Triiodothyronine administration reduces serum growth hormone levels and growth hormone responses to thyrotropin-releasing hormone in patients with anorexia nervosa

Triiodothyronine administration reduces serum growth hormone levels and growth hormone responses to thyrotropin-releasing hormone in patients with anorexia nervosa

P~choneuroendocrinology, Vol. 15, No. 4, pp. 287-295, 1990 0306-4530/90 $3.00 + 0.00 ©1991 Pergamon Press plc Printed in Great Britain TRIIODOTHYRO...

770KB Sizes 1 Downloads 2 Views

P~choneuroendocrinology, Vol. 15, No. 4, pp. 287-295, 1990

0306-4530/90 $3.00 + 0.00 ©1991 Pergamon Press plc

Printed in Great Britain

TRIIODOTHYRONINE ADMINISTRATION REDUCES SERUM GROWTH HORMONE LEVELS A N D GROWTH HORMONE RESPONSES TO THYROTROPIN-RELEASING HORMONE IN PATIENTS WITH ANOREXIA NERVOSA ROBERTO VALCAVI, MICHELE ZINI a n d ITALO PORTIOLI 2a Divisione Medicina Generale e Sezione Endocrino Metabolica, Arcispedale S. Maria Nuova, Reggio Emilia, Italy (Received 30 November 1989; in final form 16 July 1990)

SUMMARY The aim of this study was to test the hypothesis that low serum T 3 concentrations may promote an abnormal growth hormone (GH) response to thyrotropin-releasing hormone (TRH) in patients with anorexia nervosa. Eight anorexic women and two anorexic men, ages 15-25 years, with low free T3 circulating levels (mean+SEM =2.8 + 0.3 pmol/1) were studied. A TRH test (200 ~tg IV) was carded out under basal conditions and repeated following treatment with oral T a (1.5 Ixg/kg BW/day) for eight days. Following T 3 administration, GH levels dropped significantly from a baseline of 7.1+ 1.3 I.tg/l to 3.1+ 0.7 I.tg/1 (p < 0.02), as did GH peak responses to TRH (9.0 + 1.0 gg/1 vs. 4.4 + 0.8 gg/1, p < 0.01). ANOVA and analysis of area under the curve (AUC) confirmed that after T 3 treatment there was a significant reduction in TRH-induced GH release in these patients (GH AUC: 902 + 132 Ixg/l vs. 456 _+ 91 I~g/l, p<0.02). TSH responses to TRH, which were normal prior to T3 treatment, completely disappeared following it, and PRL responses to TRH also were diminished. Although our experimental approach does not permit a conclusion that low T3 levels were the primary reason for these changes, the data support the theory that low T3 circulating levels may facilitate abnormal GH secretion and the GHreleasing activity of intravenous TRH.

INTRODUCTION ANOREXIA NERVOSA is associated with elevated fasting growth hormone (GH) levels (Marks et al., 1965) and a b e r r a n t G H r e s p o n s e s to IV t h y r o t r o p i n - r e l e a s i n g h o r m o n e ( T R H ) administration (Maeda et al., 1976; M a c a r o n et al., 1978; Casper & Frohman, 1982; Kiriike et al., 1987). It is clear that the elevated basal G H levels in anorexia nervosa are an effect o f malnutrition. Plasma G H levels are constantly high in conditions of chronic starvation (Marks et al., 1965) and protein calorie malnutrition (Smith et al., 1974) and return to normal during refeeding (Smith et al., 1974). Fasting leads to m a r k e d augmentation in G H secretion as the result o f an enhanced frequency of G H pulses ( T h o m e r et al., 1986). Several factors, such as low serum somatomedins, mild hypog l y c a e m i a , alterations in the concentration o f m e t a b o l i c fuels and stress, m a y contribute to the enhanced G H secretion during starvation. It is difficult to determine whether malnutrition is involved in G H release after T R H in anorexia Address correspondence and reprint requests to: Dr. Roberto Valcavi, 2a Divisione di Medicina Generale e Sezione Endoerino Metabolica, Areispedale S. Maria Nuova, vie. Umberto I ° 50, 42100 Reggio Emilia, ITALY. 287

