Depression is among the most common and treatable illnesses encountered in medicine. Even though a growing number of strategies have become available to treat this illness effectively, the majority of patients with depression fail to receive adequate treatment, even if diagnosed correctly. Untreated depression, similar to inadequately treated depression, can result in significant morbidity and mortality, including unnecessarily high rates of suicide. Depression is expected to become the second leading cause of worldwide disability during the next 15 years.1
Studies suggest that, when depression is identified and subsequently treated, only a small percent of patients actually go on to achieve the expected full remission and recovery.2 Failure to achieve full remission during the index episode is among the most common reasons why many depressed patients fail to achieve the optimal response and many go on to present as “treatment resistant.”3 The potential to prevent a treatment-resistant depression may be possible by treating the index episode with whatever treatment modalities may be necessary to achieve a full remission. A similar analogy can be used to understand this concept, for another “treatable” illness like a bacterial infection: when using an antibiotic to treat an infection with potential significant morbidity and mortality, repeated attempts at treating the infection with multiple, inadequate trials (eg, dose/duration/combinations) of antibiotics can result in “treatment-resistant” infection that may have been preventable if adequate treatment were initiated at the onset.
When treating depression, it is not uncommon and may be necessary to use a combination of treatments (eg, pharmacologic, other somatic approaches, psychotherapeutic) to achieve and sustain the best outcomes, whether treating the first episode or when approached with a depression that has appeared to fail multiple treatment trials. At no other time in medicine have clinicians had the availability to use the many strategies to treat depression, including treatment-resistant depression, to the point of remission and ultimate recovery. While the management of treatment-resistant depression can be among the greatest challenges for a clinician, it also can be among the most rewarding and provides an opportunity to combine the art and science of clinical practice to achieve the expected outcomes.
Despite an increasing literature on treatment-resistant depression, this population remains poorly defined. This article addresses combination treatments for treatment-resistant depression; however, many of the combination strategies can be applied in many different types of cases, from a first-episode depression that shows only a partial response to an adequate trial of monotherapy to an ongoing depression that has failed to show any improvement to multiple combinations of treatment. For the purpose of the clinical strategies proposed in this article, the clinician is encouraged to individualize the definition of treatment-resistant depression to each patient and to tailor the treatments based on the multiple factors that apply to each patient, both current and historic.
Defining Treatment Resistance
While there is no consensus on what constitutes treatment-resistant depression, descriptive levels are emerging in the scientific literature, including the commonly used staging system proposed by Thase and Rush.4 In addition to the levels of treatment resistance, patients with TRD may exhibit at least four different clinical presentations:
- Partial response to treatment(s) but with persistent residual depressive symptoms;
- Good response or remission to treatment(s) but with relapse(s) during continuation treatment;
- Good response or remission to treatment(s) but with recurrence(s) during maintenance treatment; or
- Complete failure to show any meaningful response to treatment(s).
Before committing to a treatment course for treatment-resistant depression, attempts should be made to ascertain any “modifiable” factors that may account for the lack of achieving or sustaining the desired response. Many of these modifiable factors are common and may even prevent treatment resistance. These potentially modifiable factors include assuring the diagnosis and comorbid diagnosis (eg, bipolar disorder, anxiety disorders, eating disorders, substance use disorders, comorbid medical illness); assuring an adequate trial (including maximal tolerated dose and duration) for any treatment, whether it is an antidepressant, empirically based psychotherapy, electroconvulsive therapy, or nonpharmacological augmentation; problems with adherence or compliance; intolerance to treatment (and adequate attempts to mange side effects to a potentially effective treatment); psychosocial stressors; inaccurate assessment of response; and failure to achieve remission during the index episode.
