Sie sind auf Seite 1von 7


E n d o c r i n e

C a r e

Health Status, Mood, and Cognition in Experimentally

Induced Subclinical Thyrotoxicosis
M. H. Samuels, K. G. Schuff, N. E. Carlson, P. Carello, and J. S. Janowsky
Division of Endocrinology, Diabetes and Clinical Nutrition (M.H.S., K.G.S.), Division of Biostatistics (N.E.C.), and Department of Behavioral
Neurosciences (P.C., J.S.J.), Oregon Health & Science University, Portland, Oregon 97239

Objective: Our objective was to determine whether subclinical thyrotoxicosis alters health status,
mood, and/or cognitive function.
Design: This was a double-blinded, randomized, cross-over study of usual dose L-T4 (euthyroid arm)
vs. higher dose L-T4 (subclinical thyrotoxicosis arm) in hypothyroid subjects.
Patients: A total of 33 hypothyroid subjects receiving L-T4 were included in the study.
Measurements: Subjects underwent measurements of health status, mood, and cognition: Short
Form 36 (SF-36); Profile of Mood States (POMS); and tests of declarative memory (Paragraph Recall,
Complex Figure), working memory (N-Back, Subject Ordered Pointing, and Digit Span Backwards),
and motor learning (Pursuit Rotor). These were repeated after 12 wk on each of the study arms.
Results: Mean TSH levels decreased from 2.15 to 0.17 mU/liter on the subclinical thyrotoxicosis arm
(P 0.0001), with normal mean free T4 and free T3 levels. The SF-36 physical component summary
and general health subscale were slightly worse during the subclinical thyrotoxicosis arm, whereas
the mental health subscale was marginally improved. The POMS confusion, depression, and tension
subscales were improved during the subclinical thyrotoxicosis arm. Motor learning was better
during the subclinical thyrotoxicosis arm, whereas declarative and working memory measures did
not change. This improvement was related to changes in the SF-36 physical component summary
and POMS tension subscales and free T3 levels.
Conclusions: We found slightly impaired physical health status but improvements in measures of
mental health and mood in L-T4 treated hypothyroid subjects when subclinical thyrotoxicosis was
induced in a blinded, randomized fashion. Motor learning was also improved. These findings
suggest that thyroid hormone directly affects brain areas responsible for affect and motor
function. (J Clin Endocrinol Metab 93: 1730 1736, 2008)

ubclinical thyroid dysfunction [abnormal serum TSH with

normal free T4 (fT4) and free T3 (fT3) levels] may have
central nervous system effects. This has been best studied in subclinical hypothyroidism, with some (but not all) reports showing
adverse effects on mood and cognition, and improvements after
L-T4 treatment (reviewed in Refs. 1 and 2). Subclinical hyperthyroidism has been less well studied in this regard, with reports
showing increased levels of anxiety or depression and decrements in cognitive domains (3 8). However, this is not uniform
(5, 9 11), and there is little information on treatment effects.

We recently reported results of a randomized, double-blind,

cross-over study in which L-T4 treated subjects received their
usual dose of L-T4, or a slightly lower dose (12). Induction of
subclinical hypothyroidism led to decrements in specific domains of health status, mood, and working memory. In the current study, we applied a similar experimental design to induce
subclinical thyrotoxicosis. We chose to focus on memory, based
on our previous data and studies suggesting that memory is preferentially affected in subjects with thyroid dysfunction (2, 12,
13), and because animal studies support a major role for L-T4 in


Abbreviations: CV, Coefficient of variation; fT3, free T3; fT4, free T4; HSS, Hyperthyroid
Symptom Scale; MCS, mental component summary; MH, mental health; NS, not significant; PCS, physical component summary; POMS, Profile of Mood States; SF-36, Short Form
36; SOP, Subject Ordered Pointing; WAIS-R, Wechsler Adult Intelligence Scale-Revised.

Printed in U.S.A.
Copyright 2008 by The Endocrine Society
doi: 10.1210/jc.2007-1957 Received September 4, 2007. Accepted February 8, 2008.
First Published Online February 19, 2008


J Clin Endocrinol Metab. May 2008, 93(5):1730 1736

J Clin Endocrinol Metab, May 2008, 93(5):1730 1736

brain areas that subserve memory (14 18). We studied three

distinct forms of memory linked to specific brain systems: declarative memory (medial temporal lobe), working memory (prefrontal cortex), and motor learning (cerebellum and basal ganglia) (19). We included validated measures of health status and
mood because these may be altered in subjects with subclinical
thyrotoxicosis (3 8, 11, 20, 21), and may secondarily affect
cognition. We hypothesized the following:
1. Subjects with induced subclinical thyrotoxicosis have
decrements in health status and mood compared with the euthyroid state.
2. Subjects with induced subclinical thyrotoxicosis have specific decrements in memory, which may be related to changes in
health status or mood.
3. In subjects with induced subclinical thyrotoxicosis, changes
in health status, mood, and/or cognition are correlated with
changes in TSH, fT4, and/or fT3 levels.

