J. Endocrinol. Invest. 15: 437-441, 1992
Acute and chronic effects of high glucocorticoid levels on hypothalamic-pituitary-thyroid axis in man D. Rubello*, N. Sonino*, D. Casara**, M.E. Girelli*, B.Busnardo*, and M. Boscaro* *Istituto di Semeiotica Medica, Universita di Padova, **Divisione di Radioterapia e Medicina Nucleare, Ospedale Civile di Padova, Padova, Italy ABSTRACT. It is known that glucocorticoids can influence anterior pituitary hormones other than ACTH. Their effects on the hypothalamic-pituitarythyroid axis are controversial. To further investigate this issue, the acute and chronic effects of high plasma cortisol levels on TSH secretion were evaluated in 20 normal subjects and in 14 patients with Cushing's syndrome. In normals, high plasma cortisol levels were obtained by giving ACTH 250 I1g or CRH 100 I1g iv as a bolus or by an hydrocortisone 500 mg infusion over 1 h. Acute cortisol increase produced no effect both on basal and TRHstimulated TSH secretion. In patients with Cushing's syndrome, basal TSH levels, low or suppressed,
showed an impaired response to TRH, inversely correlated with urinary cortisol values. After successful surgery, TSH and its response to TRH became normal in concomitance with the normalization of plasma and urinary cortisol levels. Our data show the lack of an acute inhibitory effect of high plasma cortisol levels on TRH-TSH axis. However, after long-term exposure to high plasma cortisollevels, i.e. Cushing's syndrome, inhibition of both basal and TRH-stimulated TSH secretion was demonstrated. These findings indicate that only prolonged hypercortisolism does interfere with pituitary TSH secretion. The underlying mechanisms are still unclear.
INTRODUCTION
MATERIALS AND METHODS Patients The acute and chronic effects of high cortisol levels on TSH secretion were tested in normal subjects and in a group of patients with Cushing's syndrome. None of these patients had other pituitary abnormalities, goiter, circulating antithyroid antibodies [anti microsomal and antithyroglobulin], systemic or metabolic disorders. i) Cushing's syndrome: 14 untreated patients were investigated; 8 with a pituitary microadenoma, 6 with an adrenocortical adenoma; 3 males, 11 females, age range 16-45 yr, mean 35.4. The differential diagnosis was established by conventional hormonal tests (dexamethasone suppression, CRH stimulation, etc) and morphological evaluation (pituitary and adrenal CT-scans, adrenal scintiscan). All patients were cured by surgical removal of a pituitary or adrenal adenoma. After surgery, hormonal assessment confirmed their complete recovery. Before and 3, 6 and 12 months after treatment, baseline and TRH stimulated TSH levels were carried out. ii) Normal subjects: In 20 normal subjects, (similar age range), the effect of an acute increase of glucocorticoid levels on basal TSH (15 cases) or on TRH-stimulated TSH (5 cases) was evaluated as follows. Glucocorticoid increase was obtained by giving CRH (oCRH UCB, Belgium) 100 I1g iv as a bolus in
Glucocorticoids can interfere with the secretion of anterior pituitary hormones other than ACTH (1,2). Their effects on hypothalamic-pituitary-thyroid axis are still controversial. TSH response to TRH has been reported to be reduced (3-7) or unchanged (7-10) by high plasma glucocorticoid levels in normal subjects. Contradictory results about an inhibitory effect of exogenous glucocorticoids on basal TSH were reported both in patients with primary hypothyroidism (4,11-13) and in normal subjects in whom TSH was measured by RIA methods with low sensitivity (2-4, 6, 8, 12, 13). Similarly, different responses of pituitary-thyroid axis have been observed also in chronic endogenous hypercortisolism (14-18). The recent introduction of a specific and more sensitive immunoradiometric (IRMA) assay, by which minimal changes of TSH levels can be detected, allowed us to further investigate the influence of acute and chronic increase of glucocorticoid levels on TSH secretion in man.
