Without evidence of benefit, an intervention should not be presumed to be beneficial or safe.

- Rogue Medic

Cardioversion – I’m not doing that, you do it!

First a few ground rules about what “stable” is – from the 2005 ACLS (Advanced Cardiac Life Support) guidelines.


  • If the tachycardic patient is unstable with severe signs and symptoms related to tachycardia, prepare for immediate cardioversion.
  • If the patient with tachycardia is stable, determine if the patient has a narrow-complex or wide-complex tachycardia and then tailor therapy accordingly.
  • You must understand the initial diagnostic electrical and drug treatment options for rhythms that are unstable or immediately life-threatening.[1]


In the bulleted description of bradycardia they described unstable as:

produces signs and symptoms (eg, acute altered mental status, ongoing severe ischemic chest pain, congestive heart failure, hypotension, or other signs of shock) that persist despite adequate airway and breathing,[1]


Cardioversion is one of the skills that gets even some experienced ED personnel and medics to act like rookies.


When you took ACLS (and almost everyone working in the ED has had to pass ACLS prior to working in the ED) did they have you practice cardioversion?

Or did you just practice defibrillation?

After all they’re not much different.

The problem is that when faced with cardioversion most people want anyone else to do it. In PALS (Pediatric Advanced Life Support) classes there is a similar approach to intraosseous access.

This is not a problem if the patient is stable.

But, if the patient is stable, there is no need for cardioversion or intraosseous access.

So, this is the point where we try to come to the understanding that the patient’s health is more important than our squeamishness.

What were those criteria, again?

Acute altered mental status (not the patient’s normal level of consciousness).

Ongoing severe ischemic chest pain (not mild CP or palpitations).

Congestive heart failure (not just a previous diagnosis, but something that is causing real problems right now).

Hypotension (if hypotension is the only problem and they seem otherwise stable, I am not as aggressive as with the other criteria).

Or other signs of shock (not everything fits neatly into a list, this refers to things that activate the pucker factor).

OK, those things are bad. Why does shocking them help?

We want to cause asystole, of course!


Asystole is bad. Dead. This parrot wouldn’t voom if you put 4000 volts through him!


Yes, asystole is bad as a presenting condition, but it is the best treatment for many unstable tachycardias and for VF (Ventricular Fibrillation).

But asystole is worse than the things you say get better with asystole.

You surely are crazy.

Only in a good way and stop calling me Shirley!

So we are going to practice a bit of pseudo-homeopathy – give a tiny bit of something to cause the patient to improve a lot.

Asystole is not an improvement!

Look at it this way, the heart is beating much too quickly, we need to stop it so that the sinus node – the normal pacemaker for the heart – can regain control of things.

Like the TASER you wrote about?[2]


Not at all like the TASER.

This is designed to send enough current through the heart to stop the heart for less than a second. It is hoped that the heart starts again on its own and when the heart starts again the sinus node will be controlling the rate and rhythm.

When cardioverting (or defibrillating) we want as much of the current to go through the heart as possible – anything else is wasted and possibly dangerous.

We use pads that have conductive gel or we use paddles that we apply gel to. The purpose of this gel is to help the current go through the resistance of the skin and have the current reach the heart.

We place the pads (or paddles) so that the heart is on a line between the pads. The pads do not go on the thickest part of the right pectoral muscle, but on the thinnest part (if applied over a muscle). The more tissue the current has to travel through, the less current arrives at the heart. The other pad goes on the ribs, below the heart and near the mid-axillary line and not on any muscles (we are ignoring the intercostal muscles). The other possible placement is anterior-posterior; one pad goes below the left pectoral muscle and the other pad is placed as nearly opposite the first as possible; the current travels through the heart as much as possible, rather than through the other muscles.

Placement over saline implants does not improve conduction. Yes, saline is a wonderful conductor, but go to the side of the implant.

Another method of improving conduction through the skin is to apply about 20 pounds (9 kg) of pressure. The amount is not going to be precise, but you want to put enough pressure that you force a little bit of air out of the chest, just a little, nothing more. With the pads you do not touch the pads while shocking. Please, do not apply pressure to the pads.

Here are those unstable criteria again.

Acute altered mental status.

Ongoing severe ischemic chest pain.

Congestive heart failure.


Or other signs of shock.

