Depressive episode: does antidepressant treatment prevent the risk of depressive disorder?

Eduard Vieta, MD, PhD

Michael BAUER, MD, PhD
Professor of Psychiatry
Department of Psychiatry and Psychotherapy, University
Hospital Carl Gustav Carus Technische Universität Dresden
Dresden, GERMANY

Depressive episode: does antidepressant treatment prevent the risk of depressive disorder?

by M. Bauer, Germany

The long-term course of unipolar major depressive disorder (MDD) is characterized by high rates of recurrence and prolonged symptomatic chronicity. The primary goals of continuation andmaintenance treatment are to prevent a fast relapse into depression or new episode of depression (recurrence). Adverse prognostic indicators for recurrence include: high number of previous episodes; residual symptoms at remission; previous longer episodes and chronicity; more severe previous episodes; onset early in life; concurrent dysthymic disorder (“double depression”); concurrent substance abuse or anxiety disorders; and family history ofMDD in first-degree relatives. Key elements of long-term treatment include: pharmacotherapy; psychoeducation; and adherence monitoring. Extending treatment for an additional 6 months (continuation therapy) can reduce the likelihood of relapse by about 70%, and extending treatment for another 12 months or longer (maintenance therapy) can reduce the risk of recurrence. Most patients receive antidepressants during the acute and continuation phase, and the best treatment recommendation to prevent relapse and recurrence of depression is to continue the antidepressant medication at the same dose during these treatment phases as well. Randomized placebo-controlled efficacy studies (RCTs, usually conducted 1 or 2 years after remission) indicate that all major classes of antidepressants are effective in preventing recurrence of depression with about a twofold higher relapse rate with placebo treatment. Evidence suggests that the “newer” antidepressants have superior long-term efficacy and better tolerability compared with traditional antidepressants, eg, the tricyclics. Although concerns about the generalizability of results fromplacebo RCTs can be raised, RCTs testing the efficacy of a new antidepressant in comparison with placebo remain the gold standard for proof of efficacy requested by the regulatory authorities. Effectiveness trials studying patients in the “true world” provide additional important information, but due to methodological reasons they cannot replace placebo-controlled RCTs.

Medicographia. 2011;33:151-157 (see French abstract on page 157)

Major depressive disorder: a highly recurrent disease

Major depressive disorder (MDD) presents typically as a recurrent disorder. 50% to 85% of the patients who suffer a depressive episode will have another episode of major depression.1 The likelihood of a recurrence increases with the number of previous depressive episodes and the severity of the current episode. Patients who have had three episodes of major depression have a 90% chance of having another.2 Among other risk factors for recurrence of MDD, prior history of multiple episodes of MDD, early age at onset, persistence of dysthymic symptoms after recovery from an episode of MDD, presence of an additional, non-mood psychiatric diagnosis, and presence of a chronic physical disorder have been identified.3 Factors that have been associated with increased severity of subsequent depressive episodes include a history or a prior episode complicated by serious suicide attempts, psychotic features, or severe functional impairment.

Keeping these facts in mind, it is apparent that the primary goals of long-term, maintenance (prophylactic) treatment are to prevent a new episode of depression (a recurrence) and development of chronicity. A recurrence is an episode that appears after a completely asymptomatic period (remission) has been achieved for a 6-month period.4,5 The consideration of the patient’s course of illness and treatment history is essential for the implementation ofmaintenance phase therapy. Even though no definite recommendation can be given as to when prophylactic therapy should be initiated, it is clearly indicated in situations associated with a high risk of recurrence (Table I). In addition to the risk factors shown in Table I, patient preference, severity of functional impairments, and side effects experienced during the continuation phase also play a role in determining whether or not maintenance treatment should be implemented.6,7

