- Email: [email protected]
Thyrotrophin response to thyrotrophin-releasing hormone in unipolar and bipolar affective illness
Journal
of Affective
3 (1981) 9-16 Biomedical Press
Disorders,
0 Elsevier/North-Holland
9
THYROTROPHIN RESPONSE TO THYROTROPHIN-RELEASING HORMONE IN UNIPOLAR AND BIPOLAR AFFECTIVE ILLNESS
1 P. LINKOWSKI, 2 H. BRAUMAN and ’ J. MENDLEWICZ i Department of Psychiatry, Hospital Erasme, Free University of Brussels and 2 Radioimmunology Unit, Laboratory of Clinical Biology, Hospital Brugmann, Free University Brussels, Brussels (Belgium)
of
(Received 13 May, 1980) (Revised, received 12 August, 1980) (Accepted 14 August, 1980)
SUMMARY
The plasma levels of thyrotrophin (TSH) and the response of this hormone to 200 pg of thyrotrophin-releasing hormone (TRH) were studied in 27 unipolar and 24 bipolar depressive patients before and after amitriptyline treatment as well as in 42 normal controls. There was no significant difference in basal TSH levels between any of the groups studied according to diagnosis, menopausal status and antidepressant treatment. Before treatment, the TSH response to TRH was significantly lower in the unipolar and bipolar depressive patients than in normal controls. The TSH response to TRH did not differ significantly between the unipolar and bipolar depressives. When menopausal status was taken into account, the TSH response to TRH was significantly blunted in the unipolar postmenopausal patients when compared to postmenopausal controls. In the bipolar group, the premenopausal depressive patients had a significantly lower TSH response to TRH than premenopausal controls. After amitriptyline treatment, the TSH response to TRH, which was impaired before treatment in the bipolar premenopausal patients, improved significantly. However, no significant difference in the TSH response to TRH could be demonstrated after treatment in the unipolar postmenopausal patients in whom the TSH response was blunted before treatment. Our findings suggest that the differential TSH response to TRH in unipolar and bipolar patients may constitute biological markers of endocrine dysfunction in clinical subgroups of affective disorders.
This study has been supported by grants from the University of Brussels, the fondation De Cooman and the Association for Mental Health Research. Correspondence to: J. Mendlewicz, M.D., Ph.D., Professor and Head, Department of Psychiatry, Hospital Erasme, Free University of Brussels, 808 Route de Lennick, 1070 Brussels, Belgium.
10 INTRODUCTION
In early studies of depressed patients, Kastin et al. (1972) and Prange et al. (1972) demonstrated an impaired response of serum thyrotrophin (TSH) to thyrotrophin-releasing hormone (TRH), administered intravenously. Similar findings have been observed by several authors (Coppen et al. 1974; Kierkegaard et al. 1975; Gregoire et al. 1976; Loosen et al. 1977). Kierkegaard et al. (1975) reported the mean maximal TSH response to TRH (Amax) to be normalized in some patients after recovery but these data have been challenged by other authors (Maeda et al. 1975; Loosen et al. 1977) who were unable to confirm these results. These conflicting observations could be due to differences in sampling procedures and/or to the clinical, biochemical and genetic heterogeneity of depressive disorders. Gold et al. (1977) showed in patients with unipolar illness, a significant negative correlation between the blunting of the TSH response to TRH and the main serotonin CSF metabolite, 5-hydroxyindolacetic acid (5-HIAA) while this was not the case in bipolar patients and in normal controls. The present study was undertaken to investigate the pattern of secretion of a pituitary hormone, thyrotrophin-stimulating hormone (TSH) in response to stimulation by thyrotrophin-releasing hormone (TRH) in 51 depressive patients and 42 normal controls matched for age, sex and menopausal status. The purpose of the study was to determine if disturbances in neuroendocrine secretion could be used as biological markers in well defined subgroups of affectively ill patients, and to explore the possible influence of the menopause on TSH response to TRH in affective illness. METHODS
