Has anyone ever gone through Prozac withdrawal?

The American novelist William Styron, in his 1990 memoirs, described his mood during a depressive phase in an unadorned and terrifyingly sober manner: "He (the psychiatrist) asked me if I was thinking of suicide, and I reluctantly admitted it. I did not tell details - that It wasn't necessary either - did not tell him that the things around me were there to be killed: the roof beams were inviting to hang up, as were the maple trees; the garage was a place to breathe toxic fumes, the bathtub a vessel, my blood The kitchen knives in their drawers had only one purpose for me. A fatal heart attack seemed particularly tempting, because then I had no responsibility; and I had also toyed with the idea of ​​deliberately getting pneumonia with a long walk in the woods in cold, damp weather in shirt sleeves If you consider running in front of a truck on the nearby highway ... Such horrible fantasies for healthy people are to the deeply depressed mind what lascivious daydreams represent to people with strong sexuality. "

This description makes it clear that a clinical depression is something completely different than a total dejection and displeasure caused by life circumstances, which everyone experiences at times, and which cannot be compared with pain and grief at the loss of a close person. Morbid melancholy takes away the life force to a much greater extent, is even more threatening, and the crushing sadness is also accompanied by a number of other symptoms. In addition to thoughts of suicide, many sufferers also torment themselves with feelings of guilt and deep inferiority complexes. They often find it difficult to think clearly, to keep things in mind, or to be happy about something at all. Depressed people often feel dominated by fear and sometimes completely lacking in energy and drive; many can barely eat and sleep, others want to do so in the extreme (box on page 76).

Psychologists and neurobiologists sometimes argue about whether depression is caused by traumatic experiences and self-reproach, or by biological processes. But the human psyche does not exist without the brain. There is ample evidence that the development of the disease, whatever the impetus, ultimately involves biochemical changes in the central nervous system. It is only they that ultimately cause the deep sadness and the other noticeable symptoms. Even if these brain processes cannot yet be fully described, research into them has made rapid progress in the last few decades and especially in the last few years.

At the moment it is still as if the representatives of the various disciplines were blindfolded each feeling a different part of the body of a huge, enigmatic creature and trying to create a meaningful whole from the conclusions. It could well turn out that the individual findings do not all fit into a uniform overall picture - namely because some patients may have different biochemical deviations in the foreground than others.

Still, the rapidly growing scientific evidence gives reason to believe that the biological determinants of depression can be understood in detail, which in turn will benefit diagnosis, treatment and prevention. Among other things, it is important to find out in which characteristics the individual patients differ. In some cases, the reduced activity of a certain neurotransmitter may be essential, in others, however, the overactivity of a hormonal track (while transmitters act as messenger substances transmit signals between directly interconnected nerve cells, hormones act via the blood on more distant tissues; the boundaries are of course fluid ). Another goal would then be to classify depressives accordingly with relatively little diagnostic effort. This would require measurable variables that are relatively easy to determine (biomarkers, to use laboratory jargon). This could be, for example, deviating levels of certain molecules in the blood, as well as changes in certain brain regions that can easily be visualized with imaging methods (box on page 80).

The next step would be, so to speak, tailor-made medication, i.e. special treatment for the individual anomaly - similar to how the general practitioner may quickly take a swab for a sore throat to determine whether and which antibiotics to prescribe. So far, psychiatrists have had to rely largely on intuition when choosing antidepressants. Often they have no choice but to try several remedies one after the other with a patient. Under certain circumstances, this means weeks and months of unrelieved agony - and in many cases an acute risk of suicide. (The psychotherapeutic treatment, which is often necessary at the same time, is usually not enough to improve the critical condition, especially not in very serious cases.) More rapid treatment is extremely urgent. Today's antidepressants have fewer side effects than previous ones, and in many cases they are of crucial help; Nevertheless, the astonishingly common disease still takes a heavy toll in suffering, human life and economic productivity.

