Treatment-Resistant Unipolar Depression

Definition
Systematic integration of available data on clinical management is hampered by multiple definitions of treatment-resistant depression. It would appear that the first step to optimize treatment outcomes requires only to have patients exposed to adequate doses of antidepressants for adequate periods of time. But the specifics of the recommendation are more complex. What is an adequate dose, and how long is an adequate duration of an antidepressant trial? These issues are discussed in more detail later in this chapter.

Reasonable data correlate response with blood levels of imipramine, desipramine, amitriptyline, and nortriptyline but not the other tricyclic antidepressants (TCAs): clomipramine, trimipramine, protriptyline, doxepin. In the face of nonresponse, it is useful to check blood levels for those antidepressants with established dose-response relationships.

Data for the average patient with moderate depression, however, suggest an overall flat dose-response relationship for the specific selective serotonin reuptake inhibitors (SSRIs) (fluoxetine, sertraline, paroxetine, citalopram, and fluvoxamine), making it difficult to know from randomized clinical trials how and when (if ever) to increase these widely prescribed agents. In contrast, in one study it was found that the majority of patients who did not respond after 8 weeks of fluoxetine at a fixed dose of 20 mg responded when the dosage was increased to 40 mg. Bupropion appears to be equally effective at 150 mg or 300 mg daily. Dose-response relationships are less clear with nefazodone and mirtazapine, whereas a dose-response relationship has been found for venlafaxine. A reasonable range of minimal doses has been recommended in treatment guidelines (Depression Guideline Panel 1993). A comprehensive method to organize a complex treatment history has been proposed and validated. To ensure an adequate trial, patients should at least be exposed to the minimally effective tolerated antidepressant dose.

The definition of adequate duration is more complex than the definition of adequate dosing. Antidepressants have long been known to have a delayed onset of action for many but not all patients. Early onset and late onset but persistent patterns of response have been related to a “true” drug pattern, whereas those who respond to placebo tend to have a more fluctuating pattern during an acute-treatment trial of 6-8 weeks. Data from clinical trials suggest that an adequate trial for most patients is in the range of 4-8 weeks to demonstrate response, although the full benefit may not be realized until 12 weeks of acute treatment, especially for chronic depression.

As patients continue to exhibit an almost complete lack of response during the course of an antidepressant trial, the ultimate probability of response starts to fall. About two-thirds of patients who will have an ultimate response at 6 or 8 weeks will have evidence of response by the fourth week of treatment. One must realize, however, that if 4 weeks is used to declare response or nonresponse to an antidepressant, about one-third of ultimate responders will be misclassified as nonresponders. It might be difficult to persuade patients to continue with an antidepressant that has been ineffective for 8 weeks. It makes sense to require a minimum of 4 weeks of treatment at therapeutic doses to define an adequate trial. Guidelines for individual patients suggest a range of time from 4 to 12 weeks.

Treatment resistance—especially if one takes into consideration all patients who do not achieve full remission of mood disorders with one or more treatment trials—is an issue for the majority of patients. In this light, the understanding of factors underlying partial response and nonresponse is critical to clinical practice, as is the awareness of therapeutic strategies to improve outcome.

Despite the central importance of treatment resistance to clinical practice, the research base necessary to guide treatment selection falls short of adequacy. The reasons for this are several. The research investment in clinical trials by industry is designed to show the efficacy of new drugs in the untreated population. Federal research funding for therapeutics has historically been extremely limited. The methodology to address the therapeutics of nonresponse or incomplete response is cumbersome and costly; prospectively ascertained samples of nonresponders must be derived from very large patient subject populations to meet the requirements of statistical power to detect differences between treatments. In the end, clinicians must rely on recommendations derived from a mixture of clinical reports and series, open trials, and small controlled clinical studies.

Consequently, no definitive algorithms drive the selection of treatment, although groups are now working to define the extent of consensus among clinical research experts. For now, familiarity with the issues and options is the best underpinning for clinical practice.

