What it is:A catch-all for any heartbeat that's too fast, too slow, or too irregular. Split by where it starts: supraventricular(above the ventricles: AFib, atrial flutter, PSVT) versus ventricular(PVCs, VT, VF), plus a separate bucket of bradyarrhythmias(too slow, or blocked conduction).
The core problem:Either the heart's electrical system fires when it shouldn't (abnormal automaticity, triggered activity) or an impulse takes a wrong loop back on itself instead of dying out (reentry). Every antiarrhythmic works by slowing conduction, prolonging the refractory period, or both, in whichever tissue is misbehaving.
What you do about it:For AFib, work the ABCs: Anticoagulate to avoid stroke, get Better symptom control (rate or rhythm), optimize Comorbidities. For ventricular arrhythmias, think devices (ICD) before drugs, because antiarrhythmic drugs (AADs) have never beaten a defibrillator at preventing sudden death.
AAD use has been quietly declining for two decades. Big trials showed these drugs can increase mortalitythrough their own proarrhythmic side effects, while offering only modest efficacy. That's why ablation and ICDs have taken over so much of this chapter's real estate. When you see an AAD ordered, ask what nonpharmacologic option was considered first.
The normal circuit: SA node(60–100 bpm, the boss) → atrial myocardium → AV node(40–60 bpm, the backup and the gatekeeper) → Bundle of His → right/left bundle branches → Purkinje fibers → ventricles. Anything upstream of the AV node is "supraventricular." Anything at or below it is "ventricular."
| EKG piece | What it represents |
|---|---|
| P wave | Atrial depolarization |
| PR interval | Time from SA node firing to ventricular depolarization (conduction delay through the AV node) |
| QRS complex | Ventricular depolarization. Narrow = above the AV node. Wide = ventricular in origin or aberrantly conducted. |
| QT interval | Full ventricular depolarization + repolarization. The interval that gets you in trouble with drugs. |
| T wave | Ventricular repolarization |
At the cellular level, myocardial cells fire because of fast sodium influx (phase 0), then a plateau carried by calcium influx (phase 2) that holds the contraction, then potassium efflux resets things (phase 3). Nodal tissue (SA, AV) skips the fast sodium phase entirely and depends on a slow "funny current" (If) and calcium for its upstroke, which is exactly why non-DHP calcium channel blockers and beta blockers are so effective there and largely useless against fast-sodium-driven ventricular tissue.
Arrhythmias come from only two mechanisms: abnormal impulse formation(a cell fires when it shouldn't, i.e. enhanced automaticity or triggered activity from after-depolarizations) or abnormal impulse conduction(reentry: an impulse finds a loop with unidirectional block and just keeps circling). AFib, flutter, AVNRT, and VF are reentry problems. Sinus tach and some VT are automaticity problems. Torsades is a triggered-activity problem (early after-depolarizations from a prolonged action potential). Knowing which bucket a rhythm is in tells you why a given drug class helps.
Almost every AAD question on an exam is really asking "which class, and why does that class do what it does." Learn this table once and the rest of the chapter is just filling in drug names.
| Class | Channel blocked | Conduction velocity | Refractory period | Used for |
|---|---|---|---|---|
| Ia(quinidine, procainamide, disopyramide) | Na⁺, intermediate on/off | ↓ | ↑ | Both atrial & ventricular; niche use now, real toxicity |
| Ib(lidocaine, mexiletine) | Na⁺, fast on/off | 0/↓ | ↓ | Ventricular only; safe-ish in structural heart disease |
| Ic(flecainide, propafenone) | Na⁺, slow on/off | ↓↓ | 0 | Atrial (great for pharmacologic cardioversion); avoid in structural heart disease |
| II(beta blockers) | Indirect (anti-adrenergic) | ↓ | ↑ | Nodal tissue: rate control, suppress automaticity |
| III(amiodarone, dofetilide, dronedarone, sotalol, ibutilide) | K⁺ | 0 | ↑↑ | Both; prolongs QTc, so this is the class that causes torsades |
| IV(verapamil, diltiazem) | Ca²⁺ (non-DHP) | ↓ | ↑ | SA/AV nodal arrhythmias; contraindicated in HFrEF |
Narrow QRS (atrial) arrhythmias respond to classes Ic, II, III, IV.Wide QRS (ventricular) arrhythmias respond to classes Ib, Ia, II, III.Notice class Ib only shows up on the ventricular side (lidocaine won't touch AFib) and class Ic is your go-to for atrial rhythms but is dangerous in anyone with structural heart disease because of proarrhythmia risk.
