0021-972X/85/6106-1165$02.00/0 Journal of Clinical Endocrinology and Metabolism Copyright © 1985 by The Endocrine Society

Vol. 61, No. 6 Printed in U.S.A.

Study of Glucose and Lipid Metabolism by Continuous Indirect Calorimetry in Graves' Disease: Effect of an Oral Glucose Load J. P. RANDIN, B. SCAZZIGA, E. JEQUIER, AND J. P. FELBER Division of Endocrinology and Clinical Biochemistry, Department of Medicine, Centre Hospitaller Universitaire Vaudois, and Institute of Physiology (E.J.), 1011 Lausanne, Switzerland

ABSTRACT. Glucose and lipid metabolism were studied in 12 patients with hyperthyroid Graves' disease for 3 h during an oral glucose tolerance test (100 g) by continuous indirect calorimetry. In the postabsorptive state, glucose oxidation was not different from that in normal subjects, but Jipid oxidation was significantly increased. Impaired glucose tolerance was found, but total glucose oxidation increased after the glucose load to 47.1 ± 2.0 (±SEM) vs. 33.4 ± 1.4 g/3 h in the control group (P < 0.001). Total glucose oxidation corresponded, in hyperthyroid patients, to the highest rate obtained with progressively increasing insulin and glucose administration in normal man. Glucose

D

storage was clearly lower in hyperthyroid patients. After treatment in 7 patients, glucose tolerance improved significantly, and the metabolic patterns almost normalized. In the 12 hyperthyroid patients and the 7 patients after treatment (n = 19), a correlation was found between total serum T3 concentration and both basal lipid oxidation and suprabasal glucose oxidation. It is concluded that the decrease in glucose tolerance in hyperthyroidism cannot be explained by an alteration in glucose oxidation, but, rather, by a defect in nonoxidative glucose uptake in the periphery. (J Clin Endocrinol Metab 6 1 : 1165, 1985)

ECREASED glucose tolerance is often found in patients with hyperthyroidism (1, 2). The pathogenic mechanisms are poorly understood and probably are multiple. Possible mechanisms include an increase in the gastric emptying rate for glucose (3), abnormal insulin metabolism (4), and abnormal insulin secretion (5, 6), but abnormalities in some of these parameters have not been reported by others (7, 8). Increased glucagon secretion or action has been recently suggested as additional mechanisms for glucose intolerance in hyperthyroidism (1, 9). An important increase in lipid metabolism has been demonstrated by measurement of FFA turnover (10,11), and we consider preferential oxidation of FFA as an explanation for the decreased glucose tolerance in hyperthyroidism (8). To further address this issue, it seems essential to study different aspects of the metabolism of lipids and glucose and, in particular, to measure the rate of glucose and lipid oxidation. Variable conclusions have resulted from isotopic glucose measurements and forearm glucose uptake studies in man (12, 13) as well as from in vitro studies (14, 15). The aim of this study was to measure the amounts of glucose and lipid oxidized in patients

Received April 22,1985. Address requests for reprints to: Dr. J. P. Felber, Division of Endocrinology and Clinical Biochemistry, Centre Hospitalier Universitaire Vaudois and Institute of Physiology, Lausanne 1011, Switzerland.

