Tricyclic antidepressant toxicity

aka Toxicology Conundrum 022

A 25 year-old male (70 kg) is brought in by ambulance 30 to 60 minutes after ingesting 7 x 500mg amitriptyline. He is tachycardic (HR 120) with an otherwise ‘normal’ ECG (QRS 95 ms) but is becoming drowsy. You are called to the resuscitation room to assess him.

Questions

Q1. What is the mechanism of toxicity in tricyclic antidepressant overdose?

Tricyclic antidepressants (TCAs) overdoses are Australia’s major cause of drug ingestion fatality. TCAs are weak bases (typically with pKa of ~8.5) that act as noradrenaline and serotonin reuptake inhibitors and GABA-A receptor blockers.

Cardiotoxic effects primarily result from blockade of inactivated fast sodium channels in a use-dependent manner (blockade is higher at faster heart rates). This can result in life-threatening dysrhythmias. Of secondary importance is reversible inhibition of potassium channels and direct myocardial depression.

Other toxic effects result from blockade of muscarinic (M1), histaminergic (H1), and peripheral alpha1-adrenergic receptors.

Tricyclic antidepressant toxicity GatingCurrentF4

Sodium channel activation states

Q2. What is the risk assessment for this patient?

The patient has ingested 50 mg/kg of amitryptyline.

  • >10 mg/kg is potentially life-threatening.
  • >30 mg/kg is expected to result in severe toxicity with pH-dependent cardiotoxicity and coma lasting >24 hours.

Expected clinical manifestations include:

  • rapid deterioration within 1-2 hoursof ingestion – even if the patient is alert with a normal ECG on arrival.
    • Delayed effects may result from anticholinergic-mediated delayed gastric emptying or extended release amitriptyline.
  • central nervous system
    • sedation and coma tend to precede cardiotoxity
    • seizures
    • delirium (anticholinergic)
  • cardiovascular
    • sinus tachycardia and possible mild hypertension initially
    • hypotension (alpha-blocking effects and myocardial depression)
    • broad complex tacydysrrhythmia
    • broad complex bradycardia occurs pre-arrest
  • anticholinergic effects
    • may occur on present or may be delayed and prolonged
    • agitation, restlessness, delirium
    • mydriasis
    • dry, warm flushed skin
    • urinary retention
    • tachycardia
    • ileus
    • myoclonic jerks

Q3. What ECG findings are typical of tricyclic antidepressant overdose?

The important ECG findings suggestive of TCA toxicity are QRS widening (>100 ms) and and right axis deviation of the terminal QRS – in combination, these findings are almost pathognomic of sodium channel blockade:

  • Right axis deviation of the terminal QRS is defined by:
    • terminal R wave >3 mm in aVR, or
    • R/S ratio >0.7 in AVR
  • QRS widening
    • >100 ms is associated with seizures
    • >160 ms is associated with cardiac dysrhythmias

A right bundle branch block pattern may be found. Tachycardia is often present as a result of the anticholinergic effects of TCAs or as a reflex response to alpha1-blockade mediated hypotension. Bradycardia in the context of a massive TCA overdose is generally a pre-terminal event.

Finally, the ECG can be normal if the dose ingested was sub-toxic or if the patient has presented early.

Q4. What is the specific antidote for tricyclic antidepressant overdose?

Sodium bicarbonate.

  • most conveniently used as 50 mmol/50 mL single use pre-filled syringes for rapid administration.
  • also available as 100 mmol/ 100 mL vials.

Q5. What are the possible mechanisms of therapeutic effect of this specific antidote in tricyclic antidepressant overdose?

The mechanisms for the therapeutic effect are multifactorial and poorly understood. Any or all of the following mechanisms may have role:

1. Plasma alkalinization and TCA plasma protein binding
2. Intracellular alkalosis and TCA receptor binding
3. Intracellular hypopolarization
4. Sodium load
5. Correction of metabolic acidosis
6. Volume loading
7. Other pharmacokinetic effects

These potential mechanisms are discussed in (exhaust-ive/ing!) detail below:

1. Plasma alkalinization and TCA plasma protein binding

Plasma alkalinization promotes TCA protein binding (especially to alpha1-acid glycoprotein (AAG)), reducing the concentration of free drug available to cause sodium blockade.

