The entire central nervous system (CNS) serotonergic output originates from a compact cluster of brainstem nuclei collectively called the raphe nuclei, whose axonal projections reach virtually every region of the brain and spinal cord. Understanding this anatomy is not an academic exercise: the projection pattern of each raphe nucleus predicts which clinical functions are modulated by serotonergic drugs, which adverse effects arise from which receptor distributions, and why disruption of the serotonergic system produces such a broad range of psychiatric and somatic symptoms.
The raphe nuclei form a midline column of neurons in the brainstem, extending from the midbrain through the pons and into the medulla. The two most pharmacologically relevant nuclei are the dorsal raphe nucleus (DRN) and the median raphe nucleus (MRN). The dorsal raphe nucleus is the largest serotonergic nucleus in the brain and is the primary source of serotonergic innervation to the forebrain. Its ascending projections reach the prefrontal cortex, the striatum (caudate, putamen, and nucleus accumbens), the amygdala, the hypothalamus, and the thalamus. The median raphe nucleus projects primarily to the hippocampus and to septal nuclei. This anatomical segregation has functional implications: DRN projections to the prefrontal cortex and limbic structures are most relevant to mood, anxiety, and cognitive regulation, while MRN projections to the hippocampus are particularly relevant to memory consolidation and the neuroplastic effects of antidepressant treatment.1
The functional correlates of these projection pathways explain the broad clinical effects of serotonergic drugs. Serotonergic modulation of the prefrontal cortex influences executive function, decision-making, and impulse control; disruption of this pathway is implicated in impulsivity and certain features of depression. Serotonergic input to the amygdala modulates fear responses and anxiety; enhanced serotonergic tone in the amygdala is associated with anxiolytic effects, which is why SSRIs (selective serotonin reuptake inhibitors) are first-line treatments for generalized anxiety disorder (GAD), social anxiety disorder, panic disorder, and post-traumatic stress disorder (PTSD) in addition to major depressive disorder (MDD). Serotonergic projections to the hypothalamus regulate appetite, body temperature, and hormonal axes including the hypothalamic-pituitary-adrenal (HPA) axis; excess serotonergic drive in the hypothalamus contributes to the hyperthermia and autonomic instability of serotonin syndrome. Descending serotonergic projections from the raphe nuclei to the spinal cord dorsal horn modulate pain processing, providing the mechanistic basis for the analgesic effects of serotonin-norepinephrine reuptake inhibitors (SNRIs) in neuropathic pain and fibromyalgia.2
Serotonergic regulation of sleep architecture is mediated primarily through projections to the suprachiasmatic nucleus and brainstem sleep centers. Serotonin suppresses rapid eye movement (REM) sleep; serotonergic neurons in the DRN are maximally active during wakefulness, reduce firing during non-REM sleep, and are virtually silent during REM sleep. This explains why SSRIs frequently suppress REM sleep and increase REM latency, which can cause vivid dreams and, in some patients, exacerbate parasomnias or restless leg syndrome. Serotonergic modulation of appetite is mediated through projections to the hypothalamic ventromedial and paraventricular nuclei, where 5-HT2C receptor activation suppresses food intake; this is the mechanistic basis for the appetite-suppressant effects of serotonergic agents and the weight-promoting effects of drugs with potent 5-HT2C antagonism. Serotonergic projections also regulate sexual function through spinal cord pathways, contributing to the delayed ejaculation and anorgasmia that are among the most common adverse effects of SSRI treatment, occurring in 30–40% of patients at therapeutic doses.3
Dorsal raphe nucleus (DRN): projects to prefrontal cortex, striatum, amygdala, hypothalamus, thalamus. Mediates mood, anxiety, impulse control, appetite, and autonomic regulation. Median raphe nucleus (MRN): projects primarily to hippocampus and septal nuclei. Mediates memory consolidation and neuroplastic antidepressant responses. Both nuclei send descending projections to the spinal cord dorsal horn for pain modulation. All raphe serotonergic neurons express somatodendritic 5-HT1A autoreceptors that provide negative feedback on firing rate — the key mechanism underlying the delayed onset of SSRI action.
