Selective serotonin reuptake inhibitors (SSRIs) share a single primary mechanism: inhibition of the serotonin transporter (SERT), the plasma membrane protein responsible for reuptake of serotonin (5-hydroxytryptamine, 5-HT) from the synaptic cleft into the presynaptic neuron. Despite this shared target, clinically meaningful differences in pharmacokinetics, receptor binding, and drug interaction profiles distinguish the six agents in this class from one another in ways that directly inform prescribing decisions.
SERT is a sodium-dependent monoamine transporter encoded by the SLC6A4 gene. Under physiological conditions, SERT terminates serotonergic neurotransmission by co-transporting 5-HT, sodium, and chloride from the synaptic cleft into the presynaptic terminal, where 5-HT is either repackaged into vesicles by the vesicular monoamine transporter 2 (VMAT2) or degraded by monoamine oxidase (MAO). SSRIs bind to a recognition site on SERT that overlaps with the 5-HT binding site, competitively and non-competitively blocking transport depending on the agent.1 Acute SERT blockade elevates synaptic 5-HT within hours of the first dose, which is pharmacologically rapid but therapeutically slow, because the clinical antidepressant effect consistently requires two to four weeks of continuous treatment.
The discrepancy between the speed of SERT blockade and the delay to clinical response is explained by the somatodendritic autoreceptor mechanism described in Module 01. When acute SERT blockade raises 5-HT near serotonergic cell bodies in the dorsal raphe nucleus, 5-HT1A autoreceptors on those cell bodies are activated, suppressing serotonergic neuron firing and partially counteracting the intended increase in synaptic 5-HT output. With sustained SSRI exposure over two to four weeks, these inhibitory autoreceptors desensitize and downregulate, removing the negative feedback brake and allowing net 5-HT output into terminal synaptic fields to rise substantially.2 This autoreceptor desensitization timeline maps onto the clinical onset of antidepressant response and explains why dose escalation in the first two weeks does not accelerate the therapeutic effect.
At the postsynaptic level, chronic 5-HT elevation leads to downregulation of 5-HT2A receptors in prefrontal and limbic circuits over a similar timescale. Simultaneously, sustained SSRI treatment upregulates brain-derived neurotrophic factor (BDNF) expression and activates the tropomyosin receptor kinase B (TrkB) signaling pathway, promoting synaptic plasticity and hippocampal neurogenesis.3 These downstream adaptations, not the acute SERT blockade itself, are the primary correlates of therapeutic response. The practical consequence is that an adequate antidepressant trial requires four to six weeks at a therapeutic dose before a treatment failure determination is appropriate.
SSRIs are selective for SERT over the norepinephrine transporter (NET) and dopamine transporter (DAT), but selectivity is relative, not absolute. At high doses or in patients with reduced drug clearance, meaningful NET inhibition can occur, particularly with sertraline. Paroxetine additionally has significant muscarinic acetylcholine receptor (mAChR) antagonism at therapeutic doses, distinguishing it from the rest of the class and contributing to its anticholinergic adverse effect profile. "Selective" refers to the primary pharmacological target, not to pharmacological purity.
SSRIs are approved across a wide range of indications beyond major depressive disorder (MDD). All six agents have FDA approval for at least one anxiety disorder, and the class as a whole is considered first-line pharmacotherapy for generalized anxiety disorder (GAD), panic disorder, social anxiety disorder (SAD), post-traumatic stress disorder (PTSD), and obsessive-compulsive disorder (OCD).4 Fluoxetine and sertraline are approved for premenstrual dysphoric disorder (PMDD). Fluoxetine in combination with olanzapine is approved for treatment-resistant depression (TRD) and bipolar depression. Fluvoxamine is approved for OCD and used off-label for social anxiety disorder. The breadth of indication makes SSRIs among the most widely prescribed drug classes in medicine.12
All six SSRIs are well absorbed orally, highly protein-bound, extensively distributed into tissues, and hepatically metabolized. The clinically significant differences lie in half-life, active metabolites, cytochrome P450 (CYP) inhibition profiles, and specific receptor binding outside the primary SERT target. These differences are not minor pharmacokinetic details; they determine which agent is appropriate for a given patient, how interactions are managed, and what happens when treatment is stopped abruptly.
