Choose antidepressants by clinical goals and adverse effects, recognize toxic syndromes, and plan safe monitoring, pregnancy care, tapering, and switches.
An antidepressant choice is a prediction about benefit, tolerability, and safety in one person. Two patients with the same depression score may need different treatment because one has painful neuropathy and another has a history of bulimia. Start with the illness being treated and the adverse effect the patient can least afford, rather than treating a drug class as a list of interchangeable names.
Connect the target to the clinical effect
At a monoamine synapse, released transmitter acts on receptors and is then cleared or metabolized. SERT retrieves serotonin, NET retrieves norepinephrine, and DAT retrieves dopamine. Inhibiting a transporter reduces reuptake; it does not increase the transporter's ability to clear transmitter. MAO inhibition reduces monoamine metabolism. Presynaptic alpha-2 blockade reduces an inhibitory feedback signal. These mechanisms help organize drugs, but they do not prove that depression is simply a measured shortage of one transmitter.
Clearance at the nerve ending
SSRIs inhibit SERT. SNRIs inhibit SERT and NET. Bupropion inhibits norepinephrine and dopamine reuptake. These actions differ from direct receptor activation.
Metabolism inside cells
MAO inhibitors limit monoamine breakdown. Interactions with dietary tyramine and other drugs depend on the particular formulation and dose.
Receptor regulation
Mirtazapine blocks alpha-2 receptors and selected serotonin receptors. Histamine H1 antagonism helps explain its sedation and appetite effects.
Transporter effects occur before the full clinical response. Early nausea does not mean an SSRI has already delivered its maximum antidepressant benefit. Follow symptoms, function, adherence, and adverse effects over time. Improvement may emerge over several weeks, but a rule to do nothing until week eight is inappropriate when adverse effects, suicidality, or no response require reassessment. [1]
The monoamine framework is also incomplete. Intranasal esketamine acts as an NMDA receptor antagonist. Its US treatment-resistant depression indication now allows monotherapy or combination with an oral antidepressant. Administration still requires supervised monitoring under its restricted program, including at least two hours of observation. The indication does not establish that it prevents suicide, and improvement after a dose does not cancel a need for hospitalization. [18]
Try it here · Checkpoint 1 of 3
Make your prediction before reading the choices. A first attempt is just a starting point.
Case 27
Show answer and explanations for case 27
A. Dopamine D2 antagonism. (Why this does not fit)
D2 blockade is associated with antipsychotic motor and prolactin effects, not the characteristic TCA antimuscarinic and orthostatic pattern.
B. Antimuscarinic and alpha-1 receptor blockade, with possible additional sedating H1 effects. (Best answer)
These off-target receptor actions explain retention, dry or blurred visual symptoms, orthostasis, and sedation.
C. Selective SERT inhibition with no receptor effects. (Why this does not fit)
That description omits the TCA actions responsible for this adverse-effect pattern.
D. MAO inhibition alone. (Why this does not fit)
Amitriptyline is not an MAOI, and its anticholinergic burden better fits the findings.
Takeaway: TCA tolerability reflects several targets beyond monoamine reuptake.
Fluoxetine, sertraline, paroxetine, citalopram, escitalopram, and fluvoxamine are SSRIs. Their shared serotonergic effects can produce nausea, diarrhea, sexual dysfunction, and early activation. Platelet serotonin effects can increase bleeding risk, especially with NSAIDs or anticoagulants. SIADH and hyponatremia are particularly relevant in susceptible older adults. A class effect does not mean equal interaction profiles, half-lives, approved indications, or dosing rules. [2]
Fluoxetine and its active metabolite persist for weeks, which reduces abrupt concentration changes but complicates interactions and switching. Fluoxetine also has specific bulimia nervosa and OCD indications; an indication belongs to the individual product, not automatically to every SSRI. Paroxetine is associated with more troublesome withdrawal and greater anticholinergic burden than many other SSRIs. Sertraline does not require dose adjustment solely for renal impairment, although liver disease changes its dosing.