The most clinically significant feature of SSRI pharmacology is the disconnect between the drug's immediate molecular effect (SERT blockade begins within hours of the first dose) and its clinical antidepressant effect, which typically requires 2 to 4 weeks of continuous treatment to emerge. This therapeutic lag is not simply a matter of receptor adaptation; it reflects a cascade of neurobiological events that must occur sequentially before the drug produces its full clinical effect. Understanding this sequence is essential for counseling patients, setting realistic expectations, and avoiding the error of prematurely discontinuing effective treatment.
The acute molecular effect of SSRIs is competitive blockade of the serotonin transporter (SERT, SLC6A4) on presynaptic serotonergic terminals and somatodendritic membranes. Within hours of the first dose, SERT occupancy rises to therapeutic levels (exceeding 80% at standard doses for most SSRIs), and extracellular serotonin concentrations in the synapse begin to increase. This immediate increase in synaptic serotonin, however, does not translate directly into enhanced postsynaptic serotonergic signaling, because of a counterregulatory mechanism mediated by presynaptic autoreceptors. Somatodendritic 5-HT1A autoreceptors on raphe neuron cell bodies detect the rising synaptic serotonin and activate Gi-coupled inhibitory signaling that reduces the firing rate of the raphe neuron itself, thereby reducing serotonin release at the axon terminal. Terminal 5-HT1B autoreceptors further reduce the amount of serotonin released per action potential. The net effect is that acute SSRI treatment produces only a modest net increase in serotonergic neurotransmission at postsynaptic targets despite substantial SERT blockade.4
The resolution of this autoreceptor brake is the critical event that enables full therapeutic effect. With chronic SSRI exposure over 2 to 4 weeks, the sustained elevation of serotonin at somatodendritic 5-HT1A autoreceptors causes these receptors to desensitize: their surface expression decreases, their coupling to Gi proteins is reduced, and their ability to suppress raphe neuron firing is diminished. As 5-HT1A autoreceptor function wanes, raphe neuron firing rate recovers toward baseline, but now in the context of sustained SERT blockade, producing substantially increased serotonin release at terminal fields. The time course of this autoreceptor desensitization corresponds closely to the clinical antidepressant response latency, providing the strongest mechanistic explanation for the therapeutic lag. Pindolol, a beta-blocker with 5-HT1A antagonist properties, has been studied as an augmentation strategy to accelerate this process by blocking the autoreceptor acutely, with modest evidence of faster response onset but no consistent benefit on ultimate response rate.5
Beyond autoreceptor desensitization, chronic serotonergic enhancement activates downstream neuroplastic mechanisms that are increasingly recognized as central to antidepressant efficacy. Sustained activation of postsynaptic 5-HT receptors, particularly in the hippocampus and prefrontal cortex, upregulates brain-derived neurotrophic factor (BDNF) expression through cAMP response element-binding protein (CREB) signaling. BDNF promotes dendritic arborization, synaptic plasticity, and hippocampal neurogenesis in the dentate gyrus, reversing stress-induced reductions in hippocampal volume that are observed in chronic depression. Normalization of hypothalamic-pituitary-adrenal (HPA) axis hyperactivity, which drives excess cortisol secretion in MDD, also occurs over weeks of SSRI treatment and correlates with clinical remission. These neuroplastic changes, not the acute SERT blockade, are now understood to be the proximate therapeutic mechanism, explaining both the delayed onset and the sustained benefit that persists even after drug discontinuation in some patients.6
Patients starting an SSRI should be told explicitly that antidepressant effect requires 2 to 4 weeks and full effect may take 6 to 8 weeks. Side effects (nausea, insomnia, activation) typically emerge in the first 1 to 2 weeks, before efficacy is apparent. This temporal mismatch is the most common reason for premature discontinuation. Patients should also be told that some improvement — particularly in sleep, appetite, and energy — may emerge before mood lifts, and that this is a sign the drug is working. The absence of complete response at 2 weeks is not a signal to stop; the absence of any response at 4 to 6 weeks is the appropriate trigger for dose adjustment or augmentation.