Female patients, 20 premenopausal (15 unipolar (depressive) and 5 bipolar (manic-depressive), aged 20-50, mean 37) and 31 postmenopausal (12 unipolar and 19 bipolar, aged 50-60, mean 55) were studied during a depressive episode severe enough to warrant hospitalisation. Psychiatric diagnoses were made according to the criteria of Feighner (Feighner et al. 1972). The normal controls were 31 premenopausal (mean 35, range 20-50) and 11 postmenopausal women (mean 53, range 50-60). All subjects were in good physical health without clinical and laboratory evidence of any somatic disease. The premenopausal patients and controls were tested at random during the menstrual cycle. All patients were tested before and some also after treatment which consisted of the administration of amitriptyline ranging from 75 to 300 mg/day. Patients were retested after recovery and were still on antidepressant treatment. Informed consent was obtained from patients and controls. The TRH test was performed at 9 a.m. in a recumbent position. All patients were on a monoamine-free diet and off psychotropic drugs for at least 7 days. Synthetic TRH (Roche, 200 pg) was administered intravenously
11
at time 0. Blood was sampled through an indwelling catheter at -15,0,20, 30, 45, 60, 120 min after TRH injection for determination of serum TSH. 30-60 min were allowed between catheterization and the first sampling. T3 and T4 levels were determined on the 0 and 120 min samples. TSH, T3 and T4 were measured by radioimmunoassay using commercially available kits for TSH (IRE, Belgium) and T3, T4 (Abbott). The results for TSH are expressed in pU/ml of the hTRH research standard A, kindly supplied by the MRC (London). The TSH response to TRH was calculated as the integrated secretion above baseline in pII/ min. The integrated secretion is the area between the secretory curve and the baseline drawn between the 0 and 120 min time. Statistical evaluation was made using Student’s t-test for paired and unpaired data. RESULTS
Thyrotrophin (TSH) before treatment There was no significant difference for basal TSH levels, T3 and T4 among any of the groups of patients or controls. The data on TSH response to TRH are illustrated in Figs. 1 and 2 and in Table 1. Before treatment, the TSH response to TRH was significantly lower in the unipolar and bipolar depressed patients (t = 3.08, P < 0.001 and t = 3.29, P < 0.001, respectively) when compared to the control group. When the unipolar and bipolar group were considered together (i.e., all affective illness), the TSH response was still significantly lower (t = 4.31, P < 0.001) than in controls. However, there was no difference between the unipolar and the bipolar group (t =
l BEFORE
0 AFTER
TREAWENT
30
60
Fig. 1. TSH response
I
120 to TRH
0
3 60 MINUTES
in unipolar
and bipolar
SEn
Bipolars
Unipolars
Controls
0
TREATbENT
1200 premenopausal
30
60 women
120 vs controls.
12
.9EFORE
Controls
20
E 3
=c
1
I SEII
0 AFTER TREATMENT
TREATMENT
Bipolars
Unipolars
10
=I b-
a
0
30
r
60
Fig. 2. TSH response
120
0
1
30
0
60
120
MINUTES
to TRH
in unipolar
and bipolar
postmenopausal
women
vs controls.
0.5,P = n.s.). When menopausal status was taken into account, the TSH response to TRH was significantly lower in the unipolar postmenopausal patients, than in postmenopausal controls (t = 2.8, P < 0.01). This disturbed pattern of response was not observed in the unipolar premenopausal patients, when compared to premenopausal controls (t = 1.58, P = n.s.). Among the bipolar group, an inverse pattern of response was observed for TSH. Only the premenopausal bipolar depressives had a lower TSH response to TRH as compared to premenopausal controls (t = 2.41, P < 0.025). The response of the bipolar postmenopausal group was TABLE
1
TSH RESPONSE ~.tU/120 MIN)
TO TRH
(INTEGRATED
Before Premenopausal subjects Controls (N = 31) Unipolars (N = 15) Bipolars (N = 5) Postmenopausal subjects Controls (N = 11) Unipolars (N = 12) Bipolars (N = 19) *P
< 0.025;
**P
< 0.05;
***P
SECRETION
treatment
ABOVE
After
BASELINE
treatment
784 f 104 532+88 166f 85 *
709 + 115 3129+914**
460 f 148 147 f 31*** 332f 97
340 + 39 983k735
< 0.01.