Amiodarone has properties of all four classes: it's a sodium channel blocker with fast on/off kinetics, a nonselective beta blocker, a potassium channel blocker, and a mild calcium channel blocker. That's why it works almost everywhere and also why its side-effect profile touches the thyroid, lungs, liver, eyes, and skin. Its half-life is genuinely wild: 26–107 days, because it's so lipophilic it loads into fat and slowly leaches back out.
AFib is chaotic atrial firing at 400–600 beats/minwith total loss of the atrial kick (the ~15–20% contribution atrial contraction normally adds to ventricular filling). The AV node can't conduct most of those impulses through, so what actually reaches the ventricle is irregular and usually 90–170 bpm. On the monitor: irregularly irregular, no discernible P waves. Atrial flutter is the more organized cousin: a single reentrant circuit firing 270–330/min with regular conduction, classically producing a ventricular rate that's a clean fraction of 300 (150, 100, 75).
Symptoms: palpitations, rapid heart rate, dyspnea, dizziness, fatigue, chest discomfort. Neither AFib nor flutter is usually immediately life-threatening on its own, the two emergencies are AFib with severe HF decompensation (pulmonary edema, hypotension) or AFib complicating an acute MI.
Reentrant loops involving the AV node (AVNRT), an accessory pathway (AVRT, think Wolff-Parkinson-White), or less commonly the SA node. Classic story: sudden-onset, sudden-offset rapid heart rate/palpitations, often with no trigger. EKG shows a narrow, regular, rapid QRS tachycardia where the P wave is hiding inside the QRS or T wave because atrial and ventricular depolarization happen almost simultaneously in the reentrant loop. Life-threatening symptoms (syncope, hemodynamic collapse) point to either an extremely fast rate (>200 bpm) or AFib conducting down an accessory pathway.
In WPW, a premature atrial impulse blocks in the accessory pathway (bundle of Kent), sneaks back into the atrium, then re-enters the ventricle through the AV node and becomes self-sustaining, giving a short PR interval and a slurred QRS upstroke (delta wave). If that patient develops AFib, the accessory pathway can conduct the atrial chaos straight to the ventricle without the AV node's protective filtering, producing dangerously fast ventricular rates. This is why AV-nodal blockers (digoxin, non-DHP CCBs, sometimes beta blockers) are avoided if WPW plus AFib is suspected, they can paradoxically favor conduction down the accessory pathway.
Organize the whole workup around three sequential goals, sometimes written as the newer AF-CAREpathway: Comorbidity/risk factor management, Avoid stroke, Reduce symptoms via rate or rhythm control, Evaluate/reassess over time. Older material calls it "ABC": Anticoagulation, Better symptom management, Comorbidity optimization. Same idea, don't let the relabeling throw you.
Every new AFib diagnosis gets a stroke-risk conversation before anything else. See Anticoagulationbelow for the scoring systems and drug selection, it's dense enough to earn its own section.
