Tricyclic antidepressants (TCAs) were the dominant pharmacological treatment for depression for three decades before selective serotonin reuptake inhibitors (SSRIs) displaced them as first-line agents. Their displacement was driven not by inferior antidepressant efficacy but by a toxic-to-therapeutic ratio that makes them dangerous in overdose and poorly tolerated at antidepressant doses in many patients. Understanding TCA pharmacology remains essential because they retain genuine clinical utility in treatment-resistant depression (TRD), neuropathic pain, and migraine prophylaxis.
TCAs inhibit both the serotonin transporter (SERT) and the norepinephrine transporter (NET), producing dual monoaminergic enhancement analogous in principle to SNRIs. The antidepressant mechanism is driven by this combined reuptake inhibition, with downstream autoreceptor desensitization, upregulation of brain-derived neurotrophic factor (BDNF), and synaptic plasticity changes that parallel those seen with SSRIs and SNRIs.12 The critical pharmacological difference from modern SNRIs is not the primary mechanism but the breadth of off-target receptor binding: TCAs bind muscarinic acetylcholine receptors (mAChR), histamine H1 receptors, and alpha-1 adrenergic receptors with potencies that produce substantial adverse effects at therapeutically active concentrations, whereas SNRIs do not.
TCAs are classified as tertiary or secondary amines based on the number of methyl groups on the terminal nitrogen of the side chain, and this structural distinction has direct clinical consequences. Tertiary amines, including amitriptyline, imipramine, clomipramine, doxepin, and trimipramine, have greater SERT-to-NET selectivity and more potent muscarinic, histaminic, and alpha-1 antagonism than their secondary amine counterparts. Secondary amines, including nortriptyline, desipramine, and protriptyline, have relatively greater NET selectivity and substantially less anticholinergic and antihistaminic activity, producing a somewhat better-tolerated adverse effect profile.1 Tertiary amines are metabolized in part by demethylation to their secondary amine derivatives: amitriptyline demethylates to nortriptyline, and imipramine demethylates to desipramine, both of which are pharmacologically active. This metabolic relationship means that plasma measurements in patients on tertiary amines detect both parent and active metabolite, complicating therapeutic drug monitoring.
TCAs are well absorbed orally but undergo extensive first-pass hepatic metabolism, primarily by cytochrome P450 (CYP) 2D6, CYP1A2, and CYP3A4, with CYP2D6 being the principal isoform for most agents. Bioavailability after first-pass extraction typically ranges from 30% to 70%. Protein binding is high, typically 85% to 95%, mostly to albumin and alpha-1-acid glycoprotein (AAG). Volume of distribution is very large, in the range of 10 to 50 liters per kilogram, because TCAs partition extensively into lipid-rich tissues including the myocardium and central nervous system (CNS). This large volume of distribution has a critical safety implication: hemodialysis is not effective in removing TCAs during overdose because the fraction of total body drug in the plasma compartment is minuscule. Half-lives for most TCAs range from 12 to 36 hours with some agents extending beyond 48 hours, allowing once-daily dosing in most cases but prolonging toxicity management during overdose.2
CYP2D6 polymorphism has clinically important consequences for TCA dosing. Poor metabolizers (PMs) accumulate much higher plasma concentrations at any given dose, with toxicity risk elevated accordingly. Ultra-rapid metabolizers (UMs) may fail to achieve therapeutic concentrations at standard doses. Because CYP2D6 polymorphism affects approximately 7% to 10% of individuals of European ancestry, unexpected toxicity or treatment failure on a TCA warrants consideration of CYP2D6 genotyping. Additionally, potent CYP2D6 inhibitors including paroxetine, fluoxetine, and bupropion can double or triple TCA plasma levels when added to a stable TCA regimen, a pharmacokinetic interaction with potentially life-threatening consequences if not anticipated and monitored.
TCAs have well-established plasma concentration ranges associated with therapeutic response and toxicity risk. For nortriptyline, the therapeutic window is 50 to 150 ng/mL, with efficacy and toxicity both rising outside this range. For imipramine plus desipramine combined, the target is generally 150 to 300 ng/mL. Plasma level monitoring is recommended at steady state, after dose changes, when clinical response is unexpected, and whenever a CYP2D6-inhibiting drug is added. Levels above 500 ng/mL for most TCAs are associated with significant toxicity risk even in the absence of overdose.
