Psychoneuroendocrinology.Vol.17. No.4. pp.355-374.1992

0306-4530/92$5.00+0.00 PergamonPressLtd.

Printedin GreatBritain

EFFECTS OF THYROID H O R M O N E S ON CENTRAL NERVOUS SYSTEM IN AGING PETER T. LOOSEN Departments of Psychiatry and Medicine, Vanderbilt University, and Veterans Administration Hospital, Nashville, Tennessee, U.S.A. (Received in final form 18 March 1992)

SUMMARY Support for the many relationships between thyroid hormones and brain function comes from both laboratory and clinical studies. Studies in laboratory animals provide convincing evidence for a neuroregulatory role of thyroid hormones in the brain, suggesting that they may affect behavior. This notion is supported by human studies which have revealed that the effects of thyroid hormones on brain function are most important during the development and maturation of the brain; thereafter, age does not seem to critically affect brain-thyroid hormone relationships.

INTRODUCTION THE NOTION that thyroid gland activity influences human behavior is much older than our knowledge of the exact nature of thyroid secretions. Paracelsus (Cranefield, 1962) first mentioned the occurrence of mental retardation in patients with hypothyroidism. In 1850, Curling (Cranefield, 1962) described the first child with clinical features of congenital hypothyroidism and absence of the thyroid body at autopsy. In 1888, the Clinical Society of London (Clinical Society of London, 1888) observed that hypothyroidism is associated with a variety of behavioral symptoms, including mania, melancholia and psychosis. In 1897, Osier (Cranefield, 1962) reported the first series of patients with hypothyroidism and concluded that the thyroid deficiency was responsible for the characteristic changes encountered in sporadic cretinism. In contrast, thyroxine (T4) was identified in 1926, and triiodothyronine (T3) in 1951 (Astwood, 1970). Moreover, it has been only 20 yr since Guillemin (Burgus et al., 1969; Guillemin & Burgus, 1978) and Schally (197 8) and their colleagues identified thyrotropin-releasing hormone (TRH) as an apparent connection between the central nervous system and the hypothalamo-pituitary-thyroid (HPT) axis. Support for the many relationships between thyroid hormones and brain function comes from both laboratory and clinical studies. Studies in laboratory animals provide convincing evidence for a neuroregulatory role of thyroid hormones in the brain, suggesting that they may affect behavior. Human studies have demonstrated profound effects of thyroid hormones on brain development and maturation. After the brain has matured, however, the effects of thyroid hormones on brain function are less marked. Address correspondence and reprint requests to: Dr. Peter T. Loosen, Mental Health and Behavioral Sciences, V. A. Medical Center 116A, 1310 24th Avenue South, Nashville TN 37203, USA. 355

356

P.T. LOOSEN

In the present review, the wealth of data that pertains to relationships between thyroid hormones and brain function is evaluated, especially as associated with aging. The emphasis will be on clinical studies; studies of laboratory animals will be discussed where they contribute to our understanding of how thyroid hormones may affect behavior. Studies of adult populations will also be included, although the effects of thyroid hormones on brain function in this population are often subtle, and no clear association with age usually can be determined. In the adult human, the data will be organized by discussing the three most-often used approaches of investigation; i.e., the assessment of (1) behavioral changes during thyroid disease, (2) changes in thyroid function following behavioral disorders, especially affective disorders, and (3) the effects of hormones of the HPT axis on behavior. THE HPT AXIS The principal hormones of the HPT axis are T 4 and T 3, the latter being the more potent biologically. Although both T 4 and T 3 are released from the thyroid gland, about 90% of circulating T 3 is derived from T 4 by mono-deiodination in (mainly) liver and other tissues by the enzyme 5'-deiodinase I (5'D-I) (lngbar, 1985). As will be discussed below, a second 5'-deiodinase (5'D-II) is responsible for maintaining homeostasis of thyroid hormones in the brain. In serum, more than 99% of T 4 and T 3 are bound to such proteins as Ta-binding globulin (TBG), albumin, and T4-binding prealbumin, leaving less than 1% of T 4 and T 3 unbound and biologically active. Alterations in either deiodination and/or serum protein concentrations can therefore profoundly affect thyroid hormone economy (Israel et al., 1979; Israel & Orrego, 1984). Biosynthesis and release of T 4 and T 3 from the thyroid gland are primarily controlled by the anterior pituitary hormone thyrotropin (thyroid stimulating hormone, TSH). In turn, biosynthesis and release of TSH from thyrotroph cells are mediated principally by TRH (Fig. 1). TRH is released directly into the portal venous system, which connects the hypothalamus and the pituitary gland, from hypothalamic neurons that originate in the paraventricular nucleus. TRH also stimulates the release of prolactin (PRL) from pituitary lactotroph cells. Homeostatic control within the HPT axis is assured through negative feedback inhibition by T 4 and T 3 at the thyrotroph cell, leading to diminished synthesis and release of TSH (Fig. 1). Thyroid hormones also have been shown to selectively reduce TRH biosynthesis in the hypothalamus, which may provide another means of regulation (Segerson et al., 1987). Finally, many neurotransmitters and non-HPT axis hormones affect TRH and TSH release (Morley, 1981; Jakobowitz, 1988). The intactness of the HPT axis normally can be evaluated by measuring the concentrations of serum thyroid hormones. If such assessment provides ambiguous results, or if one suspects more subtle changes in HPT axis function (e.g., subclinical hypothyroidism), it is necessary to employ the TRH test, that is, assessment of serum TSH concentrations after TRH administration. THYROID FUNCTION AND THE BRAIN Studies in laboratory animals The effects of thyroid hormones on brain function have been extensively studied in experimental animals, usually the rat. From these studies comes supporting evidence for both a neuroregulatory role of thyroid hormones (Table I) and a homeostatic mechanism by which intracellular T 3 concentrations can be maintained within narrow limits (cf. Leonard, 1990, for a review). Thyroid hormone homeostasis in brain is maintained enzymatically. A second 5'-deiodi-

357

THYROID HORMONES AND THE CNS IN AGING

TRN-

TSH ) '~..~ )

'

/

"

Negative feedback

~

~

Positiveactions

on brain and other tissues

(ofter Frieden and Lipner)

The hypothalamo-pituitary-thyroldaxis, Reproduced,withpermission,fromLoosen,1987.

nase was discovered in the early 1980's and has been designated type II 5"-deiodinase (5"D-II). Evidence for the 5"D-II was suggested by the early finding that intracellular T 3 generation contributes substantially to the T 3 found in nuclei from brain and pituitary. 5'D-II has a more limited tissue distribution than 5'D-I (which is found in most tissues, being most abundant in liver and kidney) and has been identified in the central nervous system, anterior pituitary, brown adipose tissue, and placenta. Noteworthy in regard to 5'D-II activity are the differential influence of thyroid status on tissue enzyme content and its rapid modulation by the extranuclear action of thyroid hormone (Leonard, 1990). 5'D-II activity has been shown to increase three- to five-fold within 24 hr of thyroidectomy and to decrease by 8 0 - 9 0 % within 2 - 4 hr after injection of a saturating dose of T3, suggesting that the enzyme is responsible for a homeostatic mechanism by which intracellular T 3 is maintained within narrow limits (Leonard, 1990). Subsequent analysis has shown that (1) thyroid hormone-induced changes in 5'D-II activity in vivo and in cell cultures are due to changes in the half-life of the enzyme (in euthyroid animals the half-life of 5'D-II is about 30 min; it increases to 4 - 6 hr in hypothyroid animals) and do not depend on transcription or translation, and (2) T4 and rT 3 are more than 100-fold more effective than T 3 (Leonard, 1990). In contrast, 5'D-1 activity is decreased in thyroidectomized rats, and