Big trials (AFFIRM and friends) found rhythm control offers no mortality advantage over rate controlin most patients. So the choice comes down to symptoms and quality of life, not survival, unless the patient has new HF from AFib itself (in which case rhythm control tends to be favored) or remains symptomatic despite good rate control.
| Rate control | Rhythm control | |
|---|---|---|
| Strategy | Leave the patient in AFib, just slow the ventricular response | Restore and maintain sinus rhythm |
| Target | Resting HR <110 if asymptomatic + LVEF >40%; <80 if symptomatic or LVEF ≤40% | N/A |
| Preferred first-line agents | Beta blocker or non-DHP CCB (diltiazem/verapamil) if LVEF preserved; beta blocker or digoxin if LVEF ≤40% (avoid CCBs, negative inotropy) | Class Ic (flecainide, propafenone), class III (amiodarone, dofetilide, dronedarone, sotalol) |
| Escalation | Add amiodarone if BB/CCB/digoxin can't control rate at rest and exercise (but watch for it converting them back to sinus, raising thromboembolism risk if not anticoagulated) | DCCV if drugs fail or symptoms are severe; catheter ablation for recurrent/refractory disease |
Non-DHP CCBs (verapamil, diltiazem) are negative inotropes, they slow the heart by directly reducing contractility. Fine in a preserved-EF heart, dangerous in a heart that's already failing to squeeze. In HFrEF, rate control defaults to a beta blocker or digoxin instead.
Restoring sinus rhythm is when strokes actually happen, because a poorly adherent atrial thrombus can get dislodged the moment the atrium starts contracting normally again. The rule:
| Direct current cardioversion (DCCV) | Pharmacologic cardioversion | |
|---|---|---|
| Success rate | 80–90%, higher than drugs | Lower, most effective within 7 days of AFib onset |
| Downside | Needs periprocedural sedation; small risk of sinus arrest or ventricular arrhythmia | Risk of drug-induced torsades, drug interactions |
| Best drugs | - | Ibutilide, dofetilide (class III pure IKrblockers), flecainide/propafenone (class Ic), amiodarone (oral or IV) |
A single oral loading dose of flecainide (200 mg if <70 kg, 300 mg if >70 kg)or propafenone (450 mg if <70 kg, 600 mg if >70 kg)lets a select, structurally-normal-heart patient self-terminate a recent-onset AFib episode at home. Only offered after the same regimen was proven safe and effective for that specific patient in a monitored inpatient setting first, this is not a first-time-use-at-home drug.
Acute termination is a stepladder: vagal maneuvers(Valsalva, carotid massage) → IV adenosine(transiently blocks the AV node completely) → if hemodynamically unstable, go straight to synchronized DCCV. If stable but vagal maneuvers/adenosine fail, try an IV beta blocker or non-DHP CCB; if AVNRT still won't break, IV amiodarone. AVRT that fails first-line measures goes to synchronized cardioversion.
Long-term preventiononce reserved for AADs is now dominated by catheter ablation, which is curative, low-complication, and eliminates the need for chronic drug therapy, making it preferred over chronic AAD use for symptomatic PSVT. When drugs are used, they're AV-nodal blockers (digoxin, beta blockers, non-DHP CCBs) or class Ic agents (flecainide, propafenone).
Premature ventricular complexes are common and usually benign in a structurally normal heart, no treatment needed even if the patient feels the "skipped beat" (which is actually the compensatory pause and next sinus beat they're noticing, not the PVC itself). In patients with structural heart disease, especially multifocal PVCs or couplets, they signal higher risk of sudden cardiac death. Post-MI, beta blockers reduce mortality and suppress PVCs as a side benefit; AADs should never be used just to chase away asymptomatic PVCs.
VT = 3+ consecutive PVCs at >100 bpm, wide QRS. Sustained(>30 sec or needs intervention) vs nonsustained(self-terminates <30 sec) vs incessant(more often than sinus rhythm, dominates the tracing). Monomorphic(one consistent QRS shape, usually scar-related reentry around an old MI) vs polymorphic(varying QRS morphology, often ischemia or a repolarization abnormality). Acute causes: electrolyte derangement (hypokalemia, hypomagnesemia), hypoxia, digoxin toxicity, or most commonly, active ischemia/MI complicated by HF.
| Presentation | Management |
|---|---|
| Stable monomorphic VT, structural heart disease | Cardioversion + consider IV procainamide, amiodarone, or sotalol |
| Stable monomorphic VT, no structural heart disease | Verapamil or beta blocker first-line |
| Pulseless VT / VF | ACLS immediately: defibrillation |
For patients who survive sustained VT or cardiac arrest (secondary prevention) andfor certain high-risk patients before a first event (primary prevention), ICDs are superior to AADsfor preventing sudden cardiac death. AADs may still get used adjunctively to reduce how often the ICD has to fire, but they are not a substitute for the device.
