Original articleHigh rate of autoimmune thyroiditis in bipolar disorder: lack of association with lithium exposure
Introduction
Abnormalities in thyroid function are more prevalent in patients with mood disorder than in the general population and may interfere with treatment responsiveness (Hendrick et al 1998). Although clinical hyper- or hypothyroidism may be present, mood disorders are more frequently associated with more subtle dysfunction of the hypothalamic-pituitary-thyroid axis (Hendrick et al 1998). In bipolar disorder, some studies Bauer et al 1990, Cho et al 1979, Cowdry et al 1983, Kusalic 1992 reported that overt or subclinical hypothyroidism was more prevalent in patients with a rapid cycling course, although others Bartalena et al 1990, Joffe et al 1988, Post et al 1997, Wehr et al 1988 did not find this association. Lithium, frequently used in bipolar disorder, has antithyroid effects that may lead to clinical or subclinical hypothyroidism in up to 23% of patients (Kleiner et al 1999). Lithium interferes with the synthesis and release of thyroid hormones through various mechanisms, and it has been suggested that it induces or exacerbates the development of autoimmune thyroiditis Hassman and McGregor 1988, Myers et al 1985, Wilson et al 1991.
Autoimmune thyroiditis is a major cause of hypothyroidism and is marked by the presence of thyroid antibodies. Asymptomatic autoimmune thyroiditis with a euthyroid state and a mildly increased level of circulating antibodies is not uncommon, occurring in 5% to 15% of the normal population, predominantly women, and increasing with age (Rapoport and McLachlan 1996). Positivity for thyroid antibodies is associated with lymphocytic infiltration of the thyroid gland (Yoshida et al 1978) and constitutes a risk factor for the development of thyroid failure (Vanderpump et al 1995). After the discovery of thyroid antibodies in 1956, they were generally assayed by indirect immunofluorescence (IIF). This test yields two varieties of thyroid antibodies: those that react with the cytoplasm of thyrocytes (anticytoplasmic antibodies) and those that react with components of the colloidal spaces in the thyroid follicles (anticolloid antibodies). It had already been established at that time that anticolloid antibodies reacted mainly with thyroglobulin (anti-Tg antibodies). The anticytoplasmic antibodies later appeared to react with components in the microsomal fraction of the thyrocytes (antimicrosomal antibodies). This variety turned out to be specific for the enzyme thyroid peroxidase (TPO; Portmann et al 1985).
Many commercially available test systems for these thyroid-reactive antibodies have been developed over time. Assays for thyroid microsomal antibodies (the IIF and hemagglutination assays) have now largely been replaced by assays measuring antibodies to purified or recombinant TPO. Although there is in general a strong correlation between the various assay systems for thyroid antibodies, thyroperoxidase antibodies (TPO-Abs) are more sensitive than microsomal antibodies (Rapoport and McLachlan 1996). Probably they are also less specific to test for lymphocytic thyroiditis. Antibodies to TPO/microsomes are considered better markers for lymphocytic thyroiditis than antibodies to Tg and are therefore preferentially or solely used (Doniach et al 1982).
Studies are inconsistent as to whether thyroid antibodies are elevated in bipolar disorder, with reported rates ranging from 0% to 43%, and whether lithium exposure is associated with the occurrence of thyroid antibodies Bartalena et al 1990, Bochetta et al 1996, Calabrese et al 1985, Haggerty et al 1987, Haggerty et al 1990, Haggerty et al 1997, Hornig et al 1999, Leroy et al 1988, Lazarus et al 1981, Lazarus et al 1986, Myers et al 1985, Rapaport 1994, Wilson et al 1991. Reasons for these discrepancies may include differences in diagnostic groups, sample size, male-female ratio, antibody assessment methodology, and medication status. Of the published studies that used control groups, some Haggerty et al 1990, Haggerty et al 1997, Wilson et al 1991, but not all Bartalena et al 1990, Hornig et al 1999, Rapaport 1994, found a higher prevalence of thyroid antibodies in bipolar disorder, especially in depressed and mixed states. One study found a positive association between the occurrence of thyroperoxidase antibodies and a history of rapid cycling (Oomen et al 1996). The presence of thyroid antibodies has also been associated with a poor response to antidepressant treatment (Prange et al 1990).
To further study thyroid autoimmunity in bipolar disorder, we evaluated TPO-Abs and serum thyroid stimulating hormone (TSH) in an international, multicenter study on a large cohort of DSM-IV bipolar outpatients and compared TPO-Ab rates in these patients with a general population control group and a psychiatric inpatient group. We additionally examined thyroid parameters according to demographic, illness, and treatment variables.
Section snippets
Methods and materials
The subjects were 226 outpatients (112 men and 114 women) with bipolar I (n = 178), II (n = 42), or not otherwise specified (NOS; n = 6) disorder, aged 23 to 83 years, from the Stanley Foundation Bipolar Network, a multicenter longitudinal treatment research program performed in the United States and the Netherlands Leverich G et al in press, Suppes et al in press. All patients gave written informed consent for the study. A DSM-IV diagnosis was made by means of the SCID-I/P Research Version
Prevalence of TPO-Abs
Table 1 displays the overall occurrence of TPO-Abs in bipolar outpatients and the two control groups, differentiated by gender and age. In the general population control group TPO-Ab prevalences ranged from 3.0% to 17.8% depending on the gender and the assay system used. As expected, there was a higher prevalence in women. In psychiatric inpatient control subjects tested with the lumi-nescent system, TPO-Ab prevalences ranged from 3.7% to 15.3% depending on gender and age and were not
Discussion
To our knowledge, this is the largest controlled sample of well-diagnosed DSM-IV bipolar patients to be assessed for thyroid autoimmunity and its relationship to thyroid failure and lithium exposure. Compared with the two control groups, we found that outpatients with bipolar disorder had a higher prevalence of TPO-Abs. It has been suggested that previously reported increased thyroid antibody titers in psychiatric patients are mainly due to inadequate control for age and gender (Hornig et al
Acknowledgements
The authors acknowledge the generous support from the Theodore and Vada Stanley Foundation. We thank H.A.P.S. Oomen, M.D., Ph.D., for the collection of the psychiatric control group; H. de Wit for the processing of the blood samples; R. Docter, Ph.D., for the FT4 testing; G.W. Akkerhuis, M.S., for assistance in preparing the manuscript; and J.J. Bartko, Ph.D., for statistical advice.
References (42)
- et al.
Evaluation of thyroid function in patients with rapid-cycling and non-rapid-cycling bipolar disorder
Psychiatry Res
(1990) - et al.
The presence of antithyroid antibodies in patients with affective and non-affective psychiatric disorders
Biol Psychiatry
(1990) - et al.
Psychoendocrinology of mood disordersThe hypothalamic-pituitary-thyroid axis
Psychiatr Clin North Am
(1998) - et al.
Autoantibody disturbances in affective disordersA function of age and gender?
J Affect Disord
(1999) - et al.
Thyroid function and bipolar affective disorder
Psychiatry Res
(1988) - et al.
Effect of lithium carbonate therapy on thyroid immune status in manic depressive patientsA prospective study
J Affect Disord
(1986) - et al.
Rapid cycling bipolar disorderLack of relation to hypothyroidism
Psychiatry Res
(1997) Immune parameters in euthymic bipolar patients and normal volunteers
J Affect Disord
(1994)- et al.
Thyroid peroxidase autoantibodies
- et al.
Rapid cycling bipolar affective disorder. II. Treatment of refractory rapid cycling with high-dose levothyroxineA preliminary study
Arch Gen Psychiatry
(1990)