Subjects and Methods

Experimental subjects
A total of 33 subjects (31 women, two men), ages 24 45 yr, with
adult-onset primary hypothyroidism was recruited. There were 31 subjects that had autoimmune hypothyroidism, and two had received iodine-131 therapy for Graves disease. Duration of hypothyroidism averaged 7 yr (range 6 months to 25 yr). All subjects had elevated TSH levels
before L-T4 treatment or while on lower L-T4 doses. Subjects were receiving L-T4 as the sole treatment in stable doses for at least 3 months,
with normal TSH levels. Subjects did not have any acute or chronic
illnesses and were not receiving medications that affect thyroid hormone
levels, mood, or cognition. Stable doses of oral contraceptive or estrogen
therapy were allowed during the study. Testing was done during the first
10 d after onset of menstrual bleeding, or in the first 10 d of an oral
contraceptive cycle.

Experimental design (Fig. 1)

The protocol was approved by the Oregon Health & Science University Institutional Review Board, and subjects gave written informed consent.


vised (WAIS-R) Vocabulary subtest, which correlates with intelligence

quotient (22). Subjects were excluded if they scored less than eight
(scaled). The Symptom Checklist 90-R was administered to screen subjects for psychiatric diseases (23). Subjects who scored outside the normative range were excluded.

Baseline visit
Within 3 wk, subjects returned for a 90- to 120-min baseline visit.
Subjects refrained from taking their L-T4 dose that morning. Serum TSH,
fT4, and fT3 levels were obtained. The Hyperthyroid Symptom Scale
(HSS) was completed (24). Subjects completed validated measures of
health status and mood.
The Short Form 36 health survey (SF-36), a questionnaire about general
health (22). Higher scores on the SF-36 summary scales and subscales
reflect better health status and well-being. This instrument has been used
in studies involving thyroid disease (25), and decrements were found in
our study of subclinical hypothyroidism (12).
The Profile of Mood States (POMS), a questionnaire about mood (22).
Higher scores on the POMS subscales reflect mood decrements, except for the vigor subscale, in which higher scores reflect improved
Cognitive tests were administered by a single research assistant in a
standard fashion. Based on existent literature and our previous study, we
did not use a battery of general cognitive measures but, rather, validated
measures targeted to specific domains likely to be affected by altered
thyroid status.

Tests of declarative memory (hippocampus, medial

temporal lobe)
Paragraph Recall (verbal memory)
For this subtest of the Wechsler Memory Scale-Revised, subjects were
read two brief stories, and verbally recalled each of them immediately
and after a 30-min interval. The outcome measure was the total number
of story elements recalled at each interval (26).

Complex Figure Test (visual memory)

For the Medical College of Georgia Complex Figure Test, subjects
copied a standard complex figure, and drew it from memory immediately
and after a 30-min interval. The outcome measure was the total number
of elements correctly reproduced (22).

Tests of working memory (prefrontal cortex)

Screening visit
Subjects were screened for general health, medication use, drug
abuse, thyroid status, and mood or cognitive disorders by history, physical examination, laboratory, and electrocardiogram testing. General
intelligence was measured by the Wechsler Adult Intelligence Scale-Re-

N-back test
A series of letters was presented one at a time on a screen. Subjects
responded when a letter appeared that they had seen on the previous
screen (one back). The task was repeated with an increase in memory
load by having the subject respond when a letter appeared three back.
The outcome measure was the total number correct (27).

Subject Ordered Pointing (SOP)

Subjects were presented with stacks of cards (six, eight, 10, or 12
per set). Each card in a set showed the same array of abstract drawings,
but in a different spatial arrangement. The subject was instructed to
touch one drawing on each card, but not to touch the same drawing
on subsequent cards in the set. Subjects erred when they touched a
drawing that had been touched on a previous card. Each card set was
repeated three times. The outcome measure was the total number of
errors across each card set (22).

Digit Span Backwards

FIG. 1. Study flow diagram.

The examiner read number sequences of increasing length, and

after each sequence the subject repeated the sequence backwards.


Samuels et al.

Mood and Memory in Subclinical Thyrotoxicosis

Subjects erred when they could not successfully repeat two sequences
of the same length (22).