Key·words: Glucocorticoids, TSH, thyroid function. Correspondence: Domenico Rubollo MD .. Istituto di Semeiotica Medica. Universita di Padova. via Ospedale 105, 35100 Padova. Italy.
Received June 20, 1991 . accepted April 23, 1992.
437
0, Rube//o, N. Sonino, D. Casara, et al. D. Rubel/a, O. Casara,
Statistical analysis SE,. TSH variaThe data are expressed as mean ± SE tions after CRH CRH,, ACTH, TRH and hydrocortisone were compared with those obtained after placebo by repeated measures analysis of variance (ANOVA) using treatment as grouping factor. factor, In comparing patients with Cushing's syndrome and normal subjects group ttest was employed, employed. p values 0,05 were considered significant. < 0.05
5 subjects, ACTH (Synacthen, Ciba Geigy, Italy) 250 fl9 iv as a bolus in 5 subjects, hydrocortisone 250119 (Flebocortid, Lepetit Lepetit,, France) 500 mg iv infusion subjects. In all these 15 subjects, over 60 min in 5 subjects, blood samples for basal TSH, ACTH and plasma cortisol were drawn at -15, 0, 30, 60, 90, 120, 150, min, Two days earlier, saline was injected or 180 min. infused as control and blood samples were obtained at -15,0,30,60,90,120,150, -15 , 0, 30, 60, 90, 120, 150, 180 min, min. In the other 5 subjects, TRH test was performed 1 h after hydrocortisone 500 mg infusion over 1 h and the data were compared with those obtained two days earlier after saline infusion. infusion, TRH test was performed by giving TRH (Biodata (Biodata,, Italy) 200 Ilg fl9 iv as a bolus; blood samples for TSH were drawn at 0, 20, 60 min, min. All studies were started at 09:00 h, h.
RESULTS Cushing's syndrome In patients with Cushing's syndrome serum T3 was 7,2 ng/dl vs 144.0 144,0 ± 8.4 (105,2 ± 7.2 lower than controls (105.2 0.01), and significantly increased 3 ng/dl, p < 0,01), months after surgery compared to pretreatment levels (137,7 9.3 ng/dl, p < 0,01), 0.01). T4 and FT4 were (137.7 ± 9,3 (7,5 ± 1.3 1,3 Ilg/dl fl9/dl vs similar to those found in normals (7.5 7,9 ± 0.9 0,9Ilg/dl 0,3 ng/dl vs 1,6 0,4 ng/dl, fl9/dl and 1,5 1.5 ± 0.3 1.6 ± 0.4 7.9 respectively) respectively),, and remained unchanged after treatment. Mean basal TSH levels were lower than nor0.19 IlU/ml flU/ml vs 1,36 1.36 ± 0.20 flU/ml at 0 mal (0.29 ± 0,19 0,20 IlU/ml min, p < 0.G1) 0.Q1) and their change after TRH was bluntflU/ml vs 12,54 12.54 ± 1.82 flU/ml at 20 ed (3.23 (3,23 ± 0.91 0,91 IlU/ml 1,82 IlU/ml min,, p < 0.01) 0,01) (Fig. (Fig, 1). 1) There was an inverse corremin lation between both basal and TRH stimulated TSH and urinary free cortisol levels (r == 0,93, 0.93, P < 0,01, 0.01, and r = 0.90, 0,90, P < 001, (Figs, 2 and 3). 3), 0.01, respectively) (Figs. No correlation between TSH and plasma cortisol at 08:00 and 20:00 h was observed (data not shown). shown), surgery,, in concomitance with Three months after surgery the normalization of plasma and urinary cortisol values, basal TSH and its response to TRH significant0,01) (Fig. (Fig, 1), and did not shown ly increased (p < 0.01) months, any further variation at 6 and 12 months.
