0021-972X/78/4705-1111S02.00/0 Journal of Clinical Endocrinology and Metabolism Copyright © 1978 by The Endocrine Society
Vol. 47, No. 5 Printed in U.S.A.
Low Serum Triiodothyronine and High Serum Reverse Triiodothyronine in Old Age: An Effect of Disease Not Age THOMAS OLSEN,* PETER LAURBERG, AND J0RGEN WEEKE Second University Clinic of Internal Medicine, Kommunehospitalet, 8000 Aarhus C, Denmark ABSTRACT. Serum concentrations of T4, T:t, rT.,, free T4, free T.t, and TSH were determined in four groups of adult subjects: group A, 27 healthy young volunteers (aged 18-29 yr); group B, 24 carefully selected healthy elderly subjects (aged 70-90 yr); group C, 41 subjects living at a municipal nursing home for the elderly (aged 70-90 yr); and group D, 35 hospitalized patients (aged 70-90 yr). Identical levels of iodothyronines in serum were found in the young and in the elderly healthy subjects. Moderate and severe disease induced alterations in io-
S
ERUM levels of thyroid hormones in elderly euthyroid subjects have been investigated by several authors. Generally, serum Ts has been found to be lower in elderly subjects than in young individuals (1-9). Furthermore, serum concentrations of rT.-? have been reported to be elevated in old age (9,10), while the serum concentrations of T4 have been found to be nearly unaffected by age. The reasons for these observations are unknown. Significant changes in circulating iodothyronines are known to occur in a variety of clinical situations, including acute and chronic disease (11). Previous investigators studying age-related variation in serum iodothyronine concentrations seem not to have selected completely healthy older subjects. Thus, it is unclear whether the observed alterations in serum levels of iodothyronines have been an effect of advanced age or of concomitant moderate chronic disease. In the present study, we have carefully selected groups of normal healthy subjects and found that the concentrations of T4, Ts, and rT,3 in serum are identical in young and in elderly subjects. Received February 22, 1978. * To whom requests for reprints should be addressed.
dothyronine concentrations in serum comparable to those reported earlier. Serum T:1 and free T.i levels were low and serum rT:l levels were high in groups C and D subjects; serum free T4 was elevated in group D, while serum T4 was low. Serum TSH was lower in the healthy elderly subjects than in the young subjects. Serum TSH was higher in the elderly sick subjects (groups C and D) than in the healthy subjects (group B). We conclude that advanced age per se is not accompanied by alterations in free or total serum iodothyronine levels. (J Clin Endocrinol Metab 47: 1111, 1978)
Subjects and Methods One hundred twenty-seven subjects participated in the study. None had a history of thyroid disease. All were euthyroid by clinical examination and none had goiter. Four groups of subjects were studied. The first group (A) comprised 27 young volunteers (18 men and 9 women) from the medical staff or medical students. Their age ranged from 18-29 yr (mean age, 24 yr). They were all healthy; no medication, including oral contraceptives, was taken. The second group (B) included 24 elderly subjects (5 men and 19 women), aged 70-90 yr (mean age, 75 yr). These subjects were living at home and were contacted at a social welfare center, where they met once or twice weekly to seek company. All felt healthy, took no medication, and had had no recent contact with a hospital. The third group (C) comprised 41 subjects (16 men and 25 women) in the age range 70-90 yr (mean age, 80 yr), living at a municipal nursing home for elderly people. None of these subjects was suffering from acute disease, but some were chronically ill and receiving various medications. The last group (D) included 35 hospitalized patients (19 men and 16 women). This group comprised all patients without thyroid diseases, aged 70-90 yr, who were available in the department at the particular day of investigation. Their mean age was 77 yr. The diagnoses were: coronary insufficiency (11), acute infectious disease (5), cerebral apoplexy (6), cerebral arteriosclerosis (4), malignant neoplasms (2), diabetes mellitus (2), chronic bronchitis (2), and senile
1111
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 12 November 2015. at 13:31 For personal use only. No other uses without permission. . All rights reserved.
