Prognosis and Clinical Characteristics of Dilated Cardiomyopathy With Family History via Pedigree Analysis

Abstract

Background: The clinical characteristics and prognostic outcomes of dilated cardiomyopathy (DCM) with a familial history (FHx) via pedigree analysis are unclear.

Methods and Results: We conducted a prospective observational study of 514 consecutive Japanese patients with DCM. FHx was defined as the presence of DCM in ≥1 family member within 2-degrees relative based on pedigree analysis. The primary endpoint was a composite of major cardiac events (sudden cardiac death and pump failure death). The prevalence of FHx was 7.4% (n=38). During a median follow-up of 3.6 years, 77 (15%) patients experienced a major cardiac event. Multivariable Cox regression analysis identified FHx as independently associated with major cardiac events (hazard ratio [HR] 4.32; 95% confidence interval [CI], 2.04–9.19; P<0.001) compared with conventional risk factors such as age, QRS duration, and left ventricular volume. In the propensity score-matched cohort (n=38 each), the FHx group had a significantly higher incidence of major cardiac events (HR, 4.48; 95% CI, 1.25–16.13; P=0.022). In addition, the FHx group had a higher prevalence of a diffuse late gadolinium enhancement (LGE) pattern than the no-FHx group (32% vs. 17%, P=0.022).

Conclusions: DCM patients with FHx had a worse prognosis, which was associated with a higher prevalence of a diffuse LGE pattern, than patients without FHx.

Dilated cardiomyopathy (DCM) is a heterogeneous disease and, depending on the diagnostic approach used, 2–65% of cases are hereditable.^1–5 Since the development of next-generation sequencing, tremendous advances have been made in understanding the genetic basis of DCM and the heritability of DCM has been attracting attention. However, it remains unclear whether the progression of the underlying disease differs among patients with and without heritability. Although several studies have assessed the prognostic effect of heritability, the results are inconsistent,^6–8 which might be related to the retrospective nature of the studies or their small sample sizes. Furthermore, previous studies used different approaches to identify heritability of DCM, such as pedigree analysis,^4,9 echocardiographic and electrocardiographic investigation of relatives,^3,8 and systematic family screening including blood tests, chest X-ray, electrocardiogram (ECG), 24-h Holter monitoring, exercise stress test, or echocardiographic examination.^2,6 Among these approaches, pedigree analysis is the simplest and easiest to apply in daily practice. However, no studies have assessed the prognostic effect of a family history (FHx) of DCM identified through pedigree analysis.

Therefore, we performed a prospective study to assess the prognostic outcomes of FHx based on pedigree analysis in an established cohort of patients with DCM.¹⁰ We also performed further analysis of myocardial fibrosis in patients with FHx vs. no-FHx using late gadolinium enhancement (LGE) on cardiac magnetic resonance (CMR).

Methods

Study Population

We conducted a prospective observational study of 514 consecutive Japanese patients with DCM at the National Cerebral and Cardiovascular Center, Suita, Japan, between April 2007 and December 2015. The diagnosis of DCM was based on World Health Organization criteria and left ventricular ejection fraction (LVEF) <50%.¹¹ All patients underwent invasive coronary angiography or computed tomography angiography (CTA) to rule out significant coronary artery stenosis (>50% diameter stenosis).¹² Patients with a history of myocardial infarction or coronary revascularization, myocarditis, hypertrophic cardiomyopathy, secondary cardiomyopathy, valvular heart disease, or hypertensive heart disease were excluded. One patient with myocarditis diagnosed by endomyocardial biopsy, 2 patients with peripartum cardiomyopathy and 1 patient with Adriamycin-related cardiomyopathy were classified as secondary cardiomyopathy and excluded. Patients younger than 18 years of age were also excluded. CMR was performed while the patient was in a clinically stable, non-congested condition (New York Heart Association [NYHA] class ≤II). None of the patients had a typical subendocardial or transmural LGE pattern in the territory supplied by a coronary artery that might have resulted from myocardial damage secondary to coronary artery disease or coronary embolism. The study protocol conformed to the ethical guidelines of the 1975 Declaration of Helsinki and was approved by the institutional review board and ethics committee of the National Cerebral and Cardiovascular Center (M24-081).