Fluoxetine is unique among SSRIs for its exceptionally long effective half-life. The parent compound has a half-life of one to four days, and its active metabolite norfluoxetine has a half-life of seven to fifteen days, giving the combination an effective duration that extends weeks beyond the last dose.5 This has several clinical consequences. First, fluoxetine provides an intrinsic self-tapering effect, making abrupt discontinuation far less likely to produce discontinuation syndrome compared to shorter-acting SSRIs. Second, a five-week washout period is required before initiating an irreversible monoamine oxidase inhibitor (MAOI), the longest washout of any SSRI. Third, CYP2D6 inhibition by fluoxetine and norfluoxetine persists for weeks after the last dose, meaning drug interactions outlast apparent discontinuation. Fluoxetine is a potent inhibitor of CYP2D6 and a moderate inhibitor of CYP3A4 and CYP2C19. It is primarily metabolized by CYP2D6 and CYP2C9. Oral bioavailability is approximately 72%, and it achieves steady-state plasma concentrations in four to six weeks given its long half-life. It is approved for MDD, OCD, panic disorder, bulimia nervosa, PMDD, and bipolar depression (in combination with olanzapine).
Sertraline is widely regarded as having one of the most favorable overall pharmacokinetic profiles in the SSRI class due to its moderate half-life (approximately 26 hours for the parent compound), minimal active metabolite activity, and relatively weak CYP inhibition compared to fluoxetine and paroxetine. It is metabolized primarily by CYP2C19 and CYP2D6 to desmethylsertraline, which has weak pharmacological activity and does not contribute meaningfully to clinical effect. Oral bioavailability is approximately 44% and is modestly increased when taken with food, an effect that is clinically useful in patients who experience gastrointestinal (GI) adverse effects on an empty stomach.5 Sertraline is a weak inhibitor of CYP2D6 at standard therapeutic doses, though inhibition becomes more clinically significant at higher doses. Protein binding is approximately 98%. It is approved for MDD, OCD, panic disorder, PTSD, SAD, and PMDD, giving it the broadest FDA approval profile in the class.
Paroxetine is pharmacokinetically the most complex SSRI and the one requiring the most clinical caution. Its half-life averages 21 hours but varies substantially because paroxetine is both a substrate and a potent inhibitor of CYP2D6, meaning it inhibits its own metabolism at therapeutic doses, producing nonlinear pharmacokinetics: small dose increases produce disproportionately large plasma concentration increases.5 Paroxetine is the most potent CYP2D6 inhibitor of all SSRIs, with clinically significant effects on co-administered CYP2D6 substrates including tricyclic antidepressants (TCAs), antipsychotics, opioids, and tamoxifen. Beyond SERT, paroxetine has meaningful antagonism at muscarinic M1 receptors, producing an anticholinergic adverse effect profile (dry mouth, constipation, urinary hesitancy, cognitive effects) not seen to the same degree with other SSRIs. It also has modest inhibition of NET. Paroxetine carries the highest discontinuation syndrome risk of all SSRIs given its short effective half-life and lack of active metabolites; abrupt cessation frequently produces the FINISH syndrome (Flu-like symptoms, Insomnia, Nausea, Imbalance, Sensory disturbances, Hyperarousal). A controlled-release formulation reduces peak concentration variability but does not alter the fundamental discontinuation risk. Protein binding is approximately 95%.
Citalopram is a racemic mixture of R- and S-enantiomers. The S-enantiomer (escitalopram) is responsible for essentially all SERT inhibitory activity; the R-enantiomer contributes minimally to antidepressant effect but adds to QTc prolongation risk and adverse effect burden. Citalopram has a half-life of approximately 35 hours, is metabolized primarily by CYP2C19 and CYP3A4, and is a weak inhibitor of CYP2D6.5 Its SERT selectivity is high and it has minimal activity at other receptors, which contributes to its relatively clean adverse effect profile outside of QTc concerns. The FDA issued a dose limitation in 2011 capping citalopram at 40 mg per day in most patients (20 mg per day in patients over 60 years, those with hepatic impairment, or those who are CYP2C19 poor metabolizers) based on QTc prolongation data, which is discussed fully in Section 04. Protein binding is approximately 80%.
