13
Psychiatry Research, 34:13-17 Elsevier
Evaluation of Thyroid Function Cycling and Non-Rapid-Cycling
in Patients With RapidBipolar Disorder
Luigi Bartalena, Laura Pellegrini, Marisa Meschi, Lucia Antonangeli, Faust0 Bogazzi, Liliana Dell’Osso, Aldo Pinchera, and Gian Franc0 Placidi Received
March 19, 1990; revised version received July 10, 1990; accepted August 1 I, 1990.
Abstract. The problem of whether rapid-cycling (RC) bipolar disorder is more frequently associated than non-rapid-cycling (NRC) bipolar disorders with thyroid dysfunction was investigated in two groups of 11 women matched for age and therapy. Seven patients in each group were under chronic lithium therapy. Both RC and NRC patients, as compared to euthyroid controls, showed a reduction in mean total and free thyroid hormone concentrations, subnormal values of free thyroxine being found in four RC and three NRC patients. No patient had supranormal baseline thyroid stimulating hormone (TSH) values, but an exaggerated TSH response to thyrotropin releasing hormone was found in three RC and two NRC patients: all these patients had been receiving lithium therapy for more than one year. No differences in the prevalence of goiter and thyroid-directed autoantibodies were observed in the two groups. These data confirm that bipolar disorder, especially during treatment with lithium, is associated with at least subclinical hypothyroidism, and suggest that RC patients do not differ from NRC patients in the prevalence of spontaneous or lithiuminduced thyroid hypofunction. Lithium-induced hypothyroidism is likely to be related to the length of treatment. Key Words. Rapid-cycling lithium, thyroid stimulating
bipolar disorder, thyroid function, hypothyroidism, hormone, thyrotropin releasing hormone test.
An association between rapid-cycling (RC) bipolar disorders and thyroid dysfunction has recently been suggested, with the implication that subtle thyroid function abnormalities might predispose patients to a rapid course of bipolar illness. Several RC depressives have been reported to be, independently of treatment, either overtly or subclinically hypothyroid (Cho et al., 1979; Wehr and Goodwin, 1979; Extein et al., 1982; O’Shannick and Ellinwood, 1982; Bauer and Whybrow, 1986). Furthermore, a higher prevalence of lithium-associated hypothyroidism has been
Luigi Bartalena, M.D., is Assistant Professor of Endocrinology, lstituto di Endocrinologia. Marisa Meschi, M.D., is Intern, Istituto di Clinica Psichiatrica. Laura Pellegrini, M.D., is Intern, lstituto di Clinica Psichiatrica. Lucia Antonangeli, M.D., is Intern, Istituto di Endocrinologia. Faust0 Bogazzi, M.D., is Intern, Istituto di Endocrinologia. Liliana DelI’Osso, M.D., is Assistant Professor of Psychiatry, Istituto di Clinica Psichiatrica. Gian Franc0 Placidi, M.D., is Associate Professor of Psychiatry, Istituto de Clinica Psichiatrica. Aldo Pinchera, M.D., is Professor of Endocrinology, lstituto di Endocrinologia, University of Pisa. (Reprint requests to Dr. L. Bartalena, Istituto di Endocrinologia, University of Pisa, Viale de1 Tirreno, 64, 56018 Tirrenia-Pisa, Italy.) 0165-1781/90/$03.50
@ 1990 Elsevier Scientific
Publishers
Ireland
Ltd.