The Importance of Remission in Combination Treatments
Failure to achieve and sustain remission of the index episode remains among the most important of the “modifiable” factors associated with treatment-resistant depression. Similar to any other medical illness, the standard of treatment should always be toward “remission.” Up to 70% of patients who show a full response (greater than 50% improvement from baseline) to treatment actually fail to remit fully.2 Residual depressive symptoms, including the traditional symptoms from the Diagnostic and Statistical Manual of Mental Disorders, fourth edition (DSM-IV),5 of major depression and common somatic symptoms (eg, headache, backache, musculo-skeletal pain) significantly increase the risk of both relapse and recurrence in a depression that responds to treatment, but fails to remit.2
Additional potential consequences for depression that responds without full remission include continued psychosocial impairment, increased use of medical services, potential worsening of morbidity or mortality of comorbid psychiatric and other medical illnesses, ongoing risk of suicide, and a “theoretical risk” of developing depression less likely to respond to conventional treatments.6 Understandably, treating all depression to remission is imperative and may require the use of combination treatments, either during the index episode or during the onset of symptoms after remission has been achieved.
Basic Principles of Combination Treatment
When a patient with depression is failing to respond optimally to a particular treatment, or starts losing remission, the decision needs to be made to switch treatments or add another treatment. While there is a growing literature to provide guidelines around combining treatments for resistant depression, many of the studies used to provide information for such guidelines have methodological limitations, including consistency in controlling for baseline levels of duration and severity, possible comorbid illness, and level of treatment resistance.7 Therefore, the decision to combine treatments should always be tailored to the individual patient.
Given the paucity of combination treatments approved by the Food and Drug Administration for the management of resistant depression, it is important for the clinician to consider the treatments that are best supported in the literature in regard to both efficacy and safety. After a patient has failed the more “accepted” treatment supported by the literature in the absence of FDA approval, the choice to use less conventional treatments may be justified.
Additionally, clinicians should always obtain appropriate informed consent. Patients and their families should know both the risks and benefits of treatment, as well as the risk of not treating the depression to optimal levels. In addition, the clinician should document carefully the use of treatments that are not indicated for depression and the use of treatments that, when used in combination, have been reported to have possible safety issues (eg, stimulants with monoamine oxidase inhibitors [MAOIs]). Second opinions or consultations with a clinician who may have more expertis, or a particular expertise, may provide additional support, especially with treatments that may be associated with higher levels of possible risks. The decision to combine treatments must always be tailored to the individual patient.
Before making a decision to switch or combine treatments, assuring an adequate dose and duration of a treatment is essential, although other factors such as potential significant side effects, safety issues, worsening of symptoms, psychosis, or an increasing risk of suicide would warrant a quicker intervention to make such changes. The literature uses a variety of terminology when describing switching, combining, or augmenting antidepressant modalities.
The reamainder of this article discusses the pharmacologic, psychotherapeutic, and other somatic treatments available for management of treatment-resistant depression, concentrating on combination and augmentation therapies. Pharmacologic treatments are divided into those with documented antidepressant effects (the FDA-approved antidepressants) and those that are not in themselves antidepressants but have been documented to “augment or potentiate” an antidepressant. Psychotherapy will refer to the empirically based therapies that have been documented to show efficacy for the treatment of depression (eg, cognitive therapy, cognitive-behavior analysis systems of psychotherapy, behavioral activation, interpersonal therapy). Other somatic treatments include nonpharmacologic treatments that have documented efficacy for the treatment of depression, including electroconvulsive therapy (ECT), vagus nerve stimulation (VNS), phototherapy, repetitive transcranial magnetic stimulation (rTMS), and psychosurgery.
When to Switch, Combine, or Augment?
“Switching” refers to the intention to replace one antidepressant modality for another. Occasionally, the switch from one treatment to another may include an overlap of treatments, referred to as “bridging.” “Combination” is defined as combining two or more antidepressant modalities with the intention of using these treatments together for the desired effect. “Augmentation” is defined as adding a treatment that does not have antidepressant effects alone but is added to a modality that demonstrates antidepressant effects.