Hormone assays TSH was assayed by an IRMA method (CIS, France); normal range 0.2-4 flU/ml. The intraassay 0,2-4 IlU/ml, coefficient of variation (CV) was 2,1% 2.1 % for normal TSH values (mean TSH = = 2.4 IlU/ml) flU/ml) and 3.5% 3,5% for 0,3 IlU/ml); low values (mean TSH = 0.3 flU/ml); interassay CV 2,7% for normal TSH values (mean TSH = 2.4 was 2.7% IlU/ml) 0,3 flU/ml) and 7% for low values (mean TSH = 0.3 flU/ml). Total thyroxine (T4) was assayed by fluoIlU/ml), rescence polarization immunoassay (TOx-Abbott, (TDx-Abbott, USA), normal range 4.5-12 4,5-12 Ilg/dl; fl9/dl; free thyroxine (FT4) by RIA (Biorad, Italy), normal range 0.8-2.3 0,8-2,3 ng/dl; total triiodothyronine (T3) by RIA (Mallinckrodt, West Germany), normal range 80-200 ng/dl; ACTH by IRMA (Euro-Diagnostic, (Euro-Oiagnostic, Holland), normal range 20-80 pg/ml); plasma and urinary free cortisol by RIA (OPC, fl9/dl (DPC, West Germany), normal range 5-20 Ilg/dl and 20-120 Ilg/24 respectively, fl9/24 h, respectively.
TSR
TSH
TSH UU/l1i1
TSH uUfmJ uU/ml :EO
:W
~tO
BEFORE TREATMENT
][5
15
15 1S
10 10
10 18
'I ~: c:::::: -.--
CL. (I..
00
20 28
AFI'ER TREATMENT
5
""-
~I
3 MONTIIS 3 MONTHS AFTER TREATMENT
u U/tnl 20
BEFORE TREATMENT
:-.'."--~ ..,
6O '0
min
:1 0
00
20 10
Fig. 1 - Basal and TRH-stimulated TRH-stim ulated TSH levels in Cushing 's syndrome before and after treatment. Cushing's
438
i
60 60
min in
Effects of g/ucocorticoids on pituitary-thyroid axis
..c ..e ~ M -.:r01)
----
M :::l
= OJ)
...:l 0 ~
000 00.
E-o
.c
..,. .,.-=-.. M
N
1000
r=-0.93
1000
~ ~ ~
Acute and chronic effects of high glucocorticoid levels on hypothalamic-pituitary-thyroid axis in man D. Rubello*, N. Sonino*, D. Casara**, M.E. Girelli*, B.Busnardo*, and M. Boscaro* *Istituto di Semeiotica Medica, Universita di Padova, **Divisione di Radioterapia e Medicina Nucleare, Ospedale Civile di Padova, Padova, Italy ABSTRACT. It is known that glucocorticoids can influence anterior pituitary hormones other than ACTH. Their effects on the hypothalamic-pituitarythyroid axis are controversial. To further investigate this issue, the acute and chronic effects of high plasma cortisol levels on TSH secretion were evaluated in 20 normal subjects and in 14 patients with Cushing's syndrome. In normals, high plasma cortisol levels were obtained by giving ACTH 250 I1g or CRH 100 I1g iv as a bolus or by an hydrocortisone 500 mg infusion over 1 h. Acute cortisol increase produced no effect both on basal and TRHstimulated TSH secretion. In patients with Cushing's syndrome, basal TSH levels, low or suppressed,
showed an impaired response to TRH, inversely correlated with urinary cortisol values. After successful surgery, TSH and its response to TRH became normal in concomitance with the normalization of plasma and urinary cortisol levels. Our data show the lack of an acute inhibitory effect of high plasma cortisol levels on TRH-TSH axis. However, after long-term exposure to high plasma cortisollevels, i.e. Cushing's syndrome, inhibition of both basal and TRH-stimulated TSH secretion was demonstrated. These findings indicate that only prolonged hypercortisolism does interfere with pituitary TSH secretion. The underlying mechanisms are still unclear.
