Ž .Psychiatry Research 81 1998 293]299

Growth hormone response to growth hormone-releasinghormone stimulation in obsessive]compulsive disorder

Francesca BrambillaU , Laura Bellodi, Giampaolo Perna, Cinzia Arancio,Angelo Bertani

Centro di Psiconeuroendocrinologia, Dipartimento di Scienze Neuropsichiche, Istituto Scientifico Ospedale S. Raffaele, ViaPrinetti 29, Milano 20127, Italy

Received 15 October 1997; received in revised form 17 June 1998; accepted 22 June 1998

Abstract

Two groups of 30 patients with obsessive]compulsive disorder and 30 age- and sex-matched healthy controlŽ . Ž .subjects were given a growth hormone-releasing hormone GHRH stimulation test to determine: 1 whether the

Ž . Ž .downstream function of the somatotropic axis growth hormonesGH, somatomedin-CsSMD-C was impaired; 2Ž .what might be the central alteration responsible for such impairment; and 3 whether alterations might be linked to

the etiopathogenesis of the disease. Basal values of GH and SMD-C were the same in patients and control subjects,but GH responses to GHRH stimulation were significantly lower in patients than in control subjects. The absence ofa pathology of basal GH and SMD-C concentrations indicates that the blunted GH responses to GHRH stimulationare not due to a negative feedback mechanism and suggests that a central neurotransmitter]neuropeptide pathologymight be involved in the phenomenon. Q 1998 Elsevier Science Ireland Ltd. All rights reserved.

Keywords: Somatotropic axis; Somatomedin-C; Endocrine factors; Obsessive compulsive disorder

1. Introduction

Ž .The growth hormone GH response to stimu-lation with endogenous growth hormone-releas-

Ž .ing hormone GHRH is part of a downstreamprocess that begins in the hypothalamus and,possibly, in higher brain centers, with secretion of

U Corresponding author. Fax: q39 270122889

multiple neurotransmitters and neuropeptidesthat either stimulate or inhibit GHRH and so-

Ž .matostatin SMT production in the hypothala-mus. At the pituitary level, GHRH stimulates andSMT inhibits GH secretion, and GH stimulates

Ž .the production of somatomedin-C SMD-C bythe liver. In turn, GH and SMD-C inhibit GHRHand SMT secretion through a feedback mecha-nism in the hypothalamus and may also modulatethe secretion of central neurotransmitters]neuro-

0165-1781r98r$ - see front matter Q 1998 Elsevier Science Ireland Ltd. All rights reserved.Ž .P I I: S 0 1 6 5 - 1 7 8 1 9 8 0 0 1 2 4 - 3

( )F. Brambilla et al. r Psychiatry Research 81 1998 293]299294

Žpeptides and their receptor functions serotonin,noradrenaline, dopamine D receptors, a- and2

. Ž .b-adrenergic receptors Muller, 1987 . Thus, analteration of the GH response to GHRH stimula-tion could be due to and reveal the presence ofcomplex central and peripheral impairments.

In previous studies investigating the function ofa -adrenergic receptors and D receptors in2 2patients with obsessive]compulsive disorderŽ . Ž .OCD Brambilla et al., 1997a,b,c , we observedthat even though the mean GH responses toGHRH stimulation did not differ between patientsand control subjects, individual responses wereblunted in some of the patients. In all, 16 of the30 patients who entered the two studies hadblunted responses, compared with only two of thecontrol subjects. Since the number of patientsand control subjects in each study was rather lowŽ .15 patients and 15 control subjects in each , it ispossible that the sample sizes were too small toguard against non-significance due to type II er-ror. For that reason, we decided to combine thedata of the two previously studied groups of OCDpatients since they did not differ demographically,were diagnosed by the same criteria, were investi-gated with the same neuroendocrine procedureand were symptomatologically monitored with thesame rating scales by the same specialists. Theaim of the study was to see whether statisticalanalyses of a larger number of subjects would

Ž .reveal: 1 whether OCD is characterized by asignificant pathology of the peripheral soma-totropic axis, and if so, what might be the central

Ž .alteration responsible for it; 2 whether such analteration might be linked to the etiopathogenesis

Ž .of the disease; and 3 whether hormonal alter-ations might be correlated with specific clinicalaspects of OCD.