Citalopram has dose-dependent QT concerns; the maximum recommended daily dose is 20 mg above age 60, in hepatic impairment, and in certain CYP2C19-related circumstances. Persistent QTc above 500 ms warrants discontinuation under its label. [9][10][2][1][23]
Venlafaxine, desvenlafaxine, and duloxetine are SNRIs. Noradrenergic effects can contribute to sweating, increased pulse, and increased blood pressure. Monitor blood pressure with venlafaxine and reconsider treatment when increases are sustained. Duloxetine can address depression with diabetic peripheral neuropathic pain, fibromyalgia, or chronic musculoskeletal pain. Its US labeling advises avoiding use in severe renal impairment with GFR below 30 mL/min and in chronic liver disease or cirrhosis. [3][4]
Milnacipran belongs to the SNRI family but its US product is indicated for fibromyalgia, not major depressive disorder. Levomilnacipran is a different drug; do not transfer indications between similar names. The presence of pain can favor a drug with a relevant indication, but kidney function, blood pressure, interactions, and the patient's preferences still determine whether it is a reasonable choice. [17]
TCAs include amitriptyline, nortriptyline, imipramine, desipramine, and clomipramine. Their serotonin and norepinephrine reuptake effects coexist with varying antimuscarinic, histamine H1, and alpha-1 blockade. Dry mouth, constipation, urinary retention, sedation, and orthostasis follow from these additional actions. Secondary amines such as nortriptyline often have a lower anticholinergic burden than amitriptyline, but are not free of it. Clomipramine's role in OCD does not erase its cardiac and anticholinergic risks. [19][26]
Use distinctive adverse effects to individualize care
Benefits and liabilities belong in the same decision
Drug
Potential fit
Reason to reconsider
DrugBupropion
Potential fitDepression when sexual adverse effects or smoking cessation are priorities
Reason to reconsiderSeizure disorder, current or prior anorexia or bulimia, or abrupt withdrawal from selected sedatives
DrugMirtazapine
Potential fitDepression with insomnia and poor appetite
Reason to reconsiderWeight gain, daytime somnolence, or a high fall burden
DrugTrazodone
Potential fitAntidepressant treatment with sedating properties
Reason to reconsiderOrthostasis, daytime impairment, or priapism
DrugVortioxetine
Potential fitAn alternative for major depression after individualized review
Reason to reconsiderNausea, serotonergic interactions, and possible sexual dysfunction
Bupropion has relatively little direct serotonergic activity and generally fewer sexual adverse effects than SSRIs. [21] Its seizure risk is dose-related, and an eating-disorder history remains relevant even after purging stops. Appropriate bupropion formulations are used for smoking cessation; off-label use for adult ADHD has trial support but does not make it an immediate stimulant substitute. [22] It can worsen insomnia or increase blood pressure and is not harmless in overdose. [5]
Mirtazapine combines alpha-2 antagonism with 5-HT2, 5-HT3, and H1 blockade. Appetite increase and sedation can be useful when those symptoms are prominent, but should be discussed as tradeoffs. It is not a guaranteed weight-restoration treatment, and a simplistic promise that a higher dose always eliminates sedation is unreliable. Compared with a strongly anticholinergic TCA, it may be a more reasonable option in a patient with urinary symptoms, while still requiring monitoring. [6]
Trazodone has serotonin reuptake inhibition and 5-HT2 antagonism, with H1 and alpha-1 effects contributing to sedation and orthostasis. Low-dose use for insomnia is off-label and is not equivalent to an adequate antidepressant regimen. A prolonged erection, particularly beyond four hours, requires emergency assessment. Vortioxetine inhibits SERT and modulates several serotonin receptors. It can still cause sexual dysfunction; neither its receptor profile nor improvement in depression establishes treatment of an independent cognitive disorder. [7][8]
Ask about sexual function before and after treatment. Consider depression itself, relationship factors, endocrine or vascular illness, and other drugs. Options may include cautious dose adjustment, switching to a drug with a different adverse-effect profile, or a targeted adjunct. Sildenafil has randomized evidence in men with remitted depression and SRI-associated sexual dysfunction, subject to its own contraindications. Choose an adjunct for the specific symptom, evidence and patient contraindications; no option is guaranteed to work. [16]
Distinguish three dangerous medication patterns
Serotonin toxicity commonly follows adding, increasing, or combining serotonergic agents. Tramadol, linezolid, and IV methylene blue are relevant interaction checks. Agitation with diaphoresis, diarrhea, hyperreflexia, and clonus after such an exposure should prompt urgent cessation of implicated agents and supportive medical treatment. Neuromuscular excitation helps distinguish this pattern from uncomplicated anxiety. Neuroleptic malignant syndrome more often follows dopamine blockade or withdrawal of dopaminergic treatment and features generalized rigidity; neither syndrome should be diagnosed by fever alone. [2][3]
