic patients lowers cholesterol levels but does not improve either the metabolic control or insulin sensitivity. Therefore, the costs and benefits of guar supplementation to the diets of type 1 diabetes should be reconsidered, even if the beneficial effect on plasma cholesterol level could be potentially valuable.
Acknowledgement This work was in part supported by grants of the Natural Research Council (CNR) of Italy (n. 880066804/88), Target Project on Biotechnology and Bioinstrumentation. References 1 Jenkins, D.J.A., Wolever, T.M.S, Hockaday, R., Leeds, A.R., Howarth, R., Bacon, S., Apling, EC. and Dilawai, J. (1977) Treatment of diabetes with guar gum. Lancet ii, 779-780. 2 Jenkins, D.J.A., Leeds, A.R., Gassull, M.A., Cachet, B. and Alberti, K.G.M.M. (1977) Decrease in postprandial insulin and glucose concentrations by guar and pectin. Ann. Intern. Med., 86, 20-23. 3 Jenkins, D.J.A., Leeds, A.R., Wolever, T.M.S., Goff, D.V., Alberti, K.G.M.M., Gassull, M.A. and Hockaday, T.D.R. (1976) Unabsorbable carbohydrates and diabetes: decreased postprandial hyperglycaemia. Lancet, 1, 172-174. Jenkins, D.J.A., Wilever, T.M.S., Leeds, A.R., Gassull, M.A., Haisman, P., Dilawari, J., Goff, D.V., Metz, G.L. and Alberti, K.G.M.M. (1978) Dietary fibres, fibre analogues and glucose tolerance: importance of viscosity. Br. Med. J., 1, 1392-1394. Najemnik, C., Kritz, H., Irsigler, K., Laube, H., Knick, B., Klimm, H.D., Wahl, P., Vollmar, J. and Brauning, C. (1984) Guar and its effects on metabolic control in type II diabetic subjects. Diabetes Care, 7, 215-220. Aro, A., Uusitupa, M., Voutilainen, E., Hersio, K.,
7
8
9
10
11
12
13 14
15
16
17
18
Korhonen, T. and Siitonen, 0. (1981) Improved diabetic control and hypocholesterolemic effect induced by alongterm dietary supplementation with guar in type 2 (insulinindependent) diabetes. Diabetologia, 21 (I), 29-33. Jenkins, D.J.A., Reynold, D., Slavin, B., Leeds, A.R., Jenkins, A.L., and Jepson, E.M. (1980) Dietary fiber and blood lipids: treatment ofhypercholesterolemia with guar crispbread. Am. J. Clin. Nutr. 33, (3), 575-581. Jenkins, D.J.A., Wolever, T.M., Taylor, R.H., Reynolds, L.K., Nineham, R. and Hockaday, T.D. (1980) Diabetic glucose control, lipids and trace elements on long term guar. Br. Med. J., 280, 1353-1354. Gatti, E., Catenazzo, G., Camisasca, E., Torri, A., Denegri, E. and Sirtori, CR. (1984) Effects of guarenriched pasta in the treatment of diabetes and hyperlipidemia. Ann. Nutr. Metab. 28, l-10. Hollenbeck, C.B., Coulston, A.M. and Reaven, G.M. (1986) To what extent does increased dietary fiber improve glucose and lipid metabolism in patients with non insulin-dependent diabetes mellitus (NIDDM)? Am. J. Clin. Nutr. 43 (1), 16-24. Scott, A.R., Attenborough, Y., Peacock, I., Fletcher, E., Jeffcoate, W.J. and Tattersall R.B. (1988) Comparison of high fibre diets, basal insulin supplements, and flexible insulin treatment for non-insulin dependent (type II) diabetics poorly controlled with sulphonylureas. Br. Med. J. 297, 707-710. Venhaus, A. and Chantelau E. (1988) Self-selected unrefined and refined carbohydrate diets do not affect metabolic control in pump-treated diabetic patients. Diabetologia 31, 153-157. Jenkins, D.J.A. and Jenkins, A.L. (1986). Fiber and Guar. Alim. Nutr. Metab., 7, 9-17. Pedersen, O., Hjollund, E., Lindskov, H.O., Helms, P., Sorensens, N.S., and Ditzel, J. (1982) Increased insulinreceptor binding