MUSIC (Sudden Cardiac Death in Chronic Heart Failure)

Abstract

The MUSIC (MUerte Subita en Insuficiencia Cardiaca) study is a prospective, multicentre, longitudinal study designed to assess risk predictors of cardiac mortality and sudden cardiac death (SCD) in ambulatory patients with chronic heart failure (CHF).

The study population consisted of 992 patients with CHF consecutively enrolled from the specialized HF clinics of eight University Spanish Hospitals between April 2003 and December 2004, and followed up for a median of 44 months (until November 2008).  All patients had a 12-lead electrocardiogram (ECG),  a 24 h, 2-(4%) or 3-lead (96%) Holter ECG, chest X-ray, echocardiography, and blood laboratory parameters performed at enrolment.

Primary outcomes were cardiac death, either sudden cardiac death (SCD) or pump failure death (PFD) at the end of the follow-up period.

Background

The MUSIC (MUerte Subita en Insuficiencia Cardiaca) study, is a prospective, multi-center, longitudinal study designed to assess risk predictors of cardiac mortality and sudden cardiac death (SCD) in ambulatory patients with chronic heart failure (CHF) [1,2]. The study population consisted of 992 patients with CHF consecutively enrolled from the specialized heart failure clinics of eight University Spanish Hospitals between April 2003 and December 2004, and followed up a median of 44 months (or upon November 2008).  All patients had a 12-lead ECG,  24 h, 2-(4%) or 3-lead (96%) Holter ECG, chest X-ray, echocardiography, and blood laboratory parameters performed at enrollment. Primary outcomes were cardiac death, either sudden cardiac death (SCD) or pump failure death (PFD) at the end of the follow-up.

There have been several studies involving the MUSIC database up to date. The dataset serves as a valuable resource for developing and evaluating a wide range of prognostic ECG-derived biomarkers, focused on quantifying ventricular repolarization heterogeneity and autonomic function throughout heart rate variability and other biomarkers. In particular, the QT-interval adaptation to changes in heart rate has been assessed in terms of the QT/RR and T-peak-to-end/RR slopes, and found to be associated with both total mortality and SCD [3,4].

The index of average alternans (IAA), the long-term averaging of T-wave alternans (TWA) activity, was for the first time proposed and evaluated in the MUSIC database, and resulted to be a strong, independent predictor of SCD [5,6]. The T-wave morphology restitution (TMR) index represents another novel electrocardiogram marker that measures morphological changes in the T-wave as a response to heart rate variations, and whose clinical value was evaluated for the first time in MUSIC study. TMR specifically predicted SCD with no association with PFD [7,8].

Additionally, the results derived from MUSIC database are completed by studies on the assessment of the autonomic nervous system using heart rate information. New developed non-linear segmented symbolic dynamics (SSD) method was investigated for the suitability for risk stratification in the sub-cohort of patients with ischemic cardiomyopathy (ICM) in comparison to other heart rate variability (HRV) indices. The SSD enhances considerably risk stratification in ICM patients [9,10].

The standard heart rate turbulence (HRT) parameters, turbulence onset (TO) and slope (TS), as well as HRT categories, were also assessed for predicting total mortality and sudden death, showing that HRT is a potent risk predictor for both heart failure and arrhythmic death in patients with New York Heart Association (NYHA) class II and III [11,12]. Sympathetic modulation of ventricular repolarization, assessed by periodic repolarization dynamics (PRD) was shown to be an univariate predictor of SCD. Nonetheless, the combination of PRD with other ECG-based indices, such as TS and the IAA, improved SCD prediction. PRD could only predict PFD when combined with TS [13]. 

Restricting the analyses to the preserved left ventricular ejection fraction (LVEF > 35%) subcohort, in which risk stratification performance remains particularly challenging, traditional electrocardiographic variables, including HRV, HRT, and repolarization dynamics were evaluated through Holter monitoring. Assessing a combination of standard deviation of all normal-to-normal RR intervals, TS, and QT/RR parameters can predict increased total mortality and SCD risk in patients with CHF and LVEF>35% [14]. Moreover, risk models combining ECG markers and clinical variables to specifically assess the risk of SCD and PFD have been developed, concluding that integrating ECG markers capturing proarrhythmic and pump failure mechanisms with standard clinical variables enhances the accuracy of predicting adverse outcomes in CHF patients [15].