f SEM IN
13
similar to the one observed among the postmenopausal controls (t = 1.77, P = n.s.). TSH response
after treatment
In the bipolar premenopausal patients, the TSH response to TRH which was impaired before treatment is significantly improved, as shown in Table 1 (t = 2.55, P < 0.05). However, no significant difference in TSH response to TRH could be demonstrated after treatment in the unipolar postmenopausal patients (in whom the TSH response was blunted before treatment) (Table 1) (t = 2.14, P = n.s.). DISCUSSION
The data presented in this study demonstrate significant changes in TSH response to TRH in various clinical subgroups of depressive patients in comparison to controls matched for age, sex and menopausal status. Before treatment, the TSH response to TRH was significantly blunted in the unipolar postmenopausal patients when compared to postmenopausal controls and in the bipolar premenopausal patients when compared to premenopausal controls. This deficient endocrine response observed in some depressed patients does not seem to be related to age, severity of illness or previous drug intake. Some depressed patients show a normalisation of TSH response to TRH after recovery, while others do not as has been previously observed by Coppen et al. (1974) and Kierkegaard et al. (1975). The diagnostic and prognostic value of the TRH test in depression is still controversial, although unipolar and bipolar illnesses were not considered separately in most previous studies (Maeda et al. 1975; Prange 1976; Loosen et al. 1977). In one study, however, Gold et al. (1977) demonstrated a significant difference in TSH response to TRH in unipolar depressives when compared to either bipolar depressives or normal controls. Among the unipolars, the TSH response showed a significant negative correlation with the serotonin metabolite S-HIAA in the CSF. In the present study, the unipolar and bipolar depressive patients showed significant alterations in the TSH response to TRH before treatment when compared to normal control subjects. When menopausal status was taken into account, only postmenopausal unipolars showed a blunted TSH response to TRH, while in the bipolar group the opposite was true, i.e. bipolar premenopausal patients showed a reduced TSH response when corn-. pared to appropriate controls. After clinical recovery, striking changes were observed in the TSH response to TRH only in the bipolar group with a normalisation of the TSH response in bipolar premenopausal patients. This was not the case in the unipolar patients where the blunted TSH response was not corrected after antidepressant treatment and clinical recovery. This difference could con-
14
sequently account for previous reports by Kierkegaard et al. (1975) of a subgroup of depressive patients who do not correct their deficient TSH response after recovery and who subsequently show a higher rate of relapse in the 6-month period after discharge. Several factors could account for the blunting of the TSH response to TRH: prolonged fasting (Vinik et al. 1975), chronic renal failure (Czernichow et al. 1976), Klinefelter syndrome (Ozawa and Shishiba 1975), elevated cortisol (Otsuki et al. 1973; Re et al. 1976) and some pharmacologic agents such as aspirin (Ramey et al. 1976). Most of these factors could be controlled for in the present investigation; T3 and T4 were within the normal range for all the patients studied. Patients were off drugs for at least seven days before the TRH test and were kept on a free monoamine diet. Some earlier studies, however, have reported blunted TSH response to TRH in subgroups of depressed patients to be related to elevated cortisol production rate. Loosen et al. (1978) have shown an inverse relationship to be present between physiologic cortisol serum concentrations and TRHinduced TSH response in euthyroid normal volunteers and in unipolar depressed women. Previous reports by Otsuki et al. (1973) and Re et al. (1976) demonstrated that marked elevation of plasma corticosteroids in patients with Cushing’s syndrome or pharmacological administration of glucocorticosteroids to normals could influence the TSH response to TRH, findings consistent with the reported hypersecretion of cortisol in some depressive patients (Carroll and Mendels 1976). There is, however, some evidence against this hypothesis: in the alcohol withdrawal syndrome and in schizophrenia, there is a positive correlation between plasma cortisol and TSH peak after TRH injection (Loosen et al. 1978). Furthermore, TSH blunting to TRH has also been reported in depressive patients after recovery when the plasma cortisol had returned to normal (Carroll and Mendels 1976). These findings and ours suggest that the blunted TSH response to TRH in depression cannot solely be attributed to elevated cortisol levels nor to severity of depression. The role of the ovarian status in the alteration of the TSH response to TRH is less known; oestrogens seem to play only a minor role in the regulation of the TSH response in normal women (Rutlin et al. 1977). Menopausal status may account for the blunting of the TSH response to TRH in unipolar patients since it is only observed in postmenopausal unipolars, but the interpretation of the effect of the menopause in the bipolar group is inconclusive. The effects of tricyclic antidepressants on TSH secretion have been investigated by 2 groups. Linnoila et al. (1977) reported an increased plasma level of TSH after chronic administration of tricyclics. However, they measured only basal levels and did not standardize the duration of treatment, type or dose of tricyclic drugs. In a better designed study, Widerlov et al. (1978) found that tricyclic antidepressants did not affect the basal