Motor learning (basal ganglia, cerebellum)

Pursuit Rotor
Subjects held a photosensitive wand to maintain contact with a 2-cm
light disk rotating on a variable speed turntable (Model 30014; Lafayette
Instrument Co., Lafayette, IN). An initial block of four trials was administered at 15, 30, 45, and 60 revolutions per minute to establish the
optimal speed for further trials. Two blocks of eight 20-sec trials were
then administered, with a 20-sec rest after each trial and a 60-sec rest
period after four trials. After a 30-min interval, the two blocks were
repeated. The outcome measure was the mean total time the stylus remained on target during each trial (28).
Subjects were randomized to receive either their usual doses of L-T4
(euthyroid arm) or higher doses of L-T4, intended to lead to a TSH level
of 0.01 0.4 mU/liter (subclinical thyrotoxicosis arm), while maintaining
normal fT4 and T3 levels. Our target TSH level reflects the range seen in
clinical practice, while avoiding overtreatment and possible symptoms.
Based on average suppressive L-T4 doses in thyroid cancer patients, estimated initial L-T4 dose increases were 30%. The subjects and the physician (M.H.S.) and research assistant (P.C.) with direct subject contact
were blinded, whereas the monitoring physician (K.G.S.) was unblinded,
to adjust L-T4 doses and assess reported side effects. L-T4 pills were placed
in gel capsules to maintain blinding.
Six-week interim visit. Six weeks later, subjects returned for a brief visit.
Compliance was assessed by pill counts. Safety was assessed by questioning for symptoms, performing electrocardiograms, serum TSH, fT4,
and total T3 levels. The monitoring physician (K.G.S.) made L-T4 dose
adjustments as needed to attain target TSH levels for that arm of the
study, keeping fT4 and T3 levels within the normal range.
Cross-over visit. Twelve weeks after the baseline visit, subjects returned
for an extended visit. Serum TSH, fT4, and fT3 levels were measured.
The HSS, SF-36, POMS, and cognitive tests were repeated in the same
order as the baseline visit. Subjects were crossed over to the second arm
of the study and had their new doses of L-T4 dispensed.
Eighteen-week interim visit. This visit was identical to the 6-wk visit,
with subjects now on the alternate L-T4 dose.
End of study visit. Twelve weeks after the cross-over visit, subjects
returned for a visit identical to the cross-over visit. Subjects were
asked which arm of the study they thought was the higher dose arm

J Clin Endocrinol Metab, May 2008, 93(5):1730 1736

and which arm they preferred. They were then placed back on their
usual doses of L-T4.

Analytical methods
TSH was measured by immunochemiluminometric assay (Nichols
Institute, San Juan Capistrano, CA): sensitivity 0.003 mU/liter, normal
range 0.28 5.00, intraassay coefficients of variation (CVs) 9.5% at 0.03
mU/liter and 4.7% at 11.6 mU/liter, and interassay CVs 17% at 0.02
mU/liter and 4.6% at 14 mU/liter. fT4 was measured by immunochemiluminometric assay: sensitivity 0.08 ng/dl, normal range 0.71.8, intraassay CVs 5.7% at 0.27 ng/dl and 1% at 4.6 ng/dl, and interassay CVs
6.8% at 0.3 ng/dl and 1.6% at 3.8 ng/dl. fT3 was measured by tracer
dialysis (Nichols Institute): sensitivity 25 pg/dl, normal range 210 440,
intraassay CV 6%, and interassay CV 4%.

Statistical methods
Differences in outcomes between the two study arms were explored
by paired t tests. Sets of quality of life, mood, and cognitive measures
were then analyzed together using mixed effects models. Age, years of
education, WAIS-R, and baseline measures were covariates to reduce
residual variability. Generalized Estimating Equations (29) were used to
estimate group differences for measures with ceiling or floor effects. Each
set of tests was analyzed as a group to adjust for within-subject correlation and correlations between responses on similar tests; results are
grouped together in the results tables (see Tables 2 and 3). Bonferroni
adjustments were used for formal pairwise comparisons, when applicable. P values less than 0.05 were considered significant.
To investigate associations between changes in TSH, fT4, or fT3
and outcomes, three mixed effects models were developed for each
outcome, with one thyroid hormone as a covariate. In a final model
for each outcome, the association between a single thyroid hormone
measure was adjusted for the two other thyroid hormone measures.
The number of models was large, so we were interested in statistically
significant findings across similar measures and similar hormones
rather than in individual P values.
To assess whether differences in cognitive measures could be explained by differences in health status or mood, the model with significant cognitive differences (Pursuit Rotor) was adjusted for
changes in SF-36 and/or POMS measures. Separate models were fitted
to assess effects of each individual change in SF-36 and/or POMS score
that showed significant euthyroid-thyrotoxicosis differences. In a final analysis, we simultaneously adjusted for significant thyroid hormone and health status or mood associations to assess independent
effects on the cognitive measures.

TABLE 1. Clinical parameters and thyroid function tests at baseline and the end of each arm of the study (euthyroid and
subclinical thyrotoxicosis)

(mean SEM)

(mean SEM)

(mean SEM)

P value
(paired t test)

TSH (mU/liter)
fT4 (ng/dl)
fT3 (pg/dl)
Weight (kg)
BMI (kg/m2)
Pulse (beats/min)
Systolic blood pressure (mm Hg)
Diastolic blood pressure (mm Hg)
L-T4 dose (g/kgd)

2.58 0.27
1.33 0.04
262.7 6.3
75.86 3.83
27.1 1.2
70.8 1.8
121.0 1.9
72.2 1.8
1.74 0.30
1.62 0.09