You do know that you are repeating yourself, don’t you?

That’s OK. When the patient is behaving in one of the ways described, you need to recognize it right away, not go looking for a not-so-handy reference guide. A little repetition is a good thing.

So, if anybody has any of those unstable presentations, then I shock them?

No. We haven’t mentioned the tachycardia, yet.

Aagh! This is complicated.

What is needed tachycardia-wise is for the heart rate to be fast enough that we believe that the rate is the cause of the unstable signs and/or symptoms.

If an infant has a heart rate of 250 BPM (Beats Per Minute), that might produce signs of instability or might not.

What are the two most likely conditions to produce this heart rate?

VT is possible, but very unlikely. AFib is even less likely.

SVT (SupraVentricular Tachycardia) and ST (Sinus Tachycardia) are the main concerns.

Maybe you are not so good at ECGs – little kid hearts beat so fast that it is hard to see anything in all of that. I’m not going to try to teach you ECGs, here. Prehospital 12 Lead ECG is the place for that.

Are you telling us that we should treat the rhythm without looking at the rhythm?

No. I am recognizing the difficulty of identifying the rhythm. I am expecting you to use a bit of clinical judgment at this point.

The rhythm we want to see is a SR (Sinus Rhythm). If the patient has a very fast sinus rhythm (ST), it would not be a good idea to shock the person out of the sinus rhythm into asystole. ST is more of a symptom than a diagnosis. If the rhythm is ST, the the rhythm is telling you that there is something causing stress to the patient. This could be a lack of volume (blood loss, vomiting, excessive sweating, drugs, lack of fluid intake, . . . ), pain, agitation, fever, physical exertion, fear of you cardioverting them, . . . .

Suppose you are driving a car, you press on the gas pedal, and the gas pedal gets stuck where you pressed it. The engine starts racing, so you turn off the ignition. If you just turn the engine back on without fixing the cause of the problem (the stuck gas pedal), you will have the engine racing just as before. This is the way ST works – It isn’t a rhythm problem, but a gas problem and the heart is not very tolerant of being turned off.

If I shock unstable patients with sinus tachycardia, things will only get worse?

Pretty much.

Are there any other rhythms that should not be shocked?

Yes, but the main thing is to avoid anything that appears sinus. Those you treat by treating the cause of the tachycardia.

When in doubt – shock.

What if it is so fast that I have no idea if it is sinus?

History can be very helpful. ST is usually the result of things that develop much more slowly than the sudden onset of an arrhythmia.

There are some people who teach that if the heart rate is faster than 220 minus the patient’s age it cannot be ST. This is nonsense. A heart rate that fast is much less likely to be ST – especially at rest, but far from impossible.

So we have an unstable patient with a tachycardia, we think the rhythm is the problem (not an underlying medical condition), we have the pads on. What next?

How about some sedation?

No thanks, that makes me forget things.

Exactly the point, but let’s give it to the patient.

What kind of sedatives can we use?

There are a lot of options – remifentanyl, fentanyl, sevoflurane, ketamine, propofol, methohexital, etomidate, midazolam, . . . .

Some of these might only be available to anesthesia. Use what you are familiar with. Now is not the time to try out new drugs that you are not familiar enough with to quickly recognize the difference between bad adverse reactions and acceptable adverse reactions.

Midazolam may be most commonly used, but is likely to worsen any hypotension.

Does this mean that we shouldn’t sedate hypotensive patients prior to cardioversion?


Use caution. This is a whole discussion to itself, so I will leave it there.

What about etomidate?

It is less likely to cause, or worsen, hypotension. Etomidate may be a good choice.

How about morphine?

Morphine takes a long time to go to work, does not cause amnesia, and lasts a long time. These are good reasons to avoid morphine.

What about combining drugs for a synergistic effect?

That may work, but it is probably safer to stick to one drug when dealing with unstable patients.

If the patient is hypotensive, what about sitting the patient up quickly to use gravity for sedation?

A physician I have a lot of respect for, who is not afraid of aggressive sedation, strongly discouraged that idea. He said it is likely to cause problems with coronary steal. That doesn’t really work for me, but I don’t encourage others to try this maneuver. If you have ideas about things that might work, I would love to hear them.

