Thursday, May 8, 2014

Agitation and Tachycardia

A young male with unknown past medical history presents with AMS and tachycardia. EMS was called by a roommate after the patient was noticed to be nonverbal and lethargic. He reportedly took meth and had a recent drinking binge, but has not had alcohol for the last 2 days. His HR was 160 on arrival of EMS, and they gave him adenosine 6 mg and 12 mg and 500 cc NS, but with no response.  The preshospital ECG and strips are not available.  The patient was restless, agitated, and nonverbal on arrival to ED, with elevated HR at 150.  Here is the first ED ECG:
What is the likely diagnosis?

There is sinus tachycardia, a prolonged QRS (computer read it as 114 ms, previous ECG with 102 ms).  There is a large R-wave in lead aVR.  These findings are nearly pathognomonic of, or at least highly suspicious for, tricyclic antidepressant (TCA) overdose.

One can read about management of TCA overdose in many places, and I have put some links to lifeinthefastlane below.

A few important points:

1. An R-wave in lead aVR  greater than 3 mm, or an R/S ratio greater than 0.7, is highly suspicious for sodium channel blockade, which is the most important of the many toxicities of TCA overdose.

2. In 49 patients with known TCA overdose, a maximum limb lead QRS duration greater than 100 ms was 100% sensitive for detecting patients who will seize, and seizure is a harbinger of cardiovascular collapse.  At this cutoff of 100 ms, however, the specificity was not perfect.  Of 11 patients with an initial QRS duration of 100-119 ms, 2 (18%) had seizures, and of 22 with an initial QRS duration of 100-139 ms, 4 (18%) had seizures.  Of 14 with initial QRS of 140 ms or longer, 8 (56%) had seizures.  No patient with a QRS of less than 160 ms had ventricular dysrhythmias.

2a.  In an unselected population with suspicion of overdose, a minimally wide QRS (less than 110 ms) will be much less specific; furthermore, it is likely that frequent serial ECGs, by detecting an increasing QRS duration, will detect those at risk of toxicity.  On the other hand, administration of bicarbonate, the antidote, is relatively safe compared with a seizure.  If the diagnosis is unclear, narrowing of a widened QRS on ECG following sodium bicarbonate administration (1-2meq/kg) adds further support that pharmacologic sodium blockade is present.

3. There are multiple mechanisms of toxicity of TCAs:
a.   Blockade of cardiac fast sodium channels (leads to wide QRS, R-wave in aVR, R' wave in V1, Brugada pattern ECG, ventricular dysrhythmias.)  Sodium channel blockade in the CNS leads to seizures.
b.   Blockade of potassium channels (leads to long QT and torsade)
c.   Antagonism of central and peripheral muscarinic acetylcholine receptors (leads to delirium)
d.   Antagonism of peripheral alpha-1 adrenergic receptors (causes hypotension)
e.  Antagonism of histamine (H1) receptors (may contribute to sedation)
f.  Antagonism of CNS gamma-aminobutyric acid (GABA) A receptors (increase risk of seizures)
g.  Exaggeration of therapeutic effect of inhibiting central serotonin re-uptake. (may cause/contribute to serotonin syndrome)
h.  Exaggeration of therapeutic effect of inhibiting central norepinephrine re-uptake (increase risk of seizures)


When it became clear that the patient had sinus tach and not PSVT, his presentation was recognized as an overdose or drug toxicitiy.  But because the physicians were so focused on his tachycardia, meth use, and rhythm, they did not look for or appreciate the findings of TCA overdose.  We in emergency medicine obtain ECGs in overdoses mostly to look for TCA findings.  So when you get an ECG in this situation, look for them!

Much later, the roommate called and reported that 20 amitryptiline (unknown mg per dose) were missing.  Fortunately, the patient had not had any adverse outcome by that time.
He was given multiple amps of bicarb and a bicarbonate drip.  He remained delirious and was given 3 mg of physostigmine (after pretreatment with 2 mg of lorazepam to prophylax against seizures).  His delirium greatly improved and he was then able to follow commands.

He had a prolonged stay in the ICU requiring days of bicarbonate.  The tox screen only showed amitryptiline.

Articles on TCA

More TCA ECGs from Dr. Smith's ECG Blog

More on TCA overdose, with ECGs, from life in the fast lane.

More still on TCA overdose, from lifeinthefastlane.


  1. Thanks very nuce reminder of common pitfall. . How did you manage his agitation? Benzo only ? Physical restraint? Why he was not intubated?

    1. When I arrived and saw what was going on, we gave physostigmine (+lorazepam)) in addition to bicarb, which resolved the agitation and the patient did not need to be intubated.

  2. Such a very nice case. Unbelievable you are Dr. Smith.

    1. Thank you Maryam, You are too kind! --Steve

  3. VERY nice case Steve! Just curious - Was there no response at all to the Adenosine? (not even transient slowing of the rate to confirm sinus tachycardia?). THANKS again for presenting this illustrative case - :)

  4. Another thing with TCA overdose/toxicity: It may unmask a previously unknown Brugada syndrome.
    I saw that 8-10 years ago: A male patient took a large dose of dosulepine (i think it's not marketed in the USA)and the ECG was very typical of Brugada on admission in our CCU.
    Later pt was in cardiogenic shock with 'monstruous' QRS and repolarisation. Treatment with semimolar sodium bicarbonate was spectacular with a narrowing of QRS almost noticable beat by beat, with blood pressure restoration in the same time.
    No seizure as there was also benzodiazepine overdose.

    Alas the reference center in the nearby university hospital didn't confirm the Brugada diagnostic... Pt had a Reveal implanted. The device recorded a fatal VF.

    The story didn't end there. In an attempt to document the case for Review a previous TCA overdose related hospitalization was found for the patient . ECGs also shown the Brugada pattern, but no one noticed. Perhaps the syndrome was not well known at times.


DEAR READER: We welcome your Comments! Unfortunately — due to a recent marked increase in SPAM — we have had to restrict commenting to Users with a GOOGLE Account. If you do not yet have a Google account — it should not take long to register. Comments give US feedback on how well Dr. Smith’s ECG Blog is addressing your needs — and they help to clarify concepts of interest to all readers. THANK YOU for your continued support!

Recommended Resources