14 observed in RC than in non-rapid-cycling (NRC) patients (Cho et al., 1979; Cowdry et al., 1983). According to Cowdry et al. (1983) these differences are not related to the length of lithium treatment. This conclusion has recently been challenged by Joffe et al. (1988), who evaluated 42 patients receiving lithium therapy and found eight patients with clinical or subclinical hypothyroidism, all belonging to the NRC group: hypothyroidism appeared to be related to the female sex and to the duration of treatment. A study by Sack et al. (1988) failed to show any difference in thyroid hormone levels between RC and NRC patients. The precise relationship between thyroid function abnormalities and RC bipolar disorder is, therefore, still uncertain. To address this problem, we evaluated thyroid function in 11 consecutive RC bipolar women and 11 NRC bipolar women matched for age and type of treatment. Methods Subjects. The study group included 11 consecutive RC bipolar women (age range 22-78 years, mean age 43 years), 11 NRC bipolar women (age range 21-75 years, mean age 46 years). The diagnosis of major affective disorder was made using DSM-III-R (American Psychiatric Association, 1987) criteria. Rapid cycling was defined as four or more affective episodes per year (Dunner and Fieve, 1974); the length of a cycle is the interval between the onset of one manic episode and the onset of the next one, including both manic and depressed phases. The mean length of RC (k SD) was 65 f 32 days. The two groups were matched for type of treatment, which included tricyclic antidepressants, tranquilizers, carbamazepine, benzodiazepines, and lithium. Lithium was chronically administered (from a few months to more than 2 years) to seven RC and seven NRC patients, with no significant difference in the duration of treatment between the two groups. The diagnosis of RC was based on postlithium cycle length in patients who had been taking lithium salts. A group of 11 euthyroid women, matched for age, with no present or past history of psychiatric or thyroid diseases served as controls. The presence of goiter was assessed by palpation by one of the authors (L.B.) and classified according to World Health Organization grading. Serum total thyroxine (TT,) and triiodothyronine (TT,) were measured by ARIA II radioimmunoassay (RIA) methods; free T4 (FT.,) and free T, (FT,) were assayed by the Lisophase kits (Sclavo S.p.A., Siena, Italy); reverse T, (rT,) was determined by RIA (Radim, S.p.A., Rome, Italy); thyroid stimulating hormone (TSH) was measured by a sensitive immunoradiometric assay, with a limit of detection of 0.1 mU/l (ELSA-2 TSH, CIS, France), either basally or after iv. stimulation with 200 c(g thyrotropin releasing hormone (TRH). Antithyroglobulin (TgAb) and anti-thyroid peroxidase (TPO-Ab) antibodies were measured by ELISA methods; T,-binding globulin (TBG) was assayed by a two-site immunoradiometric assay (Corning Medical-Kontron S.p.A., Milan, Italy). As an index of the peripheral effect of thyroid hormones, we also measured, by an immunoradiometric assay (Farmos Diagnostica, Oulunsolo, Finland), sex hormone-binding globulin (SHBG), a protein whose liver synthesis is stimulated by thyroid hormones (Rosner et al., 1984). All measurements were carried out in duplicate in a single run to avoid interassay variation. Normal values for women in our laboratory were as follows: TT, 51.6-144.5 nmol/l; TT, 1.5-3.2 nmol/l; FT, 8.4-21.3 pmol/l; FT, 3.9-8.5 pmol/:l; rT, 0.1-0.5 nmol/l; TSH 0.4-4.6 mu/l; TBG 10.6-30.6 mg/l; SHBG 28-92 nmol/ 1. Data were .expressed as mean f SD. Results were analyzed by analysis of variance (ANOVA).
15 Results Table 1 summarizes results of mean serum thyroid hormone, TSH, and SHBG measurements. Mean serum TT,, TT,, FT,, FT,, and SHBG concentrations did not differ in the two groups of patients, but were significantly reduced with respect to control subjects (p < 0.001). However, no patient had clinical symptoms or signs of overt hypothyroidism. Mean serum TSH values were superimposable in the two groups and did not differ from the controls (Table 1). However, in both the RC and NRC groups, the baseline and TRH-stimulated TSH values were significantly higher in lithium-treated patients than in patients not treated with this drug. An exaggerated TSH response to TRH was observed in three RC and two NRC patients: all these patients had been treated with lithium for more than 1 year. Two patients (one from each group) had normal baseline and TRH-stimulated serum TSH levels despite the reduced serum FT, concentrations. Table 1. Mean biochemical
parameters
Rapid cyclers TT, (nmol/l) TT:, (nmol/l) FT, (pmol/l) FTJ (pmol/l) rT3 (nmol/l) TSH (mu/l) Baseline After TRH TBG (mg/l) SHBG (nmol/l)
87.2 f 2.0 It 8.9 f 4.8 f 0.4 f
21.8 0.3 2.3 0.3 0.3
1.1 f
in the study group (mean f SD) Non-rapid cyclers
Controls
92.1 f
24.1
115.3 f
16.8’
2.0 f
0.7
2.7 f
0.3’
9.1 f
2.0
12.9f
1.7*
4.8 f
0.7
6.0f
1.0’
0.4 f
0.3
0.2 f
0.3*
0.7
1.2f
0.7
1.3f
1.0
8.9
11.3f
7.6
11.4f
3.3
23.9 f
3.8
27.0 f
5.6
22.5 f
3.0
37.8 f
15.2
34.3 f
15.5
50.6 f
8.6*
11.8f
Note. TTI = total thyroxine. TTJ = total triiodothyronine. FT, = tree T4. FTI = tree Ts. rT3 = reverse TJ. TSH = thyroid stimulating hormone. TRH = thyrotropin releasing hormone. TBG = T4 binding globulin. SHBG = sex hormone-binding globulin. 1. p < 0.05 vs. rapid cyclers and non-rapid cyclers. 2. p < 0.001 vs. rapid cyclers and non-rapid cyclers.
