Maturitas 81 (2015) 266–275
Contents lists available at ScienceDirect
Maturitas journal homepage: www.elsevier.com/locate/maturitas
Review
Thyroid hormone: Influences on mood and cognition in adults Meg Ritchie a , Bu B. Yeap a,b,∗ a b
Department of Endocrinology and Diabetes, Fiona Stanley and Fremantle Hospitals, Perth, Western Australia, Australia School of Medicine and Pharmacology, University of Western Australia, Perth, Western Australia, Australia
a r t i c l e
i n f o
Article history: Received 16 March 2015 Accepted 22 March 2015 Keywords: Hypothyroidism Hyperthyroidism Subclinical Free thyroxine Mood Cognition
a b s t r a c t The association of thyroid dysfunction with alterations in mood and cognition has been recognised since some of the earliest descriptions of thyroid disease. Over the years, researchers have aimed to further define these effects throughout the spectrum of thyroid disorders, to better understand the underlying condition and refine indications for treatment. More recently, attention has turned towards examining the impact of differences in thyroid hormones within the normal reference range, particularly in older adults, providing new insights into the association of thyroid hormone with cognitive decline. This review summarises the evidence assessing the influence of thyroid hormone on mood and cognition in overt and subclinical hypothyroidism, within the reference range, and in subclinical and overt hyperthyroidism. Treatment of overt thyroid dysfunction largely resolves associated disturbances in mood and cognitive dysfunction, however in the setting of overt hypothyroidism subtle detrimental effects on cognition may not be fully reversed. Subclinical hyperthyroidism and higher free thyroxine (FT4) within the normal range have been associated with poorer cognitive outcomes. Future research including randomised controlled trials are required to confirm causality and guide the assessment of benefits vs risks of intervention in the increasing population of older adults with subclinical thyroid disease. © 2015 Elsevier Ireland Ltd. All rights reserved.
Contents 1. 2.
3. 4. 5. 6.
7.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Overt hypothyroidism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. Overt hypothyroidism and mood disorders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Overt hypothyroidism and cognitive dysfunction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. Subclinical hypothyroidism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Differences in thyroid hormones within the normal range . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Subclinical hyperthyroidism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Overt hyperthyroidism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.1. Overt hyperthyroidism and mood disorders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.2. Overt hyperthyroidism and cognition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.3. Mechanisms by which hyperthyroidism modulates brain function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Competing interest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Funding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Provenance and peer review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
267 267 267 267 269 270 271 273 273 273 273 273 274 274 274 274 274 274
∗ Corresponding author at: Harry Perkins Research Institute, Fiona Stanley Hospital, Robin Warren Drive, 102-118 Murdoch Drive, Murdoch 6150, Western Australia, Australia. Tel.: +61 8 6151 1149; fax: +61 8 6151 1199. E-mail address: [email protected] (B.B. Yeap). http://dx.doi.org/10.1016/j.maturitas.2015.03.016 0378-5122/© 2015 Elsevier Ireland Ltd. All rights reserved.
M. Ritchie, B.B. Yeap / Maturitas 81 (2015) 266–275
1. Introduction The thyroid, like other endocrine glands of the human body, has wide-ranging effects on multiple organ systems, including the brain and nervous system. Some of the earliest descriptions of thyroid disease noted a link with neuropsychiatric disturbances including mood disorders and cognitive dysfunction. A report by the Committee of the Clinical Society of London in 1888 on myxoedema observed: “Delusions and hallucinations occur in nearly half the cases, mainly where the disease is advanced. Insanity as a complication is noted in about the same proportion as delusions or hallucinations. It takes the form of acute or chronic manias, dementia, or melancholia, with a marked predominance of suspicion and self-accusation” [1]. Similarly, a previous description of thyrotoxicosis in the British Medical Journal noted a significant psychiatric component to this disorder: “One of the commonest areas of diagnostic confusion lies in the distinction between thyrotoxicosis and those anxiety states in which nervous or cardiovascular symptoms predominate. The thyrotoxic patient, often warm and losing weight despite a good appetite, is usually fidgety and hyperkinetic, and her tachycardia is accompanied by a hyperdynamic circulation. . .patients with an anxiety state are seldom hyperkinetic and despite their tachycardia usually have a normodynamic circulation” [2]. The thyroid gland produces the thyroid hormones thyroxine (T4) and liothyronine (T3) under the stimulation of pituitary secretion of thyrotrophin (TSH), and the diagnosis of hyperthyroidism is confirmed by demonstrating elevated free thyroxine (FT4) in conjunction with suppression of TSH via negative feedback to the pituitary [3]. Conversely, hypothyroidism is characterised by reduced FT4 in the setting of elevated TSH. Subclinical thyroid dysfunction is present when thyroid hormone concentrations are in the normal range yet TSH is below normal (subclinical hyperthyroidism) or above normal (subclinical hypothyroidism) [3]. In this review, we describe the relationship between mood and cognitive dysfunction in overt hypothyroidism and overt hyperthyroidism, as well as emerging evidence of corresponding associations throughout the spectrum of subclinical thyroid disorders. The pituitary-thyroid axis evolves with age and this has implications for the increasing numbers of older adults where cognitive impairment is a particular concern.
2. Overt hypothyroidism 2.1. Overt hypothyroidism and mood disorders Patients presenting with overt hypothyroidism may exhibit significant psychiatric and cognitive disturbance, classically slowness of thought and increased depressive symptoms. The original descriptions of myxoedema madness emphasise psychotic features such as delusions and hallucinations, particularly of the persecutory or suspicious type [1], which can accompany a presentation of florid hypothyroidism. However, the range of neuropsychiatric disturbances encountered in hypothyroid patients is likely to encompass a much broader spectrum. Epidemiological studies have elucidated the association of mood disorders, particularly depression, with hypothyroidism. A selection of recent reports is summarised in Table 1A. Guimaraes et al. [6] showed that high TSH levels in women were associated with an increased risk of developing depression even after adjusting for age, race, smoking and body mass index. Gulseren et al. [5] found that anxiety and depressive symptoms were more severe in patients with hypothyroidism, and that these symptoms improved with thyroxine treatment. In a small case-control study, Mowla et al. [8] compared the characteristics of depression in patients diagnosed with major depressive disorder with and without hypothyroidism.