2.15 0.31
1.40 0.04
259.3 6.2
74.45 3.79
26.7 1.2
70.6 2.2
120.0 2.7
70.2 2.1
1.94 0.40
1.66 0.09

0.17 0.06
1.70 0.06
320.2 11.4
73.78 3.67
26.6 1.2
73.5 2.3
118.7 2.5
66.7 2.0
2.06 0.39
2.45 0.09


To convert fT4 levels to International System of Units, multiply by 12.87. To convert fT3 levels to International System of Units, multiply by 0.01536. BMI, Body mass

J Clin Endocrinol Metab, May 2008, 93(5):1730 1736



Health status and mood (SF-36, POMS) (Table 2)

Clinical parameters and thyroid function tests (Table 1)

A total of 47 subjects volunteered for the study, and 33
qualified after the screening visit. Interim L-T4 dose adjustment was necessary in four subjects during the euthyroid arm
(two increased dose, two decreased dose) and in 27 during the
subclinical thyrotoxic arm (20 increased dose further, seven
decreased to intermediate dose). Per study design, mean TSH
levels decreased to 0.17 mU/liter, whereas mean fT4 and fT3
levels increased at the end of the subclinical thyrotoxicosis
arm. A total of 22 subjects had TSH levels less than 0.1 mU/
liter, whereas 11 had TSH levels more than 0.1 mU/liter. Four
subjects had slightly high fT4 levels, one with a slightly high
fT3 level. Thus, we were successful in inducing subclinical
thyrotoxicosis, with TSH levels in our target range, and almost all serum fT4 and fT3 levels in the normal range.
All subjects completed the study. There were no differences in
weight, pulse, systolic blood pressure, or HSS at the end of the
two arms of the study; diastolic blood pressure was slightly lower
at the end of the subclinical thyrotoxicosis arm (P 0.03).
Subjects did not reliably predict which arm was the
subclinical thyrotoxicosis arm: 15 guessed correctly, 10
guessed incorrectly, and eight had no opinion [P not significant (NS) by binomial calculation]. There were 15 that
preferred the actual higher dose, seven preferred the actual
lower euthyroid dose, and 11 had no preference (P NS).
Sixteen preferred the dose they perceived was the higher dose,
five preferred the dose they perceived was the euthyroid dose,
and 12 had no preference (P NS).

Physical health status. The SF-36 physical component summary (PCS) scale score was slightly worse at the end of the subclinical thyrotoxicosis arm (P 0.01), and the general health
subscale was marginally worse (P 0.06).


Mental health (MH) status. The mental component summary (MCS) scale score was slightly but not significantly better (P 0.15) at the end of the subclinical thyrotoxicosis arm,
and the MH subscale was improved (P 0.01). The confusion, depression, and tension subscales of the POMS were
improved during the subclinical thyrotoxicosis arm (P
0.04 0.06). There were no significant effects on other SF-36
or POMS subscales.
Changes in the SF-36 MCS and general health subscale
scores were directly related to changes in fT3 levels between
the two study arms (P 0.01). Changes in the SF-36 MCS
were also directly related to changes in fT4 levels (P 0.02).
There were no other associations between changes in the
SF-36 or POMS scales and changes in TSH, fT4, or fT3 levels
(data not shown).
Cognitive tests (Table 3 and Fig. 2)
Declarative memory. There were no differences between the
two study arms in Paragraph Recall (verbal memory) or performance on the Complex Figure Test (visual memory).
Working memory. There were no differences between the two
study arms in N-Back number correct, SOP, or Digit Span

Summary statistics and P values for health status and mood measures during the two treatment arms


(mean SEM)

(mean SEM)

(mean SEM)


51.72 1.02
51.91 1.33

50.69 1.46
56.59 0.61

52.71 1.16
54.75 0.83


76.00 4.09
80.84 2.65
80.38 1.60
59.84 2.83
91.72 2.82
76.56 6.54
86.33 3.86
91.67 2.99

84.43 3.01
85.03 2.23
78.40 2.28
66.83 2.57
96.33 1.10
93.33 3.16
93.33 2.05
86.67 4.40

81.87 3.14
81.55 2.28
84.13 1.55
69.03 1.96
94.52 1.56
93.33 3.38
89.58 3.55
88.89 4.33


42.13 0.89
41.78 1.07
41.16 0.73
44.72 1.37
41.66 1.06
51.84 1.10

44.50 1.11
42.44 1.54
42.66 1.06
43.38 1.16
42.53 1.17
53.16 1.39

43.16 1.21
40.00 1.03
40.44 0.65
42.59 1.03
39.16 1.64
55.75 1.28



P value
(paired t test)

P values
Treatment: 0.15
Treatment: 0.01
Treatment test: 0.008

Treatment: 0.002


P values for unadjusted paired t tests comparing the two arms of the study are shown in the fifth column. Adjusted P values in the repeated measures analysis are
shown in the sixth column. These P values were adjusted for age, years of education, WAIS-R, and baseline measure for that particular variable. Significant results are
highlighted in bold. For the adjusted analyses, treatment test P values were examined initially. If the treatment test P value was more than 0.1, indicating no treatmenttest interaction, it is not shown in the table, and only the overall treatment effect is shown. If the treatment test P value was 0.1 or less, indicating a possible treatmenttest interaction, this value is listed in the table, and subsequent analyses were done for individual measures. A, Anxiety; BP, bodily pain; C, confusion; D, depression; F,
fatigue; PF, physical functioning; RE, role emotional; RP, role physical; SF, social functioning; T, tension; V, vigor; VT, vitality.