Subnormal serum FT, values were found in four RC and three NRC patients, whereas serum FT, concentrations were, with the exception of one borderline value, within the normal range (Fig. 1). Subnormal serum SHBG values were observed in two RC and four NRC patients: a positive correlation between serum FT, and serum SHBG values was detectable (data not shown). A small goiter (OB grade) was present in three RC and three NRC patients. Tg-Ab and/or TPO-Ab tests were positive in one patient from each bipolar group and in one control subject. Discussion The results of the present study showed that a substantial proportion of all patients (7 out of 22, 32%) had subclinical hypothyroidism, as assessed by decreased serum FT, levels, exaggerated TSH response to TRH, or both. The decreased serum con-
16 Fig. 1. Individual serum free thyroxine (FT4) and free triiodothyronine (FT3) values in rapid-cycling (RC) and non-rapid-cycling (NRC) patients 15,
. = 2
.
t10. 2 a
l
-rt .. . .
p
P ; 0
,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ,,,1,,1,,,,,,,,1,,,,,~,~,,,,, ,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ;;;;'~,"',"""",,~,~,,,,,
I=
.
I f
0,
&
,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ,,,,1,,,,,,,,,,,,,,1,,,,,,,,, z ,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ,,,,,,,,,,,,,,,,,,,,,,,,,,,,~ ,,,,,~,,,,,,,,,,,,,,,~,,,,,,,, ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ,,~,,,,,,,,~,~~,~,,,~,~~,,~~~,,,,,,,,,,,,,,,,,,,,~~~, ,,,,,,,,,,,,,,,,,,,,,,,, :: .=,a,, ,,, ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ,,,,,,,,,,,,,,,,,~~~~~,,~~~~~~ ::::::::::::::::::::::::::::: ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ,,,,,,,,,,,,,11,11,,,,,,,,,,, ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ,,,,,,,,,,1,,,,,,1,1,,,,,,,,, ,,,,,,,,,,,,,,,l,~,,~,,~~~~~~~ ,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ,,,,,,,,,,,,,,,,~,,,~,,~~~,~~~ ,,,,,,,,,,,,,,,,,,,,,,,,~,,,, ,,,,,,,,,,,,,,,,,,,,,,,,,,~,,, ,,,,,,,,,,,,,,,,,,,,~,~,~,~~,~ ::::::::::::::::::::::::::::: ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,~ ,,,,,,,,,,,,,,,,,,,,,,,,,,,~,, ::::::::::::::::::::::::::::: ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ,~,,,,,,,,,,,1,,1,1,~~,~~~~~~~ ,,,,,,,,,,1,4,,,1,,1,,,,,,,,, ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ,,,,,,,,,,,,,,,,,,,,,,,,,,,,, PC
NRC
FT4
RC
NRC
FT3
Shaded area refers to subnormal values.
centrations of SHBG, whose liver synthesis is stimulated by thyroid hormones (Rosner et al., 1984), are in keeping with impending thyroid failure. It is interesting to note that two patients had inappropriately normal baseline serum TSH values (and a normal TSH response to TRH) in the presence of reduced serum thyroid hormone levels, suggesting a central defect of TSH secretion. The latter situation appears to be more common in untreated patients, in whom the central origin of hypothyroidism is also suggested by the loss of the nocturnal serum TSH surge (Bartalena et al., 1990). The main goal of this study was to investigate whether the rapid-cycling course would be related to clinical or subclinical hypothyroidism, as suggested by some authors (Cho et al., 1979; Wehr and Goodwin, 1979; Extein et al., 1982; O’Shannick and Ellinwood, 1982; Bauer and Whybrow, 1986). Our data, in keeping with the results of Joffe et al. (1988) and Wehr et al. (1988), indicate that thyroid dysfunction is not associated with the rapid-cycling course of depression. In fact, no differences were found in the prevalence of subclinical hypothyroidism, goiter, and thyroiddirected autoantibodies in RC and NRC patients. Furthermore, while lithium administration was confirmed to be associated with a relatively high prevalence of hypothyroidism, no difference emerged in the prevalence of lithium-induced hypothyroidism in RC and NRC patients, at variance with the results reported by Cowdry et al. (1983). Although the relationship between the occurrence of hypothyroidism and the duration of lithium treatment was not significant, possibly due to the small number of patients, it is worth recalling that the patients with an exaggerated TSH response to TRH had been receiving lithium treatment for more than a year. Thus, it would appear that the length of lithium treatment is relevant for the development of hypothyroidism, as suggested by Joffe et al. (1988). The question of whether T, replacement therapy affects the nature of the disorder
17 and induces a shift from rapid cycling to non-rapid cycling is unsettled. Some studies indicate that such a shift does occur when treatment with T, is initiated (Extein et al., 1982; Stancer and Persad, 1982; Leibow, 1983), but other reports have failed to show any beneficial effect of thyroid hormone supplementation (Cowdry et al., 1983). Our study does answer this question. However, in view of the similarity of thyroid function abnormalities in RC and NRC patients, it is not apparent why thyroid hormone supplementation should affect the course of the disorder. Carefully conducted prospective and controlled studies are clearly needed to expand our understanding of this controversial problem. Acknowledgments. This work was partially supported
by grants from the Minister0 della
Pubblica Istruzione (60%), Rome, Italy.