It is estimated that in the United States between 5 and 12 percent of men and between 10 and 20 percent of women will go through a major depressive episode at some point in their lives, almost half of them more than once. Around every tenth patient, i.e. around 1.0 to 1.5 percent of the population, also experiences manic phases in alternation with the depressive phases, during which the patient is bursting with entrepreneurship, speaks a lot and very quickly, hardly needs any sleep, as well as overestimating himself. There is a tendency towards megalomania and potentially self-destructive activities, such as reckless and promiscuous sexuality, extravagance, or reckless driving.

The depressive phase, on the other hand, is literally murderous: around 15 percent of depressed and manic-depressive people kill themselves each year. According to surveys by the US Centers for Disease Control and Prevention, suicide was the ninth leading cause of death in the United States in 1996 (just behind AIDS): 30,862 cases were officially recorded, an estimated half of which were depression-related. The overwhelming opinion is that the number of unreported cases is likely to be very high, as many suicides are covered up for social or insurance reasons. Undoubtedly, some of the fatal car accidents are covert suicide.

The economic loss resulting from the disease is also substantial: in the United States, it was estimated to have totaled $ 43 billion in 1992, mainly due to reduced or absent work.

The effects of major depression on other medical conditions should also be considered. There is increasing evidence that such people are more likely to succumb to heart attacks or strokes. And cancer patients often also lose quality of life and possibly life time.

Hereditary disposition

Some of the earliest evidence of a biological component in many cases of depression came from genetics. Often the affliction occurs more frequently among close blood relatives. Parents, siblings and children of severely depressed or manic-depressive patients are much more likely to develop these or similar disorders themselves than the average in the population. The connection becomes even clearer in surveys of twins: among identical couples, both partners are much more likely to be manic-depressive than among dizygoti; in pure depression the difference is less pronounced.

For around 20 years, the responsible genes have been searched for, but so far in vain. Perhaps there are several hereditary factors involved in a predisposition to depression, each of which makes only a small, difficult-to-identify contribution.

Some time ago it was said that they had struck gold in a community of the fairly isolated Amish Mennonites, in which manic-depressive - so-called bipolar - psychoses occur quite frequently. However, the preliminary results that one or more genes on chromosome 11 make the disease susceptible were not confirmed.

In some cases the X chromosome could carry a gene involved; in most of the patients examined, however, no such connection was found. Recently, various regions on chromosome 18 and one location on chromosome 21 have also been suspected - but confirmation is still pending.


Other scientists focus on neurochemistry, largely on the messenger substances that transmit signals to points of contact between brain neurons. In many cases, depression appears to be at least partially due to disorders in neuronal circuits that work with norepinephrine or serotonin. Like the well-known dopamine, both belong to the monoamines and are derivatives of amino acids.

Psychiatrists became aware of this class of drugs in the early 1950s when severe depression was a side effect of 15 percent of prescribing reserpine for high blood pressure. As it turned out later, the drug causes depletion of the monoamines. Conversely, at about the same time, an anti-tuberculosis drug was found to improve the mood of some depressed lung patients. Research has shown that its active ingredient inhibits the breakdown of monoamines (which is carried out by the enzyme monoamine oxidase), i.e. it probably ensures that these transmitters in the brain circuits remain effective for longer. All of this suggested that abnormally low monoamine levels might be causing depression. Because of this, monoamine oxidase inhibitors were developed as the first class of antidepressants.

But which monoamines were most important? Joseph J. Schildkraut from Harvard University in Cambridge (Massachusetts) advocated norepinephrine, which is chemically a catecholamine at the same time, in the 1960s. In his now classic "catecholamine hypothesis for mood disorders" he suggested that the depressive state was due to too little, while the manic state was based on too much norepinephrine in certain circuits. The theory has now been refined, because not every person's mood changes with a decrease or increase in the concentration of this neurotransmitter. But there is considerable experimental support for a connection in the case of depression. The relevant neural pathways run from the brain stem, especially from the blue core (Locus coeruleus), to many regions, including the limbic system - a group of different structures that are crucially involved in the regulation of emotions (Fig. 2).

