Methods

This was a retrospective study including all patients consecutively discharged from our hospital with diagnosis of acute myocardial infarction (AMI) from 1/2004 to 10/2010. We used the joint consensus document of the European Society of Cardiology and the American College of Cardiology for the standard diagnostic of AMI. From 4,371 patients surviving to in-hospital phase, 494 patients were identified as having LV dysfunction, defined arbitrarily as LVEF <40% at discharge on 2-dimensional echocardiography. Patients with pacemaker (n = 23) were excluded to analyze the intrinsic QRS-T angle. Thus, the final cohort was composed of 471 patients ( Figure 1 ). The study complies with the Declaration of Helsinki and was approved by the Clinical Research Ethics Committee of our hospital.

Flowchart of the patients enrolled in the study.

Demographic, clinical, and angiographic data, as well as data on management and follow-up, were prospectively collected and recorded in an electronic database. Baseline clinical variables were used to evaluate the prognostic value for predicting all-cause mortality.

For each patient, we took the electrocardiogram (ECG) at admission, similar to the methodology used in the Evaluation of Methods and Management of Acute Coronary Events study. We analyzed the computerized values of QRS and T axes. The absolute difference between the frontal QRS and frontal T-wave axes was calculated as T-wave axis − QRS axis and if >180° was subtracted from 360° to give a continuous variable ranging from 0° to 180°.

We defined the primary end point as all-cause mortality. Underlying and contributing causes of death were recorded. We used national death certification data and hospital chart or physician’s records to identify all deaths. Cause of death for each patient was classified by 2 clinicians and supported on International Classification of Diseases, 10th revision, codes. We categorized the etiology of death into 5 groups ( Figure 2 ): cardiac (acute coronary syndrome, heart failure, valvular heart disease, and cardiac arrhythmia), vascular (cerebrovascular disease and peripheral artery disease), infection-related death, malignant disease, and other causes.

Distribution of deaths by etiology.

During follow-up, the occurrence of appropriate implantable cardioverter defibrillator (ICD) therapy was also noted. In those patients, an electrophysiologist determined whether or not the ICD therapy was appropriate. All therapies, either antitachycardia pacing or shock, were classified as appropriate when they occurred in response to life-threatening arrhythmias (ventricular tachycardia or ventricular fibrillation).

The statistical analyses were performed with SPSS 20.0. The categorical or dichotomous variables are expressed as absolute values and percentages and were compared with the Pearson’s chi-square test. The continuous variables are described as mean ± SD or as median and interquartile range. Student t or Mann-Whitney U test was used for the comparisons of continuous variables between 2 groups of patients, as appropriate.

A survival analysis was performed for each end point: (1) follow-up all-cause death and (2) a composite end point of cardiac death and first appropriate device therapy, whichever occurred first. The cutoff point of 90° for the frontal QRS-T angle was defined according to a receiver-operating characteristic curve (optimal QRS-angle to predict mortality). Cumulative event rates of end points were analyzed by the method of Kaplan-Meier (log-rank test). A Cox proportional hazards model was used to estimate the hazard ratios (HRs) and 95% confidence intervals (CIs) and assess the performance of the QRS-T angle in a multivariable model. In this model, we included those variables that resulted significant predictors of mortality in the univariable model. A p value <0.05 was considered statistically significant.

Results

Complete follow-up data were available for 99.1% (n = 467) of the 471 patients over a median follow-up time of 3.9 years (interquartile range 2.1 to 5.9). The mean age was 70.0 ± 12.5, with 24.6% women and 35.3% diabetics. The rate of ST-segment elevation myocardial infarction (STEMI) was 47.3%, and the mean of LVEF was 34.4 ± 5.8, with 51.6% of patients in Killip class ≥II. From those 457 patients, 217 (47.5%) died during follow-up. Baseline characteristics according to survivors and nonsurvivors are listed in Table 1 .

