Clinical importance of long-term antidepressant treatment

Background Depression, which only a few decades ago was considered to be a short-term illness requiring short-term treatment, isnow recognised as a recurrent, sometimes chronic, long-termillness.

Aims To highlight the clinical importance of long-term antidepressanttherapy in the treatment of depression.

Method The current literature was reviewed to examine the relationshipbetween duration of antidepressant therapy and efficacy.

Results Approximately one-third to a half of patients successfully stabilised in acute-phase treatment will relapse if medicationis not sustained throughout the continuation period. Only 10-15%will relapse if medication is continued. For maintenance-phasetherapy, approximately 60% of patients at risk will experiencea recurrent episode of depression within 1 year if untreated,whereas those who continue in treatment will have a recurrencerate of between 10% and 30%.

Conclusions Risk of relapse and recurrence of depression canbe significantly reduced if adequate continuation and maintenancetherapy durations are achieved.

INTRODUCTION

TOP

INTRODUCTION

Concepts of depression have changed markedly over the past century.In the early part of the 20th century, Kraepelin describeddepression as a long-term affliction, characterised by recurrencesand chronicity. In the 1960s and 1970s came the psychopharmacologicalrevolution. Newly developed medications for depression —monoamine oxidase inhibitors (MAOIs) and tricyclic antidepressants(TCAs) — caused dramatic improvement within weeks. These developments led to a reformulation of depression as an illnessthat could be ‘cured’ quickly, like a bacterialinfection. The development of short-term, focused psychotherapiessuch as cognitive—behavioural and interpersonal therapyparalleled the psychopharmacological advances, and also fosteredthis view of depression as a short-term illness.

Research findings emerging in the 1980s challenged this view.For example, the National Institute of Mental Health CollaborativeProgram of the Psychobiology of Depression involved a multi-yearfollow-up of over 400 patients. After 15 years, only one ineight of these patients had recovered from their original episodeand stayed well (Keller et al, 1992). Over 80% had sufferedat least one recurrence, and 6% had remained chronically depressedthroughout the entire 15-year period.

These and other findings, as well as experience with antidepressant pharmacotherapy, have led physicians to return to the Kraepelineannotion of depression as a long-term, recurrent and often chronicillness. Treatment for the disorder has both short-term andlong-term aspects: short-term treatment includes acute andcontinuation phases, while long-term treatment consists of the maintenance phase (Kupfer, 1991).

The acute phase involves stabilisation of the acute symptomsand generally lasts up to 3 months (Fig. 1). However, in complicatedcases, acute treatment may take considerably longer. The continuationphase begins with stabilisation and ends at the point at whichthe depression would have ended had there been no successfultreatment (often thought to be an additional 6-12 months beyondacute symptom resolution). If depressive symptoms return duringthe continuation period, it is considered to be a relapse intothe original episode of illness. Thus, treatment for a singleepisode can take over a year, but in most cases will last 6-12months. The maintenance phase involves prevention of new episodes (recurrences) and can extend for years.

View larger version (13K):
[in this window]
[in a new window]
[as a PowerPoint slide]
Fig. 1 Response, remission, recovery, relapse and recurrence of depression. From Kupfer (1991).

ACUTE TREATMENT

TOP

ACUTE TREATMENT

The efficacy of pharmacotherapy for the acute phase of the treatmentof depression is well established. Janicak et al (1997) summarisednearly 80 studies of TCAs and 16 studies of MAOIs in the acutetreatment of major depression. He concluded that the statisticallikelihood of chance being responsible for the active drug-placebodifference was less than 10^-40 for TCAs and less than 10^-11for MAOIs. In his meta-analysis, nearly two-thirds of patientsresponded to active drug, whereas about one-third respondedto placebo. These analyses reveal a substantial drug effect,but many patients (one-third) continue to suffer from symptomsof depression even after vigorous treatment. Thus, the effectsize is less than many would hope for.

