The risk of death in heart failure (HF) is high. The electrocardiographic spatial QRS-T angle reflects changes in the direction of the repolarization sequence and predicts death in the general population. The frontal QRS-T angle is simple to measure but has not been evaluated in a large chronic HF cohort. We examined the significance of the frontal QRS-T angle in predicting the clinical outcome in a large cohort of patients with HF. The QRS-T angle was calculated from the frontal QRS and T axis of the baseline 12-lead surface electrocardiogram. The patients were followed for cardiac-related hospitalizations and death; 5,038 patients with HF were evaluated. The mean follow-up period was 576 days; 51% were men. Overall survival during the follow-up period was 83%. Cox regression analysis after adjustment for significant predictors, including age, gender, ischemic heart disease, hypertension, atrial fibrillation, body mass index, pulse, serum hemoglobin, sodium, estimated glomerular filtration rate, and urea levels, demonstrated that the QRS-T angle was an incremental predictor of increased mortality in both genders. For women, a QRS-T angle of ≥60° had a hazard ratio of 1.35 (95% confidence interval 1.04 to 1.75; p <0.05) and a QRS-T angle of ≥120° had a hazard ratio of 1.45 (95% confidence interval 1.10 to 1.92, p <0.01). For men, a QRS-T angle of ≥130° had a hazard ratio of 1.53 (95% confidence interval 1.14 to 2.06, p <0.01). For the whole cohort, a QRS-T angle of ≥125° gave a hazard ratio of 1.47 (95% confidence interval 1.20 to 1.80, p <0.0001). The QRS-T angle was also a predictor of increased cardiac-related hospitalizations. The QRS-T angle was a predictor in patients with reduced and preserved left ventricular function and in patients with a normal QRS interval. In conclusion, the QRS-T angle was a powerful predictor of outcome in patients with HF. We believe the QRS-T angle should be a part of the electrocardiographic evaluation of patients with HF.
Heart failure (HF) is a major epidemic and is associated with considerable morbidity and mortality. The 12-lead surface electrocardiogram (ECG) is a simple test available for monitoring patients with HF and provides information pertaining to mechanical and electrophysiologic cardiac pathologic features and conveys prognostic information. The electrocardiographic spatial QRS-T angle, defined as the angle between the directions of ventricular depolarization and repolarization, reflects underlying cardiac structural abnormalities and electrical heterogeneities causing abnormal changes in the direction of the repolarization sequence. It is also a powerful marker of electrical instability. The spatial QRS-T angle predicts cardiovascular death in the general population and emerged as the strongest electrocardiographic predictor of cardiovascular death, incident HF, and all-cause mortality in postmenopausal women. The frontal QRS-T angle, easily obtainable from the surface ECG, has been shown to be equivalent to the spatial QRS-T angle in the prediction of total mortality in the general population, incident cardiovascular disease in women, and sudden cardiac death in the general population. The frontal QRS-T angle has also been shown to be a predictor of arrhythmic events in patients with reduced left ventricular function. The purpose of the present study was to evaluate the frontal QRS-T angle in the prediction of clinical outcome and death in a large cohort of patients with chronic HF.
Methods
All data associated with members of the largest health maintenance organization, Clalit Health Services in Jerusalem, Israel are digitally recorded in a central computerized database. The database includes demographics, comprehensive clinical data, diagnoses, and the findings from all laboratory tests undertaken at a single centralized laboratory of the health maintenance organization. We electronically identified and retrieved from the computerized database all members with a diagnosis of chronic HF as coded by the database. A total of 6,946 patients had a diagnosis of HF. Validation of the diagnosis of HF was performed on a randomly computer-generated 5% of the patients with a diagnosis of HF (n = 338), as previously reported. In this group, the vast majority fulfilled the European Society of Cardiology criteria for the diagnosis of HF. We retrieved the first ECG performed and available for these patients that was closest to July 2008, when the database was established. Of the 6,946 patients, 5,038 (73%) had an ECG available for analysis. These patients constituted the cohort of the present study. Echocardiographic data pertaining to left ventricular function, categorized into preserved (ejection fraction ≥50%) and reduced (ejection fraction <50%), were available digitally for 33% of the cohort (n = 1,652), and a separate analysis was performed of this population. All hospitalizations in the cardiac and internal medicine departments, including the cardiac and internal intensive care units, were retrieved and analyzed. Data on mortality were retrieved from the National Census Bureau. The institutional committee for human studies of the health maintenance organization, Clalit Health Services, approved the study protocol.
