Brown Adipose Tissue Response to Cold Stimulation is Reduced in Girls With Autoimmune Hypothyroidism

James M. Law; David E. Morris; Valerie Astle; Ellie Finn; José Joaquín Muros; Lindsay J. Robinson; Tabitha Randell; Louise Denvir; Michael E. Symonds; Helen Budge


J Endo Soc. 2019;3(12):2411-2426. 

In This Article

Abstract and Introduction


Objective: The interaction between thyroid status and brown adipose tissue (BAT) activation is complex. We assessed the effect of autoimmune hypothyroidism (ATD) in female children on BAT activation, measured using infrared thermography.

Design: Twenty-six female participants (14 with ATD and 12 healthy controls) between 5 and 17 years of age attended a single study session. Thermal images were taken of the supraclavicular region before, and after, the introduction of a cool stimulus.

Results: Participants with ATD had lower resting (hypothyroid, 34.9 ± 0.7°C; control, 35.4 ± 0.5°C; P = 0.03) and stimulated (hypothyroid, 35.0 ± 0.6°C; control, 35.5 ± 0.5°C; P = 0.04) supraclavicular temperatures compared with controls, but there was no difference between groups in the temperature increase with stimulation. BAT activation, calculated as the relative temperature change comparing the supraclavicular temperature to a sternal reference region, was reduced in participants with ATD (hypothyroid, 0.1 ± 0.1°C; control, 0.2 ± 0.2°C; P = 0.04). Children with ATD were frequently biochemically euthyroid due to replacement therapy, but, despite this, increased relative supraclavicular temperature was closely associated with increased TSH (r = 0.7, P = 0.01) concentrations.

Conclusions: Girls with ATD had an attenuated thermogenic response to cold stimulation compared with healthy controls, but, contrary to expectation, those with suboptimal biochemical control (with higher TSH) showed increased BAT activation. This suggests that the underlying disease process may have a negative effect on BAT response, but high levels of TSH can mitigate, and even stimulate, BAT activity. In summary, thyroid status is a complex determinant of BAT activity in girls with ATD.


Since its rediscovery in adult humans in 2009,[1–4] interest in brown adipose tissue (BAT) has increased steadily. BAT has an integral role in adaptive thermogenesis due to its capacity to rapidly generate significant quantities of heat from fatty acids and glucose, facilitated by uncoupling protein 1 (UCP1) in the mitochondrial membrane, allowing the disassociation of ATP production from mitochondrial respiration.[5–9] As heat is eventually lost from the body, this represents a net loss of energy and has the potential to contribute to body weight management. Despite promising results in rodents demonstrating weight loss, improved metabolic profiles, and greater insulin sensitivity following BAT stimulation,[10–15] translation to human studies has been slow.[16] Human studies have been limited by high ionizing radiation associated with positron emission tomography (PET)–CT, considered to be the gold standard for imaging BAT. Thermal imaging has been established as a valid alternative.[17–20]

As a major regulator of energy expenditure, the thyroid axis can modulate the heat-generating capability of BAT.[21,22] Thyroid hormones cross the blood–brain barrier[23] and act on the hypothalamus to increase sympathetic nervous system (SNS) activation.[24] Under SNS control, brown adipocytes express elevated iodothyronine deiodinase 2, which converts intracytoplasmic free T4, taken up from the systemic circulation, into its metabolically active form, T3,[25,26] causing a localized intracellular hyperthyroid environment. Translocation and binding of T3 to its intranuclear thyroid hormone receptor-β stimulates UCP1 transcription and translation,[27] leading to heat generation. Reduced thyroid hormone concentrations may, therefore, affect BAT activity directly or by reducing SNS activation centrally. Despite this defined mechanistic pathway, in vivo studies in humans are limited and conflicting. In healthy volunteers, BAT activation is not associated with serum thyroid hormone concentrations[28,29] but is with higher TSH concentrations.[29] BAT activity is increased in patients with hyperthyroidism and returns to normal after treatment.[30] In patients with hypothyroidism, BAT remains present, and indeed may become markedly hypertrophic in the absence of replacement therapy.[31] However, it is not clear whether treatment with T4 increases[32] or decreases[33] BAT activity, although a recent small study in healthy adults demonstrated a negative correlation between plasma free T4 and BAT volume.[34]

Thyroid hormones are essential for brain and physical development in early life[35,36] and continue to be critical through childhood when BAT activity is also increased.[37,38] Adiposity patterns developed in this period can predict later obesity and metabolic health.[39–43] The influence of thyroid hormones on BAT activity, however, has not been examined in otherwise healthy children. Despite many patients achieving biochemical euthyroidaemia, physiological diurnal variation in TSH and thyroid hormone profiles is not achieved with hormone replacement therapy.[44,45]

Pediatric patients with hypothyroidism, for the most part, either have congenital hypothyroidism or autoimmune hypothyroidism, with the latter being more common in girls than boys. BAT is known to vary between sexes,[46,47] and so, to reduce heterogeneity, we compared girls with a diagnosis of autoimmune hypothyroidism (ATD) who, we hypothesized, would show reduced BAT activation in response to a cool stimulus compared with healthy age and sex-matched controls. We further hypothesized that those in the hypothyroid group who were relatively biochemically hypothyroid would have lower BAT activation than those who were relatively biochemically hyperthyroid.