About the author
Edward Shorter, PhD, FRSC is the Jason A Hannah Professor of the History of Medicine and Professor of Psychiatry, Faculty of Medicine, University of Toronto, Toronto, Ontario, Canada.
Endocrine Psychiatry in Historical Perspective
Psychiatry’s encounter with endocrinology gives a new dimension to the “tortures of Tantalus,” for whom the grapes were always so close but never quite within reach. Over the last hundred years, psychiatry has continually grasped for endocrinological riches, and failed, and given up in despair. Today, to the extent that “psychiatric endocrinology” exists, it is placed well on the back burner of the field (Fink and Shorter, 2010).
The “Powdered Organ Extract” Era
The story of psychiatric endocrinology begins with flagging male sexual ardor, “male hypoactive sexual desire disorder,” in the clunky terms of DSM-5. In 1889 Parisian physician Charles-Edouard Brown-Séquard, who decades earlier had performed the first experimental adrenalectomies, treated patients with extracts of ground-up testis. Then seventy-one years old, Séquard also treated himself with the extract, and reported, “I should add that intellectual tasks became easier for me than for some years and that I regained everything that I lost. I must say as well,” he deadpanned, “that other forces that were not lost but quite diminished, have considerably improved as well” (Brown-Séquard (1889). Brown-Séquard’s experiments aroused as much laughter as imitation. Yet it was the commencement of endocrine investigations in the field of psychiatry.
Around the turn of the century, a young Parisian physician, Maxime Laignel Lavastine, now entirely forgotten, became the true founder of scientific neuroendocrinology. Around 1908-09, he studied the range of endocrine organs, linking their hyper- or hyopoactivity to psychiatric illnesses. (Of course, various non-psychiatrists had preceded him: with the discovery of hyperthyroidism, “Grave’s disease,” in 1835; adrenal hypofunction (“Addison’s disease”) in 1849; and hypothyroidism (“myxedema”) in 1873. Yet none of the English internists who made these discoveries had primarily psychiatric interests.) Trained in internal medicine and psychiatry, Laignel-Lavastine surveyed a range of “psycho-glandular relationships,” finding that disorders of function in thyroid, pituitary, adrenals, and ovaries affected character in different ways. He treated endocrine hypofunctions with ground-up extracts, finding for example, that neurasthenia responded (or was thought to respond) to thyroid extracts (Laignel-Lavastine, 1908a). Presciently, Laignel-Lavastine associated melancholia with the endocrine organs. “It is quite evident that the melancholic syndrome is indeed mental, but simultaneously physical and psychic. It seems to me that melancholics are particularly indicated for coming research on endocrine disorders” (Laignel-Lavastine, 1908b). Laignel then drifted away from endocrine psychiatry, and the French dominance in the field gave way to the German.
Melancholia was in the center of the German radar. In 1921, Karl Kleist, originator of the concept of “bipolar disorder,” who taught psychiatry at the University of Frankfurt, speculated that the “autochthonous degeneration-psychoses” (meaning the chronic non-deteriorating psychoses, unlike those described by Emil Kraepelin as “dementia praecox”) had a large endocrine component (Kleist, 1921). A whole raft of German research along these lines was launched. In 1922 Josef Westermann, a student of Kurt Schneider’s at the University of Cologne (of “Schneiderian criteria” fame), concluded of Schneider’s “vital depression,” “It is conceivable that the basic biological disorder of vital depression, which is certainly to be conceived as endocrine, is similar to the corresponding biological mechanisms that trigger schizophrenia” (Westermann, 1922). Vital depression” was close to melancholia, and the thought that it had much in common with schizophrenia would be reprised many decades later (National Institutes of Health, 2009).
Similarly, the German-speaking Europe led the pack around the First World War in marketing commercial endocrine preparations, such as Roche’s “Pituglandol,” to the (largely unsuspecting) public (Roche-Austria, 1913). This was big business.
Yet there was genuine science in the powdered extract organ era. In 1932, U.S. neurosurgeon Harvey Cushing laid out the rudiments of what would later be called the “hypothalamic-pituitary-adrenal” (HPA) axis. As yet, the role of the hypothalamus was unknown, but Cushing associated a tumor-linked overdrive of this axis with “striking constitutional transformations,” including depression and psychosis (Cushing, 1932).
