Clinical Endocrinology (1978) 9,337-349.

T H E RELATIONSHIP BETWEEN CIRCADIAN V A R I A T I O N S IN CIRCULATING THYROTROPHIN, T H Y R O I D HORMONES A N D PROLACTIN V I V I A N CHAN, A N N J O N E S , P. LIENDO-CH, A. MCNEILLY, J. LANDON A N D G. M . BESSER

The Departments of Endocrinology and Chemical Pathology of St. Bartholomew’s Hospital, London ECIA 7BE (Received 25 November 1977; revised 4 April 1978; accepted 10 April 1978)

SUMMARY

Half-hourly blood samples were taken from six clinically euthyroid men over a continuous period of 24 h. Their concentrations of total thyroxine (T4), total triiodothyronine (T3), thyrotrophin (TSH) and prolactin (PRL) were assessed together with the degree of unsaturation of thyroid hormone binding proteins as determined by the thyroid hormone uptake test (THUT). Both T3 and T4 were also measured in urine samples collected serially during the same 24 h period. Significant circadian changes in serum TSH, THUT, serum and urine T4 and serum PRL were demonstrated in all subjects. TSH showed a reciprocal pattern t o serum T4, with higher levels during the evening and at night than the daytime. This TSH pattern did not coincide with PRL secretion. Further studies on the same subjects did not show any significant effect of posture, corticosteroid or T4 administration upon circadian changes in TSH. There appeared t o be no consistent circadian changes in serum or urinary T3. It seems likely that the TSH circadian rhythm is centrally determined and that free T 3 levels are maintained more or less constant by variation in peripheral conversion from T4. Although the circadian pattern of hormones of the hypothalamic-pituitary-adrenal axis are well-recognized, there is still controversy regarding the existence of such a pattern for the hormones of the hypothalamic-pituitary-thyroid axis. Several investigations have failed to demonstrate any rhythmic change in TSH levels in man (Utiger, 1965; Ode11 et al., 1967; Hershman & Pittman, 1971;Webster et al., 1972),while other groups who have reported circadian variations do not agree as t o the time of day at which TSH concentrations are at their maximum (Nicoloff et al., 1970; Pate1 et al., 1972; Vanhaelst et al., 1972; Weeke, 1973). Conflicting results have also been reported regarding the circadian changes in protein-bound iodine (PBI), total serum T4 and free T4 (FT4) (Lemarchand-Beraud & Vanotti, 1969; De Costre et al., 1971). Correspondence:Prof. G. M. Besser, St Bartholomew’s Hospital, London. 03004664/78/10004337$02.00 01978 Blackwell Scientific Publications

