Abstract
Background: There is paucity of data regarding conduction abnormalities in the Hispanic population with systolic heart failure (HF). We aimed to evaluate the prevalence of electrocardiogram (ECG) abnormalities in a systolic HF population, with attention to the Hispanic population.
Methods: A cross sectional study of 926 patients enrolled in a systolic HF disease management program. ECGS were obtained in patients with an ejection fraction (EF) ≤ 40% by echocardiography at enrollment. Univariate and multivariate analysis adjusted by ethnicities was performed.
Results: White patients exhibited higher prevalence of atrial fibrillation (14.7%) than black patients (8.0%, P = 0.01) whereas Hispanics presented higher prevalence of paced rhythm (14.3% in Hispanics vs. 6.5% in whites and 5.2% in blacks, P<0.01 for both comparisons), higher prevalence of left bundle branch block (LBBB, 14.5% in Hispanics vs. 8.8% in whites and 5.8% in blacks, P = 0.002) and increased frequency of abnormal QT intervals (76.7% in Hispanics) than whites (59.6%) and blacks (69%) patients (P< 0.01 for both comparisons). A QRS interval greater than 120 ms was less prevalent among blacks (15.8% vs. 26.0% in whites and 25.3% in Hispanics, P = 0.01 for both comparisons). Univariate and multivariate analysis disclosed no influence of other characteristics (age, sex, coronary artery disease, hypertension, ejection fraction, medications) in the ECG findings.
Conclusions: Hispanics with Systolic HF presented with increased prevalence of paced rhythm, LBBB, and abnormal QT intervals. Attention should be addressed to these ECG variations to recommend additional guidance for therapeutic interventions and provide important prognostic information.
Keywords: electrocardiogram, abnormalities, systolic heart failure, race and gender
Heart failure is responsible for over 1 million hospitalizations an year, with an annual cost of approximately $35 billion. 1 , 2 Systolic heart failure (HF) may have a number of etiologies, with subsequent deleterious effects to both the myocardium and its conduction pathways. The myocardial conduction system is vulnerable to ischemia, inflammation, fibrosis, and aging, with subsequent altered conduction properties. 3 , 4 The presence of such electrocardiographic conduction abnormalities as complete left bundle branch block (LBBB) and chronic atrial fibrillation (AF) have each been associated with HF, with LBBB seen in about one‐third of cases and chronic AF present in up to 25% of patients. 5 , 6 , 7
Both LBBB and AF have independently been associated with a significant increase in mortality in HF patients 6 , 7 , 8 , 9 , 10 and their appearance together have also been shown to have a cumulative effect on mortality, thus making them important markers in the complete evaluation of a patient with HF. 6
While there have been several studies evaluating the prevalence of these elecrocardiographic (ECG) conduction abnormalities such as LBBB and AF in the general population, there are none currently describing that of the Hispanic population. Given the need for examining race and ethnic differences in ECG abnormalities, the aim of this study is to compare the prevalence of ECG conductions abnormalities in three systolic heart failure populations (i.e., whites, blacks, and Hispanics) from two indigent heart failure clinics (Florida and Louisiana).
METHODS
Study Population
A cross sectional study that included 926 patients was performed in two health care facilities (Florida and Louisiana). We enrolled patients with an ejection fraction ≤ to 40% by echocardiography at both sites. Electrocardiograms were obtained closest to the date of enrollment. At the Louisiana site, recruitment of study participants took place from August 1999 to December 2007 within in the heart failure disease management program (HFDM) at Leonard J. Chabert Medical Center (LJCMC) in Houma, Louisiana (n = 536). All patients were indigent with the following ethnicity composition: white: (n = 319) and black (n = 217). LJCMC is located in rural south Louisiana and as a safety‐net hospital provides care primarily to the uninsured and underinsured patients with a severely depressed socioeconomic (SES) background (more than 55% of patients were below 200% the federal poverty level).
At the Florida facility (n = 390), we included all Hispanic patients enrolled in the Heart Failure Clinic at Jackson Memorial Hospital (JMH). JMH is a large, urban 1600 bed safety net hospital located in Miami, Florida. This hospital serves a population that is largely indigent (50% indigent, 31% had either Medicare or Medicaid, and 19% had private insurance). In addition, the patient population at this facility is largely comprised of immigrants with 98% born outside the United States. Recruitment of study participants took place from September 2007 to January 2009. The Institutional Review Board approval was obtained from the University of Miami Miller School of Medicine and the Louisiana State University Health Sciences Center. Patients signed an informed consent to enroll in an electronic data registry.
Electrocardiogram
A computerized electrocardiogram system (GE Marquette) was used to collect, store, and analyze electrocardiograms. All the ECGs were read by the electrocardiograph's built in algorithm (Marquette 12 SL Library Version 14).
Based on the most current ECG criteria from the ACC guidelines, two independent cardiologists reviewed all tracings specifically focusing on atrial fibrillation, paced rhythm, interventricular conduction delay, left anterior fascicular block, left posterior fascicular block, LBBB, right bundle branch block, PR interval, QT interval, and QRS interval.
Statistical Analyses
SPSS v18.0 (SPSS Inc., Chicago, IL, USA) was used to conduct all analyses. Categorical variables are presented as an absolute number and percentage. Continuous variables are presented as means and standard deviation. Crosstabs were used to analyze ethnic/race and gender differences in prevalence rates. For categorical variables, the chi‐square statistic was used to evaluate group differences. Analysis of variance (ANOVA) was used to evaluate group differences for continuous variables. Univariate analysis between dependent variables (chronic AF, paced rhythm, interventricular conduction delay, left posterior fascicular block, left anterior fascicular block, first degree AV block, LBBB, right bundle branch block) and independent variables (age, sex, body mass index, systolic blood pressure [SBP], diastolic blood pressure [DBP], left ventricular ejection fraction [LVEF], New York Heart Association class [NYHA], history of myocardial infarction [MI], medications) adjusted by ethnicity group was performed to assess any influence in the ECG findings. We next evaluated multiple factor interaction in the ECG's findings by a multivariate analysis. Assuming the same variable classification, we analyzed each dependent variable in combination with the overall independent variables. For all analyses a P < 0.05 was considered statistically significant.
