Thursday, August 4, 2011

A Southeast Asian with Tachycardia and Hypotension after taking a dangerous herbal medication (Bidirectional Ventricular Tachycardia from Aconite Poisoning)

I published this case in Annals of EM 6 years ago.  I figure it's ok to put it here now after all these years, especially since one cannot even read an abstract of it.  The reference is below, so as not to give away the diagnosis.

This 59 year old Hmong woman presented with palpitations, nausea, vomiting, weakness, and numbness.  She had intended to take "Hmong Medicine number 9" but unwittingly took "Hmong medicine number 12."  When she realized her mistake, she knew that she would die, and her family called 911. Her BP was 65/40.  Lungs were clear and exam was otherwise unremarkable.

This was her initial 12-lead ECG:

The family showed EMS the herbal root from which the patient ate shavings:

What is the rhythm????

Answer: Bidirectional Ventricular Tachycardia.   There are alternating ventricular beats; the frontal and transverse axes alternate because there is alternating right bundle and left bundle branch block morphologies.  This is, of course, not because there is actually RBBB and LBBB, but because the origin of the ventricular beat alternates from right ventricle (LBBB) to left ventricle (RBBB).

We treated her with esmolol without any effect.  No therapy helped her, but she did spontaneously convert to NSR 8 hours later.  A Chinese PharmD and herbal specialist identified the root as aconitum Carmichaelii.

Bidirectional Tachycardia is rare, and usually associated with digitalis toxicity.  It has been reported to be unresponsive to electrical cardioversion and lidocaine, but responsive to flecainide.  Aconite (also known as monkshood, or wolfsbane) seems to trigger automaticity by direct activation of inward sodium channels during phase II of the cardiac action potential; thus, flecainide, which blocks these sodium channels appears to be effective in rats.

I had some great questions on this from Beth Bilden, toxicologist:

1.     1.  In a patient with a wide comlex dysrhythmia accompanied by hypotension (both likely caused by the unidentified toxin), I would have been tempted to give sodium bicarb which probably would have made things worse. My answer: Although the QRS is wide, it is not wide because of delayed conduction; it is wide because of bundle branch blocks morphology which indicated a focus of dysrhythmia in the ventricle, not a delay in conduction.  Therefore, bicarb would not be indicated and was not given.  In fact, the best treatment for this (flecainide), when taken in overdose is reversed with sodium bicarb!
      2.  Esmolol was used for rate control, right?  If so, was a calcium channel blocker considered rather than a beta blocker since a sodium channel blocking toxin/toxicant would decrease inotropy and a sodium channel blocking agent had not been excluded from the differential?   My answer: Again, I don't think the ECG is consistent with a sodium channel blocking agent.  I have to admit, I had no idea what the toxin was and thus what I should do, so I tried esmolol, knowing I could shut it off if it did not work.  No one else knew what to do, so she was only observed overnight and (very luckily) survived.   Obviously, if I saw a case of this now, I'd know the differential and have a much better idea how to treat it.  As would all of you readers!

Here is another case of bidirectional tachycardia.  And another.

Smith SW et al.  Bidirectional Ventricular Tachycardia Resulting From Herbal Aconite Poisoning (Case Report). Annals of Emergency Medicine 2005; 45(1):100-101.


  1. Would a Ia like procainamide be effective, or are Ic's required in digitalis induced bidirectional tachycardia?

  2. Little data, but a 1c (Flecainide) is a better sodium channel blocker than the 1a's, and does not delay repolarization as the 1a's do. Best I can tell you.

  3. doc,

    looking at this particular tracing, how does one differentiate between bidirectional VT and an underlying BBB (or even an underlying non-BBB) with some sort of bigeminy?

    or have we been too careless in identifying bigeminy, and instead should be more suspicious of something like bidirectional VT?

    (i can't get the question to not sound stupid.)


  4. Don't worry, its a good question! Bigeminy has a sinus beat followed the PVC, so one of the beats has to be preceded by a p-wave.

  5. what about such things as, say, underlying junctional rhythm with bigeminy without P waves? can those occur? if so, how does one tell those apart from bi-directional VT?

  6. Bigeminy is always irregular, with the PVC coming early, resulting in "grouped" beating. In this case, every R-R interval is exactly the same. This is also true of atrial bigeminy. If there were p-waves in front of each complex, then it would be alternating RBBB and LBBB. But there are no p-waves. This is automaticity alternating from right to left. Then the question is this: could bidirectional tachycardia be grouped, and I have to say I don't know the answer, but I suspect that, because there could be differing rates of automaticity in the right and left ventricles, that it is likely.

    Any comments on this?

    Steve Smith

  7. Could you possibly post a link to a website explaining bidirectional tachycardia in a little bit more detail? I'm having a little trouble grasping the concept of it, thank you!

  8. Here you go:

  9. Dr. Smith,

    To add in differentiation from bigeminy, there is not the expected compensatory pause if these were PVCs, nor is there a resetting of the sinus rate if these were PACs/PJCs conducting aberrantly. This leaves bigeminal interpolated PVCs, but that just sounds crazy (if possible at all).

    Another differentiator would likely be palpable pulses with both styles of beats if this were bidirectional tachycardia versus bigeminal PVCs which wouldn't be likely to be perfusing.

    Granted, these are just observations as my experience with bidirectional tachycardia is limited to case reports (and now your blog)!

  10. It seems that's all a way of saying that, in this case, the rhythm is regular (unlike bigeminy).

  11. I would be tempted to also give a temporary pacemaker wire before giving something like flecainide in an unknown intoxication because you don't know what is underneath. Aconite can also cause bradycardia. Anyway, I would not give flec in such a case because of the long half time and because of the unknown diagnosis at that moment-fe structural heart disease? Ischemia?

  12. There is plenty of uncertainty in this arrhythmia. I don't think we have to worry about ischemia as the inciting cause, but of course the tachycardia could induce ischemia and we would be unlikely to see it on the ECG. Overdrive pacing often does not work. Here is a case in which beta blockers and amiodarone worked:

  13. Thanks a lot sir for the info ,


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