Escitalopram is the S-enantiomer of citalopram, isolated and approved separately. It has approximately twice the SERT affinity of the racemic mixture on a milligram basis, allowing effective dosing at 10 to 20 mg per day compared to citalopram's 20 to 40 mg range. Its half-life is approximately 27 to 32 hours, metabolism is primarily via CYP2C19 and CYP3A4, and CYP inhibition is minimal.13 Escitalopram retains QTc prolongation risk at higher doses, with an FDA label warning analogous to citalopram, though the dose ceiling (20 mg per day) allows a somewhat greater therapeutic range before reaching the concentration associated with QTc concern. Its overall tolerability and minimal drug interaction profile make it among the most commonly chosen SSRIs for patients on polypharmacy regimens. Protein binding is approximately 56%, the lowest in the class, making it less susceptible to protein-binding displacement interactions.
Fluvoxamine is pharmacokinetically distinctive among SSRIs for its potent inhibition of CYP1A2 and CYP2C19, and moderate inhibition of CYP3A4, producing the broadest and most clinically consequential CYP inhibition profile of the class. Its half-life is approximately 15 to 20 hours, it lacks active metabolites, and it is metabolized primarily by CYP2D6 and CYP1A2.5 The CYP1A2 inhibition is the most clinically significant consequence of fluvoxamine use: it can raise plasma concentrations of clozapine five- to tenfold, of olanzapine two- to threefold, of theophylline substantially, and of caffeine. Fluvoxamine also inhibits CYP2C19, raising levels of omeprazole, diazepam, and phenytoin. In clinical practice, fluvoxamine is used primarily for OCD and SAD rather than MDD, and its drug interaction burden limits its use in patients on complex medication regimens. Its sigma-1 receptor agonism, a property not shared by other SSRIs, has generated interest in repurposing for other indications but is not established as a primary mechanism for its antidepressant or anxiolytic effects.
Half-life and discontinuation risk: paroxetine (highest risk) vs. fluoxetine (lowest risk). CYP2D6 inhibition: paroxetine and fluoxetine (potent) vs. sertraline (weak at standard doses) vs. escitalopram and citalopram (minimal). CYP1A2 inhibition: fluvoxamine only. Active metabolite with extended duration: fluoxetine only (norfluoxetine), mandating the five-week MAOI washout.
The most clinically important drug interactions involving SSRIs arise not from pharmacodynamic competition but from pharmacokinetic inhibition of CYP enzymes. Several SSRIs are potent inhibitors of specific CYP isoforms, meaning they reduce the metabolic clearance of co-administered drugs that depend on those isoforms, raising plasma concentrations and increasing the risk of toxicity from those drugs. The interaction severity depends on the degree of CYP inhibition, the degree to which the victim drug depends on that isoform, and the therapeutic index of the victim drug. Low-therapeutic-index substrates of inhibited isoforms represent the highest-risk combinations.
Paroxetine is the most potent CYP2D6 inhibitor among SSRIs, effectively converting CYP2D6 extensive metabolizers to a phenotypic poor metabolizer state while treatment continues. Fluoxetine and its active metabolite norfluoxetine are also potent CYP2D6 inhibitors, with the additional complication that inhibition persists for several weeks after fluoxetine is discontinued given the long norfluoxetine half-life. The most clinically consequential CYP2D6 interactions for both agents involve TCAs (desipramine, nortriptyline, amitriptyline, imipramine), where CYP2D6 inhibition can double or triple TCA plasma levels, precipitating TCA toxicity including QRS prolongation and arrhythmia; antipsychotics metabolized by CYP2D6 (haloperidol, risperidone, aripiprazole); opioid analgesics including codeine (CYP2D6 converts codeine to active morphine, so inhibition reduces analgesic efficacy) and tramadol; and tamoxifen (CYP2D6 converts tamoxifen to active endoxifen; inhibition reduces anticancer efficacy).6 For patients requiring tamoxifen for breast cancer, use of paroxetine or fluoxetine is generally contraindicated; alternative SSRIs with minimal CYP2D6 inhibition, such as citalopram, escitalopram, or sertraline at lower doses, are preferred.
Fluvoxamine is a potent CYP1A2 inhibitor and the only SSRI with clinically consequential inhibition of this isoform. The most dangerous single interaction is with clozapine: fluvoxamine can raise clozapine plasma concentrations five- to tenfold, producing clozapine toxicity including severe sedation, orthostatic hypotension, and an increased risk of seizures and agranulocytosis.6 If fluvoxamine must be used in a patient on clozapine, clozapine dose should be reduced by approximately 75% and plasma concentrations monitored closely. The combination is generally avoided. Additional CYP1A2 substrates requiring monitoring include theophylline (narrow therapeutic index; fluvoxamine can precipitate theophylline toxicity), olanzapine (two- to threefold level increase), duloxetine, and tizanidine (the combination with fluvoxamine is contraindicated due to severe hypotension and bradycardia). Smoking induces CYP1A2 and reduces fluvoxamine exposure; dose reassessment is warranted at any change in smoking status.
