Abnormal growth hormone and cortisol, but not thyroid-stimulating hormone, responses to an intravenous glucose tolerance test in normal-weight, bulimic women

Psychoncurocndocrinology, Vol. 17, No. 6, pp. 639-645, 1992

0306-4530/92 $5.00-+0.00 ©1993 Pergamon Press Lad.

Printed in Great Britain

ABNORMAL GROWTH HORMONE AND CORTISOL, BUT NOT THYROID-STIMULATING HORMONE, RESPONSES TO AN INTRAVENOUS GLUCOSE TOLERANCE TEST IN NORMAL-WEIGHT, BULIMIC WOMEN V. COIRO, 1 R. VOLPI, 1 C. MARCHESI, 2 L. CAPRETI'I,3 G. SPERONI, 3 G. ROSSI,4 G. CAFFARRI,4 A . DE FERRI,5 A . MARCATO, 6 a n d P. CHIODERA 1 1Institute of Internal Medicine and 2Institute of Psychiatry, School of Medicine, University of Parma, Parma; 3Division of Internal Medicine, Hospital of Codogno, Codogno; 4Laboratory of Clinical Biochemistry, Hospital of Guastalla, Guastalla; 5Psychiatric Clinic "Villa Maria Luigia", Monticelli; and 6Division of Internal Medicine, Hospital of Fiorenzuola, Fiorenzuola, Italy (Received 30 May 1991; in final form 11 February 1992)

SUMMARY Abnormal growth hormone (OH) and adrenocorticotropic hormone (ACTH)/cortisol secretory patterns in response to a glucose load have been observed in underweight anorectic women. The present study was performed in an attempt to establish whether changes in the hypothalamic/pituitary sensitivity to hyperglycemia occur in bulimia in the absence of weight disturbance. Therefore, serum GH, plasma cortisol, and plasma insulin concentrations were measured in eight women with normal weight bulimia and in eight normal women during an intravenous glucose (0.33 g/kg as an Iv bolus) tolerance test (IGTT). In addition, since abnormal pituitary hormone responses to a glucose load might reflect alterations in somatostatin (SRIH) release, TSH secretion also was measured, in view of its sensitivity to SRIH inhibition. Both GH and cortisol levels progressively and significantly declined during IGTT in the normal subjects. In the bulimic women, cortisol levels remained unchanged, whereas GH concentrations rose significantly after glucose injection. Plasma cortisol and serum GH levels were significantly higher in the bulimic than in the control subjects. No significant differences between groups were observed in hyperglycemia-induced insulin increments or in TSH decrements. These data indicate that an altered sensitivity to hyperglycemia affects the hypothalamic/pituitary centers controlling the secretion of the counterregulatory hormones GH and ACTH/cortisol in bulimia nervosa. The lack of a simultaneous change in the TSH secretory pattern argues against a possible involvement of SRIH in the pathophysiology of this disorder.

INTRODUCTION ANOREXIA AND BULIMIA NERVOSA often have been considered together, because various clinical and p s y c h o n e u r o e n d o c r i n o l o g i c a l correlates o f these disorders overlap (Russel, 1978; C a s p e r et al., 1980; H u d s o n et al., 1983). In some patients, both syndromes m a y occur at different times (Gwirstman et al., 1983; Norris et al., 1985). Both psychiatric s y n d r o m e s are characterized by Address correspondence and reprint requests to: Dr. Vittorio Coiro, Cattedra di Clinica Medica Generale, Universit~ di Parma, Via Gramsci 14, 1-43100 Parma, ITALY. 639