References American Psychiatric Association. DSM-III-R: Diagnostic and Statistical Manual of Mental 3rd ed., revised. Washington, DC: APA, 1987. Bartalena, L.; Placidi, G.F.; Martino, E.; Falcone, M.; Pellegrini, L.; Dell’Osso, L.; Pacchiarotti, A.; and Pinchera, A. Nocturnal serum thyrotropin (TSH) surge and the TSH response to TSH-releasing hormone: Dissociated behavior in untreated depressives. Journal of Clinical Endocrinology and Metabolism, 7 1:September, 1990. Bauer, M.S., and Whybrow, P.C. The effect of changing thyroid function on cyclic affective illness in a human subject. American Journal of Psychiatry, 143:633-636, 1986. Cho, J.T.; Bone, S.; Dunner, D.L.; Colt, E.; and Fieve, R.R. The effect of lithium treatment on thyroid function in patients with primary affective disorder. American Journal of Psychiatry, 136:115-l 16, 1979. Cowdry, R.W.; Wehr, T.A.; Zis, A.P.; and Goodwin, F.K. Thyroid abnormalities associated with rapid cycling bipolar illness. Archives of General Psychiatry, 40:414420, 1983. Dunner, D.L., and Fieve, R.R. Clinical factors in lithium carbonate prophylaxis failure. Disorders,
Archives of General Psychiatry, 30~229-233, 1974.
Extein, I.; Pottash, A.L.C.; and Gold, M.S. Does subclinical hypothyroidism predispose to tricyclic-induced rapid mood cycles? Journal of Clinical Psychiatry, 43:290-293, 1982. Joffe, R.T.; Kutcher, S.; and MacDonald, C. Thyroid function and bipolar affective disorder. Psychiatry Research, 25117-121,
1988.
Leibow, D. L-Thyroxine for rapid-cycling
bipolar illness. American Journal of Psychiatry,
140:1255, 1983.
O’Shannick, G.J., and Ellinwood, E.H., Jr. Persistent elevation of thyroid stimulating hormone in women with bipolar affective disorder. American Journal of Psychiatry, 139:513514, 1982.
Rosner, W.; Aden, D.P.; and Khan, M.S. Hormonal infhuences on the secretion of steroidbinding proteins by a human hepatoma-derived cell line. Journa/of Clinical Endocrinology and Metabolism, 59:806-809, 1984. Sack, D.A.; James, S.P.; Rosenthal, N.E.; and Wehr, T.A. Deficient nocturnal surge of TSH secretion during sleep and sleep deprivation in rapid-cycling bipolar illness. Psychiatry Research, 23:179-191, 1988.
Stancer, H.C., and Persad, E. Treatment of intractable rapid-cycling manic-depressive disorder with levothyroxine. Archives of General Psychiatry, 39:311-312, 1982. Wehr, T.A., and Goodwin, F.K. Rapid cycling in”manic-depressives induced by tricyclic antidepressants. Archives of General Psychiatry, 36555-559, 1979. Wehr, T.A : Sack, D.A.; Rosenthal, N.E.; and Cowdry, R.W. Rapid cycling affective disorder: Contributing factors and treatment responses in 51 patients. American Journal of Psychiatry, 145:179-184, 1988.