Both fluvoxamine (potent) and fluoxetine (moderate) inhibit CYP2C19. Clinically significant substrates include clopidogrel (a prodrug requiring CYP2C19 activation to its active thienopyridine metabolite; inhibition reduces antiplatelet efficacy), omeprazole and other proton pump inhibitors (elevated levels, generally not dangerous given wide therapeutic index), diazepam and other benzodiazepines metabolized by CYP2C19 (elevated sedation), phenytoin (narrow therapeutic index; monitor levels), and voriconazole. The clopidogrel interaction is the most clinically consequential: patients on clopidogrel for coronary stent protection should generally not receive fluvoxamine or high-dose fluoxetine, and if an SSRI is needed, escitalopram or sertraline at standard doses are preferable given minimal CYP2C19 inhibition.
Beyond CYP-mediated pharmacokinetic interactions, all SSRIs carry a pharmacodynamic interaction risk when combined with other serotonergic agents. This risk is addressed fully in Section 05 in the context of serotonin syndrome. The most hazardous pharmacodynamic interactions involve irreversible MAOIs (phenelzine, tranylcypromine, isocarboxazid), where combined use can produce life-threatening serotonin syndrome. The standard washout before initiating an MAOI after stopping an SSRI is two weeks for most SSRIs, extended to five weeks for fluoxetine given the norfluoxetine half-life. Conversely, after stopping an MAOI, at least two weeks must elapse before any SSRI is initiated.
Tamoxifen is a prodrug converted to its active metabolite endoxifen by CYP2D6. Paroxetine is a potent CYP2D6 inhibitor. Co-administration reduces endoxifen concentrations by up to 65%, potentially reducing the breast cancer survival benefit of tamoxifen. One retrospective cohort study estimated this interaction was associated with an increased breast cancer mortality rate.11 For patients requiring adjuvant tamoxifen, SSRIs with minimal CYP2D6 inhibition (citalopram, escitalopram, venlafaxine) are preferred. Paroxetine and fluoxetine should be avoided.
The QT interval represents the total duration of ventricular depolarization and repolarization on the electrocardiogram (ECG). Corrected for heart rate (QTc), a prolonged QTc indicates delayed ventricular repolarization, which increases the risk of early afterdepolarizations and the potentially fatal arrhythmia torsades de pointes (TdP). QTc prolongation is a recognized adverse effect of citalopram and, to a lesser degree, escitalopram, attributable to dose-dependent blockade of the cardiac hERG (human ether-a-go-go-related gene) potassium channel, which is responsible for the rapid component of the delayed rectifier potassium current (IKr) and is critical for normal ventricular repolarization.7
The pharmacological basis for citalopram's greater QTc effect compared to escitalopram lies in its enantiomeric composition. The R-enantiomer of citalopram has greater hERG channel affinity than the S-enantiomer, contributing to QTc prolongation without contributing to SERT inhibition or antidepressant efficacy. Escitalopram, containing only the S-enantiomer, produces less QTc prolongation at therapeutically equivalent SERT-blocking doses, though the effect is not absent. A thorough QTc study conducted prior to the FDA safety communication demonstrated a mean QTc increase of approximately 8.5 milliseconds at the standard 20 mg dose of citalopram and approximately 18.5 milliseconds at the 60 mg dose (which is now above the FDA dose cap).7
In 2011, the FDA issued a safety communication limiting citalopram to a maximum of 40 mg per day in most patients, with a further restriction to 20 mg per day in patients over 60 years of age, those with hepatic impairment, and those who are CYP2C19 poor metabolizers, based on evidence that these populations achieve higher citalopram plasma concentrations and correspondingly greater QTc prolongation. Citalopram doses above 40 mg per day are no longer recommended by the FDA for any indication.7 Escitalopram received a similar label update limiting the maximum dose to 20 mg per day in the same higher-risk populations.