The adverse effect burden of TCAs arises almost entirely from off-target receptor binding rather than from their primary mechanism of monoamine reuptake inhibition. Three receptor systems are responsible for the majority of clinical problems: muscarinic acetylcholine receptors (mAChR), histamine H1 receptors, and alpha-1 adrenergic receptors. The binding potency at each system varies across agents and is substantially higher in tertiary amines than in secondary amines.
Blockade of muscarinic acetylcholine receptors produces the classical anticholinergic syndrome: dry mouth, constipation, urinary retention, blurred vision from loss of accommodation, tachycardia, anhidrosis, and confusion or delirium in elderly patients. Among the TCAs, amitriptyline and clomipramine have the most potent muscarinic antagonism; desipramine has the least. Urinary retention is a particularly consequential adverse effect in elderly males with benign prostatic hyperplasia and can precipitate acute urinary obstruction requiring catheterization. Confusion and cognitive impairment from central muscarinic blockade are a major reason TCAs are classified as inappropriate for routine use in elderly patients by the Beers Criteria. Constipation can be severe enough to progress to paralytic ileus with sustained high-dose TCA use.3
Histamine H1 receptor antagonism by TCAs produces sedation and weight gain. Sedation is most pronounced with amitriptyline, doxepin, and trimipramine, all tertiary amines with high H1 binding affinity. This property is sometimes exploited therapeutically: low-dose doxepin (3 to 6 mg) is FDA-approved specifically for insomnia, where its H1 antagonism produces sedation at doses too low to produce meaningful monoamine reuptake inhibition. At antidepressant doses, however, H1-mediated sedation impairs daytime function and contributes substantially to patient-initiated discontinuation. Weight gain from H1 blockade is mediated through central mechanisms involving appetite regulation in the hypothalamus and can be substantial over the course of treatment.
Blockade of postsynaptic alpha-1 adrenergic receptors by TCAs produces orthostatic hypotension, the most dangerous cardiovascular adverse effect during chronic use. NE normally maintains vasoconstriction in peripheral resistance vessels via alpha-1 receptors; when these are blocked, the normal compensatory vasoconstriction that occurs when standing is impaired, producing a precipitous drop in blood pressure and reflex tachycardia. In elderly patients, this orthostatic hypotension can produce syncope and falls, with resultant hip fractures and head injuries. The risk is compounded by concurrent use of antihypertensive medications, volume depletion, or alcohol. Secondary amines, particularly nortriptyline, have somewhat less alpha-1 antagonism than tertiary amines and produce less orthostatic hypotension, which partly accounts for nortriptyline's relatively better cardiovascular tolerability profile among the TCAs.1
Among the TCAs, nortriptyline occupies a preferred clinical position for several reasons: it has the best-characterized therapeutic window (50 to 150 ng/mL) with established plasma level-response relationships; it has relatively lower muscarinic anticholinergic and alpha-1 antagonist activity compared to tertiary amines, producing less orthostatic hypotension and less cognitive impairment; and it has established efficacy in neuropathic pain, migraine prophylaxis, and treatment-resistant depression as an augmentation agent. When the clinical situation requires a TCA, nortriptyline is the rational first choice in most patients.
TCA overdose is one of the most dangerous presentations in clinical toxicology. A quantity of drug representing only a few days' supply at therapeutic doses can be lethal. The lethality arises from two converging mechanisms: sodium channel blockade in the myocardium producing life-threatening arrhythmias, and CNS toxicity producing seizures that further exacerbate cardiac instability. Management is time-critical and requires specific interventions not applicable to other antidepressant overdoses.
TCAs block fast sodium channels (Nav1.5) in cardiac myocytes, slowing phase 0 depolarization of the action potential. This produces a characteristic constellation of electrocardiogram (ECG) changes: prolonged PR interval, widened QRS complex, and a rightward shift of the terminal QRS axis producing a prominent R wave in lead aVR and S wave in lead I. QRS duration greater than 100 milliseconds is a sensitive indicator of significant toxicity and predicts seizure risk; QRS greater than 160 milliseconds is associated with high risk of ventricular arrhythmia including ventricular tachycardia and ventricular fibrillation. The R wave amplitude in lead aVR greater than 3 millimeters independently predicts seizures and arrhythmias and is a critical ECG finding in TCA overdose.4 TCAs also block cardiac potassium channels, producing QTc prolongation that contributes to arrhythmia risk via torsades de pointes (TdP).