288

R. VALCAVIet aL

nervosa, since only a few patients with raised fasting GH levels show a TRH-induced GH release. Furthermore, the G H - r e l e a s i n g activity of T R H has also b e e n observed following weight gain (Casper & Frohman, 1982). Several mechanisms m a y be the cause of the abnormal GH response to TRH, including sensitization of somatotropes to T R H as a consequence of excessive GH-releasing hormone (GHRH) secretion (Borges et al., 1983), defective somatostatin secretion, or direct stimulation by T R H on derepressed specific T R H receptors on somatotropes (Szabo et al., 1985). In the rat, the ability to release G H in response to TRH occurs during experimentally induced h y p o t h y r o i d i s m both in vitro (Szabo et al., 1984) and in vivo (Szabo et al., 1985), due to the expression of T R H receptors on somatotrope cells. Consequently, the activity of TRH as a GH secretagogue appears to be unmasked by low thyroid hormone concentrations. P a t i e n t s w i t h a n o r e x i a n e r v o s a r e g u l a r l y d e m o n s t r a t e l o w c i r c u l a t i n g T 3 s e r u m levels (Wartofsky & B u r m a n , 1982). We thus evaluated whether low serum T 3 concentrations might enhance the GH-releasing activity of TRH in anorexic patients. METHODS Eight female and two mate patients with anorexia nervosa, aged 15-25 years (mean +SEM = 18.2+ 1.0 years), were studied. Each met the criteria for anorexia nervosa proposed by Feighner et al. (1972) and had no clinical history of either bulimia or major depression. Patients' body mass index was 15.7+ 0.7 kg/m 2 (normal values: 19-25 kg/m2). Secondary amenorrhea was present in all women. None of the patients had other diseases, and no medication had been taken in the month before testing. The demographic characteristics of the patients are presented in Table I. The study was granted ethical approval by competent local health authorities. Informed consent was obtained from all patients. The patients were hospitalized for 7--20 days prior to study and during the study period. The tests were started between 0800h and 0830h, after the subjects fasted overnight, and with the subjects reeumbenr An IV infusion was commenced and 0,9% NaC1 was administered slowly in a forearm vein. Basal samples for free thyroid hormones, reverse T3, GH, PRL, TSH and somatomedin-C measurements were obtained at -30 and 0 min. Placebo or TRH (200 gg IV over 2 rain) was then given, and blood was taken at 15, 30, 45, 60, 90 and 120 min for GH, PRL and TSH assays. The same basal hormone measurements and tests were repeated following oral T3 administration (Ti-Tre®, Glaxo, Verona, Italy), in doses of 1.5 I.tg/kg BW/day for eight days. We chose this dose in order to obtain maximal effects of T3 at the pituitary level, as assessed by TSH measurernents. M~cation other than T3 was not administered. Throughout the study the patients' habitual life style was maintained, as were their usual eating habits, and no change in BW was observed. None of the patients took strenuous exercise prior to hospitalization; however they were not confined to bed, being allowed to walk inside the hospital park according to their previous habits. Ten normal, healthy control subjects, receiving no medication, and matched for sex and age, underwent the same serum hormonal measurements during the saline and TRH tests. Hormone assays Circulating hormones wexe measured by specific commercial radioimmunoassay s (RIA). Free and total thyroid hormones were assayed by solid-phase coated-tube RIAs (Diagnostic Products, Los Angeles, CA, USA). Sensitivities were 3.8 nmol/1, 0.13 pmol/l, 0A1 nmol/t, and 0.3 pmol/1 for total T4 (tT4), free T4 (fT4), total T3 (tT3), and free T 3 (fT3), respectively. For the same thyroid hormone measurements, intra- and inter-assay coefficients of variation (CV) were, respectively, 4% and 4.9%, 3.2% and 4.9%, 6.1% mad 8.2%, and 7.3% and 9.1% at dose levels of 110 nmol/1, 23 pmol/l, 1.2 nmol/1, and 3.8 pmol/1. TSH was measured by two-site monoclonal: antibody immunoradiometric assay (IRMA) and by magnetic bound]free separation (Coming, Kontron, Medfield, MD, USA). The within-assay CVs were 6.9% and 2,3% at dose levels of 1.4 mU/1 and 5.8 mU/l, respectively; the CVs between assays were 10.9% and 7.8% at the same dose levels. The sensitivity was 0.05 mU/t. GH was measured by liquid-phase, double-antibody RIA (Sorin Bioemedia, Saluggia, Vieenza, Italy), with a sensitivity of 0.2 gg/1. The intra- and inter-assay CVs were 7.7% and 11.0% at a dose level of 5:2 ~tg/1. PRL was measured by coated tubes solid phrase RIA (Diagnostic Products). The sensitivity was 2 p.g/1. The intra- and inter-assay CVs were 5.0% and 7.2% at a dose level of 23 gg/l. Somatomedin-C was measured by liquid-phase, double'antibody RIA (Nichols Institute D i ~ t i e s , San Juan Capistrano, CA, USA), with a sensitivity of 0.1 U/mt. The inlra'and inter-assay CVs were 5 . ~ and 11.2% at a dose level of 1.4 U/ml.