267
While severity of depression was not significantly worse in those with hypothyroidism, more anxiety symptoms and agitation were present. Conversely, Wu et al. [10] found a higher prevalence and incidence of hypothyroidism in patients with major depressive disorders. There have been several case reports and case series [7,9] describing patients with primary hypothyroidism who have presented with acute mania. In these cases, patients were treated with both psychotropic medication and T4, with gradual improvement in mental state. Heinrich and Grahm [4] presented a case report of a patient presenting with acute psychosis and hypothyroidism who was treated with low-dose T4 and antipsychotic therapy. Within 2–3 weeks of therapy the patient’s psychosis had resolved and the antipsychotics were ceased, with no further recurrence of psychiatric symptoms. Therefore overt hypothyroidism is associated with depression, the presence of hypothyroidism can accentuate symptoms such as anxiety and agitation, and treatment with T4 helps improve the disturbance in mood. The wide range of psychiatric presentations ranging from depression to mania reported in overt hypothyroidism emphasises the importance of screening for thyroid dysfunction in patients presenting with an acute psychiatric disturbance [18]. 2.2. Overt hypothyroidism and cognitive dysfunction Cognitive impairment has been regarded as a possible consequence of overt hypothyroidism, although psychiatric disturbances can impact negatively on assessments of cognitive performance [19]. Studies evaluating hypothyroidism and its effects on cognitive function have shown contrasting results, with recent studies summarised in Table 1B. Parsaik et al. [13] performed a cross-sectional study evaluating the association of overt and subclinical hypothyroidism with mild cognitive impairment over a large population based cohort of older adults. They found no significant association between hypothyroidism (overt or subclinical) with mild cognitive impairment after adjusting for possible confounding factors. On the other hand, longitudinal studies evaluating cognition in patients when hypothyroid and after the restoration of euthyroid state show subtle but definable changes in cognition. Schraml et al. [12] assessed neuropsychological outcomes and thyroid function in a group of patients post thyroidectomy, whilst overtly hypothyroid and following adequate thyroid hormone replacement, as compared with controls. Hypothyroid patients performed worse than controls in the domain of working memory, and this improved following thyroid hormone replacement. Smith et al. [14] assessed clinical status, cognitive performance and driving ability in 32 patients undergoing thyroid hormone withdrawal for radioiodine scanning while hypothyroid and following restoration of euthyroid status. They found that transient profound hypothyroidism was characterised by reversible depression, decreased fine motor performance, slowed reaction times and decreased processing speed. These studies support a link between overt hypothyroidism and a degree of reduced cognitive function, although this may be limited to particular domains. The issue of reversibility of neurocognitive symptoms in treated hypothyroidism has been examined in several studies, summarised in Table 1C. In patients with diagnosed hypothyroidism, residual impairment in cognitive function despite T4 replacement has been reported [15–17]. It is possible that the effects of hypothyroidism on the brain may not be fully reversible, or that exogenous thyroxine replacement falls short of the function of the native pituitary-thyroid axis. It is also plausible that there may be other factors unique to this patient group or their treatment that predisposes them to higher rates of comorbid cognitive disorders. Saravanan e al. [15] surveyed a large group of patients in the UK who had been on T4 for at least 4 months to evaluate their psychological well-being as compared to controls. They found that
268
M. Ritchie, B.B. Yeap / Maturitas 81 (2015) 266–275
Table 1 Overt hypothyroidism. Study author and year
Study
n
Age (years)
Duration of follow up
Results
(A) Mood Heinrich 2003 [4]
Case report
1
73
–
Gulseren 2006 [5]
Longitudinal
33 with overt hypothyroidism 20 healthy controls
43.6 in overt hypothyroidism 40.5 in controls
10.2 months in overt hypothyroidism
Guimaraes 2009 [6]
Cross-sectional
1298 women
53.6 Range 35–91
–
Khemka 2011 [7]
Case series
3
56, 68 and 77
–
Mowla 2011 [8]
Cross-sectional
75 adults with major depressive disorder
34.8
–
Lin 2013 [9]
Case report
1
41
2 years
Wu 2013 [10]
Longitudinal
761,834 from general population database, 4593 with depressive disorder
≥18
5
Presentation with acute mania and hypothyroidism, improved with 2 weeks antipsychotic and continuing T4. Anxiety and depressive symptoms were more severe in patients with overt hypothyroidism. Psychological symptoms improved with treatment. Women with TSH > 4 mIU/L and T4 < 9 pmol/L (n = 20) had increased risk of depressive symptoms on questionnaire. Presentations with acute mania and hypothyroidism, improved with antipsychotic and T4. Adults with TSH > 4 mIU/L, Free T4 < 9 pmol/L (n = 7) had lower scores on depressed mood, guilt, suicidality, insomnia but higher scores on agitation and anxiety. Presentation with acute mania and hypothyroidism, improved with antipsychotic and T4. Greater prevalence and incidence of hypothyroidism (ICD coded) in persons with major depressive disorder.
