Resting Energy Expenditure and Cold-induced Thermogenesis in Patients With Overt Hyperthyroidism

Abstract

Context: Thyroid hormone (TH) is crucial for the adaptation to cold.

Objective: To evaluate the effect of hyperthyroidism on resting energy expenditure (REE), cold-induced thermogenesis (CIT) and changes in body composition and weight.

Methods: This was a prospective cohort study at the endocrine outpatient clinic of a tertiary referral center. Eighteen patients with overt hyperthyroidism were included. We measured REE during hyperthyroidism, after restoring euthyroid TH levels and after 3 months of normal thyroid function. In 14 of the 18 patients, energy expenditure (EE) was measured before and after a mild cold exposure of 2 hours and CIT was the difference between EEcold and EEwarm. Skin temperatures at 8 positions were recorded during the study visits. Body composition was assessed by dual X-ray absorption.

Results: Free thyroxine (fT4) and free triiodothyronine (fT3) decreased significantly over time (fT4, P = .0003; fT3, P = .0001). REE corrected for lean body mass (LBM) decreased from 42 ± 6.7 kcal/24 hour/kg LBM in the hyperthyroid to 33 ± 4.4 kcal/24 hour/kg LBM (-21%, P < .0001 vs hyperthyroid) in the euthyroid state and 3 months later to 33 ± 5.2 kcal/24 hour/kg LBM (-21%, P = .0022 vs hyperthyroid, overall P < .0001). fT4 (P = .0001) and fT3 (P < 0.0001) were predictors of REE. CIT did not change from the hyperthyroid to the euthyroid state (P = .96). Hyperthyroidism led to increased skin temperature at warm ambient conditions but did not alter core body temperature, nor skin temperature after cold exposure. Weight regain and body composition were not influenced by REE and CIT during the hyperthyroid state.

Conclusion: CIT is not increased in patients with overt hyperthyroidism.

Trial registration: ClinicalTrials.gov NCT03379181.

Keywords: brown adipose tissue; cold induced thermogenesis; energy expenditure; hyperthyroidism; thyroid hormone.

© The Author(s) 2021. Published by Oxford University Press on behalf of the Endocrine Society.

Figures

Figure 1.

Thyroid hormone levels during hyperthyroid state, after restoration of euthyroidism and 3 months after restoration of euthyroidism. Effect of TH status on free T4 P = .0006, P = 0.0014 for pair-wise comparison hyperthyroid vs euthyroid and P = .020 for hyperthyroid vs 3 months euthyroid. Effect of TH status on free T3 P = .0002. Pairwise comparison hyperthyroid vs euthyroid P = .0001 and hyperthyroid vs 3 months euthyroid P = .0014 (A). TSH was suppressed during hyperthyroidism and normalized subsequently, P < .0001 (mixed-effects model) (B). REE (corrected to lean body mass) was elevated in hyperthyroidism and decreased significantly during course of the study, P < .0001 (mixed-effects model) (C).

Figure 2.

(A) Tympanic temperature as a surrogate of core body temperature did not change from the hyperthyroid to the euthyroid state. (B) Skin surface temperatures measured at 8 specified locations were higher in the hyperthyroid than in the euthyroid state, P < .0001 for effect of TH status (mixed-effects model).

Figure 3.

In a subset of participants we measured EE during warm conditions (EEwarm) (A) and after a mild cold stimulus of 2 hours’ duration (EEcold) (B). EE values are corrected to lean-body mass. The increase in EE in response to cold, cold-induced thermogenesis (CIT), did not change over the course of the study (C) (P = .95, mixed-effects model). Further analysis revealed that CIT was inversely correlated to EEwarm especially during the hyperthyroid state (hyperthyroid, R2 = 0.63, P = 0.0007; euthyroid, R2=0.39, P = .031; 3 months’ euthyroid, R2 = 0.35, P = .22) (D).

Figure 4.

Tympanic (A) and skin temperatures (B-I) during the hyperthyroid state vs the euthyroid state at warm ambient temperature (red) and after mild cold exposure (blue). Cold exposure had a significant impact on skin temperature at the umbilicus (where the cooling device was placed) as well on the skin temperature of the lower arm, finger, lower leg, and foot. TH status did not significantly impact the response to cold (P for interaction > .05). **P < .01; ***P < .001; ns, P ≥ .05.

Figure 5.

Subjective cold sensation on a visual analog scale (VAS) during the course of the measurement of CIT. Red bar: warm phase; blue bar: cooling phase. 7 = hot, 6 = warm, 5 = slightly warm, 4 = neutral, 3 = slightly cool, 2 = cold, 1 = very cold. No significant difference due to TH status was observed.

Figure 6.

Body weight and body composition during the course of the study. (A) Body mass index (BMI) in kg/m2. (B) Total body weight in kg. (C) Fat mass in kg. (D) Lean mass in kg. ***P < .001 (mixed-effects model). ns, P ≥ .05.