Diagnostic Approach for FHx

Based on pedigree analysis of 3 generations,⁹ FHx was defined based on data from medical records obtained in 2 family members: the index patient and at least 1 1st- or 2nd-degree family member with DCM.⁹ We did not perform family screening in this study. We enrolled index patients with DCM who visited the Center, but not relatives with DCM who were identified by pedigree analysis in the present study.

CMR Protocol

CMR examinations were performed using a 1.5-T magnetic resonance (MR) imaging system (Magnetom Sonata, Siemens, Erlangen, Germany) with a 4-channel surface coil. The procedures used to acquire MR images in this study have been previously described.^10,13,14 Briefly, we identified LGE using segmented inversion recovery prepared true fast imaging with steady-state precession sequence with ECG triggering 10 min after the administration of 0.15 mmol/kg body weight of gadolinium diethylenetriamine pentaacetic acid. LGE was only considered present if it was visible in 2 orthogonal views.^15,16 Mid-wall LGE was only considered present if the area of LGE was confined to the intermural and/or subepicardial layers.¹⁷ In cases other than mid-wall LGE, LGE in multiple territories was defined as diffuse LGE, and isolated LGE was defined as focal LGE.¹³ Experienced radiologists (Y.M. and H.M.), who were blinded to clinical data and outcomes, independently assessed LGE presence and location. A third blinded reader adjudicated in cases with disagreement (4.9%). There was acceptable interobserver concordance for diffuse LGE (κ=0.88), mid-wall LGE (κ=0.89), and focal LGE (κ=0.83). There was also acceptable intraobserver concordance for diffuse LGE (κ=0.90), mid-wall LGE (κ=0.92), and focal LGE (κ=0.89).

Determination of B-Type Natriuretic Peptide (BNP) Levels

Blood samples were collected in tubes containing ethylenediaminetetraacetic acid. Plasma BNP levels were measured using a validated commercially available immunoassay kit (Tosoh, Tokyo, Japan).

Follow-up and Endpoints

After the CMR data were obtained, study patients were followed at 3 months, 6 months, and 12 months and annually thereafter until the occurrence of any of the following events: sudden cardiac death (SCD), aborted SCD (nonfatal ventricular fibrillation [VF], sustained ventricular tachycardia [VT], or appropriate implantable cardioverter-defibrillator [ICD] discharge for VT or VF), pump failure death (heart failure [HF] death, cardiac transplantation, or left ventricular assist device [LVAD] implantation), or rehospitalization for HF. The duration of the follow-up period was calculated from baseline CMR until an endpoint occurred or the last patient contact. The primary endpoint was a composite of major cardiac events (SCD, aborted SCD, or pump failure death). The secondary endpoint was rehospitalization for HF. Independent attending cardiologists (E.T. and H.M.) blinded to the patient’s baseline characteristics reviewed the medical records to determine if hospitalizations and deaths qualified as cardiac events. SCD was defined as unexpected death either within 1 h of cardiac symptoms in the absence of progressive cardiac deterioration, during sleep, or within 24 h of last being seen alive.¹⁸ HF death was defined as death associated with unstable, progressive deterioration of pump function despite active therapy. Aborted SCD was diagnosed in patients who received an appropriate ICD discharge for VT or VF, had nonfatal VF, or spontaneous sustained VT (>30 s) causing hemodynamic compromise requiring cardioversion.¹⁹ Rehospitalization for HF was defined as hospital admission for signs and symptoms of decompensated HF requiring treatment with an intravenous HF medication (diuretic, vasodilator, or inotropic agent). For composite endpoints, only the first event for each patient was included in the analysis.