2. Methods

2.1. Subjects

ŽThirty physically healthy patients with OCD 15women, 15 men, mean ages32.9 years, S.D.s

.10.8; ranges16]60 years , recruited over a 1-yearperiod at the Anxiety Disorder Clinic and Re-search Unit of the Dipartimento di Scienze Neuro-

psichiche, Istituto Scientifico Ospedale S. Raf-faele, Milano, Italy, entered the study. Duration

Žof illness ranged from 1 to 32 years means13.6.years, S.D.s9.3 . One subject was only 16 years

old but was included in the group because he hadfully completed pubertal maturation. Diagnosesof OCD were based on the Diagnostic Interview

Ž .Schedule-Revised DIS-R; Robins et al., 1989and the diagnostic criteria of DSM-III-RŽ .American Psychiatric Association, 1987 .

The control group comprised 15 men and 15women, recruited from the university staff byadvertising, who were age-matched to the patientsand who were physically healthy as determined bymedical history, physical examination and labora-tory tests. Control subjects were interviewed withthe DIS-R to exclude any present or past psychi-atric disorders that satisfied DSM-III-R criteria.Both patients and control subjects were excludedfrom the study if they had physical organic dis-eases, immunopathies, metabolic and endocrinedisorders, obesity or recent weight loss, organicbrain disorders, cerebral trauma, abuse of illicitdrugs or alcohol and, in the patients, any lifetimediagnosis of Axis I or II disorders other thanOCD.

Of the 30 patients, 26 had been previouslyŽtreated with psychotropic drugs short-acting

phenothiazines, haloperidol, clomipramine, flu-oxetine, fluvoxamine, and various benzodi-

.azepines, alone or in combination . All medica-tions had been discontinued before the present

Žstudy began drug washout was at least 6 monthsin 20 cases and 4 weeks in the remaining six

.cases . Twelve of the 15 women with OCD and 12control women were of reproductive age and re-ported histories of normally cycling menses. Theendocrine challenges were always given duringthe first 6 days of the menstrual cycle or in the 4days immediately preceding it to avoid cycle-re-lated estrogen fluctuations that could modify the

ŽGH response to the stimulus Erikson, 1985;.Weissberger et al., 1991 . The other female

patients and control subjects were post-menopausal.

After complete description of the study topatients and control subjects, written informedconsent was obtained. The day before the GHRH

( )F. Brambilla et al. r Psychiatry Research 81 1998 293]299 295

stimulation test, a baseline measure of the sever-ity of OC symptoms was obtained with the

ŽYale]Brown Obsessive]Compulsive Scale Y-.BOCS; Goodman et al., 1989 . The mean Y-BOCS

Žglobal score was 29.1 S.D.s6.2; mean scores forobsessiveness and compulsiveness were 14.6Ž . Ž .S.D.s3.4 and 14.4 S.D.s3.3 , respectively.

2.2. Procedure

The GHRH test was performed in the morning,with patients in the recumbent position, after 12h of fasting, refraining from smoking and drinkingalcohol or coffee, and 1 h of bed rest. At 08.30h, abutterfly needle was inserted into a forearm veinand kept patent by saline infusion. At 08.30 and09.00h, EDTA-anticoagulated blood was drawn tomeasure basal concentrations of GH and SMD-C,and the two values for each hormone were aver-

Ž .aged. At 09.00h, GHRH 1 mgrkg b.w. was in-jected i.v. as a bolus, diluted in 10 ml of saline.Thereafter, blood samples for GH assays weredrawn at 15-min intervals for 1 h and at 30-minintervals for another hour, immediately cen-trifuged and plasma frozen at y208C until as-sayed. According to the standard of our labora-

Žtory and to data in the literature Trestman et al.,.1994 , the cutoffs for significant responses of GH

to GHRH stimulation were an increase of thehormone concentrations at peak time of morethan double the basal levels and an absolute peaktime value of more than 5 ngrml. Plasma GHconcentrations were measured radioimmunologi-

Ž .cally with commercial kits from Sorin Italy andŽ .those of SMD-C with the kits of Nichols USA ,

after extraction of the plasma with alcohol-HCl.

