Heart failure is an increasingly common condition arising from a variety of different pathophysiological processes. Little is known about the unique features of Israeli Arabs who present with heart failure and who undergo cardiac device implantation. The study population comprised of 4,671 patients who were enrolled in the national Israeli Implantable Cardioverter Defibrillator registry. We compared demographic, clinical, and echocardiographic characteristics; device-related indications; and outcomes between Israeli Arabs (n = 733) and Jews (n = 3,938), who were enrolled in the registry from July 2010 through December 2013. Israeli Arabs constituted 15.7% of the study population. They were younger at presentation compared with Jews (57 ± 15 vs 66 ± 12 years, respectively; p <0.001), with a greater burden of co-morbidities, including diabetes mellitus and chronic obstructive lung disease and smoking. In addition, Arab patients had a greater frequency of non-ischemic cardiomyopathy (40.2% vs 24.6%, respectively; p <0.001), which was associated with a greater frequency of familial history of sudden cardiac death. During 15 ± 9 month follow-up, the mortality rates and appropriate device therapy were similar in both ethnic groups. In conclusion, Israeli Arab patients implanted with implantable cardioverter defibrillators display unique clinical features with greater prevalence of non-ischemic cardiomyopathy characterized by an early-onset and rapid deterioration.
The prevalence of heart failure (HF) is continuously rising, and studies describing it in the Arab population are scarce. Moreover, the cause of HF in this population is not yet established. Israeli Arabs and Jews are mutually exclusive ethnic groups that together comprise the vast majority of Israel’s population (Arabs 20.7% and Jews 75%). We assumed that Israeli Arab patients with HF have diverse characteristics because of differences in environmental and genetic background. The Israeli National Implantable Cardioverter Defibrillator (ICD) Registry is a source that could be used to delineate differences between Israeli Arab and Jewish patients with HF. Therefore, we compared clinical characteristics and associated outcomes of Israeli Arab and Jewish patients with HF who were enrolled in the Israeli National ICD Registry.
Methods
The Israeli ICD Database is a prospective multicenter registry that includes all patients who had implantation or replacement of an ICD or cardiac resynchronization therapy with defibrillator (CRT-D) for primary and secondary preventions. From July 2010 through December 2013, a total of 6,731 patients underwent ICD or CRT-D implantations or replacements in 21 centers in Israel. Ethics committees of each participating institution approved the registry, and all patients provided written informed consent. For the purpose of the present analysis, we included only patients who underwent de novo device implantation and only Israeli Arabs or Jews (n = 4,671). A total of 56 (1.2%) non-Arabs and non-Jews were excluded.
The registry, initiated in July 2010, comprises a collection of data of baseline clinical and implantation characteristics. From July 2011, prospective follow-up was initiated in all participating centers. Follow-up data for clinical and arrhythmic events were obtained from consecutively enrolled patients at 6-month intervals. To date, a total of 1,752 (37.5%) patients were followed over a mean period of 15 ± 9 months (Jews, n = 1,479; Arabs, n = 273). No significant differences were found in the clinical characteristics between registry patients with follow-up data and those who were enrolled before the initiation of follow-up. Data were prospectively collected from the index hospitalization at the time of initial device implantation (or device upgrade). A local electrophysiologist from the implanting center entered information into a secure, Web-based electronic case report form, firewall and password protected. Variables collected included demographic and clinical characteristics, indication for ICD/CRT-D, left ventricular ejection fraction (LVEF), co-morbidities, hemoglobin concentration, serum creatinine levels, and previous treatments, device manufacturer, device type, and unique device identifier. The Israeli Association for Cardiovascular Trials (IACT) at Sheba Medical Center processed the primary data and performed all statistical analyses.
The following clinical end points were evaluated during follow-up: (1) all-cause mortality; (2) cardiac death; (3) first appropriate and inappropriate device therapies (any therapy, shock or ATP); and (4) all-cause and HF hospitalizations. In the present study, we further assessed the combined end point of hospitalization, death, or device therapies. Device therapies were determined as appropriate or inappropriate by an experienced clinical electrophysiologist, who reviewed the intracardiac electrograms of all events, on a case-by-case basis. Because of the significant heterogeneity of the population, a predefined subgroup analysis by HF cause (i.e., ischemic and non-ischemic cardiomyopathy) was warranted.