INTRODUCTION
MATERIALS AND METHODS Patients The acute and chronic effects of high cortisol levels on TSH secretion were tested in normal subjects and in a group of patients with Cushing's syndrome. None of these patients had other pituitary abnormalities, goiter, circulating antithyroid antibodies [anti microsomal and antithyroglobulin], systemic or metabolic disorders. i) Cushing's syndrome: 14 untreated patients were investigated; 8 with a pituitary microadenoma, 6 with an adrenocortical adenoma; 3 males, 11 females, age range 16-45 yr, mean 35.4. The differential diagnosis was established by conventional hormonal tests (dexamethasone suppression, CRH stimulation, etc) and morphological evaluation (pituitary and adrenal CT-scans, adrenal scintiscan). All patients were cured by surgical removal of a pituitary or adrenal adenoma. After surgery, hormonal assessment confirmed their complete recovery. Before and 3, 6 and 12 months after treatment, baseline and TRH stimulated TSH levels were carried out. ii) Normal subjects: In 20 normal subjects, (similar age range), the effect of an acute increase of glucocorticoid levels on basal TSH (15 cases) or on TRH-stimulated TSH (5 cases) was evaluated as follows. Glucocorticoid increase was obtained by giving CRH (oCRH UCB, Belgium) 100 I1g iv as a bolus in
Glucocorticoids can interfere with the secretion of anterior pituitary hormones other than ACTH (1,2). Their effects on hypothalamic-pituitary-thyroid axis are still controversial. TSH response to TRH has been reported to be reduced (3-7) or unchanged (7-10) by high plasma glucocorticoid levels in normal subjects. Contradictory results about an inhibitory effect of exogenous glucocorticoids on basal TSH were reported both in patients with primary hypothyroidism (4,11-13) and in normal subjects in whom TSH was measured by RIA methods with low sensitivity (2-4, 6, 8, 12, 13). Similarly, different responses of pituitary-thyroid axis have been observed also in chronic endogenous hypercortisolism (14-18). The recent introduction of a specific and more sensitive immunoradiometric (IRMA) assay, by which minimal changes of TSH levels can be detected, allowed us to further investigate the influence of acute and chronic increase of glucocorticoid levels on TSH secretion in man.
Key·words: Glucocorticoids, TSH, thyroid function. Correspondence: Domenico Rubollo MD .. Istituto di Semeiotica Medica. Universita di Padova. via Ospedale 105, 35100 Padova. Italy.
Received June 20, 1991 . accepted April 23, 1992.
437
0, Rube//o, N. Sonino, D. Casara, et al. D. Rubel/a, O. Casara,
Statistical analysis SE,. TSH variaThe data are expressed as mean ± SE tions after CRH CRH,, ACTH, TRH and hydrocortisone were compared with those obtained after placebo by repeated measures analysis of variance (ANOVA) using treatment as grouping factor. factor, In comparing patients with Cushing's syndrome and normal subjects group ttest was employed, employed. p values 0,05 were considered significant. < 0.05
5 subjects, ACTH (Synacthen, Ciba Geigy, Italy) 250 fl9 iv as a bolus in 5 subjects, hydrocortisone 250119 (Flebocortid, Lepetit Lepetit,, France) 500 mg iv infusion subjects. In all these 15 subjects, over 60 min in 5 subjects, blood samples for basal TSH, ACTH and plasma cortisol were drawn at -15, 0, 30, 60, 90, 120, 150, min, Two days earlier, saline was injected or 180 min. infused as control and blood samples were obtained at -15,0,30,60,90,120,150, -15 , 0, 30, 60, 90, 120, 150, 180 min, min. In the other 5 subjects, TRH test was performed 1 h after hydrocortisone 500 mg infusion over 1 h and the data were compared with those obtained two days earlier after saline infusion. infusion, TRH test was performed by giving TRH (Biodata (Biodata,, Italy) 200 Ilg fl9 iv as a bolus; blood samples for TSH were drawn at 0, 20, 60 min, min. All studies were started at 09:00 h, h.
