Low sudden cardiac death risk in hypertrophic cardiomyopathy: ‘A wolf in a lamb’s skin’?

This editorial refers to ‘Long-term risk of sudden cardiac death in hypertrophic cardiomyopathy—a cardiac magnetic resonance outcome study’, by S. Greulich et al. pp. 732--41

The LGE-CMR method (late gadolinium-enhanced—cardiac magnetic resonance) is based on the principle that a tissue with an expanded extracellular space has a larger distribution volume for conventional CMR contrast agents. After intravascular administration, these extravascular and extracellular agents are distributed in the extracellular myocardial space. Minutes later, large differences between tissue with normal and expanded extracellular volumes are seen, and LGE acquisition is performed. Currently used LGE methods provide a very high spatial resolution and very good contrast to noise ratio, allowing the depiction of small amounts of myocardial fibrosis.¹

In hypertrophic cardiomyopathy (HCM) patients, LGE-fibrosis is frequent (present in about two-thirds of patients) and progressive along time. Two major location patterns are seen: intramural LGE, in the hypertrophied segments, thought to correspond to replacement fibrosis (secondary to microvascular ischaemia, leading to myocyte death); and right ventricular (RV) insertion points LGE, thought to correspond to interstitial fibrosis and/or disarray (without cellular necrosis).¹

The pathophysiological consequences of these different types of fibrosis and their prognostic role are also different. While RV insertion point fibrosis has not been shown to have prognostic impact, the presence of intramural fibrosis is strongly and independently associated with adverse outcomes in HCM patients. However, its usefulness as a binary tool (presence or absence of LGE) is clinically impractical for clinical decision-making, because up to 70% of all patients with HCM have some degree of LGE on CMR.

To overcome this limitation, quantitative LGE (the amount of LGE in terms of percentage of myocardial mass) has emerged as a tool for risk stratification in HCM. Accordingly, the extent of LGE has been associated with increased risk of sudden cardiac death (SCD) (fibrosis as the anatomical substrate for re-entry complex ventricular arrhythmias) and also with increased risk of all-cause mortality, cardiovascular-mortality, and heart failure-death [fibrosis as a surrogate of non-contractile myocardium, with left ventricular (LV) systolic dysfunction].²

SCD is a major complication of HCM and its prevention, with adequate risk assessment, is essential in the clinical management of these patients.

In the secondary prevention setting, ICD (implantable cardioverter-defibrillator) implantation is clearly indicated in patients resuscitated from SCD or with sustained ventricular tachycardia (VT). However, in the primary prevention setting, the decision is much more complex, so that individualized approaches, based on two scores are currently used:

− American Heart Association-American College of Cardiology (AHA-ACC) scores: the classical score includes five major risk factors,³ dichotomously evaluated [family history of SCD, syncope, massive left ventricular hypertrophy (LVH), non-sustained VT, abnormal blood pressure evolution with exercise]. The clinical decision regarding the implantation of ICD is classically based on the number of risk factors: zero: follow-up; one risk factor: individualized decision; more than one risk factor: ICD. This conventional score has been recently upgraded (‘enhanced ACC/AHA strategy’⁴), including now seven risk factors, four of the classical ones (exercise blood pressure was excluded) plus three ‘new’ ones (apical HCM with scar, LGE >15%, and LV systolic dysfunction). Additionally, this new strategy includes positive referees (exercise blood pressure, intermediate amounts of LGE, LV outflow tract obstruction) and a negative referee (age > 60).⁴

−European Society of Cardiology (ESC)—HCM Risk-SCD score for estimation of the 5-year risk of SCD⁵: this strategy includes seven risk factors, some evaluated in a binary way and others continuously (age, family history of SCD, syncope, non-sustained VT, anterior-posterior left atrium diameter, maximum LV outflow tract gradient, maximal wall thickness). If the risk is low (<4%), ICD is not indicated. If the risk is intermediate (4–6%) ICD may be considered; when it is high (>6%) an ICD should be considered.

It is important to acknowledge that none of these scores is perfect and both should evolve and be continuously improved and refined,⁶ with the inclusion of new risk factors and referees, or by fine-tuning the already established ones. In this context, quantitative LGE has been increasingly used to enhance their discriminative power.

Greulich et al.⁷ have assessed 203 consecutive patients with HCM at baseline and after a long follow-up period (median10.4 years).

At baseline, this population was a low ESC-HCM SCD risk population, individually and as a whole (all patients had a 5-year SCD risk <4%; median ESC-SCD risk of the population 1.84%).

The authors have shown that after 5 years of follow-up the ‘real world SCD rate’ of this population was 2.3%, demonstrating that the ESC—5 years risk score worked well in this group of low-risk patients; however, the same did not happen in the medium and long term, as the SCD rate of this same ‘low risk population’ was high (4.8% at 10 years; 15.7% at 15 years).

Remarkably, Greulich and colleagues have shown that in this population, a baseline amount of LGE >5% of LV mass was clinically useful for SCD prediction in the long term (SCD prevalence of 5.5% at 5 years; 13.0% at 10 years; 33.3% at 15 years). According to the authors, though higher cut-offs of LGE amount could also have been used, a low threshold of LGE amount is more realistic and reasonable in the clinical arena, as the number of HCM patients with massive amounts of LGE is rare.

In opposition, patients without LGE or with an amount ≤5% had favourable and similar good prognosis during follow-up (remarkably, the negative predictive value for the ≤5% cut-off is very high and similar to the patients with no LGE).

Overall, Greulich et al. conclude that in low-risk populations, the ESC score works well in the short term of 5 years, i.e. during the time interval it was designed to work; however, in the medium and long term, patients with LGE >5% at baseline should be carefully monitored because of an important SCD risk.

Though the authors should be acknowledged by this elegant study, some of their major conclusions (as they admit) must be carefully interpreted and confirmed in further studies before clinical application.

First, as it is clearly stated in its designation, the ‘ESC-5 years risk score’ only predicts risk at 5 years and should be repeated over time, because risk factors may change (and they do change) over time (age, family history of SCD, syncope, non-sustained VT, anterior-posterior left atrium diameter, maximum LV outflow tract gradient, maximal wall thickness—they all may change) modifying the risk. And the same applies to the enhanced AHA-ACC risk score, whose parameters also may evolve (and they do evolve) with time (family history of SCD, syncope, massive LVH, non-sustained VT, apical HCM with scar, LGE >15% and LV systolic dysfunction—they all may change).

So, independently of the risk score at a specific point in time, individual risk should be periodically assessed. For instance, in this study, it would be interesting to find out if at 5, 10, and 15 years, the low-risk category at baseline remained in the long term.

Second, myocardial remodelling is a feature of HCM⁸ and fibrosis may progress over time, with potential implications on risk assessment and on the occurrence of events. Accordingly, this confounder factor may have biased the results. And in this study, it would be also interesting to repeat CMR to assess if the amount of LGE at baseline remained stable.

So, please consider to periodically repeat CMR to assess LGE and other risk factors at different times.

Third, these results only apply to low-risk populations and not to the difficult grey-zone intermediate risk patients (AHA-ACC 1 risk factor or 4–6% ESC risk score). So, please, do not derive these results to other risk populations…

Low risk for SCD in HCM?

Please, quantify LGE amount at baseline for long-term prognosis …. but please, don’t forget to repeat the risk stratification score overtime.

Conflict of interest: None declared.

The opinions expressed in this article are not necessarily those of the Editors of EHJCI, the European Heart Rhythm Association or the European Society of Cardiology.

References