Baseline patient clinical characteristics and procedural data were compared between the 2 ethnic groups. Data are expressed as mean ± SD or frequency and percentage when appropriate. The chi-square test was used for dichotomous variables, and analysis of variance was used for continuous variables. Cox proportional hazard multivariate models were used to evaluate associations between ethnicity and the various outcomes. Prespecified covariates in the multivariate models included age >75 years, female gender, QRS duration >150 ms, New York Heart Association (NYHA) classes III to IV, presence of complete left bundle branch block (LBBB), non-ischemic cardiomyopathy, implantation of a CRT-D, implantation of defibrillator for secondary prevention of sudden cardiac death, and Arab ethnicity. A p value <0.05 was considered significant. All analyses were performed by the IACT using SAS statistical software version 9.4 (SAS Institute, Cary, North Carolina).
Results
From July 2010 through December 2013, a total of 4,671 patients underwent ICD or CRT-D implantations in 21 centers in Israel, of whom 733 (15.7%) were Israeli Arabs and 3,938 (84.3%) were Jews. Of note, during the study period, Israeli Arabs constituted 17.2% to 17.4% of the Israeli population. The clinical characteristics of the 2 ethnic groups are presented in Table 1 . As shown, the age at implantation was younger in Arab patients. In addition, rates of smoking and COPD were higher in Arabs, as were rates of diabetes mellitus and family history of heart disease. Notably, the proportion of patients with non-ischemic cardiomyopathy was significantly higher in Arabs (40.2% vs 24.6%, p <0.001).
| Clinical Characteristics | JEWS (n=3938) | ARABS (n=733) | P-value |
|---|---|---|---|
| Age at presentation (years) | 65.7±12.3 | 57.5±14.8 | <.001 |
| Women | 673(17%) | 152(21%) | 0.017 |
| Cardiac resynchronization therapy with defibrillator | 1521(39%) | 250(34%) | 0.021 |
| Ischemic heart disease | 2970(76%) | 438(60%) | <.001 |
| Prior myocardial infarction | 2574(87%) | 370(85%) | 0.125 |
| Non-ischemic cardiomyopathy | 968(24%) | 298(40%) | <.001 |
| Family history of sudden cardiac death | 126(3%) | 65(9%) | <.001 |
| Atrial fibrillation | 866(22%) | 121(17%) | <.001 |
| Hypertension | 2486(64%) | 459(63%) | 0.698 |
| Diabetes mellitus | 1414(36%) | 298(41%) | 0.015 |
| Smoking | 1093(28%) | 347(48%) | <.001 |
| New York Heart Association class≥3 | 1415(36%) | 270(37%) | 0.640 |
| ACE Inhibitors | 2890(74%) | 551(76%) | 0.416 |
| Diuretics | 2742(70%) | 529(72%) | 0.221 |
| Beta blockers | 3166(81%) | 603(83%) | 0.265 |
| Anti-arrhythmic | 692(18%) | 108(15%) | 0.057 |
| Ejection fraction <30% | 1964(54%) | 413(61%) | 0.003 |
| Ejection fraction, Mean± SD | 29.5±10.3 | 28.3±10.6 | <.001 |
| Left bundle brunch block | 1139(29%) | 207(28%) | <.001 |
As summarized in Table 2 , Arab patients with ischemic cardiomyopathy had a worse cardiovascular risk profile, with lower LVEF, and higher proportion of LBBB morphology. Of patients with non-ischemic cardiomyopathy, Arabs displayed higher rates of family history of sudden cardiac death (17% vs 7%, p <0.001), implying that inherited diseases may play an important role in the cause of HF in this subgroup ( Table 2 ).