Key elements of relapse prevention

Key elements of relapse/recurrence prevention treatment of recurrent depressive disorders include: (i) psychoeducation; (ii) pharmacotherapy; and (iii) adherence monitoring. Adjunctive depression-targeted psychotherapy may also be included in the treatment plan. Because maintenance treatment requires compliance with medication, education, and a close therapeutic alliance with patients and their families/friends are essential. Education does not only reduce treatment attrition, but also leads to a better outcome.8 Strategies to prepare patients and their families for maintenance treatment include education on the typical course of the illness, treatment options, medication effects and side effects, use of (daily) selfreport instruments to track mood and early warning signs of relapse or recurrence, long-term perspectives, and projected end of treatment. A relapse prevention program (a low intensity intervention including enhanced patient education, visits with a depression specialist, telephone calls, symptom monitoring) for depressed patients in primary care significantly improved antidepressant adherence and depressive symptoms outcome in a randomized controlled 12-month trial compared with usual primary care.9

Table I. Factors associated with increased risk for recurrence in
major depressive disorder.

Relapse prevention in depression with antidepressant medication

Pharmacotherapy is the most studied treatment modality in the long-term treatment of recurrent MDD. Among the therapeutic options available, antidepressant medications, and lithium have received the most study. The majority of controlled trials investigating these medications in maintenance treatment demonstrated efficacy for relapse prevention.6,10-14 The medication of first choice is either the antidepressant with which remission was achieved in the acute/continuation phase or lithium.2,6,7,15 Likely reasons why antidepressants may be preferred to lithium are that patients are usually treated with antidepressants during the acute/continuation phase and that patients usually prefer to use medication that does not require regular monitoring by blood tests. The final choice of prophylactic agent does depend on how individual patients respond to and tolerate treatment with antidepressants and lithium. Patients’ preference and their own or their family member’s experience with maintenance treatment are also helpful in the choice of the medication.

Antidepressants for relapse prevention

Randomized placebo-controlled studies (usually 1 or 2 years after remission) of antidepressants for relapse prevention treat- ment of depression indicate that literally all major available antidepressants are effective in preventing relapse of depression: tricyclic antidepressants (TCAs), eg, amitriptyline, imipramine, nortriptyline; irreversible monoamine oxidase inhibitors (MAOIs), eg, phenelzine;10,11,13,16 selective serotonin reuptake inhibitors (SSRIs), eg, citalopram, escitalopram, fluoxetine, fluvoxamine, paroxetine, sertraline);13,17-19 selective serotonin and norepinephrine reuptake inhibitors (SSNRIs), eg, duloxetine, venlafaxine;20,21 bupropion;22 and agomelatine23 have been tested with positive results by either indicating significant lower relapse rates (with placebo about twice as high as with active agents) or a longer time to relapse compared with placebo treatment. To give an example: in a most recent study, the cumulative relapse rate at 6 months for agomelatine-treated patients was 21.7% and for placebo-treated patients 46.6%.23 The calculated effect sizes (and the active treatment– placebo differences) in these studies appear to be even higher compared with acute treatment studies. Relapse rates are high in the first months after remission in particular and decline with time. From naturalistic follow-up studies, the rates of relapse following remission have been estimated at 20% to 24% by 2 months, 28% to 44% by 4 months, 27% 50% by 6 months, and 37% 54% by 12 months.14

A meta-analysis of discontinuation randomized controlled trials (RCTs) in patients with MDD reported that 60% of patients on placebo relapsed in the year after randomization and 29% relapsed in months 12 to 36.24 For clinical decision making it is also useful to express the relative benefit of an active treatment over placebo using the “number needed to treat” (NNT). The NNT concept represents one component of the complex benefit/risk estimation of antidepressant treatments. It represents the minimal number of patients who have to be treated to reach a benefit in one patient. In relapse prevention treatment studies, the active treatment–placebo difference seems to be even higher compared with acute treatment studies of antidepressants: in a pooled analysis of 31 RCTs in 4410 patients, a rounded NNT of 5 (4.34) for the relapse prevention during the first 12 months after successful acute treatment has been calculated.24