Finally, the MUSIC database has served as a proof of concept for the development of new signal processing techniques to overcome the highly irregular ventricular response and the presence of f-waves during atrial fibrillation, allowing the assessment of ventricular repolarization instability through long-term beat averaging, or using a multilead strategy based on periodic component analysis (πCA) before QT delineation, showing the prognostic value of the proposed indices [16,17]. On the other hand, studies based on high-resolution ECGs revealed that reduced atrial fibrillatory rate (AFR) was associated with an increased risk of death because of heart failure progression and reduced irregularity of RR intervals during atrial fibrillation (AF) was an independent predictor of all-cause mortality, SCD and PFD [18,19].

Methods

All patients had symptomatic CHF (NYHA class II–III) and were treated according to institutional guidelines. The study included patients with either depressed (<45%) or preserved LVEF (>45%). The latter were included if they had HF symptoms and a prior hospitalization for HF or some objective signs of HF confirmed by chest X-ray (findings of pulmonary congestion) and/or echocardiography (abnormal left ventricular (LV) filling pattern and LV hypertrophy). Patients were excluded if they had recent acute coronary syndrome or severe valvular disease amenable for surgical repair. Patients with other concomitant diseases expected to reduce life-expectancy were also excluded [1].

A follow-up period was conducted for a median of 44 months (until November 2008), including periodic visits every 6 months. At the end of the follow-up period, the study group included 94 SCDs, 111 deaths due to a different cardiac origin (PFD), 20 cardiac transplantations, 61 non-cardiac deaths, and 695 survivors. There were 11 patients lost during follow-up.

SCD was defined as (i) a witnessed death occurring within 60 min from the onset of new symptoms unless a cause other than cardiac was obvious; (ii) an unwitnessed death (<24h) in the absence of preexisting progressive circulatory failure or other causes of death; or (iii) a death during attempted resuscitation. PFD was defined as death occurring in a hospital as a result of refractory progressive end-stage heart failure.

The study protocol was approved by institutional investigation committees, and all patients signed informed consent. Protected health information (PHI) in the MUSIC database has been de-identified according to the GDPR guidelines.

Data Description

Clinical and demographic information, radiographic, echocardiographic, laboratory variables and medication are summarized in the CSV file (subject-info.csv). The spreadsheet is completed with the date of enrollment and exit of the study, together with mortality codes. The codification of categorical variables are described on a second spreadsheet (subject-info_codes.csv), and the definition of column headings in a third one (subject-info_definitions.csv).

Each 24-hour Holter recording in the database comprises a data file, named PXXXX.dat, containing the ECG signal, and the corresponding header PXXXX.hea, with information about the recording format, all contained in Holter_ECG folder. Similarly, in High-resolution_ECG folder, each high-resolution, 12-lead, short ECG recording comprises the data files PXXXX_H.dat, containing the ECG signal, and the corresponding  PXXXX_H.hea, with the header information. The full filename list of the 936 Holter ECGs and the 687 high-resolution ECGs is available in the text file RECORDS.

Usage Notes

By making this dataset available, we aim to empower the research community to create improved prognostic models for SCD and PFD risk prediction. As discussed in the Background above, there have been several studies involving the MUSIC database up to date. The paper [2] provides additional information on the dataset and the data collection process, as well as more detailed information on previous studies carried out on the MUSIC database.

Release Notes

Version 1.0.0: First release.

Ethics

In the MUSIC study, patients were consecutively enrolled from the specialized heart failure clinics of eight University Hospitals between April 2003 and December 2004. The study protocol was approved by institutional investigation committees, and all patients signed informed consent.

Acknowledgements

Patients in the MUSIC study were recruited thanks to the contribution of the following Spanish Hospitals: Arrixaca Hospital (Murcia, Clinical Investigator, CI: Mariano Valdés), Gran Canaria Insular Hospital (Las Palmas, CIs: Vicente Nieto, Ricardo Huerta), Gregorio Marañón Hospital (Madrid, CIs: Roberto Muñoz, Jesús Almendral, Marta Domínguez), Joan XXIII Hospital (Tarragona, CIs: Alfredo Bardají, Pilar Valdovinos), Santiago de Compostela University Hospital (Santiago de Compostela, CI: Pilar Mazón), Sant Pau Hospital (Barcelona, CIs: Agustina Castellví-Grisó, Maite Domingo, Mariana Noguero), Son Dureta Hospital (Mallorca, CI: Miquel Fiol, Carlos Fernández), and Valme Hospital (Sevilla, CIs: Juan Leal del Ojo, Antonio Fernández, Dolores García-Medina). Data management, storage and distribution was originally done at Polytechnic University of Cataluña (Barcelona, Principal Investigator: Pere Caminal).