RESULTS
Demographic characteristics of the three race/ethnic groups, whites, blacks, and Hispanics by sex are described in Table 1 and 2 respectively. Multiple statistically significant race/ethnic group differences were found for age, education level, SBP, DBP, BMI, prevalence of ischemic cardiomyopathy, EF, and prescription of medications (i.e., aspirin, ACE inhibitors/ angiotensin receptor blockers [ACE/ARB], spironolactone, and digoxin, Table 1), but only differences in body mass index (BMI), DBP, EF, and prevalence of ischemic cardiomyopathy were found between males and females (Table 2). Overall, black patients were younger, had more education years, presented with higher blood pressure, lower EF, were more likely to be using beta blockers, digoxin, and spironolactone, had less prevalence of using ACE/ARB than white patients and less prevalence of ischemic cardiomyopathy (P < 0.05 for all comparisons by chi‐square test, except for EF and blood pressure comparison performed by ANOVA). Conversely, Hispanic population was older had more years of formal education, lower BMI and EF, higher levels of blood pressure, and were less likely to be using spironolactone than white patients (P < 0.05 for all comparisons by chi‐square test). Hispanics were also more likely to have a previous myocardial infarction, more likely to be prescribed aspirin, beta blockers, ACE/ARB, and digoxin than white and black population (P < 0.001 for all comparisons by chi‐square test).
Table 1.
Demographic Characteristics by Race and Ethnicity
| Variable | White (n = 339) | Black (n = 364) | Hispanic (n = 223) | Total (n = 926) | P‐value |
|---|---|---|---|---|---|
| Age, years, M (SD) | 57.65 (11.31)a | 54.44 (10.76)ab | 57.19 (11.37)b | 56.28 (11.21) | 0.000 |
| Male, n (%) | 220 (64.9) | 236 (64.8) | 152 (68.2) | 608 (65.7) | 0.665 |
| Education level, years, M (SD) | 9.41 (3.30)ab | 10.53 (3.17)a | 10.53 (4.23)b | 10.15 (3.57) | 0.000 |
| BMI, kg/m2 M (SD) | 32.58 (8.85)a | 31.58 (9.21) | 30.40 (8.13)a | 31.58 (8.82) | 0.043 |
| Systolic blood pressure, mmHg, M (SD) | 130.13 (23.75)ab | 135.30 (27.09)a | 135.13 (26.72)b | 133.49 (25.98) | 0.027 |
| Diastolic blood pressure, mmHg, M (SD) | 73.01 (14.22)ac | 81.06 (16.69)ab | 84.24 (15.85)bc | 79.19 (16.30) | 0.000 |
| Ischemic cardiomyopathy, n (%) | 116 (47.2)ac | 77 (26.6)ab | 93 (44.7)bc | 286 (38.5) | 0.000 |
| EF, %, M (SD) | 31.61 (10.80)ac | 28.17 (10.91)ab | 25.11 (10.09)bc | 28.46 (10.94) | 0.000 |
| NYHA, n (%) | 0.030 | ||||
| I | 89 (32.8)abc | 70 (22.3)ab | 50 (24.8)c | 209 (26.6) | |
| II | 82 (30.3)a | 101 (32.2)a | 65 (32.2) | 248 (31.5) | |
| III | 78 (28.8)b | 100 (31.8)b | 55 (27.2) | 233 (29.6) | |
| IV | 22 (8.1)c | 43 (13.7) | 32 (15.8)c | 97 (12.3) | |
| MI, n (%) | 38 (11.2)a | 54 (14.9)b | 73 (33.3)ab | 165 (17.9) | 0.000 |
| Aspirin, n (%) | 143 (42.3)a | 142 (39.0)b | 125 (58.1)ab | 410 (44.7) | 0.000 |
| Digoxin, n (%) | 5 (3.7)ac | 31 (13.3)ab | 71 (32.6)bc | 107 (18.2) | 0.000 |
| Spironolactone, n (%) | 3 (1.8)ab | 38 (15.9)a | 48 (22.0)b | 89 (14.3) | 0.000 |
| Beta‐blocker, n (%) | 334 (98.5)ac | 344 (94.5)ab | 197 (88.3)bc | 875 (94.5) | 0.000 |
| ACE/ARB, n (%) | 336 (99.1)ac | 323 (88.7)ab | 173 (77.6)bc | 832 (89.8) | 0.000 |
M = mean; SD = standard deviation; n = number of individuals; EF = ejection fraction; NYHA = New York Heart Association functional class; ACE = angiotensin converting‐enzyme inhibitors; ARB = angiotensin receptor blockers.
(a)White versus blacks
(b)Hispanics versus blacks
(c)Hispanics versus whites
Table 2.