To better understand how norepinephrine and other monoamines work, I need to go into a little more detail. At the synaptic contacts, these molecules - like all neurotransmitters - pass through a narrow gap from the upstream cell that releases them to the downstream cell. There they attach themselves to receptor molecules (Fig. 3), and this in turn triggers a reaction that can manifest itself in an excitation or in an inhibition of the electrical activity of the recipient cell. The particular effect of a neurotransmitter depends largely on the type and density of its receptors on the recipient side. For example, there are at least 13 subtypes for serotonin, which differ in their sensitivity to the transmitter and in the effects produced.

The strength of the signal transmission can also be influenced by the amount of neurotransmitter released and by how long it remains in the synaptic gap. At least two types of molecules on the sender cell play a role here: autoreceptors and return transporters (Fig. 3). The former inform your cell how much transmitter it has released and thus signal to it to reduce the release. The return transporters, on the other hand, literally pump the messenger substance back into the cell; they are therefore also called recovery or return transports. Inside, the enzyme monoamine oxidase can break down neurotransmitter molecules, so that the amount that can be released immediately is reduced.

Indirect measurements of the messenger substance in urine or in the cerebrospinal fluid, for example, indicate a possible connection between depression and noradrenaline deficiency in brain synapses. According to a number of studies, many moody people show noticeably few degradation or remodeling products of this neurotransmitter. It is fitting that an increased density of certain noradrenaline receptors has been found in the cerebral cortex of depressed suicide victims. This is by no means a contradiction, on the contrary to be expected: the organism is apparently trying to compensate for the deficiency. In fact, such a compensation can often be observed with a permanent deficit of transmitters - as if the information-hungry cells wanted to pick up every signal, no matter how small.

The catecholamine, more precisely norepinephrine, hypothesis is further supported by the fact that new drugs which selectively inhibit the retrieval of norepinephrine from the synapse gap and thereby increase its length of stay there have an antidepressant effect in many patients. One of these compounds, reboxetine, is already approved as an antidepressant in some countries. In spite of everything, the focus of research has not been on norepinephrine anymore, but on serotonin for several years - thanks to the success of fluoxetine and related antidepressant agents that start here.


The role of this transmitter in mood disorders has been seriously investigated for almost 30 years - namely since Arthur J. Prange jr. from the University of North Carolina at Chappel Hill and Alec Coppen from the British Medical Research Council in London and their colleagues formulated the thesis that serotonin depletion promotes or "allows", as they called, a decrease in the amount of norepinephrine.

This is entirely possible, because serotonin-producing neurons move from the raphe nuclei of the brainstem to various parts of the brain (and the spinal cord), including nerve cells that release norepinephrine or control its release. Effects on other types of neurons are also likely to play a role, as serotonin-producing cells send offshoots into many areas of the brain that are believed to be involved in depressive symptoms. These include the almond kernel (important for emotions, for example), the hypothalamus (important for appetite, libido and sleep, among other things) as well as areas of the cerebral cortex that are involved in cognitive and other higher performance (Fig. 2).

In the cerebral spinal cord fluid of depressed patients, especially in suicide candidates, decreased levels of a main breakdown product of serotonin have been measured, which suggests a decreased release of the transmitter in the brain. In addition, a characteristic surface molecule, by which one recognizes serotonin-releasing cells in the brain, is found less often than usual; accordingly, the number of these cells could be abnormally small. In contrast, in the brain tissue of deceased patients, at least one form of the serotonin receptor, type 2, occurs in greater density - similar to noradrenaline, probably a measure to counter the lack of messenger substance in the synaptic gap.

The remarkable therapeutic successes with pharmaceuticals which block the return transporter for serotonin and thus prevent the released transmitter from quickly disappearing from the synaptic gap are also revealing. At the end of the 1950s, the monoamine oxidase inhibitors mentioned above were joined by tricyclic antidepressants (named after the three rings in the basic chemical structure). At that time, their mechanism of action was not yet understood. It was only later that a number of effects were discovered, including the reduced return transport of serotonin, which increases its amount in the synaptic cleft. (In fact, the organism seems to be trying a similar remedy; in depressed people, the density of the transporters is apparently lower.)

