Scarring Seen on MRI May Predict ICD Benefit Better Than LVEF

Reed Miller

August 03, 2012

August 2, 2012 (Durham, North Carolina) — Myocardial scarring assessed by MRI is an independent predictor of poor outcomes in implantable cardioverter defibrillator (ICD) candidates with low left ventricular ejection fraction (LVEF), new trial results show [1].

Senior investigator Dr Raymond Kim (Duke University, Durham, NC) told heartwire , "Our study really shows that this idea [of measuring myocardial scar] is very promising and could warrant more randomized multicenter trials."

Currently, measuring LVEF is the most common method for identifying which patients are most likely to need an implantable defibrillator to prevent sudden cardiac death, but sudden cardiac death is usually caused by ventricular tachyarrhythmia, so LVEF is only an indirect measure sudden cardiac death risk. About 70% of patients suffering sudden cardiac death have a preserved LVEF, and studies show that only one death is prevented for every 14 to 18 patients with ventricular dysfunction receiving an ICD, according to first author Dr Igor Klem (Duke University) and colleagues.

"What we have right now [for risk stratification] is not that good. The idea that function--ejection fraction--should be a key factor in determining the likelihood of arrhythmia doesn't quite make sense. We can use it clinically, but there's no real direct link from a mechanistic point of view of how muscle function or pumping function should be directly related to electrophysiology," Klem told heartwire . But several animal studies have shown that the size and shape of myocardial scarring may play an important role in the genesis of arrhythmias in these patients, he said. Also, as reported by heartwire , a 103-patient study by Dr Leah Iles (Alfred Hospital, Melbourne, Australia) and colleagues showed that cardiac MRI with late gadolinium enhancement may help predict the risk of sudden cardiac death by identifying regional fibrosis within the myocardium. Another study by Dr Jonathan Piccini (Duke University, Durham, NC) and colleagues found that myocardial perfusion defects seen with single-photon-emission computed tomography (SPECT) could also be a marker of which patients are most likely to benefit from an ICD.

"We have previously had indirect measures with echocardiography or nuclear imaging techniques, but MRI is a very high-resolution technique that can visualize myocardial scar the same way the pathologist sees the heart," Klem said. "That's the strength of this technique. You have a very high-resolution test to directly visualize the scar, and this is the anatomical substrate that is responsible for the occurrence of ventricular arrhythmias in a fairly large number of patients."

So Klem and colleagues used MRI to evaluate myocardial scarring in 137 patients being evaluated for possible ICD placement for prevention of sudden cardiac death. Results of their study are published in the July 31, 2012 issue of the Journal of the American College of Cardiology.

Over a median follow-up of two years, 39 of the patients died and/or had an appropriate ICD discharge for a sustained ventricular tachyarrhythmia. As previous trials have shown, the rate of adverse events increased steadily with decreasing LVEF, but there was a big difference in risk between those with a scar greater than 5% of left ventricular mass and those with less scarring (hazard ratio 5.2). A multivariable Cox proportional analysis showed that scar size was an independent predictor of adverse outcomes when measured either as a continuous variable or a dichotomized variable with a threshold of 5%.

Among patients with LVEFs >30%, those with scarring over 5% of the left ventricle were 6.3 times as likely to die or suffer a sustained ventricular tachyarrhythmia as those with scarring under 5%. Among patients with an LVEF <30%, those with scarring greater than 5% were 3.9 times as likely to die or suffer a sustained ventricular tachyarrhythmia as those with minimal or no scarring.

Patients with an LVEF >30% but significant scarring had the same risk as all patients with an LVEF <30%. Patients with an LVEF <30% and minimal or no scarring had a similar risk to all patients with an LVEF >30%.

Bigger Studies Are Needed

In an accompanying editorial [2], Dr Nathan Mewton and Dr Philippe Chevalier (Hospices Civils de Lyon, France) state that the study by Klem et al "confirms and adds important information to prior reports in similar settings. However, it still does not provide any evidence on the value of LV scar assessment to decide whether a patient needs an ICD. To solve this question, prospective interventional trials in larger cohorts of patients will have to be built on the relatively solid ground set by the investigations of Klem et al and previous reports."

Klem et al say that the next step in developing this technique will be to collect registry data at multiple centers and report large numbers of patients in a more heterogeneous patient group to better understand the relationship of arrhythmia risk to myocardial scar or other factors.

Kim said, "We don't want to say that this is always the best way [to assess risk] in every single situation. . . . What we're suggesting is that there could be a very important tool here for examining the heart in exquisite detail that might give you information about how that patient is going to do, but then it's incumbent on future studies to examine that for particular cohorts."

Funding for the research was provided in part by a National Institutes of Health grant. Kim is a coinventor of a US patent on delayed-enhancement MRI, which is owned by Northwestern University. Klem has no relationships to disclose. Disclosures for the coauthors are listed in the paper. Mewton and Chevalier have no conflicts of interest.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.