Discussing prevention of relapse or recurrence of major depression is beyond the scope of this paper but is clearly important. Long-term use of antidepressants is sometimes necessary particularly if patients have two or more episodes. Compliance becomes absolutely necessary, but side effects often cause patients to stop taking medications. Tricyclics and, to a lesser degree, traditional monoamine oxidase inhibitors (MAOIs) have many side effects because they act on the muscarinic, α₁-adrenergic, and histamine H₁ receptors. If an antidepressant is effective, the side effects can be managed.

The new antidepressants have become the agents of choice because they have fewer side effects. Sexual dysfunction is the most common reason for noncompliance and unfortunately is a relatively frequent problem with all antidepressants. Moclobemide and nefazodone seem to have the fewest sexual side effects.

For long-term therapy, doses that were effective for the acute episode should be continued. Clinical experience suggests a need for lifelong maintenance on antidepressants for patients older than 50 years at the time of first episode, for patients 40 years or older who have had two episodes, or for all patients with three or more episodes. If an antidepressant is discontinued, it should be withdrawn very gradually and signs of recurring depression monitored. Long-term use of lithium for prophylaxis of both bipolar disorder and major depression has also been shown to be very effective. There is a very high rate of attempted suicide among patients with major depression following lithium discontinuation.

Conclusion

Treatment-resistant depression is a relative term and depends on how far a physician is willing to go in treating a particular patient. Appropriate use of antidepressants can relieve symptoms in at least two thirds of cases. Drug combinations, electroconvulsive therapy (ECT), and augmentation strategies can, if vigorously applied, reduce the proportion of patients truly treatment resistant to about 7%. Considering the morbidity and mortality associated with depression, a vigorous approach to therapy is worthwhile. Long-term maintenance, once success is achieved, is essential.

Many less commonly used antidepressant therapies are supported by anecdotal evidence only. They include light therapy (non-seasonal affective disorder), high-dose TCA or MAOI therapy (only if serum levels can be monitored), intravenous clomipramine or maprotiline (allows for rapid perfusion, avoids first pass liver metabolism), bromocriptine, high-dose selegiline, and psychosurgery. Modern stereotaxic psychosurgical procedures offer symptom relief with minimal risk, and reports of large trials indicate that up to 60% of truly treatment-resistant patients either recover or are considerably improved.

Absolute treatment resistance

Very few patients show absolute treatment resistance. In specialized clinics, only about 7% of patients remain depressed after 1 year of extensive investigations and treatment. Extensive treatment involves many drug trials singly and in combination as well as one or more courses of electroconvulsive therapy (ECT). Patients with absolute treatment-resistant depression (TRD) are older (mean age about 55), have been depressed longer, and usually have insoluble life problems.

Even patients with absolute TRD can be helped. Antidepressants often give some relief, and carefully prescribed psychostimulants can improve mood and psychoenergize.

Supportive psychotherapy and CBT can also be of benefit. Supportive psychotherapy helps depressed patients to carry on and CBT allows patients to view the world more positively. Regular exercise and reduction of alcohol consumption also help.

Electroconvulsive therapy

An important and effective treatment for depression, electroconvulsive therapy (ECT) is effective in about 90% of cases of major depression. The success rate drops when it is used for drug-refractory cases. Using ECT depends on patient-related factors: it clearly is the treatment of choice for depression with psychomotor slowing, stupor, psychotic symptoms, or depression requiring rapid response because of suicidal risk or malnutrition. Use of ECT should be considered whenever therapeutic management is being reviewed and revised.

Triiodothyronine (T₃) augmentation

Thyroid hormone potentiation of tricyclic antidepressants (TCAs) can be useful. The thyrotropin-releasing hormone stimulation test should be done first, if available and convenient, to rule out grade three (subclinical) hypo thy roidism. If hypo thy roidism is present, it should be treated with thyroid replacement therapy. If the patient is euthyroid, low doses of T₃ (25 to 50 µg/d) can be given with the antidepressant for 10 to 14 days. Up to a third of depressed patients, particularly women, respond. Most experience with this approach has involved tricyclic antidepressants (TCAs), and the effect on the selective serotonin reuptake inhibitors (SSRIs) or other new agents is not well studied.