The choice of a particular approach to switch, combine, or augment is dependent on multiple factors for any patient at any time in the course of the depression. There are some very basic guidelines, including assessing the degree of improvement noted from the time the current treatment was initiated and maximized to the time of making the decision. Switching the primary treatment modality may be indicated for a depression that shows less than 25% reduction of symptoms and may involve bridging two treatments together, providing a period of time to ascertain efficacy and tolerability of a potentially effective combination.
If switching involves stopping the antidepressant, it is generally recommended to either overlap treatments or slowly discontinue the first antidepressant. Abrupt discontinuation can result in antidepressant discontinuation symptoms, which can aggravate the underlying depression and make it difficult to ascertain any possible efficacy that may be achieved by bridging the two treatments. However, the safety of overlapping two antidepressants should be considered for possible additive adverse events, or for potentially dangerous interactions such as a serotonin syndrome or a hypertensive reaction (eg, MAOIs combined with many antidepressants without adequate washout).8 Switching strategies are covered in more detail in the article by Thase (see page 970).
Augmentation of the primary anti-depressant modality is justified for a depression that shows greater than 50% improvement, and specific augmentation strategies can be selected based on the residual or re-emergent symptoms. For depressions that show a 25% to 50% reduction of symptoms, any or all of the three strategies, including switching, combining, or augmenting, are warranted, depending on many possible variables for each patient.
In addition to efficacy to the primary treatment, the decision to switch, combine, or augment depends on other significant factors, including safety, tolerability, cost, compliance, past treatment history, family history, severity of the current episode, residual symptoms, and comorbid illness. The art of using a particular combination also may be determined by selecting pharmacologic treatments that either enhance a particular mechanism or work differently to work synergistically.
There are innumerable possible antidepressant medication combinations, and the choice of a particular treatment will depend on the factors listed above for each patient throughout the course of their illness. When it comes to safety, it is important to be aware of possible interactions between antidepressants and any other concomitant medications. Examples include the potential for a serotonin syndrome or hypertensive reaction without adequate washout between the use of MAOIs and most antidepressants or the simultaneous use of two MAOIs together.8 The literature includes studies of the concomitant use of some tricyclic antidepressants used with MAOIs, namely low doses of amitriptyline; on the other hand, the use of clomipramine in proximity to MAOIs is absolutely contraindicated.7,8
The concomitant use of an antidepressant that potentially can increase the serum level of bupropion (eg, fluoxetine, paroxetine) or have the potential to add to the lowering of seizure threshold (eg, maprotiline, clomipramine) may increase the risk of seizures.9 The combination of a tricyclic antidepressant (TCA) and selective serotonin reuptake inhibitors (SSRIs) or other agents that affect specific cytochrome p450 pathways can result in possible elevated TCA levels, resulting in possible toxicity (eg, seizures, cardiac conduction block).10 Remaining cognizant of the possible pharmocokinetic and pharmacodynamic interactions of medications used when combining or overlapping antidepressants is imperative.
In terms of efficacy, there are reports of using various antidepressant combinations safely to achieve therapeutic effects. While many of the studies have methodological limitations (eg, acute treatment) there is evidence that specific combinations can be effective for some patients, particularly when the first antidepressant fails to show an adequate response.7 Among the more well-studied combinations are SSRI plus low-dose TCA, SSRI plus bupropion, and mirtazapine plus SSRI or serotonin norepinepherine reuptake inhibitor (SNRI).7
While psychotherapy and ECT can be primary antidepressant modalities as monotherapy, they also should be included among the possible combination strategies with antidepressants for treatment-resistant depression.11,12 VNS, with its recent FDA approval for treatment-resistant depression, was studied primarily in treatment-resistant depression as a combination with antidepressants and should be considered among combination options.13
Examples of specific patient profiles that support using specific antidepressant combinations have been described. One example might be a patient who suffers from both depression and obsessive-compulsive disorder (OCD), a comorbid illnesses demonstrated to respond to SSRIs or clomipramine. If the OCD remits but the depression fails to respond to maximal doses of several SSRIs, combining clomipramine or another antidepressant (eg, bupropion) with a different pharmacologic profile than selective serotonin reuptake may be warranted.