Sodium bicarbonate is the cornerstone of TCA overdose management and works through two distinct mechanisms. First, alkalinization of blood to pH 7.45 to 7.55 reduces TCA binding affinity for the sodium channel, directly reversing channel blockade. Second, the increase in serum sodium concentration increases the electrochemical gradient driving sodium into myocytes during depolarization, partially overcoming the channel block. The clinical target is narrowing of the QRS complex toward baseline. Sodium bicarbonate should be administered as intravenous bolus doses of 1 to 2 mEq/kg whenever QRS exceeds 100 milliseconds or arrhythmia is present, with titration to pH and ECG response rather than a fixed total dose.5
TCAs produce CNS toxicity through multiple mechanisms: direct CNS depression via H1 and muscarinic receptor antagonism, GABA-A receptor inhibition that lowers the seizure threshold, and the cardiovascular compromise that reduces cerebral perfusion. Seizures in TCA overdose are typically brief and self-limiting but are hazardous because they produce acidosis (lowering blood pH), which worsens sodium channel blockade and can precipitate ventricular arrhythmia in an already compromised myocardium. Benzodiazepines are the first-line agents for TCA-associated seizures. Phenytoin and fosphenytoin should be avoided in TCA overdose: they have their own sodium channel-blocking properties that can worsen cardiac toxicity. Physostigmine, a reversible cholinesterase inhibitor that crosses the blood-brain barrier, can temporarily reverse TCA-induced anticholinergic delirium but carries a risk of precipitating bradycardia or asystole in the setting of cardiac sodium channel blockade and is generally avoided in overdose management.5
TCA overdose can progress rapidly. The clinical sequence typically follows: initial agitation and anticholinergic syndrome (dry skin, tachycardia, urinary retention, dilated pupils) followed by CNS depression and sedation, then QRS prolongation on the ECG, then seizures, then ventricular arrhythmia and cardiovascular collapse. The transition from mild symptoms to life-threatening arrhythmia can occur within one to two hours of ingestion. All patients with suspected TCA overdose require continuous cardiac monitoring, intravenous access, and evaluation in a setting capable of immediate defibrillation and endotracheal intubation. Gastric decontamination with activated charcoal may be appropriate within one hour of ingestion if the airway is protected, given the extensive enterohepatic recirculation of some TCAs.
Sodium bicarbonate for QRS >100 ms or arrhythmia — do not wait for further deterioration. Target blood pH 7.45 to 7.55. Benzodiazepines for seizures. Avoid phenytoin and fosphenytoin. Avoid physostigmine in the setting of cardiac toxicity. Lipid emulsion therapy has been reported in severe refractory cases but lacks robust evidence. Hemodialysis is not effective due to the very large volume of distribution. Maintain close monitoring for at least 24 hours even in apparently recovering patients, as deterioration can be delayed.