Tricyclic-fluoxetine combination

Some reports indicate that fluoxetine, when added to a TCA such as desipramine, can produce a robust and rapid response in many patients. This augmentation effect could be a result of increased TCA levels due to fluoxetine inhibition of the cytochrome P 450 system. Because fluoxetine can raise blood levels of tricyclic antidepressants (TCAs), routine serum levels of TCAs are recommended. Other TCA-SSRI combinations can also be tried. All selective serotonin reuptake inhibitors (SSRIs), however, effect the cytochrome P 450 System by raising TCA levels.

Tricyclic-MAOI combination

Some evidence suggests that this combination is more effective than either drug used alone for some patients. Ideally both drugs should be started simultaneously or the MAOI added to a TCA regimen. The safest combination seems to be phenelzine with either amitriptyline or doxepin. Clomipramine, imipramine, and the new agents, such as fluoxetine, should be avoided. Give low doses initially and pay rigid attention to dietary restrictions.

The SSRI combinations

Although no literature supports the practice, clinical experience suggests that lower doses of two selective serotonin reuptake inhibitors (SSRIs) together might work better than either alone.

Augmentation with L-tryptophan

L-tryptophan is the dietary precursor of brain serotonin. Reports confirm that L-tryptophan can enhance the antidepressant effect of monoamine oxidase inhibitors (MAOIs) as well as tricyclic antidepressants (TCAs) and lithium. This might apply to the new antidepressant agents as well. The high doses required (more than 3 to 4 g/d) make this approach cumbersome because the tablets are quite large.

Psychostimulants

Dextroamphetamine, methylphenidate, and to a lesser degree magnesium pemoline all have mood-elevating, psychoenergizing properties and have a place in the management of mood disorders. These drugs can be used alone or combined with antidepressants. Apathetic, elderly, and medically ill depressed patients often respond to psychostimulants when they cannot tolerate the side effects of antidepressants or a rapid response is necessary.

Psychostimulants are also useful for patients who do not respond to any antidepressant and are truly treatment resistant. Non-response to one psychostimulant does not predict non-response to another. There is no evidence of addiction or dose escalation although clearly this class of drugs needs to be prescribed cautiously and monitored carefully. Using psychostimulants is somewhat similar to prescribing analgesics for chronic pain conditions and can be justified considering the morbidity associated with major depression.

Cognitive behavioural therapy (CBT)

This therapy can be very useful for treating chronic depression, such as dysthymia, and can be helpful as an augmentation strategy in conjunction with pharmacotherapy for treatment-resistant patients. Several studies have demonstrated the effectiveness of CBT, a technique that family physicians can easily learn.

If lithium augmentation is ineffective, it should be discontinued and the antidepressant changed (Table 5). A lack of response to one of the newer agents in a class, such as the selective serotonin reuptake inhibitors (SSRIs), does not predict lack of response to others in that class. If the first drug used was a reversible inhibitors of monoamine oxidase A (RIMA) with only one drug in the class, the switch should be to an SSRI. If the original drug was an SSRI, a second SSRI, a RIMA, or other newer agents should be tried. If a TCA or MAOI was used first, the switch should also be to one of the new classes of antidepressants.

The tricyclic antidepressants (TCAs) and monoamine oxidase inhibitors (MAOIs) should be kept as third-line antidepressants. Evidence shows that response to an SSRI does not determine response to a TCA. One study has shown that 60.5% of patients who failed to respond to SSRI monotherapy responded when switched to a noradrenergic TCA. Most tricyclic antidepressants (TCAs) have both serotonergic and noradrenergic properties. Desipramine and nortriptyline are somewhat more noradrenergic than the others. Venlafaxine, a new selective serotonin and noradrenaline reuptake inhibitor (SSNRI) might also be considered.

If atypical symptoms, such as panic attacks or anxiety, are present, an MAOI or reversible inhibitors of monoamine oxidase A (RIMA) should be considered after a suitable washout period for the previously used antidepressant (10 to 14 days for most antidepressants; up to 5 weeks for fluoxetine because of an active metabolite).

If the second antidepressant is not effective, lithium augmentation should be tried once more before the antidepressant is changed again.

Strategies before changing antidepressants

Higher dose

For the new antidepressant compounds, most manufacturers recommend standard doses. If side effects are minimal, doses usually can be increased and tolerated well. Increasing the dose should be tried before adding other antidepressants or switching drugs.