358

P. T. LOOSEN

TABLE I. EVIDENCE FOR A NEUROREGULATORY ROLE OF THYROID HORMONES • Thyroid hormone receptors are widely distributed throughout the brain (Dratman et al., 1982) • Both T4* a n d T3~ enter the brain by a high affinity saturable transport mechanism (Dratman et al., I976, 1982) • Within the brain T4 and T3 are differentially distributed regionally and highly localized in synaptosomes (Dratman et al., 1982, 1976) • The rate of conversion of T4 to T 3 is many times greater in brain than in liver (Leonard, 1990) • There is evidence that T4 may be converted into T 3 within nerve terminals (Dratman & Crutchfield, 1978) • Despite extremes of T 4 availability, brain T 4 a n d T 3 concentrations and brain T3 production and turnover rates are kept within narrow limits (Dratman et al., 1983; Leonard, 1990), suggesting that small changes in brain thyroid hormones may produce significant changes in behavior • Hyperthyroidism increases striata113-adrenoreceptors and striatal dopaminergic activity, whereas hypothyroidism reduces striatal and hypothalamic 13-adrenoreceptors (Atterwill et al., 1984); • Hyperthyroidism increases, and hypothyroidism decreases, presynaptic c~2-adrenoreceptor function (Atterwill et al., 1984) • Hypothyroidism causes a significant increase in serotonin and substance P levels in rat brain nuclei (Savard et al., 1983) *Thyroxine

tTrLiodothyronine

at least 3 - 5 days of hypothyroidism are required to observe this fall in activity (Leonard, 1990). Taken together, the data suggest that the brain maintains intracellular T 3 concentrations within narrow limits through a homeostatic mechanism involving 5'D-II, and that thyroid hormones may exert behavioral effects by acting as direct neuromodulators, by affecting other central nervous system (CNS) neurotransmitter pathways, or both. H u m a n studies The notion that hormones of the HPT axis have direct effects on human brain function is supported by the findings of significant changes in the electroencephalogram (EEG) during thyroid gland dysfunctions (Browning et al., 1954; Jackson & Renfrew, 1966; Kopell et al., 1970; Laureau et al., 1987; Pohunkova et al., 1989; Kodama et al., 1989; Kamei et al., 1990; Sulc et al., 1990). Pohunkova et al. (1989) studied five patients after total thyroidectomy. During the hypothyroid state there was an increase in percentage representation of fast frequencies in 131 and 132 bands, with reciprocal decreases in the ct band. After T4 (but not T3) replacement, there was a return to normal in the individual frequencies in the EEG spectrum. Kodama et al. (1989) reported an increase in the power spectra of the occipital ct2 band and 131 band, and a decrease in that of the 0 band after administration of T4 in hypothyroid patients. Laureau et al. (1987) studied somatosensory evoked potentials (SEP) and auditory brain-stem responses (ABR) in children with congenital hypothyroidism. SEP interpeak latencies and ABR correlated significantly with serum T 4 and TSH levels. Support for the notion that hormones of the HPT axis have direct effects on human brain function also comes of course from the observation of specific behavioral changes during

THYROIDHORMONESANDTHE CNS 1N AGING

359

thyroid disease, as well as characteristic thyroid changes during behavioral disease. These data are discussed in detail below.

Lack of thyroid hormone during brain development The most profound effects of thyroid hormones on brain function can be observed during development and maturation of the brain. Table II summarizes these effects, which usually result in impaired neuronal maturation and connectivity (Adams & DeLong, 1986). In the human, thyroid hormone deficiency in infancy will result in mental retardation, if not corrected in timely fashion.

TABLEII.

EFFECTS OF THYROID HORMONE DEFICIENCY ON BRAIN DEVELOPMENT +

Lack of thyroid hormones during infancy, if uncorrected, results in: • significantly diminished brain weight • reduced size of cortical neurons • severely retarded dendritic arborization of Purkinje cells • reduced synaptogenesis • decreased myelinization • changes in brain enzyme activity • changes in RNA transcription *Reviewed by A d a m s & De Long (1986). Thyroid h o r m o n e deficiency during infancy usually results, if uncorrected, in mental retardation. It is likely that this is due, at least in part, to these profound c h a n g e s in brain morphology, the net result of which is impaired n e u r o n a l m a t u r a t i o n and connectivity.

THE THYROID GLAND IN AGING During the human life cycle, thyroid structure and thyroid function undergo subtle changes (Table III). However, little is known of whether such changes as decreased glandular weight, increased nodularity, decreased renal iodine clearance, reduced serum T 3 concentrations, and attenuated TSH responses to TRH in men are of any medical or behavioral significance (Ingbar, 1978; Gregerman, 1986; Greenspan & Resnick, 1991). As pointed out by Ingbar (1978), "The thematic stage for speculation that aging might be associated with hypothyroidism at the tissue level was set by early observations indicating that, in the aging individual, serum cholesterol and lipoprotein concentrations rise progressively, whereas basal metabolic rate declines." It still remains speculative whether the cholesterol and lipoprotein changes are due to reduced thyroid function, and, if so, whether the tissues of aging individuals suffer from a lack of either an adequate supply of, or adequate response to, thyroid hormones (Ingbar, 1978). Moreover, in seemingly euthyroid individuals, s e r u m T 3 concentrations decrease progressively with increasing age, suggesting that old age alters thyroid hormone metabolism in peripheral tissues. However, not all studies have revealed age-related changes in serum T 3 concentrations (Ingbar, 1978), and it has been demonstrated that serum T 3 concentration is lowered in patients with caloric deprivation, as well as in patients with a great variety of acute or chronic systemic diseases or stress (Ingbar, 1978). It is therefore possible that the decreases in serum T 3

360

R T. LOOSEN

TABLEIII.

C H A N G E S IN THYROID F U N C T I O N W I T H AGE

Thyroid gland structure • Glandular weight • Microscopic and palpable nodularity • Cellular infiltrates • Autoimmune damage Thyroid disease • Hyperthyroidism (Graves' disease) • Hypothyroidism • Subclinical hypothyroidism • T3 toxicosis • Toxic multinodular goiter

Thyroid hormones in blood and hormone production • Renal iodine clearance • Rate of iodine accumulation in thyroid • MCR of T 4 • Serum T 3 • TRH-induced TSH response in men T 3 = triiodothyronine, MCR = metabolic clearance rate. T 4 = thyroxine. S e r u m T 3 c o n c e n t r a t i o n s a r e l o w e r e d in p a t i e n t s u n d e r g o i n g c a l o r i c d e p r i v a t i o n a s w e l l a s i n a g r e a t v a r i e t y o f p a t i e n t s w i t h a c u t e o r c h r o n i c s y s t e m i c i l l n e s s e s ( I n g b a r , 1985).

concentration that have been reported are the reflection of diseases associated with old age, rather than being an effect of agingper se (Kabadi & Rosman, 1988). BEHAVIORAL CHANGES FOLLOWING THYROID DISEASE

Fetal hypothyroidism Thyroid hormone deficiency during fetal life does not appear to affect growth and maturation. Growth of the athyroid fetus is normal, and nearly all athyroid infants detected in newborn screening programs and treated within 2 mo of birth have a normal intelligence quotient (IQ), suggesting that brain development in utero is not significantly thyroid-dependent (Fisher, 1986). Infantile hypothyroidism Infants with untreated congenital hypothyroidism are likely to show, in addition to the usual thermogenic (i.e., lowered body temperature), metabolic (i.e., constipation, hypotonia), and organ (i.e., enlarged tongue, dry skin) dysfunctions, growth deficiency, delayed bone maturation, and delayed dental development. Brain effects are generally profound, with mental retardation and specific motor and sensor dysfunctions being common (Fisher, 1986). How thyroid deficiency in infancy is involved in the development of mental retardation is not fully understood. It has been estimated that 7 0 - 7 5 % of total brain growth occurs after birth; of that, two-thirds occurs during the first year (Fisher, 1986). In the laboratory animal,

THYROID HORMONES AND THE C N S 1N AGING

361

thyroid deficiency during this phase of rapid brain development usually leads to profound changes in brain morphology (Table II), resulting in impaired neuronal maturation and connectivity (Adams & DeLong, 1986). Early treatment (i.e., before 3 mo of age) appears highly effective in correcting the physical and psychological sequelae of this condition (Dussault, 1986). Klein et al. (1972) showed that early treatment is associated with a normal IQ in 85% of the infants, whereas delayed treatment is responsible for a subnormal IQ in more than 80% of infants. Other investigators have found the psychological and neuropsychiatric outcomes to be more variable. The New England Congenital Hypothyroidism Collaborative (1981) reported that 63 hypothyroid infants who received treatment before recognizable clinical manifestations had normal IQ scores when studied at age 3 or 4 yr. However, four infants who were diagnosed on the day of birth because of gross clinical manifestations had IQs of 110, 50, 64, and 76. The authors concluded that it is the degree of thyroid deficiency when treatment is begun, rather than the age of the patient, that determines the intellectual prognosis. Moreover, the Quebec Screening Program demonstrated that, at age 18 and 36 mo, hypothyroid infants had lower scores in hearing-speech performance scales, compared to normal infants (Dussault, 1986). Finally, MacFaul et al. (1978)performed neurological and psychological assessments on 30 patients aged 2.7-21 yr who were being treated for hypothyroidism starting before the age of 2 yr. Their IQ scores lay in the normal range, but 77% showed at least one sign of impaired brain function. Clumsiness was found in 33%, behavior disorders in 23%, speech disorders in 20%, learning disorders in 26%, and minor motor disorders in 50%. Childhood and adolescent hypothyroidism

After the age of 3 yr, the brain's dependency on thyroid hormones wanes, most likely because its maturation is largely completed at this time. Hypothyroidism occurring after this age is thus not associated with mental retardation, but school performance may be reduced to a variable degree. Growth retardation, delayed bone maturation, and delayed sexual maturation are common (Fisher, 1986). Adult hypothyroidism