Another example might be a patient with depression who shows a 50% improvement on maximal doses of bupropion but has residual anxiety (primarily nausea and psychic anxiety), diminished appetite, and initial insomnia. Addition of mirtazapine can provide further anti-depressant and anxiolytic effects including improvement of sleep and appetite, via the effects on serotonin type 2 receptors. Nausea also may improve via effects on the serotonin type 3 receptor, in addition to any synergistic effects with adding serotonin activity. After achieving remission, the decision to combine or bridge the two treatments can be determined, based on the individual needs of the patient.
Similar to combination strategies, there are innumerable possible combinations for using augmentation strategies for treatment-resistant depression. The Sidebar (see page 999) shows the various pharmacologic and nonpharmacologic augmentation strategies that can be considered to add to a primary anti-depressant modality.7 VNS and ECT are discussed in more detail in the article by Thase (see page 970) and will not be discussed further here.
Antidepressant Augmentation Strategies
Vagus nerve stimulationElectroconvulsive therapyLithiumThyroid hormone (T3)Atypical antipsychoticsStimulantsBuspironeModafinilCarbamazepineDivalproex sodiumLamotrigineDopamine agonists (eg, pramipexole)Estrogen replacementBuprenorpineSAMeInositolPhototherapyPsychotherapy (time-limited) Cognitive-behavior therapy Cognitive-behavior analysis system Interpersonal therapy
The decision to “augment” a primary antidepressant's modality can be justified for a patient who demonstrates a significant response and tolerability to the primary treatment(s) but fails to achieve full remission. This decision is justified easily for the patient who failed to demonstrate a significant response to multiple trials of antidepressant modalities but finally shows a partial response.
The choice of augmentation should be determined by the multiple potential factors for any individual patient. Similar to combining antidepressant modalities, it is important to consider the risks and benefits, including possible drug interactions, and to assure informed consent is performed. Using augmentation strategies most supported in the literature for efficacy and safety is important, but given the paucity of data, the decision remains at the discretion of the clinician and patient. The augmentation may be a pharmacologic treatment, psychotherapy, or a nonpharmacologic somatic treatment, tailored to the symptoms of the patient.
A few examples of augmentation strategies:
- Use of a benzodiazepine or atypical antipsychotic for residual anxiety symptoms;
- Use of a stimulant or modafinil for residual symptoms of fatigue or hypersomnia;
- Use of an antipsychotic for overt psychosis or significant suspiciousness or paranoia; and
- Use of cognitive-behavior therapy (CBT) for patients who continue to show a pattern of cognitive distortion that either prohibits their remission or correlates with repeated relapse or recurrence.
Comorbid psychiatric illness also may direct specific treatment choices. Examples include the use of atomoxetine or a stimulant for a patient with comorbid attention-deficit disorder; the use of an atypical antipsychotic for a patient with bipolar disorder who remains depressed; or the addition of triiodothyronine for a patient with hypothyroid who has residual depressive symptoms (eg, fatigue) but appears to have hypothyroidism controlled with T4 supplementation.
Family or personal medical history may support a particular augmentation strategy. For example, successful use of lithium or lamotrigine in a family member with bipolar disorder might indicate a higher chance of success. In addition, a personal history of multiple “break-through” recurrences or interepisode hyperthymic affective features could point to the addition of a mood stabilizer. Finally, considering the additive, synergistic or use of a different pharmacological mechanism may support the choice of a particular augmentation.
Similar to combination treatments, the decision of how long to use the augmentation needs to be tailored to the patient. At one end of the timeline is a patient with “transient” anxiety or insomnia, where the augmentation may need to be used for a short time. Other augmentation strategies may be justified to be lifelong; examples include maintaining lithium or lamotrigine for the responder with suspected occult bipolar disorders, or the patient whose depressive symptoms and comorbid psychiatric illness both remit with an augmentation.