(B) Cognition Kramer 2009 [11]
Cross-sectional, cohort
1034
76.1
–
Schraml 2011 [12]
Case control, before and after T4
11 patients post thyroidectomy 11 healthy controls
33.0 in cases 32.9 in controls
14 weeks
Parsaik 2014 [13]
Cross-sectional, cohort
1904
81.2 (median) in hypothyroid persons
–
Smith 2015 [14]
Case series, before and after T4 deprivation
32 (20 studied after T4 replacement)
44.8 40.7 in recovery subgroup
4 months
(C) Effects of treatment Saravanan 2002 [15]
Case control Cross sectional
59.7 for cases 59.4 for controls 47.8
–
Wekking 2005 [16]
–
Samuels 2007 [17]
Case control
597 patients 551 controls 141 hypothyroid patients on T4 replacement 34 hypothyroid women on 20 control women
20–44
–
patients on T4 replacement, even with normal TSH, had much higher levels of impairment in psychological well-being compared to age and sex matched controls. Given the high numbers of people on T4 replacement, it was proposed that this could contribute to a substantial burden of psychological morbidity in the community. Samuels et al. [17] performed a cross-sectional comparison of euthyroid and treated hypothyroid patients with normal TSH levels for psychological and cognitive function, and found that the group of T4 treated patients had reduced health status, psychological function, working memory and motor learning compared to euthyroid controls. Additionally, Wekking et al. [16] assessed neurocognitive functioning and well-being of patients
No difference in cognitive function in patients with treated hypothyroidism (n = 149) compared to euthyroid (n = 885). Reduced working memory and more depressive symptoms in thyroidectomised patients when hypothyroid compared to controls, which improved with treatment. Hypothyroid (TSH > 50 mIU/L) inversely associated with word fluency and working memory. No association of previously diagnosed hypothyroidism (n = 313) with mild cognitive impairment. Increased reaction and fine motor times, and increased depressive symptoms when hypothyroid. Improvement in subgroup following T4 treatment. Reduced psychological wellbeing in hypothyroid patients on T4. Impaired scores on memory tasks and auditory attention in hypothyroid patients compared to reference group. Impaired wellbeing, motor learning and working memory in T4 treated women.
with primary hypothyroidism on adequate T4 treatment, and found poor performance in various domains of neurocognitive functioning, especially complex attention tasks and verbal memory tests, with decreased levels of well-being. The authors concluded that neurocognitive function and psychological impairment may not be completely restored in hypothyroid patients despite T4 replacement. By contrast, a cross-sectional study by Kramer et al. [11] found that patients who have been on long-term treatment for hypothyroidism had no significant difference in scores for cognitive function than euthyroid patients over three standardised cognitive function tests. Further research will be required to elucidate whether hypothyroidism leads to permanent neuropsychological
M. Ritchie, B.B. Yeap / Maturitas 81 (2015) 266–275
269
Table 2 Subclinical hypothyroidism. Study author and year
Study
n
Age (years)
Duration of follow up
Results
(A) Mood Roberts 2006 [21]
Cross-sectional
73
–
Jorde 2006 [22]
Case control
Mean 61.0–63.0 between groups
–
Chueire 2007 [23]
Cross-sectional
67
–
Subclinical hypothyroidism was not associated with depressive symptoms, anxiety or cognitive function. No difference in depressive scores between the two groups. Better performance in tests of emotional function. 9.5% of participants with elevated TSH were depressed. 31% of patients with depression had elevated TSH.
Kim 2010 [24]
Cross-sectional
5868 168 with subclinical hypothyroidism 89 with subclinical hypothyroidism (TSH 3.5–10.0 mIU/L) 154 controls 323 252 with TSH > 4.5 mIU/L 71 with primary depressive disorder 495 37 with hypothyroidism (TSH > 4.5 mIU/L)
72.4
–
Hypothyroidism (not specified as overt/subclinical) was not associated with depression.
(B) Cognition Gussekloo 2004 [25]
Longitudinal
85
4 years
Roberts 2006 [21]
Cross-sectional
73
–
Jorde 2006 [22]
Case control Subgroup – randomised placebo controlled trial (T4 vs placebo)
62.4 (subclinical hypothyroidism) 61.0 (controls)
12 months
No association of elevated TSH with depressive symptoms or global cognition. Subclinical hypothyroidism was not associated with depressive symptoms or anxiety or cognitive function. No difference in cognitive function between groups. No difference in cognition or emotional function between T4 or placebo treatment.
Bensenor 2010 [26]
Cross-sectional
>65
–
Kim 2010 [24]
Cross-sectional
72.4
–
Silva 2013 [27]
Cross-sectional
80.7
–
Parsaik 2014 [13]
Cross-sectional, cohort
(C) Effects of treatment Arinzon 2006 [28]
Cross-sectional
Parle 2010 [29]
Double-blind placebo-controlled trial of T4 vs placebo
599 67 with elevated TSH (>4.8 mIU/L) 5868 168 with subclinical hypothyroidism 89 with subclinical hypothyroidism (TSH 3.5–10.0 mIU/L) 154 controls Subgroup – 36 for T4, 34 for placebo 1276 157 subclinical hypothyroidism 495 37 with hypothyroidism (TSH > 4.5 mIU/L) 411 43 subclinical hypothyroidism 284 euthyroid 1904
57 26 with subclinical hypothyroidism 31 with overt hypothyroidism 94 with subclinical hypothyroidism
77.7
Patients reevaluated after 3 months thyroxine
73.5 (T4 group), 74.2 (placebo group)
6 and 12 months
3. Subclinical hypothyroidism Subclinical hypothyroidism is defined biochemically by elevated TSH concentration with normal FT4 concentration [3], and is more common in women and older adults [20]. Cross-sectional studies to date have shown mixed results regarding associations of subclinical hypothyroidism and altered mood, with recent studies summarised in Table 2A. Roberts et al. [21] showed no association between subclinical hypothyroidism and depression or anxiety in older adults. Kim et al. [24] reported a similar negative finding in
No difference in depressive symptoms and cognition between groups.
No association of previously diagnosed hypothyroidism (n = 141) with mild cognitive impairment.
81.7 (median) in hypothyroid persons
impairment, or whether conventional replacement therapies may not prevent or reverse more subtle cognitive dysfunction associated with hypothyroidism.
No association between subclinical hypothyroidism and cognitive dysfunction. Hypothyroidism (not specified as overt/subclinical) was not associated with cognitive impairment.