In the majority of these relapse prevention trials, however, only responders to the drug tested during the acute phase were included in the continuation/maintenance randomized, double- blind treatment phase. This design has been described as introducing a potentially significant selection bias that limits the generalizability of results.25 Strictly speaking, findings are only relevant to the population of patients who are responsive to study medication during the acute phase. To overcome this selection bias, continuation and maintenance medication should be assessed among patients achieving initial remission through other means and antidepressant medications besides the maintenance drug.18 The latter design, however, is more difficult to conduct for obvious reasons (and subsequently more expensive) and brings up the question why patients should be treated in the continuation/maintenance phase with a medication that they have not responded to in the acute phase.

Efficacy and effectiveness of antidepressants

_ Antidepressant efficacy
There is an ongoing debate as to whether randomized, placebo- controlled trials are the best and only way to test if a medication is effective in certain patient populations. There is one clear answer to that: to receive approval for a specific indication, regulatory authorities, eg, the European Medicines Agency (EMA) and the US Food and Drug Administration (FDA) request RCTs testing the efficacy of a new antidepressant in comparison with placebo or in comparison with placebo and a standard (comparator) drug.26 The term “efficacy” is usually defined as the power of an antidepressant to produce antidepressant effects under “ideal” conditions. According to the British National Institute for Clinical Excellence (NICE) it is defined “as the extent to which a specific treatment or intervention, under ideally controlled conditions (eg, in a laboratory), has a beneficial effect on the course or outcome of disease compared with no treatment or other routine care.27 The mean differences of scores on any applied depression rating scale (typically the Hamilton Rating Scale for Depression [HAM-D] or the Montgomery Åsberg Depression Rating Scale [MADRS]) between active antidepressant drugs and placebo shown in pivotal RCTs are used for decision making by the regulatory authorities to determine whether new antidepressants may receive approval or not. Unfortunately, placebo effect plays a significant role in clinical trials of MDD. Recent meta-analyses showed that the mean responder rate in placebo-treated groupswas 29.7%and that this rate had been growing over the past decades of drug development.28 The further consideration of other additional efficacy indicators such as response and remission rates and the number needed to treat (NNT) is useful.

The efficacy of antidepressant medication in the acute treatment of MDD in adults is well proven by a large number of RCTs.13,14 This has been investigated in many RCTs comparing active antidepressants with placebo treatment and with active comparators. Most publications are reporting responder and remitter rates together with end-point differences in depression rating scales such as different versions of the HAM-D or the MADRS scale using the “last observation carried forward” (LOCF) method, which follows all included patients. Usual responder rates vary between 32% and 70%,28 HAM-D17 differences between active and placebo treatment after 6 to 8 weeks of treatment often reach only 3 to 4 absolute points, but represent mean reductions in comparison to the scores before treatment of about 50% to 60%.

_ Antidepressant effectiveness
In contrast, “effectiveness “of an antidepressant drug can be achieved during the use of the substance in typical clinical circumstances and in larger populations of patients in the real world. In most studies, responder and remitter rates are higher in effectiveness studies compared with RCTs.29 Serious concerns about the generalizability of results from placebo RCTs have been raised in the literature. The standard exclusion criteria of most trials, which have become much more stringent over the past decade, do exclude a significant number of patients suffering from suicidality, comorbid axis I disorders, and medical illnesses. In the case of trials investigating the efficacy of antidepressants, patients with a history of treatment failures and long depressive index episodes are generally also excluded from participation. Furthermore, simply the use of placebo, randomization, and blinding procedures excludes many patients who refuse study participation because of these procedures that may in their view reduce the chances of improvement. Ghaemi30 has estimated that fewer than 10% of depressed patients qualify for, and agree to, participate in available RCTs of antidepressants. Subsequently, the assumption in the world of clinical trials is that the research conducted on these 10% is generalizable to the other remaining 90% of patients. Although unproven, it may well be that results of a placebo-controlled RCT would be different if a significant proportion of these 90% were included in the trial. Nevertheless, according to NICE and regulatory authorities such as the EMA, RCTs remain the most important method for establishing treatment efficacy and estimating the clinical effectiveness of antidepressants. Approval based on effectiveness studies only would introduce other biases (eg, by unblinded ratings, no placebo as a comparator drug), resulting in false results.