Conflicts of Interest

Nothing to declare.

References

1. Vazquéz R, Bayés-Genis A, Cygankiewicz I, Pascual-Figal D, Grigorian-Shamagian L, Pavon R, Gonzalez-Juanatey JR, Cubero JM, Pastor L, Ordonez-Llanos J, Cinca J, de Luna AB; MUSIC Investigators. The MUSIC Risk score: a simple method for predicting mortality in ambulatory patients with chronic heart failure. Eur Heart J. 2009 May;30(9):1088-96. doi: 10.1093/eurheartj/ehp032.
2. Martín-Yebra A, Bayés de Luna A, Vázquez R, Caminal P, Laguna P, Martínez JP. The MUSIC Database: Sudden Cardiac Death in Heart Failure Patients. Computing in Cardiology, 2024
3. Cygankiewicz I, et al. Prognostic value of QT/RR slope in predicting mortality in patients with congestive heart failure. J Cardiovasc Electrophysiol. 19(10):1066-72, 2018.
4. Ramírez J, et al. QT/RR and T-peak-to-end/RR curvatures and slopes in chronic heart failure: Relation to sudden cardiac death, J Electrocardiol, 47:842–848, 2014.
5. Monasterio V, et al. Average T-wave alternans activity in ambulatory ECG records predicts sudden cardiac death in patients with chronic heart failure. Heart Rhythm, 9(3):383- 389, 2012.
6. Martín-Yebra A, et al. Post-ventricular premature contraction phase correction improves the predictive value of average T-wave alternans in ambulatory ECG recordings. IEEE Trans Biomed Eng. 65(3):635-644, 2018.
7. Ramirez J, et al. T-wave Morphology Restitution Predicts Sudden Cardiac Death in Patients with Chronic Heart Failure. JAHA. 6:e005310;1-12, 2017.
8. Palmieri F, et al. Weighted Time Warping Improves T-wave Morphology Markers Clinical Significance, IEEE Trans Biomed Eng. 69(9):2787-2796, 2022.
9. Voss A, et al. Segmented Symbolic Dynamics for Risk Stratification in Patients with Ischemic Heart Failure. Cardiovascular Engineering and Technology Volume 1:290–298, 2010.
10. Voss A, et al, Short-term vs. long-term heart rate variability in ischemic cardiomyopathy risk stratification. Front Physiol,4;363:1-15, 2013.
11. Cygankiewicz I, et al. Heart rate turbulence predicts all cause mortality and sudden death in congestive heart failure patients. Heart Rhythm. 5(8):1095-1102, 2008.
12. Martínez JP, et al. Detection performance and risk stratification using a model-based shape index characterizing heart rate turbulence. Ann Biomed Eng, 38(10):3173-3184, 2010.
13. Palacios S, et al. Periodic Repolarization Dynamics as Predictor of Risk for Sudden Cardiac Death in Chronic Heart Failure Patients. Sci Rep, 11:20546, 2021.
14. Cygankiewicz I, et al. Risk stratification of mortality in patients with heart failure and left ventricular ejection fraction >35%. Am J Cardiol. 103:1003-1010, 2009.
15. Ramírez J, et al. Sudden cardiac death and pump failure death prediction in chronic heart failure by combining ECG and clinical markers in an integrated risk model. PLOS ONE, 12(10):e0186152;1-15, 2017.
16. Martín-Yebra A, et al. Quantification of Ventricular Repolarization Variation for Sudden Cardiac Death Risk Stratification in Atrial Fibrillation. IEEE J Biomed Health Inform. 23(3):1049-1057, 2019.
17. Martín-Yebra A, S¨ornmo L, Laguna P. QT interval Adaptation to Heart Rate Changes in Atrial Fibrillation as a Predictor of Sudden Cardiac Death. IEEE Trans Biomed Eng. 69(10):3109-3118, 2022.
18. Platonov P, et al. Low atrial fibrillatory rate is associated with poor outcome in patients with mild to moderate heart failure. Circ Arrhythm Electrophysiol, 5: 77–83, 2012.
19. Cygankiewicz I, et al. Reduced Irregularity of Ventricular Response During Atrial Fibrillation and Long-term Outcome in Patients With Heart Failure. Am J Cardiol, 116(7):1071-5, 2015.