The baseline frontal QRS-T angle was 95.9 ± 57.3°. It was higher in patients who died (116.6 ± 52.8 vs 77.9 ± 55.1, respectively, p <0.001). In receiver-operating characteristic curves ( Figure 3 ), the most accurate value of frontal QRS-T angle to predict all-cause cardiac death was 90°, with an area under the curve of 0.690 ± 0.025 (sensitivity and specificity of 73.3% and 62.2%, respectively). In 251 patients (53.7%), the frontal QRS-T angle was higher than 90°, with a higher rate of mortality in comparison with patients with QRS-T angle ≤90° (73.3% vs 26.7%, respectively, p <0.001; Figure 4 ). As listed in Table 2 , patients with a wide frontal QRS-T angle (>90°) were more likely to be older, diabetic, with previous coronary artery disease, previous heart failure, and peripheral artery disease, to present with higher rate of non–ST-segment elevation myocardial infarction, to have a lower LVEF, and to have a more percentage of left bundle branch block. After adjusting by those variables that were associated with follow-up death in univariate analysis ( Table 3 ), automatic frontal QRS-T angle remained as an excellent predictor of all-cause and cardiac deaths (HR 1.006, 95% CI 1.003 to 1.009, p <0.001), increasing the mortality 6% per each 10° ( Figure 5 ). A frontal QRS-T angle >90° increases the risk of all-cause mortality by twofold (HR 2.180, 95% CI 1.558 to 3.050, p <0.001), after adjusting by the previously described variables.

Area under the curve (AUC) for frontal QRS-T angle to predict all-cause follow-up mortality.

Survival Kaplan-Meier curves for follow-up all-cause mortality and for cardiac mortality and ICD appropriate therapy according to QRS-T angle ( red : >90°; green : ≤90°).

Table 3

Multivariate analysis to predict follow-up mortality

Variable	HR	95% CI	p Value
Age (yrs)	1.053	1.037–1.070	<0.001
Diabetes mellitus	1.237	0.931–1.645	0.143
Previous coronary artery disease	1.363	1.013–1.835	0.041
Previous heart failure	0.983	0.676–1.429	0.927
Peripheral artery disease	1.104	0.762–1.599	0.600
Chronic obstructive pulmonary disease	1.399	0.984–1.988	0.061
Previous malignant disease	1.401	0.938–2.092	0.099
STEMI	1.300	0.926–1.825	0.130
Killip ≥2	1.039	0.761–1.419	0.809
LVEF (%)	0.984	0.961–1.007	0.168
Troponine I peak (ng/dl)	0.999	0.997–1.001	0.221
Atrial fibrillation	0.874	0.612–1.249	0.461
Left bundle branch block	1.023	0.677–1.547	0.914
QRS-T angle (per each degree)	1.006	1.003–1.009	<0.001
Anterior STEMI	0.880	0.554–1.397	0.587
Percutaneous coronary intervention	0.706	0.501–0.994	0.046
Coronary artery bypass grafting	0.678	0.305–1.511	0.342
MDRD-4 (ml/min/1.73 m 2 )	0.992	0.986–0.998	0.011
Aspirin	0.697	0.466–1.040	0.077
Clopidogrel	1.467	0.998–2.155	0.051
β blocker	0.701	0.522–0.942	0.019
Angiotensin-converting enzyme inhibitors/angiotensin receptor blockers	0.930	0.658–1.315	0.682
Antialdosteronic	1.107	0.784–1.564	0.564
Statin	0.611	0.432–0.864	0.005

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. Comparison of Lipid Management in Patients With Coronary Versus Peripheral Arterial Disease
2. Impact of High-Dose Atorvastatin Therapy and Clinical Risk Factors on Incident Aortic Valve Stenosis in Patients With Cardiovascular Disease (from TNT, IDEAL, and SPARCL)
3. Left Ventricular Systolic Function Following Alcohol Septal Ablation for Symptomatic Hypertrophic Cardiomyopathy
4. Cardiotoxicity Research in Breast Cancer Patients: Past and Future

Stay updated, free articles. Join our Telegram channel

Join