Statistical Analysis

All continuous variables are presented as mean±SD. Unpaired t-tests were used to compare groups. Non-normally distributed variables are presented as median (interquartile range [IQR]). Noncontinuous and categorical variables are presented as frequencies or percentages, and were compared using the χ² test. Cumulative event-free survival curves were estimated using the Kaplan-Meier method and compared using the log-rank test. Univariable Cox proportional hazards regression models were used to calculate hazard ratios (HRs) and 95% confidence intervals (CIs) for each endpoint. Multivariable Cox regression analysis was performed using covariates that significantly predicted each endpoint in the univariable analysis or are established prognostic risk factors for chronic HF (age, sex, diabetes mellitus, BNP, estimated glomerular filtration rate [eGFR], hemoglobin, β-blocker use, and angiotensin-converting enzyme inhibitor [ACEI] or angiotensin-receptor blocker [ARB] use). Stepwise selection with a P-value of 0.05 for forward selection was used to select the best predictive model.

A propensity score was used to adjust for baseline covariates that were significantly different between groups or could influence outcomes: age; sex; BNP; β-blocker, ACEI, ARB, or aldosterone antagonist use; and LVEF. Propensity score-matched cohorts were constructed by matching patients in the 2 groups on a 1:1 basis using the nearest-neighbor matching method within a caliper of 0.01 of the propensity score. Details on propensity score matching are described in Supplementary Table 1. All statistical tests were two-sided and P<0.05 was regarded as statistically significant. All statistical analyses were performed with SPSS software (version 24.0; IBM Corp, Armonk, NY, USA) and Stata 15 (StataCorp, College Station, TX, USA).

Results

Baseline Clinical Characteristics

At baseline, 530 Japanese patients met the inclusion criteria. The follow-up rate was 96.9%; 16 patients were lost to follow-up. Ultimately, 514 patients were included in the outcome analysis (Figure 1), of whom 127 (25%) were women. Of these 514 patients, 409 (80%) underwent endomyocardial biopsy and did not have findings of secondary cardiomyopathy. Although the remaining 20% of patients did not undergo endomyocardial biopsy either because of advanced age (>80 years) or because they declined the procedure, none had clinical features specific for secondary cardiomyopathy. The mean LVEF was 26±11%. A total of 38 (7.4%) patients had FHx: 20 patients had FHx in a 1st-degree relative, 14 patients had FHx in a 2nd-degree relative, and 4 patients had FHx in both 1st- and 2nd-degree relatives.

Figure 1.

Flow chart detailing the inclusion and exclusion criteria for study patients. Enrolled patients were divided into 2 groups based on FHx of DCM. CMR, cardiac magnetic resonance; CTA, computed tomography angiography; DCM, dilated cardiomyopathy; FHx, family history; LVEF, left ventricular ejection fraction.

We divided the study patients into 2 groups according to the presence or absence of FHx: FHx (n=38) or no-FHx (n=476). Baseline characteristics are summarized in Table 1. Patients in the FHx group were younger (P=0.002), but there were no significant differences in sex; NYHA functional class; β-blocker, ACEI or ARB use; BNP levels; and LVEF and LV volumes between the 2 groups. Although the proportion of patients with presence of LGE was similar between groups (P=0.115), the prevalence of a diffuse LGE pattern, which represents severe myocardial fibrosis, was significantly higher in the FHx group than in the no-FHx group (32% vs. 17%; P=0.022, Figure 2).

Table 1. Baseline Characteristics of the Study Patients With Dilated Cardiomyopathy