640

V. COIROet al.

discontentedness with b o d y weight and appearance and p r e o c c u p a t i o n with regulation of caloric intake; therefore, both anorectic and bulimic subjects engage in abnormal patterns of eating. However, anorexia is characterized by unremitting restriction of food intake and weight loss, whereas bulimia shows alternating episodes of ingestion of conspicuous amounts of food and periods of starvation. These behaviors may reflect metabolic abnormalities at the hypothalamic level. There are close functional interactions between the hypothalarnic nuclei regulating food intake and those controlling the pituitary secretion o f the counterregulatory hormones, GH and ACTH/cortisol (Morley et al., 1985; Muller et al., 1987). These hypothalamic nuclei are sensitive to changes in blood glucose concentrations (Himsworth et al., 1972; Karteszi et al., 1982). In normal conditions, hypoglycemia stimulates, whereas hyperglycemia inhibits, food intake and secretion of the "diabetogenic" hormones, GH and ACYH/cortisol (Roth et al., 1964; Glick, 1970; JesovaRepeekowa et al., 1980; Morley & Levine, 1983; Tamai et al., 1991). An abnormal hypothalamic/pituitary sensitivity to glucose has been described in anorectic women, who showed low GH and ACTH/cortisol responses to hypoglycemia (Nakagawa et al., 1985; Hotta et al., 1986; Newman & Halmi, 1988). On the other hand, cortisol levels did not change, whereas GH concentrations showed a paradoxical increase in patients with anorexia nervosa during an intravenous glucose tolerance test (IGTT) (Tamai et al., 1991). In previous studies, we did not find significant alterations in the GH response to insulininduced h y p o g l y c e m i a in b u l i m i c w o m e n (Coiro et al., 1990). In the present study, we examined whether the GH and/or cortisol decremental responses to an IGTI" change in bulimic women. In addition, in order to test the hypothesis that abnormal h o r m o n a l responses to a glucose load reflect alterations in hypothalamic somatostatin (SRIH) release (Lengyel et al., 1984; Shibasaki et al., 1989), TSH secretion during the IGTT was evaluated, since SRIH is "known to have inhibitory effects on TSH release in man (Scanlon et al., 1979). SUBJECTS AND METHODS Eight normal women and eight women who met DSM-III criteria (American Psychiatric Association, 1980) for bulimia participated after giving informed consent. The duration of the bulimic illness ranged from 2-8 yr, with a mean (4-SEM) length of 5.3+0.7 yr. The bulimic women were tested 2 wk after admission to the hospital. All presented a history of at least 2 (range 2-4) binge-vomit episodes per week and of abuse of laxatives and/or diuretics. No other drugs had been'taken during the last year before the study. The women were matched for age (bulimics: 28.7 + 1.4 yr (mean +SEM); controls: 30.6 + 1.7 yr) and weight (bulimics: 55.4+1.3 kg; mean body mass index (BMI)=23; controls: 55.8+ 1.7 kg, mean BMI=23). There was no clinical or laboratory evidence of hepatic, renal, or other disease in any subject. The normal and bulimic women were not affected by concurrent major affective disorder, as determined with the Schedule of Affective Disorders and Schizophrenia (SADS) ('Endicott & Spitzer, 1978), and had no history of anorexia nervosa. None took any drug, including diuretics or laxatives, for at least 2 wk before each experimental day. All subjects had a history of regular mensmaal cycles of normal duration (26-30 days). Basal body temperature and serum levels of ovarian steroids were evaluated daily and were used to determine the precise phase of the cycle. All women were studied in the follicular phase (days 6-8) of a normal menstrual cycle. Body weight was measured daily for at least 7 days before the tests (day -7, 55.3-+ 1.3 kg; day -6, 55.2+ 1.3 kg; day -5, 55.3+1.3 kg; day -4, 55.34-1.3 kg; day -3, 55.44-1.3 kg; day -2, 55.44-1.3 kg; day -1, 55.44-1.3 kg; test day, 55.4 + 1.3 kg). Subjects showing changes in body weight were excluded from the study. In the 3 days before each test, all subjects were given a standard diet of 150 g carbohydrate/day. Plasma 13hydroxybutyric acid (13HBA)levels were measured on the experimental day as an index of the metabolic status of the patients, to rule out pre-test carbohydrate privation, despite assertion to the contrary by the patients. The bulimic women were induced to refrain from binge eating and vomiting for 4 days prior to testing. To ensure compliance with abstinence, the patients were controlled with the assistance of the hospital staff. Subjects who

