Prescribing Mirtazapine

Why Mirtazapine Matters

Mirtazapine is the antidepressant you reach for when the side-effect profile is the therapeutic plan. Its efficacy for depression is real and, by meta-analysis, sits near the top of the class: Cipriani's network analyses repeatedly rank it among the more efficacious agents, alongside escitalopram, sertraline, venlafaxine, and amitriptyline. But nobody prescribes mirtazapine because of a marginal efficacy edge. They prescribe it because it makes people sleep, makes people eat, and does neither of the two things patients hate most about SSRIs: it rarely causes sexual dysfunction and it rarely causes nausea.

That is the entire pitch, and it is a good one. The depressed patient who cannot fall asleep, has lost fifteen pounds, and lies awake with terminal insomnia is the patient mirtazapine was built for. Give it at bedtime, and within days, faster than the two-to-four-week antidepressant lag, the sleep and appetite improve. It is one of the few antidepressants with a fast symptomatic payoff, because the sleep and appetite effects are downstream of histamine blockade, not of any slow neuroplastic remodeling.

The flip side is equally simple, and it is the reason mirtazapine is a niche drug rather than a first-line one: it causes more weight gain than almost any other antidepressant, and its sedation can be a liability as easily as an asset. Prescribe it to a patient who is already overweight and sleeping twelve hours a day, and you have made things worse. The drug does not adapt to the patient; you adapt the patient to the drug.

A note on nomenclature: mirtazapine is the lone widely used member of the NaSSA class (Noradrenergic and Specific Serotonergic Antidepressant). It is often lumped with the "atypicals" (bupropion, trazodone, nefazodone, vortioxetine), but its mechanism is genuinely its own: it works by blocking receptors, not by inhibiting reuptake. Everything clinically distinctive about it flows from that fact.

Part 1: Indications: Who Is Mirtazapine For?

FDA-Approved Use

• Major depressive disorder (monotherapy)

That is the whole FDA label. Everything else is off-label, and the off-label uses are where the drug earns its keep.

The Sweet Spot: Depression With Insomnia and Appetite Loss

This is the single best use of mirtazapine, and it is worth being specific about the phenotype:

• Depression with prominent insomnia, especially terminal insomnia / early-morning awakening, the melancholic sleep signature.
• Depression with anorexia, poor appetite, or unintentional weight loss (the underweight, not-eating patient).
• Depression with anxious distress, particularly in older or medically frail patients where SSRI activation is unwelcome.

When these features cluster (the classic underweight, non-sleeping, anxious depressed patient), mirtazapine is often the right first choice, not a fallback. Its sleep-architecture benefit is objective: polysomnography data show it increases deep stage N3 slow-wave sleep and reduces nighttime awakenings (Karsten et al., 2017), which is more than you can say for trazodone at the low doses people actually use.

Other Off-Label Uses Worth Knowing

Antipsychotic-induced akathisia. This is a genuinely useful and underused indication. At least two RCTs and a 2024 network meta-analysis (Gerolymos et al., JAMA Netw Open) found low-dose mirtazapine (15 mg) among the most effective agents for antipsychotic-induced akathisia, arguably better than propranolol, which that analysis suggested is weaker than we'd assumed. The catch is dose-dependent and important: above 15–30 mg, mirtazapine can cause akathisia rather than treat it. Keep it at 15 mg for this indication. (Vitamin B6 / pyridoxine has a better tolerability-to-efficacy ratio and is a reasonable co-first-line for akathisia.)

SSRI-induced sexual dysfunction (add-on). Mirtazapine 15–45 mg at night can restore orgasm in patients anorgasmic on an SSRI, by antagonizing the 5-HT2A serotonergic fibers that drive the dysfunction. Evidence is open-label only, but the mechanism is sound and the risk is low.

SSRI-induced excessive sweating and SSRI-induced nausea. The same 5-HT2/5-HT3 blockade that keeps mirtazapine itself nausea-free can blunt these SSRI side effects when a low dose is added.

Comorbid anxiety in depression. Reasonable evidence for depression with anxious distress; no clear superiority over other agents, but the sedation helps.

Uses That Sound Good But Aren't: Read This Before You Prescribe

The mirtazapine literature is littered with attractive theories that large, well-designed trials have flattened. Do not fall for them.