289

L o w T 3 AND G H RELEASE IN ANOREXIA NERVOSA

Statistical analysis Statistical analysis was performed by analysis of variance (A.NOVA)and paired t-tests. Student-Newman-Keuls procedure was used for multiple comparisons. Correlations were calculated as Pearson's r. All results are presented as the mean + SEM. RESULTS A l l patients had "normal T4" nonthyroidal illness according to the classification proposed by Engler & Burger (1984). Total and free thyroxine levels were n o r m a l (tT 4, 106 + 10.5 nmol/1; f T 4, 15.8+1.7 pmol/1; n o r m a l subjects: t T 4, 1 1 4 + 7 . 4 nmol/1; f T 4, 17.0+ 0.7 p m o l / l , NS), while total and free T 3 levels were low (tT 3, 0 . 9 6 + 0.06 nmol/1; f T 3, 2.8 + 0.3 pmol/1; normal subjects: t T 3, 2.4 + 0.3 nmol/1, p < 0.0001; f T 3, 5 . 2 + 0.4 pmol/1, p < 0 . 0 0 1 ) , reverse T 3 was e l e v a t e d (0.54 + 0.08 nmol/1 vs. 0 . 3 1 + 0.02 nmol/1, p < 0 . 0 0 1 ) , and T S H was n o r m a l ( 1 . 7 + 0.2 mU/1 vs. 1.9+ 0.2 mU/1, NS). F o l l o w i n g T 3 administration, the expected changes in thyroid hormone levels took place. Free T 3 levels rose to 6.1 + 0.4 pmol/1 (p < 0.01 vs. basal), free T 4 and reverse T 3 decreased (respectively 9 . 3 + 1 . 2 pmol/1, p < 0 . 0 1 , and 0 . 2 5 + 0.03 nmol/1, p < 0 . 0 0 1 vs. basal), and T S H levels were suppressed (0.4 + 0.1 mU/1, p < 0 . 0 0 1 vs. pretreatment). S e r u m s o m a t o m e d i n - C levels were below n o r m a l a n d w e r e n o t i n f l u e n c e d b y T 3 a d m i n i s t r a t i o n (0.64 + 0.16 vs. 0.86 + 0.16 U / m l , NS; normal subjects: 1.8 + 0.2 U/ml, p < 0.01).