After Gull's (1873) report "On a cretinoid state supervening in adult life in women", the Clinical Society of London established a committee on myxedema (i.e., primary hypothyroidism) to study the relationship between this condition and behavioral disorders. In 1888, they reported that in 109 patients with myxedema "delusions and hallucinations occurred in nearly half of the cases," and that "insanity" was seen in an equal number of patients (Clinical Society of London, 1888). This "myxedematous insanity" was characterized by "mania", or "dementia melancholia, with a marked preponderance of suspicion and self-accusations." The committee also observed a general slowing of thought process in their patients. These observations are still valid today. However, it now appears that the incidence of psychosis in patients with myxedema is between 5% and 15% (Hall, 1983). Although most studies reporting on behavioral changes during primary hypothyroidism have been uncontrolled or consisted of case reports (Hall, 1983; Loosen & Prange, 1984), five studies (Crown, 1949; Schon et al., 1961; Reitan, 1963; Whybrow et al., 1969; Jain, 1972), including 80 patients, used an unselected hypothyroid population. Three studies (Crown, 1949; Schon et al., 1961; Reitan, 1963) found impaired cognition to be the major psychological disturbance during active illness. Whybrow et al. (1969), studying seven patients, noted that depression, fatigue, and anxiety were the most frequent complaints, and that one patient

362

P.T. LOOSEN

presented with psychosis. They also noted that hypothyroid patients were significantly less tense, less agitated and confused, but more depressed, than hyperthyroid patients. Jain (1972), studying 30 patients, reported that 8 patients were psychiatrically normal and that 10 patients presented with anxiety, 13 with depression, and 2 with psychosis. Drinka and Voeks (1987) compared 16 nursing home patients with normal serum fT4-index and elevated serum TSH concentrations (i.e., grade II hypothyroidism) with carefully matched controls who had normal basal TSH levels. Basal TSH elevation was not associated with an increased frequency of depressive symptoms Do behavioral symptoms abate when the hypothyroidism is corrected? If so, does the time course of the endocrine remission correspond with that of the behavioral remission? It appears that when euthyroidism was achieved (i.e., with thyroid hormone replacement), psychiatric symptoms usually resolved, though in varying degrees. Whybrow et al. (1969) reported an improvement of depressive affect, retardation, and cognitive function in three of four patients who were re-evaluated in the euthyroid state. Other investigators found the cognitive impairment to be unchanged (Reitan, 1963) or improved (Schon et al., 1961). It is possible that differences in illness duration contributed to this divergence of findings. In long-standing hypothyroidism, Whybrow et al. (1969) noted persistence of cognitive impairment after thyroid replacement therapy. Jain (1972), re-evaluating eight patients, reported that "even after successful treatment of the patient's physical condition, abnormally high scores were obtained on the scales of anxiety and depression (indicating the importance of) supplementing physical treatment with psychiatric treatment when there is evidence of affective disturbance." Although the most appropriate treatment of this condition has not been determined, Pitts and Guze (1966) pointed to the effectiveness of electroconvulsive therapy (ECT). Assessment of behavioral changes in patients with overt hypothyroidism, although important, poses certain limitations. First, one cannot be entirely certain whether the behavioral symptoms, if present, are due to the lack of thyroid hormone in target tissues, or whether they are the consequence of the massive metabolic imbalance associated with this condition (Ingbar, 1985). Denicoff et al. (1990) tried to address this question by iatrogenically producing a state of early thyroid hormone withdrawal. They assessed the mood and cognitive effects of sequential T4, T3, and withdrawal of thyroid replacement in 25 patients who had thyroidectomies for thyroid cancer. The patients experienced increased sadness and anxiety when they were without medication, but no significant difference in mood was noted between the T4 and T3 treatment conditions. The patients who experienced increased mood disturbance when not on medication were more likely to have had histories of affective illness or mood lability. Psychiatric manifestations may even precede the onset of recognizable myxedema (Asher, 1949; Logothetis, 1963; Davidoff & Gill, 1977; Reed & Bland, 1977). Reed and Bland (1977) reported a case of "masked myxedema madness" and pointed out that "failure to recognize the endocrinopathy may not only produce recovery difficulties but also psychiatric and endocrine repercussions if psychotropic medications are given in such masked cases." THYROID FUNCTION IN AFFECTIVEDISORDERS Psychiatrically, disturbances in the HPT axis have more to do with affective state than with any other aspect of mentation, save possibly cognition. As noted above, depression is the most frequently observed psychiatric symptom in patients suffering from primary hypothyroidism. Conversely, approximately 30% of euthyroid depressed patients show a blunted, i.e., attenuated TSH response after TRH administration. There also is evidence that some depressed patients

THYROID HORMONES AND THE C N S IN AGING

363

may suffer from subclinical hypothyroidism. These and other relationships between brain function and hormones of the HPT axis can be outlined, as follows: Serum thyroid h o r m o n e concentrations in depression

Assessment of serum thyroid hormone concentrations in depressed patients has produced ambiguous results. The majority of depressed patients appears to be euthyroid (Kirkegaard, 1981; Loosen, 1986, 1988a, 1988b). However, some investigators have reported serum T4 levels in the upper normal range (Dewhurst et al., 1968; Kirkegaard et al., 1975; Kolakowska & Swigar, 1977; Joffe et al., 1984; Roy Byme et al., 1984; Kirkegaard & Faber, 1986; Muller & Boning, 1988; Brady & Anton, 1989). Increased serum concentrations of reverse T3 (rT3), the hormonally inactive analogue of T3 (Ingbar, 1985), also have been reported (Linnoila et al., 1979; Kirkegaard & Faber, 1981; Kjellman et al., 1983). It is noteworthy that mild and transient s e r u m T 4 elevations are not uncommon in newly hospitalized psychiatric patients, whatever the diagnosis. Three studies, involving 1131 patients, reported elevated serum fT4index concentrations in 6% (Caplan et al., 1983), 18% (Spratt et al., 1982), and 19% (Morley & Shafer, 1982) of acutely hospitalized patients. When patients were assessed serially, serum T4 concentrations normalized quickly (Morley & Shafer, 1982; Spratt et al., 1982). Other investigators reported reduced serum fT4-index (Rybakowski & Sowinski, 1973; Rinieris et al., 1978a, 1978b) or serum T 3 concentrations (Nordgren & Von Scheele, 1976; Linnoila et al., 1979; Joffe et al., 1985; Orsulak et al., 1985) in depressed patients. These data, suggesting reduced thyroid function, have been expanded by evidence that some depressed patients may suffer from subclinical hypothyroidism. Gold et al. (1981) reported that 20 of 250 (8%) patients who complained of depression and/or anergia had "some degree of hypothyroidism." According to the criteria of Wenzel et al. (1974), two (1%) patients met criteria for grade 1 (overt) hypothyroidism, nine (4%) for grade 2 (mild) hypothyroidism, and ten (4%) for grade 3 (subclinical) hypothyroidism. Of the twenty patients mentioned above, six were later successfully treated and discharged after thyroid replacement alone, and one patient was treated and discharged after thyroid replacement and a tricyclic antidepressant. In a later report, Gold et al. (1982) reported that 15% of 100 depressed patients were identified who met criteria for either subclinical, mild, or overt hypothyroidism. Of these 15 patients, 9 (60%) had detectable titers of antimicrosomal antibodies. Other investigators also have noted detectable titers of antimicrosomal and antithyroglobulin antibodies in depressed patients, though with reduced frequency. Nemeroff et al. (1985) found antithyroid antibodies in 9 (20%) of 45 psychiatric inpatients with prominent depressive symptoms. Joffe (1987) reported that 5 (9%) of 58 patients with unipolar depression had antithyroid antibodies. The four studies support the hypothesis of subtle thyroid dysfunction in a sizable sample of depressed patients. Are there dynamic changes in thyroid function when the individual patient shifts from depression into remission? If so, are the changes related to the remission of symptoms? This questions cannot be answered with certainty. Longitudinal studies of untreated depressed patients have yet to be performed. Moreover, the interpretation of available data during treatment of depression is limited, because it is not possible to determine whether variations in thyroid hormones have been related to changes in symptoms of depression or were induced by treatment (which may or may not be related to the illness). This needs to be kept in mind if one evaluates dynamic changes in the HPT axis during the process of symptomatic recovery. Nevertheless, most (Board et al., 1957; Kirkegaard et al., 1975, 1977; Joffe et al., 1984; Roy Byrue et al., 1984; Kirkegaard & Faber, 1986; Unden et al., 1986; Joffe & Singer, 1987, 1990b; Bauer & Whybrow, 1988, 1990; Baumgartner et al., 1988; Muller & Boning, 1988;