Augmentation of a partially responsive antidepressant in treatment-resistant depression. This is the big one. "California Rocket Fuel" (venlafaxine plus mirtazapine) was oversold on the theory that adding a noradrenergic/serotonergic antagonist to an SNRI would be synergistic. The rigorous trials are negative:

• Kessler et al. (BMJ 2018, n=480): mirtazapine added to an SSRI/SNRI in primary-care non-responders gave 29% remission vs. 24% placebo (p=0.09), failed to separate, and the trivial advantage faded over the next nine months.
• CO-MED (Rush et al., 2011): venlafaxine plus mirtazapine gave 38% remission vs. escitalopram monotherapy 39%, no advantage to the combination whatsoever.
• Navarro et al. (2019): in venlafaxine non-responders, mirtazapine augmentation gave 39% remission while switching to imipramine gave 72%, switching nearly doubled the response.
• Henssler meta-analysis (JAMA Psychiatry 2022): a modest pooled SMD of ~0.37, but essentially all the large mirtazapine-combination RCTs were negative.

The lesson: for a patient who has partially responded to an SSRI/SNRI, switching, or augmenting with something with real evidence (an atypical antipsychotic, lithium, T3), beats adding mirtazapine. The one legitimate exception is the patient who also has insomnia and weight loss, where you're adding mirtazapine for its symptomatic targets, not for a synergistic antidepressant effect.

Combat-related PTSD monotherapy. Despite a mechanistically appealing story (dampening noradrenergic arousal, improving sleep), the Davis et al. trial (2020, n=78 veterans, mean 39 mg/day) failed its primary PTSD outcome, and, surprisingly for a sedating drug, it did not improve sleep and appeared to increase nightmares. Antidepressant effects do not transfer across diagnoses. Mirtazapine may still help SSRI-treated PTSD patients with residual sleep disruption, but do not use it as PTSD monotherapy.

OCD. Do not expect mirtazapine to treat OCD, and be cautious: as a potent 5-HT2A antagonist, it can theoretically provoke or worsen obsessive-compulsive symptoms. (There is one small positive augmentation-to-sertraline trial, but the signal is weak and the mechanistic concern is real.)

First-line in high-neuroticism patients. In head-to-head data, high-neuroticism patients did worse on mirtazapine (36% remission at 8 weeks) than on an SSRI (74%). If the patient's temperament is anxious-ruminative rather than melancholic-insomniac, an SSRI is the better bet.

Part 2: Before You Start: Workup and Candidacy

Mirtazapine is refreshingly light on required workup. There is no therapeutic level to chase, no narrow window, no mandatory baseline panel.

What to Actually Do Before Starting

No CBC is required at baseline despite the agranulocytosis language in the label (see Part 6). No routine ECG is required. No serum drug levels exist to monitor, this is not a tricyclic.

The Ideal Candidate

• Depressed and not sleeping and not eating.
• Underweight, or at least not weight-preoccupied.
• Bothered by (or unwilling to risk) SSRI sexual dysfunction: mirtazapine is a favorite in correctional and male-heavy populations for exactly this reason.
• Prone to SSRI/SNRI nausea.
• Elderly with insomnia and poor appetite (a classic geriatric fit, with fall-risk caveats below).

The Poor Candidate

• Overweight or weight-preoccupied patients, or anyone with metabolic syndrome you're trying not to worsen.
• Patients who are already hypersomnic or who need to be alert and functional during the day.
• High-neuroticism, anxious-ruminative depression without prominent insomnia (an SSRI is better).
• A bipolar patient without a mood stabilizer.

Part 3: How to Start and Dose

Formulations

Mirtazapine comes as standard oral tablets and an orally disintegrating tablet (ODT, "SolTab") in 15, 30, and 45 mg. The standard tablet also comes in a 7.5 mg strength. The ODT is convenient for patients who won't reliably swallow a pill, but it is not absorbed sublingually: it dissolves and is swallowed, so the pharmacokinetics are the same as the tablet.

The Paradoxical Dosing Rule: the Thing Everyone Gets Wrong

Here is the single most counterintuitive fact about mirtazapine, and it drives most dosing errors.

The mechanism is elegant. At low doses, potent H1 antihistamine blockade dominates: that's the sedation and the appetite. As the dose climbs, noradrenergic tone (from alpha-2 autoreceptor blockade releasing norepinephrine) increasingly offsets the antihistamine sedation. So a patient groggy on 7.5 mg may be less groggy on 30 mg. This is the opposite of every intuition patients (and many prescribers) bring to dose titration, and it must be explained up front.