	All patients (n=514)	No-FHx (n=476)	FHx (n=38)	P value
Age, years	53±16	53±16	45±12	0.002
Male, n (%)	387 (75)	359 (75)	28 (74)	0.811
BMI, kg/m2	22.9±4.0	22.9±4.0	23.2±4.1	0.632
Current smoker, n (%)	117 (23)	111 (23)	6 (16)	0.266
NYHA functional class, n (%)				0.813
I	148 (29)	137 (29)	11 (29)
II	226 (44)	211 (44)	15 (39)
III	81 (16)	73 (15)	8 (21)
IV	59 (11)	55 (12)	4 (11)
Medical history, n (%)
Diabetes mellitus, n (%)	100 (19)	97 (20)	3 (8)	0.061
Atrial fibrillation	157 (31)	145 (30)	12 (32)	0.886
Sustained VT	11 (2.1)	11 (2.3)	0	0.344
Nonfatal VF	4 (0.8)	4 (0.8)	0	0.571
BNP, pg/mL	200 [55–561]	196 [54–556]	155 [79–593]	0.640
Creatinine, mg/dL	0.87±0.24	0.88±0.24	0.85±0.22	0.479
eGFR, mL/min/1.73 m2	73±22	73±22	78±20	0.201
Serum sodium, mEq/L	140.1±2.9	140.1±2.9	140.1±2.0	0.955
Hemoglobin, g/dL	14.0±1.6	14.0±1.6	14.4±1.5	0.131
ECG parameters
Heart rate, beats/min	69±12	69±13	67±11	0.360
PR duration, ms (n=415)	181±31	181±31	181±33	0.912
Prolonged PR interval, n (%)	79 (19)	73 (19)	6 (19)	0.963
QRS duration, ms	118±29	118±29	121±27	0.572
Wide QRS duration, n (%)	169 (33)	155 (33)	14 (33)	0.589
CLBBB, n (%)	87 (17)	81 (17)	6 (16)	0.846
CRBBB, n (%)	30 (5.8)	26 (5.5)	4 (11)	0.200
IVCD, n (%)	49 (9.5)	45 (9.5)	4 (10.5)	0.829
QTc interval, ms	446±42	446±42	447±39	0.780
Medications at baseline, n (%)
β-blocker	460 (89)	426 (89)	34 (89)	0.997
ACEI or ARB	417 (81)	387 (81)	30 (79)	0.721
Aldosterone antagonist	244 (47)	219 (46)	25 (66)	0.019
Diuretic	286 (56)	266 (56)	20 (53)	0.698
Digoxin	98 (19)	93 (19)	5 (13)	0.335
CMR measurements
LVEDVI, mL/m2	145±54	145±55	154±51	0.310
LVESVI, mL/m2	111±54	110±54	120±52	0.275
LVSVI, mL/m2	35±12	35±12	34±10	0.761
LV mass, g	141±51	142±51	131±48	0.236
LVEF, %	26±11	27±11	24±10	0.185
RVEF, %	35±11	35±11	35±10	0.868
LGE, n (%)	303 (59)	276 (58)	27 (71)	0.115
Mid-wall pattern	163 (32)	151 (32)	12 (32)	0.985
Focal pattern	48 (9.3)	45 (9.5)	3 (7.9)	0.751
Diffuse pattern	92 (18)	80 (17)	12 (32)	0.022

ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin-receptor blocker; BMI, body mass index; BNP, B-type natriuretic peptide; CLBBB, complete left bundle branch block; CMR, cardiac magnetic resonance; CRBBB, complete right bundle branch block; DCM, dilated cardiomyopathy; ECG, electrocardiogram; EDVI, end-diastolic volume index; EF, ejection fraction; eGFR, estimated glomerular filtration rate; ESVI, end-systolic volume index; FHx, family history; IVCD, interventricular conduction delay; LGE, late gadolinium enhancement; LV, left ventricular; NYHA, New York Heart Association; RV, right ventricular; SVI, stroke volume index; VF, ventricular fibrillation; VT, ventricular tachycardia.

Figure 2.

(A–C) Representative short-axis images with LGE (arrows) and (D) prevalence of LGE patterns in the FHx and no-FHx groups. FHx, family history; LGE, late gadolinium enhancement.