2.3. Statistical analyses

Data were analyzed statistically by the fol-Ž . Ž .lowing tests: 1 Analysis of variance ANOVA

for repeated measures on logarithmic trans-formed data, to evaluate overall GH responses toGHRH stimulation. Diagnosis and Sex were thegrouping factors, and hormonal levels across the

Ž .stimulation test Time were the dependent vari-Ž .ables; 2 Mann]Whitney U-test, to compare twoŽgroups e.g. patients vs. control subjects, women

vs. men, and responders vs. non-responders to.GHRH stimulation on continuously distributed

Ž .variables; 3 Chi-square analysis, to measure thesignificance of differences for rates of positiveand negative responders to GHRH stimulation,and to assess the patterns of sex distribution

Ž .patients versus control subjects; 4 Friedman testfor repeated measures, to validate the signifi-cance of the stimulation effect at each point of

Ž .the curves; and 5 Spearman’s rank correlationtest, to evaluate the relationships between areas

Ž .under the curve AUC of GH responses toGHRH stimulation and clinical and demographicparameters. Since basal GH levels and responsesto stimuli tend to be higher in women than inmen, due to the stimulating effect of GH secre-

Ž .tion Erikson, 1985; Weissberger et al., 1991 , wealso analyzed the results separately for men andwomen.

3. Results

Basal values of GH in OCD patients and con-Žtrol subjects did not differ significantly mean"

S.D. ngrml for OCD men s 1.8 " 33; OCDwomens1.7"1.7; control mens1.3"0.9; con-

.trol womens1.9"0.9 . The hormonal values aty30 min and 0 time were not significantly dif-ferent from one another in patients or controlsubjects, and also did not differ in men andwomen. Basal SMD-C levels did not differ sig-

Žnificantly in patients and control subjects mean"S.D. ngrml for OCD mens239.5"93.0; OCDwomen s 245.6 " 85.9; control men s 259.9 "

.60.9; control womens319.3"49.5 . Values ob-tained at y30 min and 0 time did not differ fromone another and were not different in men andwomen in either group.

GH responses to GHRH stimulation wereblunted in OCD patients compared with values in

Ž .control subjects Figs. 1]3 . Friedman analysis forrepeated measures revealed that the stimulationelicited significant responses in both the patientsŽ 2 .x s50.9, d.f.s6, P-0.0001 and control sub-

Ž 2 .jects x s105.3, d.f.s6, P-0.0001 , suggestingthat the pituitary somatotrops in both groupswere able to respond to the stimulation withGHRH. ANOVA for repeated measures showed

( )F. Brambilla et al. r Psychiatry Research 81 1998 293]299296

Ža significant Sex = Time interaction Fs4.4.d.f.s6, P-0.0003 , with women having weaker

responses than men, and a significant DiagnosisŽ .= Time interaction Fs2.8, d.f.s6 P-0.02 ,

with patients showing weaker responses than thecontrol subjects. The Sex = Diagnosis = Timeinteraction was not significant. The Mann-Whit-ney U-test showed that the AUCs were signifi-

Ž .cantly lower in patients 643.5"979.6 than inŽ . Žcontrol subjects 959.2"859.3 zsy2.4, P-

.0.02 . There were significantly more non-re-sponders to the stimulus among the patients than

Ž 2among the control subjects 16 vs. 2, x s15.6,.d.f.s12, P-0.0001 , without differences between

men and women. The Mann]Whitney analysis ofthe AUC values of male patients and controlsubjects as a group vs. those of female patientsand control subjects as a group revealed a statisti-cally significant difference, with women having

Žlower values AUC in mens939.3"885.0, AUC.in womens6614"995.0, zsy2.3, P-0.03 .