337

V. Chan et al. The lack of sensitivity in the hormone assays may account for these discrepant findings. With the development of highly sensitive and precise radioimmunoassays (RIA), a more extensive study was carried out t o investigate the circadian variations in serum TSH, T3, T 4 and PRL secretions, the T3 and T4 urinary pattern, and the effect of posture, corticosteroid and T 4 administration upon these circadian patterns. S U B J E C T S , M A T E R I A L S AND METHODS Six healthy male volunteers, aged 25-35, were studied. They were not taking any drugs. 5 ml of blood were obtained every half-hour via an indwelling forearm venous cannula over a 24 h period, together with serial 2 hourly urine collections during the day and 4 hourly collections at night. The subjects carried out their normal daily activities, went t o sleep around midnight, and were rarely disturbed by the withdrawal of blood. In addition, two of the same subjects were bled as above from 09.00-20.00 h and then were recumbent (without falling asleep) from 12.00-18.00 h following an otherwise identical protocol. Another subject from the original group was given 0.5 mg of dexamethasone every 6 h starting at 09.00 h on the morning preceding this study. He was then bled at intervals from 18.00 t o 09.00 h (the following morning). A different subject was given 200 ug of T4 by mouth daily for 48 h and he went through the same protocol again. In all blood samples the serum was separated within 2 h and stored frozen until required for assay. When urine collections were made, individual volumes and time of collections were recorded. To avoid interassay variability, all samples from a single subject were run in one assay. Total T4 concentrations in 20 pl of whole serum were determined by RIA, using 8-anilino-1-naphthalene sulphonic acid (200 pg) t o inhibit T 4 binding t o endogenous thyroxine binding globulin (TBG) and charcoal t o separate antibody bound and free hormone fractions. The anti-T4 serum was raised in rabbits by multiple intra-dermal injections of T4bovine serum albumin conjugate. Total T 3 concentrations were measured in 50 p1 of whole serum by a similar RIA procedure (Chan et al., 1975), using a goat anti-T3 serum which showed 0.0125% cross-reactivity with T4. THUT (thyroid hormone uptake test) value was determined using the Thyopac-3 Kit (Radiochemical Centre, Amersham). As the radioactivity in the supernate is measured, values fall below 90% in hyperthyroidism and rise above 120% in hypothyroidism, compared with a control serum. Circulating TSH was measured by the double antibody method of Hall and his colleagues (1971), and the results expressed as mU/1 of the MRC standard 68/38. The antiserum was raised in rabbits against human TSH and was highly specific, showing no cross-reactivity with human chorionic gonadotrophin (HCG) and, therefore, no HCG was added t o the incubation mixtures. Circulating PRL was determined by RIA (McNeilly & Hagen, 1974). Urinary T 3 and T 4 excretions were measured as described elsewhere (Chan & Landon, 1972; Chan et al., 1975; Chan, 1974) and the results expressed as ng of hormone excreted per min. Urinary creatinine was measured by a standard method using the autoanalyser (Technicon handbook number N-1 Ib). RESULTS The sensitivity and within-assay precision of the assays used are outlined in Table 1. The coefficients of variation were calculated from ten analyses of three serum samples with hormone levels in the low, normal and high range respectively.

Circadian changes in TSH, T3, T4 and prolactin

339

Table 1. Characteristics of the hormone assays The within-assay coefficient of variation (%) for 3 different samples Sample volume Test Serum TSH (RIA) T4 (RIA) T3 (RIA) PRL (RIA) THUT

(PO

100 20 50

so

100

Detection limit

Low

Normal

High

0.1 mU/1 6.4 nM/1 0.04 nM/1 0.5 Pg/i -

8.7 8.9 7.5 7.2 2.2

5.0 7 .O 2.5 7 O. 2 .o

7.4 8.1 4.5 8.3 1.6

25 ng/dl 25 ng/dl

8.6 8 .O

4.6

8.6 7.5

Urine T4 (CPBA*) T 3 (RIA)

1000 100

~

3.5

~

* Competitive protein binding assay.

Absolute mean day and night circulating hormone concentrations in all subjects are presented in Table 2. Daytime represents the period between 07.30 and 24.00 h and nighttime is the interval between 00.00 and 07.30 h. Fig. 1 shows the mean changes in serum TSH, T4, T3 and THUT in all subjects expressed in the manner of Orth & Island (1969); each value is represented as per cent deviation from the subject’s own mean hormone concentration during that particular study day and the mean values for the six subjects are shown. The representative sem. is shown on the graphs, and the ranges of the SEM for all points on the graphs are presented in Table 3. Changes in circulating TSH Zevefs. A well-defined TSH circadian pattern was evident in all subjects, with lower levels during the daytime, but significantly increased values at night. The increase began approximately 4 h prior t o the onset of sleep, which was at around midnight. In addition, episodic TSH fluctuations were observed throughout the 24 h. One subject, in particular, had exaggerated fluctuations and a more marked circadian variation. He showed a mean TSH concentration for that day of 5mU/1, with maximum and minimum concentrations of 20 and 3.4 mU/1 respectively. Further investigations showed that this healthy and clinicalIy euthyroid volunteer had normal serum total T3 and T4 concentrations, and no microsomal or thyroglobulin antibodies as determined by immunofluorescence techniques (Dr W. J. Irvine). Changes in circulating thyroid hormones levels. Serum total T4 concentrations also showed a distinct circadian pattern, with minimum concentrations between midnight and 08.00 h. Table 2 shows, in all subjects, statistically significant differences between these concentrations and those recorded during the daytime. Total serum T3 concentrations did not show such a definite circadian pattern although non-significantly decreased concentrations were found between 02.00 and 08.00 h. Changes in urinaiy T4 and T3 excretions. The mean of all subjects are shown in Fig. 2. It can be seen that the urinary T4 excretion was characterized by a rise in the morning and a fall at night, but remained constant during the waking hours. The timing varied slightly between individuals, with peak values between 08.00 and 12.00 h and nadir values in the