Improvement in cognitive and functional status was found after treatment for subclinical hypothyroidism and overt hypothyroidism. No improvement in cognitive function with T4.
their evaluation of older patients with hypothyroidism (defined as TSH 4.5–17.2 but not classified as overt or subclinical hypothyroidism). By contrast, Chueire et al. [23] found that older adults with raised TSH were more likely to be depressed. With regards to cognition, the study results have been more consistent, with three recent cross sectional studies [21,24,27] showing no association between subclinical hypothyroidism and cognitive dysfunction, as summarised in Table 2B. A case control study [22] showed no neuropsychological dysfunction in patients with subclinical hypothyroidism compared with controls. Gussekloo et al. [25] performed a prospective observational study of older adults and found that those with higher TSH do not experience increased adverse events and may in fact have a prolonged life span. A longitudinal treatment study by Arinzon et al. [28] found that patients with
270
M. Ritchie, B.B. Yeap / Maturitas 81 (2015) 266–275
Table 3 Euthyroidism. Study author and year
Study type
n
Age (years)
Duration of follow up
Results
(A) Mood Medici 2014 [34]
Longitudinal
1503
70.6
8 years
Almeida 2011 [35]
Longitudinal
3932 men
75.3
5.5 years
Cross sectionally lower TSH (0.3–1.0 mIU/L) was associated with more depressive symptoms than higher TSH (1.6–4.0 mIU/L). Lower TSH was associated with higher incidence of depressive symptoms. The serum concentration of TSH and FT4 did not affect prevalence of depressive symptoms or incidence of depression.
(B) Cognition Gussekloo 2004 [25]
Longitudinal
85
4 years
Elderly individuals with increased TSH did not experience adverse events and may have a prolonged life span.
Roberts 2006 [21]
Cross-sectional
73
–
Hogervorst 2008 [36]
Longitudinal
599 67 with elevated TSH (>4.8 mIU/L) 17 with low TSH (2.1 mIU/L was associated with increased risk for Alzheimer’s dementia in women but not men. Higher total and free thyroxine was associated with an increased risk of dementia and Alzheimer’s disease. In 143 men who had an autopsy study, higher total thyroxine was associated with a higher number of neocortical plaques and neurofibrillary tangles. No association of TSH with mild cognitive impairment or Alzheimer’s disease. High TSH (>3.59 mIU/L) was associated with an increased risk of vascular dementia. High free T4 levels predicted new-onset dementia in older men. Higher free T4 or lower TSH was associated with better performance on specific tests of cognitive function. Higher TSH was associated with lower incidence of dementia.
subclinical or overt hypothyroidism experienced improvement in cognitive and functional status after T4 supplementation. By contrast Parle et al. [29] performed a double-blind placebo controlled trial of T4 in patients with subclinical hypothyroidism and followed them at 6 and 12 months assessing cognitive function. They found no evidence that T4 treatment in elderly patients with subclinical hypothyroidism improved cognitive function. Overall these studies suggest equivocal effects of subclinical hypothyroidism on depression and cognition. Further interventional studies including randomised controlled trials would be required to clarify the effect of hormonal interventions on mood and cognition in adults with subclinical hypothyroidism. 4. Differences in thyroid hormones within the normal range Recently there has been emerging interest in the impact of differences in thyroid hormones within the reference range on mood and cognition, especially in older adults. Changes in thyroid
hormone production and metabolism occur with increasing age, typically reduced production of T4 and T3, but relatively unchanged serum concentrations of total and FT4 due to decreased degradation of T4 [30]. Of note, distributions and reference limits for TSH concentrations shift higher with increasing age, contrasting with the relative stability of FT4 levels [31,32]. Therefore, even in circumstances where the log-linear relationship between FT4 and TSH is preserved, FT4 and TSH may be regarded as distinct risk predictors in older adults [33]. Studies assessing different levels of thyroid hormones within the normal range and their effects on mood have had conflicting results, with recent studies summarised in Table 3A. In a study of 3932 community-dwelling older men, Almeida et al. [35] found that the serum concentrations of TSH and FT4 did not affect the prevalence of depressive symptoms or incidence of depression. On the other hand, in their longitudinal cohort study of 1503 older adults, Medici et al. [34] found that those with low normal TSH had more concurrent depressive symptoms and a substantially increased risk for developing a depressive syndrome in subsequent years.
M. Ritchie, B.B. Yeap / Maturitas 81 (2015) 266–275
271
Table 4 Subclinical hyperthyroidism. Study author and year
Study
n
Age (years)
Duration of follow up
Results
(A) Mood Roberts 2006 [21]
Cross-sectional
5868 127 with subclinical hyperthyroidism
73
–
Kim 2010 [24]
Cross sectional
495 14 with hyperthyroidism (TSH < 0.5 mIU/mL)
72.4
–
No association of subclinical hyperthyroidism with depressive symptoms, anxiety or cognition in models adjusted for covariates. Hyperthyroidism (not specified as overt/subclinical) was not associated with depression.
(B) Cognition Kalmijn 2000 [43]
Longitudinal
1843
68.8
2.1 years
Gussekloo 2004 [25]
Longitudinal
85
4 years
Roberts 2006 [21]
Cross-sectional
73
–
Ceresini 2009 [44]
Cross-sectional
23–102
–
Bensenor 2010 [26]
Cross-sectional
>65
–
Kim 2010 [24]
Cross-sectional
72.4
–
Hyperthyroidism (not specified as overt/subclinical) was associated with cognitive impairment.
