Subcutaneous ICD: A Lesson in FDA Approval

This morning, a nice article in the Boston Globe appeared on Boston Scientific's subcutaneous implantable cardiac defibrillator (S-ICD)  that recently received FDA approval.  It was a fairly balanced article, one that touched ever-so-briefly on the pros and cons of a subcutaneous device to treat life-threatening cardiac arrhythmias using a medical device that does not require an internal wire inside the heart, but rather a sensing and shocking lead that is tunneled accross the chest under the skin (see my prior post on the details here). 

But what I found most interesting in the article was the patient who was recommended for the device: a dialysis patient, was a patient who was specifically excluded from the FDA trials to approve the device (specifically, patients with GFR < 29 were excluded). 

If you are a company that wants to get a device approved by the FDA, you want the best chances of having the fewest complications possible with new gadgets in medicine.  Because sick dialysis patients have a way of having more complications with device implants, and the device companies know this, they are not included in trials to get a device approved.  

Those of us who deal with ICDs in dialysis patients recognize the problems when ICD leads and dialysis catheters co-exist in the same vascular tree: the odds of infecting the ICD lead system is extraordinarily high.  In fact, the overall mortality advantage of ICDs in dialysis patients is much less than patients not on dialysis.  For this clinical circumstance, subcutaneous ICDs would seem to clearly be the better choice.

But dialysis patients are beset by another problem: challenges with potassium level regulation.  Periods of hyperkalemia are quite common in dialysis patients and hyperkalemia commonly causes severe bradycardia.  In these patients, pacing could maintain a patient's heart rate until their dialysis could be adjusted to lower their potassium and thereby improve their cardiac function.  Subcutaneous ICDs do not have pacing capabilities, however.

So the reality of the effectiveness of  the subcutaneous ICD to prolong life is uncertain in dialysis patients.  This theory has never been tested - we just tend to think it makes intuitive sense to apply a new technology when other medical device choices are not perfect either.

The subcutaneous ICD FDA approval process is a good example of how clinical dogma spreads amongst doctors and patients without any proof of a device's effectiveness in some subselected patient populations.  A new technology is approved by the FDA and receives a governmental stamp of approval.  Doctors, then, serve as well-meaning spokespersons.  The media gushes over the latest and greatest technology.  And sales explode...

... all while some patients in whom the device is deployed doesn't realize the safety and efficancy of that device was never tested in their particular clinical circumstance.

We should remember that exclusion criteria in FDA clinical trials are important.  They tell us who might  and who might not benefit from a new technology.  Too often we forget this. 

And so does the FDA.

-Wes