h. $ Night = 00.00-07.30. 8 = P < 0.001. II = P < 0.005.

0.161

0.22** 105 t 0.32

100 f 0.43

106 f 0.31 103 t 0.41 88 f 0.41 103 f 0.46

Dayt

f

0.828

101 f 0.453

96

99 f 0.70** 97 f 0.585 81 f 0.475 96 f 0.545

Night $

THUT (%)

f

1.54

106.19 f 1.93

91.78

116.49 f 2.45 104.00 i: 1.67 81.74 f 1.80 97.18 f 1.93

Dayt

Night $

f

1.678

88.69 f 2.325

84.82

96.41 f 1.675 84.05 f 1.165 68.73 f 2.961 84.95 f 1.935

T4 (nM/l) Night$

2.09 2 0.035 1.79 f 0.05** 1.80 t 0.028 2.04 t 0.10 NS 1.39 2 0.03 1.40 t 0.17 NS 2.44 f 0.03 2.31 2 0.041

2.54 t 0.03 1.94 f 0.03 2.00 2 0.03 2.01 f 0.03

DaYt

T3 W d l )

1 = P < 0.01.

tt = P < 0.02. NS = Not significant,P > 0.05.

** = P < 0.05.

Dayt

Night$

10.47 t 0.36

12.70 f 0.24

12.33 f 0.68tt

15.51 f 0.995

9.31 t 0.22 14.02t 1.118 9.30 f 0.43 13.16 f 0.823 5.05 f 1.68 6.69 f 1.4811 13.10 f 0.30 15.36 f 0.71 II

* Probability values for comparison between mean day and night hormone levels using student r-test (2 tailed).

f

t Day = 07.30 -24.00

f

0.13 2.92

2.27

6

f

2.65

5

1.73 f 0.101 2.78 f 0.225 1.49 f 0.06tt 8.84 f 1.101

Night$

0.17 3.39

1.55 f 0.07 1.79 f 0.09 1.28 5 0.05 5.47 f 0.40

1 2 3 4

f

Dayt

Subjects

TSH (mU/l)

Table 2. Mean day and night circulating hormone concentrations f SEM in all six subjects studied*

w

k

-+

(D

x

A

9

P 0

Circadian changes in TSH, T3, T4 and prolactin 8o

r

TSH

Fig. 1. Mean circadun changes in serum TSH, T4, T3 and THUT in all six subjects, presented as the percent deviation from the subjects' own mean hormone levels for that particular day. - - - individual mean hormone level; (&) mean + SEM for the variable.