This has long been suspected of being the reason for the antidepressant effect; However, this could only be confirmed with the introduction of preparations in the late 1980s that selectively block the serotonin return transporters without affecting other monoamines in the brain (initially fluoxetine, then, for example, paroxetine, sertraline, fluvoxamine). These active substances, known as SSRIs (for English selective serotonin reuptake inhibitors), have revolutionized the treatment of depression: They are highly effective and have far fewer side effects than older drugs. Even newer antidepressants, such as the substance venlafaxine, block both the absorption of serotonin and that of norepinephrine.

Investigations on an analogous serotonin system - that of the blood platelets - also gave rise to new clues as to why depressed people are more prone to heart attacks and strokes. This life-threatening organ damage is primarily caused by blood clots that block supplying vessels. Such a plug, usually thought of as a wound closure, forms when certain factors cause platelets to clump together. As research in my laboratory and elsewhere has shown, the platelets of depressed patients are particularly sensitive to activation signals, and apparently also to those from serotonin; it is one of their releasable ingredients and in turn makes them more responsive to other, more powerful chemical stimuli. It has also been shown that the platelets of depressed people carry fewer transporters for serotonin, so that they are probably more difficult to suck it out of their environment in order to reduce their exposure to activation signals. (These tiny blood cells, which are basically fragments of special giant cells, do not produce it themselves, but take it up from the blood in the area of ​​the intestinal wall, where serotonin producers are located.)

Hormonal disorders

Depression is often associated with a malfunction of the circuits in the brain that control the activity of certain hormones. In fact, hormonal changes in this disease have long been known.

The regulation of hormone release is organized hierarchically. At the top is the hypothalamus in the diencephalon. It produces special peptides that act on the pituitary gland (pituitary gland) at the base of the brain; these short amino acid chains stimulate or inhibit the release of various hormones into the blood, which control the release of other hormones in their various target glands. These in turn have a negative feedback effect - except on non-neural tissue and organs - on the pituitary gland and hypothalamus, which prevents excessive hormone production (Fig. 4).

As has been shown several times, depressed patients respond poorly to a number of substances that normally stimulate the release of growth hormone from the pituitary gland. The reaction to the hypothalamus substance, which otherwise induces the secretion of thyrotropin, is also abnormal; this pituitary hormone controls thyroid function, among other things. If, by the way, antidepressants don't work, an undetected hypothyroidism is often the cause.

So far, however, the most convincing evidence has been the evidence of incorrect regulation on the so-called hormonal stress axis, which is used to react to physical and psychological threats. This non-neural rail between the hypothalamus, pituitary gland and adrenal cortex is called the HPA axis for short after the English expression for the organs concerned. As soon as a danger is recognized, the hypothalamus increasingly forms the corticotropin-releasing factor (CRF), which causes the pituitary gland to release corticotropin; this is also called adrenocorticotropic hormone, or ACTH for short, because it stimulates the adrenal cortex to release the stress hormone cortisol (Fig. 4).

The entire process makes the organism ready to face the threat - for example to fight - or to evade it by fleeing; at the same time, unnecessary, obstructive or distracting activities are switched off at the moment (Spektrum der Wissenschaft, May 1993, page 92 and page 97). For example, cortisol increases the fuel supply to muscles; at the same time, CRF suppresses hunger and sex drive and increases alertness. The system is vital to managing dangers. Chronic activation of the stress axis, on the other hand, can become a breeding ground for illness - and, it seems, also for depression.

As early as the late sixties and early seventies, several research groups reported an overactive HPA axis in depressed patients who were not treated with medication. Among other things, an increased cortisol level in the urine, in the blood and in the cerebrospinal cord fluid (the liquor) spoke for it. Depressed people suffer to a considerable extent from this overactivity - especially the most severe cases. This has been confirmed by hundreds, if not thousands, of follow-up studies - arguably the most repeated evidence in biological psychiatry.