  • As up to 95% of the drug is protein bound (varies for different TCAs), sodium bicarbonate can make a large difference to its unbound fraction and hence its toxicity.
  • Thus plasma proteins can act as a “sink” that sequesters TCAs away from the sites of toxicity (the sodium channels), until they can be redistributed to peripheral tissues.

The capacity for plasma protein binding to TCAs in an overdose setting depends on many factors but their clinical significance is unknown:

  • The amount of TCA to bind.
  • The amount of TCA that binds to each AAG protein (up to 2 to 14 times the AAG concentration)
  • The amount of AAG there is in the circulation.
  • The degree to which the binding capacity (and the affinity of different binding sites) changes with change in pH.

Other factors may play a role such as variation in the distribution of different TCAs between RBCs and the plasma, the effects of age and disease-states on AAG concentration, and perhaps even lipid levels in the blood.

However, pH change is effective in the absence of protein in experimental models, so mechanisms other than the effects of protein binding must be important.

2. Intracellular alkalosis and TCA receptor binding

Intracellular alkalosis increases the unbinding rate of TCAs from the sodium channel receptor as a result of increased lipid solubility. This promotes dissociation of the neutral form of the drug from the TCA receptor site in the sodium channel.

  • The ionized form of TCAs binds the inactivated voltage-depended sodium channel and is trapped in the channel; this leads to sodium channel blockade.
  • Alkalinisation favours the nonionized state which does not become bound and trapped in the sodium channel and can thus diffuse through the plasma membrane.
  • Presumably the TCA must enter the intracellular space prior to binding the sodium channel as much of the effect of bicarbonate is lost if the cellular bicarbonate pump is blocked to prevent the intracellular accumulation of bicarbonate (Wang’s protein-free perfused heart model).

3. Intracellular hypopolarization

High bicarbonate leads to high extracellular pH. This, in turn, results in proton-potassium exchange across plasma membranes leading to low extracellular potassium concentration/ high intracellular potassium concentration and hypopolarization that decreases sodium channel blockade by voltage-dependent drug-binding changes.

4. Sodium load

Sodium load has a secondary positive effect by over-riding sodium channel blockade due to an increased sodium concentration gradient  across the cell membrane.

  • Hypertonic saline was  more efficacious than alkalinization at improving cardiac conduction and hypotension in a swine model.
  • There are case reports of good responses to rapidly administered boluses of hypertonic saline in TCA toxicity, whereas in a case report of a slow infusion there was no effect.

5. Correction of metabolic acidosis

Plasma alkalinisation also counters the metabolic acidosis caused by TCAs. Severe metabolic acidosis is potentially fatal on its own if severe. This may also help reduce tachycardia, and thus decrease use-dependent Na channel blockade.

6. Volume loading

The volume effects of sodium bicarbonate may have benefit in the shocked patient, by ameliorating the consequences of shock and allowing more widespread distribution of TCAs to tissues other than the heart and CNS.

7. Metabolism, tissue distribution, excretion, and urinary alkalinization

The effects of alkalinization on hepatic metabolism and tissue distribution are not well understood.

In the context of sodium bicarbonate use, tissue distribution is likely to be important (as alluded to above).

  • The early toxicity of TCAs results from the initially high plasma concentrations (rapid oral absorption leads to peak levels within 2 hours) and rapid distribution to highly perfused organs (brain and heart).
  • Increased protein binding may allow time for redistribution to other peripheral organs such as skeletal muscle and adipose tissue.

Metabolism and elimination are probably much less important.

  • TCAs are cytochrome P450 metabolized and undergo saturation in an overdose settling, leading to a prolonged half-life.
  • Similarly they are highly lipid-soluble and widely distributed leading to a high volume of distribution and thus a long elimination half-life.

TCAs typically undergo some degree of enterohepatic circulation.

Urine alkalinization does not confer any therapeutic benefit.

  • Renal excretion of TCAs is typically <10% as the active molecules are highly lipid-soluble and undergo extensive metabolism.
  • High pH will DECREASE ionization of TCAs, the opposite of what would be necessary to trap TCAs in the urine (and I don’t think trying to acidify the urine is a good idea!)

Q6. How and when should this specific antidote be administered in tricyclic antidepressant overdose?