Demographic Characteristics by Gender
| Variable | Male | Female | Total | P‐value |
|---|---|---|---|---|
| Age, years, M (SD) | 55.99 (11.39) | 56.84 (10.83) | 56.28 (11.21) | 0.278 |
| Race/Ethnicity, N (%) | 0.665 | |||
| White | 220 (36.2) | 119 (37.4) | 339 (36.6) | |
| Black | 236 (38.8) | 128 (40.3) | 364 (39.3) | |
| Hispanic | 152 (25.0) | 71 (22.3) | 223 (24.1) | |
| Education level, years, M (SD) | 10.21 (3.65) | 10.04 (3.39) | 10.15 (3.57) | 0.549 |
| BMI, kg/m2 M (SD) | 31.12 (8.97) | 32.63 (8.39) | 31.58 (8.82) | 0.043 |
| Systolic blood pressure, mmHg, M (SD) | 133.19 (26.13) | 134.12 (25.70) | 133.49 (25.98) | 0.628 |
| Diastolic blood pressure, mmHg, M (SD) | 80.69 (16.86) | 75.99 (14.56) | 79.19 (16.30) | 0.000 |
| Ischemic cardiomyopathy, n (%) | 217 (43.8) | 69 (27.8) | 286 (38.5) | 0.000 |
| EF, %, M (SD) | 27.61 (10.74) | 30.17 (11.14) | 28.46 (10.94) | 0.002 |
| NYHA, N (%) | 0.245 | |||
| I | 134 (25.6) | 75 (28.5) | 209 (26.6) | |
| II | 175 (33.4) | 73 (27.8) | 248 (31.5) | |
| III | 147 (28.1) | 86 (32.7) | 233 (29.6) | |
| IV | 68 (13.0) | 29 (11.0) | 97 (12.3) | |
| MI, N (%) | 122 (20.2) | 43 (13.6) | 165 (17.9) | 0.013 |
| Aspirin, N (%) | 291 (48.3) | 119 (37.8) | 410 (44.7) | 0.002 |
| Digoxin, N (%) | 69 (17.3) | 38 (20.1) | 107 (18.2) | 0.417 |
| Spironolactone, N (%) | 65 (15.3) | 24 (12.3) | 89 (14.3) | 0.330 |
| Beta‐blocker, N (%) | 572 (94.1) | 303 (95.3) | 875 (94.5) | 0.446 |
| ACE/ARB, N (%) | 550 (90.5) | 282 (88.7) | 832 (89.8) | 0.394 |
M = mean; SD = standard deviation; n = number of individuals; EF = ejection fraction; NYHA = New York Heart Association functional class; ACE = angiotensin converting‐enzyme inhibitors; ARB = angiotensin receptor blockers.
Race and Ethnic Differences in ECG Abnormalities
In Table 3 we present race and ethnic differences in ECG abnormalities. White patients presented higher prevalence of atrial fibrillation than black patients (14.7% vs. 8.0% respectively, P = 0.01 by chi‐square test), whereas Hispanics presented with higher prevalence of paced rhythm (14.3% in Hispanics vs. 6.5% in whites and 5.2% in blacks, P < 0.01 for both comparisons), higher prevalence of left bundle branch block (14.5% in Hispanics vs. 8.8% in whites and 5.8% in blacks, P = 0.002) and increased frequency of the abnormal QT intervals (76.7%) than whites (59.6%) and blacks (69%) patients (P < 0.01 for all comparisons by chi‐square test). A QRS interval greater than 120 ms was less prevalent among blacks (15.8% vs. 26.0 in whites and 25.3% in Hispanics, P = 0.01).
Table 3.
ECG Abnormalities by Race and Ethnicity
| Variable | White (n = 339) n (%) | Black (n = 364) n (%) | Hispanic (n = 223) n (%) | Total (n = 926) n (%) | P‐value |
|---|---|---|---|---|---|
| Atrial fibrillation, n (%) | 50 (14.7)a | 29 (8.0)a | 24 (10.8) | 103 (11.1) | 0.017 |
| Paced rhythm, n (%) | 22 (6.5) a | 19 (5.2)b | 32 (14.3)ab | 73 (7.9) | 0.000 |
| Interventricular conduction delay, n (%) | 26 (7.7) | 25 (6.9) | 11 (4.9) | 62 (6.7) | 0.440 |
| Left posterior fascicular block, n (%) | 5 (1.5) | 4 (1.1) | 1 (0.4) | 10 (1.1) | 0.515 |
| Left anterior fascicular block, n (%) | 46 (13.6)a | 34 (9.3) | 16 (7.2)c | 96 (10.4) | 0.037 |
| First degree AV block, n (%) | 36 (10.7) | 45 (12.4) | 23 (10.3) | 104 (11.2) | 0.681 |
| LBBB, n (%) | 30 (8.8)a | 21 (5.8) b | 32 (14.3) bc | 83 (9.0) | 0.002 |
| RBBB, n (%) | 25 (7.4) | 21 (5.8) | 10 (4.5) | 56 (6.0) | 0.357 |
| Heart rate #bpm | 0.457 | ||||
| ≤59 | 50 (19.1) | 48 (15.1) | 24 (14.1) | 122 (16.3) | |
| 60 to 100 | 192 (73.3) | 237 (74.8) | 127 (74.7) | 556 (74.2) | |
| ≥101 | 20 (7.6) | 32 (10.1) | 19 (11.2) | 71 (9.5) | |
| PR interval #msec | 0.240 | ||||
| ≤119 | 4 (1.5) | 2 (0.6) | 2 (1.2) | 8 (1.1) | |
| 121 to 200 | 215 (82.1) | 248 (78.5) | 145 (85.3) | 608 (81.3) | |
| ≥201 | 43 (16.4) | 66 (20.9) | 23 (13.5) | 132 (17.6) | |
| QT interval* | 0.000 | ||||
| Normal | 137 (40.4) ac | 113 (31.0)ab | 52 (23.3) bc | 302 (32.6) | |
| Abnormal | 202 (59.6) ac | 251 (69.0)ab | 171 (76.7)bc | 624 (67.4) | |
| QRS interval #msec | 0.013 | ||||
| ≤110 | 173 (66.0) a | 236 (74.4) ac | 118 (69.4)c | 527 (70.4) | |
| 111 to 119 | 21 (8.0) a | 31 (9.8) bd | 9 (5.3) | 61 (8.1) | |
| ≥120 | 68 (26.0) b | 50 (15.8) abc | 43 (25.3) c | 161 (21.5) |
LBBB = left bundle branch block; RBBB = right bundle branch block; bpm = beats per minute; msec = milliseconds.