OLSEN, LAURBERG, AND WEEKE
1112
.ICK&M • 1978 Vol47 • No 5
in groups A and B. Ninety-five percent confidence limits for the differences between mean values from groups A and B were (A minus B): T3, -18 and +3 ng/dl; free T3, -0.04 and +0.06 ng/dl; rT3, -1.8 and +6.2 ng/dl; T4, -0.085 and +1.3 jiig/dl; and free T4, -0.17 and +0.28 ng/dl. The mean serum TSH level was slightly lower in the old healthy subjects than in the young healthy subjects (P < 0.05). The group C subjects living at a nursing home had lower serum T3 and free T3 levels than the healthy elderly subjects (16% and 15%, respectively). The difference was statistically significant for T 3 but not for free T3. The mean serum rT 3 was 14% higher in group C than in group B, but this was not significant. This tendency towards a decrease in serum T 3 and an increase in serum rT 3 concentrations becomes significant when elderly hospitalized subjects are considered. The mean serum T 3 level was 32% lower in group D than in group B subjects (P < 0.001); mean serum free T 3 was 21% lower in group D subjects (P < 0.01). Results Serum rT 3 was 65% higher in group D than in Table 1 shows the mean levels of serum T4, group B subjects (P < 0.001). T3, rT3, free T4, free T3, and TSH in the four The variations in serum T4 and free T4 when groups. Individual values are plotted in Fig. 1. comparing group B with groups C and D subThe most impressive finding was the identical jects were more complex. The mean serum T4 levels of T 3 and rT 3 in groups A and B, the concentration was 16% higher (P < 0.05) in healthy young and the healthy old subjects. group C and 14% lower (P < 0.05) in group D There was not even a trend towards a decrease than in group B subjects. However, mean sein serum T 3 or an increase in serum rT 3 in the rum free T4 was 8% higher in group C (not group of old healthy subjects. This is further significant) and 40% higher in group D (P < illustrated by the identical rT 3 to T3 ratios in 0.001) than in group B subjects. Serum TSH groups A and B (Fig. 1). Also, the concentra- was significantly higher in groups C and D tions of T4 and free T4 in serum were similar compared to group B subjects (P < 0.05).
psychosis (3). None of the patients received a drug with a known effect on thyroid function. The numbers of men and women were different in different groups. It is not likely, however, that this had any major influence on the results, since no significant differences in the measured variables were observed between men and women in any of the four groups. All subjects had their usual breakfast and lunch. All were ambulating, except for a few patients in group C and most patients in group D who were confined to bed because of their disease. Blood samples were collected between 1100-1500 h, when the hormones are at a constant level (12). Serum was stored at —20 C until analyzed. Serum TSH, T.i, T4, and rT.-? concentrations were determined by single antibody RIA techniques (13-16). Free T 3 and free T4 were determined in the same dialysate of serum by RIAs (14, 15). Statistical differences between groups were analyzed by the Mann-Whitney U test. Ninety-five percent confidence limits were calculated for differences between mean T.i, free T-i, T4, free T4, and rT.i in groups A and B.
TABLE 1. Serum T4, T.t, rT.i, and TSH in two groups of healthy subjects of different ages and in two groups of institutionalized old subjects Free T, T, rT, Free T, T TSH (ng/dl) (ng/dl) (ng/dl) (ng/dl) (jug/dl) A 1.82 ± 0.44 117.2 ± 18.2 0.347 ± 0.084 24.9 ± 6.0 18-29 27 7.76 ± 1.13 1.61 ±0.92 B 70-90 7.17 ± 1.29" 1.76 ± 0.35" 123.7 ± 18.9" 0.335 ± 0.105" 22.6 ± 8.0" 1.11 ±0.99'' 24 41 70-90 8.34 ± 2 .18* 1.90 ± 0.47" 104.2 ± 18.9' 0.284 ± 0.102" 25.8 ± 12.0" 1.84 ± 1.64" C D 70-90 84.0 ± 26.7' 0.266 ± 0.075'' 37.2 ± 21.0' 1.81 ±0.82'' 35 6.18 ± 1.45" 2.47 ± 0.55r Groups A and B, normal healthy subjects; group C, subjects living at a nursing home for old people; group D, hospitalized subjects. Values given are mean ± SD. Statistical comparisons, group B was compared to group A, and groups C and D were compared to group B. "NS (P>0.05). * P < 0.05. r P< 0.001. Group
Age (yD
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 12 November 2015. at 13:31 For personal use only. No other uses without permission. . All rights reserved.
