Sodium Bicarbonate in Amitriptyline Overdose

Sodium Bicarbonate in Amitriptyline Overdose

A 40 year old man with pre-existing mental health problems presented after an overdose of 6g of amitriptyline. He was deeply unconscious and required invasive ventilation. He was commenced on bicarbonate therapy and hyperventilated to pH 7.5. Around 12 hours after admission he developed tonic-clonic seizures, a broad complex tachycardia and subsequently suffered a cardiac arrest that was refractory to defibrillation, adrenaline and amiodarone. He was given additional 8.4% bicarbonate and further defibrillation attempts and was successfully resuscitated after 90 minutes.

What is the rationale for the use of sodium bicarbonate in the management of amitriptyline overdose?

Helen Vollmer

Tricyclic antidepressants are well absorbed orally, rapidly distributed, highly lipophilic, and are 85-98% bound to plasma proteins and tissues. There is a large (10 to 50L/kg) volume of distribution. Metabolism is via hepatic microsomal enzymes and conjugation into active and inactive metabolites. There is significant first pass metabolism and enterohepatic recycling. In overdose, the pharmacokinetics are altered and plasma concentrations can take >12 hours to peak due to slow dissolution, poor absorption when ionised by gastric acid and the anticholinergic effects of delayed gastric emptying. Elimination half-life is 24-72 hours(1).

The toxic effects of tricyclics are caused by: 1) Inhibition of norepinephrine reuptake at nerve terminals, which causes sinus tachycardia. After a hyperadrenergic surge, it also results in hypotension. 2) Direct a-adrenergic block contributes to hypotension and tachycardia. 3) The blockade of myocardial fast-inward sodium channels leads to decreased automaticity, impaired conduction and negative inotropy. 4) Anticholinergic action including dilated pupils, urinary retention, and impaired sweating is common and aids diagnosis, but doesn’t usually cause serious problems(1,2).

Poisoning manifests as CNS disturbances (agitation, lethargy, coma – 35%, myoclonic jerks – 50% and seizures – 10%), cardiovascular derangements (hypotension, conduction block and arrhythmias) and anticholinergic effects. Dose ingested is poor at predicting outcome, the LD50 is approximately 35mg/kg but 20mg/kg has been fatal due to individual variation in pharmacokinetics. Plasma levels do not detect active metabolites, are not rapidly obtainable and correlate poorly with seizures and dysrhythmias(1).

Patients at risk of developing cardiac and neurological toxicity have sinus tachycardia, QRS complexes >100ms, prolonged QTc interval, right axis deviation of 130 to 270o and an increase in R wave amplitude (>3mm) in lead aVR. Serial ECG changes can predict impending toxicity3. This can lead to supraventricular or ventricular dysrhythmias or cardiac arrest. A terminal 40 millisecond frontal plane QRS vector of 130-270o was 100% sensitive and 98-100% specific for TCA overdose and accurately separated patients who had been poisoned from those who had not (positive and negative predictive values of 1.00). Combining sinus tachycardia, prolonged QT interval (≥ 418ms) and a terminal QRS vector between 130-270o was 97% efficient as a diagnostic aid for TCA overdose (4).

There are no RCTs evaluating treatments for cardiac arrest caused by TCAs. One case series demonstrates an improvement after sodium bicarbonate and adrenaline, however this series used physostigmine pre-arrest. The evidence for NaHCO3 in TCA cardiotoxicity is limited to animal studies and case series (5). Sodium bicarbonate has reversed metabolic acidosis and dysrhythmias associated with TCA overdose in dogs. It accelerates intraventicular conduction, which counteracts the conditions that allow re-entry. This is due to both alkalization and sodium loading (6). Because of case reports, and in the absence of any large published series, it is recommended that NaHCO3 is given, prophylactically or therapeutically, to maintain a pH of 7.40-7.507.