to monocytes from insulin-dependent diabetic patients after a low-fat, high starch, high fiber diet. Diabetes Care, 5, 284-291. Christiansen, J.S., Bonnevie-Neilsen, V., Svendsen, P.A., Rubin, P., Ronn, B. and Nerup. J. (1980) Effect of guar gum on 24-hour insulin requirements of insulin-dependent diabetic subjects as assessed by artificial pancreas. Diabetes Care 3, 659-662. Tagliaferro, V., Cassader, M., Bozzo, C., Pisu, E., Bruno, A., Marena, S., Cavallo-Perin, P., Cravero, L. and Pagano, G. (1985) Moderate guar-gum addition to usual diet improves peripheral sensitivity to insulin and lipaemic profile in NIDDM. Diabete Metab. 11 (6), 380-385. Lloyd, B., Burrin, J. and Smythe, P. (1978) Enzymic fluorometric continuous-flow assays for blood glucose, lactate, pyruvate, alanine, glycerol and 3-hydroxybutyrate. Clin. Chem. 24, 1724-1729. Price, C.P., Lloyd, B. and Alberti K.G.M.M. (1977) A kinetic spectrophotometric assay for rapid determination of acetoacetate in blood. Clin. Chem., 23n 10, 1893-1897.
21 19 Shimuzu, S., Inoue, E., Tani, Y. and Yamaha H. (1979). Enzymatic microdetermination of serum free fatty acids. Anal. Biochem. 98, 341-345. 20 Huggett, A.S.G. and Nixon, D.A. (1957) Use of glucose oxidase, peroxidase and o-dianisidine in determination of blood and urinary glucose. Lancet ii, 368-370. 21 Bio-Rad (n.d.) Hemoglobin A,, by Column test. 22 Kuzuya, T., J.S. Matzuda and S. Yoshida. (1976) Human C-peptide immunoreactivity (CPR) in blood and urine. Evaluation of a RIA method and its clinical application. Diabetologia, 12, 51 l-518. 23 Nakagawa, S., Nakayana, H., Sasaky, T., Yochico, K., Shimosaki, K., Akoi, S. and Moshino, K. (1973). A single method for the determination of serum free insulin levels in insulin treated patients. Diabetes, 22, 540-560. 24 Allain, C.C., Poon, L.S., Chan, C.S. and Richmond, W. (1974) Enzymatic determination of total serum cholesterol. Clin. Chem., 20, 470-475. 25 Wahlefeld, A.W. (1978) Triglycerides: determination after enzymatic hydrolysis. In: H.U. Bergmeyer (Ed.), Methoden der Enzymatischen Analyse, 3rd edit; Verlag Chemie, Weinheim. 26 Olson, D.C., Schmidt, G.J. and Slavin, W. (1979) The determination of amino acids in physiological fluids using
27
28
29
30
31
liquid chromatography and fluorescence detection. Chromatogr. Newslett. 7, 22-28. Rizza, R.A., Mandarino, L.J. and Gerich, J.E. (1981) Dose-response characteristics for effects of insulin on production and utilization for glucose in man. Am. J. Physiol. 240, E630-E639. Gardner, D.F., Schwartz, L., Krista, M., and Merimee, T.J. (1984) Dietary pectin and glycemic control in diabetes. Diabetes Care, 7, 143-146. Mahalko, J.R., Sandstead, H.H., Johnson, L.K., Imman, L.F., Milne, D.B., Warner, R.C. and Haunz, E.A. (1984) Effects of consuming fiber from corn bran, soy hulls, or apple powder on glucose tolerance and plasma lipids in type II diabetes. Am. J. Clin. Nutr., 39, 25-34. Simpson, H.C.R., Simpson, R.W., Lousley, S., Carter, R.D.,Geekie,M.,Hockaday,T.D. and Mann, J.I. (1981) High carbohydrate leguminous fibre diets improve all aspects of diabetic control. Lancet 28, 1 (8222) 73 l-732. Bruttomesso, D., Briani, G., Bilardo, G., Vitale, E., Lavagnini, T., Marescotti, C., Duner, E., Giorato, C. and Tiengo, A. (1989) The medium-term effect of natural or extractive dietary fibres on plasma amino acids and lipids in type 1 diabetes. Diabetes Res. Clin. Pratt. 6. 149-155.