ECGs were acquired using the MAC 5500 ECG Diagnosis System (Marquette Electronics, Milwaukee, Wisconsin), and standard 12-lead ECGs were recorded using standardized procedures. The ECGs were analyzed using the 12-SL ECG Analysis Program (Marquette 12SL ECG Physician Guide, GE Marquette), and all parameters analyzed were derived from this program. The frontal plane QRS-T angle was defined as the absolute value of the difference between the frontal plane QRS axis and T axis and was adjusted to an acute angle by (360° − angle) for an angle >180°.
SPSS, version 17.0, for Windows (SPSS, Chicago, Illinois) was used for all analyses. A comparison of the clinical characteristics was performed using the Student t test for continuous variables and the chi-square test for categorical variables. Log 10 was used for logarithmic transformations. The follow-up period was calculated using a Kaplan-Meier estimate of the potential follow-up. Kaplan-Meier curves, with the log-rank test, were used to compare survival according to the QRS-T angle categories. Multivariate Cox proportional hazards regression analysis was used to evaluate independent variables that determined survival. The parameters included in the multivariate Cox regression analysis incorporated all significant clinical and laboratory parameters on univariate analysis with the addition of drug therapy in separate models. The electrocardiographic parameters were entered into the model if significant on univariate analysis. Proportionality assumptions of the Cox regression models were evaluated using log–log survival curves and with the use of Schoenfeld residuals. An evaluation of the existence of confounding or interactive effects was made between variables and their possible collinearity. A p value of <0.05 was considered statistically significant.
Results
Because no definite cutoff has been determined for the expected QRS-T angles and differences exists in the angle between genders, we calculated the 50th and 75th percentile QRS-T angle in our cohort of patients with HF as a whole and in the women and men separately. The 50th and 75th percentile was 65° and 125° in the whole group, 60° and 120° in women, and 72° and 130° in men, respectively. We evaluated the predictive value of this parameter stratified according to these percentiles in each gender. The demographics and clinical parameters of each gender stratified according to these percentiles for comparison are listed in Table 1 . The electrocardiographic data are listed in Table 2 and the drug therapy data in Table 3 . Patients with a large QRS-T angle were the sicker patients. They had a greater prevalence of diabetes mellitus, ischemic heart disease, atrial fibrillation, and reduced renal function, hemoglobin and sodium. A smaller proportion had preserved left ventricular function. The male patients also had a lower systolic blood pressure. Almost all electrocardiographic parameters were likely to be abnormal with an increasing QRS-T angle. These parameters included the heart rate, and PR and QRS interval and corrected QT (QTc) interval. The QRS axis became more leftward and the T axis more rightward. As expected by definition, patients with an increased QRS-T angle were more likely to be ventricular paced or have a left bundle branch block. The patients were more likely to have ventricular ectopic beats with increasing QRS-T angle. More patients were likely to meet the electrocardiographic criteria for left ventricular hypertrophy. In both genders, patients with a larger QRS-T angle were more likely to be treated with standard HF medications, including angiotensin-converting enzyme inhibitors and angiotensin receptor blockers, β blockers, furosemide, and digoxin. In men, more patients received amiodorone.