Lithium augmentation

For patients who have failed to respond to an adequate course of an antidepressant, lithium augmentation is the most reasonable next step (Table 4 summarizes augmentation strategies). About 30% to 50% of patients respond to this technique, and it seems to work with all of the antidepressants (although less is known about how it works with the newer agents). Lithium, an antidepressant and mood stabilizer when used alone, appears to augment other drugs by enhancing postsynaptic receptor sensitivity.

To use this technique, the antidepressant should be continued at the current dose and lithium started at a dose of 300 mg three times a day. A positive response can occur in 5 to 12 days. If the response is positive, patients should continue to receive lithium for at least 6 months and in some cases for the duration of antidepressant therapy. Lithium levels should be monitored and kept within the therapeutic range of 0.6 to 1.0 mmol/L, and thyroid status should be closely followed.

Many physical diseases present with depressive symptoms or complicate depressive illness (Table 2). Some of these illnesses remain undetected for years and account for an apparent treatment-resistant depression (TRD). Normal thyroid function is particularly important; even mild or subclinical hypothyroidism can impair response to antidepressants. Hypothyroidism can be induced by lithium therapy, and many drugs, both medical and nonmedical, can cause depression or complicate treatment (Table 3).

——–

Are there underlying psychosocial factors?

Psychological and social factors, such as unresolved neurotic conflicts, a history of sexual abuse, current marital or work conflicts, unemployment, and poverty, can exacerbate mood disorders. Psychotherapy must be part of the treatment plan.

Is the current course of treatment adequate?

The most frequent reason for patients not responding to treatment is inadequate use of antidepressants. Surveys have shown that two thirds of correctly diagnosed patients do not receive adequate treatment, even under specialist care. Inadequate dose is the most common reason for treatment failure. One third of patients do not respond to the first course of antidepressants; this proportion decreases with use of consecutive antidepressant trials.

The dose and duration of an antidepressant drug trial must be adequate. For most antidepressants, this means the maximum recommended daily dose for at least 3 to 5 weeks. Unpleasant side effects sometimes lead to noncompliance, and serum levels are sometimes subtherapeutic despite what seems to be an adequate dose. Other factors, such as alcohol or drug use or abuse, might affect serum levels.

If the antidepressant is a tricyclic (TCA), determining serum levels could be important for management; tests are readily available at most centres. Nortriptyline has a therapeutic range of 50 to 140 ng/mL (178 to 499 nmol/L). Other tricyclic antidepressants (TCAs) have a less precise range (about 150 to 200 ng/mL or 535 to 1070 nmol/L). Establishing serum levels can help to ensure compliance; ensure adequate absorption; and assist in dose adjustment for special groups such as the elderly or medically ill, when severe or unusual side effects are present or during apparent treatment failure. Serum levels unfortunately have not been established for other classes of antidepressants.

The most common causes of noncompliance are disabling or unpleasant side effects, particularly those that cause sexual dysfunction. Most side effects are transient; if persistent, they usually can be easily managed.

Many patients are reluctant to take antidepressants. They sometimes feel they are relying on a pharmacologic “crutch,” or that they are unworthy of receiving help. They fear addiction, worry about side effects, or could have delusional beliefs about being poisoned.

Inadequate response to therapy can be iatrogenic. Lack of understanding of the biologic basis of depression, failure to educate patients and families, failure to conduct adequate treatment trials, or reluctance to prescribe the newer antidepressants, monoamine oxidase inhibitors (MAOIs), or electroconvulsive therapy (ECT) are factors.

Factors involved in treatment failure

Several questions should be considered when assessing patients with mood disorders who have not responded to treatment (Table 1).

Is the diagnosis correct?

Several psychiatric illnesses present with depressive symptoms or with comorbid or secondary depression and are mistakenly diagnosed as a primary depression. These include obsessive compulsive disorder; panic disorder or generalized anxiety disorder with demoralization; posttraumatic stress disorder, particularly in adults who have been sexually abused as children; schizophrenic defect state with apathy; uncomplicated grief; and dementia, such as early Alzheimer’s disease with cognitive decline. Although anti-depressants might benefit some of these patients, the treatment approach to each of these disorders is quite different.

Can the depression be subtyped?