364

E T. LOOSEN

Brady & Anton, 1989) but not all (Kolakowska & Swigar, 1977; Leichter et al., 1977; Karlberg et al., 1978) studies reported significant reductions in serum T4 concentrations after remission was induced by a wide range of antidepressants, lithium, or ECT. There is preliminary evidence that the reduction in serum T4 concentration was greater in treatment responders than in non-responders, suggesting that increased thyroid function may facilitate treatment response (Roy Byme et al., 1984; Joffe & Singer, 1990b). This was first noted by Whybrow et al. (1972) who showed that heightened thyroid activity before treatment was positively correlated with a prompt clinical response to imipramine. Serum thyroid hormone concentrations in rapid cycling bipolar disorder

By definition, patients with rapid cycling bipolar disorder experience more than four episodes of depression, mania, or both, per year. Evidence suggests that some of these patients have abnormal thyroid function. Wehr and Goodwin (1979) first reported that tricyclic antidepressants (TCA) induced rapid cycling in five female bipolar patients; three had a history of thyroid disorder. Cho et al. (1979) reported that 6% of lithium-treated women developed hypothyroidism; most (71%) were rapid cyclers. They also noted that significantly more rapid cycling women (31%) than non-rapid cyclic women (2%) were on post-lithium thyroid medication. Cowdry et al. (1983), conducting a retrospective chart review, found evidence of hypothyroidism in half of 24 rapid cycling patients and in none of 19 non-rapid cycling patients. Moreover, 42% of the rapid cyclers and 32% of the non-rapid cyclers presented with subclinical hypothyroidism, resulting in an overall prevalence of some variety of hypothyroidism totaling 92% in rapid cyclers compared with 32% in non-rapid cyclers. Bauer et al. (1990), studying 30 patients with rapid cycling bipolar disorder, reported that 7 (23%) were classified as having grade I hypothyroidism, while 8 (27%) had grade II and 3 (10%) had grade III abnormalities. They concluded that hypothyroidism during bipolar illness predisposes to a rapid cycling course. If hypothyroidism during bipolar illness predisposes to a rapid cycling course, what are the effects of treatment with thyroid hormones? Preliminary studies that addressed this question have produced ambiguous results. In a small sample, O'Shanick and Ellinwood (1982) first suggested that such treatment may be beneficial. However, Cowdry et al. (1983) noted that efforts to treat rapid cycling with thyroid hormones have met with inconsistent results. Bauer and Whybrow (1990) entered 11 rapid cycling bipolar patients whose symptoms were refractory to their current treatment into an open trial of high-dose T4. T4 was added to the baseline medication regimen, and the dosage was increased until clinical response occurred or until side effects precluded a further increase. When the patients received T4, their depressive and manic symptoms decreased significantly, compared with baseline. Improvement was due to response of depressive symptoms in 10 of 11 patients, with manic symptoms responding in five of the seven patients who exhibited them at baseline. Four patients then underwent single- or doubleblind placebo substitution; three relapsed into either depression or cycling. Treatment response did not depend on thyroid status at intake. In 9 of 10 responsive patients, supranormal levels of serum fT4 were necessary to induce clinical response; however, side effects were minimal, and there were no signs of T4-induced hypermetabolism. The TRH test

The TRH test, that is, measurement of serum TSH following TRH administration has been widely used in psychiatric patients. To date, more than 50 studies, involving more than 2000 patients, the majority of whom had the diagnosis of major depressive disorder, allow iden-

THYROID HORMONES AND THE C N S IN AGING

365

tification of individual patients with TSH blunting, usually defined as a A max (i.e., peak minus baseline) TSH response less than 5.0 or 7.0 ~tU/ml. Approximately 30% of patients had a blunted TSH response during depression (Loosen & Prange, 1982; Loosen, 1988a, 1988b, 1986), although methodological issues such as the use of different doses of TRH, the use of different assays, and the lack of controls in some studies make a precise estimate of prevalence uncertain. TSH blunting also has been demonstrated in some patients with alcoholism, borderline personality disorder, mania, chronic pain, panic disorder, primary degenerative dementia, and premenstrual syndrome (Loosen, 1988a, 1988b). In anorexia nervosa, TSH responses are normal in magnitude, though often delayed in their timing. Such delay may be due to starvation rather than being causally associated with the disease; delayed responses also occur in secondary amenorrhea with simple weight loss, suggesting either hypothalamic dysfunction or altered TSH plasma clearance. Although TSH blunting is not specific for any psychiatric disorder, it does not seem to be the result of mental stress p e r se. Schizophrenic patients, undoubtedly suffering from severe mental upheaval and "stress", usually do not show TSH blunting, at least not at the frequency reported for depression or alcoholism. Sources of variance: The following factors have been reported to reduce the TRH-induced TSH response: being male, increasing age (in men), acute starvation, chronic renal failure, Klinefelter's syndrome, repetitive administration of TRH, and administration of somatostatin, neurotensin, dopamine, thyroid hormones or glucocorticoids (Loosen & Prange, 1982; Loosen, 1986). Most investigators have controlled for these factors. Less well controlled is nutritional state, since profound appetite disturbances often accompany depression. In depression, such clinical and endocrine factors as (1) the patient's age, height, weight, and body surface area; (2) severity of depression; (3) previous intake of antidepressant drugs (excluding lithium); and (4) increased activity of thyroid hormones, corticosteroids, somatostatin, or dopamine do not appear to be associated with TSH blunting. Nor does the abnormality seem to aid in the distinction between primary and secondary depression, or between unipolar and bipolar subgroups. However, there is preliminary evidence suggesting that a long duration of illness and a history of violent suicidal behavior are associated with TSH blunting. A blunted response during acute illness also seems to point to an increased risk for suicide during follow-up (Loosen, 1988a, 1988b), but not all studies agree (Korner et al., 1987). State-trait considerations: In depression and alcoholism, TSH blunting was observed in both acutely sick and remitted patients (Loosen, 1988a, 1988b). In nine studies of remitted depressed patients, 33 (35%) of 93 patients showed a blunted TSH response (Loosen, 1989). In 43 patients, the rate of normalization of (initially) blunted TSH responses was studied; it ranged from 0-71% (mean_+SD= 32 +_12%). The data suggest that TSH blunting, when present during depression, may normalize in fewer than half of remitted patients. Nine studies evaluated the TSH response to TRH during abstinence from ethanol; all but one reported the occurrence of a blunted TSH response to TRH in some patients (Garbutt et al., 1992). In toto, 163 patients received TRH; of these, 43 (26%) showed a blunted TSH response. The occurrence of TSH blunting in depression and alcoholism deserves special mention. In both conditions, the abnormality has been observed both during the acute illness and after complete remission, suggesting that it may represent a trait marker, at least in some patients. Whether TSH blunting represents a biological link that parallels the epidemiologic and genetic similarities between alcoholism and depression (Cloninger et al., 1979; Winokur, 1983) remains to be determined. Prediction of outcome: Five studies have examined whether the TRH test is useful as a

366

E T. LOOSEN

correlate of improvement after treatment with tricyclic antidepressants (TCA) or ECT, and as a predictor of early relapse after TCA- or ECT-induced remission (Loosen & Prange, 1982; Loosen, 1988a, 1988b). The difference in a max TSH between the first (initial) and second (post-treatment) testing (i.e., a ,X max TSH) was used as the endocrine variable. It was shown that (1) a positive trend in a a max TSH (i.e., a a max TSH that is bigger on the second testing than on the first) was correlated with a "good" response to treatment; (2) a persistently low TSH response (i.e., a a a max TSH less than 2.0 laU/ml) predicted early relapse, usually within 6 mo; and (3) clinical relapse predicted by the TRH test could be prevented by administration of amitriptyline. The outcome data need to be interpreted with caution, because different doses of TRH were used and patients were studied with and without continued antidepressant treatment. The TRH test and other putative biological markers: In depression, no clear association between dexamethasone suppression test (DST) abnormalities and TSH blunting has been noted, suggesting that neither abnormality is an (endocrine) epiphenomenon of the other (Loosen & Prange, 1982; Loosen, 1986). However, the finding that the corticotropin-releasing hormone (CRH)-induced ACTH response and the TRH-induced TSH response are associated in depressed patients and normal controls suggests that both axes may be regulated, at least in part, by a common mechanism (Holsboer et al., 1985). When depressed patients were studied with both TRH test and EEG sleep recordings (Rush et al., 1983), 13 (59%) of 22 patients were identified by either a reduced REM latency or by a blunted TSH response. All three tests detected 15 (68%) patients. Associations between TSH response and measures of serotonergic activity have also been studied. Gold et al. (1977) found a significant negative correlation between TSH response and cerebrospinal fluid (CSF) 5-hydroxyindoleacetic acid (5-HIAA) levels in patients with primary depression. Robertson et al. (1982) reported a significantly reduced uptake of labelled serotonin in blood platelets of euthyroid depressed patients with TSH blunting, compared to normal controls. S~gnificance of TSH blunting: There is sufficient evidence to posit two preliminary endocrine hypotheses for TSH blunting. The first hypothesis suggests that TSH blunting may be due to chronic hypersecretion of (endogenous) TRH. In this condition, the thyrotroph cells in the anterior pituitary are thought to become hyporesponsive to TRH, possibly because of down-regulation of thyrotroph cell TRH receptors. After TRH challenge, these patients would then show TSH blunting. The second hypothesis regards the status of the thyrotroph cells, which may receive increased inhibitory input in patients with TSH blunting (the possibility that these ceils may be primarily disordered is presently not testable). Although, as mentioned above, thyroid hormones are unlikely causes of TSH blunting, it is possible that small alterations in the free concentrations of thyroid hormones are responsible for the TSH response changes (Kirkegaard & Faber, 1986). Although the diagnostic utility of the TRH test appears to be compromised due to disease non-specificity and low sensitivity (Loosen et al., 1987), TSH blunting has shown promising clinical utility in predicting outcome to standard antidepressant treatment and in assessing the risk for violent suicide attempts. T3

