In this ECG Cases blog we review how systematic ECG interpretation can identify toxic ECG changes and help guide patient care

Written by Jesse McLaren; Peer Reviewed and edited by Anton Helman. November 2023

7 patients presented with potential medication/drug toxicity. What do their ECGs show and how can this help guide care?

Case 1: 85 year old, history of atrial flutter, with few days of weakness, HR in 30s and BP 150

–

Case 2: 80 year old, history of hypertension, with three days of weakness, HR 60 and blood pressure 100

–

Case 3: 65 year old, history of atrial fibrillation, with vomiting and weakness, HR 80 and BP 120. Old then new ECG

–

Case 4: 25 year old found agitated with an empty bottle of pills. HR 130, BP 150, temp 38

–

Case 5: 50 year old, intoxicated, with vomiting and weakness. HR 110 and BP 130

–

Case 6: 40 year old with anxiety and palpitations after substance use. HR 120 and BP 170

–

Case 7: 45 year old, history of substance use, with chest pain, HR 90 and BP 150

–

ECG interpretation in toxicology

The new American Heart Association update on poisoning summarizes the challenges of toxicity. On the one hand, “Treatment and stabilization of critically poisoned patients often must be performed before the poison involved is known.” But on the other hand, “Management of patients with critical poisoning…often differs from standard resuscitation.”[1] The ECG plays an important role in this dilemma, as it can give clues to the type of poison ingestion and which treatments should be considered.

Like in electrolyte emergencies, remembering the cardiac action potential is helpful to understand cardiotoxic medications and their treatment.[2,3]

  • Sodium channel blockers prolong ventricular depolarization, resulting in wide QRS (disproportionately affecting right ventricle: RBBB and terminal rightward axis shift, with S wave in I and tall R wave in aVR), and risk of ventricular arrhythmias: treat with sodium bicarb for QRS >100ms
  • Potassium channel blockers prolong ventricular repolarization, resulting in prolonged QT and risk of torsades de pointes: treat with magnesium for QTc > 500ms
  • Calcium channel blockers show conduction through SA and AV node (bradycardia and block), and relax vascular smooth muscle (hypotension from peripheral vasodilation): treat with calcium, fluids +/- pressors and high-dose insulin/euglycemia
  • Beta blockers reduce intracellular calcium in SA and AV nodes, leading to bradycardia and blocks: treat with glucagon, fluids +/- pressors and high-dose insulin/euglycemia
  • Digoxin inhibits the Na/K-ATPase pump, increasing intracellular calcium and vagal tone. At therapeutic levels this can result in “digitalis effect” (long PR, short QT, scooped ST depression). At toxic levels this can result in a variety of arrhythmias including excitatory (PVCs and tachyarrhythmias) and inhibitory (including bradycardias and blocks). Treat with digoxin-specific antibodies and potassium normalization (digoxin toxicity is worsened by hypokalemia, and in severe cases can cause hyperkalemia)

See the EM cases video part 1 and part 2 on toxic bradycardias

In addition to medications there are common drugs that can produce ECG changes:

  • Alcohol intoxication can cause atrial fibrillation, from binging (“holiday heart”) or withdrawal; and prolonged QT, either acutely or chronically in association with hypomagnesemia[4]
  • Cocaine is a sympathomimetic, sodium and potassium channel blocker, so acute intoxication can cause tachyarrhythmias, coronary vasospasm or plaque rupture, as well as QRS and QT prolongation.[5]
  • Opioids can cause a variety of cardiac effects, including bradycardia, QT prolongation (especially methadone), and hypotension; as well as catecholamine surge from withdrawal [6]

Patients may present with unknown poisons, combined overdoses, overdoses of medications with combined effects, or interactions between medications/drugs and electrolyte abnormalities. ECG manifestations can reflect therapeutic range (eg digitalis effect), overdose, accumulation from reduced clearance (eg from acute renal failure) or combined mechanism of action (eg multiple conditions or medications that prolong the QT interval). Mild toxicities often resolve with supportive treatment and withholding the medication/drug, but severe toxicities require guidance from a toxicologist.