*Corrected QT interval > 460 miliseconds was considered abnormal.
(a)White versus blacks.
(b)Hispanics versus blacks.
(c)Hispanics versus whites.
Gender Differences in ECG Abnormalities
Right bundle branch block that was much more prevalent in men (7.4% in men vs. 3.5% in women, P = 0.01 by chi‐square test, Table 4). No significant gender differences within race and ethnic groups were found in the total sample (Table 5).
Table 4.
ECG Abnormalities by Gender
| Variable | Male (N = 608) | Female (N = 318) | P‐value |
|---|---|---|---|
| Atrial fibrillation | 72 (11.8) | 31 (9.7) | 0.336 |
| Paced rhythm | 55 (9.0) | 18 (5.7) | 0.069 |
| Interventricular conduction delay | 134 (22.0) | 54 (17.0) | 0.069 |
| Left posterior fascicular block | 8 (1.3) | 2 (0.6) | 0.336 |
| Left anterior fascicular block | 66 (10.9) | 30 (9.4) | 0.500 |
| First degree AV block | 71 (11.7) | 33 (10.4) | 0.562 |
| LBBB | 47 (7.7) | 36 (11.3) | 0.069 |
| RBBB | 45 (7.4) | 11 (3.5) | 0.017 |
| Heart rate# | 0.199 | ||
| ≤59 | 77 (16.0) | 45 (16.8) | |
| 60 to 100 | 365 (75.9) | 191 (71.3) | |
| ≥101 | 39 (8.1) | 32 (11.9) | |
| PR interval# | 0.194 | ||
| ≤120 | 6 (1.3) | 2 (0.7) | |
| 121 to 200 | 381 (79.4) | 227 (84.7) | |
| ≥201 | 93 (19.4) | 39 (14.6) | |
| QT interval # | 0.282 | ||
| Normal | 191 (31.4) | 111 (34.9) | |
| Abnormal | 417 (68.6) | 207 (65.1) | |
| QRS interval# | 0.210 | ||
| ≤110 | 328 (68.2) | 199 (74.3) | |
| 111 to 119 | 43 (8.9) | 18 (6.7) | |
| ≥120 | 110 (22.9) | 51 (19.0) |
LBBB = left bundle branch block; RBBB = right bundle branch block.
Table 5.
ECG Abnormalities by Gender within Race/Ethnic Groups
| Variable | White | P‐value | Black | P‐value | Hispanic | P‐value | |||
|---|---|---|---|---|---|---|---|---|---|
| Male (n = 220) n (%) | Female (n = 119) n (%) | Male (n = 236) n (%) | Female (n = 128) n (%) | Male (n = 152) n (%) | Female (n = 71) n (%) | ||||
| Atrial fibrillation | 36 (16.4) | 14 (11.8) | 0.254 | 23 (9.7) | 6 (4.7) | 0.089 | 13 (8.6) | 11 (15.5) | 0.119 |
| Paced rhythm | 15 (6.8) | 7 (5.9) | 0.738 | 14 (5.9) | 5 (3.9) | 0.407 | 26 (17.1) | 6 (8.5) | 0.086 |
| Interventricular conduction delay | 19 (8.6) | 7 (5.9) | 0.363 | 17 (7.2) | 8 (6.3) | 0.731 | 10 (6.6) | 1 (1.4) | 0.097 |
| Left posterior fascicular block | 4 (1.8) | 1 (0.8) | 0.476 | 3 (1.3) | 1 (0.8) | 0.666 | 1 (0.7) | 0 (0.0) | 0.493 |
| Left anterior fascicular block | 33 (15.0) | 13 (10.9) | 0.296 | 24 (10.2) | 10 (7.8) | 0.461 | 9 (5.9) | 7 (9.9) | 0.288 |
| First degree AV block | 23 (10.5) | 13 (11.0) | 0.873 | 32 (13.6) | 13 (10.2) | 0.346 | 16 (10.5) | 7 (9.9) | 0.879 |
| LBBB | 15 (6.8) | 15 (12.6) | 0.073 | 13 (5.5) | 8 (6.3) | 0.772 | 19 (12.5) | 13 (18.3) | 0.249 |
| RBBB | 19 (8.6) | 6 (5.0) | 0.227 | 17 (7.2) | 4 (3.1) | 0.111 | 9 (5.9) | 1 (1.4) | 0.129 |
| Heart rate# | 0.206 | 0.359 | 0.991 | ||||||
| ≤59 | 32 (19.3) | 18 (18.8) | 29 (14.4) | 19 (16.4) | 16 (14.0) | 8 (14.3) | |||
| 60 to 100 | 125 (75.3) | 67 (69.8) | 155 (77.1) | 82 (70.7) | 85 (74.6) | 42 (75.0) | |||
| ≥101 | 9 (5.4) | 11 (11.5) | 17 (8.5) | 15 (12.9) | 13 (11.4) | 6 (10.7) | |||
| PR interval# | 0.381 | 0.252 | 0.578 | ||||||
| ≤120 | 2 (1.2) | 2 (2.1) | 2 (1.0) | 0 (0) | 2 (1.8) | 0 (0) | |||
| 120 to 200 | 133 (80.1) | 82 (85.4) | 152 (76.0) | 96 (82.8) | 96 (84.2) | 49 (87.5) | |||
| ≥201 | 31 (18.7) | 12 (12.5) | 46 (23.0) | 20 (17.2) | 16 (14.0) | 7 (12.5) | |||
| QT interval# | 0.255 | 0.861 | 0.406 | ||||||
| Normal | 84 (38.2) | 53 (44.5) | 74 (31.4) | 39 (30.5) | 33 (21.7) | 19 (26.8) | |||
| Abnormal | 136 (61.8) | 66 (55.5) | 162 (68.6) | 89 (69.5) | 119 (78.3) | 52 (73.2) | |||
| QRS interval# | 0.085 | 0.754 | 0.148 | ||||||
| ≤110 | 106 (63.9) | 67 (69.8) | 148 (73.6) | 88 (75.9) | 74 (64.9) | 44 (78.6) | |||
| 111 to 119 | 18 (10.8) | 3 (3.1) | 19 (9.5) | 12 (10.3) | 6 (5.3) | 3 (5.4) | |||
| ≥120 | 42 (25.3) | 26 (27.1) | 34 (16.9) | 16 (13.8) | 34 (29.8) | 9 (16.1) | |||
We next performed independent univariate analysis of the ECG's abnormalities to assess whether the demographic characteristics (independent variables) can influence the ECG findings. We found in the univariate analysis that paced rhythm was influenced by SBP (R square = 0.025, Adjusted R square = 0.021, P < 0.01) and EF (R square = 0.021, Adjusted R square = 0.017, P < 0.05), and LBBB was only influenced by initial NYHA class (R square = 0.06, Adjusted R square = 0.05, P = 0.03, Table 6). Furthermore, in order to evaluate multiple interactions in the ECG's findings we also performed a multivariate analysis taking similar assumption of the variables. Multivariate analysis documented that intraventricular conduction delay was influenced by SBP (P = 0.04) and left anterior fascicular block was influenced by DBP (P < 0.01) but neither of the other factors disclosed an impact in the prevalence of ethnicity variances (Table 7).