SERT occupancy required for antidepressant effect has been quantified by positron emission tomography (PET) imaging studies using SERT-binding radioligands. These studies consistently show that 80% or greater SERT occupancy is required for clinically meaningful antidepressant effect, and that standard therapeutic doses of all SSRIs achieve this threshold reliably. Dose increases above the standard therapeutic dose do not substantially increase SERT occupancy (which approaches saturation at standard doses) but may engage other receptor targets or alter downstream signaling, which is why higher doses are not always more effective for antidepressant action but may offer additional benefit in anxiety disorders through partial 5-HT2A or other receptor effects. The concept of receptor occupancy also explains why the minimum effective doses established in clinical trials correspond to the pharmacokinetically determined doses that reliably achieve 80% SERT occupancy across most patients.4
Although all SSRIs share the same primary mechanism of SERT blockade, they differ substantially in their pharmacokinetic properties, CYP enzyme inhibition profiles, and secondary receptor actions. These differences are not merely academic: they determine which patients are at risk for drug interactions, which agents are safest in cardiac disease, which require the most careful discontinuation, and which can be switched most cleanly when a patient requires a change in therapy.
Fluoxetine is distinguished from all other SSRIs by its exceptionally long half-life and the presence of an active metabolite. The parent drug has a half-life of 1 to 4 days, but its active metabolite norfluoxetine has a half-life of 4 to 16 days, producing an effective elimination half-life that can exceed 2 weeks. This long half-life has two major clinical consequences. First, fluoxetine is the only SSRI that does not require a taper before discontinuation; its gradual self-taper prevents the discontinuation syndrome (dizziness, electric-shock sensations, irritability, insomnia) that characterizes abrupt cessation of shorter-acting SSRIs. Second, fluoxetine and norfluoxetine are potent inhibitors of CYP2D6 and moderate inhibitors of CYP2C9; these interactions persist for weeks after fluoxetine is stopped, which is why a 5-week washout is required before starting an MAOI (monoamine oxidase inhibitor) after fluoxetine. Fluoxetine is the most activating SSRI and is preferred in patients with prominent hypersomnia and fatigue, but its activating properties make it less suitable for patients with prominent insomnia or anxiety at initiation.7
Sertraline has the most favorable pharmacokinetic profile among the SSRIs for most clinical situations. Its half-life of approximately 26 hours permits once-daily dosing with consistent steady-state levels. It has linear pharmacokinetics across the therapeutic dose range, meaning that doubling the dose approximately doubles the plasma concentration. Sertraline is a mild inhibitor of CYP2D6 at therapeutic doses, producing fewer clinically significant drug interactions than fluoxetine or paroxetine. Its active metabolite desmethylsertraline has substantially lower pharmacological activity than the parent drug and does not contribute meaningfully to clinical effect. Sertraline is the preferred SSRI in patients on multiple medications, in post-myocardial infarction depression (supported by the SADHART trial), and in pregnancy, where its safety data are most extensive among the SSRIs.8
Paroxetine has the shortest half-life of the standard SSRIs (approximately 21 hours), no active metabolites, and the highest degree of anticholinergic activity among the SSRIs, producing dry mouth, constipation, blurred vision, and urinary hesitancy more commonly than other agents in the class. It is the most potent CYP2D6 inhibitor among the SSRIs; at standard doses, paroxetine produces near-complete inhibition of CYP2D6, raising the plasma concentrations of co-administered CYP2D6 substrates including tricyclic antidepressants (TCAs), many antipsychotics, metoprolol, and codeine (impairing its conversion to morphine and reducing analgesic efficacy). Paroxetine's short half-life and abrupt SERT de-occupancy on missed doses makes it the SSRI most strongly associated with discontinuation syndrome, characterized by flu-like symptoms, electric-shock sensations (brain zaps), dizziness, and rebound anxiety. Gradual dose tapering over weeks is mandatory before discontinuation. Despite these drawbacks, paroxetine has the most extensive evidence base for social anxiety disorder and panic disorder, and its sedating properties make it useful in patients with comorbid insomnia.7
Citalopram and escitalopram are the most selective SSRIs, binding to SERT with minimal affinity for other receptors and producing the least CYP enzyme inhibition among the class. Escitalopram is the S-enantiomer of the racemic citalopram and has approximately twice the SERT affinity of citalopram's R-enantiomer; at equipotent doses, escitalopram produces fewer adverse effects. Both agents are metabolized primarily by CYP2C19 and to a lesser extent by CYP3A4 and CYP2D6. An important safety distinction between citalopram and escitalopram is that citalopram causes QTc prolongation in a dose-dependent manner, with the FDA recommending a maximum dose of 40 mg per day in most patients and 20 mg per day in patients over 60 years of age, those with hepatic impairment, and those on CYP2C19 inhibitors such as omeprazole. Escitalopram causes substantially less QTc prolongation than citalopram at equivalent therapeutic doses and is preferred over citalopram in patients at risk for cardiac arrhythmia.9 Both citalopram and escitalopram are the preferred SSRIs for use with warfarin because their minimal CYP inhibition avoids pharmacokinetic elevation of warfarin levels.