One thing not mentioned is IV access. If the patient is unconscious and probably not going to feel the shock, why delay treatment, but if the patient is conscious and going to feel this, an IV for sedation is important. One of the problems of hypotensive people is that they tend to be very difficult to start lines on. Do not avoid shocking if you cannot get an IV. Preferably, the IV is already in place.

What kind of guideline do you recommend for adequate depth of sedation before shocking?

Almost everyone seems to agree that when the patient is slurring his words, it is a good time to start. You can add more sedation if that does not work for the patient.

The man’s very first utterance was, “If it happens again, just let me die.”

As I discovered, the reason for this patient’s terror was that he had been cardioverted in an awake state. Ventricular tachycardia had been relatively slow, he had not lost consciousness, and the physicians, in the heat of the moment, had not administered adequate anesthesia. Although the 5 mg of intravenous diazepam had made him a bit drowsy, he felt the electric current on his chest and remembered the event clearly.

The patient’s mental state complicated the case considerably.[3]


What’s left to cover before we shock the patient?

The amount of energy used.

For cardioversion the order of shocks with monophasic defibrillator/cardioverters is 50 J (Joules or Watt Seconds), then progress to 100 J, 200 J, 300 J, 360 J for SVT and AFlutter. For other rhythms start at 100 J and progress the same way. When reaching the maximum without conversion, then add medication appropriate to the rhythm and continue shocking every few minutes (single shock at maximum energy).

Cardioversion with biphasic waveforms is now available,132 but the optimal doses for cardioversion with biphasic waveforms have not been established with certainty. Extrapolation from published experience with elective cardioversion of atrial fibrillation using rectilinear and truncated exponential waveforms supports an initial dose of 100 J to 120 J with escalation as needed.133,134 [4]


For polymorphic VT, the AHA does not recommend cardioversion, even though the patient may have a pulse.

the patient with persistent polymorphic VT will probably not maintain perfusion/pulses for very long, so any attempt to distinguish between polymorphic VT with or without pulses quickly becomes moot. A good rule of thumb is that if your eye cannot synchronize to each QRS complex, neither can the defibrillator/cardioverter. If there is any doubt whether monomorphic or polymorphic VT is present in the unstable patient, do not delay shock delivery to perform detailed rhythm analysis—provide high energy unsynchronized shocks (ie, defibrillation doses).[5]


Now, what is sychronization?

The whole purpose of this method of shocking the patient is to avoid having the shock land on the relative refractory period of the heart and possibly induce VF. To synchronize is to have the monitor place a dot, or some other mark, on the QRS complex of each beat to indicate that, once the shock buttons are pressed and held the shock will be delivered at the next marked beat.

How do I get the machine to synchronize?

There should be a button that has SYNC written on it. Press it. If there is a rhythm on the monitor, you should now see a mark on the QRS complex (top or bottom, depending on whether it is a mostly positive or mostly negative complex in the lead you are viewing).

Why on the QRS complex?

Because it isn’t a T wave. The refractory period and the relative refractory period are part of the T wave.

How do I know if it is the QRS or the T that the monitor is marking?

The QRS complex generally comes to a much sharper point at the top, or bottom, of the QRS. T waves, even when somewhat pointed are still more round compared with QRS complexes.

What if I cannot tell what it is?

If you think it is synchronizing on T waves, turn off the synchronization and just defibrillate. Increase the energy levels to defibrillation doses. The main difference between synchronization and defibrillation is the timing (and how you use them).

You used bold italics for pressed and held above. Is that important?


When defibrillating, the shock is delivered as soon as both buttons are pressed at the same time.

When cardioverting the synchronization changes that. There is one more factor. Both buttons still have to be pressed at the same time, but now you need to hold them until the next marked beat comes along. For a fast rhythm that is not very long. At 180 BPM you should have 3 marked beats per second, unless the synchronizer is not working well, unless there is artifact, unless the gain is too low for the synchronizer to pick up on the beats, or unless the lead is not one with a prominent QRS complex (easier to change leads than to play with the gain).

So, I assess the patient as unstable, with a rhythm that is shockable, I sedate the patient (if possible), I place the pads on the patient appropriately, I press the SYNC button and the monitor marks the QRS complexes and ignores the T waves, I select the right amount of energy, now what?