15
or TRH stimulated levels of TSH. These data suggest that alterations in TSH response to TRH observed after recovery in some of our depressive patients are related to some feature of the depressive illness rather than being drugrelated effects. Indeed, only bipolar premenopausal patients showed statedependent changes in TSH response to TRH after treatment. The hypothesis of biological and endocrinological heterogeneity of depressive illness is given additional support by our results of differential TSH response after TRH in various clinical subgroups of depressive patients. The importance of the ovarian status in evaluating TSH response to TRH in depression has also been demonstrated for both bipolar and unipolar patients. Further research should aim at establishing a possible link between these neuroendocrine findings and monoamine as well as other neurochemical dysfunction in affective illness. REFERENCES Carroll, B.J. and Mendels, J. (1976) Neuroendocrine regulation in depression. In: E. Sachar (Ed.), Hormones, Behaviour and Psychopathology, Raven Press, New York, pp. 193-224. Coppen, A., Peet, M., Montgomery, S., Bailey, J., Marks, V. and Woods, P. (1974) Thyrotrophin releasing hormone in the treatment of depression, Lancet, i: 433-35. Czernichow, P., Dauzet, M.C., Broyer, M. and Rappoport, R. (1976) Abnormal TSH, PRL and GH response to TSH releasing factor in chronic renal failure, J. Cl. Endocr. Metab., 43: 630-37. Feighner, Y.P., Robbins, E., Guze, S.B., Woodruff, R.A., Winokur, G. and Munoz, R. (1972) Diagnostic criteria for use in psychiatric research, Arch. Gen. Psychiat., 26: 57-63. Gold, P.W., Goodwin, F.K., Wehr, T. and Rebar, R. (1978) TSH response to TRH in unipolar and bipolar affective illness - Relationship to spinal fluid amine metabolites, Amer. J. Psychiat., 134: 1028-32. Gregoire, F., Brauman, R. and Corvilain, J. (1977) Hormone release in depressed patients before and after recovery, Psychoneuroendocrinology, 2: 303-12. Kastin, A.J., Schalch, D.S., Ehrensing, R.H. and Anderson, M.S. (1972) Improvement in mental depression with decreased thyrotrophin response after administration of thyrotrophin releasing hormone, Lancet, ii: 740-42. Kierkegaard, D., Norlem, N., Lauridsen, V., Bjorum, N. and Christiansen, C. (1975) Protirelin stimulation test and thyroid function during treatment of depression, Arch. Gen. Psychiat., 32: 1115-18. Linnoila, M., Leppaluoto, J., Seppala, T. and Rauta, T. (1977) Serum gonadotrophin and TSH levels after tricyclic antidepressants in healthy males, Acta Pharmacol. (Kbh), 41: 285-88, Loosen, P.T., Prange, A.J., Wilson, I.C., Lara, P.P. and Pettus, C. (1977) Thyroid stimulating hormone response after thyrotrophin releasing hormone in depressed, schizophrenic and normal women, Psychoneuroendocrinology, 2: 137-48. Loosen, P.T., Prange, A.J. and Wilson, I.C. (1978) The thyrotrophin response to thyrotrophin releasing hormone in psychiatric patients - Relation to serum cortisol, Progr. Neuropsychopharmacol., 2: 479-86. Maeda, K., Kato, Y., Oago, S., Chihara, K., Yoshimato, Y., Yamagushi, N., Kuromaru, S. and Imura, A. (1975) Growth hormone and prolactin release after injection of thyrotrophin releasing hormones in patients with depression, J. Clin. Endocr. Metab., 40: 501-5.
16 Otsuki, M., Dakoda, M. and Baba, S. (1973) Influence of glucocorticoids on TRH mediated TSH response in man, J. Clin. Endocr. Metab., 36: 95-102. Ozawa, Y. and Shishiba, Y. (1975) Lack of TRH induced TSH secretion in a patient with Klinefelter syndrome - A case report, Endocr. Jap., 22: 269-273. Prange, Jr., A.J. (1977) Patterns of pituitary responses to thyrotrophin releasing hormone in depressed patients - A review. In: W.E. Fann, I. Karacan, A. Pikorny and R.L. Williams (Eds.), Phenomenology and Treatment of Depression, Spectrum, New York, pp. l-15. Prange, Jr., A.J., Wilson, I.C., Lara, P.P., Allsop, L.P. and Bruse, G.R. (1972) Effect of thyrotrophin releasing hormone in depression, Lancet, ii: 999-1002. Ramey, J.N., Burrow, G.N. and Spaulding, S. (1976) The effect of aspirine and indomethacin on the TRH response to synthetic thyrotrophin releasing hormone, J. Clin. Endocr. Metab., 40: 509-511. Re, R.N., Kourides, E.C., Ridgway, B.D., Weintraub, B.D. and Maloff, F. (1976) The effect of glucocorticoid administration of human pituitary secretion of thyrotrophin and prolactin, J. Clin. Endocr. Metab., 43: 338-46. Rutlin, F., Haug, E. and Torjesen, P.A. (1977) Serum thyrotrophin, prolactin and growth hormone response to TRH during oestrogen treatment, Acta Endocr., 84: 23-35. Vinik, A.I., Kalk, W.J., McLaren, H., Hendricks, S. and Pimstone, B.L. (1975) Fasting blunts the TSH response to synthetic thyrotrophin releasing hormone (TRH), J. Clin. Endocr. Metab., 40: 509-11. Widerlow, E., Wide, L. and Sjostrom, R. (1978) Effect of tricyclic antidepressants on human plasma levels of TSH, GH and Prolactin, Acta Psychiat. Stand., 59: 449-57.