GH, Corrnsot. ANDTSH R~sPONSF.STo GLUCOSELNBULIMIA

641

were unable to meet this protocol were excluded from the study. At 0800h on the experimental day, two intravenous catheters were inserted into antecubital veins of the subjects, while subjects were fasting and resting from the previous evening. One catheter was used for glucose injection, and the other served for blood sampling and was kept patent with a slow infusion of normal saline (NaC1 0.9%). At 0830h, glucose (0.33 g/kg) was injected as an IV bolus within 2 rain after withdrawal of a basal blood sample (time 0). Further blood samples were taken at 2, 5, 10, 30 and 60 rain after glucose infusion. Blood glucose levels were immediately measured, whereas serum and plasma samples were stored at -20 ° C until they were analyzed for GH, cortisol, insulin and TSH. Blood glucose concentrations were measured in all specimens with a glucose oxidase-peroxidase method, on an IL 918 autoanalyzer (Instrumentation Laboratory, Milan, Italy). Plasma I~HBA levels were measured in the basal samples, according to the method of Williamson and Mellonby (1974). In order to exclude the possibility that hypothetical major fluid shifts after withdrawal from laxatives and diuretics at the time of the patient's admission to hospital (i.e., 2 wk before the tests) might have influenced the endocrine testing, serum sodium concentrations and plasma osmolality were measured in samples taken 7 days before tests (at 0800h during a routine laboratory test) from the bulimic women, and at time 0 of the IGTT in both the bulimic and the normal subjects. Serum sodium concentrations were measured by flame photometry, and osmolality was measured with an Advanced Osmometer (Osmette Sedas s.r.l.). Radioassays were used to measure GH (Schalch & Parker, 1964), cortisol (Brock et al., 1958), and insulin (Yalow & Berson, 1960). Serum TSH (Rattle et al., 1984) concentrations were measured with a sensitive immunoradiometric method. All hormones were evaluated with commercial kits. In all radioimmunoassays, all samples from each subjects were analyzed in duplicate in the same assay. The sensitivity of the assays were 0.5 ng/ml for GH, 0.1 I.tg/dl for cortisol, 0.02 I.tU/ml for TSH, and 2 ~tU/ml for insulin. The intraassay and interassay coefficients of variation, respectively, were 3.6% and 8.0% for GH, 3.7% and 7.5% for cortisol, 4.8% and 6.7% for TSH, and 6.3% and 7.9% for insulin. The results were analyzed by analysis of variance (ANOVA) with repeated measures, Kruskall-WaUis test, Wilcoxon's matched-pairs rank-sum test, and cor50 relation coefficients as appropriate. Values are expressed as mean + SEM. 40

RESULTS

30

rr

20 10 I I

I

I !

I

I

I

I

,o

2'0

30

I

I

I

go

go

240 O~

¢/) O O

L~

200 160 120 80 I

I

,0

Minutes FIG. 1: Immunoreaetlve Insulin (l~I) and glucose levels after N injection of glucose (0.33 g/lr~l at time 0 In normal (solid line) and bulimic (dashed line) women. Each point represents the mean_+ SEM of the observaUons.

Basal blood glucose and plasma insulin c o n c e n t r a t i o n s were s i m i l a r in the control and bulimic women (by Kruskall-Wallis test). In the n o r m a l c o n t r o l s , p l a s m a osm o l a l i t y (285.4 + 0 . 7 m O s r n / k g ) and serum sodium levels (140.4+0.4 mEq/1) were similar to those o f the bulimic women, both 7 days before the IGTT (286.0+ 0.8 m O s m / k g ; 1 4 1 . 0 + 0 . 5 m E q / 1 ) a n d at t i m e 0 ( 2 8 5 . 8 + 0.7 m O s m / k g ; 140.8+0.4 mEq/1). P l a s m a 13HBA levels were slightly, but not significantly, higher in the bulimic women ( 0 . 1 6 + 0 . 0 5 ~tmol/ml) than in the n o r m a l c o n t r o l s ( 0 . 1 0 + 0 . 0 2 I.tmol/ml). G l u c o s e and insulin levels f o l l o w e d a similar pattern in all subjects after glucose loading (bulimic vs. controls: p =NS, b y ANOVA) (Fig. 1). Basal serum TSH, serum GH, and p l a s m a c o r t i s o l levels were s i m i l a r in the

642

V. COtRO et al.

control and bulimic subjects (Fig. 2). TSH, GH, and cortisol declined slowly after glucose injection in the normal controls (TSH: p<0.05 vs. baseline at 20, 30, 40, 50, and 60 min; GH: p<0.05 vs. baseline at 20, 30, 40, 50, and 60 min; cortisol: p < 0 . 0 5 at 20 and 30 min, and p < 0 . 0 2 at 40 and 60 min, vs. baseline, by Wilcoxon's matched-pairs ranksum test) (Fig. 2). In the bulimic women, TSH levels followed a similar pattern as in the normal controls (p<0.05 vs. baseline at 20, 30, 40, 50, and 60 min by Wilcoxon's matched-pairs rank-sum test; p =NS vs. the control group by ANOVA), whereas cortisol concentrations remained constant (p =NS vs. baseline) and GH levels rose slowly but significantly ( p < 0 . 0 5 at 20 min, and p < 0.01 at 30, 40, 50, and 60 rain, vs. baseline). During the IGTT, both cortisol (F= 6.52, p<0.05) and GH (F=8.715, p < 0 . 0 1 ) were significantly higher in the bulimic than in the control women (Fig. 2). When the amount of laxatives or diuretics, considered as weekly doses, taken by each bulimic woman before withdrawal was correlated with the highest GH and/or cortisot values during the IGTT, no significant relationship was found. No side effects were observed in any of the subjects during the IGTT.