Table 6.
Univariate Analysis
| Dependent Variable | Parameter | B | Standard Error | t | R‐square | Adjusted R‐square | P‐value |
|---|---|---|---|---|---|---|---|
| Atrial fibrillation | HTN | 0.01 | 0.008 | 1.159 | 0.002 | −0.002 | 0.247 |
| MI | 0.003 | 0.011 | 0.322 | 0.001 | −0.003 | 0.748 | |
| Smoker | −0.011 | 0.01 | −1.114 | 0.002 | −0.002 | 0.266 | |
| SBP | 3.93E‐05 | 0 | 0.24 | 0.001 | −0.004 | 0.81 | |
| DBP | 8.71E‐06 | 0 | 0.034 | 0 | −0.004 | 0.973 | |
| Initial NYHA | 0.002 | 0.004 | 0.564 | 0.001 | −0.003 | 0.573 | |
| EF | − 5.75E‐05 | 0 | −0.124 | 0.001 | −0.004 | 0.902 | |
| Left anterior | HTN | 0.013 | 0.021 | 0.612 | 0.004 | 0 | 0.541 |
| fascicular block | MI | −0.024 | 0.027 | −0.905 | 0.005 | 0.001 | 0.366 |
| Smoker | 0.031 | 0.024 | 1.288 | 0.006 | 0.002 | 0.198 | |
| SBP | 0 | 0 | 0.28 | 0.005 | 0.001 | 0.78 | |
| DBP | 0 | 0.001 | 0.322 | 0.005 | 0.001 | 0.747 | |
| Initial NYHA | 0.011 | 0.011 | 1.03 | 0.008 | 0.003 | 0.303 | |
| EF | −0.003 | 0.001 | −2.457 | 0.12 | 0.07 | 0.014 | |
| Left posterior | HTN | −0.005 | 0.007 | −0.688 | 0.005 | 0.001 | 0.492 |
| fascicular block | MI | −0.007 | 0.009 | −0.782 | 0.006 | 0.002 | 0.434 |
| Smoker | −0.012 | 0.008 | −1.529 | 0.007 | 0.003 | 0.126 | |
| SBP | 0 | 0 | 2.115 | 0.011 | 0.006 | 0.035 | |
| DBP | 0.001 | 0 | 3.041 | 0.016 | 0.012 | 0.002 | |
| Initial NYHA | −0.005 | 0.003 | −1.41 | 0.008 | 0.004 | 0.159 | |
| EF | 0 | 0 | 1.162 | 0.008 | 0.003 | 0.245 | |
| Intraventricular | HTN | 0.031 | 0.026 | 1.202 | 0.03 | 0.026 | 0.23 |
| conduction delay | MI | −0.065 | 0.034 | −1.894 | 0.03 | 0.026 | 0.059 |
| Smoker | −0.006 | 0.031 | −0.208 | 0.025 | 0.021 | 0.836 | |
| SBP | 0 | 0 | −0.686 | 0.03 | 0.026 | 0.493 | |
| DBP | 0 | 0.001 | −0.552 | 0.03 | 0.025 | 0.581 | |
| Initial NYHA | 0.007 | 0.014 | 0.528 | 0.029 | 0.025 | 0.597 | |
| EF | −0.001 | 0.001 | −0.854 | 0.028 | 0.023 | 0.393 | |
| LBBB | HTN | 0.01 | 0.007 | 1.402 | 0.052 | 0.048 | 0.161 |
| MI | 0.006 | 0.009 | 0.648 | 0.05 | 0.046 | 0.517 | |
| Smoker | 0.002 | 0.008 | 0.254 | 0.054 | 0.05 | 0.799 | |
| SBP | 0 | 0 | −1.007 | 0.063 | 0.058 | 0.314 | |
| DBP | 0 | 0 | −1.244 | 0.061 | 0.057 | 0.214 | |
| Initial NYHA | 0 | 0.004 | −0.034 | 0.06 | 0.055 | 0.973 | |
| EF | 0 | 0 | 0.527 | 0.058 | 0.053 | 0.598 | |
| Paced rhythm | HTN | 0.011 | 0.007 | 1.402 | 0.042 | 0.038 | 0.361 |
| MI | 0.007 | 0.009 | 0.648 | 0.029 | 0.024 | 0.417 | |
| Smoker | 0.024 | 0.008 | 0.254 | 0.022 | 0.019 | 0.899 | |
| SBP | 0.001 | 0 | 2.611 | 0.025 | 0.021 | 0.009 | |
| DBP | 0 | 0 | −1.244 | 0.062 | 0.052 | 0.314 | |
| Initial NYHA | 0 | 0.01 | 0.029 | 0.017 | 0.013 | 0.977 | |
| EF | 0.003 | 0.001 | 2.298 | 0.021 | 0.017 | 0.022 |
HTN = hypertension; MI = myocardial infarction; SBP = systolic blood pressure; DBP = diastolic blood pressure; NYHA = New York Heart Association functional class; EF = ejection fraction; LBBB = left bundle branch block.