| QRS-T angle for Women | <60° (n = 1,224) | 60°–120° (n = 632) | ≥120° (n = 607) | All Women (n = 2,463) | p Value |
|---|---|---|---|---|---|
| Age (yrs) | 77 ± 13 | 77 ± 13 | 78 ± 13 | 77 ± 13 | 0.17 |
| Diabetes mellitus | 540 (44%) | 314 (50%) | 313 (52%) | 1,167 (47%) | 0.004 |
| Hypertension | 980 (80%) | 526 (83%) | 502 (83%) | 2,008 (82%) | 0.17 |
| Hyperlipidemia | 1,045 (85%) | 556 (88%) | 524 (86%) | 2,125 (86%) | 0.30 |
| Ischemic heart disease | 734 (60%) | 446 (71%) | 456 (75%) | 1,636 (66%) | <0.001 |
| Atrial fibrillation | 347 (28%) | 199 (31%) | 208 (34%) | 754 (31%) | 0.03 |
| Body mass index (kg/m 2 ) | 31 ± 8 | 31 ± 10 | 31 ± 10 | 31 ± 9 | 0.55 |
| Systolic blood pressure (mm Hg) | 130 ± 19 | 131 ± 17 | 129 ± 20 | 130 ± 19 | 0.14 |
| Creatinine (mg/dl) | 1.1 ± 1.1 | 1.2 ± 1.0 | 1.3 ± 1.0 | 1.2 ± 1.0 | 0.02 |
| Estimated glomerular filtration rate (ml/min/1.73 m 2 ) | 65 ± 31 | 59 ± 29 | 55 ± 27 | 61 ± 30 | <0.001 |
| Urea (mg/dl) | 56 ± 34 | 61 ± 34 | 65 ± 36 | 59 ± 35 | <0.001 |
| Hemoglobin (g/dl) | 12.2 ± 1.6 | 12.0 ± 1.6 | 12.0 ± 1.5 | 12.1 ± 1.6 | 0.01 |
| Sodium (mEq/L) | 140 ± 3 | 139 ± 3 | 139 ± 4 | 139 ± 3 | 0.08 |
| Left ventricular function (ejection fraction <50%) | 90 (27%) | 84 (37%) | 142 (60%) | 316 (40%) | <0.001 |
| QRS-T Angle for Men | <72° (n = 1,285) | 72°–130° (n = 648) | ≥130° (n = 642) | All Men (n = 2,575) | p Value |
|---|---|---|---|---|---|
| Age (yrs) | 72 ± 14 | 74 ± 13 | 75 ± 12 | 73 ± 14 | <0.001 |
| Diabetes mellitus | 603 (47%) | 367 (57%) | 337 (52%) | 1,307 (51%) | <0.001 |
| Hypertension | 1,096 (85%) | 570 (88%) | 556 (87%) | 2,222 (86%) | 0.26 |
| Hyperlipidemia | 1,010 (79%) | 546 (84%) | 515 (80%) | 2,071 (80%) | 0.01 |
| Ischemic heart disease | 1,043 (81%) | 560 (86%) | 565 (88%) | 2,168 (84%) | <0.001 |
| Atrial fibrillation | 255 (20%) | 142 (22%) | 193 (30%) | 590 (23%) | <0.001 |
| Body mass index (kg/m 2 ) | 29 ± 6 | 29 ± 9 | 28 ± 5 | 29 ± 7 | 0.23 |
| Systolic blood pressure (mm Hg) | 127 ± 19 | 126 ± 19 | 124 ± 19 | 126 ± 19 | 0.01 |
| Creatinine (mg/dl) | 1.5 ± 1.4 | 1.5 ± 1.4 | 1.6 ± 1.3 | 1.5 ± 1.4 | 0.11 |
| Estimated glomerular filtration rate (ml/min/1.73 m 2 ) | 69 ± 46 | 63 ± 28 | 61 ± 45 | 65 ± 42 | <0.001 |
| Urea (mg/dl) | 57 ± 34 | 61 ± 34 | 69 ± 46 | 61 ± 38 | <0.001 |
| Hemoglobin (g/dl) | 13.2 ± 1.8 | 13.1 ± 1.9 | 12.9 ± 1.8 | 13.1 ± 1.8 | 0.002 |
| Sodium (mEq/L) | 140 ± 3 | 140 ± 3 | 139 ± 3 | 140 ± 3 | 0.01 |
| Left ventricular function (ejection fraction <50%) | 216 (61%) | 152 (66%) | 231 (85%) | 599 (70%) | <0.001 |
| QRS-T Angle for Women | <60° (n = 1,224) | 60°–120° (n = 632) | ≥120° (n = 607) | All Women (n = 2,463) | p Value |
|---|---|---|---|---|---|
| Heart rate (beats/min) | 75 ± 16 | 76 ± 16 | 79 ± 20 | 77 ± 17 | <0.001 |
| PR interval (ms) | 163 ± 32 | 166 ± 35 | 172 ± 44 | 165 ± 36 | <0.001 |
| QRS interval (ms) | 88 ± 17 | 94 ± 22 | 115 ± 33 | 96 ± 26 | <0.001 |
| Corrected QT interval (ms) | 443 ± 36 | 452 ± 42 | 465 ± 48 | 451 ± 42 | <0.001 |