with hyperthyroidism during an oral glucose tolerance test using the noninvasive method of continuous indirect calorimetry. Materials and Methods Subjects Five men and 7 women with hyperthyroidism due to Graves' disease were studied. The diagnosis of hyperthyroidism was based on clinical findings and the results of thyroid function tests. All patients were newly diagnosed and had diffuse goiter demonstrating uniform increased uptake of 131I and elevated serum total T4 [272 ± 23 (±SEM) nmol/liter] and serum total T3 (7.9 ± 0.7 nmol/liter) levels. Their age was between 20 and 57 yr, with a mean age of 37 yr. Their relative weight, as determined by dividing a patient's weight by his ideal weight (The Metropolitan Life Insurance Tables, 1959), ranged from 84-114%, with a mean of 96%. No subject had any personal or family history of diabetes. No patient took any drugs. All patients were studied on the third or fourth day after the diagnosis and had not received any antithyroid treatment. As control, 29 normal subjects were studied; 14 were men, and 15 were women. Their age ranged from 20-58 yr (mean, 37 yr), and their mean weight was similar to that of the hyperthyroid patients (61.2 ± 1.5 vs. 62.3 ± 1.5 kg). Seven patients were studied again after treatment (1 after surgical treatment and 6 after carbimazole treatment) when they were euthyroid (serum total T4, 71 ± 6 nmol/liter; serum total T3, 1.9 ± 0.08 nmol/ liter). The protocol was approved by the institutional ethical

1165

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 23 January 2017. at 00:47 For personal use only. No other uses without permission. . All rights reserved.

RANDIN ET AL.

1166

committee of the Department of Medicine of Lausanne University Hospital, and all subjects gave their informed consent before the study. Experimental protocol All normal and hyperthyroid subjects ate a weight maintenance diet containing at least 250-300 g carbohydrate/day for 3 days before each study. They were studied after an overnight fast. After a 45-min period of bed rest, the subjects were given a 100-g oral glucose load in 400 ml lemon-flavored water. Blood was withdrawn from an antecubital vein every 30 min for measurement of glucose, insulin, FFA, and glucagon starting 30 min before and for 180 min after the glucose load and at time zero for the measurement of total T4 and total T3. Urine samples were collected for the duration of the test to measure urinary glucose and nitrogen excretion. Gas exchange measurements Gas exchange measurements were performed by computerized open circuit calorimetry, as previously described (16). This method allows calculation of the total quantity of glucose oxidized basally and in response to the glucose load (suprabasal oxidation) as well as the quantity stored from the load during the 3 h of the test. The amount of suprabasal glucose oxidized was obtained by subtracting the quantity of glucose oxidized basally from the total quantity of glucose oxidized after ingestion of the substrate. The amount of glucose disposed of by nonoxidative pathways (mainly glycogen synthesis and glycolysis) during the 3-h period following the oral glucose load was calculated by subtracting the total quantity of glucose oxidized, the urinary glucose loss, and the excess glucose remaining in the glucose space from the 100 g ingested glucose (17). The glucose space was considered to represent 25% of the body weight of the subjects (18). This calculation provides quantitative information concerning the amount of glucose stored, but does not delineate the tissue sites at which the glucose is stored. It is assumed that intestinal glucose absorption is complete or close to complete after 3 h in normal or hyperthyroid patients (3,19). However, as a recent study indicated that only 80-85% of the ingested glucose is absorbed during a 3-h period (20), the absolute amount of glucose taken up in a nonoxidative way may be slightly overestimated. It should be noted that calorimetric measurements are valid even in the presence of lipogenesis (16, 21) and that the calculation of glucose storage remains quantitatively correct even under conditions of simultaneous synthesis and oxidation of lipids. When there is net gain of lipid from lipogenesis, indirect calorimetry still allows determination of the amount of glucose converted into lipid (the respiratory quotient, corrected for protein metabolism, is then higher than unity). In the presence of both lipid oxidation and lipogenesis, the method gives the net balance between these two processes (16). When net lipogenesis is occurring, at a nonprotein respiratory quotient greater than 1.00, calculations of substrate utilization/ synthesis remain valid. Gluconeogenesis may be increased (22, 23) in hyperthyroidism and may disturb the calculations (24). Moreover, even if there is substantial gluconeogenesis, the