If stopping a particular augmentation is warranted, watching for relapse and recurrence over time is imperative, and changing one medicine at a time is recommended to ascertain the effects of a specific change. For patients staying on augmentation strategies for extended periods of time, the prescribing clinician needs to be informed of any changes in the patient's medical status or the addition or removal of other medications (including over-the-counter medications) that may affect possible drug interactions or compromise the patient's medical status.
Additionally, some augmentation strategies may require specific monitoring of clinical parameters. Examples include lithium, which requires regular monitoring of serum levels, as well as baseline and follow-up thyroid and renal function testing; triiodothyronine, which calls for monitoring baseline and follow-up thyroid profiles, electrocardiogram, and bone density; atypical antipsychotics, which require monitoring of weight and other metabolic parameters; and stimulants, which call for baseline and routine monitoring of blood pressure. As with combination strategies, the clinician needs to review the risks and benefits over time with the patient periodically.
Interpersonal therapy (IPT) and cognitive-behavior therapy (CBT) also can be quite effective when used alone or in combination with pharmacotherapy in the treatment of patients with depression. IPT may improve interpersonal functioning, while CBT appears to have an enduring effect that reduces subsequent risk following treatment termination.14 In addition, ongoing treatment with either of these modalities appears to reduce risk of relapse.
Treatment with the combination of medication and either IPT or CBT retains the specific benefits of each and may enchance the probability of response when compared with either psychotherapeutic or psychopharmacologic monotherapy, especially in chronic depression.14 The ongoing Prevention of Recurrence in Depression with Drugs and Cognitive Therapy study (Hollon SD et al., unpublished data) may help further our understanding of CBT in the acute treatment setting when combined with medications and may illustrate its potential in the prevention of future episodes.
Evolving Empirical Data
Until recently, there has been a paucity of data on managing treatment-resistant depression with combination modalities. Given the complexity involved in attempting to control for the many variables that can affect outcome (eg, the variance among a group of patients), along with the growing number of possible treatment combinations, it is understandable that there are not better “guidelines.” It would require large numbers of subjects over time to control for just a few of the many possible confounding factors affecting the outcomes in performing these studies. Additionally, the ethics of using placebo for extended periods of time in a group of patients with treatment-resistant depression is another issue.
Despite these limitations, a growing number of controlled trials are occurring in this area, with more reports in the literature on acute and long-term combination strategies and a growing number of trials in “defined” treatment-resistant depression populations. The Systematic Treatment Alternative for Relieving Depression (STAR*D) protocol15 is a good example of the endeavor to ascertain the outcome of sequential treatment strategies, including specific monotherapy antidepressants, combination antidepressants, CBT, and augmentation strategies.
Another study that may complement the STAR*D outcome is the Prevention of Recurrence in Depression with Drugs and Cognitive Therapy study (Hollon SD et al., unpublished data). While one of the primary objectives of this study is to examine the role of CBT in preventing the recurrence of depression in a group of patients at high risk for recurrence, all subjects are treated with pharmacologic treatments, with the goal to achieve “remission” during acute treatment and sustain “remission” during continuation treatment. Different from the STAR*D design, pharmacotherapy is flexible and tailored to the individual patient over time. While there is a basic premise to maximize monotherapy antidepressant treatment before switching, bridging, combining, or augmenting with other pharmacotherapy, it provides the ability to tailor the treatment to the multiple factors (eg, treatment history, residual symptoms).
The protocol allows up to a year to achieve remission and is designed to provide enough time to use multiple strategies, including assuring the ability to attempt a TCA or MAOI before the year, if a patient fails to remit. Therefore, the result should be a wealth of data on various pharmacotherapies, as monotherapy or as combinations, used to achieve and sustain remission. Preliminary data of the first 358 patienss show only a minority achieve remission using monotherapy. This is a testimony to both the ability to achieve high rates of remission when treating depression and the necessity in a significant number of patients to require some type of pharmacotherapy combination to achieve these rates of remission.
