How Everyone Became Depressed (24 page)

In 1973, the Angelini Company launched trazodone, a chemically novel compound, in Italy as an antidepressant, Trittico, and by 1976 were flogging it for “nervous exhaustion” (esaurimenti nervosi). When Mead Johnson & Company acquired the American license in 1982, they decided to take it, rebranded as Desyrel, in a different direction: “It does not act by CNS stimulation, but rather, selectively inhibits serotonin uptake in the brain.”
⁴⁶
Trazodone, which has only a weak effect on serotonin, was billed as the first selective serotonin reuptake inhibitor, the first SSRI! (The drug performed poorly as an antidepressant but has recently undergone a revival as a hypnotic, or sleeping agent, and is now widely prescribed.)

The avalanche of marketing neurotransmitters to gain market share roared louder. In 1983, 4 years after its launch in the United States, Ciba started hyping its tetracyclic antidepressant maprotiline (Ludiomil) as a norepinephrine reuptake blocker: There was an impressive artist’s drawing showing serotonin and norepinephrine molecules in the end bulb of a neuron awaiting their discharge into the synapse, much as fairies await the advent of Midsummer to come and dance on the lawn.
⁴⁷
Much more colorful art, designed to grip the prescriber’s mind as powerful science, was to follow.

There were inconvenient scientific findings. How curious, said biological psychiatrist Edward Sachar, head of psychiatry at the New York State Psychiatric Institute, in 1979, that the recently discovered tricyclic mianserin is effective in depression, yet has “no effect on reuptake of serotonin or noradrenaline.”
⁴⁸

In the background, doubts grew stronger. In 1978 Seymour Kety, the dean of neuroscience in the United States and by now head of biological psychiatry at the Mailman Research Center of McLean Hospital in Belmont, Massachusetts, said that “The simplistic notion of ‘one neurotransmitter— one function’ is no longer tenable, and we are beginning to recognize that complex psychological functions can only be the result of complex interactions among neuronal circuits and the transmitters that mediate them and modulate them.”
⁴⁹

Across the Atlantic, other lights scoffed as well. At a conference in London in 1979, Laurent Maitre, a scientist in the research department of Ciba-Geigy, said that “it is very doubtful that there is a correlation between 5-HT [serotonin] uptake inhibition and clinical effects.” He was dubious about analogizing research on the inhibition of reuptake in platelets, where much previous work had been done, on a one-to-one basis to the brain.
⁵⁰
At a meeting in Monte Carlo that same year, Alain Puech, a pharmacologist in Paris, called attention to the absurdity of the notion of influencing just one neurotransmitter system: “It is inconceivable that modifying one monoaminergic [serotonin, dopamine etc] system can be done independently, because all the circuits are linked in a chain. Thus, every time there’s a change in one neurotransmitter system, there is very probably at least one other neurotransmitter system that is modified.”
⁵¹

Such reservations echo down massively from these years. A meta-analysis in 2009 led by Kathleen Merikangas at the National Institute of Mental Health found no evidence of a link between the region in the genome that controlled the reuptake of serotonin and the incidence of depression.
⁵²
Science raced on, from studying reuptake in the “presynaptic” neurons, to studying actions on the membranes of the postsynaptic neurons, to “second messenger” biochemical events involving protein kinases within the neurons themselves, to happenings within the cell nucleus, which is where we are today. But little of this science reached the ears of the community psychiatrists, internists, and family physicians who were prescribing the Prozac-style antidepressants.

What is happening here is the turning of a massive crank, to make the concept of depression scientifically acceptable to physicians, and thus indirectly appealing to their patients. Yet another crank was simultaneously grinding as well: the media crank that sold the neurotransmitter doctrine of depression directly to a mass audience. From the 1970s onward, the public was increasingly exposed to the concept of neurotransmitters as the scientific basis for confidence in the antidepressants and the whole doctrine of depression, on the logic that if it responds to Prozac, it must exist. A search of the New York Times digital database for the years 1955 to 2000 revealed the following number of hits for the word “neurotransmitters”: 1955–1960, 0; 1961–1970, 1; 1971–1980, 17; 1981–1990, 108; 1991–2000, 171.
⁵³
The number of references is not large, but the trend is clearly skyward.

Among the three main neurotransmitters in affective disorders, serotonin got the greatest play in the Times: 18 mentions in 1955–1960, rising to 67 in the 1980s and 168 in 1991–2000. Dopamine and norepinephrine displayed similar trajectories with fewer hits. There is no doubt that by 2000 readers of the New York Times had been fully exposed to the latest science on serotonin. Other important developments remained, however, quite uncovered, and a search on the dexamethasone suppression test, which would have revealed the state of the art on psychiatry’s only biological marker, turned up only three mentions for the entire period from 1955 to 2000. The coverage, therefore, reflects what readers wanted to see rather than what neuroscientists thought important—and readers wanted to know that powerful scientific developments had uncovered the key to their dysphoria, which everyone was happy to define as “depression.”