Raw scores compared using an exact McNemars test because many values were at ceiling.


Samuels et al.


Mood and Memory in Subclinical Thyrotoxicosis

J Clin Endocrinol Metab, May 2008, 93(5):1730 1736

Summary statistics and P values for cognitive measures during the two treatment arms of the study

Declarative memory
Paragraph Recall
30-min delay
Complex Figure
3 min
30 min
Working memory
N-Back number correct
1 back
3 back
SOP errors
Digit Span Backwards
Motor learning
Pursuit Rotor Trial

(mean SEM)

(mean SEM)

(mean SEM)

P value
(paired t test)

14.06 0.63
13.13 0.61

15.75 0.61
14.56 0.62

16.22 0.49
15.06 0.48


34.25 0.31
28.31 0.72
28.11 0.81

34.78 0.22
29.78 0.82
29.20 0.83

34.84 0.19
31.23 0.66
30.45 0.72


14.59 0.28
9.25 0.40

15.00 0.14
10.31 0.35

15.06 0.16
10.41 0.43


0.40 0.08
0.86 0.11
1.32 0.12
1.59 0.18
7.56 0.37

0.32 0.06
0.72 0.11
0.94 0.12
1.23 0.16
8.22 0.42

0.32 0.07
0.62 0.12
0.77 0.11
1.30 0.15
8.53 0.44


23.96 1.16
26.69 1.48
28.24 1.84
28.67 1.51

32.85 2.31
33.05 2.01
32.39 2.23
34.44 2.08

35.19 2.16
36.34 2.07
38.08 2.39
39.62 2.31

P values
Treatment: 0.25

Treatment: 0.29

Treatment: 0.67

Treatment: 0.28

Treatment: 0.33
Treatment: <0.001


Individual cognitive tests are grouped by the three memory subdomains (first column). P values for unadjusted paired t tests comparing the two treatment arms are
shown in the fifth column. Adjusted P values in the repeated measures analysis are shown in the sixth column. These P values were adjusted for age, years of
education, WAIS-R, and baseline measure for that particular variable. Significant results are highlighted in bold. For the adjusted analyses, treatment test P values were
examined initially. There were no treatment-test interactions, and, thus, only the overall treatment effect is shown.

Motor learning. The Pursuit Rotor results were significantly

better at the end of the subclinical thyrotoxicosis arm, compared
with the euthyroid arm (P 0.001 by adjusted t tests) (Fig. 2).
By unadjusted t tests, there was no difference between the two
arms for the initial Pursuit Rotor trial, whereas the subsequent
trials were all better at the end of the subclinical thyrotoxicosis
arm (P 0.001 to 0.04).
The differences in Pursuit Rotor were marginally related to
changes in fT3 levels between the two study arms (P 0.06). The
Pursuit Rotor differences were not related to changes in fT4 or
TSH levels (data not shown).
The differences in Pursuit Rotor between the two study arms
were also related to changes in the SF-36 PCS scale score (P
0.003) and the POMS tension subscale (P 0.008), but not to

the other POMS subscales. These relationships were independent of changes in fT3 levels.
There were slight differences between baseline and euthyroid
arm levels for some of the health status, mood, and cognitive
tests, likely reflecting effects due to close monitoring during the
study, and learning effects of repeated testing. We accounted for
these differences by randomized order of the two arms, and by
using baseline levels as a covariate in our mixed models.
We repeated our analysis after exclusion of four subjects with
slightly elevated fT4 levels at the end of the subclinical thyrotoxicosis arm. The results were similar, although the significance
level decreased slightly for some of the results due to the smaller
sample size. Statistical significance was lost for the SF-36 general
health subscale and the POMS depression subscale.


FIG. 2. Pursuit Rotor results at the end of the euthyroid arm (solid line)
and subclinical thyrotoxicosis arm (dashed line). The x-axis shows each of
the four trials, whereas the y-axis shows time on target for each trial (sec).
Results were better during the subclinical thyrotoxicosis arm of the study
(P 0.001 by mixed effects model).