There is a growing literature to help guide clinicians in managing treatment-resistant depression, including the data on the use of combination treatments. While the clinical approach to using combination treatments needs to be tailored to the individual patient over time, the availability of empirically based strategies should allow us to optimally treat and minimize the negative effects of treatment-resistant depression.
- Murray CJ, Lopez AD. Alternative projections of mortality and disability by cause 1990–2020: Global Burden of Disease Study. Lancet. 1997;349(9064):1498–1504. doi:10.1016/S0140-6736(96)07492-2 [CrossRef]9167458
- Nierenberg AA, Keefe BR, Leslie VC, et al. Residual symptoms in depressed patients who respond acutely to fluoxetine. J Clin Psychiatry. 1999;60(4):221–225. doi:10.4088/JCP.v60n0403 [CrossRef]10221281
- Hirschfeld RM, Keller MB, Panico S, et al. The National Depressive and Manic-Depressive Association consensus statement on the undertreatment of depression. JAMA. 1997;277(4):333–340. doi:10.1001/jama.1997.03540280071036 [CrossRef]9002497
- Thase ME. Therapeutic alternatives for difficult-to-treat depression: a narrative review of the state of the evidence. CNS Spectr. 2004;9(11):808–16, 818–21. doi:10.1017/S1092852900002236 [CrossRef]15520605
- American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders [text revision]. 4th ed. Washington, DC: American Psychiatric Publishing; 2000.
- Paykel ES, Ramana R, Cooper Z, et al. Residual symptoms after partial remission. Psychol Med. 1995;25(6):1171–1180. doi:10.1017/S0033291700033146 [CrossRef]8637947
- Zajecka JM, Goldstein C. Combining medications to achieve remission. In: Schwartz T, ed. Handbook of Treating Depression. In press.
- Physician's Desk Reference95th ed. Montvale, NJ: Thompson PDR; 2005.
- Lee KC, Finley PR, Alldredge BK. Risk of seizures associated with psychotropic medications: emphasis on new drugs and new findings. Expert Opin Drug Saf. 2003;2(3):233–247. doi:10.1517/147403220.127.116.11 [CrossRef]12904103
- Sproule BA, Naranjo CA, Brenmer KE, Hassan PC. Selective serotonin reuptake inhibitors and CNS drug interactions. A critical review of the evidence. Clin Pharmacokinet. 1997;33(6):454–471. doi:10.2165/00003088-199733060-00004 [CrossRef]
- Frank E, Novick D, Kupfer DJ. Antidepressants and psychotherapy: a clinical research review. Dialogues Clin Neurosci. 2005;7(3):263–272.16156384
- Kennedy N, Paykel ES. Treatment and response in refractory depression: results from a specialist affective disorders service. J Affect Disord. 2004;81(1):49–53. doi:10.1016/S0165-0327(03)00192-7 [CrossRef]15183599
- Groves DA, Brown VJ. Vagal nerve stimulation: a review of its applications and potential mechanisms that mediate its clinical effects. Neurosci Biobehav Rev. 2005;29(3):493–500. doi:10.1016/j.neubiorev.2005.01.004 [CrossRef]15820552
- Hollon SD, Jarrett RB, Nierenberg AA, et al. Psychotherapy and medication in the treatment of adult and geriatric depression: which monotherapy or combined treatment?J Clin Psychiatry. 2005;66(4):455–468. doi:10.4088/JCP.v66n0408 [CrossRef]15816788
- Rush AJ, Fava M, Wisniewski SR, et al. STAR*D Investigators Group. Sequenced treatment alternatives to relieve depression (STAR*D): rationale and design. Control Clin Trials. 2004;25(1):119–142. doi:10.1016/S0197-2456(03)00112-0 [CrossRef]15061154