0 1991 Elsevier Science Publishers
B.V. 016%8227/91/$03.50
15
DIABET 00509
No effects of high-fiber diets on metabolic control and insulin-sensitivity in type 1 diabetic subjects D. Bruttomesso,
G. Biolo, S. Inchiostro, C. Fongher, G. Briani, E. Duner, M.C. Marescotti, E. Iori, A. Tiengo and P. Tessari
Cattedra di Malattie de1 Ricambio. Istituto di Medicina Clinica, University of Padova, Padova, Italy
(Received 18 September 1990) (Revision received 13 February 1991) (Accepted 22 February 1991)
Summary The metabolic effects of a three-month treatment with a high-fiber diet (15 grams of guar-gum added to a standard diet) were investigated in seven type 1 diabetic subjects, with a moderately poor metabolic control. HbA,, levels, daily insulin requirement, cholesterol, triglyceride, amino acid and intermediate metabolite concentrations were evaluated before and following the high fiber diet, both in the postabsorptive state at euglycemia and during a euglycemic, hyperinsulinemic, hyperaminoacidemic clamp. Insulinmediated glucose utilization, an index of insulin-sensitivity, was also measured during the clamp. Following the diet, no differences in HbA,, levels (7.6 + 0.7% -+ 7.3 + 0.6%), daily insulin requirement (50 k 5 -_*5 1 k 3 U/d), triglyceride, amino acid and intermediary metabolite concentrations in the basal, euglycemic state, were observed. Only cholesterol concentrations decreased significantly (from 165 + 12 to 142 k 12 mg/dl, P -c 0.01) after the diet. During the clamp, the concentrations of all measured substrates were comparable before and after high fiber treatment. Insulin-mediated glucose disposal was also unchanged by guar-gum treatment. Patients’ body weights were not modified by the diet. In conclusion, our study shows that a high fiber diet, obtained with the addition of 15 grams of guar-gum to a standard diet, is of no benefit to IDDM either as regards the metabolic control or insulin sensitivity. Only cholesterol levels were decreased. Therefore, the costs and benefits of these diets in the treatment of IDDM should be reconsidered. Key words: High-fiber diet; Metabolic
Correspondence
control;
Insulin-sensitivity;
to: D. Bruttomesso, Cattedra di Malattie del Ricambio, Istituto di Medicina Clinica, University of Padova, Padova, Italy.
Diabetes mellitus
Introduction It has been shown by several authors that dietary fibers such as guar and pectins improve glucose metabolism [ l-51 and decrease serum cholesterol [6-91 in patients with diabetes mellitus.
16
Recent studies, however, have failed to support this conclusion, [lo-121 and, at this moment, the reasons for these discrepancies are not apparent. Differences in patient selection, fiber dosage and length of treatment might all explain these differences. The presumed advantage of increasing the fiber content of the diet is that fibers slow the intestinal absorption of carbohydrates and smooth the postprandial rise of plasma glucose and insulin concentrations [ 31. Furthermore, dietary fibers reduce the pool of cholesterol precursors by increasing the fecal loss of bile-salts [ 131. It has been also proposed that fibers could improve insulin sensitivity [ 14-161, but this possibility has not been extensively evaluated, in particular in insulin-dependent diabetes mellitus (IDDM). In this report we have reconsidered the longterm effects of guar-gum, added to a standard diet, on metabolic control, daily insulin requirement, insulin sensitivity, serum lipids, plasma amino acids and intermediary metabolite profiles in patients with type 1 diabetes. Insulin sensitivity was evaluated with the euglycemic hyperinsulinemic glucose clamp. The clamp technique was modified by infusing an amino acid mixture together with insulin and glucose, so as to assess insulin sensitivity under conditions that reproduce the post-prandial hyperaminoacidemia, while by-passing the possible effects of fibers on the intestinal absorption of nutrients.