But clouds were forming. The absence of solid links between endocrines and psychiatry was a source of growing dubiety, and the budding field was especially dragged down by associating presumed ovarian malfunctions with mental illness. When James Collip, who led biochemistry at McGill University (and under whose aegis ACTH had been isolated in the anterior pituitary gland in 1933), gave Heinz Lehmann, chief of Montreal’s Douglas Hospital, a supply of pituitary extract to try in schizophrenia, Lehmann believed the patient was better mainly because the extract had a high alcohol content (Lehmann, 1993).
Cortisol and Dexamethasone
The discovery of the steroid hormones of the adrenal cortex transformed endocrine psychiatry. It was Swiss chemist Tadeus Reichstein and Edward Kendall, a biochemist at the Mayo Clinic in Minnesota, who in the mid-1930s identified the corticosteroid hormones, especially cortisone and its activated version cortisol, which Reichstein isolated in 1937 (Shorter and Fink, 2010). Cortisol, also known as hydrocortisone, plays an important role in the endocrine psychiatry story because in melancholia the adrenal gland hypersecretes it.
These early successes gave a new push to endocrine psychiatry. In Bristol, England, Robert Hemphill and Max Reiss explored endocrine treatments for schizophrenia and melancholia (Hemphill, 1944). Harvard physiologist Walter Cannon put the endocrine glands and the autonomic nervous system on the behavioral map in 1914 with the “flight or fight syndrome” (Cannon, 1914). And in Montreal Hans Selye, who had little time for Cannon or Reichstein but was mainly interested in his own research, publicized a supposedly endocrine-based “general adaptation syndrome” (that today has disappeared from endocrinology) (Selye, 1950).
It was, however, a new test for detecting various adrenal cortex hormones in the urine, introduced by Wilhelm Zimmerman in 1935, that led to the direct association of cortisol with psychiatric illness (Zimmerman, 1835). Cortisol itself became available for clinical use in 1950 when two scientists at Merck developed its synthesis, and in 1956 psychiatrist Eugene Bliss at Utah, together with endocrinologist colleagues, reported that stress and emotional upset increased blood levels of cortisol in psychiatric patients (Bliss et al., 1956).
In 1956 Francis Board and David Hamburg at the Michael Reese Hospital in Chicago shone the searchlight directly on cortisol and depression, finding that serum cortisol was significantly higher in “neurotic depressions” than in controls, and that it was very much higher in psychotic depressives than in controls: There was thus a range of cortisol elevations corresponding to the type of depression, with the highest corresponding to the most severe. This is one of the most significant findings in post-World-War-II psychiatry Board, Persky and Hamburg, 1956).
Now the scene shifts from Chicago to London. In 1962 James Gibbons, a senior lecturer at the Maudsley Hospital, and Paul McHugh, a visiting fellow on a National Institutes of Health fellowship, collaborated on a study of the biochemistry of depression. They found a smooth correlation between the severity of depression and the level of plasma cortisol. Moreover, as the patients improved, their cortisol levels dropped. Their data were so striking as to make depression look like a disease of cortisol (Gibbons and McHugh, 1962)!
With this huge new interest in corticosteroids and psychiatric illness in the background, in 1968 Bernard Carroll, a psychiatrist-endocrinologist in Melbourne, together with colleagues, discovered the first biological test in psychiatry: that patients with melancholic depression fail to suppress the secretion of cortisol when administered the artificial steroid hormone dexamethasone (Carroll, 1968). Dexamethasone nonsuppression was, in other words, a biological marker of melancholia. The test was not exclusive to melancholia, because patients with other illnesses, such as dementia and anorexia nervosa, also exhibited nonsuppression in the Dexamethasone Suppression Test (DST). Yet psychiatrists usually could figure out clinically whether their depressed patients were also demented, or anorexic, and the specificity and sensitivity of the DST were just about on par with those of the interictal EEG (Shorter and Fink, 2010).