Indications for sodium bicarbonate in tricyclic antidepressant overdose:

  • severe cardiotoxicity
    • cardiac arrest
    • ventricular dysrhythmias
    • hypotension resistant to fluid challenge
  • consider for preventionof severe cardiotoxicity resulting from:
    • seizure – leads to metabolic acidosis
    • prolonged intubation attempts – leads to respiratory acidosis

Administration of sodium bicarbonate:

  • If cardiac arrest or arrhythmia and haemodynamically unstable(hypotension) then:
    • sodium bicarbonate 100 mmol (2 mmmol/kg) bolus every few minutes while monitoring the effect on ECG until haemodynamically stable.
    • the optimal total dose is “enough” (to reverse cardiotoxicity).
  • Once stableafter resuscitation:
    • consider further sodium bicarbonate to maintain pH 7.5- 7.55 based on hourly ABGs and QRS width (aim for <100 ms).
  • if there is ongoing arrhythmia, QRS >140 ms, or hypotensionthen options include:
    • repeat sodium bicarbonate boluses or
    • sodium bicarbonate infusion(100 mmol in 1L normal saline at 250 mL/h) and adjust the rate based on hourly ABGs.
      • Boluses of sodium bicarbonate are likely to be more effective than infusions because they will lead to rapid shifts in the concentration of free drug.
      • Sodium bicarbonate infusions may lead to renal compensation for metabolic alkalosis reducing their effectiveness.
  • Most patients with severe toxicity will be intubated and sodium bicarbonate infusions may be unnecessary if the patient can be hyperventilated to a target of pH 7.5-7.55.
  • It may be prudent to consider a bolus of sodium bicarbonate prior to intubation to counter the effects of increased acidosis while ventilation is ceased.

Plasma alkalinization can be stopped once the ECG and haemodynamic parameters have normalized.

  • There may be a theoretical risk of relapse of cardiotoxicity as a result of further TCA unloading from plasma proteins if sufficient time has not been given to allow TCA redistribution to the more poorly perfused tissues.
  • Ongoing monitoring is essential, although the precise duration is uncertain and requires clinical judgment.

Q7. Describe your approach to managing this patient.


  • Resuscitation -
    Manage patient in an area equipped for cardiorespiratory monitoring and resuscitation.
    Potential life threats are:

    • coma
    • respiratory acidosis
    • seizures
    • cardiac dysrhythmia
    • cardiac arrest
      • Do not stop resuscitation until intubated, treated with sodium bicarbonate, and pH >7.5 (or until the change of shift…)
      • Consider extreme measures such as extracorporeal membrane oxygenation and circulatory assist devices in extremis.
      • Good neurological outcome can be achieved even after many hours of cardiac arrest with effective CPR.

Ventricular dysrhythmias

  • treat with sodium bicarbonate
  • cardioversion and defibrillation are unlikely to be successful
  • type 1a antiarrhythmics (e.g. procainimide), amiodarone, and beta-blockers are contra-indicated.
  • hypertonic saline, intralipid and even high-dose insulin euglycemic therapy (HIET) are unproven therapeutic measures that should be considered in refractory cases.

Seizures

  • benzodiazepines (e.g. diazepam 5-10 mg IV)
  • sodium bicarbonate (seizure-induced metabolic acidosis may worsen TCA cardiotoxicity)
  • rapid sequence intubation and ventilation

Hypotension

  • IV crystalloid (10-20 ml/kg) boluses
  • vasopressors such as noradrenaline (if alpha-blockade is thought to be contributing)
  • sodium bicarbonate

CNS depression

  • prompt intubation at the onset of CNS depression (e.g. GCS<12) – consider a bolus of sodium bicarbonate prior to intubation to guard against worsening acidosis.
  • Hyperventilate intubated patients to pH 7.50-7.55
  • Supportive care and monitoring -
    general measures, including indwelling urinary catheterisation and continuous cardiac monitoring.
  • Investigations -
    Screening tests in deliberate self-poisoning – ECG, glucose, paracetamol level
    Other investigations may be indicated according to progress/ comorbidities/ possible complications (e.g. chest radiograph, ABG)
  • Decontamination -
    Activated charcoal can be given in TCA ingestions >10 mg/kg, but only after the airway is secured by endotracheal intubation.
  • Enhanced elimination – nil
  • Antidotes -
    sodium bicarbonate (see Q3-5)
  • Disposition -
    This patient will need intubation and ventilation and should be admitted to ICU.