03

1 ~ I

~ t~ E v -r ~

I

I

7654 3 2 1 0

l

I

1

T

T

T

I

I

I

/ ........

.......

.......

,

.o** ~..°"

I I

I

I I

I

I

I

1

|

I

I

50

60

15 ~

10

-6

.-~ t: oO

5

0 O25

I

I

I

I

10

20

30

40

l

I

Minutes FIG. 2: Cortisol, GH, a n d TSH l e v e l s a f t e r rv i n j e c t i o n of g l u c o s e (0.33 g / k g ) a t t i m e 0 in n o r m a l (solid line) a n d b u l i m i c ( d a s h e d line) w o m e n . E a c h p o i n t r e p r e s e n t s t h e mean_+ SEM of the o b s e r v a t i o n s .

DISCUSSION In agreement with Blouin et al. (1991), we found normal insulin responses after intravenous glucose administration in normal-weight bulimic women. However, in contrast to our results, Blouin et al. (1991) observed lower peak blood glucose levels in bulimic than in normal subjects. When glucose has been given orally, normal or markedly elevated blood glucose levels have been found in bulimic subjects (Mitchell & Bantle, 1983; Weingarten et al., 1988). The reason for these different findings is unknown. It might depend on the selection of patients without regard to their past history of anorexia nervosa, or on different carbohydrate diets and/or length of the fasting period before testing (Blouin et al., 1991). Previous studies have reported discrepant findings on the e r e c t of glucose administration on GH secretion in normal-weight bulimic women. GH failed to suppress normally after glucose administration in some subjects (Mitchell & Bantle, 1983), whereas it showed a normal suppression in others (Casper et al., 1988). However, in these studies, glucose was given orally, and thus the discrepancies in GH secretory pattern might be explained by differences among subjects in gastrointestinal function, such as gastric erupting time, gut motility, and/or

GH, CORTtSOL,~'~t)TSH RESPONSESTO GLUCOSE E~ Bt~'L[MtA

643

stimulation of neurohormonal mechanisms induced by oral glucose ingestion (Blouin et al., 1991). Our present study is the first report of the effects of an intravenous glucose administration on GH and cortisol secretion in bulimia nervosa. The results provide evidence of abnormal GH and cortisol responses to hyperglycemia in normal-weight bulimic patients. In contrast to the normal controls, the bulimic women showed significant serum GH rises and lack of change in plasma cortisol concentrations during the IGTr. These alterations might represent trait markers of the bulimic eating disorder, because they cannot be attributed to the possible interference of other nonspecific factors. In fact, our bulimic women did not fulfill criteria for major depression, as ascertained by the SADS. Furthermore, they were neither underweight nor malnourished, and they showed normal insulin responses to intravenous glucose administration. These findings indicate that the abnormal GH and cortisol responses to glucose were independent of weight disturbance or alterations in insulin secretion. However, even though in our bulimic subjects body weight was stable during the week before testing, it is possible that weight fluctuations earlier than a week, the abnormal pattern and time of eating, the type of food, or nonspecific psychological stress may have played a role in eliciting abnormal GH and ACTH/cortisol responses to glucose. Similar neuroendocrine abnormalities have been reported by other authors in anorexia nervosa (Tamai et al., 1991), and in this illness the significance of the altered GH and cortisol responses to hyperglycemia is uncertain, as well. In fact, the anorectic women were underweight, and the abnormal GH and cortisol secretory patterns might have been associated with the nutritional status and/or body weight of the subjects, rather than being a trait marker of the eating disorder. In addition, the finding of a negative correlation between GH and insulin responses to glucose in anorectic women suggested a possible link between these hormonal secretions during the IGTT. An alteration of the sensitivity to glucose at a hypothalamic level might be responsible for the altered GH and ACTH/cortisol responses to hyperglycemia in bulimia nervosa. In view of the anatomical proximity and the functional interactions between the hypothalamic nuclei regulating food intake and the hypothalamo-pituitary unit (Morley et al., 1985; Muller, 1987), it is possible that an altered sensitivity to glucose also affects satiety and hunger centers in bulimic women. Such metabolic alterations might be associated with the pathogenesis of the eating disorders; this hypothesis needs further study to be substantiated. The neuroendocrine changes underlying the abnormal GH and ACTH/cortisol responses to glucose are still unknown. Glucose inhibition of both GH (Roth et al., 1964; Tamai et al., 1991) and ACTH/cortisol (Jesova-Repeekova, 1980; Tamai et al., 1991) secretion is thought to be mediated by SRIH. A possible explanation for our findings is that alterations of both hormonal systems were on the basis of deficient glucose-dependent SRIH release. This is a provocative hypothesis, because SRIH is known to play a role in the hypothalamic control of food intake (Morley et al., 1985). Since TSH secretion also is inhibited by SRIH (Scanlon et al., 1979), alterations in bulimic women of the TSH pattern during IGTT were expected. In contrast, our results failed to show differences between groups in glucose-induced TSH decrements, arguing against the hypothesis that, in bulimia, SRIH mediates the abnormal pituitary responses to glucose. The possibility that the dose of glucose used here may induce an increase in SRIH sufficient to suppress TSH release without suppressing ACTI-I/cortisol or GH secretion must be considered; however, the normally higher sensitivity to SRIH of GH than TSH (Vale et al., 1974) makes this possibility unlikely, unless there is altered sensitivity to SRIH in bulimia. It is interesting that the hypothalamic dysfunction responsible for an abnormal sensitivity to glucose in bulimic women does not involve TSH secretion, whereas it is associated with alter-