of Affective
3 (1981) 9-16 Biomedical Press
Disorders,
0 Elsevier/North-Holland
9
THYROTROPHIN RESPONSE TO THYROTROPHIN-RELEASING HORMONE IN UNIPOLAR AND BIPOLAR AFFECTIVE ILLNESS
1 P. LINKOWSKI, 2 H. BRAUMAN and ’ J. MENDLEWICZ i Department of Psychiatry, Hospital Erasme, Free University of Brussels and 2 Radioimmunology Unit, Laboratory of Clinical Biology, Hospital Brugmann, Free University Brussels, Brussels (Belgium)
of
(Received 13 May, 1980) (Revised, received 12 August, 1980) (Accepted 14 August, 1980)
SUMMARY
The plasma levels of thyrotrophin (TSH) and the response of this hormone to 200 pg of thyrotrophin-releasing hormone (TRH) were studied in 27 unipolar and 24 bipolar depressive patients before and after amitriptyline treatment as well as in 42 normal controls. There was no significant difference in basal TSH levels between any of the groups studied according to diagnosis, menopausal status and antidepressant treatment. Before treatment, the TSH response to TRH was significantly lower in the unipolar and bipolar depressive patients than in normal controls. The TSH response to TRH did not differ significantly between the unipolar and bipolar depressives. When menopausal status was taken into account, the TSH response to TRH was significantly blunted in the unipolar postmenopausal patients when compared to postmenopausal controls. In the bipolar group, the premenopausal depressive patients had a significantly lower TSH response to TRH than premenopausal controls. After amitriptyline treatment, the TSH response to TRH, which was impaired before treatment in the bipolar premenopausal patients, improved significantly. However, no significant difference in the TSH response to TRH could be demonstrated after treatment in the unipolar postmenopausal patients in whom the TSH response was blunted before treatment. Our findings suggest that the differential TSH response to TRH in unipolar and bipolar patients may constitute biological markers of endocrine dysfunction in clinical subgroups of affective disorders.
This study has been supported by grants from the University of Brussels, the fondation De Cooman and the Association for Mental Health Research. Correspondence to: J. Mendlewicz, M.D., Ph.D., Professor and Head, Department of Psychiatry, Hospital Erasme, Free University of Brussels, 808 Route de Lennick, 1070 Brussels, Belgium.
10 INTRODUCTION
In early studies of depressed patients, Kastin et al. (1972) and Prange et al. (1972) demonstrated an impaired response of serum thyrotrophin (TSH) to thyrotrophin-releasing hormone (TRH), administered intravenously. Similar findings have been observed by several authors (Coppen et al. 1974; Kierkegaard et al. 1975; Gregoire et al. 1976; Loosen et al. 1977). Kierkegaard et al. (1975) reported the mean maximal TSH response to TRH (Amax) to be normalized in some patients after recovery but these data have been challenged by other authors (Maeda et al. 1975; Loosen et al. 1977) who were unable to confirm these results. These conflicting observations could be due to differences in sampling procedures and/or to the clinical, biochemical and genetic heterogeneity of depressive disorders. Gold et al. (1977) showed in patients with unipolar illness, a significant negative correlation between the blunting of the TSH response to TRH and the main serotonin CSF metabolite, 5-hydroxyindolacetic acid (5-HIAA) while this was not the case in bipolar patients and in normal controls. The present study was undertaken to investigate the pattern of secretion of a pituitary hormone, thyrotrophin-stimulating hormone (TSH) in response to stimulation by thyrotrophin-releasing hormone (TRH) in 51 depressive patients and 42 normal controls matched for age, sex and menopausal status. The purpose of the study was to determine if disturbances in neuroendocrine secretion could be used as biological markers in well defined subgroups of affectively ill patients, and to explore the possible influence of the menopause on TSH response to TRH in affective illness. METHODS