TABLE I. DEMOGRAPHIC CHARACTERISTICS OF I 0 UNTREATED PATIENTS WITH ANOREXIA NERVOSA AND I 0 NORMAL CONTROL SUBJECTS Patient No. 1 2 3 4 5 6 7 8 9 10

Sex

Age (yrs)

Body mass index (kg/m ~)

F F M F F M F F F F

15 16 20 16 18 16 25 18 16 22

12.8 14.4 13.4 13.2 17.9 17.2 17.3 17.3 17.9 15.1

48 30 26 30 12 10 8 8 12 14

36 26

Mean SD SEM

18.2 3.2 1.0

15.7 2.1 0.7

19.8 13.2 4.2

Normal controls (n = 10): Mean 18.5 SD 3.7 SEM 1.2 p vs. anorexic patients

NS

Duration of: disease amenorrhea (months)

free T 3 levels (pmol/1)

Basal GH levels (~g/1)

TRH-induced GH AUC (~g/1/2 hr)

8 10 9 10

1.7 1.4 2.2 2.2 3.9 2.2 3.2 4.3 3.3 3.1

14.9 9.3 6.6 12.0 7.8 6.5 4.1 3.5 4.0 2.0

1778 1052 1104 1278 452 779 512 524 614 932

17.9 11.1 3.9

2.8 1.0 0.3

7.1 4.1 1.3

902 417 132

23.1 4.2~ 1.3

5.2 1.2 0.4

1.6 1.0 0.3

146 63 20

< 0.001

< 0.001

< 0.001

< 0.001

30 14

290

R. VALCAV1et al,

The mean basal GH level was raised before T 3 treatment (7.1 + 1.3 p.g/l; normal subjects: 1.6 _+. 0.3 ~tg/1, p<0.001). Following T 3 administration, GH fell to 3.1+ 0.7 ktg/1 (p<0.02 vs. pretreatment). ANOVA showed a significant TRH-induced GH release compared with the control experiment, when placebo was administered instead of TRH. In the TRH study, the mean GH level reached a peak of 9.0+ 1.0 Ixg/1 at 45 min, while in the placebo experiment, mean GH did no~ show any increase (Fig. 1). GH AUC was 902+132 ~tg/1/2 hr following TRH, significantb higher than in the control test (595 + 47 p.g/1 / 2 hr, p < 0.02). Correlations between clinical characteristics and GH alterations showed an inverse relationship between BMI and basal and TRH-induced GH secretion, while the duration of the disease and amenorrhea were positively correlated with GH changes. Basal free T 3 levels correlated directl) with BMI and inversely with the duration of the illness and amenorrhea (Table II). In addition, there was a significant inverse relationship between basal free T 3 and both basal (r = -0.656, p < 0.05) and TRH-stimulated (r = - 0.781, p < 0.01) GH levels. In the anorexic patients treated with T 3, TRH-induced GH release was significantly decreased, both in terms of peak plasma levels (4.4 + 0.8 Ixg/1 at 90 min, p < 0.01 vs. pretreatment) and AUC (456 + 91 p.g/1 / 2 hr, p < 0.02 vs. pretreatment) (Fig. 1). Furthermore, AUC analysis confirmed that T 3 produced a reduction in basal GH circulating levels, since in the control studies with placebo, GH AUC was significantly reduced by T 3 administration (294 + 21 g g / 1 / 2 hr, p<0.001 vs. pretreatment). However, even this latter AUC was greater than the GH AUC following TRH in the normal subjects (146 + 20 ktg/1 / 2 hr, p < 0.00l), thus indicating that T 3 treatment did not completely abolish GH secretion in the anorexic patients. The TSH responses to TRH were normal and completely abolished by T 3 treatment (Fig. 2). TRH 200 p.g IV 10-

* * * I

T

T

86" --1

:l: (5

4"

.

i

-30

I

0

!