644

v. colRo et al.

ations in the c o n t r o l o f GH and A C T H / c o r t i s o l s e c r e t i o n . I n d e e d , these c o u n t e r r e g u l a t o r y hormones are closely related to glucose homeostasis and thus to the regulation of food intake. The meaning o f this finding remains to be clarified. Acknowledgements: Supported in part by a grant from the Ministero Pubblica Istruzione, Rome, Italy. REFERENCES American Psychiatric Association (1980) Diagnostic and Statistical Manual of Mental Disorders (Third Edition). American Psychiatric Association, Washington DC. Blouin AG, Blouin JH, Braaten JT, Sarwar G, Busnik T, Walker J (1991) Physiological and psychological responses to a glucose challenge in bulimia, lnt J Eating Disord 10: 285-296. Brock P, Eldred EW, Woiszillo JE, Doran M, Shoemaker MJ (1978) Direct solid phase 125-I radioimmunoassay of serum cortisol. Clin Chem 24: 1595-1598. Casper RC, Eckert ED, Halmi KA, Goldberg SC, Davis JM (1980) Bulimia. Arch Gen Psychiatry 37: 10301035. Casper RC, Pandy GN, Jaspan JB, Rubenstein AH (1988) Hormone and metabolite plasma levels after oral glucose in bulimia and healthy controls. Biol Psychiatry 24: 663-674. Coiro V, Capretti L, Volpi R, d'Amato L, Marchesi C, De Ferri A, Rossi G, Bianconi L, Marcato A, Chiodera P (1990) Growth hormone responses to growth hormone-releasing hormone, clonidine and insulin-induced hypoglycemia in normal weight bulimic women. Neuropsychobiology 23: 8-14. Endicott J, Spitzer RL (1978) A diagnostic interview. The schedule for affective disorders and schizophrenia. Arch Gen Psychiatry 35: 837-844. Glick SM (1970) Hypoglycemic threshold for human growth hormone release. J Clin Endocrinol Metab 30: 619-623. Gwirtsman HE, Roy-Byrne P, Jager J, Gerner RH (1983) Neuroendocrine abnormalities in bulimia. Am J Psychiatry 140: 559-563. Hotta M, Shibasaki T, Masuda A, Imaki T, Demura H, Ling N, Shizume K (1986) The responses of plasma adrenocorticotropin and cortisol to corticotropin-releasing hormone (CRH) and cerebrospinal fluid immunoreactive CRH in anorexia nervosa. J Clin Endocrinol Metab 62: 319-324. Himsworth RL, Camel PW, Frantz AG (1972) The location of the chemoreceptor controlling growth hormone secretion during hypoglycemia in primates. Endocrinology 91: 217-226. Hudson JL, Pope HG, Jonas JM, Yurgelhn-Todd D (1983) Family history study of anorexia nervosa and bulimia. Br J Psychiatry 142:133- r38. Jesova-Rep~ekova D, Vigas M, Klimes I (1980) Effect of glucose on plasma cortisol level after L-dopa administration in man. Horm Metab Res 12: 280-281. Karteszi M, Dallman MF, Makara GB, Stark E (1982) Regulation of the adrenocortical response to insulininduced hypoglycemia. Endocrinology 111: 535-541. Lengyel AMJ, Nieuwenhuyzen Kruseman AC, Grossman A, Rees LH, Besser GM (1984) Glucose-induced changes in somatostatin-14 and somatostatin-28 released from rat hypothalamic fragments "in vitro". Life Sci 35: 713-719. Mitchell JE, Banfle JP (1983) Metabolic and endocrine investigations in women of normal weight with the bulimia syndrome. Biol Psychiatry 18: 355-365. Morley JE, Levine AS (1983) The central control of appetite. Lancet i." 398-401. Morley JE, Levine AS, Gosnell BA, Krahn DD (1985) Peptides as central regulators of feeding. Brain Res Bull 14: 511-519. Mfiller EE (1987) Neural control of somatotropic function. Physiol Rev 67: 962-1053. Nakagawa K, Matsubara M, Obara J, Kubo M, Aldkawa K (1985) Responses of pituitary and adrenal medulla to insulin-induced hypoglycemia in patients with anorexia nervosa. Endocrinol Japon 32: 719-724. Newman MM, Halmi KA (1988) The endocrinology of anorexia nervosa and bulimia nervosa. Endocrinol Metab Clin N Arn 17: 195-212.