Female patients, 20 premenopausal (15 unipolar (depressive) and 5 bipolar (manic-depressive), aged 20-50, mean 37) and 31 postmenopausal (12 unipolar and 19 bipolar, aged 50-60, mean 55) were studied during a depressive episode severe enough to warrant hospitalisation. Psychiatric diagnoses were made according to the criteria of Feighner (Feighner et al. 1972). The normal controls were 31 premenopausal (mean 35, range 20-50) and 11 postmenopausal women (mean 53, range 50-60). All subjects were in good physical health without clinical and laboratory evidence of any somatic disease. The premenopausal patients and controls were tested at random during the menstrual cycle. All patients were tested before and some also after treatment which consisted of the administration of amitriptyline ranging from 75 to 300 mg/day. Patients were retested after recovery and were still on antidepressant treatment. Informed consent was obtained from patients and controls. The TRH test was performed at 9 a.m. in a recumbent position. All patients were on a monoamine-free diet and off psychotropic drugs for at least 7 days. Synthetic TRH (Roche, 200 pg) was administered intravenously
11
at time 0. Blood was sampled through an indwelling catheter at -15,0,20, 30, 45, 60, 120 min after TRH injection for determination of serum TSH. 30-60 min were allowed between catheterization and the first sampling. T3 and T4 levels were determined on the 0 and 120 min samples. TSH, T3 and T4 were measured by radioimmunoassay using commercially available kits for TSH (IRE, Belgium) and T3, T4 (Abbott). The results for TSH are expressed in pU/ml of the hTRH research standard A, kindly supplied by the MRC (London). The TSH response to TRH was calculated as the integrated secretion above baseline in pII/ min. The integrated secretion is the area between the secretory curve and the baseline drawn between the 0 and 120 min time. Statistical evaluation was made using Student’s t-test for paired and unpaired data. RESULTS
Thyrotrophin (TSH) before treatment There was no significant difference for basal TSH levels, T3 and T4 among any of the groups of patients or controls. The data on TSH response to TRH are illustrated in Figs. 1 and 2 and in Table 1. Before treatment, the TSH response to TRH was significantly lower in the unipolar and bipolar depressed patients (t = 3.08, P < 0.001 and t = 3.29, P < 0.001, respectively) when compared to the control group. When the unipolar and bipolar group were considered together (i.e., all affective illness), the TSH response was still significantly lower (t = 4.31, P < 0.001) than in controls. However, there was no difference between the unipolar and the bipolar group (t =
l BEFORE
0 AFTER
TREAWENT
30
60
Fig. 1. TSH response
I
120 to TRH
0
3 60 MINUTES
in unipolar
and bipolar
SEn
Bipolars
Unipolars
Controls
0
TREATbENT
1200 premenopausal
30
60 women
120 vs controls.
12
.9EFORE
Controls
20
E 3
=c
1
I SEII
0 AFTER TREATMENT
TREATMENT
Bipolars
Unipolars
10
=I b-
a
0
30
r
60
Fig. 2. TSH response
120
0
1
30
0
60
120
MINUTES
to TRH
in unipolar
and bipolar
postmenopausal
women
vs controls.
0.5,P = n.s.). When menopausal status was taken into account, the TSH response to TRH was significantly lower in the unipolar postmenopausal patients, than in postmenopausal controls (t = 2.8, P < 0.01). This disturbed pattern of response was not observed in the unipolar premenopausal patients, when compared to premenopausal controls (t = 1.58, P = n.s.). Among the bipolar group, an inverse pattern of response was observed for TSH. Only the premenopausal bipolar depressives had a lower TSH response to TRH as compared to premenopausal controls (t = 2.41, P < 0.025). The response of the bipolar postmenopausal group was TABLE
1
TSH RESPONSE ~.tU/120 MIN)
TO TRH
(INTEGRATED
Before Premenopausal subjects Controls (N = 31) Unipolars (N = 15) Bipolars (N = 5) Postmenopausal subjects Controls (N = 11) Unipolars (N = 12) Bipolars (N = 19) *P
< 0.025;
**P
< 0.05;
***P
SECRETION
treatment
ABOVE
After
BASELINE
treatment
784 f 104 532+88 166f 85 *
709 + 115 3129+914**
460 f 148 147 f 31*** 332f 97
340 + 39 983k735
< 0.01.