Primary Endpoint: Composite of Major Cardiac Events

During a median follow-up period of 3.6 years (IQR, 1.7–5.5 years), 77 (15%) patients reached the primary endpoint. Table 2 summarizes the incidence of the primary endpoint. Patients with FHx had a higher incidence of major cardiac events (29% vs. 14%; P=0.012), aborted SCD (16% vs. 6.9%; P=0.047), and cardiac transplantation or LVAD implantation (13% vs. 2.7%; P=0.001) than patients without FHx. Patients with FHx had a higher incidence of ICD implantation than patients without FHx (34% vs. 18%, P=0.012). Among patients who underwent ICD implantation, appropriate ICD discharge for VT or VF occurred in 6 of 13 (46%) patients in the FHx group and in 27 of 84 (32%) patients in the no-FHx group (P=0.321). Figure 3A shows the Kaplan-Meier curves for survival free from the primary endpoint. The FHx group had more patients who reached the primary endpoint than the no-FHx group (P=0.010). The estimated 5-year rate for the primary endpoint was higher in the FHx group than in the no-FHx group (44% vs. 15%; P=0.011) (Figure 3C). Table 3 shows the univariable and multivariable Cox regression analyses of risk factors for the primary endpoint. In model 1, a simultaneous forced entry multivariable Cox model that adjusted for factors that were significant in the univariable analysis, the following were identified as significant predictors for major cardiac events: history of sustained VT or nonfatal VF, LV end-diastolic volume index, and presence of LGE. To further investigate whether FHx was a better cardiac prognostic factor compared with conventional risk factors, we performed a multivariable Cox regression analysis that adjusted for factors that were significant in the univariable analysis as well as established risk factors for chronic HF (age, sex, diabetes mellitus, BNP, eGFR, hemoglobin level, β-blocker use, and ACEI or ARB use) (model 2). Because the number of events in this study was relatively low, we found that the best predictive model adjusted for significant predictors selected in a stepwise Cox regression analysis based on models 1 and 2. FHx (HR, 4.32; 95% CI, 2.04–9.19; P<0.001) remained a significant indicator of major cardiac events.

Table 2. Summary of Cardiac Events During the Follow-up Period

Outcome, n (%)	All patients (n=514)	No-FHx (n=476)	FHx (n=38)	P value
Primary endpoint (composite of major cardiac events)	77 (15)	66 (14)	11 (29)	0.012
SCD	11 (2.1)	11 (2.3)	0	0.343
Aborted SCD	39 (7.6)	33 (6.9)	6 (16)	0.047
Sustained VT	6 (1.2)	5 (1.1)	1 (2.6)	0.383
Nonfatal VF	2 (0.4)	2 (0.4)	0	0.689
Appropriate ICD discharge for VT/VF	31 (6.0)	26 (5.5)	5 (13)	0.055
Pump failure death	27 (5.3)	22 (4.6)	5 (13)	0.023
HF death	9 (1.8)	9 (1.9)	0	0.392
Cardiac transplantation or LVAD implantation	18 (3.5)	13 (2.7)	5 (13)	0.001
Secondary endpoint (rehospitalization for HF)	64 (13)	53 (11)	11 (29)	0.001

HF, heart failure; ICD, implantable cardiovascular defibrillator; LVAD, left ventricular assist device; SCD, sudden cardiac death. Other abbreviations as in Table 1.

Figure 3.

Kaplan-Meier curves and 5-year rates of major cardiac events and rehospitalization for HF. Kaplan-Meier curves showing survival free from the composite endpoint of major cardiac events (A) and rehospitalization for HF (B). Kaplan-Meier analysis of the composite endpoint of major cardiac events and rehospitalization for HF demonstrated that the FHx group had significantly worse long-term outcomes than the no-FHx group. Among the 2 groups, the 5-year event rates for major cardiac events (C) and rehospitalization for HF (D) were higher in the FHx group. FHx, family history; HF, heart failure.