We investigated effects of subclinical thyrotoxicosis on health

status, mood, and cognition by controlling L-T4 doses in subjects
with treated hypothyroidism. This model circumvents limitations in studying endogenous subclinical hyperthyroidism: subject recruitment and heterogeneity, variable natural history, and
subjects awareness of their thyroid status. It also circumvents
limitations in studying patients receiving suppressive L-T4 doses
for nodular thyroid disease or thyroid cancer, who may have
health status alterations due to the underlying disease. In one
previous report of a similar model, Walsh et al. (10) also safely

J Clin Endocrinol Metab, May 2008, 93(5):1730 1736

and effectively induced subclinical thyrotoxicosis in a blinded,

controlled fashion.
Our subjects did not experience changes in weight, pulse, or
systolic blood pressure during the subclinical thyrotoxicosis
arm; the slight decrease in diastolic blood pressure is consistent
with known effects of thyroid hormone on systemic vascular
resistance (30). There were no changes in the HSS, and our subjects did not accurately guess which arm included the higher L-T4
dose. This confirms adequate blinding during the study.
Our first hypothesis was that subjects would have decrements
in health status and/or mood during the subclinical thyrotoxicosis arm. We did find slight decrements in the SF-36 PCS score
and general health subscale. In contrast, there were consistent
improvements in MH measures, including the SF-36 MCS score
and MH subscale, and POMS confusion, depression, and tension
subscales. This suggests specific positive effects of mild thyrotoxicosis on MH and mood, despite decrements in physical
health indices. These effects are likely to be clinically relevant
because the magnitude of the differences in the SF-36 scales are
similar to changes reported with weight loss among overweight
adults (31), rosiglitazone treatment of polycystic ovarian syndrome (32), and gabapentin treatment of diabetic neuropathy
Previous studies reported variable effects of subclinical thyrotoxicosis on health status and mood, with most finding decrements compared with euthyroid subjects (3 8, 20, 21), or no
differences (9, 10, 34). Only a few studies reported improved
mood or well-being in subclinical thyrotoxicosis subjects (11, 35,
36). Our study had the advantage of subjects being blinded to
their treatment status. Previous findings may have been influenced by patients knowledge of their thyroid status. However,
our findings are discordant with a recent large cross-sectional
study in elderly subjects, which found no mood alterations in
subclinical thyroid disease (9). Our results are also discrepant
with the one other randomized, blinded study in hypothyroid
subjects (10). Subjects in that study were older (mean age 53 yr),
L-T4 dose alterations were less, and no specific measure of mood
such as the POMS was used, which may explain the differences.
Our second hypothesis was that subjects would have decrements in memory during the subclinical thyrotoxicosis arm. Instead, we found no changes in declarative or working memory,
in agreement with the recent large cross-sectional study (9). Using a similar study design as ours, Walsh et al. (10) also reported
no changes in working memory. However, we did find improvements in motor learning, a form of implicit or nondeclarative
memory. This was mostly independent of motor speed, and
changes in health status and mood. The magnitude of the difference in Pursuit Rotor performance was similar to that seen
with acute alcohol ingestion in healthy young men (37) and likely
has clinical relevance.
Previous studies investigated motor speed and psychomotor
function in hyperthyroidism, reporting either decrements or no
differences compared with euthyroid controls (6, 8, 33). However, to our knowledge there are no studies of motor learning in
overt or subclinical hyperthyroidism. Our findings suggest that
motor learning is an important target for future studies of mild
thyroid dysfunction.


Our final hypothesis was that changes in outcomes would be

correlated with changes in thyroid hormone levels. This hypothesis was not consistently supported. One explanation is that serum thyroid hormone and TSH levels do not adequately reflect
thyroid hormone action in the brain. Thyroid hormone levels are
tightly regulated within the brain in a tissue-specific fashion that
is not completely dependent on serum levels, and there may also
be other mediators of thyroid hormone action in those tissues
It is interesting to compare our current results with our previous study of subclinical hypothyroidism (12). Using an identical design and outcome measures, we found that subclinical
hypothyroidism was associated with decrements in the SF-36
general health and POMS fatigue subscales. Together, these
studies suggest that there is an optimal window of thyroid
function for overall health (SF-36 PCS and general health subscale) but that MH and mood may continue to improve (SF-36
MCS and POMS) as thyroid function progresses from subclinical
hypothyroidism to subclinical thyrotoxicosis. Subclinical hypothyroidism was associated with decrements in working memory,
but not motor learning. In contrast, subclinical thyrotoxicosis
was associated with improved motor learning, but no changes in
working memory. This suggests that there may be different optimal thyroid hormone levels for the brain areas that control
these functions.
There are five limitations to this study: sample size, variability
in thyroid hormone levels, the studys duration, the cross-over
design, and the age range. Our study was a priori powered on the
cognitive measures. We accounted for multiple comparisons by
statistical adjustment, and we found positive results for a number
of health status and cognitive measures. We do not feel that
variability in thyroid hormone levels is a major limitation because analysis excluding subjects with elevated fT4/fT3 levels did
not alter our results. We limited the time of subclinical thyrotoxicosis to 12 wk, and we recognize that most subjects required
interim L-T4 dose adjustment during this time. It is unclear
whether a longer time period would magnify or attenuate our
findings. The cross-over design could lead to problems with
learning effects during repeated testing, and carryover effects. To
minimize this we used alternative test forms and randomized the
treatment order. Finally, we studied younger subjects to avoid
risks of inducing subclinical thyrotoxicosis, and we recognize
our results may not be generalizable to older subjects.
In summary, we induced subclinical thyrotoxicosis in a controlled, blinded fashion in subjects with primary hypothyroidism. We found minor decrements in general and physical health
status, but improvements in mood and motor learning. Given
these mixed effects, as well as the known deleterious effects of
subclinical hyperthyroidism on other organ systems (39), we do
not recommend that hypothyroid subjects with benign thyroid
disease be treated with suppressive doses of levothyroxine.