Table 3. Absolute ranges of SEM for aU points on Fig. 1

Hormone TSH T4 T3 THUT

SEM range 18.09 - 1.89 6.03 - 0.60 5.21 - 0.21 1.53 - 0.21

Mean ~tSD 7.05 f 3.03 f 2.43 k 0.74 t

3.82 1.21 1.14 0.30

34 1

V. Chan et al.

342 Urine T3

Urine T4 a,

I

07.00

11.00

15.00 19.00 23.00 Time of day (h)

03.00

07.00

Fig, 2. Mean urinary T3 and T4 excretion in six subjects over a 24 h period (shaded areas indicate urinary hormone excretion during night-time).

second half of the night (between 04.00 and 08.00 h}. In contrast, urinary T3 excretion remained constant throughout the 24 h period, that is within t 0.1 ng/min of the mean. Urinary creatinine excretion did not show any circadian fluctuation. The mean excretion was 6.42 0.54 nM/min. Changes in THUT value. As shown in Fig. 1, it is evident that the degree of unsaturation of the thyroid hormone binding proteins, as reflected by the THUT value, shows a welldefined circadian pattern which is almost identical t o the one found for serum T4. _+

4 1h

I

I

I

I

I

04.00

08.00

12.00

16.00

20.00

24.00

Time of day (h)

Fig. 3. Circadian changes in concentration of serum PRL (solid symbols) and TSH (open symbols) in one subject shown as absolute values.

343

Circadian changes in TSH, T3, T4 and prolactin 80

-

40

-

0-

-a, 40-

E E

=

80

I,

1

I

04.00

08.00

I

I

I

12.00

16.00

20.00

I

m

E E

24.00

Time of day (h)

Fig. 4. Circadian changes in serum PRL concentrations in remaining five subjects (the values are expressed as percentage deviation from each subject’s own mean daily concentration).

Changes in circulating PRL. Fig. 3 shows the relationship between the circadian changes in circulating PRL and TSH concentrations in one representative subject, although all six subjects studied showed the same relationship. The mean day and night serum PRL concentrations are given in Table 2 . The PRL patterns were characterized by episodic fluctuations, with marked increases at night. The period of maximal PRL secretion varied in different individuals, either from midnight t o 04.00 or 04.00 to 07.00 h (Fig. 4). It always occurred after the TSH peak. Effect of posture. Effects of changes in posture were studied in two subjects. There was no significant variation in the TSH circadian pattern in either subject, whether in a vertical or horizontal position (Fig. 5). The circulating T4 and THUT circadian patterns also remained unchanged (see Appendix).

V. Chan et al.

344

Vertical

~

i

~~

10.00

12.00

14.00

16.00

18.00

20.00

Time of day (h) Fig. 5. Effect of posture on circadian changes in serum TSH concentrations in one subject (the shaded area indicates the period when the subject was recumbent; the lower graph shows the TSH circadian pattern, in the Same subject, during a normal day).

Effect of corticosteroid. The circadian variation in circulating endogenous corticosteroids was abolished by dexamethasone, t o non-cycling levels of less than 55 nmol/l. Although the mean concentrations of serum TSH, T3, and PRL were significantly lowered compared with predexamethasone treatment (P< 0.05, < 0.005 and < 0.05 respectively), their circadian patterns remained very much the same (Fig. 6). Effect of 7'4administration. Fig. 7 shows the circadian TSH pattern in one subject either basally or after T4 administration. No significant change was seen although the mean TSH concentration was lowered from 2.9 t o 0.9 mU/1. DISCUSSION Using highly specific, sensitive and precise RIA, it has been possible t o measure with confidence, small changes in the concentrations of circulating hormones. Thus, we have been able t o measure, throughout a continuous period of 24 h , simultaneous variations in the serum concentration of TSH, T3 and T4, making it easier to interpret the possible role of each of these hormones in the feedback mechanisms controlling the hypothalamic-pituitarythyroid axis. In addition, since only small volumes of serum are required in the measurement

Circadian changes in TSH, T3, T4 and prolactin

345

c

8 0

- 40

,...\.,

..'