Starting Dose and Titration

• Start: 15 mg at bedtime (some clinicians start 7.5 mg; 15 mg is the standard and, per the biphasic curve, often no more sedating than 7.5).
• Titrate by 15 mg every 1–2 weeks as needed for antidepressant effect, guided by response and tolerability.
• Therapeutic range: 15–45 mg/day, all at bedtime.
• Maximum: 45 mg/day.

The Dose-Response Ceiling: Don't Chase 45 mg

The efficacy data (Furukawa meta-analysis, Lancet Psychiatry 2019) are clear and clinically useful: antidepressant efficacy increases up to about 30 mg/day and then plateaus or even declines above 30 mg, while adverse-effect dropouts rise sharply.

The practical target for depression is 15–30 mg. Going to 45 mg buys you little additional antidepressant benefit and costs you tolerability. Reserve 45 mg for patients who have clearly tolerated 30 mg and plausibly need more, not as a routine destination.

Dose by the Goal

Timing

Dose at bedtime, always. If daytime grogginess (a "hangover") is a problem, moving the dose to early evening can help by shifting the peak earlier, but the counterintuitive alternative is to raise the dose slightly, letting the noradrenergic offset kick in.

Onset

Unlike the classic two-to-four-week antidepressant lag, the sleep and appetite effects appear within days. The full mood effect still takes the usual 4–8 weeks; reassess response at that point. Meta-analyses credit mirtazapine with a somewhat faster onset of antidepressant action than citalopram, fluoxetine, paroxetine, or sertraline.

Part 4: Monitoring

Mirtazapine's monitoring burden is minimal, the opposite of lithium. There are no levels, no mandatory labs, and no ECG requirement.

If there's no meaningful response by 4–8 weeks at an adequate dose (15–30 mg), switch agents rather than push higher or augment.

Part 5: Side Effects and How to Manage Them

The governing principle: mirtazapine's side effects are dominated by two things: sedation and weight gain, both driven by H1 antihistamine blockade. Nearly everything else about its tolerability is a plus. The management art is deciding whether the sedation and appetite are features or bugs for your particular patient.

Weight Gain and Increased Appetite: the Defining Liability

This is the reason mirtazapine isn't first-line for most depression. It is, by reputation and by data, among the worst antidepressants for weight gain, as one Carlat expert put it bluntly, "probably the worst." Reported figures vary widely across the noisy literature (some sources cite very high proportions gaining clinically significant weight), but the honest summary is: a large share of patients gain weight, often on the order of several kilograms over months, and it is frequently the reason they quit. Carbohydrate craving is a specific and common complaint.

• Mechanism: H1 antagonism (with a contribution from 5-HT2C blockade) drives appetite.
• Set expectations up front: tell every patient this will likely increase appetite and may add weight, and that you'll be tracking it.
• Dietary and exercise counseling from day one, not after ten pounds are on.
• Lower doses may attenuate it, though the appetite effect is present even at low doses.
• Monitor weight every visit.
• If weight gain is intolerable, switch: to bupropion (weight-neutral to weight-reducing), or to trazodone/nefazodone if you need a sedating agent with less weight liability.

Sedation and Drowsiness

• Prominent at low doses, may diminish as the dose rises (the biphasic curve).
• Dose at bedtime: make the sedation work for you.
• If daytime grogginess persists, move the dose earlier in the evening, or paradoxically increase the dose (30+ mg) to bring in the noradrenergic offset.
• Consider a lower dose (7.5 mg) if sedation is excessive and you only need the sleep/appetite effect.
• As a last resort for otherwise-good responders, an activating agent (a stimulant or modafinil-class drug) can be layered, but reconsider the drug choice first.
• In the elderly, sedation translates directly into fall risk (see Special Populations).

Sexual Function: a Major Advantage

Mirtazapine has low rates of sexual dysfunction, one of the best profiles in the class, because it lacks the serotonin-reuptake blockade that drives SSRI/SNRI dysfunction and because its 5-HT2A/5-HT2C antagonism actively counteracts it. This is a primary reason to choose it, and a reason it's used to rescue SSRI-induced dysfunction.

GI: Another Advantage

Mirtazapine is essentially nausea-neutral. In a 15-antidepressant meta-analysis (Oliva et al., 2021), it was the only agent not associated with nausea/vomiting, courtesy of H1 antagonism plus 5-HT3 blockade (the same mechanism ondansetron uses). It is also not a diarrhea offender. If a patient can't tolerate SSRI/SNRI GI effects, this is a natural landing spot.

Anticholinergic Effects (Mild)

Mild dry mouth and constipation can occur.