A polymorphic VT where the QRS complexes appear to twist around the baseline, caused by delayed repolarization from potassium channel blockade, i.e. a prolonged QT. Clinically significant prolongation is defined as QTc >500 msec, or an increase of >60–70 msec from baselineafter starting a drug, that's your trigger to stop or reduce the offending agent.
| Step | Action |
|---|---|
| 1 | Defibrillate if the patient is unstable (TdP often recurs rapidly even after successful shock) |
| 2 | IV magnesium sulfate, drug of choice to prevent recurrence, even if serum magnesium is normal |
| 3 | If magnesium fails: increase heart rate to shorten repolarization, temporary pacing at 105–120 bpm, or isoproterenol |
| 4 | Stop every QT-prolonging drug; correct hypokalemia and hypomagnesemia |
Lidocaine is usually ineffective. IV procainamide is absolutely contraindicated, it's a class Ia agent that further prolongs repolarization and will make TdP worse, not better.
The AAD itself creates or worsens an arrhythmia. Two named patterns worth knowing cold:
Sinus bradycardia (HR <60) is common and usually benign, especially in fit young people, and doesn't need treatment if asymptomatic. Sick sinus syndromeis diffuse SA nodal disease from aging or structural disease, and can alternate with paroxysmal tachycardias (AFib) as tachy-brady syndrome. Drugs that treat the tachy side (AV nodal blockers) have to be used cautiously here since they can worsen the brady side, unless the patient already has a functioning pacemaker.
| Degree | Pattern |
|---|---|
| First | 1:1 conduction, just a prolonged PR interval |
| Second, Mobitz I (Wenckebach) | PR progressively lengthens until a beat drops. Usually benign, at the AV node. |
| Second, Mobitz II | Beats drop suddenly and unpredictably, no progressive PR lengthening. More concerning, often infranodal. |
| Third (complete) | Total AV dissociation, atria and ventricles beat independently |
Common drug culprits:beta blockers, digoxin, and non-DHP CCBs cause AV block primarily at the AV node itself; class I AADs can worsen conduction delay below the node.
| Scenario | Management |
|---|---|
| Symptomatic 2nd/3rd degree block (hypotension, altered mental status, chest pain) | Atropine 0.5 mg IV q3–5min, up to 3 mg total → transcutaneous pacing if unresponsive → epinephrine 2–10 mcg/min or dopamine 2–10 mcg/kg/min infusion |
| Block below the AV node (Mobitz II, trifascicular) | Atropine usually doesn't work, escalate straight to pacing |
| Chronic symptomatic block | Permanent pacemaker |
Vasovagal syncope:education on trigger avoidance (prolonged standing, heat) and counter-pressure maneuvers (squatting, lying down) first. Then salt/fluid loading. Drug of choice is midodrine, an alpha-agonist that limits venous pooling. Beta blockers are an option in patients ≥42 years old. Fludrocortisone, anticholinergics, and SSRIs are other described options.