Thus, just as popular culture was turning toward neurotransmitters as the explanation of what ailed everybody, academic culture was turning away from them. In 1985, Eugene Paykel, among the most distinguished international figures in the field as professor of psychiatry at the University of London and editor of the Journal of Affective Disorders, asked rhetorically at a conference: “Is the amine hypothesis of depression dead?” meaning the hypothesis that viewed depression as a disorder of serotonin and norepinephrine. “Diabetes comes down very clearly,” he said, “to a relative deficiency in insulin. Fifteen years ago depression was like that; we thought it was a clear-cut deficiency of noradrenaline or serotonin in a particular neurotransmitter system in the brain. But it’s not as simple as that.”
⁵⁴
Was the whole line of inquiry dead?

Three years later, in 1988, the response of the pharmaceutical industry to this massive scientific dubiety was to launch Prozac (fluoxetine), the first of the drug class that shortly became known as SSRIs, the most popular drug class in history. “The first highly specific . . . highly potent blocker of serotonin uptake,” screamed Lilly in its launch ad in the Journal of the American Medical Association in June.
⁵⁵
Using a model of drug action that had been largely discredited, Lilly and the other SSRI manufacturers went on to promote a drug class—and reap billions of dollars in profit—that captured the imagination of the public and the medical profession not just in the United States but in the whole world: Depression, a disease suffered sooner or later by half of the entire population (it was argued),
⁵⁶
could be cured with a drug that made your serotonin right again! Is it any wonder that depression spread epidemically within the medical profession as a diagnosis and within the public as a supposed disorder? You are wondering why everyone became depressed? This is why.

Motto: “A patient whose high numbers are for things like sleep and fatigue has a different kind of ‘depression’ than somebody who mainly endorses hopelessness,

anhedonia, and guilt.”
Herbert Peyser, senior psychiatrist, New York City, post to a psychopharmacology list-serve, January 8, 2010

Before 1980 there had been two depressions, melancholia—also called endogenous depression—and nonmelancholia, called a number of terms such as reactive depression and neurotic depression.
¹
DSM-III flattened this distinction, abolishing the clinical distinction between the two with the homogenizing term major depression. To be sure, DSM-III reinserted the term melancholia in the discussion as a subtype of major depression, but only in letter, not in spirit. In the decades after 1980 melancholia returned, but to a landscape of mood disorder that had been leveled and laid waste by the concept of “depression.” In a world where everybody is depressed, nobody is melancholic.

Emil Kraepelin had sent the diagnosis of melancholia into a death spiral. The psychoanalysts had little interest in the concept, aside from venerating a single essay of Freud, and the only people interested in keeping melancholia alive as a notion after the 1930s were the British who, with their admixture of Heidelberg science and homegrown common sense, had turned into shrewd psychopathologists. The textbook that Willi Mayer-Gross, a Heidelberg refugee, published in 1954 together with Eliot Slater and Martin Roth gave pride of place to involutional melancholia as the serious melancholic illness that often affected people at midlife and afterward.
³

But world psychiatry after W orld War II marched to an increasingly American beat, and the Americans had little use for the antique term melancholia. The glossary of Alfred Freedman’s Comprehensive Textbook of Psychiatry, the world’s leading textbook first published in 1967, had scads of psychoanalytic terms but claimed of melancholia: “Old term for depression that is rarely used at the present time.”
⁴
In Europe after World War II, endogenous depression was the serious variety and melancholia was deemed as “contaminated by Freud and the 19th century novels that degraded it to grief,” as Tom Ban, who trained in Budapest in the early 1950s, put it. “For Kraepelinians, grief and depression were not the same, and they excluded each other. Anyone who had an identifiable precipitating factor could not be labeled as having a depressive state.”
⁵
In 1978 Pierre Pichot, one of France’s leading psychopharmacologists, found the term “m é lancolie” useless and urged that it be abandoned.
⁶
So on both sides of the Atlantic, melancholia was toast.

Yet melancholia as an illness has such a distinctive clinical profile that images of it remained in the minds of experienced clinicians: the expressions of utter hopelessness and despair, the slouched shoulders, and the tight little smile intended to conceal firm plans of suicide. And then a week later the patient is found swinging from the underside of a bridge. This was a special horror that engraved into the collective memory of the profession the engram that there was, after all, such a thing as melancholia, and that it was different from every other illness entity in psychiatry. So even though official academic psychiatry had long lost interest, a subterranean stirring of curiosity remained.

