Night Falls Fast Read online

  Some of the marginalization of lithium is due to other important advances in medical research. Many new medications used for treating mood disorders—the anticonvulsant drugs (which were first used to treat epilepsy, but now are used to treat manic-depression, as well) and the newer antidepressants, for example, the selective serotonin reuptake inhibitors, such as citalopram, fluvoxamine, paroxetine, fluoxetine, and sertraline (Celexa, Luvox, Paxil, Prozac, and Zoloft, respectively)—are more easily administered than lithium by general practitioners, internists, and psychiatrists. This ease of prescription is largely to the good, although it makes it more likely that highly effective and relatively inexpensive drugs such as lithium—which, in fact, is generally not that difficult to prescribe or monitor properly—will be bypassed for other, better-marketed drugs. It also increases the likelihood that the more popular and easier-to-prescribe antidepressants may be given to patients who would benefit more from a mood-stabilizing drug such as lithium and who may actually get worse on antidepressants (that is, their episodes may increase in frequency and intensity, and they may experience severely agitated or mixed states). Often antidepressants and mood-stabilizing drugs need to be used together in order to obtain the best therapeutic results.

  In recent years, advances in psychiatric research have made the highly profitable marketplace for mood-altering drugs far more competitive. Patients unresponsive to lithium or unwilling to take it now have good alternatives available to them. The most commercially successful of these, valproate (Depakote), an anticonvulsant, has now overtaken lithium as the most widely prescribed, and often the first prescribed, medication for bipolar disorder, or manic-depressive illness. This is a striking reversal in prescription patterns. There has also been a marked increase in the total number of prescriptions written for depression and bipolar disorder over the course of the past five years (a trend even more dramatic for the prescription of antidepressants), which reflects an increase in media and public awareness of the availability of effective medical treatment for mood disorders; impressive educational work on the part of patient advocacy groups; and highly effective physician and public marketing campaigns financed by the major pharmaceutical companies.

  The ability of the anticonvulsant medications (valproate, carbamazepine, gabapentin, lamotrigine, and topiramate) to prevent suicide is unproven, however. Hypothetically, because they stabilize moods and have an impact on agitated and aggressive states, they should have an impact on suicide rates as well. In the one direct comparison of lithium and an anticonvulsant (carbamazepine, or Tegretol), however, this did not prove to be true. German doctors studied 378 hospitalized patients who suffered acute major depression (one-half suffered from bipolar manic-depression) and, at the time they were discharged from the hospital, randomly assigned them to lithium, carbamazepine, or amitriptyline (an antidepressant) treatment. Over the subsequent two and one-half years, five patients killed themselves and four made very serious suicide attempts. All the patients who killed themselves, or tried to, were in the carbamazepine or antidepressant treatment group. Despite the fact that there had been more prior, pretreatment suicide attempts in the group that received lithium, none of the suicides or suicide attempts that occurred during the treatment period were made by patients taking lithium. The authors of the study concluded that their research demonstrated that lithium had “antisuicidal effects which may have been specific, which markedly exceeded its prophylactic efficacy, and which were superior to the effects of carbamazepine and antidepressants on suicidal behavior.”

  The same researchers also believe that lithium appears to protect against suicide even in those patients who do not show a good mood-stabilizing response to the drug. In a recent study, patients who had attempted suicide at least once were classified as excellent, questionable, or poor responders to lithium when it was given to prevent severe depressive episodes. There was in each group, despite the variability in recurrence rates, a significant decrease in the number of suicide attempts. Bruno Müller-Oerlinghausen of the Free University of Berlin summarized his group’s findings: “Discontinuing lithium or switching to other medications in apparently nonresponding patients may be considered a rational step toward optimizing medication, but it may result in the death of the patient.”

  It may well be that future research will show an antisuicidal effect of the anticonvulsant medications. Certainly they provide a real and important alternative to lithium for many patients. But in light of the many studies demonstrating lithium’s ability to prevent suicide in high-risk patients and the utter dearth of studies documenting this for the anticonvulsants, caution is in order. The clinical problem is complex, however. Not everyone who has depression or manic-depression is suicidal. If a patient refuses to take lithium or does not respond to it, anticonvulsants provide an important and often more agreeable treatment alternative. Lithium is effective in preventing suicide only if patients are willing to take it and if they respond to it. Not everyone will take it. Not everyone will respond to it.