THUT

t1,02- ,t . a , -' C L - + 8 - * - d \ +

-&

-5

'0-0'

-40!iO0 20100 22100 24100 02100 04f00 06100 OS!OO Time of day (h)

Fig. 6. Circadian changes in TSH, T4, T3. THUT and PRL in one subject given dexamethasone (0.5 mg 6 hourly) starting 9 h before beginning sampling at 18.00 h.

of THUT and PRL, we were able t o include them in this study, keeping the total amount of individual blood loss down t o a reasonable minimum of approximately 240 ml over a period of 24 h , which represents about 4.9 f 0.5% of the estimated total blood volume of 4.99 k 0.55 1 (Gray & Frank, 1953). This loss would be insufficient t o account for the circadian changes found, since differences between maximal and minimal hormone concentrations of up t o 140% were observed. This study clearly confirms the existence of circadian patterns for serum TSH, PRL, T 4 and THUT whilst there is no evidence of any rhythmic variations in the serum concentration of total T3. The circadian pattern of serum TSH is in agreement with the findings of Pate1 and his colleagues (1972) and Weeke (1973) who reported higher TSH levels at night in all their subjects, while Vanhaelst and coworkers (1972) found higher

V. Chan et al.

346 Sleep I

'

-801 00.00

I

I

04.00

08.00

I 12.00

I 16.00

I

I

20.00

24.00

Time of day (h)

Fig. 7. Circadian changes in serum TSH concentrations, before and after thyroid suppression by T4 administration (200 rg/day for 48 h ) in one subject.

nocturnal TSH levels in women, but not in men. Present observations showed that the increase in TSH occurred approximately 4 h prior t o the onset of sleep, and maximum concentrations were attained between 23.00 and 02.00 h. Weeke (1973) found higher TSH levels at night, between 20.00 and 08.00 h, but was able t o detect a high TSH peak in only one subject (between 07.00 and 09.00 h). A11 our subjects showed a well-defined circadian pattern for serum total T4 and THUT. These patterns were similar to each other, but reciprocal to TSH. Lemarchand-Beraud and Vanotti (1 969) showed the same inverse relationship between TSH and free T4 patterns. We have confirmed the findings of Robyn and coworkers (Robyn et al., 1973; Vanhaelst et al., 1973) that TSH and PRL release are not concomitant and may well be unrelated. Changes in posture during daytime, from the vertical to horizontal position, have no apparent effect on the circadian variations in any of the thyroid hormones, TSH or PRL. In contrast, DeCostre and his colleagues (1971) found that a reversed circadian pattern of serum total T4 was produced as soon as the patient's sleepwake cycle was reversed. Alterations in hepatic clearance or haemoconcentration are not likely t o explain the circadian variations in TSH, since the changes in plasma proteins, haemoglobin and haematocrit are such that the lowest values occur at night (Renbourn, 1947), whereas TSH is at its maximum at this time. Nicoloff et al. (1970) demonstrated a circadian pattern of thyroidal iodine release with maximum levels around 04.00 h and postulated the presence of a negative feedback regulation of TSH by circulating corticosteroids and this concept received some support from the work of Patel et al. (1974), who reported abolition of the circadian rhythm of TSH by pharmacological amounts of cortisol (200 mg/24 h). However, it is at variance with our finding that the abolition of the circadian rhythm of endogenous plasma corticosteroids with 2 mg per day of dexamethasone for 24 h did not alter the circadian pattern of TSH. It seemed possible that the primary event was a fall in T4 due t o altered peripheral clearance of the hormone and that the TSH changes were secondary to this. We have attempted t o exclude this by showing that administration of exogenous T4, maintaining the lowered serum T4 concentrations constant, did not abolish the TSH circadian rhythm. Similarly posture was not responsible, either through central TSH or peripheral T4 effects

Circadian changes in TSH, T3, T4 and prokctin

347

since acute changes in posture during daytime, have no apparent effect on the circadian variations observed. Finally, in our studies the TSH circadian pattern was not dependent on the corticosteroid circadian rhythm, as abolition of this with dexamethasone did not alter the changes in TSH levels. The practical clinical consequences of these circadian studies is that the collection of blood for PRL, TSH and oth& thyroid hormone estimations should be performed at a particular time of day (e.g. between 09.00 and noon) when hormone concentrations have attained the mean basal level. It would be wise to bear in mind the possible circadian changes of hormone concentrations when interpreting results of any in vivo test performed over a period of several hours. ACKNOWLEDGEMENTS

We wish t o thank Dr W. J. Irvine, Royal Infirmary, Edinburgh, for estimating thyroid antibodies, Dr J. s. Glover and Dr G. Smith of the Radiochemical Centre, Amersham, for generous supply of materials.