For several subtypes of depression, a standard approach to treatment likely will be ineffective, and patients will appear treatment resistant. More information on these subtypes is available elsewhere.

Bipolar affective disorder, depressed phase

The depressed phase of bipolar disorder clinically resembles an episode of major depression. Introducing lithium or other anticycling drugs and avoiding drugs that might induce cycling becomes important. Rapid onset of symptoms, hypersomnia, a postpartum trigger, a history of manic or hypomanic episodes, or a positive family history are clues to bipolarity.

Delusional (psychotic) depression

A major depressive episode with psychotic features, such as mood-congruent delusions or auditory hallucinations, usually responds poorly to antidepressant monotherapy and requires an antipsychotic agent as well. Electroconvulsive therapy, however, is the treatment of choice.

Major depression with obsessional features

Major depression presenting clinically with obsessions and compulsions responds better to antidepressants that are strong inhibitors of the serotonin transport system, such as the tricyclic clomipramine or any of the selective serotonin reuptake inhibitors (SSRIs).

Atypical depression

Atypical depression is characterized by symptoms such as panic attacks, severe generalized anxiety, agoraphobia, or social phobia. Vegetative features are also atypical and include initial insomnia, increased appetite, hypersensitivity to rejection, and attention-seeking behaviours. Monoamine oxidase inhibitors (MAOIs) or the reversible inhibitors of monoamine oxidase A (RIMA) are probably the agents of choice.

Postpartum depression

About 80% of depressions that begin for the first time postpartum are bipolar. Postpartum depression often resists treatment. Psychosocial issues, such as bonding with the infant and the marital relationship, are very important and need to be addressed as part of the treatment plan. If the mother is hospitalized, physicians should consider admitting the infant once the mother has improved enough to show interest in and look after the baby, even if assistance from nursing staff is needed.

Double depression

Double depression, as the name suggests, is a major depression superimposed on chronic dysthymia. Chronic dysthymia usually requires psychotherapy and lifestyle changes, although pharmacotherapy can help. A major depression that responds to antidepressants might seem unresolved if the dysthymia continues untreated.

Seasonal affective disorder

In this disorder, probably a variant of bipolar disorder, patients suffer from a winter-onset depression for at least 2 consecutive years and have either a normal mood or mild hypomania during the summer. Seasonal affective disorder (SAD) is further characterized by carbohydrate craving, decreased energy, and increased need for sleep during the winter depressive phases. Some patients with SAD respond to traditional antidepressants or lithium, but the treatment that seems most effective is light therapy. Patients with variants of seasonal affective disorder (SAD), such as summer depression (and winter highs), sometimes respond to temperature manipulation rather than light.

In a study of collaborative care for depression treatment at the University of Washington Medical School, only 63.5% of the control group – which received “usual” care – rated their quality of care as good to excellent.

Kenneth B. Wells, M.D., of RAND, a Santa Monica, Calif.-based think tank, and the University of California, Los Angeles, Neuropsychiatric Institute, was more emphatic. In reporting the results of the Patient Outcomes Research Team (PORT) on depression in JAMA this past January, Wells et al. (2000) wrote, “Quality of care for depression in managed primary care settings is moderate to poor with resultant poor outcomes.”

Both the University of Washington and PORT studies, however, have shown that enhanced interventions can measurably improve both outcomes and patient satisfaction. These findings may offer new fuel to the managed care debate over costs versus efficacy.

Collaborative Care

In the University of Washington study, a team led by Wayne Katon, M.D., compared the effects of a stepped, collaborative care intervention in a primary care setting with a control group receiving usual care. In most cases, usual care consisted of a prescription for antidepressant medication, two or three visits during the first three months of treatment and the option of referral to a mental health program.

Intervention patients were provided with a book and videotape prepared by the study team. These tools reviewed the biology of depression, depression’s relationship to stress, how medications and psychotherapy help depression, and how patients and their significant others could be active partners with their physician in caring for their depressive illness. In addition, patients with severe psychosocial stressors were encouraged to seek psychotherapy or were referred to community-based support groups.

A psychiatrist worked with all of the intervention patients’ primary care physicians to optimize medication usage and to find alternatives if side effects developed. The psychiatrist also monitored the patients’ adherence to the medication regimen through the use of automated pharmacy records and alerted the primary care physician if it appeared the patient had discontinued using the drugs.