CONTINUATION TREATMENT

TOP

CONTINUATION TREATMENT

The second phase of short-term anti-depressant treatment isthe continuation phase. The purpose of continuation treatmentis to prevent relapse, the re-emergence of symptoms, followingpremature withdrawal of medication after acute stabilisationhas been achieved. The number of studies on continuation therapyis small compared with those on acute therapy. The key questionin these studies is whether continued treatment on medicationis necessary after acute stabilisation of symptoms.

Six continuation studies of TCAs, primarily amitriptyline, wereidentified (Table 1). Overall, less than one-fifth of thoseon active drug therapy relapsed compared with over half of those on placebo.

View this table:
[in this window]
[in a new window]

Table 1 Continuation studies in depression

Several similarly designed studies of selective serotonin reuptake inhibitors (SSRIs) and other newer agents have also been conducted.For example, in an international, multi-centre study, 467 patientswith major depressive disorder were enrolled in an open-labeltrial of sertraline for 8 weeks (Doogan & Caillard, 1992).Of the 300 responders, 185 agreed to be randomised either tostay on sertraline or to be blindly switched to placebo. After approximately 10 months of follow-up, nearly 13% of patientstaking active medication relapsed compared with 46% on placebo(P<0.001) (Doogan & Caillard, 1992). Most of the relapsesin this study occurred within the first 6 months of continuationtreatment (Fig. 2).

View larger version (12K):
[in this window]
[in a new window]
[as a PowerPoint slide]
Fig. 2 Patients not relapsing during continuation study of sertraline. From Doogan & Caillard (1992).

Additional continuation studies involving paroxetine, nefazodone, citalopram, reboxetine, amineptine and fluoxetine, with nearlyidentical protocol designs, gave very similar results (Montgomery & Dunbar, 1993;Montgomery et al, 1993; Robert & Montgomery, 1995;Ferreri et al, 1997; Reimherr et al, 1998;Feiger et al, 1999; Versiani et al, 1999). Approximately one-thirdto one-half of patients who did not continue active treatmentafter stabilisation (i.e. were switched to placebo) relapsed,whereas about 15% relapsed if they continued on active medication(Table 1).

The continuation study conducted by Reimherr et al was designedto assess not only rates of relapse, but also the questionof how long continuation therapy should last. Patients withmajor depression were treated in an open-label trial with 20mg per day of fluoxetine for 12 weeks (Reimherr et al, 1998).Patients who had sustained remission during this period wererandomised either to continue on fluoxetine or to be blindlyswitched to placebo. Groups of patients in the fluoxetine armwere then randomised to placebo after a further 14 or 18 weeks(for a total of 26 or 50 weeks of treatment), or were maintainedon fluoxetine for the entire 50 weeks of the continuation period(for a total of 62 weeks of treatment) (Fig. 3).

View larger version (12K):
[in this window]
[in a new window]
[as a PowerPoint slide]
Fig. 3 Fluoxetine (FLX) continuation study design. From Reimherr et al (1998).

The results for the patients who were randomised at week 12to placebo compared with those staying on fluoxetine were consistentwith the earlier reported studies. Approximately 20% of thoseon fluoxetine relapsed, in comparison with nearly half of thoseon placebo (Fig. 4).

View larger version (15K):
[in this window]
[in a new window]
[as a PowerPoint slide]
Fig. 4 Cumulative probability of remaining well during continuation treatment with fluoxetine (solid lines) or placebo (dashed lines). (a) Comparison interval 1 (weeks 12-24); (b) interval 2 (weeks 26-38); (c) interval 3 (weeks 50-62). From Reimherr et al (1998).

A second group of patients were blindly discontinued at week26 from fluoxetine. The rate of relapse for these patients(approximately 21%) was less than that found in patients discontinuedat week 12 (approximately 50%), but was still significantlyworse than the approximately 5% relapse rate in patients whohad remained on fluoxetine throughout the 12-week period. There was no significant difference in relapse between those randomisedto placebo at week 50 compared with those who stayed on fluoxetinefor an additional 12 weeks (both approximately 12%). Theseresults suggest that continuation therapy should be given forat least 3 months following remission, but in most patientsneed not be given beyond 9 months following remission.