Comparative efficacy of antidepressants

A relatively small number of studies have directly compared different medications for maintenance treatment in recurrent unipolar depression.10 A meta-analysis of studies comparing lithium with other antidepressants showed no conclusive advantage for lithium in the prophylaxis of unipolar illness.31 In one relatively small randomized, placebo-controlled 2-year maintenance study, lithium (serum level 0.8 to 1.2 mmol/L) was superior to imipramine (100 to 150 mg/day); the combination of lithium and imipramine was not superior to lithium alone.32 Another, larger randomized, placebo-controlled 2-year study reported better maintenance effects for imipramine (the mean daily dosage at the start of the maintenance phase was 137 mg, range 75 to 150 mg/day) than lithium (the mean serum lithium level at the start of the maintenance phase was 0.66 mmol/L, range 0.43 to 1.05 mmol/L).33 In the latter study, the combination of imipramine and lithium did not provide any advantage over imipramine alone in preventing depressive recurrences.

However, in a later reanalysis of the data, the same authors concluded that the results of the latter study could be accounted for by alternative explanations that are a consequence of the study design.25 One randomized, prospective, open, 2.5-year trial comparing lithium (average serum lithium level 0.59 mmol/L) with amitriptyline (average dosage 98 mg/day) found significantly better prophylactic efficacy for lithium.34

The question of dosing and additional psychotherapy in relapse prevention

The majority of patients with a moderate/severe depressive episode receive antidepressants during the acute and continuation phase, and the best treatment recommendation to prevent recurrence of depression is to continue the antidepressant medication that was effective during the acute and continuation phase of treatment at the same dose during the maintenance phase.35,36 In two studies, the group of patients who received only half of the acute-phase dose of imipramine35 or paroxetine36 rather than the full dose, showed a significantly higher recurrence rate. Of course, in case of intolerance or disturbing side effects, the dose needs adaptation (decrease) in the clinical setting.

In the largest and probably most influential study of the use of antidepressants in maintenance treatment, a randomized 3-year placebo-controlled trial, survival analysis showed fulldose imipramine (mean dose at randomization 215 mg/day) withor without Interpersonal Therapy (IPT;weekly for12weeks, then biweekly for 8 weeks, and then monthly) to be the best maintenance treatment, followed by IPT with or without placebo, and then placebo.37 In this study of highly recurrent unipolar depression, all patients enrolled in the 3-year study, had remitted on the combination of imipramine and IPT, and all had remained well for 4 months of continuation therapy prior to randomization. A subsequent additional 2-year placebo- controlled study of the patients who completed the 3- year study37 showed that imipramine (average dose of 200 mg/day) was significantly better than placebo in preventing recurrence.38

Side effect profile: relevance for long-term treatment

Side effects and tolerability of medications are key considerations in maximizing adherence to treatment, and they should be as minimal as possible. Even mild-to-moderate side effects during maintenance treatment may lead to noncompliance with the consequence of symptom worsening and increased risk of recurrence. Using medications with a more favorable side effect profile than the TCAs may facilitate patient compliance with pharmacotherapy, as long as these agents are effective in the maintenance treatment of depression. A number of more recent studies suggest that the “newer” antidepressants have superior long-term efficacy and better tolerability compared with traditional TCAs and tetracyclics.11,13,39 A randomized, placebo-controlled 2-year study comparing the efficacy of mirtazapine with that of amitriptyline found that the time to relapse was significantly longer in the mirtazapine group.40 Similarly, a double-blind 1-year study reported significantly greater improvement in some of the out- come measures in the venlafaxine group compared with imipramine.41 Ideally, the rate and severity of side effects of the antidepressant medication should be at the level of placebo as demonstrated for agomelatine23 and other “newer” antidepressants.