Vadiveloo 2011 [45]
Longitudinal
599 17 with low TSH (20% increased risk of dementia and >30% increased risk of Alzheimer’s disease. In a sub-sample who underwent autopsy, it was also shown that those with higher total thyroxine had a greater number of neocortical plaques and neurofibrillary tangles [38], suggesting a pathophysiological mechanism for their clinical findings. In a longitudinal study of 3401 older men Yeap et al. [40] found that higher FT4 levels predicted new-onset dementia with an 11% increase in risk per 1 pmol/L increment in FT4 and a 70% higher hazard ratio for developing dementia in men with FT4 greater than the 25th percentile. In that study TSH was not associated with risk of dementia. In an Italian cohort study of 660 elderly subjects, Forti et al. [39] found baseline TSH was not associated with mild cognitive impairment or Alzheimer’s disease, while higher TSH was associated with an increased risk of vascular dementia. More recently in a large longitudinal study of older adults Cappola et al. [42] found an association of higher TSH concentrations with lower incidence of dementia, which was attenuated by adjustment for covariates. Therefore, although the link between differences in thyroid hormones within the reference range and mood changes has been inconsistent, the data support an association of exposure to thyroid hormone, particularly higher FT4, with
Reduced TSH (
Contents lists available at ScienceDirect
Maturitas journal homepage: www.elsevier.com/locate/maturitas
Review
Thyroid hormone: Influences on mood and cognition in adults Meg Ritchie a , Bu B. Yeap a,b,∗ a b
Department of Endocrinology and Diabetes, Fiona Stanley and Fremantle Hospitals, Perth, Western Australia, Australia School of Medicine and Pharmacology, University of Western Australia, Perth, Western Australia, Australia
a r t i c l e
i n f o
Article history: Received 16 March 2015 Accepted 22 March 2015 Keywords: Hypothyroidism Hyperthyroidism Subclinical Free thyroxine Mood Cognition
a b s t r a c t The association of thyroid dysfunction with alterations in mood and cognition has been recognised since some of the earliest descriptions of thyroid disease. Over the years, researchers have aimed to further define these effects throughout the spectrum of thyroid disorders, to better understand the underlying condition and refine indications for treatment. More recently, attention has turned towards examining the impact of differences in thyroid hormones within the normal reference range, particularly in older adults, providing new insights into the association of thyroid hormone with cognitive decline. This review summarises the evidence assessing the influence of thyroid hormone on mood and cognition in overt and subclinical hypothyroidism, within the reference range, and in subclinical and overt hyperthyroidism. Treatment of overt thyroid dysfunction largely resolves associated disturbances in mood and cognitive dysfunction, however in the setting of overt hypothyroidism subtle detrimental effects on cognition may not be fully reversed. Subclinical hyperthyroidism and higher free thyroxine (FT4) within the normal range have been associated with poorer cognitive outcomes. Future research including randomised controlled trials are required to confirm causality and guide the assessment of benefits vs risks of intervention in the increasing population of older adults with subclinical thyroid disease. © 2015 Elsevier Ireland Ltd. All rights reserved.
Contents 1. 2.
3. 4. 5. 6.
7.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Overt hypothyroidism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. Overt hypothyroidism and mood disorders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Overt hypothyroidism and cognitive dysfunction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. Subclinical hypothyroidism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Differences in thyroid hormones within the normal range . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Subclinical hyperthyroidism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Overt hyperthyroidism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.1. Overt hyperthyroidism and mood disorders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.2. Overt hyperthyroidism and cognition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.3. Mechanisms by which hyperthyroidism modulates brain function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Competing interest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Funding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Provenance and peer review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
267 267 267 267 269 270 271 273 273 273 273 273 274 274 274 274 274 274
∗ Corresponding author at: Harry Perkins Research Institute, Fiona Stanley Hospital, Robin Warren Drive, 102-118 Murdoch Drive, Murdoch 6150, Western Australia, Australia. Tel.: +61 8 6151 1149; fax: +61 8 6151 1199. E-mail address: [email protected] (B.B. Yeap). http://dx.doi.org/10.1016/j.maturitas.2015.03.016 0378-5122/© 2015 Elsevier Ireland Ltd. All rights reserved.
M. Ritchie, B.B. Yeap / Maturitas 81 (2015) 266–275
1. Introduction The thyroid, like other endocrine glands of the human body, has wide-ranging effects on multiple organ systems, including the brain and nervous system. Some of the earliest descriptions of thyroid disease noted a link with neuropsychiatric disturbances including mood disorders and cognitive dysfunction. A report by the Committee of the Clinical Society of London in 1888 on myxoedema observed: “Delusions and hallucinations occur in nearly half the cases, mainly where the disease is advanced. Insanity as a complication is noted in about the same proportion as delusions or hallucinations. It takes the form of acute or chronic manias, dementia, or melancholia, with a marked predominance of suspicion and self-accusation” [1]. Similarly, a previous description of thyrotoxicosis in the British Medical Journal noted a significant psychiatric component to this disorder: “One of the commonest areas of diagnostic confusion lies in the distinction between thyrotoxicosis and those anxiety states in which nervous or cardiovascular symptoms predominate. The thyrotoxic patient, often warm and losing weight despite a good appetite, is usually fidgety and hyperkinetic, and her tachycardia is accompanied by a hyperdynamic circulation. . .patients with an anxiety state are seldom hyperkinetic and despite their tachycardia usually have a normodynamic circulation” [2]. The thyroid gland produces the thyroid hormones thyroxine (T4) and liothyronine (T3) under the stimulation of pituitary secretion of thyrotrophin (TSH), and the diagnosis of hyperthyroidism is confirmed by demonstrating elevated free thyroxine (FT4) in conjunction with suppression of TSH via negative feedback to the pituitary [3]. Conversely, hypothyroidism is characterised by reduced FT4 in the setting of elevated TSH. Subclinical thyroid dysfunction is present when thyroid hormone concentrations are in the normal range yet TSH is below normal (subclinical hyperthyroidism) or above normal (subclinical hypothyroidism) [3]. In this review, we describe the relationship between mood and cognitive dysfunction in overt hypothyroidism and overt hyperthyroidism, as well as emerging evidence of corresponding associations throughout the spectrum of subclinical thyroid disorders. The pituitary-thyroid axis evolves with age and this has implications for the increasing numbers of older adults where cognitive impairment is a particular concern.
2. Overt hypothyroidism 2.1. Overt hypothyroidism and mood disorders Patients presenting with overt hypothyroidism may exhibit significant psychiatric and cognitive disturbance, classically slowness of thought and increased depressive symptoms. The original descriptions of myxoedema madness emphasise psychotic features such as delusions and hallucinations, particularly of the persecutory or suspicious type [1], which can accompany a presentation of florid hypothyroidism. However, the range of neuropsychiatric disturbances encountered in hypothyroid patients is likely to encompass a much broader spectrum. Epidemiological studies have elucidated the association of mood disorders, particularly depression, with hypothyroidism. A selection of recent reports is summarised in Table 1A. Guimaraes et al. [6] showed that high TSH levels in women were associated with an increased risk of developing depression even after adjusting for age, race, smoking and body mass index. Gulseren et al. [5] found that anxiety and depressive symptoms were more severe in patients with hypothyroidism, and that these symptoms improved with thyroxine treatment. In a small case-control study, Mowla et al. [8] compared the characteristics of depression in patients diagnosed with major depressive disorder with and without hypothyroidism.