JCE & M • 1985 Vol 61 • No 6

influence on glucose and fat oxidation is small. Although the assessment of the protein oxidation rate from urinary nitrogen excretion might not be entirely correct, variations in urinary nitrogen measurements do not significantly affect the calculation of glucose and lipid oxidation rates (25). Change in ventilation is the factor that may most influence the calorimetric measurements. This problem was circumvented by ensuring that all subjects were fully at rest for a long period (1 h) before the test and by using a ventilated hood with prolonged and continuous integrated periods of measurement. Analytical procedures Plasma and urinary glucose were measured by the hexokinase method, and total serum T4 and total serum T 3 were determined by automated RIA ARIA II (Becton Dickinson, Rutherford, NJ). Plasma immunoreactive insulin (IRI) concentrations were measured by the method of Herbert et al. (26). Plasma FFA were extracted using the method of Dole and Meinertz (27) and determined according to the procedure of Heindel et al. (28). Plasma glucagon was determined according to the method of Aguilar-Parada et al. (29). Urinary nitrogen was measured by the method of Kjeldahl (30). Statistical methods All data are presented as the mean ± SEM. Statistical comparisons between hyperthyroid and control groups were calculated using the unpaired t test and those between the treated and untreated patients were determined using the paired t test.

Results Biochemical analysis Mean fasting plasma glucose was significantly increased in the hyperthyroid group [5.3 ±0.1 (±SEM) vs. 4.8 ± 1 mmol/liter in the control group; P < 0.001]. The hyperthyroid patients had impaired glucose tolerance, with significant elevations of plasma glucose concentrations at all times during the test (Fig. 1). The plasma glucose area above the fasting level was markedly increased in the hyperthyroid patients (592 ± 83 us. 219 ± 36 mmol/liter • min in the control group; P < 0.001; Table 1). Fasting plasma IRI levels were not different (13.1 ± 2 us. 13.4 ± 1 /iU/ml in the control group). After the glucose load, the IRI curve was elevated. Expressed as the area above the fasting level, the IRI response to the glucose load was not significantly enhanced compared with that in the control group (13,400 ± 1,800 vs. 10,200 ± 900 jtU/ml-min; Table 1). Plasma fasting FFA levels were significantly higher in the hyperthyroid group (828 ± 70 vs. 581 ± 47 /imol/liter in the control group; P < 0.01); they decreased to control values 30 min after the load (Fig. 1). In the seven patients who were restudied after treatment, the mean fasting plasma glucose level was normal (4.6 ±0.1 vs. 5.1 ± 0.2 mmol/liter before treatment; P


JL

JL

_L

40

20

FIG. 2. Mean ± SEM glucose oxidation (milligrams per min) and lipid oxidation (milligrams per min) in the fasting state and during the 3 h after the glucose load (100 g) in hyperthyroid, treated hyperthyroid, and normal subjects.

JU tkh 0000 0©G©

0000

i

120

ooo p < 0.01 ) , . .. ~~« Hyperthyroid versus control oooo p < 0.001 ) e@© p < 0.01 ) ( Hyperthyroid versus treated ©see p < 0.001 )

100

•

p < 0.05

Treated versus control

80

I 60

I,

I,

40

1

o CO

•p

20

'x o d -30-0

glycogen content of hyperthyroid patients (37, 38). Similar observations have been made in animals given thyroid hormones (39, 40), and dogs treated with thyroid hormones had a reduced rate of muscle resynthesis of glycogen during postexercise recovery (39). Nolte et al. (37) reported a decrease in the activity of the enzymes involved in glycogenesis (glucokinase and phosphoglucomutase) in hyperthyroid patients. The deficiency in nonoxidative glucose uptake might be explained by increased glucose oxidation together with glycogenolysis.

0-30

30-60

60-90

90-120

120-150

150-180 min

In man, inhibition of hepatic glucose production after a glucose load is blunted in hyperthyroid patients compared to that in a control group (41), and splanchnic glucose production is increased in the basal state (22, 41). In the present study, the residual hepatic glucose production after glucose ingestion was not taken into account in the calculation of nonoxidative glucose uptake; the values might, therefore, be somewhat underestimated. Glucagon is believed to maintain hepatic glucose pro-

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 23 January 2017. at 00:47 For personal use only. No other uses without permission. . All rights reserved.