When the AUC values of men and women wereanalyzed separately for OCD and control sub-jects, there were no statistically significant dif-

Žferences male patients s1059.0"874.7 vs. fe-male patients s859.0"934.8; male control sub-jectss819.6"909.2 vs. female control subjects s

.467.5"1046.3 .When the AUC values of patients vs. control

subjects were analyzed by the Mann]Whitney

Ž .Fig. 1. Growth hormone responses mean"SE to growthhormone-releasing hormone stimulation in 30 men, 15 with

Ž .obsessive]compulsive disorder OCD and 15 control subjects.

Ž .Fig. 2. Growth hormone responses mean"SE to growthhormone-releasing hormone stimulation in 30 women, 15 with

Ž .obsessive-compulsive disorders OCD and 15 control subjects.

Fig. 3. Growth hormone responses to growth hormone-releas-Žing hormone stimulation expressed as areas under the curves;

.mean"S.D. of 30 patients with obsessive]compulsive dis-Ž .order OCD and 30 control subjects, 15 men and 15 women

Ž .for each group mensblack column, womensgray column .

U-test, the women’s values were statistically dif-Žferent for both groups OCD patients s467.5"

1046.3 vs. control subjects s859.4"934.8, zs.y2.6, P-0.02 , while those of men were not

ŽOCD patients s1059.0"874.7 vs. control sub-.jectss819.6 909.2 . Finally, in patients, Spear-

man’s rank correlation tests between AUC valuesand age, age of onset, duration of previous thera-pies, length of drug washout time, Y-BOCS global

Žscores and subscale scores obsessions, compul-.sions showed no significant values. The same was

true when the patients were divided into respon-ders and non-responders to the GHRH test.

( )F. Brambilla et al. r Psychiatry Research 81 1998 293]299 297

4. Discussion

Even though our results are preliminary andobtained from too small a group of patients topermit definite conclusions to be drawn, theblunted GH responses to GHRH stimulation thatwe observed in a substantial group of patientssuggest an impairment in OCD of the regulatorymechanism of secretion of the somatotropic axis.This cannot be due to a peripheral negative feed-back mechanism, since basal plasma concentra-tions of GH and SMD-C were normal in all ofour patients.

A primary, isolated, hypothalamus-independentalteration of pituitary somatotrop function is dif-ficult to envisage and explain. A prolonged hy-pothalamic deficiency of GHRH secretion isknown to be followed by reduced pituitary re-sponses to the stimulus with the releasing hor-

Ž .mone Rh , and to be normalized after repeatedŽadministration of exogenous GHRH Muller,

1987; Blanchard et al., 1989; Pavia et al., 1993;.Skare et al., 1994 . Reduced GHRH secretion

could be due to a catecholaminergic deficiency,Ž . Ž .either of norepinephrine NE or dopamine DA ,

each of which is a physiological stimulator of RhŽ .production Muller, 1987 . However, in OCD, DA

secretion, as expressed by cerebrospinal fluidŽ .CSF , plasma and urinary concentrations of the

Ž .main DA metabolite homovanillic acid HVAand by responses of post-synaptic DA receptorsto pharmacological stimuli, has been repeatedly

Žreported to be normal or increased Thoren et al.,´1980; Benkelfat et al., 1991; Swedo et al., 1992;

.Marazziti et al., 1992; Brambilla et al., 1997a .ŽOnly Hollander et al. 1991; Hollander et al.,

.1992 have suggested, but not proved, that DAproduction might be reduced as a consequence ofupward regulation of serotonin receptors. NE se-cretion has also been reported to be either nor-mal or increased, when examined as CSF orplasma concentrations of NE, or of its mainmetabolite 3-methoxy-4-hydroxy-phenylglycolŽ .MHPG , or through investigation of presynapticand post-synaptic central and peripheral adrener-

Žgic receptor sensitivity Siever et al., 1983; Ras-mussen et al., 1987; Khanna et al., 1988; Lee etal., 1990; Hollander et al., 1989; Hollander et al.,

.1991; Lucey et al., 1992, Brambilla et al., 1997a,b .