| Clinical Characteristics | Ischemic cardiomyopathy | Non-ischemic cardiomyopathy | ||||
|---|---|---|---|---|---|---|
| JEWS (n=2970%) | ARABS (n=438%) | P-value | JEWS (n=968%) | ARABS (n=295%) | P-value | |
| Age at presentation (years) | 68.0±10.0 | 62.6±10.5 | <.001 | 58.9±15.8 | 50.0±16.9 | <.001 |
| Women | 327(11%) | 55(13%) | 0.338 | 346(36%) | 97(33%) | 0.367 |
| Cardiac resynchronization therapy with defibrillator | 1122(38%) | 142(32%) | 0.030 | 399(41%) | 108(37%) | 0.157 |
| Prior myocardial infarction | 2574(87%) | 370(85%) | 0.116 | 0 | 0 | 0 |
| Prior percutaneous coronary intervention | 2178(74%) | 359(82%) | <.001 | 0 | 0 | 0 |
| Heart failure | 2451(83%) | 384(88%) | 0.008 | 666(69%) | 204(69%) | 0.964 |
| Non-ischemic dilated | 0 | 0 | 0 | 562(58%) | 188(64%) | 0.078 |
| Hyperthrophic cardiomyopathy | 101(3%) | 17(4%) | 0.596 | 138(14%) | 22(7%) | 0.002 |
| Arrhythmogenic right ventricular cardiomyopathy | 12(0%) | 1(0%) | 1.000 | 18(2%) | 6(2%) | 0.848 |
| Congenital heart disease | 54(2%) | 5(1%) | 0.299 | 31(4%) | 21(8%) | 0.004 |
| History of ventricular arrhythmia | 880(30%) | 124(28%) | 0.574 | 310(32%) | 101(34%) | 0.498 |
| Family history of sudden cardiac death | 55(2%) | 15(3%) | 0.030 | 71(7%) | 50(17%) | <.001 |
| Atrial fibrillation | 633(21%) | 73(17%) | 0.027 | 233(24%) | 48(16%) | 0.005 |
| Permanent pacemaker | 160(5%) | 6(1%) | <.001 | 74(8%) | 14(5%) | 0.087 |
| Diabetes mellitus | 1191(40%) | 220(50%) | <.001 | 223(23%) | 78(27%) | 0.232 |
| Smoking | 911(31%) | 261(60%) | <.001 | 182(19%) | 86(30%) | <.001 |
| New York Heart Association class≥3 | 1089(37%) | 169(39%) | 0.437 | 326(34%) | 101(34%) | 0.859 |
| ACE Inhibitors | 2305(79%) | 359(83%) | 0.050 | 585(61%) | 192(65%) | 0.177 |
| Diuretics | 2177(74%) | 350(80%) | 0.004 | 565(59%) | 179(61%) | 0.552 |
| Beta blockers | 2445(83%) | 387(89%) | 0.002 | 721(75%) | 216(74%) | 0.672 |
| Anti-arrhythmic | 545(19%) | 70(16%) | 0.225 | 147(15%) | 38(13%) | 0.319 |
| Aspirin | 1634(78%) | 262(83%) | 0.023 | 325(48%) | 104(48%) | 0.913 |
| Ejection fraction <30% | 1520(54%) | 269(65%) | 0.001 | 444(51%) | 144(56%) | 0.071 |
| Ejection fraction | 28.4±8.0 | 26.9±8.3 | <.001 | 33.0±14.9 | 30.7±13.2 | 0.064 |
| Left bundle brunch block | 831(71%) | 122(76%) | 0.003 | 308(38%) | 85(29%) | 0.130 |
A total of 1,752 consecutive patients, of whom 273 (15.5%) were Arabs, were prospectively followed over a mean period of 15 ± 9 months. Rates of unadjusted all-cause mortality at 16 months were similar between the 2 ethnic groups, as were rates of cardiac mortality ( Figures 1 and 2 , respectively). Also, there were no statistically significant differences in the overall rates of device therapies. However, the rate of appropriate shocks during follow-up was higher in Arabs compared with Jews (3% vs 2%, respectively; p = 0.05).