RESULTS Cushing's syndrome In patients with Cushing's syndrome serum T3 was 7,2 ng/dl vs 144.0 144,0 ± 8.4 (105,2 ± 7.2 lower than controls (105.2 0.01), and significantly increased 3 ng/dl, p < 0,01), months after surgery compared to pretreatment levels (137,7 9.3 ng/dl, p < 0,01), 0.01). T4 and FT4 were (137.7 ± 9,3 (7,5 ± 1.3 1,3 Ilg/dl fl9/dl vs similar to those found in normals (7.5 7,9 ± 0.9 0,9Ilg/dl 0,3 ng/dl vs 1,6 0,4 ng/dl, fl9/dl and 1,5 1.5 ± 0.3 1.6 ± 0.4 7.9 respectively) respectively),, and remained unchanged after treatment. Mean basal TSH levels were lower than nor0.19 IlU/ml flU/ml vs 1,36 1.36 ± 0.20 flU/ml at 0 mal (0.29 ± 0,19 0,20 IlU/ml min, p < 0.G1) 0.Q1) and their change after TRH was bluntflU/ml vs 12,54 12.54 ± 1.82 flU/ml at 20 ed (3.23 (3,23 ± 0.91 0,91 IlU/ml 1,82 IlU/ml min,, p < 0.01) 0,01) (Fig. (Fig, 1). 1) There was an inverse corremin lation between both basal and TRH stimulated TSH and urinary free cortisol levels (r == 0,93, 0.93, P < 0,01, 0.01, and r = 0.90, 0,90, P < 001, (Figs, 2 and 3). 3), 0.01, respectively) (Figs. No correlation between TSH and plasma cortisol at 08:00 and 20:00 h was observed (data not shown). shown), surgery,, in concomitance with Three months after surgery the normalization of plasma and urinary cortisol values, basal TSH and its response to TRH significant0,01) (Fig. (Fig, 1), and did not shown ly increased (p < 0.01) months, any further variation at 6 and 12 months.
Hormone assays TSH was assayed by an IRMA method (CIS, France); normal range 0.2-4 flU/ml. The intraassay 0,2-4 IlU/ml, coefficient of variation (CV) was 2,1% 2.1 % for normal TSH values (mean TSH = = 2.4 IlU/ml) flU/ml) and 3.5% 3,5% for 0,3 IlU/ml); low values (mean TSH = 0.3 flU/ml); interassay CV 2,7% for normal TSH values (mean TSH = 2.4 was 2.7% IlU/ml) 0,3 flU/ml) and 7% for low values (mean TSH = 0.3 flU/ml). Total thyroxine (T4) was assayed by fluoIlU/ml), rescence polarization immunoassay (TOx-Abbott, (TDx-Abbott, USA), normal range 4.5-12 4,5-12 Ilg/dl; fl9/dl; free thyroxine (FT4) by RIA (Biorad, Italy), normal range 0.8-2.3 0,8-2,3 ng/dl; total triiodothyronine (T3) by RIA (Mallinckrodt, West Germany), normal range 80-200 ng/dl; ACTH by IRMA (Euro-Diagnostic, (Euro-Oiagnostic, Holland), normal range 20-80 pg/ml); plasma and urinary free cortisol by RIA (OPC, fl9/dl (DPC, West Germany), normal range 5-20 Ilg/dl and 20-120 Ilg/24 respectively, fl9/24 h, respectively.
TSR
TSH
TSH UU/l1i1
TSH uUfmJ uU/ml :EO
:W
~tO
BEFORE TREATMENT
][5
15
15 1S
10 10
10 18
'I ~: c:::::: -.--
CL. (I..
00
20 28
AFI'ER TREATMENT
5
""-
~I
3 MONTIIS 3 MONTHS AFTER TREATMENT
u U/tnl 20
BEFORE TREATMENT
:-.'."--~ ..,
6O '0
min
:1 0
00
20 10
Fig. 1 - Basal and TRH-stimulated TRH-stim ulated TSH levels in Cushing 's syndrome before and after treatment. Cushing's
438
i
60 60
min in
Effects of g/ucocorticoids on pituitary-thyroid axis
..c ..e ~ M -.:r01)
----
M :::l
= OJ)
...:l 0 ~
000 00.
E-o
.c
..,. .,.-=-.. M
N
1000
r=-0.93
1000
~ ~ ~
