Sudden cardiac arrest (SCA), due mainly to coronary artery disease (CAD), is a leading cause of death. To identify electrocardiographic and clinical differences between patients with CAD with and without SCA, 87 victims of SCA with CAD were compared with 131 patients with CAD without SCA. Patients’ latest routine electrocardiograms and clinical variables were compared. Patients with CAD with and without previous myocardial infarctions (MIs) were included. Patients with SCA had a higher incidence of echocardiographic evidence of left ventricular hypertrophy and/or heart failure than controls. The median left ventricular ejection fractions for patients with SCA with and without previous MIs were 0.30 (interquartile range 0.24 to 0.41) and 0.41 (interquartile range 0.25 to 0.56). The median time between the last electrocardiographic assessment and SCA was 59 days (interquartile range 29 to 137). Regarding electrocardiographic characteristics, in patients with and without previous MIs, QRS width (odds ratio 1.032, 95% confidence interval 1.012 to 1.053, p = 0.002, and odds ratio 1.035, 95% confidence interval 1.015 to 1.056, p = 0.001) was the only significant predictor of SCA. In conclusion, in patients with CAD, regardless of a previous MI, a longer QRS width and echocardiographic parameters consistent with heart failure are associated with SCA, even in patients with ischemic cardiomyopathy currently not eligible for an implantable cardioverter-defibrillator.
Sudden cardiac arrest (SCA) is a leading mode of death in industrialized countries, frequently with ventricular fibrillation (VF) as the underlying arrhythmogenic mechanism. VF may occur in the acute and chronic phases of myocardial infarction (MI) or in ischemic conditions without apparent infarction. Using 12-lead electrocardiography, we recently showed longer conduction intervals, independent of the amount of ischemia, in patients with first ST-segment elevation MI (STEMI) developing ischemic VF.
Inhomogeneity of intramyocardial conduction velocity plays a role as a substrate for reentrant ventricular arrhythmias and sudden death during acute ischemia, and we found QRS widening related to the infarct location as a possible expression of such inhomogeneity of conduction velocity. To develop strategies to prevent SCA, it is important to correctly select patients at highest risk for SCA. In the present study, we studied victims of SCA with known coronary artery disease (CAD) and evaluated possible differences in clinical and electrocardiographic (ECG) characteristics, the latter possibly pointing to inhomogeneity of conduction velocity, already present before the event in nonischemic conditions.
Methods
A registry of victims of SCA, described previously, was used. Briefly, all victims of out-of-hospital SCA aged 20 to 75 years living in the Maastricht region were included. The Maastricht region is 203 km 2 in size and has approximately 181,500 inhabitants, of whom approximately 133,000 (73%) are aged 20 to 75 years. Served by 1 hospital, 1 emergency medical service, and a network of cooperative general practitioners, the region is suitable for population-based studies. All witnessed and unwitnessed (when circumstances were suggestive of unexpected SCA) events of out-of-hospital SCA were included in the registry. Patients with circulatory arrest caused by trauma or intoxication or SCA occurring in the terminal phase of a chronic disease were not included. For all victims of SCA included in the registry, the last available routine ECG and clinical characteristics before the events were collected from the hospital, general practitioners, family members, ambulance staff members, or a combination of these.
For the present study, 87 victims of SCA from the SCA registry with known CAD were compared to 1.5 times as many randomly derived patients with CAD without SCA (controls), providing a total of 131 control patients. These control patients were random patients with CAD drawn from the patient population from the Department of Cardiology, applying the same age restriction, but not matched in another way with the patients with SCA.
To study the characteristics of patients with CAD without previous MIs, patients were stratified according to the ECG presence of a previous MI.
Clinical characteristics, including baseline characteristics, cardiovascular risk factors and history, use of medication, and echocardiographic characteristics including ejection fraction, left ventricular dimensions, valvular disease, and left atrial dilatation, were analyzed.
For SCA and control patients, the most recent standard 12-lead 10-second paper electrocardiogram, recorded during the last routine visit, was analyzed. The electrocardiograms were recorded using a Marquette MAC VU recorder, providing heart rate, PR interval, QRS width, QRS axis, and axis of the T wave in the frontal plane. Additional ECG parameters collected on the single electrocardiograms included rhythm, QT interval measured in lead V 5 , corrected QT interval (using Fridericia’s formula), signs of a previous MI according to the Minnesota criteria, bundle branch block, left, right, and biatrial hypertrophy, left ventricular hypertrophy, and presence of ventricular ectopic beats.
Data analysis was performed using SPSS for Windows release 16.0.1 (SPSS, Inc., Chicago, Illinois). Continuous variables are expressed as medians and interquartile ranges and categorical variables as percentages. For comparison of continuous variables, Student’s t tests for normally distributed data and Mann-Whitney U tests for non-normally distributed data were used. For comparison of categorical variables, Fisher’s exact tests were used. All statistical tests were 2 tailed, and p values <0.05 were considered statistically significant. ECG characteristics showing significant univariate relations with the occurrence of SCA, but lacking multicollinearity, were included in multivariate logistic regression. Variables were removed stepwise from the model when the p value exceeded 0.10. Variables with p values <0.05 in the final model were considered main effect predictors. The predictive accuracy of the final model is reported as the area under the receiver-operating characteristic curve. Cut-off values for ECG characteristics by which most patients with SCA can be correctly classified are identified by applying the Pythagorean theorem to receiver-operating characteristic curves, which is a mathematical determination of the cut-off value closest to sensitivity and specificity of 1.