• Dry mouth: sugar-free gum/lozenges, saliva substitutes; bethanechol 25 mg BID in stubborn cases.
• Constipation: fiber, fluids, docusate, polyethylene glycol; avoid chronic stimulant laxatives.

Orthostatic Hypotension / Dizziness

Possible, especially on standing quickly and especially in the elderly. Bedtime dosing helps; counsel slow position changes; monitor in older patients.

Vivid Dreams / Nightmares

Can occur; notably prominent (and problematic) in the PTSD trial data. If distressing, this is a reason to reconsider the drug in that patient.

Rare but Worth Knowing

• Elevated lipids / glucose dysregulation: monitor in metabolically vulnerable patients.
• Mania/hypomania induction in bipolar-spectrum patients.
• Agranulocytosis (see Part 6; extremely rare).

Part 6: Overdose, Toxicity, and Boxed Warnings

Overdose: One of the Safer Antidepressants

This is a genuine advantage, especially in a population with suicide risk. Mirtazapine is markedly safer in overdose than tricyclics. Overdose typically produces sedation, tachycardia, and mild disorientation; it is not associated with the lethal cardiac conduction toxicity of TCAs and is not a significant lethal-overdose agent on its own. Management is supportive care. As always, danger rises with co-ingestants (alcohol, benzodiazepines, opioids, additive CNS depression).

The Boxed Warnings

Agranulocytosis is also a labeled serious warning. Premarketing trials recorded a small number of cases. In practice this is extremely rare, unpredictable, and does not warrant routine CBC monitoring. The prescribing information does not recommend scheduled blood counts. The correct posture: check a CBC only if the patient develops signs of infection (fever, sore throat, stomatitis, or other flu-like illness), and discontinue if agranulocytosis is confirmed.

QTc

Mirtazapine appears on lists of QTc-prolonging drugs, but the real-world signal is small. In a hospital study of adding a second QTc-active drug, only ~2% developed a prolonged QTc and there were no cases of torsades or sudden death. No routine ECG is required; use ordinary caution when stacking multiple QTc-active agents or in patients with baseline prolongation.

Seizure

Mirtazapine sits in the middle-to-low range for seizure risk, lower than clomipramine, amitriptyline, venlafaxine, and several SSRIs, though higher than escitalopram and fluoxetine. It is not contraindicated in seizure history and carries no bupropion-like seizure warning.

Part 7: Drug Interactions

Mirtazapine's interaction profile is clean, which is a real selling point in polypharmacy and in the medically complex or elderly patient.

Pharmacokinetic Interactions: Minimal

• Metabolized by CYP2D6, CYP3A4, and CYP1A2, but it is not a meaningful inhibitor or inducer of these enzymes. It does not raise the levels of other drugs the way fluoxetine, paroxetine, or fluvoxamine can.
• Strong CYP3A4/2D6 inhibitors or inducers can modestly shift mirtazapine levels, but this rarely requires action given the wide safety margin.
• With warfarin: monitor INR; a minor interaction is possible.

Pharmacodynamic Interactions: the Ones That Matter

• CNS depressants (alcohol, benzodiazepines, opioids, antihistamines, Z-drugs): additive sedation. This is the practically important interaction: counsel patients, and take special care in the elderly and in those on opioids.
• MAOIs: contraindicated within 14 days (serotonin-syndrome risk). Notably, some experts consider mirtazapine one of the safer agents to combine with an MAOI when this is done deliberately and under close supervision, but standard practice is to observe the 14-day washout.
• Other serotonergic agents (SSRIs/SNRIs): additive serotonergic effect is theoretically possible but not clinically significant in practice; serotonin syndrome from mirtazapine combinations is rare.
• No tyramine interaction (unlike MAOIs).
• Low bleeding risk: unlike SSRIs/SNRIs, mirtazapine does not meaningfully impair platelet aggregation, an advantage in patients on anticoagulants/antiplatelets or NSAIDs.

Part 8: Special Populations

Pregnancy

Data are limited. Historically FDA Pregnancy Category C; no consistent signal of major teratogenicity has emerged, but the evidence base is thin. Some clinicians reach for it in pregnancy specifically when a depressed patient also has nausea/hyperemesis and poor intake with insomnia, leveraging its antiemetic and appetite effects. Decisions should be individualized, weighing the substantial risk of untreated maternal depression; coordinate with obstetrics.

Lactation

Mirtazapine passes into breast milk in low amounts; infant exposure is generally low but data are limited. Discuss risks and benefits; if used, watch the infant for sedation and poor feeding. Better-characterized agents (e.g., sertraline) are often preferred if the clinical picture allows.

