Ineffective or inappropriate ICD shocks - Part I

Implantable cardioverter defibrillators (ICDs)

Once upon a time, to receive an implantable defibrillator required that you survive not one, but two episodes of sudden cardiac death. You had to have ventricular arrhythmias refractory to drug therapy, and you had to be strong enough to undergo a thoracotomy.

It's astonishing that anyone qualified for the device!

Since then, the technology has come a long way. The device has been miniaturized, allowing a transvenous approach. The devices are also now highly programmable, and usually integrated with a pacemaker.

Thanks to a series of clinical trials (MADIT I, MADIT II, DEFINITE, SCD-HeFT) the number of patients for whom the device is indicated has grown exponentially.

Got Class II or Class III heart failure and impaired left ventricular function? Then there's a good chance you qualify for an ICD.

What this means is that paramedics are seeing more of these devices in the field. It also means that more of our heart failure patients are going to contact 9-1-1 when they get shocked. We also may be called to a heart failure patient who presents with a wide complex tachycardia, leaving the paramedic to wonder why the ICD is not firing.

Here's a recent case to illustrate the point.

A 70 year old male contacts 9-1-1 because his ICD has fired several times.

EMS arrives on the scene and assesses the patient.

Vital signs:

Pulse: 100 and irregular
BP: 139/79
Resp: 18
SpO2: 98 on RA

The cardiac monitor is attached.


We see a borderline wide complex tachycardia at a rate of 100 (probably paced) and a rhythm change at the end of the strip.

A 12 lead ECG is captured.


The 12 lead ECG shows a paced rhythm with concordant ST segment depression in leads V4 and V5.

As a side note, we need to remember that this patient has been shocked several times with an ICD, which could cause ST/T wave abnormalities.

This patient was shocked at least 12 more times while he was with EMS.

When I asked the treating paramedic if he captured a rhythm strip of the patient being shocked (so we could determine whether or not the device was malfunctioning) the first thing he showed me was the second 12 lead ECG.


I'm not trying to be critical here. This is an opportunity for learning. What we see in this 12 lead ECG is a loose electrode in the V2 position. Had this been an actual ICD shock, leads V1 and V3 would also have been affected. Also, the duration of the shock would have shorter.

However, the paramedic in question did document a few ICD shocks because he wisely pressed the "print" button and left the printer running.

Good call!

Here's a rhythm strip of the ICD firing.


It shows a vertical takeoff that goes straight off the paper and comes back down to baseline after about 200 ms. (Note: it will also generally be associated with a "Yelp!")

Here's the post-shock rhythm.


How would you treat this patient and why?

See also:

Ineffective or inappropriate ICD shocks - Part II

Ineffective or inappropriate ICD shocks - Part III