REFERENCES CHAN, V., BESSER, G.M., LANDON, J. & EKINS, R.P. (1972) Urinary triiodothyronine excretion as index of thyroid function. Lancet, 2,253-256. CHAN, V. (1974) The assay of urinary thyroid hormones for assessing thyroid function. Annals Clinical Biochemistry, 11, 120-129. CHAN, V. & LANDON, J. (1972) Urinary thyroxine excretion as index of thyroid function. Lance?, 1, 4-6. CHAN, V., MERRETT, T., LANDON, J., LINDEN, A-M. & JOUSTRA, J. (1975) A simple solid-phase radioimmunoassay for triiodothyronine. Annals Clinical Biochemistry, 12, 173-1 75. DE COSTRE, P., BUHLER, U., DEGROOT, LJ. & REFETOFF, S. (1971) Diurnal rhythm in total serum thyroxine levels. Metabolism, 20,782-791. GRAY, S.J. & FRANK, H. (1953) Simultaneous determination of red-cell mass and plasma volume in man with radioactive sodium chromate and chromic chloride. Journal of Clinical Investiyaiion, 32, 1000-1004. HALL, R., AMOS, J.& ORMSTON, B.J. (1971) Radioimmunoassay of human serum thyrotrophin. British Medical Journal, 1,582-585. HERSHMAN, J.M. & PITTMAN, J.A. Jr. (1971) Utility of the radioimmunoassay of serum thyrotropin in man. Annals o f InternalMedicine, 74,481-490. LEMARCHAND-BERAUD, T.H. & VANOTTI, A. (1969) Relationship between blood thyrotrophin level, protein bound iodine and free thyroxine concentration in man under normal physiological conditions. Acta Endocrinologica, 60,315-326. McNEILLY, A.S. & HAGEN, C. (1974) Prolactin, TSH, LH and FSH responses t o a combined LHRH/ TRH test at different stages of the menstrual cycle. Clinical Endocrinology, 3,427-435. NICOLOFF, J.T., FISHER, D.A. & APPLEMAN, M.D. Jr. (1970) The role of glucocorticoids in the regulation of thyroid function in man. Journal of Clinical Investigation, 49, 1922-1929. ODELL, W.D., WILBER, J.F. & UTIGER, R.D. (1967) Studies of thyrotropin physiology by means of radioimmunoassay. Recent Progress in Hormone Research, 23,47-85. ORTH, D. & ISLAND, D.P. (1969) Light synchronisation of the circadian rhythm in plasma cortisol (17-OHCS) concentration in man. Journalof Clinical Endocrinology & Metabolism, 29,479-486. PATEL, Y.C., ALFORD, F.P. & BURGER, H.G. (1972) The 24-hOUr plasma thyrotrophin profile. Clinical Science, 43,71-77. PATEL, Y.C., BAKER, H.W.G., BURGER, H.G., JOHNS, M.W. & LEDINEK, J.E. (1974) Suppression of the thyrotrophin circadian rhythm by glucocorticoids. Journal of Endocrinology, 62,421-422.