Figure 1. Therapeutic options for maintenance
treatment of major depressive disorder.

Abbreviations: CBZ, carbamazepine; LAM, lamotrigine;
MT, maintenance treatment; VAL, valproate.
Modified after reference 13: Bauer et al. World J Biol Psychiatry.
2007;8:67-104. © 2007, Informa, Plc.

Maintenance treatment options

There is growing recognition that prophylactic treatment of depressive disorders may be inadequate in a substantial proportion of patients. The maintenance treatment of patients with recurrent depression who experience recurrences during prophylactic treatment with standard agents, eg, antidepressants and/or lithium, is one of the most challenging issues in the treatment of these disorders. However, little data from formal studies are available to guide physicians in the maintenance treatment of patients suffering from recurrences during standard prophylactic treatment.12 Combination therapy administering two or even three antidepressants, maybe combined with lithium (or in case of refractoriness or intolerance lamotrigine or valproate), are treatment options for the clinician although there are little controlled data to support such polypharmacy. Figure 1 gives an algorithm for treatment options during the maintenance treatment phase.

Duration and discontinuation of maintenance treatment

The optimal moment to discontinue a longer- term medication is difficult to predict. Current evidence suggests that maintenance treatment should be continued as long as the risk of recurrence persists. That risk is often difficult to assess in the individual patient, particularly after a long period (years) of absence of symptoms/recurrence. It appears that the likelihood of a recurrence increases with the number of previous depressive episodes. However, some authors have argued that there is a similar risk of recurrence whether medication is discontinued after months or years of pharmacotherapy.42 There is good evidence from a controlled 5-year study that patients who benefited the most from continued prophylaxis were those receiving active full-dose medication for at least 5 years.38 Thus, for some patients, maintenance treatment is required for very long periods (eg, a decade) and for others it is required indefinitely.43 Three years maintenance therapy is appropriate almost as a routine for recurrent patients, particularly where an episode prior to the present one has occurred in the last 5 years or where remission has been difficult to achieve. Maintenance for 5 years or indefinitely is recommended for those patients at greater risk, particularly where two or three attempts to withdraw medication have been followed by another episode within 1 year. Only with long-term antidepressant treatment can the risk of development of serious depressive illness with a high relapse and suicide rate be stopped or at least reduced. _