Protocol Seven male insulin-dependent diabetic patients, of a mean age of 30 k 2 (m k SE) years, were studied. Their body mass index was normal (Table 1). In all, the plasma level of C-peptide stimulated by i.v. glucagon was undetectable (< 0.3 ng/ml). In four of them the metabolic control was fairly good (HbA,. 4-7.7% ; normal range : 4-6.5 %), in three it was poor (HbA,, > 9.0%). No patients had any evidence of autonomic neuropathy. They were admitted to our metabolic unit to obtain a baseline metabolic profile some days before the study, while following their usual diet (about 2000 Kcal, containing - 50% carbohydrates, - 30% lipids, - 20% proteins and -30 g of natural fibers). Blood samples were obtained for the assay of fasting plasma cholesterol, triglycerides, HbA,, and amino acid levels. On the evening before the clamp, the patients received at dinner one half of their usual dose of regular insulin, while longacting and intermediate insulins were withheld. During the night, euglycemia was maintained with the intravenous infusion of regular insulin at the rate of l-2 U/h. In the morning of the study, four basal samples for the determination of substrate and hormone concentrations were taken over 30 min with the patients kept at euglycemic conditions. Thereafter, a euglycemic, hyperinsulinemic, hyper-
TABLE 1 Clinical characteristics Subjects
of the diabetic patients Age (yrs)
BMI @g/m* )
Duration of diabetes
HbA,, (%I
Ws) M.A. F.C. T.V. C.L. C.M. C.S. M.A.
20 25 28 32 28 35 40
24.5 24.2 27.1 23.9 21.8 22.6 27.7
9 2 10 9 1.5 7 7
M f SE
29.7 k 2.4
24.5 t 0.8
6.5 + 1.3
Insulin requirement w/24
6.8 4.6 6.1 10.2 9.0 7.7 9.0 7.6 + 0.7
63 50 63 63 53 35 32 50+ 5
h)
17
aminoacidemic clamp was started. Insulin (Actrapid, Novo, Bagswaert, Denmark), was infused intravenously at the rate of 0.05 U/m?&’ for 3 h (O-180 min) by means of a calibrated Harvard pump (Harvard Apparatus, South Nartick, MA, U.S.A.). Insulin was dissolved in 0.9% saline solution along with 2 ml of the patient’s blood, to prevent absorption of the hormone to the plastic tubing. Blood glucose was kept at -90 mg/dl by means of a Biostator GCIIS (Miles Laboratories, Elkhart, IN). During the clamp, an amino acid solution was also infused intravenously, at a rate corresponding to a total nitrogen infusion rate of 0.23 mg/kg * min by means of a Harvard pump. The composition of the amino acid mixture is reported on Table 2. This amino acid infusion rate was chosen in order to increase the concentration of most amino acids by SO-150%, i.e., similar to post-prandial values. Blood samples were drawn every 30 min for assay of glucose, amino acids, intermediary metabolites and free insulin. Four samples were also taken over 30 min at the end of the clamp. These values were averaged and used for the comparison to the basal data. At the end of the study day, the patients were dismissed with their usual diet. In addition, they were asked to take 5 g of guar-gum, as suspension in water, 3 times a day immediately before breakfast, lunch and dinner. Therefore, the total fiber content of the diet was increased from 30 g of natural fibers per day to a total of 45 g (30 g/natural and 15 g/extractive). The patients were contacted by telephone by one of the investigators about compliance to diet TABLE 2 Amino acid concentrations L-alanine L-arginine L-phenylalanine Glycine L-isoleucine L-histidine L-leucine
84.3 34.5 6.1 120.0 68.7 15.5 83.4
in the infused solution (mmol/l) L-lysine L-methionine L-proline L-serine L-threonine L-tryptophan L-valine
41.6 6.1 69.6 41.6 37.8 3.1 71.8
and drug prescription at least once a month. The patients were also recommended not to change their usual insulin dosage, except for major derangement in the metabolic control. The whole study, i.e., the metabolic profile and the euglycemic, hyperaminoacidemic, hyperinsulinemic clamp was repeated after 3 months of treatment during a short admission to our metabolic ward. Methods