One of the 2 case series cited in the 2010 International Consensus on Cardiopulmonary Resuscitation is Hoffman et al. A retrospective review of 91 patients admitted to the ICU with TCA overdose who had hypertonic sodium bicarbonate and another 43 patients who had not. 91% had rate, rhythm or morphology abnormalities on ECG. 54% had QRS complexes >0.16secs. There were no demographic differences between the groups; the NaHCO3 group were more likely to have prolonged QRS complexes (54% v 29%). NaHCO3 was associated with improvement in blood pressure (96% within 1 hour) and ECG changes (resolution of QRS in 80%). No physical or ECG parameter worsened after bicarbonate. One patient died, who was moribund on arrival. ICU stay was a median of 2 days compared to 4 days in the non-bicarbonate group (7). It is unclear the contribution to blood pressure improvement that NaHCO3 has independent of other factors. The authors note that the patients treated without NaHCO3 were treated in the earlier part of the 8 year data collection period, which may account for part of the difference in length of stay.

The 2010 International Consensus states that there is insufficient evidence to suggest that CPR algorithms should be altered for patients with cardiac arrest as a result of TCAs. Sodium bicarbonate should be used in the post-arrest period when associated with wide QRS complexes5.

Case reports e.g. Engels suggest that intra-lipid emulsion has a role in amitriptyline overdose management (8). A patient, requiring adrenaline and noradrenaline after reversal of conduction abnormalities with NaHCO3, had a rapid reduction in vasopressor requirements following intra-lipid administration. Intra-lipid emulsion has an established role in local anaesthetic toxicity. Proposed mechanisms include a ‘sink’ for lipophilic drugs, augmentation of cardiac energy supplies and direct activation of cardiac voltage-gated calcium channels.

A rat study, comparing intravenous lipid emulsion to Hartmann’s solution or 8.4% bicarbonate in amitriptyline poisoning, found an increase in QRS interval, and a fall in blood pressure for both intra-lipid and Hartmann’s solution compared to bicarbonate. Survival was worse in the intra-lipid group. It is possible that intra-lipid emulsion has a beneficial effect in addition to sodium bicarbonate but this has not been investigated (9).


1) Newton EH, Shih RD, Hoffman S. Cyclic Antidepressant Overdose: A Review of Current Management Strategies. American Journal of Emergency Medicine 1994; 12(3): 376-9
2) Kerr GW, McGuffie AC, Wilkie S. Tricyclic antidepressant overdose: A review. Emergency Medicine Journal 2001; 18: 236-41
3) Singh N, Singh HK and Khan IA. Serial electrocardiographic changes as a predictor of cardiovascular toxicity in acute tricyclic antidepressant overdose. American Journal of Therapeutics 2002; 9: 75-9
4) Groleau G, Jotte R and Barish R. The electrocardiographic manifestations of cyclic antidepressant therapy and overdose: a review. The Journal of Emergency Medicine 1990; 8: 597-605
5) Deakin CD, Morrison LJ, Morley PT et al. Part 8: Advanced life support. 2010 International Consensus on Cardiopulmonary Resuscitation and Cardiovascular Care Science with Treatment Recommendations. Resuscitation 2010: 81S; e93-174
6) Sasyniuk BI, Jhamandas V, Valois M. Experimental Amitriptyline Intoxication: Treatment of Cardiac Toxicity with Sodium Bicarbonate. Annals of Emergency Medicine 1986; 15(9): 1052-9
7) Hoffman JR, Votey SR, Bayer M et al. Effect of Hypertonic Sodium Bicarbonate in the Treatment of Moderate-to-Severe Cyclic Antidepressant Overdose. American Journal of Emergency Medicine 1993; 11(4): 336-41
8) Engels PT, Davidow JS. Intravenous fat emulsion to reverse haemodynamic instability from intentional amitriptyline overdose. Resuscitation 2010; 81: 1037-9
9) Perichon D, Turfus S, Gerostamoulos D et al. An assessment of the in vivo effects of intravenous lipid emulsion on blood drug concentration and haemodynamics following oro-gastric amitriptyline overdose. Clinical Toxicology 2013; 51: 208-15

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s