Monoamine oxidase inhibitors (MAOIs) produce antidepressant effects by blocking the primary intraneuronal degradation pathway for serotonin (5-HT), norepinephrine (NE), and dopamine (DA). MAO exists as two isoforms with distinct substrate preferences and tissue distributions. MAO-A preferentially deaminates 5-HT, NE, and DA and is found predominantly in noradrenergic and serotonergic neurons, the intestinal mucosa, and the liver. MAO-B preferentially deaminates DA and phenylethylamine (PEA) and is found predominantly in serotonergic neurons, platelets, and glial cells. Tyramine, which is the substrate responsible for the dietary interaction, is a substrate for both isoforms but is preferentially metabolized by MAO-A in the gut and liver during first-pass extraction.6
The classical MAOIs used for depression, phenelzine and tranylcypromine, are irreversible non-selective inhibitors of both MAO-A and MAO-B. They form covalent bonds with the flavin adenine dinucleotide (FAD) cofactor of MAO, permanently inactivating the enzyme. Recovery of MAO activity after discontinuing an irreversible MAOI depends entirely on synthesis of new enzyme, a process that takes approximately two weeks. This two-week recovery period is the pharmacological basis for the required two-week washout between an irreversible MAOI and any serotonergic agent or sympathomimetic: the MAOI effect persists long after the drug has been cleared from plasma, because the mechanism of action is covalent and enzyme-dependent rather than concentration-dependent.7
Selegiline, originally developed as a selective MAO-B inhibitor for Parkinson's disease at low doses (5 to 10 mg orally), loses its MAO-B selectivity at the higher doses required for antidepressant effect. The transdermal formulation of selegiline (Emsam) delivers drug systemically while substantially reducing first-pass gut and hepatic MAO-A inhibition, which is the basis for the reduced dietary tyramine restriction at the lowest transdermal dose (6 mg per 24 hours). At higher transdermal doses (9 and 12 mg per 24 hours), dietary tyramine restrictions are still required because systemic MAO-A inhibition is sufficient to impair tyramine metabolism at peripheral sympathetic terminals.9
Moclobemide is a reversible inhibitor of MAO-A (RIMA) used clinically in Europe, Canada, and Australia but not FDA-approved in the United States. Because it inhibits MAO-A reversibly, it can be displaced from the enzyme by high concentrations of tyramine, which competitively outcompetes moclobemide for the MAO-A binding site. This competitive displacement mechanism means that MAO-A activity is partially preserved during tyramine ingestion, substantially reducing the risk of the tyramine pressor response. Moclobemide requires much less stringent dietary restrictions than irreversible MAOIs, though avoidance of very large tyramine loads is still prudent. The washout period before serotonergic drugs is 24 hours rather than two weeks, reflecting the reversible mechanism.9
Irreversible MAOIs (phenelzine, tranylcypromine): require two weeks after stopping before initiating any serotonergic drug or indirect sympathomimetic. Two weeks is the time for MAO enzyme resynthesis to restore baseline activity. Selegiline transdermal: same two-week washout required when stopping before starting serotonergic drugs. Moclobemide (RIMA): 24-hour washout only, reflecting reversible binding and rapid functional recovery of MAO-A. When starting an irreversible MAOI after stopping a serotonergic drug, the washout is drug-dependent: two weeks for most SSRIs and SNRIs; five weeks for fluoxetine due to the long half-life of norfluoxetine.
The tyramine pressor response is the interaction that defines the clinical use of MAOIs. Understanding its mechanism precisely is necessary for appropriate patient counseling and for recognizing the situations in which MAOI use is justified despite the interaction risk.
Tyramine is a dietary amine formed by bacterial decarboxylation of tyrosine during fermentation and aging of foods. Under normal circumstances, tyramine ingested in food is almost completely metabolized by MAO-A in the intestinal mucosa and liver during first-pass extraction and never reaches systemic circulation in appreciable concentrations. When MAO-A is irreversibly inhibited, this first-pass extraction fails entirely, and dietary tyramine enters the systemic circulation intact. Tyramine is an indirect sympathomimetic: it is transported into adrenergic nerve terminals by the norepinephrine transporter (NET), where it enters vesicles and displaces NE stores into the synapse in massive quantities. The resulting massive NE release produces acute hypertension that can be sudden, severe, and potentially fatal. The hypertensive crisis from tyramine ingestion during MAOI treatment classically presents as a severe pounding headache, diaphoresis, flushing, nausea, vomiting, and markedly elevated blood pressure, sometimes exceeding 200/120 mmHg. Intracerebral hemorrhage is the most feared consequence.10
Foods with high tyramine content that must be avoided by patients on irreversible MAOIs include aged cheeses (the highest-risk single food category), cured and fermented meats, fermented fish and soy products, certain wines and draft beers, sauerkraut, and broad bean (fava bean) pods, which contain dopamine precursors that are themselves vasopressor substrates. Fresh cheeses (cottage cheese, ricotta, cream cheese) and fresh meats are generally safe. The tyramine content of foods varies with fermentation time and bacterial load rather than with the food category alone; the same food item can have widely varying tyramine content depending on production and storage conditions.10 Drug interactions that must be avoided include all serotonergic agents (risk of serotonin syndrome), indirect sympathomimetics including pseudoephedrine, phenylephrine, and amphetamines (risk of hypertensive crisis), meperidine (risk of a serotonin syndrome variant with hyperthermia and excitation), and dextromethorphan (serotonergic risk).