267
While severity of depression was not significantly worse in those with hypothyroidism, more anxiety symptoms and agitation were present. Conversely, Wu et al. [10] found a higher prevalence and incidence of hypothyroidism in patients with major depressive disorders. There have been several case reports and case series [7,9] describing patients with primary hypothyroidism who have presented with acute mania. In these cases, patients were treated with both psychotropic medication and T4, with gradual improvement in mental state. Heinrich and Grahm [4] presented a case report of a patient presenting with acute psychosis and hypothyroidism who was treated with low-dose T4 and antipsychotic therapy. Within 2–3 weeks of therapy the patient’s psychosis had resolved and the antipsychotics were ceased, with no further recurrence of psychiatric symptoms. Therefore overt hypothyroidism is associated with depression, the presence of hypothyroidism can accentuate symptoms such as anxiety and agitation, and treatment with T4 helps improve the disturbance in mood. The wide range of psychiatric presentations ranging from depression to mania reported in overt hypothyroidism emphasises the importance of screening for thyroid dysfunction in patients presenting with an acute psychiatric disturbance [18]. 2.2. Overt hypothyroidism and cognitive dysfunction Cognitive impairment has been regarded as a possible consequence of overt hypothyroidism, although psychiatric disturbances can impact negatively on assessments of cognitive performance [19]. Studies evaluating hypothyroidism and its effects on cognitive function have shown contrasting results, with recent studies summarised in Table 1B. Parsaik et al. [13] performed a cross-sectional study evaluating the association of overt and subclinical hypothyroidism with mild cognitive impairment over a large population based cohort of older adults. They found no significant association between hypothyroidism (overt or subclinical) with mild cognitive impairment after adjusting for possible confounding factors. On the other hand, longitudinal studies evaluating cognition in patients when hypothyroid and after the restoration of euthyroid state show subtle but definable changes in cognition. Schraml et al. [12] assessed neuropsychological outcomes and thyroid function in a group of patients post thyroidectomy, whilst overtly hypothyroid and following adequate thyroid hormone replacement, as compared with controls. Hypothyroid patients performed worse than controls in the domain of working memory, and this improved following thyroid hormone replacement. Smith et al. [14] assessed clinical status, cognitive performance and driving ability in 32 patients undergoing thyroid hormone withdrawal for radioiodine scanning while hypothyroid and following restoration of euthyroid status. They found that transient profound hypothyroidism was characterised by reversible depression, decreased fine motor performance, slowed reaction times and decreased processing speed. These studies support a link between overt hypothyroidism and a degree of reduced cognitive function, although this may be limited to particular domains. The issue of reversibility of neurocognitive symptoms in treated hypothyroidism has been examined in several studies, summarised in Table 1C. In patients with diagnosed hypothyroidism, residual impairment in cognitive function despite T4 replacement has been reported [15–17]. It is possible that the effects of hypothyroidism on the brain may not be fully reversible, or that exogenous thyroxine replacement falls short of the function of the native pituitary-thyroid axis. It is also plausible that there may be other factors unique to this patient group or their treatment that predisposes them to higher rates of comorbid cognitive disorders. Saravanan e al. [15] surveyed a large group of patients in the UK who had been on T4 for at least 4 months to evaluate their psychological well-being as compared to controls. They found that
268
M. Ritchie, B.B. Yeap / Maturitas 81 (2015) 266–275
Table 1 Overt hypothyroidism. Study author and year
Study
n
Age (years)
Duration of follow up
Results
(A) Mood Heinrich 2003 [4]
Case report
1
73
–
Gulseren 2006 [5]
Longitudinal
33 with overt hypothyroidism 20 healthy controls
43.6 in overt hypothyroidism 40.5 in controls
10.2 months in overt hypothyroidism
Guimaraes 2009 [6]
Cross-sectional
1298 women
53.6 Range 35–91
–
Khemka 2011 [7]
Case series
3
56, 68 and 77
–
Mowla 2011 [8]
Cross-sectional
75 adults with major depressive disorder
34.8
–
Lin 2013 [9]
Case report
1
41
2 years
Wu 2013 [10]
Longitudinal
761,834 from general population database, 4593 with depressive disorder
≥18
5
Presentation with acute mania and hypothyroidism, improved with 2 weeks antipsychotic and continuing T4. Anxiety and depressive symptoms were more severe in patients with overt hypothyroidism. Psychological symptoms improved with treatment. Women with TSH > 4 mIU/L and T4 < 9 pmol/L (n = 20) had increased risk of depressive symptoms on questionnaire. Presentations with acute mania and hypothyroidism, improved with antipsychotic and T4. Adults with TSH > 4 mIU/L, Free T4 < 9 pmol/L (n = 7) had lower scores on depressed mood, guilt, suicidality, insomnia but higher scores on agitation and anxiety. Presentation with acute mania and hypothyroidism, improved with antipsychotic and T4. Greater prevalence and incidence of hypothyroidism (ICD coded) in persons with major depressive disorder.