Psychiatry Research, 34:13-17 Elsevier
Evaluation of Thyroid Function Cycling and Non-Rapid-Cycling
in Patients With RapidBipolar Disorder
Luigi Bartalena, Laura Pellegrini, Marisa Meschi, Lucia Antonangeli, Faust0 Bogazzi, Liliana Dell’Osso, Aldo Pinchera, and Gian Franc0 Placidi Received
March 19, 1990; revised version received July 10, 1990; accepted August 1 I, 1990.
Abstract. The problem of whether rapid-cycling (RC) bipolar disorder is more frequently associated than non-rapid-cycling (NRC) bipolar disorders with thyroid dysfunction was investigated in two groups of 11 women matched for age and therapy. Seven patients in each group were under chronic lithium therapy. Both RC and NRC patients, as compared to euthyroid controls, showed a reduction in mean total and free thyroid hormone concentrations, subnormal values of free thyroxine being found in four RC and three NRC patients. No patient had supranormal baseline thyroid stimulating hormone (TSH) values, but an exaggerated TSH response to thyrotropin releasing hormone was found in three RC and two NRC patients: all these patients had been receiving lithium therapy for more than one year. No differences in the prevalence of goiter and thyroid-directed autoantibodies were observed in the two groups. These data confirm that bipolar disorder, especially during treatment with lithium, is associated with at least subclinical hypothyroidism, and suggest that RC patients do not differ from NRC patients in the prevalence of spontaneous or lithiuminduced thyroid hypofunction. Lithium-induced hypothyroidism is likely to be related to the length of treatment. Key Words. Rapid-cycling lithium, thyroid stimulating
bipolar disorder, thyroid function, hypothyroidism, hormone, thyrotropin releasing hormone test.
An association between rapid-cycling (RC) bipolar disorders and thyroid dysfunction has recently been suggested, with the implication that subtle thyroid function abnormalities might predispose patients to a rapid course of bipolar illness. Several RC depressives have been reported to be, independently of treatment, either overtly or subclinically hypothyroid (Cho et al., 1979; Wehr and Goodwin, 1979; Extein et al., 1982; O’Shannick and Ellinwood, 1982; Bauer and Whybrow, 1986). Furthermore, a higher prevalence of lithium-associated hypothyroidism has been
Luigi Bartalena, M.D., is Assistant Professor of Endocrinology, lstituto di Endocrinologia. Marisa Meschi, M.D., is Intern, Istituto di Clinica Psichiatrica. Laura Pellegrini, M.D., is Intern, lstituto di Clinica Psichiatrica. Lucia Antonangeli, M.D., is Intern, Istituto di Endocrinologia. Faust0 Bogazzi, M.D., is Intern, Istituto di Endocrinologia. Liliana DelI’Osso, M.D., is Assistant Professor of Psychiatry, Istituto di Clinica Psichiatrica. Gian Franc0 Placidi, M.D., is Associate Professor of Psychiatry, Istituto de Clinica Psichiatrica. Aldo Pinchera, M.D., is Professor of Endocrinology, lstituto di Endocrinologia, University of Pisa. (Reprint requests to Dr. L. Bartalena, Istituto di Endocrinologia, University of Pisa, Viale de1 Tirreno, 64, 56018 Tirrenia-Pisa, Italy.) 0165-1781/90/$03.50
@ 1990 Elsevier Scientific
Publishers
Ireland
Ltd.