RANDIN ET AL.

1170

TABLE 3. Lipid oxidation in normal, hyperthyroid, and treated hyperthyroid subjects Hyperthyroid Basal lipid oxidation (g/3 h) Postload lipid oxidation

Normal

Treated

20.8 ± I0-6

9 ± 0.5

11.6 ± l c

8.9 ± l°'d

4.2 ± 0.4

4.6 ± 0.6

Values are the mean ± SEM. ° P < 0.001 compared to normal. b P < 0.01 as compared to treated. c P < 0.025 compared to normal. d P < 0.005 compared to normal.

JCE & M • 1985 Vol 61 • No 6

insulin release (44). In conclusion, the increase in energy expenditure in hyperthyroid Graves disease is accompanied by an important increase in lipid oxidation in the fasting state and in glucose oxidation after a glucose load. The decrease in glucose tolerance in hyperthyroidism cannot, therefore, be explained by an alteration in glucose oxidation, but, rather, by a defect in nonoxidative glucose uptake in the periphery and/or a decreased suppression of glucose hepatic production after glucose ingestion (41).

Acknowledgments

50-i

The authors are indebted to Mrs. Penseyres and Mrs. E. Temler for their technical assistance, and to Misses M. Donner and M. C. Evraere for their secretarial assistance. We thank Prof. T. Lemarchand for the determinations of thyroid hormones.

References B30 CD

T3

3 $20 o o r . 0.79 p< 0.001

10

nO

n-19

CO

Q. 3

0

2

4

6 8 10 12 Serum total T3 nmol/l

FIG. 3. Relationship between suprabasal glucose oxidation and total serum T3 concentrations in 12 untreated and 7 treated hyperthyroid patients.

duction (42). Contrarily to the results of others (1), we did not find fasting hyperglucagonemia or an abnormal decrease in glucagon levels after a glucose load in hyperthyroid patients. Therefore, we cannot explain the increased hepatic glucose production in hyperthyroidism (22, 41) by hyperglucagonemia, but the elevated level of thyroid hormones may increase hepatic responsiveness to glucagon (9). The plasma insulin response to the oral glucose load, expressed as the insulin area above the fasting level, was significantly elevated in the hyperthyroid patients before therapy. Therefore, inadequate insulin secretion cannot be evoked in hyperthyroid patients to explain glucose intolerance. Recently, it was suggested that impaired hepatic insulin extraction would contribute to hyperinsulinism in hyperthyroidism (43). The significantly higher peak of plasma insulin 30 min after the glucose load may reflect the increased gastric emptying rate in hyperthyroidism (3), as the gastric emptying rate may be important factor regulating postprandial