Whole blood pyruvate, lactate, 3/$hydroxybutyrate @-OH), acetoacetate (AcAc) and glycerol were analyzed using an enzymatic fluorometric method [ 17,181. Plasma free fatty acid concentrations were measured according to Shimuzu [ 191. Plasma glucose was determined using the glucose oxidase method of the Beckman Glucose Analyzer II [20]. The Bio-Rad column test [21] was used for determining HbA,. Plasma C-peptide concentration was evaluated according to Kuzuya et al. [22]. Free insulin was determined according to Nakagawa et al. [24] and triglycerides [ 251 were determined using enzymatic methods. Plasma amino acids were determined liquid chromatography by high-performance (HPLC) [26]. Insulin-mediated glucose disposal (M) was determined by averaging the exogenous glucose infusion rates (mg/kg . min), at 10 min intervals, in the last hour of the clamp. At the plasma insulin our achieved in study concentrations (N 60 pU/ml), endogenous glucose production is virtually suppressed [ 271. Therefore, the M value represents total glucose utilization. The statistical analysis was performed using the two-tailed Student c-test for paired data in comparison between the two groups before and after treatment. A P value equal or less than 0.05 was considered as statistically significant. Results Basal euglycemic state
The patients
showed a good tolerance
to guar-
18
TABLE 3
unchanged
Haemoglobin A,, and insulin mediated glucose (mean k SE) before and after high fiber diet HbA,, Subject
disposal
M (mg/kg min)
(%)
Diet Standard
High fiber
Standard
High fiber
M.A. F.C. T.V. C.L. C.M. C.S. M.A.
6.84 4.6 6.12 10.2 9 9
8.13 5.2 1.1 5.81 1.8 8 8.2
6.5 1.5 5.1 6.1 3.1 9.6 1.3
5.3 I 8 1.9 10.1 9.2 6.1
Mean SE
1.6 0.1
1.3 0.5
6.5 0.1
1.1 0.6
1.1
(50 + 5 before,
51 f 3 after guar-
gum). In the postabsorptive state, at comparable insulin infusion rates, plasma glucose (102 + 6 mg/dl before, 102 k 10 mg/dl after guar-gum) and free insulin concentrations (Table 4) were identical before and after the high fiber diet as well as plasma amino acid and whole-blood intermediary metabolite concentrations (Tables 4,5). The only change observed after treatment was a significant decrease in the fasting levels of cholesterol, from 165 ? 12 to 142 k 12 mg/dl (P < 0.005, Fig. 1). In contrast, serum triglycerides remained unchanged (Fig. 1). Clamp studies
gum. In general, they complained of flatulence and meteorism only during the first few days of treatment. Only one patient had to reduce the guar-gum to 10 g/day because of abdominal discomfort. No patient had to stop guar-gum intake. Patients’ body weights did not change significantly during the study (75.5 kg before, to 76.8 kg, after guar-gum, NS). The addition of guar-gum failed to improve the metabolic control of diabetes, as judged by HbA,, levels (Table 3). Insulin dosage was also
During the clamp, insulin was raised to comparable levels in the two studies (Table 4). Plasma glucose was also clamped at comparable values before (88 + 3 mg/dl, coefficient of variation : 7.5 k 2% and 83 + 3% mg/dl, coefficient of variation 10 + 2%) after guar-gum treatment. Plasma amino acid and lactate concentrations (Tables 4, 5) increased (P < 0.05) to values commonly found after a meal, while acetoacetate, FFA and glycerol decreased. P-OH and pyruvate concentrations did not change (Table 4). Fiber treatment failed to alter the concentrations of these metabolites following the euglycemic, hyperinsulinemic, hyperaminoacidemic clamp
TABLE 4 Free-insulin and metabolite concentrations postabsorptive state and during the clamp
(means k SE) in diabetic subjects
Normal fiber Basal Free-insulin (uU/ml) Lactate (mM/l) Pyruvate (mM/l) BGH (mM/l) AcAc (mM/l) Glycerol (mM/l) FFA (mM/l) * P < 0.05 vs basal.
