ESC 2012: Could a simple blood test predict the risk of sudden cardiac death?

Aug 27, 2012 00:46 EDT

Forget what you know about implantable cardiac defibrillators (ICDs) for a moment and imagine yourself in the shoes of a learner—often a patient. Do you think the fact that 70% of implanted ICDs never get used would cause confusion?

This is the problem with sudden cardiac death (SCD): you don't know when it's going to happen. Everything about it is unpredictable, including choosing which 30% of heart-failure patients will need a life-saving shock. And worse, there are many more patients who would benefit from an ICD but don't get one.

Now imagine this: a simple blood test that could identify patients that really need the device. This would be something. Perhaps a $100 blood test could avoid a $50 000 procedure?

Enter Dr Samuel Dudley from the University of Illinois. He and his colleagues have developed a simple blood test, based on a polymerase chain reaction (like hepatitis C and HIV assays), that identifies changes in the gene message (mRNA) for the cardiac sodium channel—a critical ion channel in cardiac action potentials. They have named the assay PulsePredic.

The cell biology is over my head, but what I understood was that in heart failure, SCN5A (the gene encoding for the cardiac Na channel) has two truncated mRNA splicing variants that are upregulated. These variants encode for abnormally functioning sodium channels. The secret sauce of the study is that these variants are detectable on both heart muscle and white blood cells. You can see the connection: abnormalities in signaling to the cardiac sodium channel causes altered sodium current. Altered sodium current predisposes to arrhythmia and an increased risk of sudden death.

This wasn't just a bench study. Dr Dudley and colleagues tested the assay in real patients. The study cohort was a small group of heart-failure patients who had ICDs implanted for standard primary-prevention indications. They then compared the results of the assay in patients with and without shocks. The results were striking:

Patients with ICD shocks had significantly higher number of splicing variants than those without shocks. They appeared as easily visible spikes on a graph. Both the negative and positive predictive values of the test were orders of magnitude greater than LVEF, approaching 96%.

To a therapeutic minimalist, this is all very exciting. But it's very early.

A few cautionary words:

As Dr Dudley candidly stated at the outset, he has important conflicts. His employer, the University of Illinois, holds the patent, and he is the CEO of a small biotech company whose goal it is to monetize the concept. This test is not commercially available. It's just a promising idea that needs more testing. A 700-patient study is in the planning stage. Also a limitation is that these same signals do not occur in non–heart-failure models. This means the test may not predict sudden death in other important groups, such as hypertrophic cardiomyopathy and normal hearts.

A final thought about early ideas. Although many don't play out, some do. I can remember where I was when I first heard that a coronary sinus lead might correct left bundle-branch block (LBBB) and improve CHF. No way that could work, I thought.

We shall see.

JMM

Here is the link to the ESC 2012 abstract .