More than 95% of patients with out-of-hospital cardiac arrest die during the episode. 30-50% of sudden cardiac deaths that are due to coronary artery disease occur as the first cardiac event. Only one third of sudden cardiac death occur in patients with known coronary artery disease or risk markers. That means that we are unable to predict about two thirds of the sudden cardiac death.

Problems with risk factors of sudden cardiac death

The so called risk factors lack of specificity, sensitivity, predictive accuracy. They are able to identify populations at risk but not the individual at risk. Present risk factors identify risk of developing sudden cardiac death rather than the proximate precipitator of sudden cardiac death. What we need is individual-specific predisposition: single patient probabilities, not population predictions, in order to prevent sudden cardiac death in an individual.

Familial clustering of sudden cardiac death

Familial clustering of sudden cardiac death is statistically distinct and separable from familial risks of myocardial infarction. Parental history of sudden cardiac death increased relative risk of sudden cardiac death to about 1.9 times. Relative risk rises up to 9.4 times when both maternal and paternal history of sudden cardiac death is present.

Risk factors for sudden cardiac arrest

Standard risk factors for sudden cardiac arrest

Standard risk markers are history of previous sudden cardiac arrest, prior ventricular tachycardia, decreased left ventricular ejection fraction, frequent premature ventricular complexes on coronary artery disease or during recovery from a treadmill test and electrophysiological parameters like corrected QT interval, QRS duration, heart rate variability, baroreceptor sensitivity, T wave alternans, signal averaged electrocardiogram, , QT dispersion and inducible ventricular arrhythmias during EP study.

Ejection fraction
Low ejection fraction predicts sudden cardiac death from various observational studies. In
MADIT II (Multicenter Automatic Defibrillator Implantation Trial), ICD was found to reduce mortality in patients with ejection fraction less than 30%. But ejection fraction lacks specificity as a predictor of arrhythmia events.

Ambient ventricular arrhythmias
PVCs predicted mortality of post-AMI patients in certain observational studies. But suppression PVCs by antiarrhythmic drugs does not reduce mortality, possibly because of the chance for proarrhythmia.

Non sustained ventricular tachycardia (NSVT) predicted mortality of post-AMI patients in observational studies. Multicenter Postinfarction Research Group study found a 2-fold increase in total and arrhythmic deaths. Patients with NSVT, low ejection fraction and inducibility of ventricular tachycardia benefit from ICD implantation as per the MADIT and MUSTT trials. But in GISSI-2 it was not an independent predictor of cardiac mortality.

QRS duration
A retrospective analysis of MADIT II suggests that prolonged QRS duration predicts arrhythmic mortality. But the prognostic power has not been tested in prospective trials.

QT dispersion
QT dispersion is the interlead difference in QT interval. It is a measure of inhomogeneity of repolarization. Controversial results from observational and case-control studies and methodological problems in measurement question the reliability of QT dispersion as a useful marker.

Signal averaged ECG
Normal signal averaged ECG (SAECG) has high negative predictive accuracy and predicted outcome of patients in the MUSTT trial. But the positive predictive accuracy is low. CABG Patch trial did not show a mortality benefit among patients with low ejection fraction and positive SAECG who underwent CABG.

T wave alternans
Positive T wave alternans (TWA) predicts arrhythmia events among high-risk patients. But the predictive power has not been tested in randomized control trials. Max predictive accuracy is at heart rates between 100 – 120. TWA can be measured during exercise, pharmacological stress, or atrial pacing. Overt TWA in the ECG is not common. Microvolt TWA detected by digital signal processing. TWA cannot be measured in patients with atrial fibrillation.

Autonomic markers: Heart rate variability (HRV) and baroreceptor sensitivity (BRS)
Low HRV and BRS predict both nonsudden and SCD in post-AMI patients. In the ATRAMI (Autonomic Tone and Reflexes after MI) study, the combination of NSVT, BRS and HRV increased the risk of death by 22 times.

In DINAMIT (The Defibrillator in Acute Myocardial Infarction Trial), ICD reduced arrhythmic death by more than 50% in the those with abnormal HRV (SDNN < 70 ms), heart rate more than 80 and ejection fraction less than 35. But the advantage was offset by a significant increase in non arrhythmic death.

Newer risk factors for sudden cardiac arrest

Newer risk markers for sudden cardiac arrest are markers of inflammation like CRP, molecular markers like beta receptor subtypes, genetic markers in inherited arrhythmias and heart rate turbulence.

High-sensitivity CRP
C reactive protein (CRP) is a nonspecific marker produced by in liver in response to multiple inflammatory challenges. CRP levels were highly predictive in identifying SCD victims in the Physician Health Study with 22 071 participants. Risk increased independent of traditional markers of CAD.

Heart rate turbulence
Hear rate turbulence (HRT) is the physiological, bi-phasic response of the sinus node to PVC. It is a short initial acceleration followed by a deceleration. It is probably an autonomic baro-reflex in response to the brief disturbance of blood pressure caused by the PVC i. e., the low amplitude of the premature beat and the high amplitude of the ensuing normal beat. If autonomic control is intact, it responds with heart rate turbulence, if impaired, HRT is weakened or missing.

Two important parameters of HRT are turbulence onset and turbulence slope. Turbulence Onset (TO) is the percentage difference between the RR interval immediately following PVC and the RR immediately preceding PVC. Positive values indicate deceleration and negative values indicate acceleration. Turbulence Slope (TS) corresponds to the steepest slope of the linear regression line for each sequence of five consecutive normal intervals in the local tachogram.

Three categories of HRT have been described:

Category 0: Normal TO and TS
Category 1: Either TO or TS abnormal
Category 2: Both TO and TS abnormal

HRT category 2 was as powerful as left ventricular ejection fraction as a risk predictor of SCD in one study.

Composite index of cardiac autonomic function

A composite index of cardiac autonomic function is calculated by combining the parameters from HRT, HRV and BRS (TO and TS, BRS and SDNN) has been shown to be the strongest risk predictor of various combinations studied in a study.

EP testing

Inducibility of VT has been shown to predict mortality in patients with clinical presentation of VT event or syncope plus low ejection fraction (0.35) in the MADIT study. CAD patients with asymptomatic NSVT and EF 0.40 and inducible VT has been shown to predict mortality in MUSTT. The invasive nature of EP testing limits it widespread use. The prognostic power of EP testing among patients without a clinical arrhythmia event is unknown.