Table 3. Univariable and Multivariable Cox Regression Analyses of Risk Factors for Composite of Major Cardiac Events

	Univariable analysis			Multivariable analysis
				Model 1*			Best predictive model‡
	HR	95% CI	P value	HR	95% CI	P value	HR	95% CI	P value
Age, per 10 year increment	0.99	0.86–1.14	0.897				1.28	1.06–1.54	0.010
Male sex	1.63	0.91–2.97	0.106
BMI, per kg/m2 increment	0.98	0.93–1.04	0.530
NYHA class ≥II	3.60	1.80–7.22	<0.001	2.01	0.94–4.33	0.073
Current smoker	1.06	0.62–1.80	0.843
Diabetes mellitus	1.02	0.58–1.79	0.958
Atrial fibrillation	1.43	0.90–2.27	0.128
History of sustained VT or nonfatal VF	3.38	1.55–7.35	0.002	3.17	1.30–7.72	0.011
FHx of DCM	2.38	1.26–4.52	0.008	1.91	0.99–3.68	0.055	4.32	2.04–9.19	<0.001
ECG parameters
Heart rate, per 10 beats/min increment	1.01	0.84–1.20	0.958
QRS duration, per 10 ms increment	1.18	1.10–1.26	<0.001	1.06	0.97–1.15	0.192	1.00	1.01–1.21	0.022
QTc, per 10 ms increment	1.09	1.04–1.14	<0.001	1.01	0.95–1.08	0.69
Log (BNP), per 1 pg/mL increment	2.14	1.48–3.09	<0.001	1.11	0.70–1.76	0.651
eGFR, per 1 mL/min/1.73 m2 increment	1.00	0.99–1.01	0.407
Serum sodium, per 1 mEq/L decrement	1.07	1.02–1.13	0.013	0.98	0.91–1.05	0.580
Hemoglobin, per 1 g/dL decrement	1.08	0.82–1.07	0.328
Medications
β-blocker	0.91	0.42–1.98	0.807
ACEI or ARB	1.40	0.74–2.65	0.305
Diuretic	2.65	1.56–4.50	<0.001	1.40	0.77–2.53	0.272	2.55	1.29–5.05	0.007
Amiodarone	4.04	2.46–6.64	<0.001	1.27	0.72–2.23	0.409
Digoxin	0.98	0.56–1.69	0.931
CMR parameters
LVEDVI, per 10 mL/m2 increment	1.14	1.10–1.18	<0.001	1.09	1.04–1.14	<0.001	1.09	1.05–1.14	<0.001
LVSVI, per 10 mL/m2 increment	0.96	0.79–1.16	0.647
LV mass, per 1 g increment	1.00	0.99–1.01	0.572
Presence of LGE	6.09	3.13–11.85	<0.001	4.09	2.05–8.17	<0.001

*Multivariable Cox models were selected using a simultaneous forced entry method with factors that were significant in the univariable analysis. QRS duration, aldosterone antagonist, LVESVI, LVEF, and RVEF were dropped due to high collinearity, as suggested by variance inflation factors >10. ^‡Best predictive model, adjusted for significant predictors selected by stepwise Cox regression analysis using factors that were significant in the univariable analysis and established risk factors for prognosis (age, sex, diabetes mellitus, BNP, eGFR, hemoglobin, β-blocker use, and ACEI or ARB use). CI, confidence interval; HR, hazard ratio. Other abbreviations as in Table 1.

Secondary Endpoint: Rehospitalization for HF

Rehospitalization for HF occurred in 64 (13%) patients (Table 2). A higher proportion of patients in the FHx group reached this outcome (P<0.001) (Figure 3B). The FHx group had a higher estimated 5-year rate of rehospitalization for HF (42% vs. 16%; P<0.001) (Figure 3D) than the no-FHx group. In the multivariable model based on stepwise Cox regression analysis, FHx was a significant predictor of rehospitalization for HF (HR, 4.32; 95% CI, 2.04–9.19; P<0.001) (Table 4).