AUGMENTATION OF ANTIDEPRESSANTS

There is evidence that a small amount of thyroid hormone accelerates the response to TCA in women and converts TCA non-responders into responders in both sexes. Augmentation of

THYROID HORMONES AND THE C N S ZN AGING

367

the antidepressant effect of TCA with thyroid hormones appears clinically useful with a variety of TCA and, preferably, should utilize T 3 rather than T4. E n h a n c e m e n t o f treatment response

Delayed onset of action, at times as long as 6 wk (Prien & Kupfer, 1986), is a significant limiting factor of standard antidepressant therapy. It is thus understandable that means were sought to shorten this delay. Augmentation of TCA with sleep deprivation (Wehr et al., 1985; Wu & Bunney, 1990; Kuhs & Toelle, 1991), lithium (Joffe, 1988a; Price, 1989), or thyroid hormone (Prange et al., 1969; Goodwin et al., 1982; Stein & Avni, 1988; Joffe & Singer, 1990a) all achieve this goal. Most (Prange et al., 1969; Wilson et al., 1970; Coppen et al., 1972; Wheathley, 1972), but not all (Feighner et al., 1972; Steiner et al., 1978) studies have shown that when TCA are given in usual doses and accompanied by as little as 25 ~tg T 3 per day, the antidepressant effect of the regimen occurs much faster than with TCA alone. This acceleration of the therapeutic response has been more marked in women than in men. In all six of the above-cited double-blind, controlled studies, patients were euthyroid at the outset and remained euthyroid throughout the course of treatment. T 3 has not increased toxicity; in one study (Coppen et al., 1972) the hormone appeared to reduce toxicity. Conversion o f treatment failures T 3 augmentation also has been applied to a related problem, the problem of TCA inefficacy. Most (Cavalca et al., 1974; Ogura et al., 1974; Banki, 1975, 1977; Earle, 1979; Tsutsui et al., 1979; Goodwin et al., 1982; Hullett & Bidder, 1983; Schwarcz et al., 1984; Targum et al., 1984; Joffe, 1988a, 1988b; Browne et al., 1990; Joffe & Singer, 1990a) but not all (Garbutt et al., 1986; Gitlin et al., 1987) studies have demonstrated that the addition of T 3 will convert TCA non-responders into responders, even when blood levels of TCA are in the therapeutic range (Goodwin et al., 1982). The studies have involved a variety of TCA in a variety of doses, with various amounts of T 3. The conversion rate - - therapeutic failure to therapeutic success is in the order of 66%; it appears about as frequent in men as in women. Joffe and Singer (1990a) demonstrated that T 3 was significantly more efficacious than T 4 in converting TCA non-responders into responders. The mechanism(s) by which T 3 accelerates the TCA response in pharmacologically naive women and converts TCA failures to success in both sexes are not known. -

-

BEHAVIORAL EFFECTS OF TRH Studies in laboratory animals

TRH has a wide range of effects in laboratory animals. They include (1) reversal of naturally-induced (i.e., hibernation) and narcotic-induced (i.e., barbiturates) CNS sedation; (2) stimulation of locomotor activity in rats; (3) effects on thermoregulation; (4) effects on the cardiovascular and respiratory systems, including increases in blood pressure and respiratory rate; (5) antinococeptive actions; (6) anorexic effects, including suppression of both food and water intake; and (7) increase in gastric mobility. For all these TRH effects, there appears to be a complex interaction among a variety of central neurotransmitters (i.e., dopamine, acetylcholine, 5-HT) and other neuropeptides (i.e., opioid peptides, neurotensin) (Nemeroff et al., 1984; Griffith, 1985).

368

E T. LOOSEN

Clinical studies In humans, TRH appears to increase the sense of well-being, motivation, relaxation, and coping capacity. This was observed in normal subjects and patients with neurological (i.e., Parkinson's disease) and psychiatric (i.e., depression, schizophrenia, childhood autism) disease (Prange et al., 1979; Loosen & Prange, 1984; Loosen, 1988b). Its behavioral effects are thus not disease-specific; rather, they appear to address a functional syndrome that includes normal behavior but also may be of importance in certain states of psychopathology. For example, in all instances of mental disorder, functions such as well-being, motivation, and coping capacity are impaired to some extent. It is noteworthy that schizophrenic patients who displayed social withdrawal, apathy, and anhedonia benefited most from TRH administration (Inanaga et al., 1978). In addition, these functions are not the specific foci of rating scales designed for assessing one or another mental illness. Betts et al. (1976) could identify TRH effects from watching videotapes in a distributive way, but not from the discrete information contained in rating scales. It is thus not surprising that attempts to use TRH as a treatment for depression or schizophrenia have been disappointing. In summary, compared with its clear effects in laboratory animals (Nemeroff et al., 1984; Griffith, 1985), there is less convincing evidence that exogenous TRH influences human behavior. It is possible that the putative behavioral effects of TRH have not been capitalized upon for a variety of reasons. First, peripherally injected TRH does not cross the blood-brain barrier to any appreciable extent. Second, trying to determine whether TRH has a direct behavioral effect is rendered difficult by the endocrine effects the peptide exerts in the periphery. Third, TRH is rapidly degraded in plasma, the half-life time being approximately 5 min (Loosen et al., 1985). Fourth, many interactions with other neuropeptides and neurotransmitters further complicate the matter (Nemeroff et al., 1984; Griffith, 1985). For future studies it thus would be advantageous to develop TRH analogues which are both hydrophobic-lipophilic (to cross the blood-brain barrier in appreciable quantities) and metabolically stable (to prevent rapid degradation). CONCLUSION Support for the many relationships between hormones of the HPT axis and brain function comes from both laboratory and clinical studies. Studies in laboratory animals provide convincing evidence for a neuroregulatory role of thyroid hormones in brain, suggesting that they may affect behavior. This notion is supported by clinical studies (Table IV) which can be most usefully separated into three major approaches: (1) Assessment of behavioral changes during thyroid disease. For example, infantile hypothyroidism, if uncorrected, usually is associated with mental retardation and motor and sensory dysfunctions. In contrast, adult hypothyroidism is associated with depression, fatigue and cognitive decline; these symptoms usually abate when the condition is corrected. (2) Evaluation of thyroid function during behavioral disease. For example, depressed patients, in varying degrees, have been reported to show serum T4 elevation, subclinical hypothyroidism, or blunted TSH response after TRH administration. The clinical significance of these findings is unknown. (3) Measurement of the effects of HPT axis hormones on behavior. For example, T 3 accelerates the response to TCA in depressed women and converts TCA non-

369

THYROID HORMONES AND THE C N S IN AGING

responders into responders in both sexes. Moreover, TRH has been shown to be behaviorally active in healthy normal volunteers and patients with neurologic and psychiatric disease. Taken in their entirety, the data point to complex relationships between thyroid hormones and behavior although, to be sure, there exists no conceptual framework within which these relationships can be understood. The data also emphasize that the effects o f thyroid hormones on brain function are most important during development and maturation of the brain; thereafter, age does not seem to critically affect brain-thyroid hormone relationships (Table IV).

TABLE IV. HORMONESOF THE HYPOTHALAMO-PITUITARY-THYROID AXISr BRAIN FUNCTIONAND AGING Diagnosis/Findings

Clinical Significance

Age Effect

Behavioral changes during thyroid disease Infantile hypothyroidism Mental retardation, cretinism, specific motor and sensory dysfunctions Adult hypothyroidism Depression, cognitive decline, mental slowing

Unknown

Thyroid function in depression Serum T 4 elevation at intake TSH blunting

May facilitate treatment response May predict treatment outcome and identify patients at risk for violent suicide

Unknown None

T3 augmentation of TCA

T3 accelerates the response to TCA in women and converts TCA non-responders into responders

Unknown

Behavioral effects of TRH

TRH appears to increase the sense of wellbeing, motivation, and coping capacity in normal subjects and patients with neurological and psychiatric disease

Unknown

Definite

Subclinical hypothyroidism

T3= trilodothyronine. TCA= tricyclic antidepressants.