Fluvoxamine occupies a distinct niche among the SSRIs due to its disproportionate CYP enzyme inhibition relative to its antidepressant use. Fluvoxamine is a potent inhibitor of CYP1A2 and CYP3A4 and a moderate inhibitor of CYP2C19, producing clinically significant interactions with a wide range of co-administered drugs. The CYP1A2 interaction is particularly important in patients on clozapine (an antipsychotic with a narrow therapeutic index metabolized by CYP1A2), theophylline, and caffeine; fluvoxamine can raise clozapine plasma concentrations by 5 to 10-fold, producing toxicity at previously tolerated doses. Fluvoxamine's primary regulatory indication in the United States is obsessive-compulsive disorder (OCD) rather than major depressive disorder (MDD); it is also used for social anxiety disorder. Its sedating properties are useful in patients with prominent anxiety and sleep disturbance but may be problematic when daytime alertness is required.7
From fluoxetine to an MAOI: 5-week washout required (norfluoxetine half-life). From any other SSRI to an MAOI: 2-week washout. From an MAOI to any SSRI: 2-week washout after stopping irreversible MAOI. From moclobemide (reversible MAOI) to an SSRI: 24 hours washout.
Switching between SSRIs: For most switches (e.g., sertraline to escitalopram), a direct switch or brief cross-taper is appropriate. Switching from fluoxetine to a short-acting SSRI may require waiting for fluoxetine/norfluoxetine washout if drug interactions are a concern, but direct switching is often used in practice given the self-tapering property of fluoxetine.
Discontinuation syndrome risk: Highest with paroxetine and venlafaxine; moderate with sertraline and fluvoxamine; lowest with fluoxetine (self-tapering). Always taper short-acting SSRIs over at least 2 to 4 weeks when discontinuing after prolonged use.
SNRIs (serotonin-norepinephrine reuptake inhibitors) add norepinephrine transporter (NET) inhibition to the SERT blockade shared with SSRIs. This dual mechanism has meaningful clinical consequences beyond treating depression: enhanced noradrenergic tone in descending pain pathways provides analgesic benefit in neuropathic pain syndromes, noradrenergic enhancement in the prefrontal cortex improves attention and working memory, and peripheral noradrenergic effects include increases in heart rate and blood pressure that require monitoring. Understanding which agents achieve meaningful NET inhibition at therapeutic doses, and under what conditions, is necessary for choosing between agents in the class.