Supportive treatment remains central in serotonin toxicity; severe cases may need sedation and airway support. Serotonin antagonists such as cyproheptadine have been used as adjuncts under specialist direction. An adjunct must not delay resuscitation or treatment of dangerous hyperthermia. [25]
TCAs can block fast cardiac sodium channels in overdose, causing conduction delay, wide QRS complexes, hypotension, ventricular dysrhythmias, and seizures. Sodium bicarbonate provides sodium loading and alkalinization; it does not pull sodium away from a channel. AHA recommends bicarbonate for life-threatening TCA cardiotoxicity. For persistent wide-complex dysrhythmia despite appropriate initial treatment, a class Ib agent such as lidocaine can be considered with toxicology expertise. A blanket statement that lidocaine is forbidden is incorrect. [13][19]
Phenelzine and tranylcypromine inhibit MAO and can cause a dangerous pressor response to excessive tyramine or sympathomimetic exposure. Severe headache and hypertension after a high-tyramine meal require emergency evaluation and controlled treatment, potentially including a short-acting agent such as phentolamine. This is a different mechanism from sodium-channel toxicity. The selegiline patch has dose-specific dietary rules. At 6 mg per 24 hours a modified tyramine diet is not required by its label, whereas 9 and 12 mg per 24 hours require restrictions that continue for two weeks after stopping or reducing those doses. [11][12]
Hyponatremia is another emergency pathway. Confusion or a seizure after starting an SSRI warrants sodium measurement and assessment of the cause. Severe or moderately severe symptoms attributable to hyponatremia require urgent monitored hypertonic saline treatment, irrespective of an arbitrary sodium cutoff such as 120 mmol/L. Stop the suspected causative drug as appropriate, prevent overly rapid correction, and distinguish SIADH from hypovolemia before choosing ongoing fluid treatment. [14]
Try it here · Checkpoint 2 of 3
Make your prediction before reading the choices. A first attempt is just a starting point.
Case 14
Show answer and explanations for case 14
A. Give procainamide for the persistent wide-complex dysrhythmia. (Why this does not fit)
Class Ia agents can aggravate the sodium-channel conduction toxicity.
B. Give propranolol to slow the ventricular rhythm. (Why this does not fit)
Beta blockade is not the recommended rescue here and can worsen cardiovascular compromise.
C. Consider lidocaine in this refractory setting. (Best answer)
A class Ib antiarrhythmic can be considered for persistent dysrhythmia from sodium-channel-blocker poisoning after initial treatment.
D. Use magnesium as the only next antiarrhythmic despite no torsades or prolonged-QT pattern. (Why this does not fit)
Magnesium is not the targeted rescue for this refractory TCA sodium-channel dysrhythmia; rhythm-specific toxicology management is needed.
Takeaway: Drug-class names alone do not determine an antidotal or rescue role.
Before prescribing, assess suicide risk, prior treatment response, bipolar symptoms, substance use, interactions, and relevant medical illness. Review an antidepressant usually within two weeks, and within one week for a person aged 18 to 25 or when suicide risk is a particular concern. Early intense restlessness, new suicidal thoughts, reduced need for sleep, or expansive behavior require active reassessment, not automatic reassurance that activation always settles. [1][9]
If there is no response after four weeks at a recognized therapeutic dose, check adherence, tolerability, diagnosis, ongoing stressors, and comorbid illness before deciding among further treatment options. A partial response differs from no response. Dose adjustment within recommended limits, a switch, psychotherapy, or specialist augmentation may be appropriate. After remission, treatment often continues for at least six months, with longer prevention based on relapse risk and preference. [1]
Withdrawal can cause flu-like symptoms, insomnia, nausea, imbalance, sensory disturbances such as electric-shock sensations, and hyperarousal. These may begin soon after a missed dose or rapid reduction, particularly with paroxetine or venlafaxine. Relapse is not the only explanation for distress after stopping medication. Taper in agreed stages, often using smaller proportional reductions as the dose falls, and adapt to symptoms. Fluoxetine's long persistence can delay withdrawal rather than making it impossible.
Washout is directional and drug-specific
Sequence
Label information
Practical interpretation
SequenceFluoxetine to an MAOI
Label informationAt least five weeks
Practical interpretationActive drug and metabolite persist after the final dose.
SequenceMAOI to fluoxetine
Label informationAt least 14 days
Practical interpretationAllow enzyme activity to recover after an irreversible MAOI.
SequenceVenlafaxine to an MAOI
Label informationEffexor XR specifies at least seven days
Practical interpretationAlso check the receiving MAOI label, which may require longer.
SequencePhenelzine and serotonergic antidepressant switches
Label informationNardil generally specifies at least 14 days, with the longer fluoxetine exception
Practical interpretationFollow both product labels and the individualized specialist plan.