  Ultimately, the best course of treatment for many patients may be a combination of lithium, used as a hedge against suicide, with another mood stabilizer or with an antipsychotic, antidepressant, or antianxiety medication. Because the cost of lithium is far less than that of valproate, the economic factor is a further issue, although the additional expense for the newer antidepressant, anticonvulsant, and antipsychotic medications is often cost-effective and clinically warranted due to increased compliance and greater safety and efficacy.

  Antidepressant use at time of suicide

  Antidepressant medications are not as obviously effective in lowering suicide rates as lithium is, although there are many practical problems in showing a clear-cut impact of antidepressant medications on suicide rates. (One reason for this is simple: suicidal patients are almost always excluded from clinical drug trials.) But there is persuasive evidence that the newer antidepressants, the selective serotonin reuptake inhibitors (or SSRIs), not only alleviate and prevent depression but also decrease angry, aggressive, and impulsive behaviors. These effects on such dangerous risk factors for suicide are immensely important. Some epidemiological and clinical studies have shown a decrease in the number of suicides and serious suicide attempts in patients taking antidepressants, but the extent of the decrease remains equivocal.

  What is unequivocal, however, is that in every investigation of individuals who have committed suicide, researchers have demonstrated that depression has been underdiagnosed and antidepressants have been underprescribed. Even when antidepressants have been prescribed, they have been given at inadequate dosages or for too short a period of time for them to take effect. This gross undertreatment of depression is shown in the graph on the opposite page, a summary of seven American and European toxicology and autopsy investigations that calculated the percentage of depressed individuals actually taking antidepressants at the time they committed suicide. The majority of patients had not been taking antidepressants at all, and far fewer still had been taking a therapeutic amount. The undertreatment of depression is consistent with research showing that doctors in general woefully underprescribe antidepressants and lithium for patients who could benefit from them.

  Several additional interpretations are possible. In the earlier studies, more individuals with depressive illnesses were taking antidepressants at the time they committed suicide than those in recent years. This may reflect the fact that the earlier antidepressants worked less well, or it may reflect the fact that, because the tricyclic antidepressants were far more toxic than the modern SSRIs, very suicidal patients had access to a more deadly means of suicide.

  There have been other difficulties in demonstrating a clear effect of antidepressant medications on patterns of suicidal behavior. There is, we know, a high rate of suicide in patients with bipolar II disorder, a variant of manic-depressive illness characterized by extended periods of depression and shorter, mild episodes of mania. These patients are frequently misdiagnosed as sufferin
g from depression only; this is in part because the mild manias are not experienced by the patients as pathological and in part because doctors are inadequately trained to make the differential diagnosis. Many clinicians are unaware that mood volatility and irritability are often signs of bipolar illness, and they do not inquire specifically enough about other sleep, mood, and behavioral symptoms that are important in making the diagnosis. Doctors also make the diagnostic error because depression is far more likely than mild mania to motivate a person to seek clinical care. This underdiagnosis of bipolar manic-depressive illness is widespread—perhaps one-third of patients are inaccurately diagnosed as depressed, rather than bipolar—and it can result in treatment that makes the illness worse over time. Antidepressants, if prescribed alone, rather than with mood stabilizers such as lithium or the anticonvulsants, can precipitate mania and, occasionally, highly agitated and potentially suicidal mixed states.