More detailed research into the phenomenon has now revealed deviations from the norm at all levels of the HPA axis. The adrenal glands, like the pituitary gland, are enlarged and release too much cortisol. The main culprit that this axis is overactive and symptoms of depression appear, lies with another entity: the CRF-producing neurons in the hypothalamus and elsewhere. In any case, many researchers are now convinced of this, including us at Emory University in Atlanta (Georgia).

It is impressive how one study after the other showed increased CSF values ​​for CRF - in comparison to healthy people as well as people with other mental illnesses. When treated with antidepressants, the values ​​decrease - by the way, even with successful electroconvulsive therapy. As the examination of the brain tissue of deceased patients showed, both the number of CRF-producing neurons in the hypothalamus and the activity of the CRF gene are significantly increased in comparison to healthy persons; each individual cell therefore forms a particularly large amount of the factor. When it was applied to animals' brains, they too showed behavioral symptoms that are among the cardinal features of depression in humans: insomnia, low hunger, reduced sex drive and excessive anxiety (see box in Figure 4).

It is not yet entirely clear how the various findings on genetic disposition, neurotransmitters and hormones fit together into a unified picture - if they ever do. At least a partial scenario can be devised that could explain in part how people with traumatic childhood are prone to depression. I call it the stress diathesis model of mental illnesses, because it contains the interaction between stressful experience (stress) and predisposition to illness (diathesis).

Disposition and stress

In families with a hereditary predisposition to depression, certain genetic traits are likely to somehow lower the critical threshold for the onset of the disease. They could have the direct or indirect consequence that synapses have less serotonin or noradrenaline available or that the HPA stress axis overreacts. A genetic disposition does not necessarily lower the individual threshold from the outset to such an extent that depression would arise even without stressful influences; However, through early traumatic life experiences it could be further degraded and thus brought closer to the critical limit.

My colleagues and I imagine that neglect or abuse not only simply activates the stress response, but also causes the activity of the CRF neurons to increase in the long term (the cells that respond to stress and that are hyperactive in depression ). If these remained hypersensitive into adulthood, they would respond violently to even mild stressors (burdensome factors). In people with a corresponding genetic disposition, both together could ultimately cause the behavioral changes and neurohormonal deviations typical of depression.

To test the stress diathesis hypothesis, we performed a series of experiments on social neglect on newborn rats. During the first three weeks of life, we took her away from her mother briefly on about ten days; after they were weaned, they then grew up in a standard rat colony without further interference. When these occasionally uterine deprived animals grew up, changes were actually found in the CRF neurons, all of which went in the same direction as in depressed patients: for example, the pituitary gland released excessive amounts of ACTH in stressful situations, and there was ACTH in some areas of the brain the CRF concentration increased; The same applied to the blood level of corticosterone (which in rats functions as the human stress hormone cortisol). Accordingly, the young animals, repeatedly separated from their mothers, seemed to produce too much CRF all the time, as if the responsible gene was now permanently overactive. My colleague Paul M. Plotsky has now been able to confirm this in a further experiment.

Even the density of CRF receptors in certain brain regions was higher than normal in the deprived rats. In this case, however, this could hardly be an attempt to compensate for a deficiency - there was even too much CRF available. Still increasing the number of receptors is the worst possible reaction; because a permanently higher density should increase the effect of the hormone - and thus its depression-inducing effects as well as those of stress.

A preliminary finding by Plotsky is downright exciting, according to which the CRF content normalizes in such rats when they are treated with the selective serotonin reuptake inhibitor Parotexin. Apparently, too high a sensitivity or number of CRF receptors is fully compensated; because the corticosterone level also normalizes, i.e. the hormone secretion of the adrenal glands. The same applies to behavior: the rats, for example, are no longer so fearful.

We don't know exactly what this success is based on, so how the drug calms the HPA axis, so to speak. But the finding suggests that serotonin reuptake inhibitors could be particularly helpful in depressed patients with childhood trauma. In the rats, however, any abnormalities related to the stress axis and CRF recur after drug withdrawal. This could mean that patients with comparable symptoms would have to continue taking the drug even after the depression has subsided if relapses are to be avoided.

