Description of the condition
Major depressive disorder is a psychiatric condition characterised by depressed mood and diminished interest or pleasure [
1]. Major depressive disorder is associated with cognitive deficits leading to functional and occupational impairment [
2]. The prevalence of major depressive disorder is estimated to be more than 264 million people globally, making it one of the leading contributors to functional disability [
3]. Additionally, the high prevalence of major depressive disorder leads to an extensive economic burden estimated at more than 210 billion US dollars annually in the US alone, deriving from direct medical costs as well as costs related to occupational disability and comorbidities [
4]. Furthermore, major depressive disorder is associated with an increased risk of suicidal behaviour, with an estimated 15% of patients having attempted suicide at least once in their lifetime [
5‐
7].
Description of interventions
Tricyclic antidepressants are a group of first-generation antidepressants commonly used for treating major depressive disorder, obsessive–compulsive disorder and chronic pain [
8,
9]. The first tricyclic antidepressant, imipramine, was developed in the 1950s by modifying the phenothiazine ring and substituting sulphur with an ethylene bridge [
9]. The majority of tricyclic antidepressants function as serotonin and norepinephrine reuptake reuptake inhibitors [
10,
11]. By blocking the reuptake of monoamine neurotransmitters in the presynaptic neuron, tricyclic antidepressants theoretically increase the levels of serotonin and norepinephrine in the synaptic cleft [
10,
12]. However, the role of monoamines in major depression is unclear and the exact mechanism of action of tricyclic antidepressants is uncertain [
10,
13‐
15].
Whilst selective serotonin reuptake inhibitors are generally recommended as first-line treatment for major depressive disorder, tricyclic antidepressants are amongst recommended treatments for patients whose condition does not improve after treatment with newer medications [
16‐
18]. The World Health Organisation Model List of Essential Medicines includes the tricyclic antidepressant amitriptyline as one of just two essential antidepressants for the treatment of major depressive disorder [
19].
Why is it important to do this review?
Several systematic reviews with meta-analysis have assessed the beneficial effects of tricyclic antidepressants and have concluded that tricyclic antidepressants reduce depressive symptoms with a statistically significant effect for patients with major depressive disorder [
20‐
23]. Some systematic reviews have concluded that tricyclic antidepressants, either as a drug class [
20] or as an individual drug [
21], are indeed the most effective antidepressants [
20,
21]. However, the effect sizes of tricyclic antidepressants were small and may not be important to the average patient [
24]. Furthermore, trials comparing antidepressants with ‘active placebo’ (a placebo that mimics the adverse effects of the experimental intervention) indicate that the beneficial effects may in fact be inflated due to the unblinding effects of using an inert placebo [
25].
Tricyclic antidepressants are associated with a broad spectrum of adverse effects, but the serious and non-serious adverse events associated with all types of tricyclic antidepressants have not been systematically assessed in adults with major depressive disorder. A recent network meta-analysis published in The Lancet in 2018 included placebo-controlled and head-to-head trials to assess the effects of 21 commonly used antidepressants, including two tricyclic antidepressants, amitriptyline and clomipramine [
21]. The results showed that antidepressants compared with placebo seemed to reduce depressive symptoms with a statistically significant effect (standardised mean difference (SMD) 0.30, 95% credibility interval 0.26 to 0.34) [
21]. The results also showed that amitriptyline was the most effective antidepressant for reducing depressive symptoms (odds ratio (OR) 2.30, 95% credibility interval 1.89 to 2.41), and that clomipramine was one of the least effective antidepressants for reducing depressive symptoms in the meta-analysis (OR 1.49, 95% credibility interval 1.21 to 1.85) [
21]. However, neither serious nor non-serious adverse events were assessed. Instead, the authors assessed lack of ‘acceptability’ (treatment discontinuation measured by the proportion of participants who withdrew for any reason) and the proportion of participants who dropped out early because of adverse effects [
21]. Such data on withdrawals as surrogate markers for safety or tolerability should, however, be interpreted with caution due to a number of issues that include difficulty attributing reasons for discontinuation, pressures on patients and investigators to reduce the number of withdrawals, and unblinding that often precedes decisions to withdraw [
26].
A Cochrane review published in 2003 investigated effects of low dosage tricyclic antidepressants compared with placebo or standard dosage tricyclic antidepressants in the acute-phase treatment of depressive disorder [
23]. Thirty-five trials (2013 participants) compared low dosage tricyclic antidepressants with placebo, and six trials (551 participants) compared low dosage tricyclic antidepressants with standard dosage tricyclic antidepressants [
23]. The authors found that low dosage tricyclic antidepressants were more effective in reducing depressive symptoms than placebo, and that standard dosage tricyclic antidepressants were not significantly more effective in reducing depressive symptoms compared with low dosage tricyclic antidepressants [
23]. Low dosage tricyclic antidepressants were found to be more likely than placebo to cause at least one adverse effect, and standard dosage was more likely than low dosage tricyclic antidepressants to cause dropouts due to adverse effects [
23]. Serious adverse events, suicides and suicide attempts were not assessed. Additionally, this review did not compare standard dosage tricyclic antidepressants with placebo, and not all types of tricyclic antidepressants were included [
23].
A meta-analysis of 15 randomised clinical trials published in 2005 assessed the efficacy and tolerability of tricyclic antidepressants and selective serotonin reuptake inhibitors compared with placebo for treatment of depression in primary care [
22]. The results showed that tricyclic antidepressants compared with placebo reduced depressive symptoms with a statistically significant effect (SMD − 0.42, 95% confidence interval − 0.55 to − 0.30) [
22]. The authors also found that tricyclic antidepressants increased the risk of withdrawal from the trial due to drug-related adverse events [
22]. However, the meta-analysis only assessed drug-related adverse events (adverse reactions) and did not assess all adverse effects including serious adverse events. Furthermore, the risk of suicide and suicide attempts were not assessed, and the meta-analysis was limited by only including trials in a primary care setting [
22].
Given the limitations of extant systematic reviews, we aim to investigate the beneficial effects and serious and non-serious adverse effects of tricyclic antidepressants for major depressive disorder in adults including both published and unpublished data. Our systematic review will take bias risk (systematic errors), play of chance (random errors) and certainty of the findings into consideration. This systematic review will be conducted as part of a larger project investigating the beneficial and harmful effects of all antidepressants for major depressive disorder [
27]. In addition to this systematic review, we will also publish separate systematic reviews for selective serotonin reuptake inhibitors, duloxetine [
28], venlafaxine and mirtazapine [
27]. These systematic reviews will ultimately provide data for a systematic review investigating the effects of all antidepressants for major depressive disorder [
27]. We chose to publish the present protocol and systematic review separately to investigate the effects of tricyclic antidepressants in more detail (i.e. more outcomes) [
27].