348

V. Chan et al.

RENBOURN, E.T. (1947) Variation, diural and over longer periods of time, in blood haemoglobin, haematocrit, plasma protein, erythrocyte sedimentation rate, and blood chloride. Journal ofHygiene (Camb), 4 5 , 4 5 5 4 6 7 . ROBYN, C., DELVOYE, P., NOKIN, J., VEKEMANS, M., BADAWI, M., PEREZ-LOPEZ, F.P. & L’HERMITE, M. (1973) Prolactin and human reproduction. International Symposium on Human Prolactin (Ed. by J.L. Pasteels & C. Robyn), pp. 98-119. Excerpta Medica, Amsterdam. UTIGER, R.D. (1965) Radioimmunoassay of human plasma thyrotrophin. Journal of Clinical Investigation, 44, 1277-1286. VANHAELST, L, VAN CAUTER, E., DEGAUTE, J.P. & GOLDSTEIN, J. (1972) Circadian variations of serum thyrotrophin levels in man. Journal of Clinical Endocrinology & Metabolism, 3 5 , 4 7 9 4 8 2 . VANHAELST, L, GOLDSTEIN, J., VAN CAUTER, E., L’HERMITE, M. & ROBYN, C. (1973) Etude simultanee des variations circadiennes des taux sanguins de la thyrotropine (TSH) et de la prolactine hypophysaires chez lhomme. C R . A c a d e m y of Science (Paris), 276, 1875-1 877. WEBSTER, B.R., GUANSING, A.R. & PAICE, J.C. (1972) Absence of diurnal variation of Serum TSH. Journal of Clinical Endocrinology & Metabolism, 34,899-901. WEEKE, J. (1973) Circadian variations of the serum thyrotrophin level in normal subjects. Scandinavian Journal of Clinical Laboratory Investigation, 31,337-342.

See Appendix on page 349,

I1

I .6 1.4 99.1 132.6 2.69 2.23 103 99 9.5 2.4

10.30

1.4 1.1 101.7 132.6 2.69 2.00 103 101 11.0 1.8

10 pm

1.1 1.4 101.7 130.0 2.78 2.07 106 102 11.6 2.2

Time

TSH hU/C T4 (nM/l) T3 (nM/I) THUT (%) PRL Wl)

1 1.4 0.8 124.9 105.5 2.49 1.95 I04 98 9.8 2.4

12.30 pm 1.2 0.9 108.1 122.3 2.40 1.98 106 99 8.0 3.1

12 1.4 1.2 121.0 128.7 2.61 2.09 105 103 9.7 3.2

11.30

1.4 1.2 95.3 132.6 2.61 2.04 102 102 8.4 1.6

APPENDIX

1.1 0.9 105.5 122.3 1.97 2.01 98 100 11.8 3.9

1.30

2 1.1 1.2 108.1 119.7 2.40 1.89 98 97 10.2 4.3

1.0 0.8 95.3 108.1 2.29 1.92 98 101 10.8 3.7

2.30 1.1 0.8 108.1 121.0 2.14 1.83 98 101 8.8 2.0

3 1.4 0.8 105.5 106.8 2.14 1.90 102 102 10.0 1.9

3.30

Subjects recumbent

1.3 1.3 99.1 108.1 2.23 1.84 105 100 8.9 1.2

4

0.6 86.2 104.3 2.14 1.98 102 100 8.8 1.8

1.3

4.30

Effects of posture on circulating hormone levels in two subjects

5 1.2 0.8 124.9 109.4 2.03 1.83 101 99 9.5 9.6

I .o 0.7 112.0 100.4 2.29 1.75 92 100 7.5 7.3

0.8 117.1 101.7 2.14 1.86 106 103 7.7 4.9

6 1.5

5.30

1.4 0.9 117.1 101.7 1.97 1.61 104 100 6.9 5.1

6.30

7 1.5 0.9 128.7 104.3 2.32 1.81 107 102 8.7 7.9

1.6 0.7 123.6 101.7 2.32 1.86 103 99 7.1 12.0

7.30

8 1.6 0.8 126.1 101.7 2.15 1.72 103 97 7.5 6.6

1.4 0.7 109.4 101.7 2.23 1.84 104 100 7.4 6.3

8.30