Identification of a left ventricular scar by cardiac MRI may help select which patients with non-ischaemic cardiomyopathy will benefit from an implantable cardioverter defibrillator (ICD).

Speaking at CSANZ 2018 in Brisbane, Dr Sarah Gutman said the 2016 DANISH study was the first to challenge the idea that all patients experience a survival benefit from an ICD.

“For a long time it’s been in the guidelines that for non-ischaemic cardiomyopathy with an ejection fraction ≤ 35%, optimal medical therapy and good functional capacity… put in a defibrillator.”

“DANISH came along and challenged everything. You don’t need to put a defibrillator in all these patients.”

Those findings inspired Dr Gutman, a cardiologist who specialises in imaging, to further investigate which patients benefit most from an ICD.

The observational study comprised 450 consecutive patients at the Alfred Hospital who fulfilled the American Heart Association guidelines for ICD implantation. The decision to implant was however at the treating physician’s discretion.

Patients, mostly men with a median age of 53 years and a New York Heart Association (NYHA) Class II, were followed for a mean of 37.9 months for a primary outcome of all-cause mortality.

The study found that in patients without a LV scar, implantation of an ICD was not associated with improved mortality (HR 1.2).

However in patients with a LV scar, ICD implantation was associated with a significant reduction in mortality (HR 0.45).

“We’ve believed for a long time that the benefit was restricted to patients with a LV scar.”

“We strongly believe that the ones with scar would receive the mortality benefit from an ICD because we know that patients with non-ischaemic cardiomyopathy with scar have more arrhythmias and a higher rate of sudden cardiac death. So it made sense that they were the ones that would receive a benefit from defibrillators and that’s what was borne out in our study.”

Dr Gutman, a finalist in the Ralph Reader Prize for Clinical Science, told the limbic the LV scar represented replacement myocardial fibrosis.

“There are cardiomyopathy patients that do well – they go on and on and on. They don’t have arrhythmias, their devices never discharge and it makes sense that if we can identify the nidus for reentry – which we can with scar – that would be the cause of death in most patients.”

“LV scar is something we can identify on cardiac MRI with contrast. It’s a routine test that gets done with every cardiac MRI as long as the renal function is good enough to receive the contrast.”

She said the findings should be a stepping-stone to attract funding for a randomised controlled trial, which could help inform future guidelines.

Dr Gutman added that better patient selection was vital given the morbidity and cost associated with defibrillators.

“They get infected, they need to be changed all the time, they go off inappropriately and cause significant psychological damage to people. And then there is the cost on a broader level.”

However there was also an ethical consideration with an RCT given the demonstrated strong mortality benefit from an ICD in the patients with LV scar.

“There is an issue with randomising scar positive patients to no ICD. But then we have to decide as a community – is this enough evidence to change the guidelines? And if we say yes, then we don’t need to do a RCT. But if it’s not, then we need to just wear that ethical consideration.”