Risk stratification for SSRI-related QTc prolongation requires identification of additive risk factors. Baseline QTc above 450 milliseconds in men or 470 milliseconds in women warrants caution or avoidance of QTc-prolonging medications. Additional risk factors include hypokalemia, hypomagnesemia, bradycardia, congenital long QT syndrome, female sex (women have longer baseline QTc than men), advanced age, structural heart disease, and concurrent use of other QTc-prolonging drugs including antipsychotics, antiarrhythmics (class Ia and III), macrolide antibiotics, fluoroquinolones, and methadone.8 For patients with multiple QTc risk factors who require an SSRI, sertraline, paroxetine, or fluoxetine are preferable choices, as they have not demonstrated clinically significant QTc prolongation at therapeutic doses, though all prescribing decisions in high-risk cardiac patients warrant cardiology input.
Obtain a baseline ECG before initiating citalopram or escitalopram in patients with known cardiac disease, electrolyte abnormalities, or concurrent QTc-prolonging medications. Recheck QTc four to six weeks after reaching the target dose. If QTc exceeds 500 milliseconds at any point, the drug should be discontinued or the dose reduced and alternative therapy considered. Electrolyte correction, particularly potassium and magnesium, should accompany any QTc-prolonging drug initiation in patients at risk of depletion.
Serotonin syndrome (SS) is a potentially life-threatening adverse drug reaction resulting from excess serotonergic activity at central and peripheral 5-HT receptors, most commonly the 5-HT1A and 5-HT2A subtypes. It is almost always iatrogenic, arising from the combination of two or more serotonergic agents or from severe overdose of a single agent. Serotonin syndrome is not an idiosyncratic reaction; it is a predictable pharmacological consequence of excessive serotonergic stimulation and is dose-dependent in nature. The clinical triad consists of neuromuscular abnormalities (clonus, hyperreflexia, myoclonus, tremor, ataxia), autonomic instability (hyperthermia, diaphoresis, tachycardia, hypertension, diarrhea), and altered mental status (agitation, anxiety, disorientation).9 The presence of all three elements of the triad is not required for diagnosis.
The Hunter Serotonin Toxicity Criteria are the most clinically useful diagnostic tool for serotonin syndrome, with higher sensitivity (84%) and specificity (97%) than the Sternbach criteria.9 The Hunter criteria require exposure to a serotonergic agent plus one of the following: spontaneous clonus; inducible clonus plus agitation or diaphoresis; ocular clonus plus agitation or diaphoresis; tremor plus hyperreflexia; or hypertonia plus temperature above 38 degrees Celsius plus ocular or inducible clonus. The emphasis on clonus, particularly ocular clonus (a rhythmic oscillation of the eyes), distinguishes serotonin syndrome from neuroleptic malignant syndrome (NMS), which produces lead-pipe rigidity rather than clonus and develops more slowly over 24 to 72 hours rather than within 24 hours of the precipitating pharmacological change.
Serotonin syndrome exists on a severity spectrum. Mild cases present with tremor, tachycardia, diaphoresis, myoclonus, and mydriasis but normal temperature and preserved orientation. Moderate cases add hyperthermia (temperature typically 38 to 40 degrees Celsius), agitation, hyperreflexia, and clonus. Severe cases are characterized by extreme hyperthermia (above 41 degrees Celsius), muscle rigidity, rhabdomyolysis, metabolic acidosis, renal failure, and seizures, representing a medical emergency with significant mortality if not treated promptly.9 Hyperthermia in severe SS is primarily metabolic, driven by continuous muscle hyperactivity rather than a disorder of thermoregulation, which has implications for treatment: aggressive neuromuscular paralysis and sedation are required to halt heat generation, not merely antipyretic therapy.
The highest-risk precipitating combinations involve agents that increase 5-HT synthesis or release combined with agents that reduce 5-HT degradation or reuptake. SSRI combined with an irreversible MAOI is the prototypical and most dangerous combination, capable of producing severe or fatal serotonin syndrome; this combination is absolutely contraindicated and requires the washout periods described in Section 03. Other high-risk combinations include SSRIs with linezolid (a reversible MAO-A inhibitor and weak MAOI used as an antibiotic) or intravenous methylene blue (which inhibits MAO and can precipitate SS even when given perioperatively to a patient on an SSRI).10 Moderate-risk combinations include SSRIs with triptans (5-HT1B/1D agonists; the risk is debated and probably lower than once thought but warrants monitoring), tramadol (which inhibits SERT and NET in addition to its opioid activity), meperidine (SERT inhibition), dextromethorphan, fentanyl, and St. John's Wort (hypericum perforatum, which contains a weak SERT inhibitor).
