Assessment of Noninvasive Methods to Identify Patients at Risk of Serious Arrhythmias After a Heart Attack

Background. Sudden cardiac death (SCD) kills >450,000 North Americans each year. Patients with a history of myocardial infarction (MI) are at particular risk. Concurrent alterations in myocardial structure \& autonomic tone appear important for the development of the arrhythmias leading to SCD.

Given the grave consequences of SCD, an ideal testing procedure should identify most of those at risk (sensitive) \& correctly classify risk (accurate). Since >90% of patients who suffer serious arrhythmias post-MI have at least mild left ventricular (LV) dysfunction (ejection fraction [EF] <0.50) this is an optimal group to study.

While noninvasive tests have been developed to estimate SCD risk, prior approaches have failed to: 1) identify the majority of patients at risk for serious arrhythmias (insensitive), 2) evaluate temporal changes in parameters, 3) identify the optimal timing for risk assessment post-MI, \& 4) develop a widely-applicable screening tool. This has resulted in a failure to delivery effective therapies (e.g., defibrillator) in a cost-effective manner.

Hypotheses. Primary: Concurrent evaluation of electrical structure \& autonomic tone will accurately identify most post-MI patients at risk of serious arrhythmic events. Secondary: 1) assessment later (16 weeks) provides more prognostic information than assessment early post-MI (4 weeks), 2) a single multi-parameter test procedure can be developed, and 3) individually, repolarization alternans provides the most prognostic information.

Methods. 350 persons with a recent MI (<31 days) \& EF <0.50 will undergo testing early (4 weeks), intermediate (8 weeks) \& late (16 weeks) post-MI.

Four techniques assess cardiac structure (spectral T-wave alternans [TWA], modified moving average TWA; signal-averaged [SA] ECG) \& nuclear ejection fraction. Three others evaluate autonomic tone (baroreceptor sensitivity [BRS], heart rate variability [HRV], and Heart Rate Turbulence [HRT]).

Data Collection \& Outcomes. Patients will be recruited over 24 months \& followed biannually for four years. Committee (blinded) endpoint classification \& central laboratory data analysis will be utilized. A composite of resuscitated cardiac arrest and cardiac mortality is the primary outcome. The components (resuscitated / non-resuscitated cardiac arrest and cardiac death) are secondary outcomes.

Statistical Aspects \& Sample Size. Standard methods of description \& analysis will be used. The capacity to accurately identify most patients at risk for serious arrhythmias will be evaluated using Cox multivariate models. The primary model will include age, sex, EF at 8 wks \& important baseline medication use.

Since multivariate modeling requires lower (more sensitive) dichotomy limits the following will be used: spectral TWA positivity will be defined as a non-negative test. SA-ECG QRS width >104 msec will be labeled as abnormal. For HRV, SDNN values <105 msec will be considered abnormal. For BRS, values <6.1 msec per mmHg will indicate impairment. For HRT abnormalities in T-onset \& / or T-slope will be considered abnormal. Receiver operating characteristic curves will be used to identify a cut-point for modified moving average TWA.

Assuming a 5 year 20% rate of the composite arrhythmias in patients with positive test results, we have 85% power to detect a 2.5-fold higher risk in patients with abnormalities than those without these abnormalities.

Relevance. This is the first large prospective study to evaluate the utility of concurrent structural \& autonomic tone assessment in predicting the development of serious arrhythmias after an MI.