Oral maintenance doses for the class I and III AADs you'll actually be asked about, plus IV dosing for acute rhythm emergencies.
| Drug | Oral dose | Dose-adjust for |
|---|---|---|
| Class Ia | ||
| Disopyramide | IR 100–150 mg q6h; CR 200–300 mg q12h | Hepatic, renal |
| Quinidine | 200–300 mg sulfate q6h, or 324–648 mg gluconate q8–12h | Hepatic |
| Class Ib | ||
| Mexiletine | 200–300 mg q8h | Hepatic |
| Class Ic | ||
| Flecainide | 50–200 mg q12h | Hepatic, renal |
| Propafenone | 150–300 mg q8h (IR) or 225–425 mg q12h (SR) | Hepatic |
| Class III | ||
| Amiodarone | 400 mg 1–3x/day until 10 g total load, then 200 mg/daymaintenance for AFib (300–400 mg/day for ventricular) | None |
| Dofetilide | Dosed strictly by CrCl: 500 mcg BID (CrCl >60), 250 mcg BID (40–60), 125 mcg BID (20–40); avoid if CrCl <20 | Renal - must be initiated inpatient with QTc monitoring |
| Dronedarone | 400 mg BID with meals | Hepatic; avoid in severe hepatic impairment |
| Sotalol | 80–160 mg q12h (max 320 mg/day atrial use) | Hepatic; must be initiated inpatient |
| Drug | Clinical situation | IV dose |
|---|---|---|
| Amiodarone | Pulseless VT/VF | 300 mg IV/IO push (may repeat 150 mg), then 1 mg/min ×6h, then 0.5 mg/min ×18h |
| Amiodarone | Stable VT or AFib termination | 150 mg IV over 10 min, then same taper as above |
| Amiodarone | AFib rate control | 300 mg IV over 1 h, then 10–50 mg/h ×24h |
| Diltiazem | PSVT / AFib rate control | 0.25 mg/kg IV over 2 min (may repeat 0.35 mg/kg), then 5–15 mg/h infusion |
| Verapamil | PSVT / AFib rate control | 2.5–5 mg IV over 2 min (max cumulative 20 mg), then 2.5–10 mg/h infusion |
| Ibutilide | AFib termination | 1 mg IV over 10 min, may repeat once after 10 min |
| Lidocaine | Pulseless VT/VF | 1–1.5 mg/kg IV/IO push (repeat 0.5–0.75 mg/kg q5–10min, max 3 mg/kg), then 1–4 mg/min (1–2 mg/min if liver disease/HF) |
| Procainamide | AFib termination, stable VT | 15–18 mg/kg IV over 60 min, then 1–4 mg/min infusion |
All three subclasses block the fast Na⁺ current responsible for phase 0 depolarization, they just differ in on/off kinetics, which is what determines their conduction and refractoriness effects.
Ia (intermediate on/off):quinidine, procainamide, disopyramide slow conduction andprolong refractoriness, widening the QRS and prolonging the QTc. Effective in both atrial and ventricular arrhythmias but relegated to niche/2nd-3rd line use now. Quinidine causes cinchonism (tinnitus, headache, GI upset) and is a CYP2D6 inhibitor. Procainamide's active metabolite NAPA carries more potassium-blocking effect than the parent drug. Disopyramide brings prominent anticholinergic effects (dry mouth, urinary retention, blurred vision) plus significant negative inotropy, avoid in HF.
Ib (fast on/off):lidocaine and mexiletine act almost exclusively on depolarized (ischemic/diseased) ventricular tissue and barely touch normal atrial tissue, so they only work for ventricular arrhythmias. Considered relatively safe in structural heart disease compared to the other subclasses. CNS effects (dizziness, tremor, confusion, at high levels seizures) are the dominant toxicity.
Ic (slow on/off):flecainide and propafenone hit conduction velocity hard while barely touching refractoriness, prolonging the PR and widening the QRS. Highly effective for atrial arrhythmias and pharmacologic cardioversion.
Class Ic drugs are contraindicated in structural heart disease, including CAD and reduced EF. Landmark trials found flecainide-class drugs used to suppress PVCs after MI actually increased mortalitythrough proarrhythmia. This is the historical reason AADs fell out of favor broadly and why "any structural heart disease" is a near-automatic exclusion for class Ic on an exam question.