Management of serotonin syndrome begins with identification and discontinuation of all serotonergic agents. Mild cases often resolve within 24 hours of drug cessation with supportive care including benzodiazepines for agitation and myoclonus, hydration, and cardiac monitoring. Moderate cases require hospital admission, continuous monitoring, active cooling if temperature exceeds 39 degrees Celsius, and consideration of cyproheptadine, a histamine H1 antagonist with 5-HT2A antagonist properties that can reduce serotonergic tone.9 Cyproheptadine is available only in oral form, which limits its use in severely agitated or intubated patients. Severe cases require intensive care unit (ICU) admission, aggressive benzodiazepine sedation, endotracheal intubation, neuromuscular paralysis to halt hyperthermia from muscle hyperactivity, cooling measures, and management of rhabdomyolysis and its renal consequences. Dantrolene, which is used in malignant hyperthermia (MH), is not indicated for serotonin syndrome, as the hyperthermia in SS is metabolic rather than due to uncontrolled muscle calcium release via the ryanodine receptor.
Both produce hyperthermia and altered mental status but differ in time course, neuromuscular findings, and precipitant. Serotonin syndrome: rapid onset (within 24 hours), clonus and hyperreflexia, precipitated by serotonergic agents. NMS: slow onset (24 to 72 hours or more), lead-pipe rigidity and bradyreflexia, precipitated by dopamine antagonists or abrupt dopaminergic withdrawal. Treatment also differs: cyproheptadine and benzodiazepines for SS; bromocriptine and dantrolene for NMS. Misidentification can result in inappropriate treatment with serious consequences.
Stahl SM. Mechanism of action of serotonin selective reuptake inhibitors: serotonin receptors and pathways mediate therapeutic effects and side effects. J Affect Disord. 1998;51(3):215–235.
doi:10.1016/S0165-0327(98)00221-3Blier P, de Montigny C. Current advances and trends in the treatment of depression. Trends Pharmacol Sci. 1994;15(7):220–226.
doi:10.1016/0165-6147(94)90315-8Casarotto PC, Girych M, Fred SM, et al. Antidepressant drugs act by directly binding to TRKB neurotrophin receptors. Cell. 2021;184(5):1299–1313.e19.
doi:10.1016/j.cell.2021.01.034American Psychiatric Association. Practice Guideline for the Treatment of Patients with Major Depressive Disorder. 3rd ed. Washington, DC: American Psychiatric Publishing; 2010.
Hiemke C, Hartter S. Pharmacokinetics of selective serotonin reuptake inhibitors. Pharmacol Ther. 2000;85(1):11–28.
doi:10.1016/S0163-7258(99)00048-0Hemeryck A, Belpaire FM. Selective serotonin reuptake inhibitors and cytochrome P-450 mediated drug-drug interactions: an update. Curr Drug Metab. 2002;3(1):13–37.
doi:10.2174/1389200023338017U.S. Food and Drug Administration. FDA Drug Safety Communication: Revised recommendations for Celexa (citalopram hydrobromide) related to a potential risk of abnormal heart rhythms with high doses. 2012. Available at: https://www.fda.gov/drugs/drug-safety-and-availability/fda-drug-safety-communication-revised-recommendations-celexa-citalopram-hydrobromide-related
Woosley RL, Heise CW, Romero KA. QTDrugs List. AZCERT, Inc. 1822 Innovation Park Dr., Oro Valley, AZ 85755. Available at: https://www.crediblemeds.org. Accessed 2025.
Boyer EW, Shannon M. The serotonin syndrome. N Engl J Med. 2005;352(11):1112–1120.
doi:10.1056/NEJMra041867Pedavally S, Fugate JE, Rabinstein AA. Serotonin syndrome in the intensive care unit: clinical presentations and precipitating medications. Neurocrit Care. 2014;21(1):108–113.
doi:10.1007/s12028-013-9914-2Kelly CM, Juurlink DN, Gomes T, et al. Selective serotonin reuptake inhibitors and breast cancer mortality in women receiving tamoxifen: a population based cohort study. BMJ. 2010;340:c693.
doi:10.1136/bmj.c693Stahl SM. Stahl's Essential Psychopharmacology: Neuroscientific Basis and Practical Applications. 5th ed. Cambridge: Cambridge University Press; 2021. Chapter 8.
Søgaard B, Mengel H, Rao N, Larsen F. The pharmacokinetics of escitalopram after oral and intravenous administration of single and multiple doses to healthy subjects. J Clin Pharmacol. 2005;45(12):1400–1406.
doi:10.1177/0091270005280860