REFERENCES Adams RD, DeLong GR (1986) Thyroid diseases: disorders that cause thyrotoxicosis. The neuromuscular system and brain. In: Ingbar SH, Braverman LE (Eds) The Thyroid. Lippincott, Philadelphia, pp 11681180. Asher R (1949) Myxodematous madness. Brit Med J 2: 555-562. Astwood EB (1970) Hormones and agonists. In: Goodman LS, Gilman A (Eds) The Pharmacological Basis of Therapeutics. MacMillan, London, pp 1464-1500. Atterwill CK, Bunn SJ, Atkinson DJ, Smith SL, Heal DJ (1984) Effects of thyroid status on presynaptic alpha-2 adrenoreceptor function and beta-adrenoreceptor binding in the rat brain. J Neural Transm 59: 43-55. Banki CM (1975) Triiodothyrone treatment in depression. Orvieti Hetilia 116: 2543-2546. Banki CM (1977) Cerebrospinal fluid amine metabolites after combined amitriptyline-triiodothyronine treatment of depressed women. EurJPharmaco141: 311-315. Bauer MS, Whybrow PC (1988) Thyroid hormones and the central nervous system in affective illness: interactions that may have clinical significance, lntegrPsychiatry 6: 75-100.

370

P.T. LOOSEN

Bauer MS, Whybrow PC (1990) Rapid cycling bipolar affective disorder. II. Treatment of refractory rapid cycling with high-dose levothyroxine: a preliminary study. Arch Gen Psychiatry 47: 435-440. Bauer MS, Whybrow PC, Winokur A (1990) Rapid cycling bipolar affective illness. I. Association with grade I hypothyroidism. Arch Gen Psychiatry 47: 427-432. Baumgartner A, Graf KJ, Kurten I, Meinhold H (1988) The hypothalamic-pituitary-thyroid axis in psychiatric patients and healthy subjects: parts 1-4. Psychiatry Res 24: 271-332. Betts TA, Smith J, Pidds S, Mackintosh J, Harvey P, Funicane J (1976) The effects of thyrotropin releasing hormone on measures of mood in normal women. Brit J Clin Pharmacol 3: 469-473. Board F, Wadeson R, Persky H (1957) Depressive affect and endocrine functions. Arch Neurol Psychiat 78: 612-620. Brady KT, Anton RF (1989) The thyroid axis and desipramine treatment in depression. Biol Psychiatry 25: 703 -709. Browne JL, Rice JL, Evans DL, Prange AJ (1990) Triiodothyronine augmentation of the antidepressant effect of the nontricyclic antidepressant trazodone. J Nerv Ment Dis 178: 598-599. Browning T, Atkins R, Weinger H (1954) Cerebral metabolic disturbances in hypothyroidism: clinical and encephalographic studies in the psychoses of myxedema and hypothyroidism. Arch Intern Med 93: 938942. Burges R, Dunn TF, Desiderio D, Guillemin R (1969) Structure moleculaire du facteur hypothalamique hypophysiotrope TRF d'origine ovine: mise en evidence spectrometrie de mass de la sequence, PCA-HisPro-NH2. Cr Acad Sci 269: 1870-1873. Caplan RH, Pagliara AS, Wickus G, Goodlund LS (1983) Elevation of free thyroxine index in psychiatric patients. J Psychiat Res 17: 267-174. Cavalca GG, Covezzi E, Boncinelli A (1974) Clinical experiences with the combination of thyroid extract and tricyclics in the treatment of depressed patients. Riv Sper Freniat 98: 271-300. Cho JT, Bone S, Dunner DL, Colt E, Fieve RR (1979) The effect of lithium treatment on thyroid function in patients with primary affective disorder. Am J Psychiatry 136:115-116. Clinical Society of London (1888) Report on myxedema. Trans Clin Soc London 21(Suppl): 18-38. Cloninger CR, Reich T, Wetzel R (1979) Alcoholism and affective disorders: familial associations and genetic models. In: Goodwin DW, Erickson CK (Eds) Alcoholism and Affective Disorders. SP Medical, New York, pp 57-86. Coppen A, Whybrow PC, Noguera R, Maggs R, Prange AJ Jr (1972) The comparative antidepressant value of L-tryptophan and imipramine with and without attempted potentiation by liothyronine. Arch Gen Psychiatry 26: 234-241. Cowdry RW, Wehr TA, Zis AP, Goodwin FK (1983) Thyroid abnormalities associated with rapid-cycling bipolar illness. Arch Gen Psychiatry 40: 414-420. Cranefield PF (1962) The discovery of cretinism. Bull Hist Med 36:489-511. Crown S (1949) Notes on an experimental study of intellectual deterioration. Brit Med J 2: 684-685. Davidoff F, Gill J (1977) Psychosis as an early manifestation of hypothyroidism. Conn Med 41: 618-621. Denicoff KD, Joffe RT, Lakshmanan MC, Robbins J, Rubinow DR (1990) Neuropsychiatric manifestations of altered thyroid state. Am J Psychiatry 147: 94-99. Dewhurst KE, El Kabir DJ, Exley D, Harris GW, Mandelbrote BM (1968) Blood levels of TSH, protein-bound iodine, and cortisol in schizophrenia and affective states. Lancet ii: 1160-1162. Dratman MB, Crutchfield FL (1978) Synaptosomal [125I]triiodothyronine after intravenous [125I]thyroxine. Am J Physio1235(6): E638-E647. Dratman MB, Crutchfield FL, Axelrod J, Colburn RW, Nguyen T (1976) Localisation of triiodothyronine in nerve ending fractions of rat brain. Proc Natl Acad Sci 73: 941-944. Dratman MB, Futaesaku Y, Crutchfield FL, Berman N, Payne B, Sar M, Stumpf WE (1982) Iodine-125-labeled triiodothyronine in rat brain: evidence for localization in discrete neural systems. Science 215: 309-312. Dratman MB, Crutchfield FL, Gordon JT, Jennings AS (1983) Iodothyronine homeostasis in rat brain during hypo- and hyperthyroidism. Am J Physio1245: E185-E193. Drinka PJ, Voeks SK (1987) Psychological depressive symptoms in grade II hypothyroidism in a nursing home. Psychiatry Res 21: 199-204. Dussault JH (1986) The thyroid in infancy and childhood. Congenital hypothyroidism. In: Ingbar SH, Braverman LE (Eds) The Thyroid. Lippincott, Philadelphia, pp 1399-1404.