SERUM Ta AND rT 3 IN OLD AGE Ifjg/dD
lng/dl)
Tq
1113
15
150
75
10
100
50
50
25
A
B
C
D
oL
free
B
C
A
D
0.5
1,0
0.25
0,5
0
fl
a
B
C
D
rT3/T3
free
1
0L
A
= 128
Ing/dl)
L
D
FIG. I. Individual serum values of TV T.t, rT.i, free T\, and free T:(, and of the ratios between rT:t and Tn in four groups of subjects. A, Healthy young subjects; B, healthy old subjects; C, old subjects living at a nursing home; D, old subjects submitted to a medical department.
levels of rT 3 in elderly subjects. No age-correlated variations in serum rT 3 were observed The present study clearly demonstrates by Kaplan et al. (17). The only difference that when care is taken to select elderly sub- between young and old healthy subjects objects without debilitating diseases, their serum served in the present study was the slightly concentrations of both free and total thyroid lower mean serum TSH concentrations in the hormones as well as rT 3 do not differ from older subjects (group B). values found in young adults. Like many other investigators (11), we have This is in contrast to several earlier studies found that T 3 is low and rT 3 high in sick in which it was concluded that there is a patients. Free T4 was significantly elevated in decrease in serum T 3 with advancing age in hospitalized patients when they were comadult subjects (1-9). The influence of increas- pared to healthy elderly subjects. This is in ing age on serum rT 3 is less well studied. Nicod accordance with earlier findings in patients et al. (9) and Rudorff et al. (10) found high suffering from acute or chronic illness (18-23). Discussion
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 12 November 2015. at 13:31 For personal use only. No other uses without permission. . All rights reserved.
1114
OLSEN, LAURBERG, AND WEEKE
Total T4 in serum is dependent on the concentration of thyroid hormone-binding proteins. A decrease in the serum level of these binding proteins is probably the reason for the low levels of serum total T4 in group D despite a high level of free T4 (24, 25). The most likely explanation for the discrepancy between our results in elderly subjects and those of other investigators is the qualitative difference in the populations. The clinical data on the old subjects studied in previous investigations have been rather limited. For example, the subjects studied by Snyder and Utiger (1) were "ambulatory volunteers who had no history of thyroid disease, and who had no illness or were taking any medication known, or suspected to affect thyroid hormone economy," while those of Rubenstein et al. (3) were "ambulatory elderly patients, free of obvious debilitating disease." Our group of healthy elderly subjects consisted of individuals who felt well, had no recent contact with a hospital, and were without medication. Such a group of apparently healthy elderly subjects seems not to have been studied previously. The idea that the age-dependent variations in serum T4, T3, and rT 3 in adult subjects described earlier could be due to varying degrees of chronic disease in the elderly subjects studied is supported by recent findings of Burrows et al. (26). In groups of geriatric in-patients and geriatric out-patients, these authors found a low mean serum T3 in the in-patients, and a normal serum T 3 in the out-patients.
Acknowledgments The gifts of human TSH and anti-TSH serum from the Pituitary Agency of the NIH, Bethesda, MD, and of human TSH standard 68/38 from the National Institute of Medical Research, London, are gratefully acknowledged. The authors are indebted to Ketty Jensen, Anni Larsen, Karen Mathiassen, and Lene Grundtvig Sorensen for expert technical assistance.
References 1. SNYDER, P. J., AND R. D. UTIOER, Response to thyrotropin
releasing hormone (TRH) in normal man, J Clin Endocrinol Metab 34: 380, 1972. 2. BRUNELL, PH., AND C. BOHUON, Baisse de la triiodothyronine
serique avec l'age, Clin Chim Acta 42: 201, 1972.
-JCE & M Vol47
1978 No 5
3. RUBENSTEIN, H. A., V. P. BUTLER, AND S. C. WERNER,
Progressive decrease in serum triiodothyronine concentrations with human aging: radioimmunoassay following extraction of serum, J Clin Endocrinol Metab 37: 247, 1973. 4. HERMANN, J., H. J. RUSCHE, H. J. KROLL, P. HILGER, AND H.