15

I

I

I

I

30

45

60

90

Time

I

120

(minutes)

FIG. i: Effect of oral T 3 a d m i n i s t r a t i o n (I.5 l~g/kg BW daffy for e i ght days) on m e a n + SEM GH r e s p o n s e s to IV TRH (200 lag) in 10 p a t i e n t s w i t h a nore xi a nervosa. The s h a d e d a re a r e p r e s e n t s the m e a n GH r e s p o n s e + 2 SD to IV TRH (200 ~ i n 10 m a t c h e d , n o r m a l control subjects. TRH ( o - - e ) a n d placebo ( A - - 4 ) before T 3 a d m i n i s t r a t i o n . TRH {O--O) a n d placebo ( A - - A ) following T 3 a d m i n i s t r a t i o n . *p<0.05, **p<0.02, ***p<0.01 vs. T 3 therapy.

L O W T 3 A N D G H RELEASE IN ANOREXIA NERVOSA

291

TABLE II. CORRELATIONS B E T W E E N CLINICAL C H A R A C T E R I S T I C S A N D G H A N D F R E E T 3 A L T E R A T I O N S IN I 0 P A T I E N T S W I T H A N O R E X I A N E R V O S A Basal GH

TRH-induced

G H AUC

Basal free T s levels

r =-0.66 p < 0.05

r =-0.92 p < 0.001

r =+0.75 p < 0,02

Duration of disease

r =+0.87 p < 0.001

r =+0.95 p < 0.001

r =-0.76 p < 0.02

Duration of amenorrhea

r =+ 0.96 p < 0.001

r =+ 0.90 p < 0.01

r = - 0,82 p < 0.02

Body mass index

E

v

m

-30

m

0

I

i

n

I

15 30 45 60 Time

|

m

90

120

(minutes)

FIo. 2: M e a n + SEM TSH r e s p o n s e s to IV TRH (200 ~ in I 0 p a t i e n t s with anorexia n e r v o s a befor (O--O) a n d following ( O - - O ) oral T 3 a d m i n i s t r a t i o n (1.5 p g / k g BW daffy for eight days). ***/7<0,001 vs. T s t r e a t m e n t . Saline control t e s t s before (A--A) a n d after (Z~--A) T 3 t r e a t m e n t also are s h o w n . T h e s h a d e d area r e p r e s e n t s t h e m e a n TSH r e s p o n s e + 2 SD to IV TRH 2 0 0 fig in 10 m a t c h e d , n o r m a l control subjects.

292

R. VALCAVIet al. TRH 900 p.g lV

80-

"" 6O n

-

=1.

'~' o

40-

a.

20-

I

-30

I

0

I

15

I

I

a

I

I

30

45

60

90

120

Time (minutes) FIG. 3: Mean ± SEM PRL r e s p o n s e s to IV TRH (200 ~ in 10 patients wlth anorexla nervosa befor ( O - - O ) a n d following ( O - - O ) oral T 3 administration (1.5 ~ g / k g BW daily for eight days). *p< 0.05, **p< 0.02 vs. T 3 therapy. Saline control t e s t s before (A--A) a n d after (Z~--A) T s t r e a t m e n t also are shown. The s h a d e d area r e p r e s e n t s the m e a n PRL r e s p o n s e + 2 SD to IV TRH 200 Iag In 10 matched, normal control subjects.

Basal PRL levels were normal (mean basal value: 7.9+ 1.3 btg/1; normal subjects: 11.0+_ 1.0 Ixg/1, NS) and not influenced b y T 3 administration (8.5+2.4 lxg/1, NS vs. pretreatment). However, as shown in Fig. 3, serum PRL responses to TRH tended to decrease following T 3 administration, as indicated by AUC analysis (3943_+ 890 vs. 2420 + 434 lag/1 / 2 hr, p < 0.05), T 3 treatment was well toler~ed and without any noticeable side effects. Resting pulse rate increased from 58 + 8 to 825:5 BPM (p < 0.01), while blood pressure remained unchanged. DISCUSSION