1. Mueller TI, Leon AC, Keller MB, et al. Recurrence after recovery from major depressive disorder during 15 years of observational follow-up. Am J Psychiatry. 1999;156:1000-1006.
2. NIMH Consensus Development Conference Consensus Development Conference Statement. Mood disorders: pharmacologic prevention of recurrences. Am J Psychiatry. 1985;142:469-476.
3. Kovacs M, Devlin B, Pollock M, et al. A controlled family history study of childhood- onset depressive disorder. Arch Gen Psychiatry. 1997;54:613-623.
4. Frank E, Prien RF, Jarrett RB, et al. Conceptualization and rationale for consensus definitions of terms in major depressive disorder. Remission, recovery, relapse, and recurrence. Arch Gen Psychiatry. 1991;48:851-855.
5. Kupfer DJ. Managment of recurrent depression. J Clin Psychiatry. 1993;54(2): 29-33.
6. AHCPR (Agency for Health Care Policy and Research). Depression Guidelines Panel. Depression in Primary Care: Clinical Practice Guideline No. 5. AHCPR pub. No. 93-0550. Rockville, Md; 1993.
7. American Psychiatric Association. Practice guideline for the treatment of patients with major depressive disorder (revision). Am J Psychiatry. 2000;157(4):1-45.
8. Rush AJ. Strategies and tactics in the management of maintenance treatment for depressed patients. J Clin Psychiatry. 1999;60(14):21-26.
9. Katon W, Rutter C, Ludman EJ, et al. A randomized trial of relapse prevention of depression in primary care. Arch Gen Psychiatry. 2001;58:241-247.
10. Solomon DA, Bauer MS. Continuation and maintenance pharmacotherapy for unipolar and bipolarmood disorders. Psychiatr Clin North Am. 1993;16:515-540.
11. AHCPR (Agency for Health Care Policy and Research). Evidence Report on Treatment of Depression: Newer Pharmacotherapies. San Antonio Evidence- Based Practice Center. Washington, DC, AHCPR, Evidence-Based Practice Centers. AHCPR pub. No. 99-E014; 1999.
12. Bauer M, Whybrow PC, Angst J, et al. World Federation of Societies of Biological Psychiatry (WFSBP) Guidelines for Biological Treatment of Unipolar Depressive Disorders, Part 2: Maintenance treatment of major depressive disorder and treatment of chronic depressive disorders and subthreshold depressions. World J Biol Psychiatry. 2002;3:69-86.
13. Bauer M, Bschor T, Pfennig A, et al. World Federation of Societies of Biological Psychiatry (WFSBP) Guidelines for Biological Treatment of Unipolar Depressive Disorders in Primary Care. World J Biol Psychiatry. 2007;8:67-104.
14. Anderson IM, Ferrier IN, Baldwin RC, et al. Evidence-based guidelines for treating depressive disorders with antidepressants: a revision of the 2000 British Association for Psychopharmacology guidelines. J Psychopharmacol. 2008;22 (4):343-396.
15. Paykel ES. Continuation and maintenance therapy in depression. Br Med Bull. 2001;57:145-59.
16. Bauer M, Whybrow PC, Angst J, et al. World Federation of Societies of Biological Psychiatry (WFSBP) Guidelines for biological treatment of unipolar depressive disorders, Part 1: Acute and continuation treatment of major depressive disorder. World J Biol Psychiatry. 2002;3:5-43.
17. Hochstrasser B, Isaksen PM, Koponen H, et al. Prophylactic effect of citalopram in unipolar, recurrent depression: placebo-controlled study of maintenance therapy. Br J Psychiatry. 2001;178:304-310.
18. Lépine JP, Caillard V, Bisserbe JC, et al. A randomized, placebo-controlled trial of sertraline for prophylactic treatment of highly recurrent major depressive disorder. Am J Psychiatry. 2004;161(5):836-842.
19. Kornstein SG, Bose A, Li D, et al. Escitalopram maintenance treatment for prevention of recurrent depression: a randomized, placebo-controlled trial. J Clin Psychiatry. 2006;67(11):1767-1775.
20. Kocsis JH, Thase ME, Trivedi MH, et al. Prevention of recurrent episodes of depression with venlafaxine ER in a 1-year maintenance phase from the PREVENT Study. J Clin Psychiatry. 2007;68:1014-1023.
21. Perahia DG, Maina G, Thase ME, et al. Duloxetine in the prevention of depressive recurrences: a randomized, double-blind, placebo-controlled trial. J Clin Psychiatry. 2009;70(5):706-716.