G H , CORTISOL AND TSH RESPONSES TO GLUCOSE tN BULIMIA

645

Norris PD, O'Malley BP, Palmer RL (1985) The TRH test in bulimia and anorexia nervosa: a controlled study. J Psychiat Res 19: 215-219. Rattle SJ, Purnell DR, Williams P, Siddle K, For'rest GC (1984) New separation method for monoclonal immunoradiometric assays and its application to assays for thyrotropin and human choriogonadotropin. Clin Chem 30: 1457-1462. Roth J, Glick SM, Yalow RS, Berson SA (1964) The influence of blood glucose on the plasma concentrations of growth hormone. Diabetes 13: 355-361. Russel GFM (1978) Bulimia nervosa: an ominous variant of anorexia nervosa. Psychol Med 9: 939-948. Scanlon MF, Pourmand M, McGregor AM, Rodriguez-Arnao MD, Hall K, Gomez-Pan A, Hall R (1979) Some current aspects of clinical and experimental neuroendocrinology with particular references to growth hormone, thyrotropin and prolactin. J Endocrinol Invest 2: 307-331. Schalch DS, Parker ML (1964) A sensitive double antibody immunoassay for human growth hormone in plasma. Nature 203:1141-1142. Schweiger U, Poellinger J, Laessle R, Wolfram G, Ficher MM, Pirke KM (1987) Altered insulin response to a balanced test meal in bulimic patients, lnt J Eating Disord 6: 455-468. Shibasaki T, Masuda A, Hotta M, Yamauchi N, Hizuka N, Takano K, Demura H, Shizume K (1989) Effects of ingestion of glucose on GH and TSH secretion: evidence for stimulation of somatostatin release from the hypothalamus by acute hyperglycemia in normal man and its impairment in acromegalic patients. Life Sci 44: 431-438. Tamai H, Kiyohara K, Mukuta T, Kobayashi N, Komaki G, Nakagawa T, Kumagai LF, Aoki TT (1991) Responses of growth hormone and cortisol to intravenous glucose loading test in patients with anorexia nervosa. Metabolism 40: 31-34. Vale W, Rivier C, Brazeau P, Guillemin R (1974) Effects of somatostatin on the secretion of thyrotropin and prolactin. Endocrinology 95: 968-977. Weingarten HP, Hendler P, Rodin J (1988) Metabolism and endocrine secretion in response to a test meal in normal weight bulimic women. Psychosom Med 50: 273-285. Williamson DH, Mellonby J (1974) Betahydroxy-butirat. In: Bergmeyer HU (Ed) Methoden der Enzymatischen Analyse. Verlag Chemie, Weilheim, pp 1883-1886. Yalow RS, Berson SA (1960) Immunoassay of endogenous plasma insulin in man. J Clin Invest 39: 11571175.