Venlafaxine illustrates the concept of dose-dependent dual inhibition. At low doses (37.5–75 mg per day), venlafaxine predominantly blocks SERT, with NET inhibition minimal. As the dose increases toward 150 mg per day and above, NET inhibition becomes pharmacologically significant, producing the noradrenergic component of the drug's effect. At doses of 225 mg per day and higher, venlafaxine also begins to inhibit the dopamine transporter (DAT), though this contribution is pharmacologically minor at clinically used doses. This dose-dependent mechanism means that switching a patient from low-dose venlafaxine (functioning essentially as an SSRI) to high-dose venlafaxine (achieving dual inhibition) is a pharmacologically distinct step. Venlafaxine is extensively metabolized by CYP2D6 to its primary active metabolite O-desmethylvenlafaxine (desvenlafaxine), which itself is now available as a standalone agent. CYP2D6 poor metabolizers achieve higher venlafaxine and lower desvenlafaxine plasma concentrations, though both parent and metabolite are pharmacologically active, so the clinical consequence is modest. Venlafaxine is associated with dose-dependent increases in blood pressure and heart rate; regular monitoring is warranted at higher doses, and it should be used with caution in uncontrolled hypertension.11
Duloxetine achieves more balanced dual SERT and NET inhibition across its full therapeutic dose range (60–120 mg per day), without the dose-dependent progression seen with venlafaxine. This balanced profile makes duloxetine particularly useful in clinical situations requiring both antidepressant and analgesic effects simultaneously, such as major depressive disorder (MDD) with comorbid diabetic peripheral neuropathy, fibromyalgia, or chronic musculoskeletal pain; duloxetine has regulatory approval for each of these indications. Duloxetine is metabolized primarily by CYP1A2 and CYP2D6; CYP1A2 inducers such as tobacco smoking can significantly reduce duloxetine plasma concentrations, and CYP1A2 inhibitors such as fluvoxamine can markedly increase them. Duloxetine is a moderate inhibitor of CYP2D6 and can raise plasma concentrations of co-administered CYP2D6 substrates, though its inhibitory potency is less than that of paroxetine or fluoxetine. It should not be used in patients with significant hepatic impairment due to extensive hepatic metabolism and the risk of hepatotoxicity at elevated concentrations.12
Desvenlafaxine, the active metabolite of venlafaxine, is available as a once-daily extended-release formulation. Because it is not extensively metabolized by CYP enzymes (it undergoes primarily glucuronidation), desvenlafaxine produces fewer pharmacokinetic drug interactions than venlafaxine or duloxetine. Approximately 45% of desvenlafaxine is excreted unchanged in urine, making dose adjustment necessary in severe renal impairment. Levomilnacipran is the active enantiomer of milnacipran and has a uniquely high ratio of NET to SERT inhibition among the approved SNRIs, with NET inhibition substantially exceeding SERT inhibition at therapeutic doses. This makes levomilnacipran the SNRI with the strongest noradrenergic profile and the most pronounced effects on heart rate and blood pressure. It requires dose adjustment in moderate to severe renal impairment. The clinical implication of the strong noradrenergic profile is that levomilnacipran may be particularly useful in patients with prominent fatigue, hypersomnia, and psychomotor slowing as features of their depression, but requires careful cardiovascular monitoring.13
Duloxetine: MDD (major depressive disorder), GAD (generalized anxiety disorder), diabetic peripheral neuropathy, fibromyalgia, chronic musculoskeletal pain. Venlafaxine: MDD, GAD, social anxiety disorder, panic disorder, PTSD (post-traumatic stress disorder). Both agents are used for chemotherapy-induced peripheral neuropathy (off-label), menopausal vasomotor symptoms (off-label), and as adjuncts in chronic pain management. The noradrenergic component of SNRIs activates descending inhibitory pain pathways in the spinal cord dorsal horn via alpha-2 adrenoceptors and contributes to analgesic benefit independently of antidepressant effect.
The ADME profiles of SSRIs and SNRIs differ in ways that are directly clinically relevant. Protein binding is high for most agents in both classes (fluoxetine approximately 94%, sertraline approximately 98%, paroxetine approximately 95%, venlafaxine approximately 27%). The low protein binding of venlafaxine means it is less likely to be displaced from binding sites by co-administered highly protein-bound drugs, reducing one class of pharmacokinetic interaction. Volume of distribution (Vd) is large for most agents (fluoxetine Vd approximately 20–40 L/kg; paroxetine approximately 8–28 L/kg), reflecting extensive tissue distribution and meaning that dialysis is ineffective for managing overdose. Renal impairment requires dose adjustment for desvenlafaxine, levomilnacipran, and venlafaxine (which has significant renal excretion of active metabolite); SSRIs are generally safer in renal impairment because their elimination is primarily hepatic. Hepatic impairment affects all agents in both classes to varying degrees, with paroxetine and duloxetine requiring the most caution in severe hepatic disease.7
SSRIs and SNRIs produce clinically significant drug interactions through two distinct mechanisms: pharmacokinetic interactions arising from CYP enzyme inhibition that alter the metabolism of co-administered drugs, and pharmacodynamic interactions that arise from the combination of their biological effects with those of other agents. The most consequential pharmacokinetic interactions involve CYP2D6, CYP1A2, and CYP2C9 inhibition; the most clinically common pharmacodynamic interactions involve increased bleeding risk and, in some combinations, serotonin syndrome. Recognizing these interactions is a mandatory component of prescribing these agents safely.
