Myocardial Fibrosis Predicts Appropriate Device Therapy in Patients With Implantable Cardioverter-Defibrillators for Primary Prevention of Sudden Cardiac Death

OBJECTIVES: The purpose of this study was to evaluate the association between regional myocardial fibrosis and ventricular arrhythmias in patients with cardiomyopathy. Patients with heart failure are at risk of sudden cardiac death (SCD). Current guidelines recommend implantable cardioverter-defibrillator (ICD) devices for a subgroup based on impaired left ventricular function. A significant proportion of devices never discharge, hence a more accurate method for targeting those at risk is desirable.

METHODS: We prospectively enrolled 103 patients meeting criteria for ICD implantation for primary prevention of SCD. Cardiac magnetic resonance imaging was performed before device implantation. Regional fibrosis was identified with late gadolinium enhancement (LGE). 

RESULTS: Median follow-up was 573 days (interquartile range: 379 to 863 days). The LGE identified regional fibrosis in 31 of 61 (51%) patients with nonischemic cardiomyopathy (NICM) and in all 42 patients with ischemic cardiomyopathy (ICM). There was a 29% (9 of 31) discharge rate in the NICM group with LGE compared with a 14% (6 of 42) discharge rate in the ICM group (p = NS). There were no ICD discharges in the NICM group without LGE, which was significantly lower than the rate observed in both the ICM patients (p = 0.04) and the NICM patients with LGE (p < 0.01). Left ventricular ejection fraction was similar in patients with and without device therapy (24 ± 12% vs. 26 ± 8%, p = NS) and those with or without LGE (25 ± 9% vs. 26 ± 9%, p = NS).

CONCLUSIONS: Patients with advanced cardiomyopathy and myocardial fibrosis demonstrated by LGE on cardiac magnetic resonance imaging have a high likelihood of appropriate ICD therapy. Correspondingly, absence of LGE may indicate a lower risk for malignant ventricular arrhythmias.  

PMID: 21310318

Posted in Magnetic Resonance Imaging and tagged , , , , , , .


  1. Interesting manuscript! None of the patients that had NO LGE had an ICD discharge. This raises some good questions. Should we be screening patients prior to ICD placement?

    This, in addition to the recently published data that shows that 22.5% of these procedures did not meet the criteria for an ICD implant in the 2006 or 2008 guidelines may also further promote that concept. Especially since that paper also showed that ICD implants not recommended by the professional guidelines are more likely to result in in-hospital complications.

    Larger randomized studies needed.

    In the meantime, see the editorial comment published:
    – Willerson J. Imaging in electrical cardiac devices. J Am Coll Cardiol 2011; 57:829-830. []

  2. See also:

    Profiling cardiac resynchronization therapy patients: responders, non-responders and those who cannot respond-The good, the bad and the ugly?
    Zacà V, Mondillo S, Gaddi R, Favilli R.
    Int J Cardiovasc Imaging. 2011 Jan;27(1):51-7.

Leave a Reply

Your email address will not be published. Required fields are marked *