14 observed in RC than in non-rapid-cycling (NRC) patients (Cho et al., 1979; Cowdry et al., 1983). According to Cowdry et al. (1983) these differences are not related to the length of lithium treatment. This conclusion has recently been challenged by Joffe et al. (1988), who evaluated 42 patients receiving lithium therapy and found eight patients with clinical or subclinical hypothyroidism, all belonging to the NRC group: hypothyroidism appeared to be related to the female sex and to the duration of treatment. A study by Sack et al. (1988) failed to show any difference in thyroid hormone levels between RC and NRC patients. The precise relationship between thyroid function abnormalities and RC bipolar disorder is, therefore, still uncertain. To address this problem, we evaluated thyroid function in 11 consecutive RC bipolar women and 11 NRC bipolar women matched for age and type of treatment. Methods Subjects. The study group included 11 consecutive RC bipolar women (age range 22-78 years, mean age 43 years), 11 NRC bipolar women (age range 21-75 years, mean age 46 years). The diagnosis of major affective disorder was made using DSM-III-R (American Psychiatric Association, 1987) criteria. Rapid cycling was defined as four or more affective episodes per year (Dunner and Fieve, 1974); the length of a cycle is the interval between the onset of one manic episode and the onset of the next one, including both manic and depressed phases. The mean length of RC (k SD) was 65 f 32 days. The two groups were matched for type of treatment, which included tricyclic antidepressants, tranquilizers, carbamazepine, benzodiazepines, and lithium. Lithium was chronically administered (from a few months to more than 2 years) to seven RC and seven NRC patients, with no significant difference in the duration of treatment between the two groups. The diagnosis of RC was based on postlithium cycle length in patients who had been taking lithium salts. A group of 11 euthyroid women, matched for age, with no present or past history of psychiatric or thyroid diseases served as controls. The presence of goiter was assessed by palpation by one of the authors (L.B.) and classified according to World Health Organization grading. Serum total thyroxine (TT,) and triiodothyronine (TT,) were measured by ARIA II radioimmunoassay (RIA) methods; free T4 (FT.,) and free T, (FT,) were assayed by the Lisophase kits (Sclavo S.p.A., Siena, Italy); reverse T, (rT,) was determined by RIA (Radim, S.p.A., Rome, Italy); thyroid stimulating hormone (TSH) was measured by a sensitive immunoradiometric assay, with a limit of detection of 0.1 mU/l (ELSA-2 TSH, CIS, France), either basally or after iv. stimulation with 200 c(g thyrotropin releasing hormone (TRH). Antithyroglobulin (TgAb) and anti-thyroid peroxidase (TPO-Ab) antibodies were measured by ELISA methods; T,-binding globulin (TBG) was assayed by a two-site immunoradiometric assay (Corning Medical-Kontron S.p.A., Milan, Italy). As an index of the peripheral effect of thyroid hormones, we also measured, by an immunoradiometric assay (Farmos Diagnostica, Oulunsolo, Finland), sex hormone-binding globulin (SHBG), a protein whose liver synthesis is stimulated by thyroid hormones (Rosner et al., 1984). All measurements were carried out in duplicate in a single run to avoid interassay variation. Normal values for women in our laboratory were as follows: TT, 51.6-144.5 nmol/l; TT, 1.5-3.2 nmol/l; FT, 8.4-21.3 pmol/l; FT, 3.9-8.5 pmol/:l; rT, 0.1-0.5 nmol/l; TSH 0.4-4.6 mu/l; TBG 10.6-30.6 mg/l; SHBG 28-92 nmol/ 1. Data were .expressed as mean f SD. Results were analyzed by analysis of variance (ANOVA).
15 Results Table 1 summarizes results of mean serum thyroid hormone, TSH, and SHBG measurements. Mean serum TT,, TT,, FT,, FT,, and SHBG concentrations did not differ in the two groups of patients, but were significantly reduced with respect to control subjects (p < 0.001). However, no patient had clinical symptoms or signs of overt hypothyroidism. Mean serum TSH values were superimposable in the two groups and did not differ from the controls (Table 1). However, in both the RC and NRC groups, the baseline and TRH-stimulated TSH values were significantly higher in lithium-treated patients than in patients not treated with this drug. An exaggerated TSH response to TRH was observed in three RC and two NRC patients: all these patients had been treated with lithium for more than 1 year. Two patients (one from each group) had normal baseline and TRH-stimulated serum TSH levels despite the reduced serum FT, concentrations. Table 1. Mean biochemical
parameters
Rapid cyclers TT, (nmol/l) TT:, (nmol/l) FT, (pmol/l) FTJ (pmol/l) rT3 (nmol/l) TSH (mu/l) Baseline After TRH TBG (mg/l) SHBG (nmol/l)
87.2 f 2.0 It 8.9 f 4.8 f 0.4 f
21.8 0.3 2.3 0.3 0.3
1.1 f
in the study group (mean f SD) Non-rapid cyclers
Controls
92.1 f
24.1
115.3 f
16.8’
2.0 f
0.7
2.7 f
0.3’
9.1 f
2.0
12.9f
1.7*
4.8 f
0.7
6.0f
1.0’
0.4 f
0.3
0.2 f
0.3*
0.7
1.2f
0.7
1.3f
1.0
8.9
11.3f
7.6
11.4f
3.3
23.9 f
3.8
27.0 f
5.6
22.5 f
3.0
37.8 f
15.2
34.3 f
15.5
50.6 f
8.6*
11.8f
Note. TTI = total thyroxine. TTJ = total triiodothyronine. FT, = tree T4. FTI = tree Ts. rT3 = reverse TJ. TSH = thyroid stimulating hormone. TRH = thyrotropin releasing hormone. TBG = T4 binding globulin. SHBG = sex hormone-binding globulin. 1. p < 0.05 vs. rapid cyclers and non-rapid cyclers. 2. p < 0.001 vs. rapid cyclers and non-rapid cyclers.