Both classes work on SA/AV nodal tissue where the action potential upstroke depends on calcium, not fast sodium, so they slow conduction and prolong refractoriness specifically where reentrant SVTs and rate control of AFib live. Beta blockers add antiadrenergic suppression of automaticity; only propranolol among beta blockers has any real cardioversion capability, the rest are purely rate-control tools.
Non-DHP CCBs (verapamil, diltiazem) are effective for PSVT and AFib rate control but are contraindicated in HFrEFbecause of negative inotropy, and should not be combined with a beta blocker in a rate-control regimen without real caution, the additive AV nodal blockade risks symptomatic bradycardia or heart block.
| Amiodarone | Sotalol | Dofetilide | Dronedarone | Ibutilide | |
|---|---|---|---|---|---|
| Other channels hit | Na⁺, Ca²⁺, β1 | β1 | None (clean K⁺ blocker) | Na⁺, Ca²⁺, β1 | Na⁺ |
| Clearance | CYP3A4/2C8, active metabolite | Renal | Renal | CYP3A4 | Renal |
| Use in HF? | Yes | Maybe | Yes | No | No |
| Signature risk | Thyroid, pulmonary fibrosis, hepatotoxicity, optic neuropathy | TdP, bradycardia, bronchospasm | TdP - must load inpatient | Hepatotoxicity, worsens HF | TdP - cardioversion only, supervised |
Amiodaroneis the most effective AAD for maintaining sinus rhythm, but its multi-organ toxicity means it's reserved for when other options fail or are contraindicated. Its extreme lipophilicity gives it a 26–107 day half-life, so drug interactions and toxicity can show up (or persist) long after a dose change.
Dronedaronewas designed as a "safer amiodarone" (no iodine, so no thyroid toxicity), but it also lost potency and picked up a heart failure warning, it's specifically contraindicated in HFafter a trial showed excess mortality.
Dofetilide and sotalolboth require inpatient initiationwith continuous QTc monitoring because of torsades risk; dofetilide dosing is driven entirely by CrCl.
Ibutilideis IV-only, used strictly for acute pharmacologic cardioversion of AFib/flutter in a monitored setting, never a chronic maintenance drug.
Digoxin doesn't fit Vaughan Williams (some course material tags it "class V" along with magnesium and adenosine). It increases vagal (parasympathetic) tone and indirectly suppresses AV nodal conduction, making it useful for rate control, especially at rest, but it does nothing to control rate during exertion because sympathetic tone during exercise overrides its vagal mechanism. That's exactly why it's never used as monotherapy for rate control, it gets paired with a beta blocker or non-DHP CCB.
A patient on metoprolol tartrate(immediate-release, twice daily) for AFib/HF had "controlled" resting heart rates in clinic but developed palpitations and a racing heart during exercise, because IR metoprolol's peaks and troughs leave the patient under-blocked between doses, and exercise exposes that gap. Fix: switch to metoprolol succinate(long-acting), which smooths the plasma concentration curve and gives steadier beta-blockade around the clock. This is a favorite distractor: tartrate and succinate are chemically related but not interchangeable in HF/AFib dosing regimens.
Digoxin is renally cleared and P-glycoprotein is its main transporter, so verapamil (inhibits P-gp) raises digoxin levels while rifampin (induces P-gp) lowers them. Toxicity can be reversed with digoxin immune Fab (DigiFab) in severe cases.
Before touching rate or rhythm, every AFib patient gets risk-stratified for stroke. This is arguably the highest-yield part of the whole chapter.
| Risk factor | Points |
|---|---|
| Congestive heart failure | 1 |
| Hypertension | 1 |
| Age ≥75 | 2 |
| Diabetes | 1 |
| Stroke / TIA / systemic embolism history | 2 |
| Vascular disease (PAD, prior MI, aortic plaque) | 1 |
| Age 65–74 | 1 |
| Sex category (female) | 1 |
Score 0 (male) or 1 (female, from sex category alone):no anticoagulation. Score 1 (male) or 2 (female):consider a DOAC (preferred) or warfarin, clinical judgment call. Score ≥2 (male) or ≥3 (female):anticoagulation recommended. The female sex point is why the male and female thresholds shift by exactly one, it isn't scored as an independent risk factor on its own without at least one other point present.