Radioimmune assay (RIA) became available in the 1960s for the study of peptide hormones that mediate endocrine signaling. In the early 1970s researchers led by Edward Sachar, first at Montefiore Hospital in the Bronx and then at Columbia University where Sachar became head of psychiatry in 1976, pursued the new RIA technology with studies of prolactin and growth hormone in schizophrenia and depression. From the late 1960s to the mid-1970s, Sachar and his group also documented elevated cortisol production and especially abnormal nocturnal cortisol secretion in depression (Sachar et al., 1973). This last discovery led Sachar to abandon his early formulation of cortisol hypersecretion in depression as resulting from a breakdown of ego defenses against anxiety. Sachar’s larger role in the history of psychoneuroendocrinology, however, was the recruitment of a talented next generation of researchers. His life was sadly ended by suicide in a post-stroke depression in 1984 before his concept of a neuroendocrine “window on the brain” could come to fruition.
The 1970s were really the heyday of endocrine psychiatry. A broad picture of the underlying neurochemical differences between depressive illness and schizophrenia began to emerge. As Carroll pointed out in 1976, depressive patients have high cortisol levels in serum, urine, and cerebrospinal fluid (CSF), as well as abnormal DST results, whereas schizophrenic patients with similar rated levels of depressive symptoms and ego defense breakdown have normal HPA function (Carroll, 1976). At the time, psychiatry was still struggling to strike itself free of psychoanalysis: Diseases were not just constructs “mediated by a nonspecific breakdown of psychological defense mechanisms,” as Carroll put it tongue in cheek, but were actually real phenomena.
I wish I could report that from these triumphs the endocrine approach went on to become dominant in psychiatry, surpassing the study of the neurotransmitters and pointing the way to future advances. Unfortunately, the opposite happened: endocrine approaches went downhill, essentially vanishing from psychiatry by the 1990s. The germinative event was an assault from the towering heights of psychiatry upon Carroll’s dexamethasone suppression test. Early in its trajectory the DST had become touted, somewhat unwisely (and not by Carroll), as a “screening test for depression.” Just emerging from the swamp of psychoanalysis, many psychiatrists felt uneasy around brain biology, and clung to the DST as a biological measure that could tell them what they seemed incapable of learning clinically: whether their patients were “depressed.” It was unhelpful that DSM-III in 1980 elided the difference between melancholic illness, for which the DST is a reliable guide, and non-melancholic illness, for which it is not. Instead, DSM-III created “major depression,” embodying a highly heterogeneous clinical population that a biological test such as the DST would individuate poorly.
In July 1982 the National Institute of Mental Health held a workshop ostensibly on “neuroendocrine tests,” but in fact it was really on the DST. Few attendees had much clinical experience with the DST. And the report about the specificity of the DST that emerged from the meeting was unremittingly negative (Hirschfeld, Koslow and Kupfer, 1983). (This illustrates one of the great problems in the “consensus approach” in U. S. psychiatry: the consensus may be ill informed.)
Worse was to come. In 1983 the American Psychiatric Association asked a similar task force to assess the DST, and the APA report in 1985, emitted after much infighting and acrimony, was even harsher. “The task force found no incontrovertible role for the DST in current clinical practice” (APA Task Force, 1987).
In retrospect, the entire DST affair was a tragedy for psychiatry. The field had permitted its attention to be diverted to the issue of screening tests, whereas the importance of the DST lay in identifying an underlying biochemical abnormality in serious depression. Not all depressive patients had this abnormality. How did those who did have it differ clinically? What was their distinctive biochemistry? Their molecular genetics? These questions were never asked, as the fledgling biological psychiatrists, fresh from the wreckage of psychoanalysis, were too inexperienced with the physicality of the body to discern where future study should go. Instead, bewitched by pharmaceutical largesse, they leapt into neurotransmitters.
Psychiatry has long known that thyroid disturbances play a big role in mood disorders. As English physician Helen Boyle pointed out in 1930, “. . . . There is too high a percentage of manic-depressives who show evidence of thyroid disturbance for the connection to be due to coincidence . . . Out of 100 consecutive cases at the Lady Chichester Hospital, six were cyclothymics, and in four of these the patients had large thyroids. Of the ninety-four others only eight had large thyroids” (Boyle, 1930).