You have to charge the paddles. The initial amount of energy needed for cardioversion is much less than for defibrillation, so it takes much less time to fully charge. If the patient is intubated, disconnect the oxygen from the tube and point the oxygen away from the chest (possible fire hazard). Make sure everyone is clear of the patient and anything conductive that may be touching the patient. Make sure you are clear. With everyone remaining clear press and hold the buttons until the shock is delivered. You should hear a clunk (or whatever sound effect you feel is more descriptive) and the patient should contract a lot of muscles.

Great that was a lot. It is good to be through with this.

Hold on.

You have just treated your patient, you need to reassess.

What if you see VF on the monitor?

I would defibrillate at 360 J (monophasic).


But the patient is in VF.

Are you sure?

The monitor may be picking up artifact from poorly connected electrodes or pads. How much time does it take to check a pulse to confirm that the patient is pulseless?

When applying electrodes to unstable patients, Benzoin is a great way of helping the electrodes to stick. Unstable patients are often very sweaty. A loose electrode can produce artifact that can mislead you into believing the patient is in VF.

If this truly is VF, defibrillate immediately and follow VF treatments. If the defibrillator/cardioverter you are using remains in cardioverter mode after each shock, then turn the cardioverter off by pressing SYNC again. If it turns off automatically with each shock, then just charge up to 360 J (monophasic).

If the rhythm has not changed, press SYNC again (if necessary) and charge up to the next energy level, repeating everything as necessary.

If the rhythm has changed, reassess and treat as appropriate for the new rhythm and presentation.

Anything else?

Practice with your equipment. If you do not know how to use it in an emergency, you may only make things worse. That is not what you are there for.

No horse was actually flogged in the writing of this post – it may have fallen asleep on its feet, but that is a different matter.


[1] Principles of Arrhythmia Recognition and Management
2005 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care
Part 7.3: Management of Symptomatic Bradycardia and Tachycardia
Free Full Text from Circulation.

[2] TASER or Glock – Which treatment do you choose?
Tue, 18 Mar 2008
Rogue Medic

[3] The calamity of cardioversion of conscious patients.
Kowey PR.
Am J Cardiol. 1988 May 1;61(13):1106-7. No abstract available.
PMID: 3364364 [PubMed – indexed for MEDLINE]

[4] Supraventricular Tachycardias (Reentry SVT)
2005 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care
Part 5: Electrical Therapies
Automated External Defibrillators, Defibrillation, Cardioversion, and Pacing
Free Full Text from Circulation.

[5] Ventricular Tachycardia
2005 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care
Part 5: Electrical Therapies
Automated External Defibrillators, Defibrillation, Cardioversion, and Pacing
Free Full Text from Circulation.

Updated links and formating and switched to footnotes 10-26-10.



  1. Hahaha…there are few things that I enjoy more than zapping a conscious patient…is that evil? Oh…I enjoy adenosine quite a bit, as well ;)If any medic doesn’t want to do it…I’d be more than glad to do it for them.

  2. Hey there Rogue,For the past hour I've been reading some of your posts and strangely I felt like a void was filled… one shaped much like another medic blogger from Boston? Refreshing to hear people genuinely interested in medicine, not just being a no-it-all. As a brand new medic (awaiting licensure in the mail) it will take time to form my own opinions, but I think I may be a Fentanyl advocate right out of the box thanks to you! As far as CV, I'd have to agree with Scott. In ER clinicals I was very vocal about getting to press the bolt button! (Then again I wasn't liable then) Maybe you could write about the loose interpretations of Joules, stacking shocks, etc. I ECVed a semi-conscious pt. with 100J, then 200J in the ER per docs orders. Didn't work, and boy was that pt mad at me! Luckily he was so snowed he had no idea when he came to. Anyway, great blog. Stay safe, keep us posted.Jeff

  3. I’m an individual who was cardioverted while awake – during transport by yellow ambulance I had a pulse was breathing and a rapid heartbeat, I begged the worker not to shock me. It was a hell I never want to experience again. My eyes saw silver then black my ears popped and started ringing, I started convulsing, my face dropped on the right for two days. Its been a year. I suffer from severe tinnitus, falling, headaches, hearing loss, facial numbness loss of sensation in my fingertips on the right…DONT EVER SHOCK SOMEONE WHILE THEY ARE AWAKE. Get this – a doc from a hospital I wasn’t being transported to gave the order at 70 joules…my life is forever changed.