A washout interval is not a taper schedule, and a taper does not eliminate an incompatible overlap. Do not reduce these distinctions to a universal two-week rule. In urgent infections requiring linezolid, coordinate with the treating service and pharmacy rather than simply adding it or delaying essential antimicrobial care without a plan. [9][3][11]
Reassess the diagnosis and the treatment setting
Pregnancy does not make untreated depression benign, and it does not automatically require stopping or switching an effective medication. Review severity, relapse history, prior response, gestational timing, medication-specific risks, and patient preferences. ACOG advises against withholding or discontinuing treatment solely because of pregnancy or lactation. A stable patient taking paroxetine needs individualized counseling rather than a forced abrupt switch; no drug should be described as having zero fetal or neonatal risk.
Formulation also matters. Zoloft oral solution contains 12% alcohol and its label does not recommend that solution during pregnancy; this is distinct from whether sertraline itself remains an appropriate treatment. [2][15]
In an older adult, consider falls, anticholinergic burden, sodium, QT risk, bleeding, renal function, and the rest of the medication list. Renal dosing is product-specific. Sertraline's renal labeling does not mean every other SSRI is prohibited in kidney disease. Emotional blunting likewise needs a differential diagnosis, including residual depression and medication effects, before a dose change or switch. Do not promise that changing to one named drug restores a particular feeling.
Depression with psychotic symptoms often warrants specialist care and combined antidepressant and antipsychotic treatment. ECT can be considered when a rapid response is needed, the illness is severe or life-threatening, or other treatments have failed. A patient with a history of mania requires a bipolar treatment plan; routine antidepressant monotherapy is not an appropriate default for bipolar I depression. New mania during antidepressant treatment requires reassessment and antimanic treatment. [1][20]
For depression with panic or generalized anxiety, select a treatment with evidence for the actual disorders and start cautiously when activation is a concern. NICE does not recommend benzodiazepines for panic disorder and limits their GAD use to short-term crises; bridging is not compulsory. Atypical depressive features do not automatically mandate an MAOI. Specialist options for inadequate response include psychological treatment, switching and selected augmentation such as lithium or an antipsychotic after reviewing risks. [1][24]
Esketamine's treatment-resistant depression indication permits monotherapy or an oral-antidepressant combination. Its separate indication for depressive symptoms in adults with MDD and acute suicidal ideation or behavior still requires an oral antidepressant. It has not been shown to prevent suicide and does not replace hospitalization when indicated. [18]
Try it here · Checkpoint 3 of 3
Make your prediction before reading the choices. A first attempt is just a starting point.
Case 6
Show answer and explanations for case 6
A. Continue the same prescription without discussing the pregnancy or prior relapses. (Why this does not fit)
Continuing treatment may be appropriate, but informed review of medication-specific and untreated-illness risks is still needed.
B. Automatically cross-taper to sertraline based only on the positive pregnancy test. (Why this does not fit)
A switch may be considered in some circumstances, but a stable patient with recurrent illness needs individualized evaluation rather than an automatic switch.
C. Review individualized maternal and fetal risks and prior response before agreeing on a treatment plan. (Best answer)
Pregnancy alone does not require stopping an effective treatment, and the relapse history is directly relevant.
D. Stop paroxetine today and arrange review after the first trimester. (Why this does not fit)
Abrupt cessation can cause withdrawal and relapse and leaves the current treatment decision unaddressed.
Takeaway: Pregnancy decisions require individualized counseling rather than a forced switch.
These original educational scenarios test selection, adverse effects, interactions, and follow-up. Each has one best answer based on the information provided.
Case 1
Show answer and explanations for case 1
A. Clomipramine. (Why this does not fit)
This serotonergic TCA is not a particularly suitable choice for avoiding sexual adverse effects and adds anticholinergic and cardiac burdens.
B. Paroxetine. (Why this does not fit)
Changing to another SSRI, particularly one with a substantial sexual-adverse-effect burden, does not best fit this patient preference to reduce the effect.
C. A higher SSRI dose solely to treat the sexual symptoms. (Why this does not fit)
Increasing serotonergic exposure is not a dependable treatment for an adverse effect that began on the SSRI.
D. Bupropion. (Best answer)
Its different reuptake profile generally carries fewer sexual adverse effects, making it a reasonable switch to discuss with a taper and monitoring plan.
Takeaway: Use the patient's treatment preference and adverse-effect pattern to guide a safe change.
A. IV lidocaine as the first treatment before bicarbonate. (Why this does not fit)
Lidocaine can be considered for refractory dysrhythmia after initial treatment; bicarbonate is the first targeted therapy for life-threatening TCA cardiotoxicity.
B. IV sodium bicarbonate with resuscitative monitoring. (Best answer)
Sodium loading and alkalinization counter TCA sodium-channel toxicity in this wide-QRS, hypotensive presentation.