14.8 0.63 0.03 0.14 0.13 0.05 0.340
High fiber Clamp
f 2.5 f 0.1 5 0.006 * 0.05 f 0.02 & 0.006 f 0.10
before and after high fiber diet, in the basal
59.2 0.91 0.025 0.03 0.01 0.04 0.051
Basal f 6.4* + 0.06* k 0.005 + 0.003 + 0.01* & 0.002* It: 0.01*
13.18 0.53 0.04 0.08 0.11 0.06 0.225
Clamp k f f f f + k
2.4 0.05 0.002 0.03 0.01 0.009 0.01
5.1* 0.01* 0.01 0.006 + 0.008* 0.01 0.04 + 0.009* 0.043 + 0.01*
51.6 1.09 0.06 0.03
+ * + f
19 TABLE 5 Plasma amino acid concentrations (means + SE) in the diabetic subjects before and after high fiber diet, in the basal post-absorptive state and during the clamp Standard
Valine @M/l) Leucine Isoleucine Phenyl-Alanine Tryptophan Lysine Proline Alanine Glycine
diet
Basal
Clamp
Basal
Clamp
301 + 20 145+ 11 65 + 7 40 f 4 52+ 7 121 f 13 238 + 27 17Ok 26 178k13
497 f 12 218_+ 7 196 + 12 36+ 7 62+ 7 170 f 24 476 f 40 29Ok21 322k25
270 k 17 120_+ 6 53 f 3 32+ 3 45+ 8 945 7 195 + 27 158 f 20 158+13
483 k 35 202_+12 191 f 10 30* 2 53+ 9 155& 17 409 k 63 296 + 24 322k31
< ’ ’ 0.005
200
High fiber diet
T
m
STANDARD
DIET
HIGH
DIET
FIBER
150 n
100
z) 50
CHOLESTEROL
TRIGLYCERIDES
Fig. 1. Postabsorptive plasma cholesterol and triglyceride concentrations before and after high fiber diet.
(Tables 4, 5). Therefore also insulin sensitivity was measured at comparable circulating substrate levels. Insulin-mediated glucose disposal, an index of peripheral insulin sensitivity, did not significantly change after treatment (Table 3). Discussion This study indicates that a long-term treatment with guar-gum does not improve glucose metabolism in type 1 diabetes, as reflected by both metabolic control and peripheral insulin sensitivity. This finding is in agreement with the results of
some studies performed both in IDDM and NIDDM [ 10-121, although other authors have found favorable effects following the supplementation of the diet with extractive fibers [ 1,2,4]. The discrepancy among the various studies could be due to differences in patient selection, in the length of treatment or in the amount of fiber administered. The patients selected in our study had, on average, a moderately poor metabolic control (HbA,, = 7.6 + 0.7x, normal range 46.5%). In contrast, most studies that reported a beneficial effect of fibers included patients with much worse HbA,, levels [5]. Also two out of three of our patients with the highest basal level of HbA,, (10.2% and 9%) showed a marked improvement of both HbA,, (5.8% and 7.8%, respectively) and insulin sensitivity (6.1-+ 7.9 and 3.7 --* 10.1 mg/kg emin). Therefore, it is possible that high fiber diets may be useful only in poorlycontrolled diabetic patients. As regards the duration of treatment, it appears also from previous studies [28,29] that a possible initial beneficial effect of high dietary fibers is lost with a more prolonged treatment. As concerns the total amount of dietary fibers, some of the authors that reported favorable results have raised dietary fiber content to a higher level than that achieved in our study [ 301. In fact, we have previously observed in type 1 diabetes a significant decrease of the plasma levels of HbA,, after doubling, for one month, the content of natural fibers of the diet, from 30 to 70 g per day [31]. However, since large amounts of dietary fiber are not tolerated in more prolonged studies because of increased side effects, like flatulence and meteorism, we have used in the present longterm study lower amounts of extractive fibers (15 g/day). According to our patients’ reports, this is probably the maximum amount of fiber supplement that they would accept on a long term basis and is the likely limit of feasibility of such studies. The only positive effect observed in our patients was the lowering of serum cholesterol. It is interesting to note that both in this and in other studies, this beneficial effect has been obtained
20 with relatively low doses of fibers and is apparently persistent for a long period [ 10,301. Therefore, a small increase in dietary fiber may be particularly beneficial as regards cholesterol also in type 1 diabetes, a disease at increased risk for macrovascular complications and at increased prevalence of risk factors for macroangiopathy. In conclusion, the addition of a relatively low amount of extractive fiber to the usual diet of type 1 diabetic patients lowers cholesterol levels but does not improve either the metabolic control or insulin sensitivity. Therefore, the costs and benefits of guar supplementation to the diets of type 1 diabetes should be reconsidered, even if the beneficial effect on plasma cholesterol level could be potentially valuable.