Yet the crucial event in reviving the concept of melancholia as not just a severe form of depression but a different illness was Roland Kuhn’s discovery in the mid-1950s of a drug specific for what Kuhn was calling “vital depression” but was in fact melancholic illness. The drug was the first tricyclic antidepressant (TCA), imipramine, marketed by the Geigy Company in Switzerland in 1957 and in the United States in 1959 under the brand name Tofranil.

For Kuhn, an experienced psychiatrist, the effect of imipramine on hospital depression was nothing less than extraordinary. On August 11, 1956, he wrote to Robert Domenjoz, chief of pharmacology at Geigy, on the workings of the drug. “In the beginning, the effects are sedative. The patients sleep better and are calmer during the daytime.” Yet the sedation, Kuhn said, was less intense than with the antipsychotic drug chlorpromazine, and “during the day, the patients feel less sleepy and are less lethargic. In many cases, the hallucinations become softer and seem to retreat into the distance. Illusions and delusions lose their threatening character. The train of thoughts often runs more smoothly and the patients become more relaxed when dealing with other people.” In contrast to chlorpromazine, Kuhn continued, imipramine “has an obvious effect on depression. The vital depression visibly improves. The patients feel less tired, the sensation of weight decreases, the inhibitions become less pronounced, and the mood improves. . . . The patients also become more open to psychotherapeutic efforts and it even becomes possible to establish psychotherapeutic contact with people who have very severe depressions, with whom one could not have a meaningful conversation prior to the treatment.”
⁷

The chart of the first patient treated with imipramine at the Münsterlingen Mental Hospital, Paula G, 49 years old, has survived. The entry for January 21, 1956: “For three days now,” Kuhn wrote in the chart, “it is as if the patient had undergone a transformation. All of her restlessness and agitation have vanished. Yesterday, she herself observed that she had been in a complete muddle, that she had never acted so dumb in all her life. She did not know what had caused her behavior, but she was just glad to be better again . . . The patient is also different now from before the depressive phase. She is no longer so aggressive and quarrelsome. She is energetic and friendly, likes to read and work. Her sleep is now also markedly better.”
⁸

Kuhn’s observations mark the opening of a remarkable new chapter in the history of the treatment of mood disorders, and emphasize at the same time that his patients at Münsterlingen had a very special kind of depression in which they were psychotic, unreachable, and gravely disabled. Imipramine’s launch sent the message that these kinds of depressions were now treatable with something other than electroconvulsive therapy (ECT), which had been introduced in 1938. Indeed, several of Kuhn’s patients had failed to respond to ECT.

Vital depression did not readily catch on as a term, but neither did melancholia in these years. The term that psychiatry used for these serious, often psychotic, mood disorders was endogenous depression. And opinions drifted back and forth about whether endogenous and reactive depressions were just two different levels of severity or quite different illnesses.

Bernard Carroll, his training Australian but his scientific career passed largely in the United States, was at the University of Michigan as he attempted to shore up his initial discovery in 1968 that melancholic depression had a biochemical marker in the form of the dexamethasone suppression test (DST, see p. xxx).
⁹
Patients with what Carroll called “serious depression” and “endogenous depression” failed to suppress the secretion of cortisol after the administration of a dose of the artificial steroid dexamethasone the night before; the adrenal glands of patients with lesser depressions and other psychiatric disorders suppressed cortisol secretion whether they had received dexamethasone or not. This is the most fundamental biological finding in psychiatry, and it would be pleasant to argue that Carroll did much to reestablish the concept of melancholia as a distinctive entity. But things did not work out that way. Carroll began using the term melancholia only around 1981,
¹⁰
and his work on the DST, that by rights should have merited him great honor, was forgotten in psychiatry in the 1980s in a dismal episode that Max Fink and I have described elsewhere.
¹¹
The bottom line is that Carroll’s work was elbowed aside right about the time that melancholia began its great resurgence.

What caused the resurgence? It was partly the inclusion of the diagnosis of melancholia in DSM-III in 1980 and partly the progression of scientific curiosity.

The APA’s Diagnostic and Statistical Manual did reimport the term melancholia into psychiatry, albeit as a subcategory of major depression, simultaneously abolishing endogenous depression. This put melancholia again on the radar, although, as we saw in a previous chapter, in an attenuated form. You would qualify as melancholic if you had low appetite, awakened early in the morning, felt really guilty about something, and good news did not make you feel better. This is a pale shadow of the real McCoy, with its psychotic desperation, deep psychic pain, and massive changes in bodily function. DSM-III was also silent on other kinds of information useful in making the diagnosis, such as past history, family history, biological measures, and nonverbal behavior.

From the outset, it was clear that the DSM criteria of melancholia did not aptly characterize the illness. If, for example, you were looking for “a distinct quality of depressed mood,” as the Manualspecified, you might miss melancholia in your patients. Robert Priest, a psychiatrist at St. Mary’s Hospital Medical School in London, remembered a patient, “a chubby person,” whose symptoms were “of melancholic depression, some of them delusion.” “I’ve committed the most awful sins,” she told him. “I’m riddled with disease.”

