Diurnal metabolic profiles in hyperthyroidism

Insulin resistance in hyperthyroidism: the role of IL6 and TNF alpha

OBJECTIVE

Although insulin resistance is a common finding in hyperthyroidism, the implicated mechanisms are obscure. The aim of this study was to investigate whether interleukin 6 (IL6) and tumour necrosis factor alpha (TNFalpha) are related to the development of insulin resistance in hyperthyroidism of nonautoimmune origin.

DESIGN AND METHODS

A meal was given to ten hyperthyroid (HR) and ten euthyroid (EU) women. Plasma samples were taken for 360 min from the radial artery for measurements of glucose, insulin, and nonesterified fatty acids (NEFA). IL6 and TNFalpha were measured preprandially from the superficial epigastric vein and from the radial artery.

RESULTS

i) In HR versus EU: (a) arterial glucose was similar (AUC(0-360) 2087+/-57 vs 2010+/-43 mM x min), but insulin was increased (AUC(0-360) 17 267+/-2447 vs 10 331+/-666 microU/ml x min, P=0.01), (b) homeostasis model assessment (HOMA) was increased (2.3+/-0.4 vs 1+/-0.1 kg/m(2), P=0.007), (c) arterial NEFA were increased (AUC(0-360) 136+/-18 vs 89+/-7 mmol/lxmin, P=0.03), (d) arterial IL6 (2+/-0.3 vs 0.9+/-0.1 pg/ml, P=0.0009) and TNFalpha (4.2+/-0.8 vs 1.5+/-0.2 pg/ml, P=0.003) were increased, and (e) IL6 production from the subcutaneous adipose tissue (AT) was increased (18+/-6 vs 5+/-1 pg/min per 100 ml tissue, P=0.04). ii) (a) Subcutaneous venous IL6 was positively associated with HOMA (beta-coefficient=1.7+/-0.7, P=0.049) and (b) although TNFalpha was not produced by the subcutaneous AT, arterial TNFalpha was positively associated with NEFA (AUC(0-360); beta-coefficient=0.045+/-0.01, P=0.005).

CONCLUSIONS

In hyperthyroidism: i) glucose and lipid metabolism are resistant to insulin, ii) subcutaneous AT releases IL6, which could then act as an endocrine mediator of insulin resistance, iii) although there is no net secretion of TNFalpha by the subcutaneous AT, increased systemic TNFalpha levels may be related to the development of insulin resistance in lipolysis.

Forearm muscle metabolism in primary hypothyroidism

Thyrotoxicosis is associated with increased skeletal muscle glucose metabolism and lipid oxidation in the fasted state. Conversely, in hypothyroidism, glucose metabolism is reduced. The present study has assessed in patients the effects of chronic thyroid hormone deficiency, on insulin-stimulated skeletal muscle metabolism, using a combination of the forearm and hyperinsulinaemic euglycaemic clamp techniques. Seven female patients presenting with primary auto-immune hypothyroidism were studied at presentation and 4-9 months after being rendered biochemically euthyroid. No significant differences were demonstrated between the hypothyroid and euthyroid states in forearm muscle blood flow, or in glucose or lipid metabolism.

Thyroid hormone action on lipid metabolism in humans: a role for endogenous insulin

The effects of moderate hyperthyroidism on lipid metabolism were investigated in six healthy subjects before and after thyroxine treatment (300 micrograms/d). T4-treatment increased basal metabolic rate (+8%) and glucose oxidation (+87%), without affecting lipid oxidation, plasma free fatty acids, glycerol, and beta-hydroxybutyrate. During euthyroidism, a hypoinsulinaemic-euglycaemic 150-minute clamp protocol increased energy expenditure (+3%), lipid oxidation (+42%), plasma free fatty acids (+254%), glycerol (+232%), and beta-hydroxybutyrate (+343%), but decreased glucose oxidation (-20%). Similar effects were observed after T4-treatment, but hyperthyroidism induced disproportionate increases in energy expenditure (+7%), plasma glycerol (+310%), and ketone body levels (+436%). We conclude that moderate hyperthyroidism enhances hypoinsulinemia-induced increases in lipolysis, free fatty acid recycling, and ketogenesis without affecting lipid oxidation. Thus basal insulin may camouflage some of thyroid hormone action on lipid metabolism.