Acknowledgement This work was in part supported by grants of the Natural Research Council (CNR) of Italy (n. 880066804/88), Target Project on Biotechnology and Bioinstrumentation. References 1 Jenkins, D.J.A., Wolever, T.M.S, Hockaday, R., Leeds, A.R., Howarth, R., Bacon, S., Apling, EC. and Dilawai, J. (1977) Treatment of diabetes with guar gum. Lancet ii, 779-780. 2 Jenkins, D.J.A., Leeds, A.R., Gassull, M.A., Cachet, B. and Alberti, K.G.M.M. (1977) Decrease in postprandial insulin and glucose concentrations by guar and pectin. Ann. Intern. Med., 86, 20-23. 3 Jenkins, D.J.A., Leeds, A.R., Wolever, T.M.S., Goff, D.V., Alberti, K.G.M.M., Gassull, M.A. and Hockaday, T.D.R. (1976) Unabsorbable carbohydrates and diabetes: decreased postprandial hyperglycaemia. Lancet, 1, 172-174. Jenkins, D.J.A., Wilever, T.M.S., Leeds, A.R., Gassull, M.A., Haisman, P., Dilawari, J., Goff, D.V., Metz, G.L. and Alberti, K.G.M.M. (1978) Dietary fibres, fibre analogues and glucose tolerance: importance of viscosity. Br. Med. J., 1, 1392-1394. Najemnik, C., Kritz, H., Irsigler, K., Laube, H., Knick, B., Klimm, H.D., Wahl, P., Vollmar, J. and Brauning, C. (1984) Guar and its effects on metabolic control in type II diabetic subjects. Diabetes Care, 7, 215-220. Aro, A., Uusitupa, M., Voutilainen, E., Hersio, K.,
7
8
9
10
11
12
13 14
15
16
17
18
Korhonen, T. and Siitonen, 0. (1981) Improved diabetic control and hypocholesterolemic effect induced by alongterm dietary supplementation with guar in type 2 (insulinindependent) diabetes. Diabetologia, 21 (I), 29-33. Jenkins, D.J.A., Reynold, D., Slavin, B., Leeds, A.R., Jenkins, A.L., and Jepson, E.M. (1980) Dietary fiber and blood lipids: treatment ofhypercholesterolemia with guar crispbread. Am. J. Clin. Nutr. 33, (3), 575-581. Jenkins, D.J.A., Wolever, T.M., Taylor, R.H., Reynolds, L.K., Nineham, R. and Hockaday, T.D. (1980) Diabetic glucose control, lipids and trace elements on long term guar. Br. Med. J., 280, 1353-1354. Gatti, E., Catenazzo, G., Camisasca, E., Torri, A., Denegri, E. and Sirtori, CR. (1984) Effects of guarenriched pasta in the treatment of diabetes and hyperlipidemia. Ann. Nutr. Metab. 28, l-10. Hollenbeck, C.B., Coulston, A.M. and Reaven, G.M. (1986) To what extent does increased dietary fiber improve glucose and lipid metabolism in patients with non insulin-dependent diabetes mellitus (NIDDM)? Am. J. Clin. Nutr. 43 (1), 16-24. Scott, A.R., Attenborough, Y., Peacock, I., Fletcher, E., Jeffcoate, W.J. and Tattersall R.B. (1988) Comparison of high fibre diets, basal insulin supplements, and flexible insulin treatment for non-insulin dependent (type II) diabetics poorly controlled with sulphonylureas. Br. Med. J. 297, 707-710. Venhaus, A. and Chantelau E. (1988) Self-selected unrefined and refined carbohydrate diets do not affect metabolic control in pump-treated diabetic patients. Diabetologia 31, 153-157. Jenkins, D.J.A. and Jenkins, A.L. (1986). Fiber and Guar. Alim. Nutr. Metab., 7, 9-17. Pedersen, O., Hjollund, E., Lindskov, H.O., Helms, P., Sorensens, N.S., and Ditzel, J. (1982) Increased insulinreceptor binding to monocytes from insulin-dependent diabetic patients after a low-fat, high starch, high fiber diet. Diabetes Care, 5, 284-291. Christiansen, J.S., Bonnevie-Neilsen, V., Svendsen, P.A., Rubin, P., Ronn, B. and Nerup. J. (1980) Effect of guar gum on 24-hour insulin requirements of insulin-dependent diabetic subjects as assessed by artificial pancreas. Diabetes Care 3, 659-662. Tagliaferro, V., Cassader, M., Bozzo, C., Pisu, E., Bruno, A., Marena, S., Cavallo-Perin, P., Cravero, L. and Pagano, G. (1985) Moderate guar-gum addition to usual diet improves peripheral sensitivity to insulin and lipaemic profile in NIDDM. Diabete Metab. 11 (6), 380-385. Lloyd, B., Burrin, J. and Smythe, P. (1978) Enzymic fluorometric continuous-flow assays for blood glucose, lactate, pyruvate, alanine, glycerol and 3-hydroxybutyrate. Clin. Chem. 24, 1724-1729. Price, C.P., Lloyd, B. and Alberti K.G.M.M. (1977) A kinetic spectrophotometric assay for rapid determination of acetoacetate in blood. Clin. Chem., 23n 10, 1893-1897.