22. Weihs KL, Houser TL, Batey SR, et al. Continuation phase treatment with bupropion SR effectively decreases the risk for relapse of depression. Biol Psychiatry. 2002;51(9):753-761.
23. Goodwin GM, Emsley R, Rembry S, et al. Agomelatine prevents relapse in patients with major depressive disorder without evidence of a discontinuation syndrome: a 24-week randomized, double-blind, placebo-controlled trial. J Clin Psychiatry. 2009;70(8):1128-1137.
24. Geddes JR, Carney SM, Davies C, et al. Relapse prevention with antidepressant drug treatment in depressive disorders: a systematic review. Lancet. 2003; 361:653-661.
25. Greenhouse JB, Stangl D, Kupfer DJ, et al. Methodologic issues in maintenance therapy clinical trials. Arch Gen Psychiatry. 1991;48:313-318.
26. Baldwin D, Broich K, Fritze J, et al. Placebo-controlled studies in depression: necessary, ethical and feasible. Eur Arch Psychiatry Clin Neurosci. 2003;253: 22-28.
27. National Institute for Clinical Excellence. Depression: management of depression in primary and secondary care. Clinical Guideline 23. 2004.
28. Walsh BT, Seidman SN, Sysko R, et al. Placebo response in studies of major depression: variable, substantial, and growing. JAMA. 2002;287:1840-1847.
29. Hegerl U, Mergl R. The clinical significance of antidepressant treatment effects cannot be derived from placebo-verum response differences. J Psychopharmacol. 2010;24(4):445-448.
30. Ghaemi SN. A Clinician’s Guide to Statistics and Epidemiology in Mental Health. Cambridge, UK: Cambridge University Press; 2009.
31. Souza FGM, Goodwin GM. Lithium treatment and prophylaxis in unipolar depression: a meta-analysis. Br J Psychiatry. 1991;158:666-675.
32. Kane JM, Quitkin FM, Rifkin A, et al. Lithium carbonate and imipramine in the prophylaxis of unipolar and bipolar II illness: a prospective, placebo-controlled comparison. Arch Gen Psychiatry. 1982;39:1065-1069.
33. Prien RF, Kupfer DJ, Mansky PA, et al. Drug therapy in the prevention of recurrences in unipolar and bipolar affective disorders. Report of the NIMH Collaborative Study Group comparing lithium carbonate, imipramine, and a lithium carbonate-imipramine combination. Arch Gen Psychiatry. 1984;41:1096-1104.
34. Greil W, Ludwig-Mayerhofer W, Erazo N, et al. Comparative efficacy of lithium and amitriptyline in the maintenance treatment of recurrent unipolar depression: a randomised study. J Affect Disord. 1996;40:179-190.
35. Frank E, Kupfer DJ, Perel JM, et al. Comparison of full-dose versus half-dose pharmacotherapy in the maintenance treatment of recurrent depression. J Affect Disord. 1993;27:139-145.
36. Franchini L, Gasperini M, Perez J, et al. Dose-response efficacy of paroxetine in preventing depressive recurrences: a randomized, double-blind study. J Clin Psychiatry. 1998;59:229-232.
37. Frank E, Kupfer DJ, Perel JM, et al. Three-year outcomes for maintenance therapies in recurrent depression. Arch Gen Psychiatry. 1990;47:1093-1099.
38. Kupfer DJ, Frank E, Perel JM, et al. Five-year outcome for maintenance therapies in recurrent depression. Arch Gen Psychiatry. 1992;49:769-773.
39. Montgomery SA. New developments in the treatment of depression. J Clin Psychiatry. 1999;60(14):10-15.
40. Montgomery SA, Reimitz PE, Zivkov M. Mirtazapine versus amitriptyline in the long-term treatment of depression: a double-blind placebo-controlled study. Int Clin Psychopharmacol. 1998;13:63-73.
41. Shrivastava RK, Cohn C, Crowder J, et al. Long-term safety and clinical acceptability of venlafaxine and imipramine in outpatients with major depression. J Clin Psychopharmacol. 1994;14:322-329.
42. Thase ME. Long-term nature of depression. J Clin Psychiatry. 1999;60(14):3-9.
43. Rush AJ, Kupfer DJ. Strategies and tactics in the treatment of depression. In: Gabbard GO, ed. Treatment of Psychiatric Disorders. Third Edition. Washington, DC: American Psychiatric Publishing, Inc; 2001:1417-1439.

Keywords: depressive episode; depressive disorder; recurrence; relapse; antidepressant; prevention; tolerability