  Accurate diagnosis and the appropriate treatment of psychiatrically ill children and adolescents are material problems. A 1999 survey of pediatricians and family doctors found that only 8 percent of those prescribing antidepressants for children felt they had received adequate training in treating childhood depression. Many children with early-onset manic-depression, or bipolar disorder, are mistakenly diagnosed as suffering from attention deficit disorder with hyperactivity, either because doctors do not recognize the symptoms of manic-depression in children or because they are unduly sensitive to subtle pressures from parents and teachers who feel there is less stigma attached to attention deficit disorder than there is to a major psychiatric illness. Although there are overlapping symptoms—hyperactivity, distractibility, and irritability, for instance—and the differential diagnosis can be difficult, there are many distinguishing features: bipolar children are more likely to have a family history of bipolar illness or depression, to have mood lability, euphoria, grandiosity, hypersexuality, less need for sleep, and racing thoughts, and to be suicidal. Their pre-illness social and academic histories tend to be good and their illness is often a sharp departure from their normal level of functioning. The correct diagnosis is important because the primary treatment for attention deficit disorder is stimulant medication, which may aggravate the condition of a child with bipolar disorder (a disorder that generally requires a mood stabilizer such as lithium or an anticonvulsant). The long-term effect of the combined use of antidepressants and stimulants in a child or adolescent with bipolar illness is unclear but problematic.

  The role of antidepressants in actually precipitating suicidal behavior is controversial and unresolved. From a broad clinical and public health perspective, the evidence is strong that antidepressants do not increase the number of suicide attempts or actual suicide. On the contrary. But there is almost certainly a vulnerable subgroup of individuals who become agitated, restless, and virulently sleepless as a result of taking antidepressants. Although this is an uncommon reaction, it is potentially dangerous, and all patients need to be advised in advance of taking antidepressants that these side effects can occur, and if they do, they should be reported to their doctors. (The manufacturers of antidepressants generally include a warning to this effect in their product information in the Physicians’ Desk Reference: “The possibility of a suicide attempt is inherent in depression and may persist until significant remission occurs. Close supervision of high risk patients should accompany initial drug therapy. Prescriptions … should be written for the smallest quantity of capsules consistent with good patient management, in order to reduce the risk of overdose.” The manufacturers also include in the list of adverse effects anxiety, nervousness, insomnia, agitation, akathisia, and central nervous system stimulation.)

  The newer SSRIs, although no more effective against depression than the older tricyclic antidepressants, tend to be better tolerated by patients because their side effects (occasional insomnia, agitation, nausea, or sexual problems) are less difficult to live with than those caused by the tricyclics (dry mouth, blood pressure changes, constipation, or dizziness). The SSRIs have also been used to good effect in suicidal patients who are dependent on alcohol. Their major clinical advantage, however, is that they are much less toxic and therefore less likely to cause death in deliberate overdose. Lethal drug overdoses in British women, for example, have declined by a third over the past two decades, despite an increase in the rate of nonfatal overdoses during this period. Most of this decline is attributable to the wider use of the newer, less toxic antidepressants.

  There are other treatments for depression. Some existing drugs have effects on both norepinephrine and serotonin reuptake, while others exert their primary influence on serotonergic neurotransmission. Many new antidepressant medications are under development. (The Pharmaceutical Research Manufacturers of America reported that at the end of 1998 there were eighty-five psychiatric drugs in the research pipe-line: twenty-three for Alzheimer’s disease, nineteen for substance abuse, eighteen for depression, fifteen for schizophrenia, and ten for other disorders.) Some of these drugs, like the existing SSRIs, are designed to affect the serotonergic pathways in the brain; others have been developed to focus not only on serotonin but a variety of other neurotransmitters; yet others target different neurochemical systems, including norepinephrine. One drug in the pipeline, for instance, decreases the level of “Substance P,” a chemical that is highly concentrated in the amygdala and the hypothalamus, areas of the brain that are intricately tied up with the regulation of mood and emotion.

  Additional drugs that may ultimately have an effect on suicidal behaviors are under development. Some focus on the neurotransmitter glutamate and will, it is hoped, decrease alcohol and drug cravings in individuals who are dependent on those substances. Yet other drugs, known as CRH receptor antagonists, have been developed to reduce the stress response, which, if it is severe, and if it occurs in vulnerable persons, can trigger suicide.

