• Late Gadolinium Enhancement Imaging a ‘Significant, Consistent and Strong’ Predictor of Ventricular Arrhythmias and Sudden Death in Dilated Cardiomyopathy

    Late gadolinium enhancement (LGE) imaging was a “significant, consistent, and strong predictor” of ventricular arrhythmias (VA) and sudden death in a large new retrospective cohort study of non-ischemic dilated cardiomyopathy (DCM) patients.

    The study, written by Andrea Di Marco, MD, PhD, from L’Hospitalet de Llobregat, Barcelona, Spain, and the University of Manchester, England, and colleagues, was published online Monday ahead of the June 15 issue of the Journal of the American College of Cardiology.

    The data was used to create a new simple algorithm combining LGE and left ventricular ejection fraction (LVEF) strata, which the researchers said was “significantly superior” to existing “suboptimal” risk stratification methods for VA and sudden death in DCM, which have largely been based on LVEF alone.

    LVEF is “is not a very accurate predictor of arrhythmic endpoints,” they said, adding that randomized trials with inclusion criteria for LVEF ≤35% in DCM have failed to detect significant survival benefits with primary prevention implantable cardioverter-defibrillators (ICDs).

     

    Thesis

    Recent studies have flagged localized myocardial fibrosis – detected by cardiac magnetic resonance (CMR) with LGE imaging – as a predictor of VA and sudden death. However, the researchers noted that data have been scarce concerning the prognostic impact of LGE across the different strata of LVEF, while there is no simple risk prediction model to integrate LGE status and LVEF.

    The current study, therefore, was aimed at developing a new algorithm for risk stratification of VA and sudden death in DCM, assessing the impact of LGE across a range of LVEF strata.

    The study included 1,165 consecutive patients with hypokinetic DCM in a retrospective cohort who underwent CMR between August 2008 and June 2018 at the North West Heart Centre in Manchester and between September 2013 and June 2018 at Bellvitge University Hospital, Barcelona.

    LGE was present in 486 (42%) of the patients and was significantly associated with older age (median age of 60 vs. 57 years for those without LGE) and male sex (76%).

    In general, at baseline, the patients were majority male (66%), had a median age of 58 years (interquartile range [IQR]: 48 to 68 years), 27% had atrial fibrillation, and 81% were on beta-blockers, 83% on angiotensin-converting enzyme (ACE) inhibitors/ angiotensin receptor blockers (ARBs), 49% on mineralocorticoid receptor antagonists (MRA) treatment, and 41% on loop diuretics.

    Their median LVEF was 39% (IQR: 30% to 46%), the median elevated left ventricular end-diastolic volume (LVEDV) index was 118 ml/m2 (IQR: 99 - 142 ml/m2), median left ventricular end-systolic volume indexed to body surface area (LVESV) index was 69 ml/m2 (IQR: 55 - 95 ml/m2). The highest percentage (40%) were in NYHA class I, followed by 34% in class II, 22% in class III and 4% in class IV.

    With a median follow-up of 36 months (IQR: 20 to 58 months), the combined arrhythmic endpoint included appropriate implantable cardioverter-defibrillator therapies, sustained ventricular tachycardia, resuscitated cardiac arrest (rCA) and sudden death.

    Seventy-four patients (6%) reached the primary endpoint during follow-up – 33 of whom had received appropriate ICD implantation, 26 had sustained monomorphic ventricular tachycardia before any ICD implantation, rCA happened in eight cases and sudden death in seven cases. In addition, sudden death occurred in five patients who had already experienced episodes of VA.

    LGE was found to be an “independent and strong” predictor of the arrhythmic endpoint (hazard ratio [HR]: 9.7; 95% confidence interval [CI]: 4.6 - 20.4; p < 0.001), the researchers said. This association was consistent across all strata of LVEF, they added, noting that epicardial LGE, transmural LGE, and combined septal and free-wall LGE were all associated with heightened risk.

    LVEF was another independent predictor, they said (HR: 0.96; 95% CI: 0.93 - 0.98; p < 0.001).