However, in a study of patients referred to a poison centre, 1 in 4 referrals provided inaccurate ECG interpretation to the toxicologist that was clinically significant or would have resulted in a change in recommendations– the most common being not identifying prolonged QRS or QT interval, as well as overcalling prolong QT interval. The authors emphasized “the importance of careful ECG interpretation in the care of poisoned or potentially poisoned patients by all clinicians who participate in their care. Telephone recommendations by poison center consultants can only be as good as the information they are based on.”[7] ECG predictors for adverse cardiovascular events in poisoned patients include non-sinus rhythm, ectopy, QTc>500ms, and ischemic ST changes.[8,9]

A systematic approach to ECG interpretation can identify toxic ECG changes and help guide patient care.[10]

Heart rate/rhythm

  • Bradycardia: consider antidotes for beta-blockers, calcium channel blockers, and digoxin
  • Sinus tachycardia: consider fluids for vasodilation, and benzodiazepines for agitation
  • Arrhythmias: consider benzodiazepines for sympathetic surge, and antidotes (eg magnesium for torsades, sodium bicarb for wide complex tachycardia with rightward axis shift, digoxin antibodies for digoxin-related arrhythmias).

Electrical conduction and axis

  • PR: consider antidotes for beta blockers, calcium channel blockers, digoxin
  • QRS: consider sodium bicarbonate for QRS >100ms, especially with RBBB and terminal rightward axis shift (deep S in I and tall R in aVR)
  • QT: consider magnesium for QTc >500ms

ST/T changes

  • ST elevation: cocaine can cause coronary vasospasm, but in the patient with chest pain and ST elevation this is a diagnosis of exclusion after ruling out acute coronary occlusion
  • ST depression: diffuse ST depression could be a sign of subendocardial demand ischemia in a patient with hypotension, or a sign of hypokalemia
  • Brugada pattern: since Brugada syndrome is a sodium channellopathy, sodium channel blockers can induce or mimic Brugada pattern

Back to the cases

Case 1: 85 year, history of atrial flutter, with few days of weakness, HR in 30s and BP 150. Beta-blocker toxicity from renal failure

  • Heart rate/rhythm: atrial flutter (upright flutter waves in V1) with slow ventricular response
  • Electrical conduction: normal QRS/QT
  • Axis: normal
  • R-wave progression:
  • Tall/small voltages: low voltage
  • ST/T: no changes

Impression: atrial flutter with slow ventricular response. Was on beta-blocker, found be to be in renal failure. Treated with fluids and beta-blocker held, did not need glucagon or high-dose insulin/euglycemia. Repeat ECG had resolution of bradycardia:

–

Case 2: 80 year old, history of hypertension, with three days of weakness, HR 60 and blood pressure 100. Calcium and beta blocker toxicity from renal failure

  • H: junctional bradycardia with premature complexes
  • E: normal QRS/QT
  • A: normal axis
  • R: small R wave in V2-3
  • T: low limb lead voltages
  • S: no ST/T changes

Impression: junctional bradycardia in context of beta blocker and calcium channel blocker, patient found to be in renal failure. Improved with fluids and holding medications, did not require calcium or high-dose insulin/euglycemia. Repeat ECG showed sinus rhythm:

–

Case 3: 65 year old, history of atrial fibrillation, with vomiting and weakness, HR 80 and BP 120. Old then new ECG. Digoxin toxicity exacerbated by hypokalemia

  • H: sinus rhythm
  • E: 2nd degree AV block type 1 (prolonging PR before dropped beat), prolong QT vs QU
  • A: normal axis
  • R: normal R wave
  • T: normal voltages
  • S: scooped ST depression

Impression: digitalis effect with AV block, and more pronounced ST depression and long QT/QU. Potassium was low (2.7) and dig level was high (2.8). Treated with fluids, potassium replacement and holding digoxin, did not require dig antibodies. Repeat ECG had resolution of changes:

–

Case 4: 25 year old agitated with empty bottle of pills. HR 130, BP 150, temp 38. Diphenhydramine overdose with combined anticholinergic, sodium channel and potassium channel blockade – similar to TCA toxicity

  • H: sinus tach not SVT (biphasic P waves in V1, upright P waves superimposed on long QT in II)
  • E: wide QRS with incomplete RBBB, long QTc
  • A: large S wave in I and tall R wave in aVR
  • R: slight delayed R wave progression
  • T: normal voltages
  • S: no ST/T changes

Impression: TCA-type pattern with overdose (from diphenhydramine) with toxicity from anticholinergic (tachycardia), sodium channel blockade (wide QRS with terminal rightward axis), and potassium channel blockade (long QTc). Treated with fluids, benzodiazepines, sodium bicarb and magnesium. Follow up ECG had resolution of changes (except for baseline incomplete RBBB):