Yet it was only modern techniques in the study of neurochemistry that permitted the systematic investigation of the hypothalamic-pituitary-thyroid axis. A team at the University of North Carolina led by Arthur Prange made thyroid their particular mission. And in 1969, Prange and colleagues reported an interesting finding: administration of L-triiodothyronine (T3) augments the benefits of imipramine in depression (Prange et al., 1969). As research on the peptide hormones began to rage, in 1972, the Prange group reported another important biological finding: that the TSH response to TRH is blunted in depression. Depression was, in other words, marked by “hypothalamic underactivity” (Prange et al., 1972). Prange later called this surmise (which may actually not be true) an “endocrine scar” (Prange, 1998).
This research led to much international interest in thyroid metabolism and in TRH as factors in depression, or milestones on the pathway of pharmaceutical intervention. In 1990 Prange’s student Peter Whybrow and Mark Bauer reported on the benefits of augmenting lithium treatment of bipolar disorder with thyroid hormones (Bauer an Whybrow, 1990). Yet on the whole psychiatry’s interest in thyroid proved readily exhaustible.
Sprigs of Hope
In the absence of pharmaceutical treatments for endocrine abnormalities, psychiatry lost interest in the hormones. As Paul McHugh, who had become chair of psychiatry at Johns Hopkins University, said in 2007, “You can’t persuade people to do DSTs. I do them but I can’t get anyone else to do them” (McHugh, 2007). Attention now shifted to genetics and neurotransmitters. Endocrine psychiatry seemed dead.
Yet not quite. On the therapeutic side, the postmodern trail to endocrine psychiatry commenced in 1984 when Charles Nemeroff at Duke University identified a possible pharmacological target: high CRF concentrations in the CSF of depressed patients who were studied in Sweden by Eric Widerlov; Nemeroff postulated that “CRF hypersecretion is, at least in part, responsible for the hyperactivity of the hypothalamo-pituitary-adrenal axis characteristic of major depression” (Nemeroff et al., 1984).
This appealing suggestion, that would have opened new therapeutic avenues in the treatment of depression, has not been confirmed. Several studies have converged on the finding that M Wong and associates reported in 2000: despite their hypercortisolism, “depressed patients had normal levels of plasma ACTH and CSF CRH” (Wong et al., 2000).
Research at the National Institute of Mental Health has nuanced slightly these negative findings. As Philip Gold and George Chrousos concluded in 2003: “There is evidence of abnormal CRH secretion in the hyperactive HPA axis function of melancholics.” Yet, “. . . There is hyperactivity of the CRH system without there necessarily being hypersecretion of CRH per se.” (Gold and Chrousos, 2013).
Despite high hopes, it is now apparent that CRH receptor antagonists are not the magic bullet for depressive disorders (or anxiety, for that matter). One example: a trial of a selective CRHR1 antagonist “failed to demonstrate efficacy in the treatment of major depression,” as Brendon Binneman et al at Pfizer reported in 2008 (Binneman et al., 2008). High hopes had been placed also in the glucocorticoid receptor antagonist RU486 (mifepristone) that Alan Schatzberg’s group at Stanford -- where, in the words of Kenneth Davis the “neuroendocrine window was just explode in” (Davis, 2011) -- had hoped to develop. In trials for psychotic depression that were somewhat puzzling (because of a high placebo response in that illness), RU486 failed to demonstrate efficacy in the treatment of depression, although the results for psychosis were slightly more encouraging (DeBattista et al., 2004). Bernard Carroll and Robert Rubin were discouraging about the entire enterprise of treating psychotic depression with RU486 (Rubin and Carroll, 2006). On the whole, as the first decade of the new millennium closed, the prospects of pharmacological interventions in the HPA axis were yet remote. As two investigators at Organon Laboratories concluded in 2008, “. . . Drug treatments based on manipulation of the HPA axis have not yet entered clinical practice” (Thomson a Craighead, 2008).
A second intriguing avenue leads from the secretion of the peptide hormone oxytocin in the posterior hypothalamus to autism. In 1998, Charlotte Modahl and associates at the Boston University School of Medicine found lower levels of plasma oxytocin in autistic children than in normal controls (Modahl, et al., 1998). In 2003 Eric Hollander and associates at the Mount Sinai School of Medicine reported that oxytocin infusions reduced the stereotypies characteristic of autism and intellectual disability (Hollander et al., 2003). (The stereotypies constitute a form of catatonia, and in fact respond readily to anticatatonic remedies) (Wachtel et al., 2008). These findings have produced a great deal of speculation – notably from Thomas Insel, director of the National Institute of Mental Health – that the “dark matter” of social neuroscience might lead to endocrine breakthroughs in autism (Insel, 2010). Whether these dizzying theoretical perspectives will be realized remains open.