Although both post-synaptic D receptor and2a -adrenergic receptor sensitivities might be2

Ždown-regulated in OCD Brambilla et al.,.1997a,b,c , it seems rather unlikely that this could

reduce the catecholaminergic stimulation ofGHRH secretion so profoundly as to lead to theblunted GH responses to the stimulus observed.It has also been reported that repeated injectionsof GHRH into experimental animals result indown-regulation of the Rh receptors at the pitu-

Ž .itary somatotrop level Muller, 1987 . Our patientsshowed increased noradrenergic and dopaminer-

Ž .gic secretory tonus Brambilla et al., 1997a,b,c ,which could have hyperstimulated GHRH secre-tion and induced a pituitary receptor subsensitiv-ity.

Alternatively, the blunted GH responses mightbe due to increased hypothalamic secretion of

Ž .somatostatin SMT , which inhibits GHRH-in-Ž .duced GH rises Muller, 1987 . Increased SMT

secretion might be expected in OCD patients as aconsequence of hypersecretion of the NE-DA-de-pendent GHRH or of acetylcholine, which stimu-

Žlates SMT rises Lucey et al., 1993; Reichard et. Ž .al., 1996 . Altemus et al. 1993 reported in-

creased SMT concentrations in the CSF of OCDŽ .patients. However, Roy et al. 1994 observed

high plasma levels of antisomatostatin-14 anti-bodies, which should antagonize SMT function,with ensuing disinhibition of GHRH-GH produc-tion. Other possible causes of altered GH respon-ses to stimuli are obesity, starvation with recentweight loss, hyperglycemia or diabetes, which wereall criteria for exclusion from the study for both

Žpatients and control subjects Williams et al.,1984; Masuda et al., 1985; Harel and Tannen-

.baum, 1993 .The explanation for the occurrence of blunted

GH responses to GHRH in only 16 of the 30patients is unclear, since no significant demo-graphic or symptomatologic differences wereobserved between responders and non-re-sponders. Our research design did not permit usto determine whether the impaired somatotropicfunction was a state or trait characteristic of asubgroup of patients. To resolve this question, avery longitudinal study encompassing both phasesof active symptomatology and remission would benecessary. This type of study might be of interest,

( )F. Brambilla et al. r Psychiatry Research 81 1998 293]299298

since impairments of somatotropic function havebeen observed in major depressive disordersŽ . Ž .MDD and in generalized anxiety disorder GADŽLesch et al., 1987; Lesch and Erb, 1988; Abelson

.et al., 1991 . Even though no comorbidities withMDD and GAD were observed in the subgroupof OCD patients who were characterized byblunted GH responses to GHRH stimulation,these subjects could be more prone to the laterdevelopment of MDD or GAD.

It is also unclear why blunted GH responsesoccurred more frequently in women than in men.Estrogens are known to stimulate GH secretion

Žmore than testosterone does Erikson, 1985; Ho.et al., 1987; Lang et al., 1987 . However, Martin

Ž . Ž .et al. 1968 , Pieters et al. 1980 , and Smals et al.Ž .1986 reported higher GH responses to GHRHstimulation in normal men than in women, butwere unable to explain the observed pheno-menon. Administration of high doses of estrogensto castrated female rats decreased the GH secre-tory response to GHRH, either by increasinghypothalamic SMT secretion or by decreasing so-matotrop GHRH receptor number or affinityŽ .Schulman et al., 1987 . Even though we did notmeasure estrogen levels in our patients, theirnormal menstrual histories suggest that estrogenconcentrations reaching the hypothalamus andpituitary should have been in the normal rangeand thus ought not to have interfered with theGH secretion. In conclusion, our data indicatethat in OCD there is a complex dysfunction ofsomatotropic axis secretion, which might possiblybe linked to or be an expression of the alterationsof catecholaminergic function that have been re-peatedly found in the course of the disorder. Theetiopathogenetic significance of this alteration forOCD is still obscure and needs further and morecomplex studies, including a long-term simultane-ous investigation of multiple neurotransmitter]neuropeptide systems.