In experiments with Indian hat monkeys - which, as primates, are more likely to be more like humans emotionally than rodents - the results were similar to those of the rats. In the first three months after the birth of a child, the mother monkeys were kept under three different conditions: some were always fed abundantly, others always scarce; the third group, at irregular intervals, received a lot of food, and then little. This unmanageable situation frightened and preoccupied the female monkeys to such an extent that they no longer took proper care of their offspring. As predicted by our model, the young animals in this test group were less active and enterprising than those of the others, avoided playing together and fell into a state of fright when something unusual happened - typical behavior in social deprivation. As adults, they had significantly increased levels of CRF in the spinal fluid.

Studies like this are of general concern. More than three million cases of child abuse and neglect were reported in the United States in 1995 alone, and at least a third of these are officially confirmed. If the effects in children are actually similar to those in the animals examined, changes under these circumstances would also be expected, which would manifest themselves in the maturing brain. The consequence of this would be that the release of CRF and the sensitivity to the substance are chronically too high, and would make those affected more susceptible to depression for the rest of their lives.

If these assumptions prove to be true, it will certainly be an important task to find out whether the later risk of depression in abused or abused children can be identified with non-invasive examinations by examining the activity of the CRF-producing neurons or the number of CRF receptors with imaging Procedure determined (box on page 80). It would also be desirable to know whether children at particular risk can be protected from depression with antidepressants or other measures such as psychotherapy. Likewise, one should research whether sick adults with a childhood trauma should best take antidepressants constantly and whether the medication or perhaps psychotherapy can really normalize the activity of the CRF-producing neurons.

The stress diathesis model does not, of course, explain all cases of depression; not everyone or every sick person was ultimately abused, ill-treated or neglected as a child. However, where a badly coped childhood experience comes together with a family history, there seems to be an unusual risk. Those who do not have a genetic disposition (which can be assumed if the condition does not occur in any relative) should be relatively well protected from severe depression, even if they had to go through bad traumatic experiences in their youth or later. Conversely, some who have a pronounced predisposition to the disease will later suffer from it, although nothing of the sort has happened to them.

Further studies on the neurobiological background of the disease are undoubtedly necessary, but at least the knowledge gained so far is already being reflected in new ideas about where the medicinal lever could be applied. Inhibitors for CRF receptors are under development at several pharmaceutical companies (such a substance, which has shown its antidepressant potential in animal experiments, is soon to be tested on humans in Germany, for example). Another promising class of drugs activates specific types of serotonin receptors; this could possibly be very effective in treating depression without inevitably stimulating the other, non-involved receptor types at the same time. With certainty, new findings will enable better and better therapies with fewer side effects. The more one understands the neurobiological basis of emotional disorders, the more purposefully and efficiently one should be able to intervene.


- Network people. On the trail of the connections between body and soul. By G. Miketta. Rowohlt, Reinbek 1994.
- Biology of Depressive Disorders, Parts A and B. Edited by J. John Mann and David J. Kupfer. Plenum Press, 1993.
- The Role of Serotonin in the Pathophysiology of Depression: Focus on the Serotonin Transporter. By Michael J. Owens and Charles B. Nemeroff in: Clinical Chemistry, Volume 40, Issue 2, pages 288-295, February 1994.
- Biology of Mood Disorders. By K. I. Nathan, D. L. Musselman, A. F. Schatzberg and C. B. Nemeroff. APA Press, Washington D.C., 1995.
- The Corticotropin-Releasing Factor (CRF) Hypothesis of Depression: New Findings and New Directions. By C. B. Nemeroff in: Molecular Psychiatry, Volume 1, Issue 4, pages 336 to 342, September 1996.
- Psyche, stress and defense against illness. By Esther M. Sternberg and Philip W. Gold. Spectrum of Science, November 1997, pages 64 to 71

From: Spektrum der Wissenschaft 8/1998, page 74
© Spektrum der Wissenschaft Verlagsgesellschaft mbH

This article is contained in Spectrum of Science 8/1998