–

Case 5: 50 year old, intoxicated, with vomiting and weakness. HR 110 and BP 130. Long QT and diffuse ST depression from hypomagnesemia and hypokalemia in context of alcohol use disorder

  • H: sinus tach
  • E: long QT
  • A: normal axis
  • R: delay R wave progression
  • T: low precordial voltages
  • S: mild diffuse ST depression

Impression: long QT and diffuse ST depression from alcohol/vomiting induced electrolyte abnormalities. Potassium was low (2.8) and Magnesium was low (0.32). Treated with fluids and electrolyte replacement, with normalization of ECG:

–

Case 6: 40 year old with anxiety and palpitations after substance use. HR 120 and BP 170. Transient sodium channel blocker effect from cocaine

  • H: sinus tach
  • E: RBBB
  • A: terminal rightward axis: S in I and tall R in aVR
  • R: early R wave from RBBB
  • T: normal voltages
  • S: no primary ST/T changes

Impression: features of sodium channel blockade from cocaine ingestion. Sympathetic surge treated with benzodiazepine, and repeat ECG showed narrowing of QRS and resolution of right axis, with baseline incomplete RBBB, without the need for sodium bicarb:

–

Case 7: 45 year, history substance use, chest pain, HR 90 and BP 150. History of cocaine use but presenting with LAD occlusion, not vasospasm

  • H: sinus rhythm
  • E: normal conduction
  • A: normal axis
  • R: Q wave V2
  • T: normal voltages
  • S: convex ST elevation V1-2 with reciprocal ST depression V5-6, and ST depression inferior which is reciprocal to subtle high lateral injury

Impression: proximal LAD occlusion in context of cocaine, where vasospasm is diagnosis of exclusion. Cath lab activated: 100% LAD occlusion. First troponin 9 (normal) and peak 75,000 ng/L. Follow up ECG showed anterolateral reperfusion T wave inversion and ongoing Q waves

–

Take home points for ECG interpretation in toxicology

  • Heart rate/rhythm: consider antidotes for brady/tachy-arrhythmias, and for sinus tachycardia consider fluids for vasodilation and benzodiazepines for agitation
  • Electrical conduction and axis: consider sodium bicarb for QRS > 100 especially if RBBB or terminal rightward shift, and magnesium for QTc> 500
  • ST/T changes: consider the differential including demand ischemia, associated electrolyte abnormalities, Brugada pattern from sodium channel blockade, and acute coronary occlusion vs vasospasm from cocaine

References for ECG Cases 47 – ECG interpretation in toxicology

  1. Lavonas EJ, Akpunonu PD, Arens AM, et al. 2023 American Heart Association focused update on the management of patients with cardiac arrest of life-threatening toxicity due to poisoning: an update to the American Heart Association guidelines for cardiopulmonary resuscitation and emergency cardiovascular care. Circulation 2023;148:e149-e184
  2. Holstege CP, Eldridge DL, Rowden AK. ECG manifestations: the poisoned patient. Emerg Med Clin N Am 2006 Feb;24(1):159-77
  3. Delk C, Holstege CP, Brady WJ. Electrocardiographic abnormalities associated with poisoning. Am J Emerg Med 2007 Jul;25(6):672-87
  4. Raheja H, Namana V, Chopra K, et al. Electrocardiogram changes with acute alcohol intoxication: a systematic review. Open Cardiovasc Med J 2018 Feb 12;12:1-6
  5. Hoffman RS. Treatment of patients with cocaine-induced arrhythmias: bringing the bench to the bedside. Brit J of Clin Pharm 2010 May;69(5):448-57
  6. Krantz MJ, Palmer RB, Haigney MCP. Cardiovascular complications of opioid use: JACC state-of-the-art review. 2021 Jan 19;77(2):205-223
  7. Prosser JM, Smith SW, Rhim ES, et al. Inaccuracy of ECG interpretations reported to the poison center. Ann Emerg Med 2011 Feb;57(2):122-7
  8. Manini AF, Nelson LS, Skilnick AH, et al. Electrocardiographic predictors of adverse cardiovascular events in suspected poisoning. J Med Toxicol 2010 Jun;6(2):106-15
  9. Manini AF, Nair AP, Vedanthan R, et al. Validation of the prognostic utility of the electrocardiogram for acute drug overdose. J Am Heart Ass 2017 Feb 3;6(2):e004320
  10. Yates C and Minini AF. Utility of the electrocardiogram in drug overdose and poisoning: theoretical considerations and clinical implications. Curr Cardiol Rev 2012 May;8(2):137-51