(B) Cognition Kramer 2009 [11]
Cross-sectional, cohort
1034
76.1
–
Schraml 2011 [12]
Case control, before and after T4
11 patients post thyroidectomy 11 healthy controls
33.0 in cases 32.9 in controls
14 weeks
Parsaik 2014 [13]
Cross-sectional, cohort
1904
81.2 (median) in hypothyroid persons
–
Smith 2015 [14]
Case series, before and after T4 deprivation
32 (20 studied after T4 replacement)
44.8 40.7 in recovery subgroup
4 months
(C) Effects of treatment Saravanan 2002 [15]
Case control Cross sectional
59.7 for cases 59.4 for controls 47.8
–
Wekking 2005 [16]
–
Samuels 2007 [17]
Case control
597 patients 551 controls 141 hypothyroid patients on T4 replacement 34 hypothyroid women on 20 control women
20–44
–
patients on T4 replacement, even with normal TSH, had much higher levels of impairment in psychological well-being compared to age and sex matched controls. Given the high numbers of people on T4 replacement, it was proposed that this could contribute to a substantial burden of psychological morbidity in the community. Samuels et al. [17] performed a cross-sectional comparison of euthyroid and treated hypothyroid patients with normal TSH levels for psychological and cognitive function, and found that the group of T4 treated patients had reduced health status, psychological function, working memory and motor learning compared to euthyroid controls. Additionally, Wekking et al. [16] assessed neurocognitive functioning and well-being of patients
No difference in cognitive function in patients with treated hypothyroidism (n = 149) compared to euthyroid (n = 885). Reduced working memory and more depressive symptoms in thyroidectomised patients when hypothyroid compared to controls, which improved with treatment. Hypothyroid (TSH > 50 mIU/L) inversely associated with word fluency and working memory. No association of previously diagnosed hypothyroidism (n = 313) with mild cognitive impairment. Increased reaction and fine motor times, and increased depressive symptoms when hypothyroid. Improvement in subgroup following T4 treatment. Reduced psychological wellbeing in hypothyroid patients on T4. Impaired scores on memory tasks and auditory attention in hypothyroid patients compared to reference group. Impaired wellbeing, motor learning and working memory in T4 treated women.
with primary hypothyroidism on adequate T4 treatment, and found poor performance in various domains of neurocognitive functioning, especially complex attention tasks and verbal memory tests, with decreased levels of well-being. The authors concluded that neurocognitive function and psychological impairment may not be completely restored in hypothyroid patients despite T4 replacement. By contrast, a cross-sectional study by Kramer et al. [11] found that patients who have been on long-term treatment for hypothyroidism had no significant difference in scores for cognitive function than euthyroid patients over three standardised cognitive function tests. Further research will be required to elucidate whether hypothyroidism leads to permanent neuropsychological
M. Ritchie, B.B. Yeap / Maturitas 81 (2015) 266–275
269
Table 2 Subclinical hypothyroidism. Study author and year
Study
n
Age (years)
Duration of follow up
Results
(A) Mood Roberts 2006 [21]
Cross-sectional
73
–
Jorde 2006 [22]
Case control
Mean 61.0–63.0 between groups
–
Chueire 2007 [23]
Cross-sectional
67
–
Subclinical hypothyroidism was not associated with depressive symptoms, anxiety or cognitive function. No difference in depressive scores between the two groups. Better performance in tests of emotional function. 9.5% of participants with elevated TSH were depressed. 31% of patients with depression had elevated TSH.
Kim 2010 [24]
Cross-sectional
5868 168 with subclinical hypothyroidism 89 with subclinical hypothyroidism (TSH 3.5–10.0 mIU/L) 154 controls 323 252 with TSH > 4.5 mIU/L 71 with primary depressive disorder 495 37 with hypothyroidism (TSH > 4.5 mIU/L)
72.4
–
Hypothyroidism (not specified as overt/subclinical) was not associated with depression.
(B) Cognition Gussekloo 2004 [25]
Longitudinal
85
4 years
Roberts 2006 [21]
Cross-sectional
73
–
Jorde 2006 [22]
Case control Subgroup – randomised placebo controlled trial (T4 vs placebo)
62.4 (subclinical hypothyroidism) 61.0 (controls)
12 months
No association of elevated TSH with depressive symptoms or global cognition. Subclinical hypothyroidism was not associated with depressive symptoms or anxiety or cognitive function. No difference in cognitive function between groups. No difference in cognition or emotional function between T4 or placebo treatment.
Bensenor 2010 [26]
Cross-sectional
>65
–
Kim 2010 [24]
Cross-sectional
72.4
–
Silva 2013 [27]
Cross-sectional
80.7
–
Parsaik 2014 [13]
Cross-sectional, cohort
(C) Effects of treatment Arinzon 2006 [28]
Cross-sectional
Parle 2010 [29]
Double-blind placebo-controlled trial of T4 vs placebo
599 67 with elevated TSH (>4.8 mIU/L) 5868 168 with subclinical hypothyroidism 89 with subclinical hypothyroidism (TSH 3.5–10.0 mIU/L) 154 controls Subgroup – 36 for T4, 34 for placebo 1276 157 subclinical hypothyroidism 495 37 with hypothyroidism (TSH > 4.5 mIU/L) 411 43 subclinical hypothyroidism 284 euthyroid 1904
57 26 with subclinical hypothyroidism 31 with overt hypothyroidism 94 with subclinical hypothyroidism
77.7
Patients reevaluated after 3 months thyroxine
73.5 (T4 group), 74.2 (placebo group)
6 and 12 months
3. Subclinical hypothyroidism Subclinical hypothyroidism is defined biochemically by elevated TSH concentration with normal FT4 concentration [3], and is more common in women and older adults [20]. Cross-sectional studies to date have shown mixed results regarding associations of subclinical hypothyroidism and altered mood, with recent studies summarised in Table 2A. Roberts et al. [21] showed no association between subclinical hypothyroidism and depression or anxiety in older adults. Kim et al. [24] reported a similar negative finding in
No difference in depressive symptoms and cognition between groups.
No association of previously diagnosed hypothyroidism (n = 141) with mild cognitive impairment.
81.7 (median) in hypothyroid persons
impairment, or whether conventional replacement therapies may not prevent or reverse more subtle cognitive dysfunction associated with hypothyroidism.
No association between subclinical hypothyroidism and cognitive dysfunction. Hypothyroidism (not specified as overt/subclinical) was not associated with cognitive impairment.