THYROID HORMONES AND THE CNS IN AGING

371

Earle BV (1979) Thyroid hormone and tricyclic antidepressants in resistant depressions. Am J Psychiatry 126: 1667-1669. Feighner JP, King LJ, Schuckit MA, Croughan J, Briscoe W (1972) Hormonal potentiation of imipramine and ECT in primary depression. Am J Psychiatry 128: 1230-1238. Fisher DA (1986) Acquired juvenile hypothyroidism. In: Ingbar SH, Braverman LE (Eds) The Thyroid. Lippincott, Philadelphia, pp 1404-1411. Garbutt JC, Mayo JP Jr, Gillette GM, Little KY, Mason GA (1986) Lithium potentiation of tricyclic antidepressants following lack of T 3 potentiation. Am J Psychiatry 143: 1038-1039. Garbutt JC, Loosen PT, Prange AJ Jr (1992) Interrelationships between the hypothalamic-pituitary-thyroid axis and alcoholism. In: Joffe R (Ed) Psychiatric Disorders and the Thyroid. American Psychiatric Press, Washington DC. Gitlin MJ, Weiner H, Fairbanks L, Hershman JM, Friedfeld N (1987) Failure ofT 3 to potentiate tricyclic antidepressant response. J Affect Disord 13:267-272. Gold MS, Pottash ALC, Extein I (1981) Hypothyroidism and depression. JAMA 245: 1919-1922. Gold MS, Pottash ALC, Extein I (1982) "Symptomless" autoimmune thyroiditis in depression. Psychiatry Res 6: 261-269. Gold PW, Goodwin FK, Wehr T, Rebar R (1977) Pituitary thyrotropin response to thyrotropin-releasing hormone in affective illness: relationship to spinal fluid amine metabolities. Am J Psychiatry 134: 1028-1031. Goodwin FK, Prange AJ Jr, Post RM, Muscettola G, Lipton MA (1982) Potentiation of antidepressant effect by L-triiodothyronine in tricyclic nonresponders. Am J Psychiatry 139: 34-38. Greenspan SL, Resnick NM (1991) Geriatric endocrinology. In: Greenspan FS (Ed) Basic and Clinical Endocrinology. Appleton and Lange, Norwalk CT, pp 741-756. Gregerman RI (1986) Factors that control thyroid function. Intrinsic physiologic variables. In: Ingbar SH, Braverman LE (Eds) The Thyroid. Lippincott, Philadelphia, pp 361-381. Griffith EC (1985) TRH: endocrine and central effects. Psychoneuroendocrinology 10: 225-235. Guillemin R, Burgus R (1978) Peptides in the brain: the new endocrinology of the neuron. Science 202: 390402. Gull WW (1873) On a cretinoid state supervening in adult life in women. Trans Clin Soc Lond 7: 180-185. Hall RCW (1983) Psychiatric effects of thyroid hormone disturbance. Psychosomatics 24: 7-18. Holsboer F, Gerken A, von Bardeleben U, Grimm W, Stalla GK, Muller OA (1985) Relationship between pituitary responses to human corticotropin-releasing factor and thyrotropin-releasing hormone in depressives and normal controls. Eur J Pharmacol 110:153-154. Hullett FJ, Bidder TG (1983) Phenelzine plus triiodothyronine combination in a case of refractory depression. J Nerv Merit Dis 171:318 - 320. Inanaga K, Nakano T, Nagato T, Tanaka M, Ogawa N (1978) Behavioral effects of protirelin in schizophrenia. Arch Gen Psychiatry 35: 1011-1014. lngbar SH (1978) The influence of aging on the human thyroid hormone economy. In: Greenblatt RB (Ed) Geriatric Endocrinology. Raven Press, New York, pp 13-31. lngbar SH (1985) The thyroid gland. In: Wilson JD, Foster DW (Eds) Textbook of Endocrinology. Saunders, Philadelphia, pp 682- 815. Israel Y, Orrego H (1984) Hypermetabolic state and hypoxic liver damage. Rec Dev Alcohol 2:119-133. Israel Y, Walfish PG, Orrego H, Blake J, Kalant H (1979) Thyroid hormones in alcoholic liver disease. Gastroenterology 76:116-122. Jackson IMD, Renfrew S (1966) The diagnostic value of the EEG in thyrotoxicosis. Acta Endocrinol 52: 399- 403. Jain VK (1972) A psychiatric study of hypothyroidism. Psychiat Clin 5: 121-130. Jakobowitz DM (1988) Multifactorial control of pituitary hormone secretion: the "wheels" of the brain. Synapse 2: 186-192. Joffe RT (1987) Antithyroid antibodies in major depression. Acta Psychiatr Scand 76: 598-599. Joffe RT (1988a) T 3 and lithium potentiation of tricyclic antidepressants. Am J Psychiatry 145: 1317-1318. Joffe RT (1988b) Triiodothyronine potentiation of the antidepressant effect of phenelzine. J Clin Psychiatry 49: 409-410.

372

E T. LOOSEN

Joffe RT, Singer W (1987) Effect of phenelzine on thyroid function in depressed patients. Biol Psychiatry 22: 1033-1035. Joffe RT, Singer W (1990a) A comparison of triiodothyronine and thyroxine in the potentiation of tricyclic antidepressants. PsychiatryRes 32: 241-251. Joffe RT, Singer W (1990b) The effect of tricyclic antidepressants on basal thyroid hormone levels in depressed patients. Pharmacopsychiatry 23: 67-69. Joffe RT, Roy Byrne PP, Uhde TW, Post RM (1984) Thyroid function and affective illness: a reappraisal. Biol Psychiatry 19: 1685-1691. Joffe RT, Blank DW, Post RM, Uhde TW (1985) Decreased triiodothyronines in depression: a preliminary report. Biol Psychiatry 20: 922-925. Kabadi UM, Rosman PM (1988) Thyroid indices in adult healthy subjects: no influence of aging. J Am Geriatr Soc 36: 312-316. Kamei H, Sasaki H, Abe H, Nishimaru K, Okumura M (1990) The quantitative analysis of electroencephalography in primary hypothyroidism. Rinsho Shinkeigaku 30" 891-893. Karlberg BE, Kjellman BF, Kagedol B (1978) Treatment of endogenous depression with oral thyrotropin. Acta Psychiatr Scand 58: 389- 400. Kirkegaard C (1981) The thyrotropin response to thyrotropin releasing hormone in endogenous depression. Psychoneuroendocrinology 6: 189-212. Kirkegaard C, Faber J (1981) Altered serum levels of thyroxine, triiodothyronines and diiodothyronines in endogenous depression. Acta Endocrino196: 199-207. Kirkegaard C, Faber J (1986) Influence of free thyroid hormone levels on the TSH response to TRH in endogenous depression. Psychoneuroendocrinology 11: 491-497. Kirkegaard C, Norlem N, Lauridsen UB, Bjorum N, Christiansen C (1975) Protirelin stimulation test and thyroid function during treatment of depression. Arch Gen Psychiatry 32:1115-1118. Kirkegaard C, Bjorum N, Cohn D, Faber J, Lauridsen UB, Nerup J (1977) Studies on the influence of biogenic amines and psychoactive drugs on the prognostic value of the TRH stimulation test in endogenous depression. Psychoneuroendocrinology 2:131-136. Kjellman BF, Ljunggren JG, Beck Friis J, Wetterberg L (1983) Reverse T 3 levels in affective disorders. Psychiatry Res 10:1-9. Klein AH, Meltzer S, Kenny FN (1972) Improved diagnosis of congenital hypothyroidism treated before age 3 months. J Pediatr 81: 912-915. Kodama S, Tanaka K, Konishi H, Momota K, Nakasako H, Nakayama S, Yagi J, Koderazawa K (1989) Supplementary thyroxine therapy in patients with hypothyroidism induced by long-term anticonvulsant therapy. Acta Paediatr Jpn Overseas Ed 31: 555-562. Kolakowska T, Swigar ME (1977) Thyroid function in depression and alcohol abuse. Arch Gen Psychiatry 34: 984-988. Kopell B, Wittner W, Lunde D, Warrick G, Edwards D (1970) Influence of triiodothyronine on selective attention in mood as measured by the visual averaged evoked potential. Psychosom Med 32: 495-502. Korner A, Kirkegaard C, Larsen JK (1987) The thyrotropin response to thyrotropin-releasing hormone as a biological marker of suicidal risk in depressive patients. Acta Psychiat Scand 76: 355-358. Kuhs H, Toelle R (1991) Sleep deprivation therapy. Biol Psychiatry 29: 1129-1148. Laureau E, Hebert R, Vanasse M, Letarte J, Glorieux J, Desjardins M, Dussault JH (1987) Somatosensory evoked potentials and auditory brain-stem responses in congenital hypothyroidism. II. A cross-sectional study in childhood. Correlations with hormonal levels and developmental quotients. EEG Clin Neurophysio167: 521-530. Leichter SB, Kirstein L, Martin ND (1977) Thyroid function and growth hormone secretion in amitriptylinetreated depression. Am J Psychiatry 134" 1270-1272. Leonard JL (1990) Identification and structure analysis of iodothyronine deiodinases. In: Greer MA (Ed) The Thyroid Gland. Raven Press, New York, pp 285-305. Linnoila M, Lamberg BA, Rosberg G, Karonen SL, Welin MG (1979) Thyroid hormones and TSH, prolactin and LH responses to repeated TRH and LRH injections in depressed patients. Acta Psychiatr Scand 59: 536-544. Logothetis J (1963) Psychotic behavior as the initial indicator of adult myxedema. J Nerv Ment Dis 136: 561-568.