     

    A new algorithm

    In view of the “highly significant association” with the primary endpoint for both LGE and LVEF, the researchers considered which LVEF stratification could best describe the arrhythmic risk in the cohort and evaluated the performance of predictive models combining LGE status and LVEF strata.

    They devised a simplified clinical algorithm by grouping patients into four categories of arrhythmic risk, which they said could be of relevance in clinical practice:

    1) Low risk (yearly event rate of 0.2%): Patients without LGE (LGE-) but with LVEF >20%.

    2) Intermediate to low risk (yearly event rate of 1.6%): Patients with LGE (LGE+), without high-risk LGE distribution and with LVEF >35%.

    3) Intermediate to high risk (yearly event rate of 2.8%): LGE+ patients with high-risk LGE distribution and LVEF >35%, as well as LGE- patients with LVEF ≤20%. 4) High-risk (yearly event rate of 7.2%): LGE+ patients with LVEF ≤35%.

    The researchers said the algorithm was “significantly superior to LVEF with the 35% cutoff” (Harrell’s C statistic: 0.8 vs. 0.69; area under the curve: 0.82 vs. 0.7; p < 0.001), and had “excellent predictive ability” with “easy clinical applicability.”

    Applied, the algorithm reclassified the arrhythmic risk of 34% of patients with DCM.

    It identified a “true low-risk group,” representing 54% of the study who experienced just 5% of the total arrhythmic events. “By contrast, the supposedly low-risk group of patients with LVEF >35% experienced 24% of the total arrhythmic events,” the researchers noted.

    While LGE-negative patients with LVEF 21% to 35% had low risk (annual event rate 0.7%), those with high-risk LGE distributions and LVEF of more than 35% had significantly higher risk, with a 3% annual event rate ( p = 0.007).

    “In other words, the low-risk patients of the new algorithm had significantly lower risk compared with the supposedly low-risk group identified by the >35% LVEF cut off (p = 0.007).

    “In addition, the high-risk patients of the new risk stratification model had significantly higher arrhythmic risk compared with the high-risk group selected by the ≤35% LVEF cut off (p = 0.009).”

    The researchers concluded that the algorithm “might help in refining the selection of patients for primary prevention ICD”.

    They added that randomized trials are now needed to enhance identification of patients with non-ischemic DCM who benefit from implantation of automatic defibrillators.

    In an accompanying editorial, Anne B. Curtis , MD, and Hassan A. Khan, MD, from the University of Buffalo, New York, stressed that the study does have limitations in that it has a retrospective design, “thus, despite strong internal validation and accuracy, applying this analysis to other patient populations may yield different results”.

    Nevertheless, they noted that the study adds to the increasing evidence that the presence, amount and location of LGE on cardiac magnetic resonance imaging provides “important information” that can be used with LVEF to determine the DCM patients most likely to have life threatening arrhythmic events in follow-up.

    “Further clinical trials should help clarify whether patients with LGE and more preserved LV function should be offered ICDs,” they said.

    “Equally important, determining if and when (sudden cardiac death) risk is low enough to forego ICD implantation in patients with DCM and heart failure is a critical subject for future guideline consensus and shared decision-making with patients.”

     

    Sources:

    Di Marco A, Brown PF, Bradley J, et al. Improved Risk Stratification for Ventricular Arrhythmias and Sudden Death in Patients With Nonischemic Dilated Cardiomyopathy. J Am Coll Cardiol 2021;77:2890-905.

    Curtis AB, Khan HA. Refining the Approach to Risk Stratification in Patients With Dilated Cardiomyopathy. J Am Coll Cardiol 2021;77:2906-8.

    Image Credit: D. P. Germain: Fabry disease. In: Orphanet journal of rare diseases Vol. 5, 2010, 30, PMID 21092187. PMC 300961. https://commons.wikimedia.org/wiki/File:Morbus_Fabry_MRI_01.jpg

This site uses cookies. By continuing to browse the site you are agreeing to our use of cookies. Review our Privacy Policy for more details