Address all correspondence and requests for reprints to: Mary H. Samuels, M.D., CR107, Oregon Health & Science University, 3181 SW Sam


Samuels et al.

Mood and Memory in Subclinical Thyrotoxicosis

Jackson Park Road, Portland, Oregon 97239. E-mail: samuelsm@
This work was supported in part by R21 DK062787 (to M.H.S.) and
UL1 RR024120 (Oregon Health & Science University Clinical and
Translational Research Center).
Disclosure Summary: The authors have nothing to declare.

1. Boelaert K, Franklyn JA 2005 Thyroid hormone in health and disease. J Endocrinol 187:115
2. Davis JD, Tremont G 2007 Neuropsychiatric aspects of hypothyroidism and
treatment reversibility. Minerva Endocrinol 32:49 65
3. Gulseren S, Gulseren L, Hekimsoy Z, Cetinay P, Ozen C, Tokatlioglu B 2006
Depression, anxiety, health-related quality of life, and disability in patients
with overt and subclinical thyroid dysfunction. Arch Med Res 37:133139
4. Rockel M, Teuber J, Schmidt R, Kaumeier S, Hafner H, Usadel KH 1987
[Correlation of latent hyperthyroidism with psychological and somatic
changes]. Klin Wochenschr 65:264 273 (German)
5. Schlote B, Nowotny B, Schaaf L, Kleinbohl D, Schmidt R, Teuber J, Paschke
R, Vardarli I, Kaumeier S, Usadel KH 1992 Subclinical hyperthyroidism: physical and mental state of patients. Eur Arch Psychiatry Clin Neurosci 241:357
6. Bommer M, Eversmann T, Pickardt R, Leonhardt A, Naber D 1990 Psychopathological and neuropsychological symptoms in patients with subclinical
and remitted hyperthyroidism. Klin Wochenschr 68:552558
7. Sait Gonen M, Kisakol G, Savas Cilli A, Dikbas O, Gungor K, Inal A, Kaya A
2004 Assessment of anxiety in subclinical thyroid disorders. Endocr J 51:311
8. Botella-Carretero JI, Galan JM, Caballero C, Sancho J, Escobar-Morreale HF
2003 Quality of life and psychometric functionality in patients with differentiated thyroid carcinoma. Endocr Relat Cancer 10:601 610
9. Roberts LM, Pattison H, Roalfe A, Franklyn J, Wilson S, Hobbs FD, Parle JV
2006 Is subclinical thyroid dysfunction in the elderly associated with depression or cognitive dysfunction? Ann Intern Med 145:573581
10. Walsh JP, Ward LC, Burke V, Bhagat CI, Shiels L, Henley D, Gillett MJ,
Gilbert R, Tanner M, Stuckey BG 2006 Small changes in thyroxine dosage do
not produce measurable changes in hypothyroid symptoms, well-being, or
quality of life: results of a double-blind, randomized clinical trial. J Clin Endocrinol Metab 91:2624 2630
11. Eustatia-Rutten CFA, Corssmit EPM, Pereira AM, Frolich M, Bax JJ, Romijn
JA, Smit JWA 2006 Quality of life in longterm exogenous subclinical hyperthyroidism and the effects of restoration of euthyroidism, a randomized controlled trial. Clin Endocrinol (Oxf) 64:284 291
12. Samuels MH, Schuff KG, Carlson NE, Carello P, Janowsky JS 2007 Health
status, mood and cognition in experimentally-induced subclinical hypothyroidism. J Clin Endocrinol Metab 92:25452551
13. Samuels MH, Schuff KG, Carlson NE, Carello P, Janowsky JS 2007 Health
status, psychological symptoms, mood and cognition in L-thyroxine treated
hypothyroid subjects. Thyroid 17:249 258
14. Sinha AK, Pickard MR, Kim KD, Ahmed MT, al Yatama F, Evans IM, Elkins
RP 1994 Perturbation of thyroid hormone homeostasis in the adult and brain
function. Acta Med Austriaca 21:35 43
15. Gould E, Woolley CS, McEwen BS 1991 The hippocampal formation: morphological changes induced by thyroid, gonadal and adrenal hormones. Psychoneuroendocrinology 16:67 84
16. Madeira MD, Sousa N, Lima-Andrade M, Calheiros F, Cadete-Leite A, PaulaBarbosa MM 1992 Selective vulnerability of the hippocampal pyramidal neurons to hypothyroidism in male and female rats. J Comp Neurol 322:501508
17. Tejani-Butt SM, Yang J 1994 A time course of altered thyroid states on the
noradrenergic system in rat brain by quantitative autoradiography. Neuroendocrinology 59:235244