References

  • Blackman K, Brown SG, Wilkes GJ. Emerg Med (Fremantle). Plasma alkalinization for tricyclic antidepressant toxicity: a systematic review. 2001 Jun;13(2):204-10. PMID: 11482860
  • Harvey M, Cave G. Intralipid outperforms sodium bicarbonate in a rabbit model of clomipramine toxicity. Ann Emerg Med. 2007 Feb;49(2):178-85, 185.e1-4. PMID: 17098328
  • Liebelt EL, et al. Serial electrocardiogram changes in acute tricyclic antidepressant overdoses. Crit Care Med. 1997 Oct;25(10):1721-6. PMID: 9377889
  • Liebelt EL, et al. ECG lead aVR versus QRS interval in predicting seizures and arrhythmias in acute tricyclic antidepressant toxicity. Ann Emerg Med. 1995 Aug;26(2):195-201. PMID: 7618783
  • McCabe JL, et al. Experimental tricyclic antidepressant toxicity: a randomized, controlled comparison of hypertonic saline solution, sodium bicarbonate, and hyperventilation. Ann Emerg Med. 1998 Sep;32(3 Pt 1):329-33. PMID: 9737495
  • Murray L, Daly FFS, Little M, and Cadogan M. Chapter 3.65 TCAs: Tricyclic Antidepressants; in Toxicology Handbook, Elsevier Australia, 2007. [Google Books Preview]
  • Murray L, Daly FFS, Little M, and Cadogan M. Chapter 4.24 Sodium bicarbonate; in Toxicology Handbook,  Elsevier Australia, 2007. [Google Books Preview]

Related posts:

About Chris Nickson

An oslerphile suffering from a bad case of knowledge dipsosis. Key areas of interest include: emergency medicine, critical care, toxicology, tropical medicine, clinical epidemiology, history, literature and the internet-learning revolution. @precordialthump | + Chris Nickson | Contact

Comments

  1. I am surprised that this is such a big problem in Australia. In the US, almost everybody seems to get the SSRIs or variants of them. However, when they were more commonly prescribed, they were the overdose most likely to be successful. Part of that due to the various actions of the drugs in overdose.

    I was taught to intervene early. At the first observation of tachycardia related to a TCA OD. Before any widening of the QRS complex, if possible. I was under the impression that the sodium helped to prevent the delay in ventricular conduction.

    Now that we have waveform capnography, it is easier to observe changes in the CO2 levels and respiratory drive, since these patients tend to lose their ability to protect their airways pretty quickly, once the toxic effects start to present themselves. The bicarb won’t do a thing for the level of consciousness, but it is easier to manage the level of consciousness if the vital signs are not rapidly deteriorating.

    It has been years since I have had a patient with a TCA OD. This is a great refresher on mechanism and treatment.

  2. Thanks for the comment Rogue Medic.

    I agree with you in highlighting the need to intervene early. In a significant TCA overdose the patient usually goes downhill rapidly. A tachycardia in itself may be a result of the anticholininergic effects in combiination with a reflex response to the alpha blocking affects of TCAs, so NaHCO3 is not strictly indicated as it primarily targets sodium channel blockade.

    However, a drop in GCS (e.g. to about 12 if you need to put a number on it) means the patient should be intubated for airway protection and to facilitate hyperventilation if required for impending QRS widening. A good tip is to give an NaHCO3 bolus prior to intubation, because respiratory acidosis (resulting from absent ventilation during a rapid sequence intubation) may lead to greater ionisation of the TCA and thus greater cardiotoxicity and all round badness…

    Quetiapine and olanzepine are our most common drug overdoses requiring intubation and ventilation. TCAs are much less common these days but have greater lethality than the antipsychotics and other anti-depressants. Calcium channel blockers (i.e. verapamil and diltiazem) are generally regarded as even more lethal than TCAs, but are even less common -- and in our neck of the woods deaths have been very rare or non-existent since the more widespread use of of high-dose insulin euglycemic therapy (HIET)…. But that’s another story…

Speak Your Mind

*