Hypertension (uncontrolled), abnormal renal/hepatic function (1–2 pts), stroke history, bleeding history/predisposition, labile INRs, elderly (>65), drugs (aspirin/NSAIDs) or alcohol (1–2 pts). A score ≥3 is high bleed risk.
The tempting-but-wrong approach is "high CHA₂DS₂-VASc, low HAS-BLED → anticoagulate; both high → don't." That's not how it works in practice. Most HAS-BLED risk factors are modifiable: control the blood pressure, stop the NSAID, address the labile INR. High bleed risk is a prompt to fix what's fixable and choose the safest agent, not a reason to withhold anticoagulation from a patient who needs stroke prevention.
DOACs (apixaban, rivaroxaban, dabigatran, edoxaban) are preferred over warfarin for most patients. If warfarin is used, target INR 2–3with time-in-therapeutic-range ideally >70%.
| Situation | Preferred agent | Why |
|---|---|---|
| Mechanical heart valve | Warfarin only | DOACs are contraindicated, proven harmful in this population |
| Hepatic disease | Warfarin, growing evidence for apixaban | DOACs generally okay above Child-Pugh B/C |
| CrCl <25–30 | Apixaban or warfarin | Apixaban is the only DOAC with real evidence at this level of renal impairment |
| History of GI bleed | Apixaban | Lowest GI bleed risk of the DOACs |
| Age >90 | Apixaban first, then other DOACs | Lowest bleeding risk vs warfarin |
| Poor medication adherence | Warfarin | INR checks catch nonadherence; missed DOAC doses are silent |
| Poor INR control / low TTR | DOAC | Removes the monitoring-dependence problem entirely |
Aspirin with or without clopidogrel is not effectivefor AFib-related stroke prevention and just adds bleeding risk without the benefit. If a patient truly cannot take any anticoagulant, that's a left atrial appendage occlusion device conversation, not an aspirin substitution.
Warfarin monitoring quick reference:INR <1.5 → increase weekly dose 10–20%, recheck 4–8 days. INR 1.5–1.9 → increase 5–10%, recheck 7–14 days. INR 2–3 → no change, extend recheck interval. INR 3.1–3.9 → decrease 5–10%. INR 4–4.9 → hold 0–2 doses, decrease 10–20%. INR ≥5 → assess bleed risk directly, don't just adjust the dose on paper.
| Parameter | When | Watching for |
|---|---|---|
| Heart rate / rhythm | Every visit; continuous if inpatient or on a new AAD | Adequate rate control at rest and exertion; recurrence |
| QTc interval | Baseline and after starting/titrating any class Ia or III agent | QTc >500 msec or Δ >60–70 msec from baseline → hold or reduce the drug |
| Electrolytes (K⁺, Mg²⁺) | Regularly, especially on diuretics or any QT-prolonging drug | Hypokalemia/hypomagnesemia lower the torsades threshold |
| INR (if warfarin) | Per the titration schedule above, then monthly once stable | Staying in the 2–3 therapeutic window |
| Renal function | Baseline, then periodically, always before dofetilide dosing | DOAC and dofetilide dose adjustments; digoxin accumulation |
| Amiodarone-specific labs | TFTs and LFTs every 6 months; annual ophtho exam; baseline + periodic chest X-ray/PFTs | Thyroid dysfunction, hepatotoxicity, pulmonary fibrosis, corneal deposits |
| Digoxin level | Steady state (~1 week), with any dose or renal function change | Target <2 ng/mL for rate control; toxicity signs (visual changes, nausea, arrhythmia) matter more than the number alone |
| Bleeding signs (any anticoagulant) | Every visit, patient self-monitoring at home | Gum bleeding, blood in urine/stool, unusual bruising |