The goals for acute treatment of depression are full symptom remission and return to baseline functioning. Among patients undergoing primary care treatment for depression, 60 to 80 per cent achieve symptom remission. Remission typically is defined as having normal mood and minimal symptoms, but it is best evaluated using a rating scale (e.g., a score within the normal range on the PHQ-9, Hamilton Depression Rating Scale or Beck Depression Inventory).
Antidepressant medications are the first-choice treatment for moderate to severe depression. (For treatment of mild to moderate depression, see “Psychotherapy.”) The newer antidepressants—selective serotonin reuptake inhibitors (SSRIs), serotonin and norepinephrine reuptake inhibitors (SNRIs), bupropion, mirtazapine and vortioxetine—are all first- line medications, offering improved tolerability and a better safety profile compared with tricyclic antidepressants (TCAs) and monoamine oxidase inhibitors (MAOIs) (Bezchlibnyk-Butler et al., 2014; Stahl, 2017). The following tables summarize antidepressant characteristics.
Table 1: First-Line Antidepressant Medications, Doses and Profiles
Medication (Brand Name)
Usual Daily Dose (mg)
Evidence for superior efficacy is based on meta-analyses (Kennedy et al., 2016).
Evidence for better tolerability is based on meta-analyses (Cipriani et al., 2009).
Evidence for efficacy in anxiety disorders is based on Canadian guidelines (Katzman et al., 2014).
Table 2: Second-and-Third-Line Antidepressants Medications and Doses
Medication (Brand Name)
Usual Daily Dose
Second-line antidpressant a
Levomilnacipran (Fetzima) b
Vilazodone (Viibryd)) b
Third-line antidepressant d
AAP = atypical antipsychotic.
Second-line due to side-effect or safety profile, except where noted.
Second-line due to lack of comparator and maintenance studies.
Take with food.
Use with caution due to dietary restrictions and drug–drug interactions.
Choosing an antidepressant
Usually there is not one definite choice of antidepressant for any given patient because there is so much individual variability in efficacy and side- effects. The choice is based primarily on individual profiles of efficacy, tolerability and anxiety indications. See Table 1 for guidance, and also consider these findings:
There is some evidence for slightly increased efficacy with escitalopram, mirtazapine, sertraline and venlafaxine, particularly in patients with more severe depression.
There is some evidence for better short-term tolerability with bupropi- on, citalopram, desvenlafaxine, escitalopram and sertraline. Bupropion, desvenlafaxine, mirtazapine, vilazodone and vortioxetine have fewer sexual side-effects than other antidepressants.
A broad-spectrum medication (indicated for both depressive and anxiety disorders) is recommended due to high comorbidity of these disorders.
Efficacy for most anxiety disorders is demonstrated with escitalopram, paroxetine, sertraline and venlafaxine. Other antidepressants also may be effective for anxiety disorders, but they have not been studied.
Cognitive dysfunction is commonly seen in depression. There is some evidence that vortioxetine improves depression-associated cognitive dysfunction.
Antidepressant side-effects vary by drug class and specific medication:
SSRIs can cause sexual dysfunction (most commonly decreased libido and delayed orgasm/ejaculation), insomnia, headaches, nausea and weight gain or loss. Some medications (fluvoxamine, paroxetine) may cause sedation.
SNRIs have a side-effect profile similar to that of SSRIs, but they can also cause hypertension, agitation, tremors and sweating.
Newer medications (desvenlafaxine, vilazodone, vortioxetine) and bupropion have fewer sexual side-effects than SSRIs and SNRIs.
Common adverse effects of MAOIs and TCAs include sedation, hypotension and anticholinergic effects (dry mouth, constipation, ocular side-effects, urinary hesitancy).
Uncommon but serious side-effects of SSRIs and SNRIs include hyponatremia and osteoporosis/fractures (most often seen with serotonergic medications), gastrointestinal bleeding (particularly when given with ASA or another NSAID) and QTc prolongation (primarily with TCAs and citalopram). TCAs and MAOIs may cause arrhythmias and postural hypotension, the latter being a risk factor for falls.
Antidepressants and suicide
Antidepressants have been associated with a small increased risk of suicidal thoughts, particularly in adolescents and young adults, and a decreased risk in older age groups. The relative risk may be elevated in some patient groups, but the absolute risk remains very small.