Despite their interaction burden, irreversible MAOIs retain a specific and important clinical niche. Multiple randomized controlled trials and a substantial body of clinical experience support the superiority of MAOIs over TCAs and over placebo in atypical depression, a subtype characterized by mood reactivity, hypersomnia, hyperphagia, leaden paralysis, and rejection sensitivity.8 Phenelzine has demonstrated superiority to imipramine in atypical depression in landmark trials, an effect that has been replicated.11 MAOIs are also among the most effective pharmacological agents for treatment-resistant depression in patients who have failed multiple adequate trials of other antidepressant classes. The reluctance to use MAOIs in routine practice reflects appropriate concern about the dietary and drug interaction risks rather than any deficiency in antidepressant efficacy. In patients with atypical depression or TRD who are reliable and capable of following dietary restrictions, MAOIs represent a genuine and underutilized therapeutic option.
Before initiating an MAOI, verify that the patient understands and can adhere to dietary restrictions, can carry a list of contraindicated medications including over-the-counter sympathomimetics in cold preparations, and can communicate MAOI status to all treating providers. Provide written dietary and drug interaction guidance. For emergency hypertensive crisis from tyramine ingestion, phentolamine (alpha-adrenergic blocker) or sublingual nifedipine have been used, though patients should be instructed to seek emergency care immediately. The selegiline transdermal patch reduces dietary restriction burden at the lowest dose and is a reasonable choice for patients who would benefit from an MAOI but for whom strict dietary adherence is uncertain.
Richelson E. Pharmacology of antidepressants. Mayo Clin Proc. 2001;76(5):511–527.
doi:10.4065/76.5.511Preskorn SH, Irwin HA. Toxicity of tricyclic antidepressants: kinetics, mechanism, intervention. J Clin Psychiatry. 1982;43(4):151–156.
American Geriatrics Society 2023 updated AGS Beers Criteria for potentially inappropriate medication use in older adults. J Am Geriatr Soc. 2023;71(7):2052–2081.
doi:10.1111/jgs.18372Liebelt EL, Francis PD, Woolf AD. ECG lead aVR versus QRS interval in predicting seizures and arrhythmias in acute tricyclic antidepressant toxicity. Ann Emerg Med. 1995;26(2):195–201.
doi:10.1016/S0196-0644(95)70151-6Kerr GW, McGuffie AC, Wilkie S. Tricyclic antidepressant overdose: a review. Emerg Med J. 2001;18(4):236–241.
doi:10.1136/emj.18.4.236Youdim MB, Edmondson D, Tipton KF. The therapeutic potential of monoamine oxidase inhibitors. Nat Rev Neurosci. 2006;7(4):295–309.
doi:10.1038/nrn1883Finberg JP, Rabey JM. Inhibitors of MAO-A and MAO-B in psychiatry and neurology. Front Pharmacol. 2016;7:340.
doi:10.3389/fphar.2016.00340Liebowitz MR, Quitkin FM, Stewart JW, et al. Antidepressant specificity in atypical depression. Arch Gen Psychiatry. 1988;45(2):129–137.
doi:10.1001/archpsyc.1988.01800260037004Stahl SM, Felker A. Monoamine oxidase inhibitors: a modern guide to an unrequited class of antidepressants. CNS Spectr. 2008;13(10):855–870.
doi:10.1017/S1092852900016965McCabe-Sellers BJ, Staggs CG, Bogle ML. Tyramine in foods and monoamine oxidase inhibitor drugs: a crossroads where medicine, nutrition, pharmacy, and food technology intersect. J Food Compost Anal. 2006;19(Suppl):S58–S65.
doi:10.1016/j.jfca.2005.12.008Henkel V, Mergl R, Allgaier AK, Kohnen R, Moller HJ, Hegerl U. Treatment of atypical depression: meta-analysis of clinical studies. Psychiatry Res. 2006;141(1):89–101.
doi:10.1016/j.psychres.2005.07.012Stahl SM. Stahl's Essential Psychopharmacology: Neuroscientific Basis and Practical Applications. 5th ed. Cambridge: Cambridge University Press; 2021. Chapters 11–12.