Elderly

Mirtazapine is a strong option in late-life depression with insomnia and poor appetite/weight loss, a common and hard-to-treat geriatric presentation where SSRIs may worsen appetite and sleep.

• Start low, go slow: 7.5–15 mg QHS; titrate by 7.5–15 mg every 1–2 weeks.
• Target 15–30 mg; the full 45 mg is rarely needed.
• Watch closely for: falls (sedation plus orthostasis), cognitive dulling from oversedation, and hyponatremia (SIADH, a class effect of antidepressants in the elderly).
• Advantages here: no sexual-dysfunction concern, fixes appetite and sleep, and a low interaction burden in the polypharmacy patient.

Hepatic Impairment

Mirtazapine is hepatically metabolized; clearance is reduced in hepatic impairment. Reduce the dose and titrate cautiously, watching for exaggerated sedation.

Renal Impairment

Clearance is also reduced in significant renal impairment. Mild-to-moderate impairment usually tolerates standard dosing with monitoring; in severe renal disease, reduce the dose and go slowly.

Children and Adolescents

Limited data; not FDA-approved for pediatric depression (only fluoxetine and escitalopram are). Use is off-label and non-standard, reserved for select cases (e.g., depression with severe insomnia) after first-line SSRIs. If used, start 7.5 mg and titrate carefully. The pediatric suicidality boxed warning applies with full force.

Bipolar Disorder

Can precipitate mania/hypomania like any antidepressant. Do not use as monotherapy in bipolar depression: pair it with a mood stabilizer or choose a different strategy.

Part 9: Discontinuation

Here mirtazapine shines: it is one of the cleanest antidepressants to stop.

• Discontinuation syndrome is minimal to none, markedly gentler than the short-half-life serotonergic agents (paroxetine, venlafaxine, duloxetine).
• The ~20–40 hour half-life (mean ~26 hours) provides a modest built-in self-taper.
• A gradual taper over 1–2 weeks is reasonable for comfort and to watch for relapse, but abrupt discontinuation rarely produces the flu-like, dizzy, "brain-zap" withdrawal that plagues SSRI/SNRI cessation.

As with all antidepressants, discontinuation of successful treatment still risks relapse of the underlying depression: a maintenance trial found higher relapse on discontinuation (~56%) than on continuation (~39%) over a year. The taper protects against withdrawal; it does not protect against recurrence, which is a separate clinical decision about how long to maintain treatment.

Part 10: Mirtazapine vs the Alternatives

Since efficacy across modern antidepressants is broadly similar (~60% response), the choice is a side-effect and symptom-target decision. Mirtazapine's position:

Where mirtazapine uniquely or best delivers: the fast fix for depression with insomnia and appetite loss; the lowest sexual-dysfunction and nausea profiles in practical use; a clean drug-interaction sheet; an easy discontinuation; and reliable relief of antipsychotic-induced akathisia at 15 mg.

Where it clearly loses: anyone weight-sensitive, anyone who needs daytime alertness, high-neuroticism anxious depression, TRD augmentation (the "Rocket Fuel" myth), PTSD monotherapy, and OCD.

Mechanism: The Short Version

Mirtazapine is a NaSSA: it works by blocking receptors, not by inhibiting monoamine reuptake, which is what makes it distinctive.

• Alpha-2 adrenergic autoreceptor/heteroreceptor antagonism is the antidepressant core: blocking these presynaptic "brakes" disinhibits release of both norepinephrine and serotonin. More transmitter is released, rather than reuptake being blocked.
• 5-HT2A and 5-HT2C antagonism: routes the extra serotonin preferentially onto 5-HT1A receptors, and accounts for the low sexual dysfunction, improved slow-wave sleep, and appetite/weight effects.
• 5-HT3 antagonism: the built-in antiemetic effect (why it doesn't cause nausea).
• Potent H1 antihistamine antagonism: the sedation and appetite stimulation, dominant at low doses, progressively offset by rising noradrenergic tone at higher doses (the biphasic dose-sedation curve).

The half-life is ~20–40 hours (mean ~26 h), comfortably supporting once-daily bedtime dosing. Metabolism is hepatic via CYP2D6, 3A4, and 1A2, with negligible enzyme inhibition.

Everything clinically important about mirtazapine (the sleep, the appetite, the weight, the clean GI and sexual profile, the paradoxical dosing) falls directly out of this receptor map.

The Bedside Cheat Sheet