Decreased insulin receptor binding in hyperthyroidism

The binding of 125I-insulin to insulin receptors on circulating mononuclear leukocytes was studied in ten patients with hyperthyroidism and 20 euthyroid normal volunteers. The hyperthyroid patients demonstrated significantly elevated glucose levels following an oral glucose load, despite normal insulin secretion. The infusion of insulin resulted in a delayed hypoglycaemic effect in the hyperthyroid patients; however, the inhibition of the endogenous insulin secretion as indicated by suppression of C-peptide levels was not different from euthyroid control subjects. Insulin binding to monocytes was significantly decreased in the hyperthyroid patients. Scatchard analysis of binding data indicates that a decrease of receptor number rather than receptor affinity seems to be the cause of the lowered insulin binding in hyperthyroid patients with diffuse toxic goitre. The findings of decreased insulin receptor number, mild degree of glucose intolerance despite normal insulin secretion and the delayed hypoglycaemic effect following insulin infusion suggest that peripheral insulin resistance could be involved in the highly complex pathophysiology of glucose intolerance in hyperthyroidism.

Evidence that thyroid hormones regulate gluconeogenesis from glycerol in man

We have previously reported that glucose production assessed using radioisotopic methods is 50% increased in hyperthyroidism but 30% decreased in hypothyroidism. These studies, however, do not distinguish between glycogenolysis and gluconeogenesis. In fasting man more than 80% of circulating glycerol is cleared by the liver and enters the gluconeogenic pathway. We have therefore measured glycerol clearance following bolus intravenous glycerol administration as an indirect assessment of gluconeogenic capacity. Hyperthyroid and hypothyroid subjects were compared with separate matched controls after an overnight fast. In hyperthyroid subjects blood glucose and blood glycerol were increased but lactate, pyruvate, and alanine concentrations were normal. Glycerol clearance was increased in hyperthyroidism and followed a double exponential decay with a shortened second component half-time. Endogenous glycerol production was increased three-fold. In hypothyroidism fasting circulating levels of glucose, lactate, pyruvate, alanine, and glycerol were normal but glycerol clearance was diminished. Both first and second component half-times were prolonged in hypothyroidism and endogenous glycerol production was decreased by 50%. Thus in hyperthyroidism glycerol clearance is greatly enhanced whilst in hypothyroidism glycerol clearance is diminished. The magnitude of the changes suggests that alterations in gluconeogenesis are probably the major factors concerned in the reported increase and decrease in glucose production in hyperthyroidism and hypothyroidism respectively.

Insulin sensitivity in hyperthyroidism: measurement by the glucose clamp technique

Sensitivity to porcine insulin has been compared in overnight fasted hyperthyroid and control subjects using a euglycaemic clamp technique. Basal values for blood glucose, lactate, pyruvate, alanine, serum insulin and C-peptide were similar in the two groups, whilst blood glycerol (hyperthyroid 0.11 +/- 0.02 (mean +/- S.E.) vs. control 0.06 +/- 0.01 mmol/l, P less than 0.01) and blood 3-hydroxybutyrate (0.28 [0.03-0.79, range ]vs 0.09 [0.01-0.29 ]mmol/l, P less than 0.05) were increased in hyperthyroidism. During the 2 hour insulin infusion (0.05 U/kg/h), serum insulin plateaued at the same level (44 +/- 4 vs 44 +/- 1 mU/l) and insulin metabolic clearance rates were similar (1.21 +/- 0.10 vs 1.25 +/- 0.03 l/min). Serum C-peptide levels also decreased by similar amounts (40 +/- 8 vs 47 +/- 6%). The amount of glucose infused to maintain euglycaemia was identical during the second hour of insulin infusion (290 +/- 50 vs 330 +/- 30 mg/kg) as were the increments in lactate and pyruvate concentrations. Blood glycerol values decreased in both groups although values in hyperthyroid patients remained significantly higher than in controls. 3-Hydroxybutyrate concentrations fell to similar values in the two groups. These findings suggest that insulin-stimulated glucose metabolism and inhibition of ketogenesis are normal in hyperthyroidism.