Subnormal serum FT, values were found in four RC and three NRC patients, whereas serum FT, concentrations were, with the exception of one borderline value, within the normal range (Fig. 1). Subnormal serum SHBG values were observed in two RC and four NRC patients: a positive correlation between serum FT, and serum SHBG values was detectable (data not shown). A small goiter (OB grade) was present in three RC and three NRC patients. Tg-Ab and/or TPO-Ab tests were positive in one patient from each bipolar group and in one control subject. Discussion The results of the present study showed that a substantial proportion of all patients (7 out of 22, 32%) had subclinical hypothyroidism, as assessed by decreased serum FT, levels, exaggerated TSH response to TRH, or both. The decreased serum con-
16 Fig. 1. Individual serum free thyroxine (FT4) and free triiodothyronine (FT3) values in rapid-cycling (RC) and non-rapid-cycling (NRC) patients 15,
. = 2
.
t10. 2 a
l
-rt .. . .
p
P ; 0
,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ,,,1,,1,,,,,,,,1,,,,,~,~,,,,, ,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ;;;;'~,"',"""",,~,~,,,,,
I=
.
I f
0,
&
,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ,,,,1,,,,,,,,,,,,,,1,,,,,,,,, z ,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ,,,,,,,,,,,,,,,,,,,,,,,,,,,,~ ,,,,,~,,,,,,,,,,,,,,,~,,,,,,,, ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ,,~,,,,,,,,~,~~,~,,,~,~~,,~~~,,,,,,,,,,,,,,,,,,,,~~~, ,,,,,,,,,,,,,,,,,,,,,,,, :: .=,a,, ,,, ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ,,,,,,,,,,,,,,,,,~~~~~,,~~~~~~ ::::::::::::::::::::::::::::: ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ,,,,,,,,,,,,,11,11,,,,,,,,,,, ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ,,,,,,,,,,1,,,,,,1,1,,,,,,,,, ,,,,,,,,,,,,,,,l,~,,~,,~~~~~~~ ,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ,,,,,,,,,,,,,,,,~,,,~,,~~~,~~~ ,,,,,,,,,,,,,,,,,,,,,,,,~,,,, ,,,,,,,,,,,,,,,,,,,,,,,,,,~,,, ,,,,,,,,,,,,,,,,,,,,~,~,~,~~,~ ::::::::::::::::::::::::::::: ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,~ ,,,,,,,,,,,,,,,,,,,,,,,,,,,~,, ::::::::::::::::::::::::::::: ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ,~,,,,,,,,,,,1,,1,1,~~,~~~~~~~ ,,,,,,,,,,1,4,,,1,,1,,,,,,,,, ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ,,,,,,,,,,,,,,,,,,,,,,,,,,,,, PC
NRC
FT4
RC
NRC
FT3
Shaded area refers to subnormal values.
centrations of SHBG, whose liver synthesis is stimulated by thyroid hormones (Rosner et al., 1984), are in keeping with impending thyroid failure. It is interesting to note that two patients had inappropriately normal baseline serum TSH values (and a normal TSH response to TRH) in the presence of reduced serum thyroid hormone levels, suggesting a central defect of TSH secretion. The latter situation appears to be more common in untreated patients, in whom the central origin of hypothyroidism is also suggested by the loss of the nocturnal serum TSH surge (Bartalena et al., 1990). The main goal of this study was to investigate whether the rapid-cycling course would be related to clinical or subclinical hypothyroidism, as suggested by some authors (Cho et al., 1979; Wehr and Goodwin, 1979; Extein et al., 1982; O’Shannick and Ellinwood, 1982; Bauer and Whybrow, 1986). Our data, in keeping with the results of Joffe et al. (1988) and Wehr et al. (1988), indicate that thyroid dysfunction is not associated with the rapid-cycling course of depression. In fact, no differences were found in the prevalence of subclinical hypothyroidism, goiter, and thyroiddirected autoantibodies in RC and NRC patients. Furthermore, while lithium administration was confirmed to be associated with a relatively high prevalence of hypothyroidism, no difference emerged in the prevalence of lithium-induced hypothyroidism in RC and NRC patients, at variance with the results reported by Cowdry et al. (1983). Although the relationship between the occurrence of hypothyroidism and the duration of lithium treatment was not significant, possibly due to the small number of patients, it is worth recalling that the patients with an exaggerated TSH response to TRH had been receiving lithium treatment for more than a year. Thus, it would appear that the length of lithium treatment is relevant for the development of hypothyroidism, as suggested by Joffe et al. (1988). The question of whether T, replacement therapy affects the nature of the disorder
17 and induces a shift from rapid cycling to non-rapid cycling is unsettled. Some studies indicate that such a shift does occur when treatment with T, is initiated (Extein et al., 1982; Stancer and Persad, 1982; Leibow, 1983), but other reports have failed to show any beneficial effect of thyroid hormone supplementation (Cowdry et al., 1983). Our study does answer this question. However, in view of the similarity of thyroid function abnormalities in RC and NRC patients, it is not apparent why thyroid hormone supplementation should affect the course of the disorder. Carefully conducted prospective and controlled studies are clearly needed to expand our understanding of this controversial problem. Acknowledgments. This work was partially supported
by grants from the Minister0 della
Pubblica Istruzione (60%), Rome, Italy.