THYROID HORMONES AND THE CNS IN AGING

373

Loosen PT (1986) Hormones of the hypothalamic-pituitary thyroid axis: a psychoneuroendocrine perspective. Pharmacopsychiatry 19: 401-415. Loosen PT (1987) The TRH-induced TSH response in psychiatric disorders: a possible neuroendocrine marker. In: Nemeroff CB, Loosen PT (Eds) Handbook of Clinical Psychoneuroendocrinology. Guilford Press, New York, pp 336-360. Loosen PT (1988a) Thyroid function in affective disorders and alcoholism. Endocrinol Metab Clin NA 17: 55-82. Loosen PT (1988b) TRH: behavioral and endocrine effects in man. Prog Neuropsychopharm Biol Psychiatry 12(Suppl): $87-Sl17. Loosen PT (1989) TRH test in psychiatric disorders. In: Bunney WE, Hippius H, Laakmann G, Schmaus M (Eds) Neuropsychopharmacology. Springer Verlag, New York, pp 328-336. Loosen PT, Prange AJ Jr (1982) The serum thyrotropin (TSH) response to tbyrotropin-releasing hormone (TRH) in psychiatric patients: a review. Am J Psychiatry 139: 405-416. Loosen PT, Prange AJ Jr (1984) Hormones of the thyroid axis and behavior. In: Nemeroff CB, Dunn AJ (Eds) Peptides, Hormones and Behavior. Spectrum, New York, pp 533-577. Loosen PT, Youngblood B, Dew B (1985) Plasma levels of exogenous TRH in normal subjects: relationship to endocrine and behavioral effects. Pharmacopsychiatry 18: 325-329. Loosen PT, Garbutt JC, Prange AJ (1987) Evaluation of the diagnostic utility of the TRH-induced TSH response in psychiatric disorders. Pharmacopsychiatry 20: 90-95. MacFaul R, Domer S, Brett EM, Grant DB (1978) Neurological abnormalities in patients treated for hypothyroidism from early life. Arch Dis Child 53:611-619. Morley JE (1981) Neuroendocrine control of thyrotropin secretion. Endocr Rev 2: 396-436. Morley JE, Shafer RB (1982) Thyroid function screening in new psychiatric admissions. Arch Intern Med 142: 591-593. Muller B, Boning J (1988) Changes in the pituitary-thyroid axis accompanying major affective disorders. Acta Psychiatr Scand 77: 143-150. Nemeroff CB, Kalivas PW, Golden RN, Prange AJ Jr (1984) Behavioral effects of hypothalamic hypophysiotropic hormones, neurotensin, substance P and other neuropeptides. Pharmacology 24: 1-56. Nemeroff CB, Simon JS, Haggerty JJ Jr, Evans DL (1985) Antithyroid antibodies in depressed patients. Am J Psychiatry 142: 840-843. New England Congenital Hypothyroidism Collaborative (1981) Effects of neonatal screening for hypothyroidism: prevention of mental retardation by treatment before clinical manifestation. Lancet ii: 1095-1098. Nordgren L, Von Scheele C (1976) Myxedema madness without myxedema. Acta Med Scand 199: 233-236. O'Shanick G J, Ellinwood EH Jr (1982) Persistent elevation of thyroid-stimulating hormone in women with bipolar affective disorder. Am J Psychiatry 139:513-514. Ogura C, Okuma T, Uchida Y, Imai S, Yogi H, Sunami Y (1974) Combined thyroid (triiodothyronine)-tricyclic antidepressant treatment in depressive states. Folia PsychiatNeurolJpn 28: 179-186. Orsulak PJ, Crowley G, Schlesser MA, Giles D, Fairchild C, Rush AJ (1985) Free triiodothyronine (T3) and thyroxine (T4) in a group of unipolar depressed patients and normal subjects. Biol Psychiatry 20: 10471054. Pohunkova D, Sulc J, Vana S (1989) Influence of thyroid hormone supply on EEG frequency spectrum. Endocrinol Exp 23: 251-258. Prange AJ Jr, Wilson IC, Rabon AM, Lipton MA (1969) Enhancement of imipramine antidepressant activity by thyroid hormone. Am J Psychiatry 126: 457-469. Prange AJ Jr, Loosen PT, Nemeroff CB (1979) Peptides: application to research in nervous and mental disorders. In: Fielding S (Ed) New Frontiers of Psychotropic Drug Research. Futura, Mount Kisko NY, pp 117189. Price LH (1989) What is the duration for augmentation with T 3 o r lithium in resistant depression? J Clin Psychopharmacol 9: 7333. Prien RE Kupfer DJ (1986) Continuation drug therapy for major depressive episodes: how long should it be maintained? Am J Psychiatry 143: 18-23. Reed K, Bland RD (1977) Masked "myxedema madness". Acta Psychiatr Scand 56: 421-526. Reitan RM (1963) Intellectual functions in myxedema. Arch Neurol Psychiat 69: 436-449.

374

E T. LOOSEN

Rinieris PM, Christodoulou GN, Souvatzoglou A, Koutras DA, Stefanis CN (1978a) Free thyroxine index in mania and depression. Compr Psychiatry 19- 561-564. Rinieris PM, Christodoulou GN, Souvatzoglou A, Koutras DA, Stefanis CN (1978b) Free-thyroxine index in psychotic and neurotic depression. Acta PsychiatScand58: 56-60. Robertson AG, Tricou BJ, Fang VS, Meltzer HY (1982) Biological markers for depressive illness. Psychopharmacol Bull 18:120-122. Roy-Byrne PP, Joffe RT, Uhde TW, Post RM (1984) Carbamazepine and thyroid function in affectively ill patients. Clinical and theoretical implications. Arch Gen Psychiatry 41:1150-1153. Rush AJ, Schlesser MA, Roffwarg HP, Giles DE, Orsulak PJ, Fairchild C (1983) Relationships among the TRH, REM latency, and dexamethasone suppression tests: preliminary findings. J Clin Psychiatry 44: 2 3 29. Rybakowski J, Sowinski J (1973) Free thyroxine index and absolute free thyroxine in affective disorders. Lancet i: 889. Savard P, Merand Y, Di Paolo T, Dupont A (1983) Effects of thyroid state on serotonin, 5-hydroxyindoleacetic acid and substance P in discrete brain nuclei of adult rat. Neuroscience I0: 1399-1404. Schally AV (1978) Aspects of hypothalamic regulation of the pituitary gland. Science 202: 18-28. Schon M, Sutherland AM, Rawson RW (1961) Hormones and neuroses: the psychological effects of thyroid deficiency. In: Proc 111 Worm Congr Psychiatry. McGill University Press, Montreal. Schwarcz G, Halaris A, Baxter L, Escobar J, Thompson M, Young M (1984) Normal thyroid function in desipramine nonresponders converted to responders by the addition of L-triiodothyronine. Am J Psychiatry 141: 1614-1616. Segerson TP, Kauer J, Wolfe HC, Mobtaker H, Wu P, Jackson IMD, Lechan RM (1987) Thyroid hormone regulates TRI0 biosynthesis in the paraventricular nucleus of the rat hypothalamus. Science 238: 78-80. Spratt DI, Pont A, Miller MB, McDougall IR, Bayer MF, McLaughlin WT (1982) Hyperthyroxinemia in patients with acute psychiatric disorders. Am J Med 73: 41-48. Stein D, Avni J (1988) Thyroid hormones in the treatment of affective disorders. Acta Psychiatr Scand 77: 623-636. Steiner M, Radwan M, Elizur A, Blum I, Atsmon A, Davidson S (1978) Failure of L-triiodothyronine (T3) to potentiate tricyclic antidepressant response. CurrTherRes 23:655-659. Sulc J, Pohunkova D, Vana S (1990) EEG frequency spectrum and the saturation with thyroid hormones. Acta Nerv Superior 32: 74XXIS. Targum SD, Greenberg RD, Harmon RL, Kessler K, Salerian AJ, Fram DH (1984) Thyroid hormone and the TRH stimulation test in refractory depression. J Clin Psychiatry 45: 345-346. Tsutsui S, Yamazaki Y, Namba Y, Tsushima M (1979) Combined therapy of T 3 antidepressants in depression. JlntMedRes 7: 138-146. Unden F, Ljunggren JG, Kjellman BF, Beck Friis J, Wetterberg L (1986) Twenty-four-hour serum levels of T4 and T 3 in relation to decreased TSH serum levels and decreased TSH response to TRH in affective disorders. Acta Psychiatr Scand 73: 358-365. Wehr TA, Sack DA, Rosenthal NE (1985) Antidepressant effects of sleep deprivation and phototherapy. Acta PsychiatrBelg 85: 593-602. Wenzel KW, Meinhold H, Herpach M (1974) TRH stimulationstest mit alters- und geschlechtsabhangigem TSH Anstieg bei Normalpersonen. Klin Wschr 52: 722-727. Wheathley D (1972) Potentiation of amitryptyline by thyroid hormone. Arch Gen Psychiatry 26: 229-233. Whybrow PC, Prange A J, Treadway CR (1969) The mental changes accompanying thyroid gland dysfunction. Arch Gen Psychiatry 20: 48-63. Whybrow PC, Coppen A, Prange AJ Jr, Noguera R, Bailey JE (1972) Thyroid function and the response to liothyronine in depression. Arch Gen Psychiatry 26: 242-245. Wilson IC, Prange AJ Jr, McClane TK, Rabon AM, Lipton MA (1970) Thyroid hormone enhancement of imipramine in nonretarded depression. N Engl J Med 282:1063-1067. Winokur G (1983) Alcoholism and depression. Subst Alcohol Actions Misuse 4:11 I-119. Wu JC, Bunney WE (1990) The biological basis of an antidepressant response to sleep deprivation and relapse: review and hypothesis. Am J Psychiatry 147: 14-21.