Table 4. Univariable and Multivariable Cox Regression Analyses of Risk Factors for Rehospitalization for HF

	Univariable analysis			Multivariable analysis
				Model 1*			Best predictive model‡
	HR	95% CI	P value	HR	95% CI	P value	HR	95% CI	P value
Age, per 10 year increment	1.21	1.03–1.42	0.024	1.15	0.94–1.40	0.185	1.28	1.06–1.54	0.010
Male sex	0.81	0.47–1.40	0.455
BMI, per kg/m2 increment	0.98	0.93–1.05	0.615
NYHA class ≥II	2.24	1.17–4.29	0.015	0.85	0.40–1.78	0.663
Current smoker	0.48	0.24–0.98	0.044	0.49	0.40–1.78	0.066
Diabetes mellitus	1.46	0.83–2.57	0.193
Atrial fibrillation	5.67	0.68–47.14	0.109
History of sustained VT or nonfatal VF	0.42	0.06–3.05	0.393
FHx of DCM	3.50	1.83–6.72	<0.001	4.36	1.96–9.68	<0.001	4.32	2.04–9.19	<0.001
ECG parameters
Heart rate, per 10 beats/min increment	0.87	0.81–1.08	0.200
QRS duration, per 10 ms increment	1.20	1.17–1.28	<0.001	1.10	0.99–1.21	0.075	1.11	1.01–1.21	0.022
QTc, per 10 ms increment	1.10	1.05–1.16	<0.001	1.01	0.95–1.08	0.687
Log (BNP), per 1 pg/mL increment	2.31	1.54–3.26	<0.001	1.48	0.82–2.70	0.196
eGFR, per 1 mL/min/1.73 m2 increment	1.00	0.98–1.01	0.433
Serum sodium, per 1 mEq/L decrement	1.05	0.97–1.14	0.236
Hemoglobin, per 1 g/dL decrement	1.17	1.02–1.36	0.029	1.12	0.93–1.35	0.234
Medications
β-blocker	3.14	0.77–12.86	0.111
ACEI or ARB	1.64	0.78–3.44	0.191
Diuretic	3.72	1.99–6.98	<0.001	2.57	1.26–5.24	0.009	2.25	1.29–5.05	0.007
Amiodarone	2.09	1.12–3.91	0.022	0.70	0.33–1.46	0.342
Digoxin	1.09	0.60–1.98	0.768
CMR parameters
LVEDVI, per 10 mL/m2 increment	1.12	1.08–1.16	<0.001	1.08	1.02–1.14	0.012	1.09	1.05–1.15	<0.001
LVSVI, per 10 mL/m2 increment	1.07	0.89–1.28	0.469
LV mass, per 1 g increment	1.01	1.00–1.01	0.036	1.00	1.00–1.01	0.757
RVEF, per 5% decrement	1.12	0.99–1.26	0.052
Presence of LGE	2.75	1.54–4.90	0.001	1.97	1.05–3.69	0.035

*Multivariable Cox models were selected using a simultaneous forced entry method with factors that were significant in the univariable analysis. QRS duration, Aldosterone antagonist, LVESVI, and LVEF were dropped due to high collinearity, as suggested by variance inflation factors >10. ^‡Best predictive model, adjusted for significant predictors selected by stepwise Cox regression analysis based on factors that were significant in the univariable analysis and established risk factors for prognosis (age, sex, diabetes mellitus, BNP, eGFR, hemoglobin, β-blocker use, and ACEI or ARB use). Abbreviations as in Tables 1,3.

Propensity Score-Matched Analyses for Composite Endpoint of Major Cardiac Events and Rehospitalization for HF

In further analysis, a propensity score-matched cohort consisting of 38 patients with FHx and 38 patients without FHx was constructed (Supplementary Tables 1,2). In this model that matched for age, sex, BNP, β-blocker, ACEI, ARB, or aldosterone antagonist use, and LVEF, the composite endpoint of major cardiac events (HR, 4.48; 95% CI, 1.25–16.13; P=0.022) and rehospitalization for HF (HR, 3.38; 95% CI, 1.08–10.63; P=0.037) occurred more frequently in the FHx group than in the no-FHx group. A higher proportion of patients in the FHx group reached the composite endpoint of major cardiac events (P=0.004, Figure 4A), and rehospitalization for HF (P=0.027, Figure 4B). In addition, the FHx group had a higher 5-year event rate for the composite endpoint of major cardiac events (44%, P=0.011, Figure 4C) and rehospitalization for HF (42%, P=0.027, Figure 4D).