f SEM IN
13
similar to the one observed among the postmenopausal controls (t = 1.77, P = n.s.). TSH response
after treatment
In the bipolar premenopausal patients, the TSH response to TRH which was impaired before treatment is significantly improved, as shown in Table 1 (t = 2.55, P < 0.05). However, no significant difference in TSH response to TRH could be demonstrated after treatment in the unipolar postmenopausal patients (in whom the TSH response was blunted before treatment) (Table 1) (t = 2.14, P = n.s.). DISCUSSION
The data presented in this study demonstrate significant changes in TSH response to TRH in various clinical subgroups of depressive patients in comparison to controls matched for age, sex and menopausal status. Before treatment, the TSH response to TRH was significantly blunted in the unipolar postmenopausal patients when compared to postmenopausal controls and in the bipolar premenopausal patients when compared to premenopausal controls. This deficient endocrine response observed in some depressed patients does not seem to be related to age, severity of illness or previous drug intake. Some depressed patients show a normalisation of TSH response to TRH after recovery, while others do not as has been previously observed by Coppen et al. (1974) and Kierkegaard et al. (1975). The diagnostic and prognostic value of the TRH test in depression is still controversial, although unipolar and bipolar illnesses were not considered separately in most previous studies (Maeda et al. 1975; Prange 1976; Loosen et al. 1977). In one study, however, Gold et al. (1977) demonstrated a significant difference in TSH response to TRH in unipolar depressives when compared to either bipolar depressives or normal controls. Among the unipolars, the TSH response showed a significant negative correlation with the serotonin metabolite S-HIAA in the CSF. In the present study, the unipolar and bipolar depressive patients showed significant alterations in the TSH response to TRH before treatment when compared to normal control subjects. When menopausal status was taken into account, only postmenopausal unipolars showed a blunted TSH response to TRH, while in the bipolar group the opposite was true, i.e. bipolar premenopausal patients showed a reduced TSH response when corn-. pared to appropriate controls. After clinical recovery, striking changes were observed in the TSH response to TRH only in the bipolar group with a normalisation of the TSH response in bipolar premenopausal patients. This was not the case in the unipolar patients where the blunted TSH response was not corrected after antidepressant treatment and clinical recovery. This difference could con-
14
sequently account for previous reports by Kierkegaard et al. (1975) of a subgroup of depressive patients who do not correct their deficient TSH response after recovery and who subsequently show a higher rate of relapse in the 6-month period after discharge. Several factors could account for the blunting of the TSH response to TRH: prolonged fasting (Vinik et al. 1975), chronic renal failure (Czernichow et al. 1976), Klinefelter syndrome (Ozawa and Shishiba 1975), elevated cortisol (Otsuki et al. 1973; Re et al. 1976) and some pharmacologic agents such as aspirin (Ramey et al. 1976). Most of these factors could be controlled for in the present investigation; T3 and T4 were within the normal range for all the patients studied. Patients were off drugs for at least seven days before the TRH test and were kept on a free monoamine diet. Some earlier studies, however, have reported blunted TSH response to TRH in subgroups of depressed patients to be related to elevated cortisol production rate. Loosen et al. (1978) have shown an inverse relationship to be present between physiologic cortisol serum concentrations and TRHinduced TSH response in euthyroid normal volunteers and in unipolar depressed women. Previous reports by Otsuki et al. (1973) and Re et al. (1976) demonstrated that marked elevation of plasma corticosteroids in patients with Cushing’s syndrome or pharmacological administration of glucocorticosteroids to normals could influence the TSH response to TRH, findings consistent with the reported hypersecretion of cortisol in some depressive patients (Carroll and Mendels 1976). There is, however, some evidence against this hypothesis: in the alcohol withdrawal syndrome and in schizophrenia, there is a positive correlation between plasma cortisol and TSH peak after TRH injection (Loosen et al. 1978). Furthermore, TSH blunting to TRH has also been reported in depressive patients after recovery when the plasma cortisol had returned to normal (Carroll and Mendels 1976). These findings and ours suggest that the blunted TSH response to TRH in depression cannot solely be attributed to elevated cortisol levels nor to severity of depression. The role of the ovarian status in the alteration of the TSH response to TRH is less known; oestrogens seem to play only a minor role in the regulation of the TSH response in normal women (Rutlin et al. 1977). Menopausal status may account for the blunting of the TSH response to TRH in unipolar patients since it is only observed in postmenopausal unipolars, but the interpretation of the effect of the menopause in the bipolar group is inconclusive. The effects of tricyclic antidepressants on TSH secretion have been investigated by 2 groups. Linnoila et al. (1977) reported an increased plasma level of TSH after chronic administration of tricyclics. However, they measured only basal levels and did not standardize the duration of treatment, type or dose of tricyclic drugs. In a better designed study, Widerlov et al. (1978) found that tricyclic antidepressants did not affect the basal