  Omega-3 fatty acids, implicated by some (but by no means all) researchers in both depression and suicide, have been tested in recent clinical studies at Harvard. On discharge from a psychiatric hospital, patients with bipolar illness were, in addition to their regular dosages of valproate or lithium, given either omega-3 fatty acids or a placebo. After four months, 64 percent of those taking fatty acids were in remission but only 19 percent of those on placebo remained well. The results were sufficiently significant that the researchers were obliged to “break the blind” of the experimental condition, in order that those who were on the placebo could be treated with the omega-3 fatty acids. To date, although the research is very preliminary, there have been no serious adverse effects from the fatty acids given to the patients in the study. A seventeen-year epidemiological study of fish consumption in 265,000 Japanese adults—which found a 19 percent reduction in suicides in those who consumed large amounts of fish rich in omega-3 fatty acids—adds further suggestive evidence to the fatty-acid hypothesis of depression. Still, the theory remains unproven until the research findings are replicated.

  St. John’s wort, a mild to moderate antidepressant taken from the yellow-flowered plant Hypericum, is currently under study in a large clinical investigation coordinated by Duke University Medical Center. It is widely used as an antidepressant in Europe and more recently in the United States, but its usefulness in preventing suicide is unknown. Because it is not considered a drug, its purity and potency are unregulated by the Food and Drug Administration. Without doubt, it is helpful to some individuals who are depressed, but because it is usually taken without clinical supervision, there are several potential difficulties. Many people assume that St. John’s wort and other herbal treatments are safe simply because they are “natural” substances (as indeed, one should add, are lithium and arsenic), but there have been reports, although rare, of rapid mood swings, mania, and suicidal thinking induced by the herb. (These adverse reactions have also been rarely observed in another unregulated treatment for depression, light therapy.) More problematically, buying over-the-counter treatmen
ts for potentially lethal medical conditions, such as severe depression, may give the illusion of treatment and thereby prevent people from seeking out more effective drugs if the depression or suicidal thoughts do not abate.

  The antipsychotic medications used to treat schizophrenia and, occasionally, manic-depressive illness share some of the same problems and promise in preventing suicide that the antidepressants do. Used ineptly or without adequate clinical supervision, they can cause akathisia, an extremely uncomfortable state of agitation, muscle discomfort, and a difficulty in sitting still (often described by patients as feeling as if they are “jumping out of their skin”). But used carefully, in moderate doses, the antipsychotic medications—especially the more recently developed ones that have fewer side effects, such as clozapine, risperidone, and olanzapine (Clozaril, Risperdal, and Zyprexa, respectively)—may reduce the rate of suicide in psychotically ill patients.

  Electroconvulsive therapy (ECT), sometimes known as “shock therapy,” has been used for decades to treat severely suicidal patients. Although there is little evidence of a long-term effect on suicide, profound and often rapidly induced short-term improvements in suicidal mood are common in deeply depressed patients. Keeping patients alive through an acute suicidal crisis is the most important clinical priority; ECT not only can save lives, but it buys time to work out the best long-term treatment.

  Although ECT is the most effective and rapid treatment for severe depression, it remains controversial and, particularly in the United States, underutilized. This is due in part to the highly negative media coverage of ECT (some of which was clearly justified several decades ago, when abuses of shock treatments were rampant) and in larger part to the availability and ease of use of alternative treatments for depression. For many doctors, ECT remains the treatment of last resort, even with highly suicidal patients. This has led Jonathan Himmelhoch, a psychiatrist at the University of Pittsburgh, to write, “The narcissistic ponderings of psychiatrists whose political agenda supersedes their clinical experience must not be allowed to keep patients who are suffering the most severe form of pain from relief.” This is some truth in this remark, although the controversy over ECT is unlikely to disappear. A new, noninvasive treatment technology, transcranial magnetic stimulation, which involves placing a small, powerful electromagnetic coil on the scalp and sending repetitive pulses of high-intensity current to the brain, is now being tested in depressed patients. Transcranial magnetic stimulation has antidepressant capabilities, but clinical efficacy and safety studies are still in the very early stages. Unlike ECT, it does not require anesthesia and does not involve inducing a seizure; so far at least, memory impairment has not been reported as a significant adverse effect. It may or may not have an effect on suicidal thinking and behavior.