In patients with anorexia nervosa we found the expected elevation in serum GH levels and an abnormal responsivity to IV TRH. The severity of GH changes was correlated with the severity el the disease, which in turn was associated with low circulating free T 3 levels. In those patients T 3 administration significantly reduced basal serum GH levels and GH responses to IV TRH. The physiologic mechanism of the abnuHnal GH response to TRH in anorexia nervosa is unknown at present. TRH receptors may constitute a normal phenotype of the somatotrope, and either hypothalamic GHRH excess and/or defective somatostatin secretion may reveal the activity of these receptors at the level of the anterior pituitary gland (Thomer et al., 1986). TRH receptors are thyroid hormone-sensitive and are consequently increased in hypothyroidism. This might explain the paradoxical GH response to TRH in hypothyroidism (Collu et al., 1977), d~pite the well-described reduced GH response to standard secretagogues, including GHRH, in this illness (Valcavi et al., 1986).

LOW T 3 AND G H RELEASE IN ANOREXIA NERVOSA

293

Low circulating T 3 levels in anorexia nervosa probably represent the peripheral tissue adaptation to the altered nutritional status. When anorexia nervosa patients are given a sub=replacement dose of oral T 3, a greater reduction in their ankle reflex relaxation time than in normal controls has been observed (Croxson & Ibbertson, 1977), indicating that true thyroid hormone deficiency occurs.

Many pathological conditions associated with abnormal GH responses to TRH, such as anorexia nervosa, primary hypothyroidism, insulin-dependent diabetes, liver cirrhosis and chronic renal failure (Dieguez et al., 1988), are also recognized as "low T3" states (Wartofsky & Burman, 1982). Our finding that the "nonspecific" GH response to TRH was reduced by T 3 administration in anorexic patients, all of whom showed low T 3 levels, supports the possibility that thyroid hormone deficiency may facilitate the GH-releasing activity of TRH. T 3 may act directly on the pituitary, modulating the numbers and affinity of TRH receptors on the surface of somatotrope cells (Szabo et al., 1984). Therefore, the administration of T 3 in anorexia nervosa patients may decrease the activity of TRH receptors and reduce the TRH-induced GH release. However, hypothalamic mechanisms are also involved, since thyroid hormone deficiency at the pituitary level could not explain the elevated basal GH levels; nor was GH secretion completely abolished by T 3 treatment. There is clear evidence that abnormalities in the hypothalamic control of GH secretion occur in anorexia nervosa. GH responses to insulin-induced hypoglycemia are reduced (Marks et al., 1965; Nakagawa et al., 1985), whereas arginine- and GHRH-induced GH release are intact (Nakagawa et al., 1985; Casanueva et al., 1987) or even excessive (Mtiller et al., 1987). Furthermore, abnormal GH responsiveness to glucose (Gold et al., 1980) and GnRH (Tamai et al., 1986) has been documented. Compromised somatomedin synthesis may reduce the inhibitory effects of the feedback action of somatomedins on GH release and raise plasma GH basal levels (Tanaka et al., 1985). In our patients, the administration of T 3 did not lead to any change in serum somatomedin levels; nevertheless, a significant reduction in basal GH levels occurred following T 3 treatment, indicating that the inhibitory effects of T 3 on GH are not mediated through somatomedin feedback. A decrease in the hypothalamic content of somatostatin has been found in hypothyroid rats (Berelowitz et al., 1980), and it has been reported that T 3 stimulates the acute release of somatostatin from the perfused rat hypothalamus (Berelowitz et al., 1980). However, the latter finding has not been confirmed (Femandez-Durango et al., 1978; Martin et al., 1985). Although only tentative, one may speculate that T 3 administration to anorexic patients produces an enhancement in endogenous somatostatinergic tone and, in turn, a decrease in plasma basal GH levels and in the GH response to TRH. It should be noted, however, that the association between low T 3 and aberrant GH responsivity does not always occur. In major affective disorders, unusual GH responses to TRH may be present (Dieguez et al., 1988) in the face of normal serum T 3 levels. Therefore, thyroid alterations in anorexia nervosa might not play a primary pathogenetic role, and the possible effect of T 3 administration could be secondary to the influences exerted by the thyroid hormones on central somatostatin activity or other neurotransmitter pathways, whose changes might be the reason for the altered GH secretion. Basal levels of serum PRL were normal and there was a decrease in PRL responses to TRH following T 3 administration, in the anorexic patients. Thyroid hormones may influence PRL neuroregulation by increasing dopamine utilization in the median eminence of the hypothalamus (Andersson & Eneroth, 1985) and by direct inhibitory effects at the anterior pituitary level (Vale et al., 1973). In summary, in patients with anorexia nervosa and "low T3" nonthyroidal illness, T 3 administration produced a significant reduction in basal serum GH levels and in the GH response to TRH. Although we cannot conclude that low T 3 levels were the primary reason for these changes, our