References

Abelson, J.L., Glitz, D., Cameron, O.G., Lee, M.A., Bronzo,M., Curtis, G.C., 1991. Blunted growth hormone responseto clonidine in patients with generalized anxiety dsorder.Archives of General Psychiatry 48, 157]162.

Altemus, M., Pigott, T., L’Heureux, F., Davis, C.L., Rubinow,D.R., Murphy, D.L., Gold, P.N., 1993. CSF somatostatin inobsessive]compulsive disorder. American Journal of Psy-chiatry 150, 460]464.

American Psychiatric Association, 1987. DSM-III-R: Diagnos-tic and Statistical Manual of Mental Disorders, 3rd reviseded. American Psychiatric Press, Washington, DC.

Benkelfat, C., Mefford, I.N., Masters, C.F., Nordahl, T.E.,King, A.C., Cohen, R.M., Murphy, D.L., 1991. Plasma cate-cholamines and their metabolites in obsessive]compulsivedisorder. Psychiatry Research 37, 321]331.

Blanchard, M., Charrier, L., Kann, G.N., Garandel, V., Fraysse,A., Barenton, B., 1989. Influence d’un traitement chronique

Ž .au GRF 1]21 ou au clenbuterol chez la brebis en fin degestation sur I’activite de l’axe somatotrope fetale et´

Ž .maternel. Abstract Annales d’Endocrinologie 50, 256.Brambilla, F., Perna, G., Bellodi, L., Arancio, C., Bertani, A.,

Perini, G., Carraro, C., Gava, F., 1997a. Noradrenergicreceptor sensitivity in obsessive]compulsive disorders:growth hormone response to clonidine stimulation. Psychi-atry Research 69, 155]162.

Brambilla, F., Bellodi, L., Perna, G., Arancio, C., Bertani, A.,Perini, G., Carraro, C., Gava, F., 1997b. Noradrenergicreceptor sensitivity in obsessive]compulsive disorders: II.Cortisol response to acute clonidine administration. Psychi-atry Research 69, 163]168.

Brambilla, F., Bellodi, L., Perna, G., Arancio, C., Bertani, A.,1997c. Dopamine function in obsessive]compulsive dis-order: growth hormone response to apomorphine stimula-tion. Biological Psychiatry 42, 889]897.

Erikson, E., 1985. Brain a-2-adrenoceptor function and growthhormone release. In: Experimental Psychoneuroen-docrinology. Medi Press, Goteborg.

Goodman, W.K., Price, L.H., Rasmussen, S.A., Mazure, C.,Fleischman, R.L., Hill, C.L., Heninger, G.R., Charney, D.S.,1989. The Yale]Brown Obsessive]Compulsive Scale. I.Development, use and reliability. Archives of General Psy-chiatry 46, 1006]1011.

Harel, Z., Tannenbaum, G.S., 1993. Dietary protein restrictionimpairs both spontaneous and growth hormone-releasingfactor-stimulated growth hormone release in the rat. En-docrinology 133, 1035]1043.

Ho, K.Y., Evans, W.R., Blizzard, R.M., Veldhuis, J.D., Mer-riam, G.R., Samoljlik, E., Furlanetto, R., Rogol, A.D.,Kaiser, D.L., Thorner, M.O., 1987. Effects of sex and ageon the h profile of growth hormone secretion in man:importance of endogenous estradiol concentrations. Jour-nal of Clinical Endocrinology and Metabolism 64, 51]58.