Phases of Treatment
The goal of antidepressant treatment for patients with mood disorders is for these patients to achieve or return to and remain in a state of wellness (i.e., no symptoms) and optimal functioning. This goal is approached, at least initially, in 50%-65% of depressed patients, who improve with adequate antidepressant treatment, leaving 35%-50% who do not improve (as measured by at least 50% reduction in standard depression scale scores). However, even those patients who experience substantial improvement with antidepressant treatment may still fall short of complete resolution of depressive symptoms; furthermore, between 10% and 25% of responders relapse within the first year, even with continuation treatment. At best, therefore, only about 58.5% (90% of 65% who improve) and, at worst, 37.5% (75% of 50%) of depressed patients get better and stay better with antidepressant treatment. Even fewer achieve remission, and many have residual symptoms. Depressed patients, their families, and doctors are all too familiar with insufficient improvement, incomplete recovery, and relapse and recurrence.

How to address suboptimal treatment responses is determined in part by when in the course of treatment the inadequate response occurs (i.e., during the acute, continuation, discontinuation, or maintenance phase). The acute phase is defined as the period of diagnosis of a mood disorder episode and initiation of treatment for the disorder. Problems that may arise during the acute phase include misdiagnosis, treatment intolerance, inadequate dosage, partial response, and nonresponse.

The continuation phase begins when the patient experiences maximal response or remission of symptoms. Loss of the initial response with full or partial return of symptoms is a problem frequently encountered during this phase of treatment. Determination of partial or full relapse depends on whether or not a patient’s symptoms meet the full criteria for an episode. The continuation phase runs through a period of time judged to be longer than the likely (natural) duration of the episode in the absence of treatment. Whether the estimated duration of the continuation phase is determined on the basis of the patient’s prior episodes or on the basis of a more general estimate of the likely duration of depressive or manic episodes, the purpose of the continuation phase is to prevent relapse, which is usually accomplished by continuing treatment at the levels required to induce remission.

Some treatment-responsive patients benefit from dose reduction during the continuation phase if medication side effects substantially compromise the benefit. Many patients intermittently experience depressive symptoms during the continuation phase, a phenomenon known as “roughening.” Although these symptoms may not meet the full criteria for a major depressive episode, they may warrant an increase in the intensity of antidepressant treatment, such as increasing the drug dosage, augmenting the current drug with another agent(s), or switching to a different drug.

The end of the continuation phase and the beginning of either the discontinuation or maintenance phase occurs when the patient is deemed to have recovered from the acute episode. The discontinuation phase follows a decision to discontinue the treatment and involves monitoring for relapse while withdrawing medication. A decision to redirect the therapeutic focus away from treatment of the index episode toward maintaining recovery or preventing the recurrence of future acute episodes initiates the maintenance phase.

Regardless of whether clinical judgment calls for discontinuation or maintenance, if the patient has been in remission for 8 weeks or longer, the reappearance of symptoms is considered a recurrence (new episode) at this point rather than a relapse. Patients frequently experience intermittent subsyndromal symptoms (i.e., roughening) during the discontinuation or maintenance phase. The significance of roughening depends on whether it is the harbinger of an impending acute episode or is merely a brief period of mild symptoms. Studies of interepisode symptoms (Keller et al. 1992a) suggest that roughening with features of depression often resolves without intervention. Roughening with symptoms of mood elevation in patients with bipolar disorder, on the other hand, should be monitored more closely, because this type of symptom recurrence appears more likely to develop into a full affective episode.

Thus, patients who do not achieve and sustain euthymia with “adequate” antidepressant treatment can be considered treatment-resistant. However, the definitions of adequate antidepressant dosages and durations of treatment change over time. Due in part to the evolving nature of current psychiatric practice, which is continually refined by consensus or clinical research, there exists no universally accepted definition of treatment resistance in depression. Thase and Rush (1994), however, proposed definitions that may prove useful in allowing consistency in future studies and reports. In the absence of universally accepted standard criteria for refractoriness, both clinical and research practice are served by a clear and concise description of treatments tried and failed and of how response or nonresponse was measured. Despite patient reports of treatment history, a prospective adequate trial of an antidepressant is generally necessary to confirm resistance. When a trial of adequate duration and dosage fails, one should consider the possibility of insufficient doses or blood levels of the antidepressant medication.