294

R. VALCAVIet al.

data support the hypothesis that low T 3 circulating levels may facilitate an abnormal G H secretkm and the GH-releasing activity of IV T R H in these patients. REFERENCES Andersson K, Eneroth P (1985) The effects of acute and chronic treatment with triiodothyronine and thyroxine on the hypothalamic and telencephalic catecholamine nerve terminal systems of the hypophysectomized male rat. Chronic treatment modulates catecholamine utilization in discrete eatecholamine nerve terminal systems. Neuroendocrinology 40: 398-408. Berelowitz M, Maeda K, Harris S, Frohman LA (1980) The effect of alterations in the pituitary-thyroid axis on hypothalamic content and in vitro release of somatostatin-like immunoreactivity. Endocrinology 107: 24-29. Borges JLC, Uskavitch DR, Kaiser DL, Cronin MJ, Evans WS, Thorner MO (1983) Human pancreatic growth hormone-releasing factor-40 (hpGRF-40) allows stimulation of GH release by TRH. Endocrinology 113: 1519-1521. Casanueva FF, Borras CG, Burguera B, Lima L, Mutuals C, Tresguerres JAF, Devesa J (1987) Growth hormone and prolactin secretion after growth hormone-releasing hormone administration, in anorexia nervosa patients, normal controls and tamoxifen pretreated volunteers. Clin Endocrino127: 517-523. Casper RC, Frohman LA (1982) Delayed TSH release in anorexia nervosa following injection of thyrotropinreleasing hormone (TRH). Psychoneuroendocrinology 7: 59-68. Collu R, Leboeuf G, Letarte J, Ducharme JR (1977) Increase in plasma growth hormone levels following thyrotropin-releasing hormone injection in children with primary hypothyroidism. J Clin Endocrinol Metab 44: 743 -747. Croxson MS, Ibbertson HK (1977) Low serum triiodothyronine (T3) and hypothyroidism in anorexia nervosa. J Clin Endocrinol Metab 44: 167-174. Dieguez C, Page MD, Scanlon MF (1988) Growth hormone neuroregulation and its alterations in disease states. Clin Endocrino128: 109-143. Engler D, Burger AG (1984) The deiodination of the iodothyronines and of their derivatives in man. Endocr Rev 5: 151-184. Feighner JP, Robins E, Guze SB, Woodruff RA Jr, Winokur G, Munoz R (1972) Diagnostic criteria for use in psychiatric research. Arch Gen Psychiatry 26: 57-63. Fernandez-Durango R, Arimura A, Fishbak J, Schally AV (1978) Hypothalamic somatostatin and LHRH afte~ hypophysectomy in hyper- and hypothyroidism, and during anesthesia in rats. Proc Soc Exp Biol Med 157: 235 -240. Gold MS, Pottash AL, Sweeney DR, Martin DM, Davies RK (1980) Further evidence of hypothalamic-pituitary dysfunction in anorexia nervosa. Am J Psychiatry 137: 101-102. Kiriike N, Nishiwaki S, Izumiya Y, Maeda Y, Kawaldta Y (1987) Thyrot~opin, prolactin, and growth hormone responses to thyrotropin releasing hormone in anorexia nervosa and bulimia. Biol Psychiatry 22: 167-176~ Macaron C, Wilber JF, Green O, Freinkel N (1978) Studies of growth hormone (GH), thyrotropin (TSH) and prolactin (PRL) secretion in anorexia nervosa. Psychoneuroendocrinology 3: 181-185. Maeda K, Kato Y, Yamaguchi N, Chihara K, Ohgo S, Iwasaki Y, Yoshimoto Y, Moridera K, Kuromaru S, Imura H (1976) Growth hormone release following thyrotropin-releasing hormone injection into patients with anorexia nervosa. Acta Endocrino181: 1-8. Marks V, Howorth N, Greenwood FC (1965) Plasma growth hormone levels in chronic starvation in man. Nature 208: 686-687. Martin D, Epelbaum J, Bluet-Payot MT, Prelot M, Kordon C, Durand D (1985) Thyroidectomy abolishes pulsatile GH secretion without affecting hypothalamic somatostatin. Neuroendocrinology 41: 476-481. Miiller EE, Cavagnini E Panerai AE, Massironi R, Ferrari E, Brambilla F (1987) Neuroendocrine measures in anorexia nervosa: comparisons with primary affective disorders. In: Nerozzi D, Goodwin FK, Costa E (Eds) Hypothalamic Dysfunction in Neuropsychiatric Disorders. Raven Press, New York, pp 261-271. Nakagawa K, Matsubara M, Obara T,: Kubo M, Akikawa K (1985) Responses of pituitary and adrenal medulla to insulin-inducedhypoglycaemia in patients with anorexia nervosa. EndocrinolJapon 32: 719-724. Smith SR, Edgar PJ, Pozefsky T, Chhetri MK, Prout TE (1974) Growth hormone in adults with protein-calorie malnutrition. J Clin Endocrinol Metab 39: 53-62. Szabo M, Stachura ME, Paleologos N, Bybee DE, Frohman LA (1984) Thyrotropin-releasing hormone stimulates growth hormone release from the anterior pituitary of hypothyroid rats in vitro. Endocrinology 114: 13441351. Szabo M, Ruestow PC, Kramer DE (1985) Growth hormone response to thyrotropin-releasing hormone in the urethane-anesthetized rat: effect of thyroid status. Endocrinology 117: 330-337.