15
or TRH stimulated levels of TSH. These data suggest that alterations in TSH response to TRH observed after recovery in some of our depressive patients are related to some feature of the depressive illness rather than being drugrelated effects. Indeed, only bipolar premenopausal patients showed statedependent changes in TSH response to TRH after treatment. The hypothesis of biological and endocrinological heterogeneity of depressive illness is given additional support by our results of differential TSH response after TRH in various clinical subgroups of depressive patients. The importance of the ovarian status in evaluating TSH response to TRH in depression has also been demonstrated for both bipolar and unipolar patients. Further research should aim at establishing a possible link between these neuroendocrine findings and monoamine as well as other neurochemical dysfunction in affective illness. REFERENCES Carroll, B.J. and Mendels, J. (1976) Neuroendocrine regulation in depression. In: E. Sachar (Ed.), Hormones, Behaviour and Psychopathology, Raven Press, New York, pp. 193-224. Coppen, A., Peet, M., Montgomery, S., Bailey, J., Marks, V. and Woods, P. (1974) Thyrotrophin releasing hormone in the treatment of depression, Lancet, i: 433-35. Czernichow, P., Dauzet, M.C., Broyer, M. and Rappoport, R. (1976) Abnormal TSH, PRL and GH response to TSH releasing factor in chronic renal failure, J. Cl. Endocr. Metab., 43: 630-37. Feighner, Y.P., Robbins, E., Guze, S.B., Woodruff, R.A., Winokur, G. and Munoz, R. (1972) Diagnostic criteria for use in psychiatric research, Arch. Gen. Psychiat., 26: 57-63. Gold, P.W., Goodwin, F.K., Wehr, T. and Rebar, R. (1978) TSH response to TRH in unipolar and bipolar affective illness - Relationship to spinal fluid amine metabolites, Amer. J. Psychiat., 134: 1028-32. Gregoire, F., Brauman, R. and Corvilain, J. (1977) Hormone release in depressed patients before and after recovery, Psychoneuroendocrinology, 2: 303-12. Kastin, A.J., Schalch, D.S., Ehrensing, R.H. and Anderson, M.S. (1972) Improvement in mental depression with decreased thyrotrophin response after administration of thyrotrophin releasing hormone, Lancet, ii: 740-42. Kierkegaard, D., Norlem, N., Lauridsen, V., Bjorum, N. and Christiansen, C. (1975) Protirelin stimulation test and thyroid function during treatment of depression, Arch. Gen. Psychiat., 32: 1115-18. Linnoila, M., Leppaluoto, J., Seppala, T. and Rauta, T. (1977) Serum gonadotrophin and TSH levels after tricyclic antidepressants in healthy males, Acta Pharmacol. (Kbh), 41: 285-88, Loosen, P.T., Prange, A.J., Wilson, I.C., Lara, P.P. and Pettus, C. (1977) Thyroid stimulating hormone response after thyrotrophin releasing hormone in depressed, schizophrenic and normal women, Psychoneuroendocrinology, 2: 137-48. Loosen, P.T., Prange, A.J. and Wilson, I.C. (1978) The thyrotrophin response to thyrotrophin releasing hormone in psychiatric patients - Relation to serum cortisol, Progr. Neuropsychopharmacol., 2: 479-86. Maeda, K., Kato, Y., Oago, S., Chihara, K., Yoshimato, Y., Yamagushi, N., Kuromaru, S. and Imura, A. (1975) Growth hormone and prolactin release after injection of thyrotrophin releasing hormones in patients with depression, J. Clin. Endocr. Metab., 40: 501-5.
16 Otsuki, M., Dakoda, M. and Baba, S. (1973) Influence of glucocorticoids on TRH mediated TSH response in man, J. Clin. Endocr. Metab., 36: 95-102. Ozawa, Y. and Shishiba, Y. (1975) Lack of TRH induced TSH secretion in a patient with Klinefelter syndrome - A case report, Endocr. Jap., 22: 269-273. Prange, Jr., A.J. (1977) Patterns of pituitary responses to thyrotrophin releasing hormone in depressed patients - A review. In: W.E. Fann, I. Karacan, A. Pikorny and R.L. Williams (Eds.), Phenomenology and Treatment of Depression, Spectrum, New York, pp. l-15. Prange, Jr., A.J., Wilson, I.C., Lara, P.P., Allsop, L.P. and Bruse, G.R. (1972) Effect of thyrotrophin releasing hormone in depression, Lancet, ii: 999-1002. Ramey, J.N., Burrow, G.N. and Spaulding, S. (1976) The effect of aspirine and indomethacin on the TRH response to synthetic thyrotrophin releasing hormone, J. Clin. Endocr. Metab., 40: 509-511. Re, R.N., Kourides, E.C., Ridgway, B.D., Weintraub, B.D. and Maloff, F. (1976) The effect of glucocorticoid administration of human pituitary secretion of thyrotrophin and prolactin, J. Clin. Endocr. Metab., 43: 338-46. Rutlin, F., Haug, E. and Torjesen, P.A. (1977) Serum thyrotrophin, prolactin and growth hormone response to TRH during oestrogen treatment, Acta Endocr., 84: 23-35. Vinik, A.I., Kalk, W.J., McLaren, H., Hendricks, S. and Pimstone, B.L. (1975) Fasting blunts the TSH response to synthetic thyrotrophin releasing hormone (TRH), J. Clin. Endocr. Metab., 40: 509-11. Widerlow, E., Wide, L. and Sjostrom, R. (1978) Effect of tricyclic antidepressants on human plasma levels of TSH, GH and Prolactin, Acta Psychiat. Stand., 59: 449-57.