Table 7.
Multivariate Analysis
| Dependent Variable | Parameter | B | Standard Error | t | R‐square | Adjusted R‐square | P‐value |
|---|---|---|---|---|---|---|---|
| Atrial fibrillation | Intercept | 0.07 | 0.092 | 0.762 | 0.018 | 0.008 | 0.447 |
| NYHA | −0.001 | 0.012 | −0.116 | 0.908 | |||
| EF | 2.04E−05 | 0.001 | 0.014 | 0.989 | |||
| DBP | 0.001 | 0.001 | 0.61 | 0.542 | |||
| SBP | 0 | 0.001 | −0.243 | 0.808 | |||
| MI | 0.014 | 0.029 | 0.461 | 0.645 | |||
| Paced rhythm | Intercept | 0.077 | 0.087 | 0.88 | 0.049 | 0.039 | 0.379 |
| NYHA | 0.008 | 0.012 | 0.66 | 0.509 | |||
| EF | 0.003 | 0.001 | 2.248 | 0.025 | |||
| DBP | −0.004 | 0.001 | −4.029 | 0 | |||
| SBP | 0.002 | 0.001 | 3.708 | 0 | |||
| MI | −0.014 | 0.028 | −0.517 | 0.605 | |||
| Interventricular conduction | Intercept | 0.017 | 0.075 | 0.226 | 0.012 | 0.001 | 0.821 |
| delay | NYHA | −0.002 | 0.01 | −0.182 | 0.856 | ||
| EF | 0.001 | 0.001 | 0.819 | 0.413 | |||
| DBP | −0.002 | 0.001 | −1.93 | 0.054 | |||
| SBP | 0.001 | 0.001 | 2.024 | 0.043 | |||
| MI | 0.012 | 0.024 | 0.509 | 0.611 | |||
| Left posterior fascicular | Intercept | −0.01 | 0.028 | −0.342 | 0.005 | −0.005 | 0.732 |
| block | NYHA | −0.001 | 0.004 | −0.331 | 0.741 | ||
| EF | 0 | 0 | 0.597 | 0.551 | |||
| DBP | 0 | 0 | 0.641 | 0.522 | |||
| SBP | −1.61E‐05 | 0 | −0.085 | 0.933 | |||
| MI | −0.01 | 0.009 | −1.129 | 0.259 | |||
| Left anterior fascicular block | Intercept | 0.218 | 0.091 | 2.402 | 0.032 | 0.022 | 0.017 |
| NYHA | 0.007 | 0.012 | 0.561 | 0.575 | |||
| EF | 0 | 0.001 | 0.268 | 0.789 | |||
| DBP | −0.003 | 0.001 | −3.247 | 0.001 | |||
| SBP | 0.001 | 0.001 | 1.256 | 0.209 | |||
| MI | −0.014 | 0.029 | −0.473 | 0.637 | |||
| First degree AV block | Intercept | 0.156 | 0.092 | 1.697 | 0.008 | −0.003 | 0.09 |
| NYHA | 0.005 | 0.012 | 0.391 | 0.696 | |||
| EF | 0 | 0.001 | −0.316 | 0.752 | |||
| DBP | 0 | 0.001 | −0.321 | 0.749 | |||
| SBP | 0 | 0.001 | −0.232 | 0.817 | |||
| MI | −0.024 | 0.029 | −0.816 | 0.415 | |||
| LBBB | Intercept | 0.309 | 0.09 | 3.429 | 0.022 | 0.012 | 0.001 |
| NYHA | −0.006 | 0.012 | −0.535 | 0.593 | |||
| EF | −0.001 | 0.001 | −0.711 | 0.477 | |||
| DBP | 0 | 0.001 | 0.169 | 0.865 | |||
| SBP | −0.001 | 0.001 | −1.567 | 0.117 | |||
| MI | −0.024 | 0.029 | −0.845 | 0.398 | |||
| Right bundle branch block | Intercept | 0.093 | 0.072 | 1.297 | 0.007 | −0.003 | 0.195 |
| NYHA | −0.011 | 0.01 | −1.178 | 0.239 | |||
| EF | −0.001 | 0.001 | −0.599 | 0.549 | |||
| DBP | 0 | 0.001 | −0.364 | 0.716 | |||
| SBP | 6.01E‐05 | 0 | 0.124 | 0.901 | |||
| MI | 0.031 | 0.023 | 1.334 | 0.183 |
NYHA = New York Heart Association functional class; EF = ejection fraction; SBP = systolic blood pressure; DBP = diastolic blood pressure; MI = myocardial infarction; LBBB = left bundle branch block.
DISCUSSION
This study describes the prevalence of ECG abnormalities in three systolic heart failure populations: whites, blacks, and Hispanics. Significant race and ethnic differences were found for atrial fibrillation, paced rhythm, LBBB, QRS interval greater than 120 ms, and QT interval that were not statistically significant related to other demographic characteristics or comorbidities of the population. Consistent with results from past studies, we found that white patients had significantly more atrial fibrillation than black patients. 11 , 12 , 13 , 14 , 15 Ruo and colleagues also reported a higher prevalence of atrial fibrillation among whites when compared to blacks. 16 This study also found that blacks were more likely to have some risk factors for atrial fibrillation including hypertension and prior diagnosed HF. In addition, the study found that blacks were less likely to be diagnosed with coronary disease, revascularization, hypothyroidism, or valve replacement than whites.