Hollander, E., DeCaria, C.M., Nitescu, A., Cooper, T.B.,Stover, B., Gully, R., Klein, D.F., Liebowitz, M.R., 1991.Noradrenergic function in obsessive]compulsive disorder:behavioral and neuroendocrine responses to clonidine andcomparison to healthy control subjects. Psychiatry Re-search 37, 161]179.

Hollander, E., Liebowitz, M.R., DeCaria, C.M., 1989. Concep-tual and methodological issues in studies of obsessive]com-

( )F. Brambilla et al. r Psychiatry Research 81 1998 293]299 299

pulsive and Tourette’s disorders. Psychiatric Development4, 276]296.

Hollander, E., Stein, D.J., Saoud, J.B., DeCaria, C.M., Cooper,T.B., Trungold, S., Stanley, M., Liebowitz, M.R., 1992.Effects of fenfluramine on plasma HVA in OCD. Psychia-try Research 42, 185]188.

Khanna, S., Reddy, P.L., Subhash, M.N., Rao, B.S.S.S.,Channabasavanna, S.M., 1988. Neuroendocrine profile ofobsessive]compulsive disorder. Neuroendocrinology Let-ters 10, 250.

Lang, I., Schernthaner, G., Pietschmann, P., Kurz, R.,Stephenson, J.M., Templ, H., 1987. Effect of sex and age ongrowth hormone in healthy individuals. Journal of ClinicalEndocrinology and Metabolism 65, 535]540.

Lee, M.A., Cameron, O.G., Gurguis, G.N.M., Glitz, D., Smith,C.B., Hariharan, M., Abelson, L.L., Curtis, G., 1990. Alpha-2-adrenoceptor status in obsessive]compulsive disorder.Biological Psychiatry 27, 1083]1093.

Ž .Lesch, K.P., Erb, A., 1988. Growth hormone GH responsesto GH-releasing hormone in depression: correlation withGH release following clonidine. Psychiatry Research 25,301]310.

Lesch, K.P., Laux, G., Pfuller, H., Erb, A., Beckmann, H.,Ž .1987. Growth hormone GH response to GH-releasing

hormone in depression. Journal of Clinical Endocrinologyand Metabolism 65, 1278]1281.

Lucey, J.V., Barry, S., Webb, M.G., Dinan, T.G., 1992. Thedesipramine-induced growth hormone response and thedexamethasone suppression test in obsessive]compulsivedisorder. Acta Psychiatrica Scandinavica 86, 367]370.

Lucey, J.V., Butcher, G., Clare, A.W., Dinan, T.G., 1993.Elevated growth hormone responses to pyridostigmine inobsessive]compulsive disorder: evidence of cholinergic su-persensitivity. American Journal of Psychiatry 150, 961]962.

Martin, L.G., Clark, J.W., Conner, T.B., 1968. Growth hor-mone secretion enhanced by androgens. Journal of ClinicalEndocrinology and Metabolism 28, 425]429.

Masuda, A., Shibasaki, T., Nakahara, M., Imaki, T., Kiyosawa,Y., Jibiki, K., Demura, H., Shizume, K., Ling, N., 1985. The

Ž .effect of glucose on growth hormone GH releasing hor-mone-mediated GH secretion in man. Journal of ClinicalEndocrinology and Metabolism 60, 523]526.

Marazziti, D., Hollander, E., Lensi, P., Ravagli, S., Cassano,L.B., 1992. Peripheral markers of serotonin and dopaminefunction in obsessive]compulsive disorder. Psychiatry Re-search 42, 41]52.

Muller, E.E., 1987. Neural control of somatotropic function.Physiological Review 3, 962]1053.

Pavia, C., Valls, C., Gallart, M., 1993. Priming with GHRHŽ .1]29 -NH-2 as a useful tool in the origin diagnosis of

Ž .growth hormone GH deficiency. Acta Medica Auxologica25, 133]138.

Pieters, G.F.F.M., Smals, A.G.H., Kloppenborg, P.W.C., 1980.Defective suppression of growth hormone after oral glu-cose loading in adolescence. Journal of Clinical En-docrinology and Metabolism 51, 265]270.