C. IV procainamide. (Why this does not fit)
Additional class Ia sodium-channel blockade can worsen the TCA-associated conduction disturbance.
D. IV magnesium alone to correct the wide QRS. (Why this does not fit)
Magnesium is used for torsades in an appropriate long-QT context; it does not replace bicarbonate for TCA sodium-channel cardiotoxicity.
Takeaway: Wide QRS and hypotension after TCA overdose require urgent bicarbonate-based treatment.
A. A cross-taper eliminates the need for washout. (Why this does not fit)
Overlapping incompatible serotonergic treatment can cause severe toxicity.
B. The washout can be counted from the first taper reduction while full doses continue. (Why this does not fit)
Washout follows discontinuation of the relevant drug, not merely the beginning of a taper.
C. Effexor XR specifies at least seven days after stopping, but Nardil generally requires at least 14 days after serotonergic agents; follow both labels. (Best answer)
The apparent conflict reflects the departing and receiving product requirements, so the shorter interval alone should not dictate this switch.
D. Seven days is always adequate for every antidepressant-to-MAOI switch. (Why this does not fit)
Fluoxetine and receiving-product requirements make a universal seven-day rule incorrect.
Takeaway: Check both ends of an antidepressant switch.
A. Delay all antimicrobial treatment for a full antidepressant washout without considering infection urgency. (Why this does not fit)
A serious infection may require immediate treatment. Coordinate antibiotic selection and the antidepressant plan rather than delaying essential care automatically.
B. Coordinate the urgent antibiotic and antidepressant plan with the treating team. (Best answer)
Linezolid has MAOI activity; the need for effective infection treatment and serotonin-toxicity prevention must be addressed together.
C. Continue sertraline with linezolid and defer interaction review until routine follow-up. (Why this does not fit)
Linezolid inhibits MAO and can interact with sertraline. The plan and monitoring must be addressed before or during urgent initiation.
D. Replace sertraline with fluoxetine immediately to eliminate the interaction. (Why this does not fit)
Fluoxetine is also serotonergic and has prolonged persistence; this does not solve the interaction.
Takeaway: Medication reconciliation must include nonpsychiatric serotonergic interactions.
A. Testosterone despite normal testosterone testing. (Why this does not fit)
Hormone replacement is not appropriate without a supported deficiency; the trial-supported symptom-targeted option here is sildenafil.
B. Aripiprazole to augment the remitted depression. (Why this does not fit)
Augmentation for inadequately treated depression is a different indication; this patient needs assessment and treatment of erectile difficulty.
C. Increase sertraline to treat the erectile symptom as residual depression. (Why this does not fit)
Depression is in remission and the erectile difficulty began on treatment. Increasing the suspected drug is not the symptom-targeted adjunct supported by the cited trial.
D. Sildenafil. (Best answer)
A randomized placebo-controlled trial in men with remitted depression and SRI-associated sexual dysfunction supports benefit. This does not guarantee improvement in every sexual symptom or population.
Takeaway: Match the intervention and its evidence to the specific sexual symptom and patient.
A. Arrange an alcohol-free formulation plan while preserving effective depression care. (Best answer)
The product label does not recommend the alcohol-containing oral solution during pregnancy; that does not itself require abandoning sertraline treatment.
B. Use the solution unchanged because only the active ingredient matters in pregnancy. (Why this does not fit)
The alcohol-containing formulation creates a separate labeled concern even though the active drug is familiar.
C. Dilute the measured solution and describe it as alcohol-free. (Why this does not fit)
Dilution changes concentration but does not eliminate the amount of alcohol in the administered dose.
D. Abruptly stop all antidepressant treatment solely because this formulation is unsuitable. (Why this does not fit)
A formulation problem should prompt a coordinated alternative, not an automatic interruption of effective care.
Takeaway: Formulation ingredients can change prescribing even when the active drug remains appropriate.
A. The acute-suicidality MDD indication also permits monotherapy in the same way as TRD. (Why this does not fit)
The separate acute-suicidality depressive-symptom indication still requires an oral antidepressant.
B. TRD treatment may be monotherapy or combined with an oral antidepressant, with supervised administration and at least two hours of monitoring. (Best answer)
The 2025 expansion permits monotherapy for TRD while retaining the restricted monitored setting.
C. Observation can routinely end after 30 minutes if dissociation has resolved. (Why this does not fit)
The label requires at least two hours of monitoring and assessment of clinical stability before discharge.
D. Its approval proves that it prevents suicide. (Why this does not fit)
The label specifically states that suicide-prevention effectiveness has not been demonstrated.
Takeaway: A broader indication does not remove monitoring requirements or establish suicide prevention.