In order to bring you the best possible user experience, this site uses Javascript. If you are seeing this message, it is likely that the Javascript option in your browser is disabled. For optimal viewing of this site, please ensure that Javascript is enabled for your browser.
Did you know that your browser is out of date? To get the best experience using our website we recommend that you upgrade to a newer version. Learn more.
COVID-19 and Cardiology Read more

A novel clinical risk prediction model for sudden cardiac death in hypertrophic cardiomyopathy (HCM Risk-SCD)

This European multicenter study developed and validated a new risk prediction model for sudden cardiac death (SCD) in patients with hypertrophic cardiomyopathy (HCM). The prognostic model was based on data derived from a retrospective study of 3675 consecutive HCM patients from six different tertiary referral centers. Only adult HCM patients without a previous history of ventricular fibrillation or sustained ventricular tachycardia (VT) were included in the cohort. The follow-up period included 24313 patient years with a median follow-up period of 5.7 years (IQR: 2.8, 9.2) and during follow-up 198 patients (5%) reached the study endpoint defined as SCD, aborted SCD, or appropriate therapy from an implantable cardioverter defibrillator (ICD). The annual rate of SCD was 0.81% (95% CI: 0.71, 0.94) and the 5-year cumulative incidence of SCD of 3.8% (CI: 3.1, 4.5).

Based on previous risk stratification reports a list of eight predictor variables were defined: age at primary evaluation; family history of SCD; maximal left ventricular (LV) wall thickness (MWT); fractional shortening (FS); left atrial (LA) diameter; maximal LV outflow tract (LVOT) gradient; non-sustained VT; and unexplained syncope.
Patients with and without SCD endpoints were compared by univariable Cox regression analysis. The analysis showed that presence of non-sustained VT and unexplained syncope were highly significant predictors of SCD. In addition, LA diameter, MWT, a family history of SCD, and an increased LVOT gradient were also predictors of SCD while age and FS were not significantly associated with SCD (table below).

Predictor variable

SCD group

Hazard ratio

95% CI


Age (yearsa)

42 ± 15


0.979, 0.997


MWT (mma)

21.5 ± 6


1.025, 1.071


FS (%a)

41.0 ± 10


0.977, 1.008


LA diameter (mma)

46.2 ± 9


1.018, 1.052


LVOT gradient (mmHGb)

18 (6-58)


1.001, 1.008


Family history of SCD

73 (37%)


1.318, 2.235


Non-sustained VT

62 (31%)


1.849, 3.469


Unexplained syncope

52 (26%)


1.693, 3.195


amean ± standard deviation, bmedian (interquartile range)

Based on these risk estimates of hazard ratios a statistical model called “prognostic index” predicting the 5-year risk of SCD in an individual HCM patient was developed. Importantly, the clinical implications of using the model were examined in 3066 HCM patients with a complete data-set, and the prognostic model showed a very good correlation between predicted and observed probabilities of SCD. Furthermore, the results indicated that clinical use of the model would have minimized ICD implantation in patients not meeting the SCD endpoint. The authors suggested touse a cut-off value for SCD of ≥4% risk of SCD in 5 years. When this threshold was used, the model identified 60 (71%) of 84 SCD endpoints and resulted in ICD implantations in 30% of patients without SCD at 5 years. The authors concluded that for every 16 ICD implantations in HCM patients with ≥4% SCD risk, 1 patient will can potentially be saved from SCD at 5 years.


This investigation study is by far the largest risk stratification study ever conducted in HCM retrospective study of HCM patients with a quality of data, and since the cohort includes patients from different tertiary referral centers across Europe.  the results are of high clinical value and can be generalised the majority of European HCM patientsCompared with the establishedtraditional binary SCD risk factors in HCM patients (NSVT, severe left ventricular hypertrophy, unexplained syncope, family history of SCD, and abnormal blood pressure response), this model enables a more fairly accurate risk assessment afor SCD in HCM and, additionally, lowers the number of ICD implantations in patients not meeting the SCD endpoint. When the cut-off level of ≥4% SCD risk in 5 years was used the number of patients needed to treat (NNT) with ICD therapy to postpone one SCD endpoint in 5 years was 16. This number A NNT of 16 is comparable to the absolute 6.9% reduction in mortality (NNT 14) observed in the Sudden Cardiac Death in Heart Failure Trail (SCD-HeFT).
The best risk model for SCD in HCM currently available - enables fairly accurate risk assessment.



In conclusion, the results of the current HCM Risk-studySCD study will definitely have a major clinical impactimpact on management of HCM patients by improving risk assessment. Although it is demanding and difficult to conduct it would be desirable to repeat the study in yet another HCM cohort to provide the definitive proof of concept, and the next rational step will be validation the prognostic model in a different HCM cohort. Most valuable if the risk model could be validated in a different HCM cohort.

Notes to editor

O’Mahony C, Jichi F, Pavlou M, Monserrat L, Anastasakis A, Rapezzi C, Biagini E, Gimeno JR, Limongelli G, McKenna WJ, Omar RZ, Elliott PM, for the Hypertrophic Cardiomyopathy Outcomes Investigators. European Heart J 2013, epub October 14.

Commented by:
Torsten Bloch Rasmussen1, MD, PhD and Jens Mogensen2, MD, PhD, DMSc.
1 Department of Cardiology, Aarhus University Hospital, Denmark; 2 Department of Cardiology, Odense University Hospital, Denmark.
The content of this article reflects the personal opinion of the author/s and is not necessarily the official position of the European Society of Cardiology.