L. KRUSKEMPER, Free triiodothyronine (TO- and thyroxine (T
Vol. 47, No. 5 Printed in U.S.A.
Low Serum Triiodothyronine and High Serum Reverse Triiodothyronine in Old Age: An Effect of Disease Not Age THOMAS OLSEN,* PETER LAURBERG, AND J0RGEN WEEKE Second University Clinic of Internal Medicine, Kommunehospitalet, 8000 Aarhus C, Denmark ABSTRACT. Serum concentrations of T4, T:t, rT.,, free T4, free T.t, and TSH were determined in four groups of adult subjects: group A, 27 healthy young volunteers (aged 18-29 yr); group B, 24 carefully selected healthy elderly subjects (aged 70-90 yr); group C, 41 subjects living at a municipal nursing home for the elderly (aged 70-90 yr); and group D, 35 hospitalized patients (aged 70-90 yr). Identical levels of iodothyronines in serum were found in the young and in the elderly healthy subjects. Moderate and severe disease induced alterations in io-
S
ERUM levels of thyroid hormones in elderly euthyroid subjects have been investigated by several authors. Generally, serum Ts has been found to be lower in elderly subjects than in young individuals (1-9). Furthermore, serum concentrations of rT.-? have been reported to be elevated in old age (9,10), while the serum concentrations of T4 have been found to be nearly unaffected by age. The reasons for these observations are unknown. Significant changes in circulating iodothyronines are known to occur in a variety of clinical situations, including acute and chronic disease (11). Previous investigators studying age-related variation in serum iodothyronine concentrations seem not to have selected completely healthy older subjects. Thus, it is unclear whether the observed alterations in serum levels of iodothyronines have been an effect of advanced age or of concomitant moderate chronic disease. In the present study, we have carefully selected groups of normal healthy subjects and found that the concentrations of T4, Ts, and rT,3 in serum are identical in young and in elderly subjects. Received February 22, 1978. * To whom requests for reprints should be addressed.
dothyronine concentrations in serum comparable to those reported earlier. Serum T:1 and free T.i levels were low and serum rT:l levels were high in groups C and D subjects; serum free T4 was elevated in group D, while serum T4 was low. Serum TSH was lower in the healthy elderly subjects than in the young subjects. Serum TSH was higher in the elderly sick subjects (groups C and D) than in the healthy subjects (group B). We conclude that advanced age per se is not accompanied by alterations in free or total serum iodothyronine levels. (J Clin Endocrinol Metab 47: 1111, 1978)
Subjects and Methods One hundred twenty-seven subjects participated in the study. None had a history of thyroid disease. All were euthyroid by clinical examination and none had goiter. Four groups of subjects were studied. The first group (A) comprised 27 young volunteers (18 men and 9 women) from the medical staff or medical students. Their age ranged from 18-29 yr (mean age, 24 yr). They were all healthy; no medication, including oral contraceptives, was taken. The second group (B) included 24 elderly subjects (5 men and 19 women), aged 70-90 yr (mean age, 75 yr). These subjects were living at home and were contacted at a social welfare center, where they met once or twice weekly to seek company. All felt healthy, took no medication, and had had no recent contact with a hospital. The third group (C) comprised 41 subjects (16 men and 25 women) in the age range 70-90 yr (mean age, 80 yr), living at a municipal nursing home for elderly people. None of these subjects was suffering from acute disease, but some were chronically ill and receiving various medications. The last group (D) included 35 hospitalized patients (19 men and 16 women). This group comprised all patients without thyroid diseases, aged 70-90 yr, who were available in the department at the particular day of investigation. Their mean age was 77 yr. The diagnoses were: coronary insufficiency (11), acute infectious disease (5), cerebral apoplexy (6), cerebral arteriosclerosis (4), malignant neoplasms (2), diabetes mellitus (2), chronic bronchitis (2), and senile
1111
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 12 November 2015. at 13:31 For personal use only. No other uses without permission. . All rights reserved.