| P axis (°) | 51.9 ± 27.0 | 50.0 ± 26.0 | 48.5 ± 31.6 | 50.7 ± 27.8 | 0.08 |
| QRS axis (°) | 20 ± 34 | 6 ± 49 | −3 ± 64 | 10 ± 48 | <0.001 |
| T axis (°) | 32 ± 35 | 67 ± 66 | 124 ± 70 | 64 ± 66 | <0.001 |
| QRS-T angle (°) | 25.8 ± 16.3 | 86.3 ± 17.6 | 152.2 ± 18.0 | 72.5 ± 54.7 | <0.001 |
| Atrial fibrillation | 122 (10%) | 93 (15%) | 130 (21%) | 345 (14%) | <0.001 |
| Pacemaker | 26 (2%) | 23 (4%) | 88 (14%) | 137 (6%) | <0.001 |
| Left bundle branch block | 3 (0%) | 27 (4%) | 113 (19%) | 143 (6%) | <0.001 |
| Left ventricular hypertrophy | 170 (14%) | 131 (21%) | 138 (23%) | 439 (18%) | <0.001 |
| Ventricular ectopic complexes | 89 (7%) | 77 (12%) | 62 (10%) | 228 (9%) | 0.002 |
| QRS-T Angle for Men | <72° (n = 1,285) | 72°–130° (n = 648) | ≥130° (n = 642) | All Men (n = 2,575) | p Value |
|---|---|---|---|---|---|
| Heart rate (beats/min) | 73 ± 16 | 76 ± 17 | 76 ± 19 | 75 ± 17 | <0.001 |
| PR interval (ms) | 163 ± 33 | 172 ± 38 | 173 ± 45 | 168 ± 38 | <0.001 |
| QRS interval (ms) | 96 ± 19 | 105 ± 26 | 130 ± 35 | 107 ± 29 | <0.001 |
| Corrected QT interval (ms) | 438 ± 34 | 450 ± 42 | 472 ± 47 | 449 ± 42 | <0.001 |
| P axis (°) | 47.3 ± 26.1 | 47.0 ± 30.6 | 49.1 ± 33.9 | 47.6 ± 29.1 | 0.44 |
| QRS axis (°) | 19 ± 41 | 4 ± 62 | −10 ± 79 | 8 ± 59 | <0.001 |
| T axis (°) | 36 ± 43 | 79 ± 66 | 118 ± 64 | 67 ± 65 | <0.001 |
| QRS-T angle (°) | 31.4 ± 19.8 | 100.6 ± 17.3 | 157.4 ± 14.6 | 80.2 ± 55.7 | <0.001 |
| Atrial fibrillation | 93 (7%) | 57 (9%) | 76 (12%) | 226 (9%) | <0.001 |
| Pacemaker | 28 (2%) | 26 (4%) | 132 (21%) | 186 (7%) | <0.001 |
| Left bundle branch block | 14 (1%) | 26 (4%) | 107 (17%) | 147 (6%) | <0.001 |
| Left ventricular hypertrophy | 189 (15%) | 106 (16%) | 141 (22%) | 436 (17%) | <0.001 |
| Ventricular ectopic complexes | 134 (10%) | 70 (11%) | 87 (14%) | 291 (11%) | <0.001 |
| QRS-T Angle for Women | <60° (n = 1,224) | 60°–120° (n = 632) | ≥120° (n = 607) | All Women (n = 2,463) | p Value |
|---|---|---|---|---|---|
| Angiotensin-converting enzyme inhibitors/angiotensin receptor blockers | 883 (72%) | 484 (77%) | 462 (76%) | 1,829 (74%) | 0.06 |
| β Blockers | 768 (63%) | 408 (65%) | 433 (71%) | 1,609 (65%) | 0.001 |
| Spironolactone | 314 (26%) | 195 (31%) | 238 (39%) | 747 (30%) | <0.001 |
| Furosemide | 787 (64%) | 454 (72%) | 468 (77%) | 1,709 (69%) | <0.001 |
| Thiazide | 337 (28%) | 162 (26%) | 133 (22%) | 632 (26%) | 0.03 |
| Digoxin | 98 (8%) | 68 (11%) | 112 (18%) | 278 (11%) | <0.001 |
| Amiodorone | 193 (16%) | 100 (16%) | 98 (16%) | 391 (16%) | 0.97 |
| Aspirin | 722 (59%) | 388 (61%) | 381 (63%) | 1,491 (61%) | 0.26 |
| QRS-T Angle for Men | <72° (n = 1,285) | 72°–130° (n = 648) | ≥130° (n = 642) | All Men (n = 2,575) | p Value |
|---|---|---|---|---|---|
| Angiotensin-converting enzyme inhibitor/angiotensin receptor blocker | 948 (74%) | 495 (76%) | 513 (80%) | 1,956 (76%) | 0.01 |
| β Blockers | 872 (68%) | 475 (73%) | 431 (67%) | 1,778 (69%) | 0.02 |
| Spironolactone | 344 (27%) | 214 (33%) | 250 (39%) | 808 (31%) | <0.001 |