But from the live interview he got a rather different impression. “She had a ready smile and after each proclamation of disaster she would grin, giggle or in some way suggest that she was far from feeling miserable.” Priest took her statements to be fictitious, “falsehood at an unconscious level” and “self-deception of a hysterical nature.” Having missed the diagnosis of smiling depression, he took no particular action.

“The outcome was tragic. I forget exactly how many days afterwards . . . she was found dead, floating in a nearby pond. The consultant, a Scotswoman, was horrified at my diagnosis of hysteria.”
¹²

So although clinical diagnosis is often pattern recognition, and the classic melancholic pattern is instantly recognizable, not all cases are classic and if you have only DSM “phenomenology,” as it is called, you might tragically miss the diagnosis.

Research started to be conducted using the new DSM criteria. In 1981, Craig Nelson and Dennis Charney, both at Yale, singled out lack of reactivity to good news as the most characteristic symptom of the melancholia subtype of DSM’s major depression (psychomotor change also ranked highly).
¹³
These were pretty generous criteria: A study of 800 women at midlife in Gothenburg, Sweden, in 1984 found 6.9% of them currently had major depression: 2.9% had melancholia and 4.0% had nonmelancholia.
¹⁴
(Informed observers, mindful of the tremendous difficulty of gaining true prevalence rates, estimate the lifetime prevalence of unipolar melancholic and psychotic depression in the community to be about 2–3%.
¹⁵
)

How else was melancholia different? It was discovered, or rediscovered, to be quite unresponsive to psychotherapy and placebos. This was actually something long known. In 1930 Henry Yellowlees, medical superintendent of The Retreat at York and lecturer in psychological medicine at St. Thomas’s Hospital in London, told the Section of Psychiatry of the Royal Society of Medicine that, in the words of the secretary, “Melancholia was a psychosis, and it was, ipso facto, not amenable to psychotherapy. Neurasthenia, on the other hand, was amenable to psychotherapy in general, and was often especially amenable to [psycho]analysis in particular.”
¹⁶
This insight was then forgotten in the decades ahead as melancholia went out of style and psychotherapy became the treatment of choice for many conditions in psychiatry. In 1988 Duane Spiker and David Kupfer at the Western Psychiatric Institute in Pittsburgh revived this classical truth, finding the response rate to placebo in psychotic depression zero.
¹⁷

Then there was the revival of sheer scientific curiosity about melancholia, stimulated perhaps by the discovery that some depressions were unresponsive (“treatment resistant”) to conventional antidepressants. Many of these resistant depressions responded to ECT, and ECT in the 1980s was beginning a comeback.
¹⁸
It is actually quite remarkable that ECT needed a comeback because the treatment, launched in 1938 and entailing the induction of therapeutic convulsions with electricity, is the most powerful therapy that psychiatry has to offer. It was almost driven off the boards in the 1960s by the flower children (and by a systematic campaign of the Church of Scientology). Yet ECT did hang on because some patients will respond to nothing else and will kill themselves if they do not get shock treatment. In the 1980s common sense conduced to a revival of ECT, and this reanimation of convulsive therapy doubtlessly spurred a revival of curiosity about melancholia.

There was also growing awareness that melancholia responded better to the TCAs than to the Prozac-style drugs that, from the late 1980s, began to run the board in psychiatric treatment. In 1996 Paul Perry at the University of Iowa, in a review of controlled trials, found that the TCAs were “consistently more effective than the SSRIs” in “endogenous/melancholic depression.”
¹⁹

Thus, research began to edge away from the DSM subtype and toward a restoration of the concept of two depressions, one of which would be melancholia independent of major depression, although the term was not yet current. Danish investigators were leading figures in this work. As early as 1979 Per Bech and Ole Rafaelsen at the Psychochemistry Institute of the university psychiatric hospital in Copenhagen, both major figures in international circles, had proposed at an academic meeting a “Bech-Rafaelsen Melancholia Scale.” They published a comparison with the famed Hamilton Depression Scale, the field’s standard, in 1980.
²⁰
Eight years later, Bech, by now in the department of psychiatry at Frederiksborg General Hospital in Hillderod, felt the evidence on the dissimilarity of outpatient and inpatient depression was sufficient that they should be treated as “two different diagnostic entities.”
²¹
These Danish neuroscientists disliked the term depression. Tom Ban, who knew Rafaelsen well, said of him later, “Ole was quit explicit that the word depression should be dismissed because people on the street associate it with something different from psychiatrists. Psychoanalysts extended it in a way that one can use it practically for any somatic complaint, and biologically oriented psychiatrists use it inevitably to mean that depression is a disorder of arousal or of the EEG [electroencephalogram].”
²²