1. Kabadi UM, Eisenstein AB 1980 Glucose intolerance in hyperthyroidism: Role of glucagon. J Clin Endocrinol Metab 50:392 2. Nieschlag E, Wordehoff B, Gilfrich HJ, Michaelis J, Overzier C 1971 Kohlenhydratestoffwechsel bei Hyperthyreose: Einfluss einer thyreostatischen Therapie auf orale und intravenose Glucosetoleranz und Insulinaktivitat. Acta Endocrinol (Copenh) 67:277 3. Holdsworth CD, Besser GM 1968 Influence of gastric emptying rate and of insulin response on oral glucose tolerance in thyroid disease. Lancet 2:700 4. Orsetti A, Collard F, Jaffiol C 1974 Abnormalities of carbohydrate metabolism in experimental and clinical hyperthyroidism: studies on plasma insulin and on the A and B chains of insulin. Acta Diabetol Lat 11:486 5. Andreani D, Menzinger G, Fallucca F, Aliberti G, Tamburrano G, Cassano C 1970 Insulin levels in thyrotoxicosis and primary myxoedema: response to intravenous glucose and glucagon. Diabetologia 6:1 6. Andersen OO, Friis TH, Ottesen B 1977 Glucose tolerance and insulin secretion in hyperthyroidism. Acta Endocrinol (Copenh) 84:576 7. Wajchenberg BL, Cesar FP, Leme CE, Didio R, Kieffer J 1978 Plasma insulin disappearance curve after intravenous insulin injection in human hyperthyroidism. Clin Endocrinol (Oxf) 8:233 8. Hales CN, Hyams DE 1964 Plasma concentrations of glucose nonesterified fatty acid, and insulin during oral glucose-tolerance tests in thyrotoxicosis. Lancet 2:69 9. Madsen SN, Sonne O 1976 Increase of glucagon receptors in hyperthyroidism. Nature 262:793 10. Saunders J, Hall SE, Sonksen PH 1980 Glucose and free fatty acid turnover in thyrotoxicosis and hypothyroidism before and after treatment. Clin Endocrinol (Oxf) 13:33 11. Hagenfeldt L, Wennlund A, Felig P, Wahren J 1981 Turnover and splanchnic metabolism of free fatty acids in hyperthyroid patients. J Clin Invest 67:1672 12. Gordon ES, Goldberg M 1964 Carbon-14 studies of energy metabolism on various thyroid states. Metabolism 13:591 13. Butterfield WJ, Wichelow MJ 1964 Are thyroid hormones diabetogenic? A study of peripheral glucose metabolism during glucose infusion in normal subjects and hyperthyroid patients before and after treatment. Metabolism 13:620 14. Bressler R, Wittels B 1966 The effect of thyroxine on lipid and animal in vitro carbohydrate metabolism in the heart. J Clin Invest 45:1326 15. Renauld A, Foglia VG 1972 Studies on the effects of hyperthyroidism in vitro on glucose oxidation in adipose tissue. Acta Physiol Lat Am 22:45 16. Jequier E 1981 Long-term measurement of energy expenditure in

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 23 January 2017. at 00:47 For personal use only. No other uses without permission. . All rights reserved.

GLUCOSE AND LIPID METABOLISM IN GRAVES' DISEASE

17.

18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29.

man: direct of indirect calorimetry? In: Bjorntorp P, Cairella M, Howard AN (eds), Recent Advances in Obesity Research III. Libbey, London, p 130 Felber JP, Meyer HU, Curchod B, Iselin HU, Rousselle J, Maeder E, Pahud P, Jequier E 1981 Glucose storage and oxidation in different degrees of human obesity measured by continuous indirect calorimetry. Diabetologia 20:39 Steele R, Wall JS, De Bodo RL, Altszuler N 1956 Measurement of size and turnover rate of body glucose pool by the isotope dilution method. Am J Physiol 187:15 Perley JJ, Kipnis DM 1967 Plasma insulin response to oral and intravenous glucose: studies in normal and diabetic subjects. J Clin Invest 46:1954 Radziuk J, McDonald TJ, Rubenstein D, Dupre J 1978 Initial splanchnic extraction of ingested glucose in normal man. Metabolism 27:657 Acheson KJ, Schutz Y, Bessard T, Ravussin E, Jequier E 1984 Nutritional influences on lipogenesis and thermogenesis after a carbohydrate meal. Am J Physiol 246:62 Sandier MP, Robinson RP, Rabin D, Lacy WW, Abumrad NN 1983 The effect of thyroid hormones on gluconeogenesis and forearm metabolism in man. J Clin Endocrinol Metab 56:479 Wahren J, Wennlund A, Nilsson LH, Felig P 1981 Influence of hyperthyroidism on splanchnic exchange of glucose and gluconeogenic precursors. J Clin Invest 67:1056 Frayn KN 1983 Calculation of substrate oxidation rates in vivo from gaseous exchange. J Appl Physiol 55:628 Burzstein S, Saphar P, Glaser P, Faitehman U, de Myttenaere S, Nedey R 1977 Determination of energy metabolism from respiratory function alone. J Appl Physiol 42:117 Herbert V, Lau KS, Gottlieb CW, Bleicher SL 1965 Coated charcoal immunoassay of insulin. J Clin Endocrinol Metab 25:1375 Dole V, Meinertz H 1960 Microdetermination of long-chain fatty acids in plasma and tissues. J Biol Chem 235:2595 Heindel JJ, Cushman SW, Jeanrenaud B 1974 Cell-associated fatty acid levels and energy requiring processes in mouse adipocytes. Am J Physiol 226:16 Aguilar-Parada E, Eisentraut AM, Unger RH 1969 Pancreatic