Improvement in cognitive and functional status was found after treatment for subclinical hypothyroidism and overt hypothyroidism. No improvement in cognitive function with T4.
their evaluation of older patients with hypothyroidism (defined as TSH 4.5–17.2 but not classified as overt or subclinical hypothyroidism). By contrast, Chueire et al. [23] found that older adults with raised TSH were more likely to be depressed. With regards to cognition, the study results have been more consistent, with three recent cross sectional studies [21,24,27] showing no association between subclinical hypothyroidism and cognitive dysfunction, as summarised in Table 2B. A case control study [22] showed no neuropsychological dysfunction in patients with subclinical hypothyroidism compared with controls. Gussekloo et al. [25] performed a prospective observational study of older adults and found that those with higher TSH do not experience increased adverse events and may in fact have a prolonged life span. A longitudinal treatment study by Arinzon et al. [28] found that patients with
270
M. Ritchie, B.B. Yeap / Maturitas 81 (2015) 266–275
Table 3 Euthyroidism. Study author and year
Study type
n
Age (years)
Duration of follow up
Results
(A) Mood Medici 2014 [34]
Longitudinal
1503
70.6
8 years
Almeida 2011 [35]
Longitudinal
3932 men
75.3
5.5 years
Cross sectionally lower TSH (0.3–1.0 mIU/L) was associated with more depressive symptoms than higher TSH (1.6–4.0 mIU/L). Lower TSH was associated with higher incidence of depressive symptoms. The serum concentration of TSH and FT4 did not affect prevalence of depressive symptoms or incidence of depression.
(B) Cognition Gussekloo 2004 [25]
Longitudinal
85
4 years
Elderly individuals with increased TSH did not experience adverse events and may have a prolonged life span.
Roberts 2006 [21]
Cross-sectional
73
–
Hogervorst 2008 [36]
Longitudinal
599 67 with elevated TSH (>4.8 mIU/L) 17 with low TSH (2.1 mIU/L was associated with increased risk for Alzheimer’s dementia in women but not men. Higher total and free thyroxine was associated with an increased risk of dementia and Alzheimer’s disease. In 143 men who had an autopsy study, higher total thyroxine was associated with a higher number of neocortical plaques and neurofibrillary tangles. No association of TSH with mild cognitive impairment or Alzheimer’s disease. High TSH (>3.59 mIU/L) was associated with an increased risk of vascular dementia. High free T4 levels predicted new-onset dementia in older men. Higher free T4 or lower TSH was associated with better performance on specific tests of cognitive function. Higher TSH was associated with lower incidence of dementia.
subclinical or overt hypothyroidism experienced improvement in cognitive and functional status after T4 supplementation. By contrast Parle et al. [29] performed a double-blind placebo controlled trial of T4 in patients with subclinical hypothyroidism and followed them at 6 and 12 months assessing cognitive function. They found no evidence that T4 treatment in elderly patients with subclinical hypothyroidism improved cognitive function. Overall these studies suggest equivocal effects of subclinical hypothyroidism on depression and cognition. Further interventional studies including randomised controlled trials would be required to clarify the effect of hormonal interventions on mood and cognition in adults with subclinical hypothyroidism. 4. Differences in thyroid hormones within the normal range Recently there has been emerging interest in the impact of differences in thyroid hormones within the reference range on mood and cognition, especially in older adults. Changes in thyroid
hormone production and metabolism occur with increasing age, typically reduced production of T4 and T3, but relatively unchanged serum concentrations of total and FT4 due to decreased degradation of T4 [30]. Of note, distributions and reference limits for TSH concentrations shift higher with increasing age, contrasting with the relative stability of FT4 levels [31,32]. Therefore, even in circumstances where the log-linear relationship between FT4 and TSH is preserved, FT4 and TSH may be regarded as distinct risk predictors in older adults [33]. Studies assessing different levels of thyroid hormones within the normal range and their effects on mood have had conflicting results, with recent studies summarised in Table 3A. In a study of 3932 community-dwelling older men, Almeida et al. [35] found that the serum concentrations of TSH and FT4 did not affect the prevalence of depressive symptoms or incidence of depression. On the other hand, in their longitudinal cohort study of 1503 older adults, Medici et al. [34] found that those with low normal TSH had more concurrent depressive symptoms and a substantially increased risk for developing a depressive syndrome in subsequent years.
M. Ritchie, B.B. Yeap / Maturitas 81 (2015) 266–275
271
Table 4 Subclinical hyperthyroidism. Study author and year
Study
n
Age (years)
Duration of follow up
Results
(A) Mood Roberts 2006 [21]
Cross-sectional
5868 127 with subclinical hyperthyroidism
73
–
Kim 2010 [24]
Cross sectional
495 14 with hyperthyroidism (TSH < 0.5 mIU/mL)
72.4
–
No association of subclinical hyperthyroidism with depressive symptoms, anxiety or cognition in models adjusted for covariates. Hyperthyroidism (not specified as overt/subclinical) was not associated with depression.
(B) Cognition Kalmijn 2000 [43]
Longitudinal
1843
68.8
2.1 years
Gussekloo 2004 [25]
Longitudinal
85
4 years
Roberts 2006 [21]
Cross-sectional
73
–
Ceresini 2009 [44]
Cross-sectional
23–102
–
Bensenor 2010 [26]
Cross-sectional
>65
–
Kim 2010 [24]
Cross-sectional
72.4
–
Hyperthyroidism (not specified as overt/subclinical) was associated with cognitive impairment.
Vadiveloo 2011 [45]
Longitudinal
599 17 with low TSH (20% increased risk of dementia and >30% increased risk of Alzheimer’s disease. In a sub-sample who underwent autopsy, it was also shown that those with higher total thyroxine had a greater number of neocortical plaques and neurofibrillary tangles [38], suggesting a pathophysiological mechanism for their clinical findings. In a longitudinal study of 3401 older men Yeap et al. [40] found that higher FT4 levels predicted new-onset dementia with an 11% increase in risk per 1 pmol/L increment in FT4 and a 70% higher hazard ratio for developing dementia in men with FT4 greater than the 25th percentile. In that study TSH was not associated with risk of dementia. In an Italian cohort study of 660 elderly subjects, Forti et al. [39] found baseline TSH was not associated with mild cognitive impairment or Alzheimer’s disease, while higher TSH was associated with an increased risk of vascular dementia. More recently in a large longitudinal study of older adults Cappola et al. [42] found an association of higher TSH concentrations with lower incidence of dementia, which was attenuated by adjustment for covariates. Therefore, although the link between differences in thyroid hormones within the reference range and mood changes has been inconsistent, the data support an association of exposure to thyroid hormone, particularly higher FT4, with
Reduced TSH (