Patients in the intervention group were significantly more likely to adhere to the medication regimen than were patients in the control group. At the end of three months, 78.6% of the intervention patients and 62.1% of the control patients had adhered to the medication schedule; at six months, 73.2% of the intervention group versus 50.5% of the control group were still following the medication schedule. Pharmacy records indicated that 68.8% of the intervention patients versus 43.8% of the usual care group received antidepressant medication (at or above the lowest recommended dosage) for at least 90 days.

Response to treatment was based on a Structured Clinical Interview for DSM-IV finding of 0 or 1 of the nine major depressive symptoms at three and six months. At three months, 40% of the intervention patients versus 23% of the usual care patients were asymptomatic (p=0.01); at six months, 44% versus 31% were asymptomatic (p=0.05).

Katon et al. concluded, “The cost per case successfully treated was lower for collaborative care than for usual care because the success rate of treatment was increased more than the total costs of treatment per case.”

Katon also assisted the PORT team, lead author Wells told Mental Health Economics. “Wayne helped on the medication arm of our study. A lot of what we did was learn from what he’s learned about how to support primary care practices and how to package and disseminate the tools without exerting a lot control.”

The PORT Depression Study

The PORT study tested the use of evidence-based materials including training guides, slides, brochures and videos for clinicians, nurse specialists, psychotherapists and patients. Researchers compared the results of quality improvement (QI) programs with usual care at 46 primary care clinics in six managed care organizations.

The research team studied two variant QI programs: one with enhanced resources for supporting medication management (QI-meds) and the other with enhanced resources for providing psychotherapy for depression (QI-therapy). The common elements for each variant included: 1) institutional commitment on the part of the health plan; 2) training of local leaders, including a primary care clinician, a nursing supervisor and a mental health specialist, in each clinical setting; 3) training for local staff; and 4) patient identification.

In QI-therapy, local psychotherapists were trained to provide manualized individual and group therapy for 10 to 16 sessions. In QI-meds, nurse specialists were trained to provide follow-up assessments and support adherence through monthly contacts with the patients for six or 12 months.

At the end of six months, 50.9% of the QI patients and 39.7% of the controls had counseling or used antidepressant medication at an appropriate dosage; at 12 months, 59.2% of the QI patients versus 50.1% (p=0.006) of the controls had done so.

The QI patients showed a markedly greater rate of improvement than the controls. At six months, only 39.9% of the QI patients still met the criteria for probable depressive disorder, compared with 49.9% of the control group (p=0.001). QI patients were 8% to 10% less likely to have probable disorder at six and 12 months; in addition, QI patients showed a 5% increase in employment retention.

“To our knowledge,” the team wrote, “no QI study has demonstrated improved employment, although perceived interpersonal work functioning improves with efficacious treatment for major depression.”

Wells told Mental Health Economics the PORT study showed “when practices that are not academically based make a modest effort to do the right thing and organize their resources to support doctors’ and patients’ decisions, the patients will benefit over a long period of time”

Answer. I am not aware of any controlled, persuasive evidence that Thorazine (chlorpromazine) or any standard antipsychotic – in low doses or high – is particularly effective as an antidepressant. The exception to this claim may be in so-called delusional depression, in which psychotic features are present; antipsychotics such as Thorazine may be somewhat effective for the agitation and delusional thinking in this condition, but probably not for the core depressive features, such as lack of pleasure, lack of energy, and slowing of mental functions. On the other hand, depression is often encountered as a side effect of standard antipsychotics (neuroleptics).

Having said all this, I would hasten to add that new, atypical antipsychotics, such as risperidone and clozapine, do, indeed, seem to have antidepressant properties. However, the standard treatment of depression remains the antidepressants.

Answer. Music has been used to treat various anxiety and stress-related states, including postoperative pain. However, in my literature search, I was unable to find information on specific types of music used in the treatment of depression. However, you may find details on this in the article, “A musical assessment of psychiatric states in adults,” by Pavlicevic M & Trevarthen C, in Psychopathology, 1989;22(6):325-34. Also, Dr. Herbert Dandes, Professor of Counseling and Psychology, University of Miami, Fla., has presented a talk on the use of music to stimulate the therapeutic process and may be able to give you some detailed information on this topic.