References American Psychiatric Association. DSM-III-R: Diagnostic and Statistical Manual of Mental 3rd ed., revised. Washington, DC: APA, 1987. Bartalena, L.; Placidi, G.F.; Martino, E.; Falcone, M.; Pellegrini, L.; Dell’Osso, L.; Pacchiarotti, A.; and Pinchera, A. Nocturnal serum thyrotropin (TSH) surge and the TSH response to TSH-releasing hormone: Dissociated behavior in untreated depressives. Journal of Clinical Endocrinology and Metabolism, 7 1:September, 1990. Bauer, M.S., and Whybrow, P.C. The effect of changing thyroid function on cyclic affective illness in a human subject. American Journal of Psychiatry, 143:633-636, 1986. Cho, J.T.; Bone, S.; Dunner, D.L.; Colt, E.; and Fieve, R.R. The effect of lithium treatment on thyroid function in patients with primary affective disorder. American Journal of Psychiatry, 136:115-l 16, 1979. Cowdry, R.W.; Wehr, T.A.; Zis, A.P.; and Goodwin, F.K. Thyroid abnormalities associated with rapid cycling bipolar illness. Archives of General Psychiatry, 40:414420, 1983. Dunner, D.L., and Fieve, R.R. Clinical factors in lithium carbonate prophylaxis failure. Disorders,
Archives of General Psychiatry, 30~229-233, 1974.
Extein, I.; Pottash, A.L.C.; and Gold, M.S. Does subclinical hypothyroidism predispose to tricyclic-induced rapid mood cycles? Journal of Clinical Psychiatry, 43:290-293, 1982. Joffe, R.T.; Kutcher, S.; and MacDonald, C. Thyroid function and bipolar affective disorder. Psychiatry Research, 25117-121,
1988.
Leibow, D. L-Thyroxine for rapid-cycling
bipolar illness. American Journal of Psychiatry,
140:1255, 1983.
O’Shannick, G.J., and Ellinwood, E.H., Jr. Persistent elevation of thyroid stimulating hormone in women with bipolar affective disorder. American Journal of Psychiatry, 139:513514, 1982.
Rosner, W.; Aden, D.P.; and Khan, M.S. Hormonal infhuences on the secretion of steroidbinding proteins by a human hepatoma-derived cell line. Journa/of Clinical Endocrinology and Metabolism, 59:806-809, 1984. Sack, D.A.; James, S.P.; Rosenthal, N.E.; and Wehr, T.A. Deficient nocturnal surge of TSH secretion during sleep and sleep deprivation in rapid-cycling bipolar illness. Psychiatry Research, 23:179-191, 1988.
Stancer, H.C., and Persad, E. Treatment of intractable rapid-cycling manic-depressive disorder with levothyroxine. Archives of General Psychiatry, 39:311-312, 1982. Wehr, T.A., and Goodwin, F.K. Rapid cycling in”manic-depressives induced by tricyclic antidepressants. Archives of General Psychiatry, 36555-559, 1979. Wehr, T.A : Sack, D.A.; Rosenthal, N.E.; and Cowdry, R.W. Rapid cycling affective disorder: Contributing factors and treatment responses in 51 patients. American Journal of Psychiatry, 145:179-184, 1988.