Results
The median time between the last available electrocardiogram and SCA was 59 days (interquartile range 29 to 137). In 25 of the 87 patients with SCA (29%), the initial rhythm documented by the emergency medical services was VF. In 22 patients with SCA, other rhythms were documented, whereas in 40 patients with SCA, no initial rhythm was documented.
A total of 120 patients had electrocardiographically documented previous MIs, and of these patients, 34 showed signs of previous anterior MIs, 66 of nonanterior MIs, and 20 of anterior and nonanterior MIs.
Significantly different clinical characteristics are listed in Table 1 . SCA and control patients were not significantly different regarding baseline characteristics.
| Variable | Infarction on Electrocardiogram | |||||
|---|---|---|---|---|---|---|
| Yes | No | |||||
| SCA Patients (n = 59) | Controls (n = 61) | p Value ⁎ | SCA Patients (n = 28) | Controls (n = 70) | p Value ⁎ | |
| Hypertension | 50% | 27% | 39% | 20% | 0.073 | |
| No sports | 93% | 79% | 0.036 | 86% | 60% | 0.017 |
| Body mass index (kg/m 2 ) | 25 (23–27) | 27 (24–28) | 0.007 | 24 (22–27) | 26 (24–28) | 0.032 |
| Heart failure | 37% | 10% | <0.001 | 32% | 3% | <0.001 |
| Medications | ||||||
| Digoxin | 27% | 5% | 0.001 | 25% | 4% | 0.005 |
| Diuretics | 54% | 15% | <0.001 | 39% | 13% | 0.006 |
| Nitrates | 58% | 44% | 0.150 | 68% | 44% | 0.045 |
| Antiplatelet agents | 20% | 51% | 0.001 | 39% | 63% | 0.043 |
| Echocardiography | ||||||
| Left atrial dilation | 48% | 29% | 0.051 | 46% | 18% | 0.015 |
| Left ventricular hypertrophy | 31% | 11% | 0.018 | 39% | 18% | 0.058 |
| Left ventricular dilation | 18% | 6% | 0.074 | 27% | 2% | 0.001 |
| Tricuspid regurgitation | 41% | 24% | 0.068 | 46% | 17% | 0.007 |
| Left ventricular ejection fraction (%) | 30 (24–41) | 45 (38–59) | <0.001 | 41 (25–56) | 60 (45–64) | <0.001 |
| Left ventricular end-diastolic diameter (mm) | 55 (52–64) | 50 (46–57) | <0.001 | 56 (48–63) | 49 (46–52) | 0.002 |
| Left ventricular end-systolic diameter (mm) | 42 (38–51) | 34 (29–38) | <0.001 | 38 (32–51) | 32 (28–34) | <0.001 |
⁎ Student’s t test, Mann-Whitney U test, or Fisher’s exact test as appropriate.
Echocardiographic data were available in 111 of 120 patients with previous MIs and in 86 of 98 patients without previous MIs. Given the large number of significantly different variables relative to the number of patients with SCA, no multivariate analysis was applied for the clinical characteristics.
Table 2 lists the significantly different ECG variables, as well as the nonsignificant difference in corrected QT intervals.
| Variable | Infarction on Electrocardiogram | |||||
|---|---|---|---|---|---|---|
| Yes | No | |||||
| SCA Patients (n = 59) | Controls (n = 61) | p Value ⁎ | SCA Patients (n = 28) | Controls (n = 70) | p Value ⁎ | |
| Heart rate (beats/min) | 76 (69–96) | 74 (67–86) | 0.128 | 75 (68–87) | 68 (58–76) | 0.020 |
| QRS width (ms) | 112 (100–124) | 92 (85–105) | <0.001 | 100 (89–146) | 94 (86–102) | 0.013 |
| Fridericia corrected QT interval (ms) | 424 (393–461) | 396 (382–424) | 0.085 | 412 (383–428) | 399 (388–413) | 0.735 |
| Left bundle branch block | 10% | 2% | 0.059 | 29% | 4% | 0.002 |
| Right bundle branch block | 9% | 7% | 0.741 | 14% | 6% | 0.220 |
| T axis intermediate | 37% | 72% | <0.001 | 50% | 88% | <0.001 |
| T axis left | 20% | 7% | 0.033 | 0% | 0% | NA |
| T axis right | 27% | 21% | 0.525 | 25% | 10% | 0.106 |
| T axis extreme right | 15% | 0% | 0.001 | 25% | 1% | 0.001 |
| Left atrial hypertrophy | 52% | 30% | 0.021 | 39% | 17% | 0.054 |
| Left ventricular hypertrophy | 17% | 5% | 0.042 | 21% | 16% | 0.561 |
| Sinus rhythm | 88% | 100% | 0.006 | 89% | 96% | 0.344 |
| Atrial fibrillation | 10% | 0% | 0.012 | 11% | 4% | 0.344 |
| Ventricular ectopic beats | 25% | 7% | 0.006 | 25% | 0% | <0.001 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