32. 33. 34. 35.

36. 37.

38. 39. 40. 41.

42.

43.

1171

human adipose tissue in hyperthyroidism. J Clin Endocrinol Metab 48:415 Ostman J, Arner P, Bolinder J, Engfeldt P, Wennlund A 1981 Regulation of lipolysis in hyperthyroidism. Int J Obesity 5:665 Wennlund A, Arner P, Ostman J 1981 Changes in the effects of insulin on human adipose tissue metabolism in hyperthyroidism. J Clin Endocrinol Metab 53:631 Sestoft L 1980 Metabolic aspects of the calorigenic effect of thyroid hormone in mammals. Clin Endocrinol (Oxf) 13:489 Thiebaud D, Jacot E, DeFronzo RA, Maeder E, Jequier E, Felber JP 1982 The effect of graded doses of insulin on total glucose uptake, glucose oxidation, and glucose storage in man. Diabetes 31:957 Rousselle J, Biickert A, Pahud P, Jequier E, Felber JP 1982 Relationship between glucose oxidation and glucose tolerance in man. Metabolism 31:866 Nolte J, Pette D, Bachmaier B, Kiefhaber P, Schneider H, Scriba PC 1972 Enzyme response to thyrotoxicosis and hypothyroidism in human liver and muscle: comparative aspects. Eur J Clin Invest 2:141 Levy LJ, Adesmann JJ, Spergel G 1971 Studies on the carbohydrate and lipid metabolism in thyroid disease: effects of glucagon. J Clin Endocrinol Metab 30:372 Brzezinska Z, Kaciuba-Uscilko H 1979 Low muscle and liver glycogen contents in dogs treated with thyroid hormones. Horm Metab Res 11:675 Paul AK, Dhar A 1980 Effect of thyroxine on the hepatic glycogen and glucose-6-phosphatase activity of developing rats. Horm Metab Res 12:261 Baker B, Dimitriadis G, Marsh M, Rizza R, Bergman R, Gerich J 1983 Effects of experimental hyperthyroidism on insulin action in man: induction of insulin resistance due to selective impairment of suppression of glucose production at a postreceptor site. Clin Res 3L499A (Abstract) Liljenquist JE, Mueller GL, Cherrington AD, Keller U, Chiasson JL, Perry JM, Lacy WW, Rabinowitz D 1977 Evidence for an important role of glucagon in the regulation of hepatic glucose production in normal man. J Clin Invest 59:369 Osei K, Cataland S, O'Dorisio TM, Adam DR, Falko JM 1983

glucagon secretion in normal and diabetic subjects. Am J Med Sci

Impaired hepatic insulin extraction contributes to hyperinsulinism

257:415 30. Hawk P 1947 The Kjeldahl method. In: Practical Physiology Chemistry, ed 12. Blakiston, Toronto, p 814 31. Arner P, Wennlund A, Ostman J 1979 Regulation of lipolysis by

in hyperthyroid patients. Clin Res 31:719A (Abstract) 44. Schusdziarra V, Henrichs I, Rouiller D, Unger RH, Pfeiffer EF 1980 Gastric emptying: an important regulatory factor in postprandial insulin release. Biochem Soc Trans 8:55

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 23 January 2017. at 00:47 For personal use only. No other uses without permission. . All rights reserved.