On the diagnostic side, endocrine perspectives have shown more promise. The DST emerges as a possible predictor of suicide, as William Coryell and colleagues at the University of Iowa found in 2001. Of 78 patients with major depression or psychotic depression (“schizoaffective disorder”) entered in a long-term follow-up study between 1978 and 1981, 32 had abnormal DST results. Of these patients, the suicide risk was 26.8 percent, compared to only 2.9 percent among patients with normal DSTs (Coryell and Schlesser, 2001). Studies in the VA Greater Los Angeles Healthcare System (Yerevenian, et al., 2004), and the Karolinska Institute (Yokinen, et al., 2008) confirmed these results.
What will become of Tantalus? Will these diagnostic and therapeutic perspectives encourage psychiatry to cool its love affair with neurotransmitters and reexamine the beguilements of the endocrine approach?
APA Task Force on Laboratory Tests in Psychiatry. The dexamethasone suppression test: An overview of its current status in psychiatry. AJP 1987; 144: 1253 – 62.
Bauer MS, Whybrow PC. Rapid cycling bipolar affective disorder II: Treatment of refractory rapid cycling with high-dose levothyroxine: A preliminary study. Archives of General Psychiatry 1990; 47: 435–40.
Binneman B, et al. A 6-week randomized, placebo-controlled trial of CP-316,311 (a selective CRH1 antagonist) in the treatment of major depression. American Journal of Psychiatry 2008; 165: 617-20.
Board F, Persky H, Hamburg DA. Psychological stress and endocrine functions: blood levels of adrenocortical and thyroid hormones in acutely disturbed patients. Psychosomatic Medicine 1956; 18: 324–33.
Boyle H. Discussion in Farquhar Buzzard E, “Discussion on the diagnosis and treatment of the milder forms of the manic-depressive psychosis.” Proceedings of the Royal Society of Medicine 1930; 23: 890.
Brown-Séquard C-E. Des effets produits chez l’homme par des injections sous-cutanées d’un liquide retiré des testicules frais de cobaye et de chien. Comptes rendus hebdomadaires des séances et mémoires de la Société de Biologie et de Ses Filiale 1889; 41: 415–22.
Cannon WB. The interrelations of emotions as suggested by recent physiological researches. American Journal of Physiology 1914; 25: 256–82.
Carroll BJ. Limbic system-adrenal cortex regulation in depression and schizophrenia. Psychosomatic Medicine 1975; 38: 106-21.
Carroll BJ, Martin FIR, Davies B. Resistance to suppression by dexamethasone of plasma 11-OH CS levels in severe depressive illness. BMJ 1968; 2 (Aug. 3): 285–7
Coryell W, Schlesser M. The dexamethasone suppression test and suicide prediction. American Journal of Psychiatry 2001; 158 : 748-53.
Cushing H. The basophil adenomas of the pituitary body and their clinical manifestations (pituitary basophilism). Johns Hopkins Hospital Bulletin 1932; 50 : 137–95.
Davis KL interviewed by Stanley J Watson. In Thomas A Ban, editor, An Oral History of Neuropsychopharmacology, The First Fifty Years: Peer Interviews. Volume 8 (volume editor: Carl Salzman C). Brentwood TN: ACNP; 2011, p. 75-86.
DeBattista C, et al. Mifepristone versus placebo in the treatment of psychosis in patients with psychotic major depression, Biological Psychiatry 2006; 60: 1343–9.
Gibbons Jl, McHugh PR. Plasma cortisol in depressive illness. Journal of Psychiatric Research 1962; 1: 162–71.
Gold PW, Chrousos GP. Melancholic and atypical subtypes of depression represent distinct pathophysiological entities: CRH, neural circuits, and the diathesis for anxiety and depression. Molecular Psychiatry 2013; 18 : 632-4.
Hemphill R. Endocrinology in clinical psychiatry. Journal of Mental Science 91944; 90: 410–34.
Hirschfeld MS, Koslow SH, Kupfer DJ. The clinical utility of the dexamethasone suppression test in sychiatry: Summary of a National Institute of Mental Health workshop. JAMA, 1983; 250: 2172–4.