Rasmussen, S.A., Goodman, W.K., Woods, S.W., Heninger,G.R., Charney, D.S., 1987. Effects of yohimbine in obses-

sive]compulsive disorder. Psychopharmacology 93,308]313.

Reichard, B., Schrader, M., Mojto, J., Mehltretter, G., Muller,O.A., Schopohl, L., 1996. The decrease in growth hormoneŽ .GH response after repeated stimulation with GH-releas-ing hormone is partly caused by an elevation of so-matostatin tonus. Journal of Clinical Endocrinology andMetabolism 81, 1994]1998.

Robins, L., Helzer, L.E., Cottler, L., Goldring, E., 1989. NIMHŽ .Diagnostic Interview Schedule: Version III-R DIS-III-R .

Washington University Medical School, St. Louis, MO.Roy, B.F., Benkelfat, C., Hill, L.L., Pierce, P.F., Dauphin,

M.M., Kelly, T.M., Sunderland, T., Weinberger, D.R., Bres-lin, N., 1994. Serum antibody for somatostatin-14 and pro-dynorphin 209]240 in patients with obsessive]compulsivedisorder, schizophrenia, Alzheimer’s disease, multiple scle-rosis and advanced HIV infection. Biological Psychiatry 35,335]344.

Schulman, D.I., Sweetland, M., Duckett, G., Root, A.W., 1987.Ž .Effect of estrogen on the growth hormone GH secretory

response to GH-releasing factor in the castrate adult fe-male rat in vivo. Endocrinology 120, 1047]1051.

Siever, L.J., Insel, T.R., Jimerson, D.C., Lake, R.C., Uhde,T.W., Aloi, J., Murphy, D.L., 1983. Growth hormone re-sponse to clonidine in obsessive]compulsive patients.British Journal of Psychiatry 142, 184]187.

Skare, S.S., Dysken, M.W., Billington, C.J., 1994. A review ofthe GHRH stimulation test in psychiatry. Biological Psychi-atry 36, 249]265.

Smals, D.I., Sweetland, M., Duckett, G., Root, A.W., 1986.Ž .Effect of estrogen on the growth hormone GH secretory

response to GH-releasing factor in the castrate adult fe-male rat in vivo. Endocrinoogy 120, 1047]1051.

Swedo, S.E., Leonard, H.L., Kruesi, M.J.P., Rette, D.C., List-vak, S.J., Berrettini, W., Stipetic, M., Hamburger, S., Gold,P.W., Potter, W.Z., Rapoport, J.L., 1992. Cerebrospinalfluid neurochemistry in children and adolescents withobsessive]compulsive disorder. Archives of General Psy-chiatry 49, 29]36.

˚Thoren, P., Asberg, M., Bertilsson, L., Mellstrom, B., Sjoqvist,´ ¨ ¨R., Traskman, L., 1980. Chlomipramine treatment of obses-¨sive]compulsive disorders. II Biochemical aspects. Archivesof General Psychiatry 37, 1289]1294.

Trestman, R.L., Coccaro, E.F., Lawrence, T., Knott, P.,Gabriel, S.M., Siever, L.J., 1994. The stability of plasmagrowth hormone and X1HPG responses to repeated cloni-dine challenge in normal males. Psychoneuroendocrinology19, 13]20.

Weissberger, A.J., Ho, K.Y., Lazarus, L., 1991. Contrastingeffects of oral and transdermal routes of estrogen replace-

Ž .ment therapy on 24-h growth hormone GH secretion,insulin-like growth factor I, and GH-binding protein inpost-menopausal women. Journal of Clinical Endocrinologyand Metabolism 72, 374]381.

Williams, T., Berelowitz, M., Joffe, S.N., Thorner, M.O., Riv-ier, J., Vale, W., Frohman, L.A., 1984. Impaired growthhormone responses to growth hormone-releasing factor inobesity. New England Journal of Medicine 211, 1403]1407.