CYP2D6 inhibition by fluoxetine and paroxetine produces the most clinically impactful pharmacokinetic interactions in this drug class. CYP2D6 metabolizes a wide range of commonly prescribed drugs including tricyclic antidepressants (TCAs), many antipsychotics (haloperidol, risperidone, aripiprazole, thioridazine), beta-blockers (metoprolol, carvedilol), antiarrhythmics (flecainide, propafenone), opioids (codeine, tramadol), and tamoxifen. The clinical consequences of these interactions are drug-specific: co-administration of fluoxetine or paroxetine with a TCA raises TCA plasma concentrations by 2 to 5-fold, producing TCA toxicity (QTc prolongation, anticholinergic effects, cardiac arrhythmia) at previously tolerated TCA doses. Fluoxetine or paroxetine combined with tamoxifen significantly impairs conversion of tamoxifen to its active metabolite endoxifen by CYP2D6, reducing endoxifen plasma concentrations by 40–65% and potentially diminishing the breast cancer survival benefit of tamoxifen therapy; guidelines recommend avoiding fluoxetine and paroxetine in patients receiving tamoxifen for hormone receptor-positive breast cancer, with sertraline, venlafaxine, or escitalopram as preferred alternatives.14
Fluvoxamine's potent inhibition of CYP1A2 produces a distinct set of interactions. CYP1A2 is the primary enzyme metabolizing clozapine, and co-administration of fluvoxamine with clozapine can raise clozapine plasma concentrations by 5 to 10-fold, producing clozapine toxicity including sedation, hypersalivation, hypotension, seizures, and potentially fatal agranulocytosis at previously tolerated doses. This combination must be avoided unless clozapine dose is substantially reduced and plasma levels are closely monitored. Other CYP1A2 substrates affected by fluvoxamine include theophylline (narrow therapeutic index; toxicity risk), caffeine (prolonged half-life and effects), ramelteon (melatonin receptor agonist; contraindicated with fluvoxamine), and several antipsychotics including olanzapine. Fluvoxamine also inhibits CYP2C19, reducing the metabolism of omeprazole, diazepam, and phenytoin, and is a moderate CYP3A4 inhibitor, potentially elevating levels of benzodiazepines metabolized by CYP3A4 (alprazolam, triazolam) and some statins.10
Antidepressants interact with warfarin through two distinct mechanisms that are pharmacokinetically additive. SSRIs and SNRIs deplete platelet serotonin through SERT inhibition, impairing platelet aggregation and potentiating warfarin's anticoagulant effect on the clotting cascade without changing warfarin pharmacokinetics. Fluvoxamine, fluoxetine, and to a lesser degree paroxetine inhibit CYP2C9, which metabolizes the more potent S-warfarin enantiomer, pharmacokinetically increasing warfarin concentrations and international normalized ratio (INR).15 The combined pharmacodynamic and pharmacokinetic interaction requires INR monitoring within one to two weeks of antidepressant initiation or dose change in patients on warfarin. Among SSRIs, citalopram and escitalopram have the least CYP2C9 inhibitory activity and are the preferred choices in patients on warfarin when SSRI therapy is indicated.
Fluoxetine or paroxetine + tamoxifen: use an alternative SSRI. Fluvoxamine + clozapine: contraindicated in practice. Any SSRI + warfarin: INR check within 2 weeks. St. John's wort + any prescription antidepressant: avoid combination due to induction of CYP3A4 and SERT expression, reducing SSRI plasma levels and risking pharmacodynamic serotonergic interactions simultaneously. Carbamazepine + antidepressant: expect reduced antidepressant concentrations; consider dose adjustment and concentration monitoring.