The increased prevalence of atrial fibrillation among white patients with systolic HF presents a concern for this population for several reasons. First, AF is characterized by irregular and often rapid ventricular rate, loss of atrial contraction, loss of atrioventricular synchrony, elevated filling pressures causing atrial dilatation and reduction in stroke volumes, 17 and is the most common cardiac arrhythmia, responsible for approximately one‐third of all hospital admissions with a cardiac rhythm disturbance. Second, AF is associated with an increased risk of stroke and similarly is systolic HF which is responsible for an increased number of emergency department visits and hospitalizations. 18 Finally, HF and AF frequently coexist carrying severe hemodynamic consequences for the patient. 19 , 20
There are several well‐known risk factors for AF such as hypertension, coronary disease, prior diagnosed of HF, mitral stenosis, valvular repair, chronic lung disease, hyperthyroidism, left ventricular systolic function status, and medication. Differences in these risk factors and in baseline characteristics, when comparing white patients with black patients, may account for the differences found in AF prevalence, but was not found in our study. A lower prevalence of AF in blacks compared to whites with HF could also be explained by intrinsic racial differences in atrial membrane stability and conduction pathways, or even by genetic polymorphisms leading to different susceptibility to the development of AF. 16
We found that Hispanic patients had more paced rhythm and evidence of LBBB than whites and blacks. By comparison, Perez et al. 21 recently reported a lower prevalence of LBBB in black and Hispanic patients compared to white patients. The mean age of the Hispanic cohort in Perez study is 54 years, slightly lower than in our study, considering that age is a risk factor for the development of LBBB. 22 Another explanation for this difference could be the low‐SES background. Our group of patients with low SES may not receive consistent reliable medical therapy or medical assistance and therefore have more advanced disease. Our population, in both Florida and Louisiana, is predominately indigent. The patients enrolled in Perez study received medical assistance at Palo Alto Veterans Affairs Health Care System that offers “equal access to healthcare regardless of race or socioeconomic status and provide a distinct advantage for studying race‐based differences.” No specific information about the SES of the patients was provided.
Likewise, we found that abnormal QT interval was more prevalent in the Hispanic group compared to blacks and whites. Indeed, Perez et al. reported a higher prevalence of prolonged QTc among Hispanics comparing to non‐Hispanics. The QT interval measured in the ECG is a representation of global repolarization duration in the ventricular myocardium. Abnormal QTc prolongation on the electrocardiogram should be viewed as an independent risk factor for sudden cardiac death (SCD) and seems to be associated with fivefold increased odds of SCD. 23
Intraventricular conduction disturbances are commonly found in HF patients and seem to be associated with increased mortality. 6 Patients associated with intraventricular conduction disturbances are at an increased risk and could potentially benefit from electrophysiologic therapies such as atrial synchronized biventricular pacing or cardiac resynchronization therapy (CRT). 24 , 25 CRT has been found to be beneficial by improving the left ventricular filling, decreasing mitral regurgitation, and reducing septal dyskinesis. 24 Indeed, we found 122 patients in this study that met the criteria for CRT.
A QRS interval greater than 120 ms was less prevalent among blacks compared to whites and Hispanics. QRS prolongation is an independent predictor of both increased total mortality and SCD. 26 A duration greater than 120 ms has been shown to have 99% specificity for left ventricular dysfunction and may be a potent marker for adverse outcome. 26 Early identification of patients at risk can allow precocious therapeutic interventions such as CRT thereby improving morbidity and mortality rates.
Limitations
The composition of our population is variable and therefore we could not compare the same ethnicity group between the two HFDM program sites. We do not have full access to the past medical history of these patients nor all medications they were prescribed or taking which clearly could influence the results. It would be important to have a better knowledge of the patient's medical and surgical history, selected medications which affect conduction and an accurate physical examination report. A more advanced cardiovascular disease could be the explanation for the difference in the prevalence and not necessarily the race or ethnicity. ECGs were obtained at enrollment of the HFMD programs but we could not gather data of a second ECG after any particular event or hospital re‐admission, which could have shown ECG variations and progression of the disease. We also do not have a full disclosure of the specific causes of mortality or hospital readmission which could be useful for establishing risk‐prediction models and hazard ratios.
Clinical Implications
Currently there are available several pharmacological and nonpharmacological therapeutic options for patients with atrial fibrillation. Several recently published studies have had promising results regarding catheter ablation as an option for a broad spectrum of patients with atrial fibrillation. Keeping in mind that these treatment options are expensive and these patients are from low SES, many without insurance, early detection to identify patients at risk could prove to result in cost savings by diminishing further development of the pathology and its consequences.
CONCLUSION
Hispanics with systolic heart failure showed more prevalence of baseline paced rhythm, LBBB, and abnormal QT intervals that were not statistically related to other demographic factors. Adequate medical attention should be addressed to these variations since relevant ECG findings can provide additional guidance for therapeutic interventions, provide important prognostic information, and potentially modify morbidity and mortality rate.