OLSEN, LAURBERG, AND WEEKE
1112
.ICK&M • 1978 Vol47 • No 5
in groups A and B. Ninety-five percent confidence limits for the differences between mean values from groups A and B were (A minus B): T3, -18 and +3 ng/dl; free T3, -0.04 and +0.06 ng/dl; rT3, -1.8 and +6.2 ng/dl; T4, -0.085 and +1.3 jiig/dl; and free T4, -0.17 and +0.28 ng/dl. The mean serum TSH level was slightly lower in the old healthy subjects than in the young healthy subjects (P < 0.05). The group C subjects living at a nursing home had lower serum T3 and free T3 levels than the healthy elderly subjects (16% and 15%, respectively). The difference was statistically significant for T 3 but not for free T3. The mean serum rT 3 was 14% higher in group C than in group B, but this was not significant. This tendency towards a decrease in serum T 3 and an increase in serum rT 3 concentrations becomes significant when elderly hospitalized subjects are considered. The mean serum T 3 level was 32% lower in group D than in group B subjects (P < 0.001); mean serum free T 3 was 21% lower in group D subjects (P < 0.01). Results Serum rT 3 was 65% higher in group D than in Table 1 shows the mean levels of serum T4, group B subjects (P < 0.001). T3, rT3, free T4, free T3, and TSH in the four The variations in serum T4 and free T4 when groups. Individual values are plotted in Fig. 1. comparing group B with groups C and D subThe most impressive finding was the identical jects were more complex. The mean serum T4 levels of T 3 and rT 3 in groups A and B, the concentration was 16% higher (P < 0.05) in healthy young and the healthy old subjects. group C and 14% lower (P < 0.05) in group D There was not even a trend towards a decrease than in group B subjects. However, mean sein serum T 3 or an increase in serum rT 3 in the rum free T4 was 8% higher in group C (not group of old healthy subjects. This is further significant) and 40% higher in group D (P < illustrated by the identical rT 3 to T3 ratios in 0.001) than in group B subjects. Serum TSH groups A and B (Fig. 1). Also, the concentra- was significantly higher in groups C and D tions of T4 and free T4 in serum were similar compared to group B subjects (P < 0.05).
psychosis (3). None of the patients received a drug with a known effect on thyroid function. The numbers of men and women were different in different groups. It is not likely, however, that this had any major influence on the results, since no significant differences in the measured variables were observed between men and women in any of the four groups. All subjects had their usual breakfast and lunch. All were ambulating, except for a few patients in group C and most patients in group D who were confined to bed because of their disease. Blood samples were collected between 1100-1500 h, when the hormones are at a constant level (12). Serum was stored at —20 C until analyzed. Serum TSH, T.i, T4, and rT.-? concentrations were determined by single antibody RIA techniques (13-16). Free T 3 and free T4 were determined in the same dialysate of serum by RIAs (14, 15). Statistical differences between groups were analyzed by the Mann-Whitney U test. Ninety-five percent confidence limits were calculated for differences between mean T.i, free T-i, T4, free T4, and rT.i in groups A and B.
TABLE 1. Serum T4, T.t, rT.i, and TSH in two groups of healthy subjects of different ages and in two groups of institutionalized old subjects Free T, T, rT, Free T, T TSH (ng/dl) (ng/dl) (ng/dl) (ng/dl) (jug/dl) A 1.82 ± 0.44 117.2 ± 18.2 0.347 ± 0.084 24.9 ± 6.0 18-29 27 7.76 ± 1.13 1.61 ±0.92 B 70-90 7.17 ± 1.29" 1.76 ± 0.35" 123.7 ± 18.9" 0.335 ± 0.105" 22.6 ± 8.0" 1.11 ±0.99'' 24 41 70-90 8.34 ± 2 .18* 1.90 ± 0.47" 104.2 ± 18.9' 0.284 ± 0.102" 25.8 ± 12.0" 1.84 ± 1.64" C D 70-90 84.0 ± 26.7' 0.266 ± 0.075'' 37.2 ± 21.0' 1.81 ±0.82'' 35 6.18 ± 1.45" 2.47 ± 0.55r Groups A and B, normal healthy subjects; group C, subjects living at a nursing home for old people; group D, hospitalized subjects. Values given are mean ± SD. Statistical comparisons, group B was compared to group A, and groups C and D were compared to group B. "NS (P>0.05). * P < 0.05. r P< 0.001. Group
Age (yD
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 12 November 2015. at 13:31 For personal use only. No other uses without permission. . All rights reserved.