The bleeding risk associated with SSRIs and SNRIs deserves particular clinical attention in surgical and procedural settings. The mechanism is depletion of platelet serotonin stores over days to weeks of SERT blockade, impairing the serotonin-mediated amplification of platelet aggregation without producing a coagulation factor abnormality detectable on standard prothrombin time or activated partial thromboplastin time. Studies consistently show an approximately 2 to 3-fold increased risk of upper gastrointestinal bleeding with SSRIs alone, rising to 10 to 15-fold when combined with non-steroidal anti-inflammatory drugs (NSAIDs), which impair the prostaglandin-mediated mucosal protection that partially compensates for platelet dysfunction. In patients requiring chronic NSAID use, a proton pump inhibitor (PPI) co-prescribed with the SSRI substantially mitigates but does not eliminate the GI bleeding risk. The decision to continue or hold an SSRI before elective surgical procedures requires individualized risk assessment balancing bleeding risk against the risk of psychiatric decompensation during the perioperative period; most guidelines do not recommend routine discontinuation of SSRIs before non-cardiac surgery, though the evidence base for cardiac surgery is less clear.15
Jacobs BL, Azmitia EC. Structure and function of the brain serotonin system. Physiol Rev. 1992;72(1):165–229.
doi:10.1152/physrev.1992.72.1.165Millan MJ. Descending control of pain. Prog Neurobiol. 2002;66(6):355–474.
doi:10.1016/S0301-0082(02)00009-6Montejo AL, Llorca G, Izquierdo JA, Rico-Villademoros F. Incidence of sexual dysfunction associated with antidepressant agents: a prospective multicenter study of 1022 outpatients. J Clin Psychiatry. 2001;62(Suppl 3):10–21.
Sorensen A, Ruhe HG, Munkholm K. The relationship between dose and serotonin transporter occupancy of antidepressants: a systematic review. Mol Psychiatry. 2022;27(1):192–201.
doi:10.1038/s41380-021-01285-wBlier 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)90315-8Duman RS, Monteggia LM. A neurotrophic model for stress-related mood disorders. Biol Psychiatry. 2006;59(12):1116–1127.
doi:10.1016/j.biopsych.2006.02.013Stahl SM. Stahl's Essential Psychopharmacology: Neuroscientific Basis and Practical Applications. 4th ed. Cambridge, UK: Cambridge University Press; 2013:284–352.
Glassman AH, O'Connor CM, Califf RM, et al. Sertraline treatment of major depression in patients with acute MI or unstable angina. JAMA. 2002;288(6):701–709.
doi:10.1001/jama.288.6.701Vieweg WV, Hasnain M, Howland RH, et al. Citalopram, QTc interval prolongation, and torsade de pointes: how should we apply the recent FDA ruling? Am J Med. 2012;125(9):859–868.
doi:10.1016/j.amjmed.2012.03.014Spigset O, Hagg S, Soderstrom E, Dahlqvist R. Fluvoxamine pharmacokinetics in relation to cytochrome P-450 phenotype. Ther Drug Monit. 1997;19(3):297–302.
doi:10.1097/00007691-199706000-00008Harvey AT, Rudolph RL, Preskorn SH. Evidence of the dual mechanisms of action of venlafaxine. Arch Gen Psychiatry. 2000;57(5):503–509.
doi:10.1001/archpsyc.57.5.503Wohlreich MM, Acharya N, Strombom I, et al. Answers to the most common questions about the hepatic safety profile of duloxetine. Postgrad Med. 2008;120(2):111–118.
doi:10.3810/pgm.2008.07.1803Asnis GM, Henderson MA. Levomilnacipran for the treatment of major depressive disorder: a review. Neuropsychiatr Dis Treat. 2015;11:125–135.
doi:10.2147/NDT.S54710Kelly CM, Juurlink DN, Gomes T, et al. Selective serotonin reuptake inhibitors and breast cancer mortality in women receiving tamoxifen. BMJ. 2010;340:c693.
doi:10.1136/bmj.c693Anglin R, Yuan Y, Moayyedi P, Tse F, Armstrong D, Leontiadis GI. Risk of upper gastrointestinal bleeding with selective serotonin reuptake inhibitors with or without concurrent nonsteroidal anti-inflammatory use: a systematic review and meta-analysis. Am J Gastroenterol. 2014;109(6):811–819.
doi:10.1038/ajg.2014.82