J Clin Endocrinol Metab, May 2008, 93(5):1730 1736

18. Hemmings SJ, Shuaib A 1998 Hypothyroidism-evoked shifts in hippocampal

adrenergic receptors: implications to ischemia-induced hippocampal damage.
Mol Cell Biochem 185:161169
19. Budson AE, Price BH 2005 Memory dysfunction. N Engl J Med 352:692 699
20. Biondi B, Palmieri EA, Fazio S, Cosco C, Nocera M, Sacca L, Filetti S, Lombardi G, Perticone F 2000 Endogenous subclinical hyperthyroidism affects
quality of life and cardiac morphology and function in young and middle-aged
patients. J Clin Endocrinol Metab 85:4701 4705
21. Bianchi GP, Zaccheroni V, Solaroli E, Verscini F, Cerutti R, Zoli M,
Marchesini G 2004 Health-related quality of life in patients with thyroid disorders. Qual Life Res 13:4554
22. Spreen O, Strauss EA 1998 A compendium of neuropsychological tests: administration, norms, and commentary. 2nd ed. New York: Oxford University
23. Wetzler S, Marlowe DB 1993 The diagnosis and assessment of depression,
mania, and psychosis by self-report. J Pers Assess 60:131
24. Klein I, Trzepacz PT, Roberts M, Levey GS 1988 Symptom rating scale for
assessing hyperthyroidism. Arch Intern Med 148:387390
25. Razvi S, McMillan CV, Weaver JU 2005 Instruments used in measuring symptoms, health status and quality of life in hypothyroidism: a systematic qualitative review. Clin Endocrinol (Oxf) 63:617 624
26. Lezak MD 1995 Neuropsychological assessment. 3rd ed. New York: Oxford
University Press
27. Cohen JD, Forman S, Braver TS, Casey BJ, Servan-Schreiber D, Noll DC 1993
Activation of prefrontal cortex in a nonspatial working memory task with
functional MRI. Hum Brain Map 1:293304
28. van Gorp WG, Altshuler L, Theberge DC, Mintz J 1999 Declarative and
procedural memory in bipolar disorder. Biol Psychiatry 46:525531
29. Shults J, Whitt MC, Kumanyika S 2004 Analysis of data with multiple sources
of correlation in the framework of generalized estimating equations. Stat Med
23:3209 3226
30. Danzi S, Klein I 2004 Thyroid hormone and the cardiovascular system. Minerva Endocrinol 29:139 150
31. Blissmer B, Riebe D, Dye G, Ruggiero L, Greene G, Caldwell M 2006 Healthrelated quality of life following a clinical weight loss intervention among overweight and obese adults: intervention and 24 month follow-up effects. Health
Qual Life Outcomes 4:43 48
32. Hahn S, Benson S, Elsenbruch S, Pleger K, Tan S, Mann K, Schedlowski M, van
Halteren WB, Kimmig R, Janssen OE 2006 Metformin treatment of polycystic
ovary syndrome improves health-related quality-of-life, emotional distress and
sexuality. Hum Reprod 21:19251934
33. Backonja M, Beydoun A, Edwards KR, Schwartz SL, Fonseca V, Hes M,
LaMoreaux L, Garofalo E 1998 Gabapentin for the symptomatic treatment of
painful neuropathy in patients with diabetes mellitus: a randomized controlled
trial. JAMA 280:18311836
34. Larisch R, Kley K, Nikolaus S, Sitte W, Franz M, Hautzel H, Tress W, Muller
HW 2004 Depression and anxiety in different thyroid function states. Horm
Metab Res 36:650 653
35. Carr D, McLeod DT, Parry G, Thornes HM 1988 Fine adjustment of thyroxine replacement dosage: comparison of the thyrotrophin releasing hormone
test using a sensitive thyrotrophin assay with measurement of free thyroid
hormones and clinical assessment. Clin Endocrinol (Oxf) 28:325333
36. Kathmann N, Kuisle U, Bommer M, Naber D, Muller OA, Engel RR 1994
Effects of elevated triiodothyronine on cognitive performance and mood in
healthy subjects. Neuropsychobiology 29:136 142
37. Harrison EL, Fillmore MT 2005 Social drinkers underestimate the additive
impairing effects of alcohol and visual degradation on behavioral functioning.
Psychopharmacology (Berl) 177:459 464
38. Bernal J 2007 Thyroid hormone receptors in brain development and function.
Nat Clin Pract Endocrinol Metab 3:249 259
39. Biondi B, Palmieri EA, Klain M, Schlumberger M. Biletti S, Lombardi G 2005
Subclinical hyperthyroidism: clinical features and treatment options. Eur J
Endocrinol 152:19