21 19 Shimuzu, S., Inoue, E., Tani, Y. and Yamaha H. (1979). Enzymatic microdetermination of serum free fatty acids. Anal. Biochem. 98, 341-345. 20 Huggett, A.S.G. and Nixon, D.A. (1957) Use of glucose oxidase, peroxidase and o-dianisidine in determination of blood and urinary glucose. Lancet ii, 368-370. 21 Bio-Rad (n.d.) Hemoglobin A,, by Column test. 22 Kuzuya, T., J.S. Matzuda and S. Yoshida. (1976) Human C-peptide immunoreactivity (CPR) in blood and urine. Evaluation of a RIA method and its clinical application. Diabetologia, 12, 51 l-518. 23 Nakagawa, S., Nakayana, H., Sasaky, T., Yochico, K., Shimosaki, K., Akoi, S. and Moshino, K. (1973). A single method for the determination of serum free insulin levels in insulin treated patients. Diabetes, 22, 540-560. 24 Allain, C.C., Poon, L.S., Chan, C.S. and Richmond, W. (1974) Enzymatic determination of total serum cholesterol. Clin. Chem., 20, 470-475. 25 Wahlefeld, A.W. (1978) Triglycerides: determination after enzymatic hydrolysis. In: H.U. Bergmeyer (Ed.), Methoden der Enzymatischen Analyse, 3rd edit; Verlag Chemie, Weinheim. 26 Olson, D.C., Schmidt, G.J. and Slavin, W. (1979) The determination of amino acids in physiological fluids using
27
28
29
30
31
liquid chromatography and fluorescence detection. Chromatogr. Newslett. 7, 22-28. Rizza, R.A., Mandarino, L.J. and Gerich, J.E. (1981) Dose-response characteristics for effects of insulin on production and utilization for glucose in man. Am. J. Physiol. 240, E630-E639. Gardner, D.F., Schwartz, L., Krista, M., and Merimee, T.J. (1984) Dietary pectin and glycemic control in diabetes. Diabetes Care, 7, 143-146. Mahalko, J.R., Sandstead, H.H., Johnson, L.K., Imman, L.F., Milne, D.B., Warner, R.C. and Haunz, E.A. (1984) Effects of consuming fiber from corn bran, soy hulls, or apple powder on glucose tolerance and plasma lipids in type II diabetes. Am. J. Clin. Nutr., 39, 25-34. Simpson, H.C.R., Simpson, R.W., Lousley, S., Carter, R.D.,Geekie,M.,Hockaday,T.D. and Mann, J.I. (1981) High carbohydrate leguminous fibre diets improve all aspects of diabetic control. Lancet 28, 1 (8222) 73 l-732. Bruttomesso, D., Briani, G., Bilardo, G., Vitale, E., Lavagnini, T., Marescotti, C., Duner, E., Giorato, C. and Tiengo, A. (1989) The medium-term effect of natural or extractive dietary fibres on plasma amino acids and lipids in type 1 diabetes. Diabetes Res. Clin. Pratt. 6. 149-155.