Low T3 ANDGH RELEASEIN ANOREXIANERVOSA

295

Tamai H, Kobayashi N, Fukata S, Hirota Y, Matsubayashi S, Nakagawa T, Okimura MC, Walter RM Jr, Kumagai LF (1986) Paradoxical responses of plasma cortisol, adrenoeorticotropic hormone and growth hormone to thyrotropin-releasing hormone and luteinizing hormone-releasing hormone in anorexia nervosa patients. Psychother Psychosom 46: 147-151. Tanaka T, Maesaka H, Suwa S (1985) Changes in somatomedin activity in anorexia nervosa. EndocrinolJapon 32: 891-897. Thorner MO, Vance ML, Evans WS, Blizzard RM, Rogol AD, Ho K, Loong DA, Borges JLC, Cronin MJ, MacLeod RM, Kovacs K, Asa S, Horvath E, Frohman L, Furlanetto R, Klingensmith GJ, Brook C, Smith P, Reichlin S, Rivier J, Vale W (1986) Physiological and clinical studies of GRF and GH. Rec Progr Horm Res 42: 589-640. Valcavi R, Jordan V, Dieguez C, John R, Manicardi E, Portioli I, Scanlon MF (1986) Growth hormone responses to GRF 1-29 in patients with primary hypothyroidism before and during replacement therapy with thyroxine. Clin Endocrino124: 693-698. Vale W, Blackwell R, Grant G, Guillemin R (1973) TRF and thyroid hormones on prolactin secretion by raI anterior pituitary ceils in vitro. Endocrinology93: 26-33. Wartofsky L, Burman KD (1982) Alterations in thyroid function in patients with systemic illness: the "euthyroid sick syndrome". EndocrRev 3: 164-217.