REFERENCES
- 1. Rosamond W, Flegal K, Furie K, et al Heart disease and stroke statistics–2008 update: A report from the American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Circulation 2008;117(4):e25–146. [DOI] [PubMed] [Google Scholar]
- 2. Effect of enalapril on survival in patients with reduced left ventricular ejection fractions and congestive heart failure. The SOLVD Investigators. N Engl J Med 1991;325(5):293–302. [DOI] [PubMed] [Google Scholar]
- 3. Jessup M, Brozena S. Heart failure. N Engl J Med 2003;348(20):2007–2018. [DOI] [PubMed] [Google Scholar]
- 4. Farmakis D, Filippatos G, Lainscak M, et al Anticoagulants, antiplatelets, and statins in heart failure. Cardiol Clin 2008;26(1):49–58, vi. [DOI] [PubMed] [Google Scholar]
- 5. Shenkman HJ, Pampati V, Khandelwal AK, et al Congestive heart failure and QRS duration: Establishing prognosis study. Chest 2002;122(2):528–534. [DOI] [PubMed] [Google Scholar]
- 6. Baldasseroni S, Opasich C, Gorini M, et al Left bundle‐branch block is associated with increased 1‐year sudden and total mortality rate in 5517 outpatients with congestive heart failure: A report from the Italian network on congestive heart failure. Am Heart J 2002;143(3):398–405. [DOI] [PubMed] [Google Scholar]
- 7. Clark AL, Goode K, Cleland JG. The prevalence and incidence of left bundle branch block in ambulant patients with chronic heart failure. Eur J Heart Fail 2008;10(7):696–702. [DOI] [PubMed] [Google Scholar]
- 8. Hawkins NM, Wang D, McMurray JJ, et al Prevalence and prognostic impact of bundle branch block in patients with heart failure: Evidence from the CHARM programme. Eur J Heart Fail 2007;9(5):510–517. [DOI] [PubMed] [Google Scholar]
- 9. Mehra MR, Uber PA, Park MH, et al Obesity and suppressed B‐type natriuretic peptide levels in heart failure. J Am Coll Cardiol 2004;43(9):1590–1595. [DOI] [PubMed] [Google Scholar]
- 10. Davis RC, Hobbs FD, Kenkre JE, et al Prevalence of left ventricular systolic dysfunction and heart failure in high risk patients: Community based epidemiological study. BMJ 2002;325(7373):1156. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Sacco RL, Boden‐Albala B, Abel G, et al Race‐ethnic disparities in the impact of stroke risk factors: The northern Manhattan stroke study. Stroke 2001;32(8):1725–1731. [DOI] [PubMed] [Google Scholar]
- 12. Conway DS, Lip GY. Ethnicity in relation to atrial fibrillation and stroke (the West Birmingham Stroke Project). Am J Cardiol 2003;92(12):1476–1479. [DOI] [PubMed] [Google Scholar]
- 13. Upshaw CB, Jr. Reduced prevalence of atrial fibrillation in black patients compared with white patients attending an urban hospital: An electrocardiographic study. J Natl Med Assoc 2002;94(4):204–208. [PMC free article] [PubMed] [Google Scholar]
- 14. Rho RW, Page RL. Asymptomatic atrial fibrillation. Prog Cardiovasc Dis 2005;48(2):79–87. [DOI] [PubMed] [Google Scholar]
- 15. Hajat C, Tilling K, Stewart JA, et al. Ethnic differences in risk factors for ischemic stroke: A European case‐control study. Stroke 2004;35(7):1562–1567. [DOI] [PubMed] [Google Scholar]
- 16. Ruo B, Capra AM, Jensvold NG, et al Racial variation in the prevalence of atrial fibrillation among patients with heart failure: The epidemiology, practice, outcomes, and costs of heart failure (EPOCH) study. J Am Coll Cardiol 2004;43(3):429–435. [DOI] [PubMed] [Google Scholar]
- 17. Clark DM, Plumb VJ, Epstein AE, et al. Hemodynamic effects of an irregular sequence of ventricular cycle lengths during atrial fibrillation. J Am Coll Cardiol 1997;30(4):1039–1045. [DOI] [PubMed] [Google Scholar]
- 18. Pappone C, Santinelli V. Atrial fibrillation ablation: State of the art. Am J Cardiol 2005;96(12A):59L–64L. [DOI] [PubMed] [Google Scholar]
- 19. Li D, Fareh S, Leung TK, et al. Promotion of atrial fibrillation by heart failure in dogs: Atrial remodeling of a different sort. Circulation 1999;100(1):87–95. [DOI] [PubMed] [Google Scholar]
- 20. Shinbane JS, Wood MA, Jensen DN, et al. Tachycardia‐induced cardiomyopathy: A review of animal models and clinical studies. J Am Coll Cardiol 1997;29(4):709–715. [DOI] [PubMed] [Google Scholar]
- 21. Perez MV, Yaw TS, Myers J, et al. Prognostic value of the computerized ECG in Hispanics. Clin Cardiol 2007;30(4):189–194. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Eriksson P, Hansson PO, Eriksson H, et al. Bundle branch block in a general male population: The study of men born 1913. Circulation 1998;98(22):2494–2500. [DOI] [PubMed] [Google Scholar]
- 23. Tomaselli GF, Zipes DP. What causes sudden death in heart failure? Circ Res 2004;95(8):754–763. [DOI] [PubMed] [Google Scholar]
- 24. Bradley DJ, Bradley EA, Baughman KL, et al Cardiac resynchronization and death from progressive heart failure: A meta‐analysis of randomized controlled trials. JAMA 2003;289(6):730–740. [DOI] [PubMed] [Google Scholar]
- 25. McAlister FA, Ezekowitz JA, Wiebe N, et al Systematic review: Cardiac resynchronization in patients with symptomatic heart failure. Ann Intern Med 2004;141(5):381–390. [DOI] [PubMed] [Google Scholar]
- 26. Iuliano S, Fisher SG, Karasik PE, et al. Department of Veterans Affairs Survival Trial of Antiarrhythmic Therapy in Congestive Heart Failure. QRS duration and mortality in patients with congestive heart failure. Am Heart J 2002;143(6):1085–1091. [DOI] [PubMed] [Google Scholar]