The risk of suicide is highest at the onset of treatment, when neurovegetative symptoms subside without the accompanying improvement in mood. Close monitoring is essential during the first few weeks of treatment. Given the consistent link between depression and suicidality, clinicians should screen for and treat depression attentively.
The goal for maintenance treatment of depression is preventing relapse and recurrence.
For uncomplicated episodes of depression, patients should continue on anti- depressants for at least four to six months after achieving symptom remission.
Patients with risk factors or complicated episodes (chronic, recurrent, severe, comorbid or difficult-to-treat depressive episodes) should continue on antidepressants for at least two years. Some patients will require lifetime treatment.
When stopping a patient’s medications, gradually taper the dose whenever possible (e.g., at least one week for each dose reduction). Caution patients about, and monitor for, discontinuation symptoms, which are usually mild and transient. Use the mnemonic FINISH to identify discontinuation symptoms:
Sensory disturbances (tingling, electric shocks)
Discontinuation symptoms are more likely with paroxetine and venlafaxine (less so with their extended-release formulations), and less likely with fluoxetine.
No response to treatment
If the patient does not respond to treatment, consider these options:
Check the diagnosis (any bipolarity; missed comorbidity, such as substance use problems).
Optimize the antidepressant. Increase to the maximum tolerable dose within the dose range; ensure adherence for at least several weeks and manage side-effects.
Switch to another antidepressant. There is no difference whether switching within the same drug class (e.g., SSRI to another SSRI) or to a different class (e.g., SSRI to SNRI).
Add an augmenting agent. For example, add triiodothyronine (Cytomel; 25–50 mcg/day) or lithium (600–900 mg/day, or to therapeutic serum levels of 0.6–1.0 meq/L).
Alternatively, add an atypical antipsychotic (e.g., aripiprazole 2–10 mg/day; quetiapine-XR 150–300 mg/day; risperidone 0.5–3 mg/day; brexpiprazole 1–3 mg/day; note: olanzapine 2.5–10 mg/day is a second-line medication due to metabolic side-effects). Be careful about increased side-effect burden, especially with lithium and atypical antipsychotics.
Combine with another antidepressant in a different drug class. For example, add bupropion to an SSRI. Be careful about drug–drug interactions and increased side-effect burden.
Although there is little evidence to guide decisions when there is only partial response to treatment, most clinicians would augment or combine medications in order not to lose gains from the first antidepressant.
Switching between antidepressants
A washout period usually is not necessary when switching between most antidepressants, except to and from MAOIs and RIMAs (reversible monoamine oxidase inhibitors). Generally, the second antidepressant can be started at a low dose while tapering off the first antidepressant (the X approach).
Be careful about increased side-effect burden when starting a medication before stopping another. In some patients, for example those who appear sensitive to side-effects, taper off the first antidepressant before starting the second (the V approach).
A useful online tool that guides switching patients from one antipsychotic or antidepressant to another is SwitchRx. It provides tapering and titration schedules, clinical tips and detailed information on drug pharmacokinetics and other precautions (see Resources).
Avoiding drug–drug interactions
Be aware of the effects that combining certain medications can have:
Fluoxetine and paroxetine can markedly inhibit cytochrome P450 (CYP) isoenzyme 2D6 and can increase blood levels of medications that are metabolized primarily by that isoenzyme (caution with TCAs, beta-blockers, codeine [reduces effect]).
Fluvoxamine can markedly inhibit CYP 1A2 and can increase blood levels of medications that are metabolized primarily by that isoenzyme (caution with statin drugs, warfarin, quinidine, phenytoin, cyclosporine, sildenafil, vardenafil, clozapine, buspirone, diazepam).
Duloxetine is extensively metabolized by CYP 1A2, so avoid using it with potent inhibitors of CYP 1A2 (such as fluvoxamine and quinolones).
Vortioxetine is metabolized by many isoenzymes, but primarily by CYP 2D6, so avoid using it with potent inhibitors of CYP 2D6 (e.g., fluxoxetine and paroxetine), and reduce the dose with moderate inhibitors of CYP 2D6 (e.g., bupropion, duloxetine, sertraline).
MAOIs (and moclobemide) should not be combined with other antidepressants or with serotonergic medications such as meperidine (Demerol) due to the risk of potentially fatal interactions.
Use caution with atypical antipsychotics because they have more interactions.
Drug interaction mobile apps, such as RxTx from the Canadian Pharmacists Association, or Epocrates and Medscape, are helpful, but they usually make very conservative recommendations.