Hollander E, et al. Oxytocin infusion reduces repetitive behaviors in adults with autistic and Asperger’s disorders. Neuropsychopharmacology 2003; 28: 193-8.
Insel T. The challenge of translation in social neuroscience: A review of oxytocin, vasopressin, and affiliative behavior. Neuron 2010; 65: 768-79.
Jokinen J, et al. Hypothalamic-pituitary-adrenal (HPA) axis and suicide risk in mood disorders. J Affective Disorders 2008; 107: S65-S66.
Kleist K. Autochthone Degenerationspsychosen. Zeitschrift für die gesamteNeurologie und Psychiatrie 1921; 69: 1–11.
Laignel-Lavastine M. Les Troubles psychiques dans les syndromes thyroidiens. Congrès des Aliénistes et Neurologistes de France, Rapports et comptes rendus 1908a; 17: 204–30.
Laignel-Lavastine M. Les Troubles de glandes à sécrétion interne chez les mélancoliques. Revue de psychiatrie et de psychologie expérimentale 1908b; 12: 429–33.
Lehmann HE. Before they called it psychopharmacology. Neuropsycho-pharmacolog 1993: 83: 291–303.
McHugh P interview with Max Fink, Dec. 10, 2007, Stony Brook, N.Y
Modahl C, et al. Plasma oxytocin levels in autistic children. Biol Psyhiatry 1998; 43: 270-7.
National Institutes of Health: NIH News, “Schizophrenia and Bipolar Disorder Share Genetic Roots. http//www.nih.gov/news/health/jul 2009/ nimh-01.htm, July 1, 2009.
Nemeroff CB, et al. Elevated concentrations of CSF corticotropin-releasing factor-like immune-reactivity in depressed patients. Science 1984; 226: 1342-4.
Prange AJ. Psychoneuroendocrinology: A commentary. Psychiatric Clinics of North America 1998; 21: 491–505.
Prange AJ, Wilson IC, Rabon AM, Lipton MA. Enhancement ofimipramine antidepressant activity by thyroid hormone. Am J Psychiatry 1969; 126: 457 - 69.
Prange AJ, et al. Effects of thyrotropin-releasing hormone in depression. Lancet, 1979; 2 (Nov. 11): 999–1002.
Roche-Wien. Pituglandol advertisement, Wiener Medizinische Wochenschrift 1913 (June 21); 63: 1591.
Rubin RT, Carroll BJ. Claims for mifepristone in neuropsychiatric disorders: Commentary on DeBattista and Belanoff, and Neigh and Nemeroff. Trends in Endocrinology and Metabolism 2006; 17 : 384–5.
Sachar EJ, et al. Disrupted 24-hour patterns of cortisol secretion in psychotic depression. Archives of General Psychiatry 1973; 28: 19–24.
Selye H. The Physiology and Pathology of Exposure to Stress: A Treatise Based on the Concepts of the General-Adaptation-Syndrome and the Diseases of Adaptation. Montreal: Acta; 1950.
Shorter E, Fink M. Endocrine Psychiatry: Solving the Riddle of Melancholia. New York: Oxford University Press; 2010.
Thomson F, Craighead M. Innovative approaches for the treatment of depression: targeting the HPA axis. Neurochem Res 2008; 33: 691-707.
Wachtel LE, et al. ECT for catatonia in an autistic girl. AJP 2008; 165: 329-33.
Westermann J. Ueber die vitale Depression. Zeitschrift für die gesamte Neurologie und Psychiatrie 1922; 77: 391–422.
Wong M, et al. Pronounced and sustained central hypernoradrenergic function in major depression with melancholic features: relation to hypercortisolism and corticotropin-releasing hormone. Proceedings of the National Academy of Sciences 2000; 97: 325-30.
Yerevanian BI, et al. The dexamethasone suppression test as a predictor of suicidal behavior in unipolar depression. J Affective Disorders 2004; 83: 103-8.
Zimmermann W. Eine Farbreaktion der Sexualhormone und ihre Anwendung zur quantitativen colorimetrischen Bestimmung. Hoppe-Seyler’s Zeitschrift für physiologische Chemie 1935; 233: 257–64.