In summary, results of continuation studies of both older andnewer classes of antidepressant consistently demonstrate thatapproximately one-third to half of all patients will relapseif medication is not sustained throughout the continuationperiod. Only 10-15% will relapse if medication is continued. The results from the study by Reimherr et al indicate that continuation therapy should last at least 3-9 months followingremission.

MAINTENANCE TREATMENT

TOP

MAINTENANCE TREATMENT

The objective of maintenance therapy, the long-term aspect of antidepressant treatment, is to prevent the occurrence of anew episode of depression after full recovery from a previousepisode. Maintenance therapy should be strongly consideredfor all patients at risk of recurrence. Such patients are thosewith a history of three or more prior episodes of major depression,pre-existing dysthymia, severe episodes, seasonal patterns,a familial history of affective disorder, a poor response tocontinuation therapy, and comorbid anxiety (or similar disorders)or substance misuse problems (Hirschfeld & Schatzberg, 1994).

A number of maintenance therapy studies have been performedusing TCAs and one has been performed using an MAOI (Table 2).Although the recurrence rates for active drug versus placebo differ in these studies, the data consistently demonstrate asignificantly higher rate of recurrence in patients given placebocompared with those who continue on active drug therapy. Whilethese older classes of antidepressants have proven clinicalefficacy, their use in the real world is hampered by concernsabout side-effects (including anticholinergic, cardiovascularand histaminic effects), which can lead to non-compliance,as well as the risk of fatal overdose in patients who stockpilemedication.

View this table:
[in this window]
[in a new window]

Table 2 Maintenance studies in depression

Maintenance studies involving the SSRIs fluoxetine, fluvoxamineand sertraline have also been performed. In one study, 456patients with DSM—III (American Psychiatric Association, 1980)recurrent major depression were treated with open-labelfluoxetine for 6 weeks (Montgomery et al, 1988). Those whoresponded were continued on fluoxetine therapy for an additional18 weeks. Those who completed this 24 weeks of treatment wererandomised to continue with fluoxetine (n=88) or to switch blindly to placebo (n=94). Recurrence of depression was definedby a score above 18 on the Hamilton Rating Scale for Depression (Hamilton, 1967). A recurrence rate of 26% occurred in thefluoxetine group during the 1-year follow-up compared with57% in the placebo group (P<0.001) (Fig. 5).

View larger version (10K):
[in this window]
[in a new window]
[as a PowerPoint slide]
Fig. 5 Patients without recurrence of depression during 1-year maintenance study; ^*P<0.01. From Montgomery et al (1988).

A similarly designed study using fluvoxamine gave results consistentwith those observed in the fluoxetine study. A mean daily doseof 100 mg of fluvoxamine was shown to be effective in preventingrecurrence of DSM—III—R (American Psychiatric Association, 1987)major depressive disorder. Recurrence was defined as a reappearance of depressive symptoms in the opinionof the investigator. The fluvoxamine group had a recurrencerate of 13%, whereas the placebo group had a recurrence rateof 35% (P<0.001) (Terra & Montgomery, 1998).

The studies described above and summarised in Table 2 suggestthat, among patients at risk, depression will recur in approximately60% within 1 year if untreated. If patients continue to receivetreatment with newer antidepressants, the recurrence rate dropsto between 10% and 30% (Table 2). This clearly supports theefficacy and advisability of maintenance treatment if patients are among those at risk.

CONCLUSION

TOP

CONCLUSION

Depression is a recurrent and often a lifelong illness, whichrequires long-term treatment. The notions prevalent in the1960s and 1970s that depression could be treated successfullywithin a period of weeks have been modified in the light ofsubstantial evidence to the contrary. All patients with majordepression who have successfully responded to an antidepressant drug should be continued on medication for a period of 3-6 monthsbeyond acute symptom resolution. If not, up to half of themwill relapse back into their original episode. Patients atrisk of recurrence should be considered for maintenance therapy.Such patients will almost certainly experience a recurrenceof depression within several years without continued treatment.

REFERENCES

TOP