| Furosemide | 733 (57%) | 416 (64%) | 459 (71%) | 1,608 (62%) | <0.001 |
| Thiazide | 253 (20%) | 159 (25%) | 129 (20%) | 541 (21%) | 0.04 |
| Digoxin | 91 (7%) | 59 (9%) | 79 (12%) | 229 (9%) | <0.001 |
| Amiodorone | 180 (14%) | 119 (18%) | 142 (22%) | 441 (17%) | <0.001 |
| Aspirin | 905 (70%) | 475 (73%) | 443 (69%) | 1,823 (71%) | 0.22 |
The median follow-up period was 576 days. The overall mortality rate during this period was 17.2% (865 of 5,038). An increasing QRS-T angle category was associated with an increased mortality rate in both genders and in the whole cohort. The estimated cumulative survival rate at the median follow-up point was reduced with an increasing QRS-T angle category, evident in both genders and the whole cohort ( Figure 1 ). An increasing QRS-T angle category was also associated with an increase in cardiac-related hospitalizations in both genders and the whole cohort ( Figure 2 ).
Multivariate Cox regression analysis after adjustment for significant predictors demonstrated that the QRS-T angle was a significant incremental predictor of increased mortality in women and in the whole cohort ( Table 4 ). In men, it was predictive when comparing the lowest and highest QRS-T angle categories. Analysis, after the exclusion of patients with left bundle branch block or a ventricular-paced rhythm (n = 4,543 [90% of the cohort]), demonstrated that the QRS-T angle was a significant incremental predictor of increased mortality ( Table 4 ). Inclusion of HF medications did not significantly change the result in either cohort, and the QRS-T angle remained a significant incremental predictor of increased mortality.
| QRS-T Angle Category | p Value | |||
|---|---|---|---|---|
| Low | Intermediate | High | ||
| Mortality | ||||
| Women (n = 2,463) | 1.0 (Ref) | 1.35 (1.05–1.75); 0.02 | 1.35 (1.04–1.74); 0.02 | 0.03 |
| Men (n = 2,575) | 1.0 (Ref) | 1.01 (0.76–1.34); 0.95 | 1.53 (1.18–1.97); 0.001 | 0.002 |
| All patients (n = 5,038) | 1.0 (Ref) | 1.21 (1.00–1.46); 0.05 | 1.42 (1.18–1.70); <0.001 | 0.001 |
| Cardiac-related hospitalization | ||||
| Women | 1.0 (Ref) | 0.93 (0.81–1.06); 0.3 | 1.16 (1.01–1.33); 0.03 | 0.01 |
| Men | 1.0 (Ref) | 1.04 (0.91–1.19); 0.6 | 1.11 (0.97–1.27); 0.1 | 0.3 |
| All patients | 1.0 (Ref) | 0.95 (0.86–1.05); 0.3 | 1.13 (1.02–1.24); 0.01 | 0.005 |
| Analysis after exclusions (mortality) ∗ | ||||
| Women (n = 2,240) | 1.0 (Ref) | 1.35 (1.04–1.75); 0.02 | 1.45 (1.10–1.92); 0.01 | 0.01 |
| Men (n = 2,303) | 1.0 (Ref) | 1.00 (0.74–1.34); 0.97 | 1.53 (1.14–2.06); 0.01 | 0.01 |
| All patients (n = 4,543) | 1.0 (Ref) | 1.19 (0.98–1.45); 0.08 | 1.47 (1.20–1.80); <0.001 | 0.001 |
| Subanalysis for left ventricular function (mortality) † | ||||
| Women (n = 795) | 1.0 (Ref) | 2.28 (1.44–3.61); <0.001 | 2.14 (1.32–3.46); 0.002 | 0.001 |
| Men (n = 857) | 1.0 (Ref) | 1.32 (0.82–2.12); 0.26 | 2.14 (1.35–3.37); 0.001 | 0.004 |
| All patients (n = 1,652) | 1.0 (Ref) | 1.76 (1.27–2.44); <0.001 | 2.03 (1.47–2.81); <0.001 | <0.001 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