SERUM Ta AND rT 3 IN OLD AGE Ifjg/dD
lng/dl)
Tq
1113
15
150
75
10
100
50
50
25
A
B
C
D
oL
free
B
C
A
D
0.5
1,0
0.25
0,5
0
fl
a
B
C
D
rT3/T3
free
1
0L
A
= 128
Ing/dl)
L
D
FIG. I. Individual serum values of TV T.t, rT.i, free T\, and free T:(, and of the ratios between rT:t and Tn in four groups of subjects. A, Healthy young subjects; B, healthy old subjects; C, old subjects living at a nursing home; D, old subjects submitted to a medical department.
levels of rT 3 in elderly subjects. No age-correlated variations in serum rT 3 were observed The present study clearly demonstrates by Kaplan et al. (17). The only difference that when care is taken to select elderly sub- between young and old healthy subjects objects without debilitating diseases, their serum served in the present study was the slightly concentrations of both free and total thyroid lower mean serum TSH concentrations in the hormones as well as rT 3 do not differ from older subjects (group B). values found in young adults. Like many other investigators (11), we have This is in contrast to several earlier studies found that T 3 is low and rT 3 high in sick in which it was concluded that there is a patients. Free T4 was significantly elevated in decrease in serum T 3 with advancing age in hospitalized patients when they were comadult subjects (1-9). The influence of increas- pared to healthy elderly subjects. This is in ing age on serum rT 3 is less well studied. Nicod accordance with earlier findings in patients et al. (9) and Rudorff et al. (10) found high suffering from acute or chronic illness (18-23). Discussion
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 12 November 2015. at 13:31 For personal use only. No other uses without permission. . All rights reserved.
1114
OLSEN, LAURBERG, AND WEEKE
Total T4 in serum is dependent on the concentration of thyroid hormone-binding proteins. A decrease in the serum level of these binding proteins is probably the reason for the low levels of serum total T4 in group D despite a high level of free T4 (24, 25). The most likely explanation for the discrepancy between our results in elderly subjects and those of other investigators is the qualitative difference in the populations. The clinical data on the old subjects studied in previous investigations have been rather limited. For example, the subjects studied by Snyder and Utiger (1) were "ambulatory volunteers who had no history of thyroid disease, and who had no illness or were taking any medication known, or suspected to affect thyroid hormone economy," while those of Rubenstein et al. (3) were "ambulatory elderly patients, free of obvious debilitating disease." Our group of healthy elderly subjects consisted of individuals who felt well, had no recent contact with a hospital, and were without medication. Such a group of apparently healthy elderly subjects seems not to have been studied previously. The idea that the age-dependent variations in serum T4, T3, and rT 3 in adult subjects described earlier could be due to varying degrees of chronic disease in the elderly subjects studied is supported by recent findings of Burrows et al. (26). In groups of geriatric in-patients and geriatric out-patients, these authors found a low mean serum T3 in the in-patients, and a normal serum T 3 in the out-patients.
Acknowledgments The gifts of human TSH and anti-TSH serum from the Pituitary Agency of the NIH, Bethesda, MD, and of human TSH standard 68/38 from the National Institute of Medical Research, London, are gratefully acknowledged. The authors are indebted to Ketty Jensen, Anni Larsen, Karen Mathiassen, and Lene Grundtvig Sorensen for expert technical assistance.
References 1. SNYDER, P. J., AND R. D. UTIOER, Response to thyrotropin
releasing hormone (TRH) in normal man, J Clin Endocrinol Metab 34: 380, 1972. 2. BRUNELL, PH., AND C. BOHUON, Baisse de la triiodothyronine
serique avec l'age, Clin Chim Acta 42: 201, 1972.
-JCE & M Vol47
1978 No 5
3. RUBENSTEIN, H. A., V. P. BUTLER, AND S. C. WERNER,
Progressive decrease in serum triiodothyronine concentrations with human aging: radioimmunoassay following extraction of serum, J Clin Endocrinol Metab 37: 247, 1973. 4. HERMANN, J., H. J. RUSCHE, H. J. KROLL, P. HILGER, AND H.
L. KRUSKEMPER, Free triiodothyronine (TO- and thyroxine (T
