The management of treatment-resistant depression (TRD) remains a vexing clinical problem for a large population of patients and their clinicians. An estimated 32 to 35 million adults in the United States experience an episode of major depression during their lifetime.1 When depressed patients present for treatment, the results are often less than satisfactory. Even under the relatively ideal treatment conditions of the recent NIMH-funded Sequenced Treatment Alternatives to Relieve Depression (STAR*D) study, only 32.9% of patients achieved remission in level 1 with citalopram(Drug information on citalopram).2 An analysis of 36 open and double-blind antidepressant trials found a 36% rate of partial response or nonresponse.3 About 15% of patients had depression that failed to respond to multiple treatment trials.4 Depressed patients report impaired function and they overuse medical services.5,6
Patients with TRD may represent a biologically unique subtype of depressed persons.7 Unfortunately, the longer a patient remains depressed, the lower his or her chance of recovery—a fact that lends a sense of urgency to finding appropriate therapy.8
This article focuses on recent innovations in diagnostic issues, tactics, and strategy, and takes a brief look at the future.
Defining treatment resistance
The lack of consensus about how to define treatment resistance and how it should be classified are major methodological problems.4,9,10 The term “treatment resistance” obviously implies an inadequate response to antidepressant therapy. Sackheim9 notes that 4 of the following conditions must be met before the adequacy of an antidepressant trial can be judged:
- Drug dosage was titrated to the maximum when appropriate.
- The drug was administrated for an adequate duration and at maximal dosage as appropriate.
- The adequacy of patient adherence to therapy was monitored.
- The degree of nonresponse (partial vs complete) was recorded.
Without such detail, one cannot be certain that patients are not pseudo- resistant (ie, inappropriately classified as having TRD, when the real issue is suboptimal treatment). Up to half of treatment nonresponse may be a result of poor adherence to medication regimens and/or poor tolerability.11
Diagnostic and assessment issues
What constitutes an adequate clinical trial? In general, 6- to 8-week trials are recommended, with 2 to 4 weeks of medication at the upper end of the standard dosage.12 However, one of the most important findings of STAR*D was that many patients required longer than 8 weeks to respond, and thus an adequate trial may be 12 weeks or longer.2 Minimum target dosages for an adequate duration have been recommended.9 Some data suggest that without some clinical improvement after 2 to 4 weeks, the odds of a response are greatly diminished.13
As noted by Fava,10 a patient is not considered to have TRD until he or she experiences a poor response to at least one adequate clinical trial. Berlim and Turecki4 concluded that there is a “general sense” that the definition of treatment resistance requires the failure of 2 adequate trials of antidepressants from different classes. These researchers note, however, that there is little validation that patients who experience 2 failed trials with an antidepressant from the same class are less resistant that those who fail to respond to different classes.
Beyond the general issue of treatment resistance, attempts have been made to classify the degree of treatment resistance. Factors such as the number of antidepressant drug classes used, the trial duration, and the response to electroconvulsive therapy have been included. Berlim and Turecki4 have recently published a review of 3 of the most widely used staging methods.
From my perspective, the most useful definition of TRD for a clinician is simple: it is the failure to achieve a response to a medication to any degree short of remission. Any response short of remission is associated with impairment and an increased risk of relapse and should therefore trigger some kind of treatment intervention. It is important for clinicians to be aware of the research issues regarding the classification of resistance so that they may properly evaluate the results of clinical trial data, particularly when comparing one study with another. More specifically, the greater the degree of treatment resistance in a patient population, the more difficult it becomes to demonstrate a response to any agent.