Figure 4.

Kaplan-Meier curves and 5-year rates of (A,C) major cardiac events and (B,D) rehospitalization for heart failure (HF) in propensity score-matched groups. Kaplan-Meier analysis of the composite endpoint of major cardiac events and rehospitalization for HF demonstrated that the FHx group had significantly worse long-term outcomes than the no-FHx group. Among the 2 groups, the 5-year event rates were higher in the FHx group. FHx, family history.

Discussion

This prospective cohort study of DCM from Japan is the first to assess the prognostic effect of FHx identified by pedigree analysis. We found that patients with FHx had a high risk for major cardiac events and rehospitalization for HF. In particular, the propensity score-matched model confirmed FHx as a significant prognostic factor for major cardiac events and rehospitalization for HF. We also showed that the FHx group had a higher prevalence of a diffuse LGE pattern than the no-FHx group.

Several retrospective studies have assessed the prognostic effect of the heritability of DCM (Supplementary Table 3). In a study of 240 Hungarian patients with DCM, including 31 patients with hereditary DCM identified through population screening using chest X-rays, and 209 patients with idiopathic DCM, survival at 6 years was significantly lower in the hereditary DCM group than in the idiopathic DCM group (6% vs. 23%; P<0.05).⁷ Michels et al conducted a study of 30 patients who had hereditary DCM identified through familial screening of echocardiographic findings and 71 patients with idiopathic DCM. During the follow-up period of 6.0 years, there was no significant difference in the survival rate between the hereditary DCM (51.5%) and idiopathic DCM (55.7%) groups.⁸ Moretti et al conducted a matched cohort study of 48 patients with asymptomatic hereditary DCM identified through a systematic clinical and echocardiographic screening program and 96 idiopathic DCM patients. They found that the rate of survival free from heart transplant at 5 years was lower in the hereditary DCM group than in the idiopathic DCM group (76% vs. 91%, P=0.04). However, after stratification by NYHA class, the difference disappeared.⁶ Thus, it remained unclear whether or not heritability of DCM affected disease progression. Therefore, we conducted a prospective observational study that included 514 patients with DCM. Indeed, the incidence of major cardiac events and rehospitalization for HF was higher in the FHx group than in the no-FHx group. Furthermore, a propensity score-matched model showed that major cardiac events (HR, 4.48; 95% CI, 1.25–16.13; P=0.022) and rehospitalization for HF (HR, 3.38; 95% CI, 1.08–10.63; P=0.037) occurred in a significantly higher proportion of patients in the FHx group than in the no-FHx group. In addition, we showed that the FHx group had a higher prevalence of a diffuse LGE pattern than the no-FHx group. Diffuse LGE represents severe myocardial fibrosis, which is associated with poor prognosis in patients with DCM.^20,21 Therefore, our study showed that patients with FHx identified through pedigree analysis are at high risk of having diffuse myocardial fibrosis and future cardiac events.

DCM is associated with remarkable genetic heterogeneity. The cumulative effect of multiple genetic variants may affect DCM;²² to date, >40 genes have been shown to cause DCM in humans.²³ For example, variants in LMNA or RBM20 are associated with high penetrance and poor prognosis.^24,25 Variants in TTN are the most common cause of DCM, and are associated with incomplete penetrance and often milder phenotypes.²⁶ A recent study of a Japanese cohort showed that LMNA variants were associated with a higher prevalence of hereditary DCM (92% vs. 25%; P<0.001) and worse outcomes, including life-threatening arrhythmic events, than TTN variants.²⁷ Thus, the DCM patients with FHx in our study might have genetic variants with high penetrance and poor prognosis. Further study with genetic testing is warranted to assess the mechanisms responsible for poor prognosis in patients with FHx.

The worse outcomes observed in patients with FHx than in those without FHx could be explained by genetic anticipation. In order to assess for the presence of genetic anticipation, further study is needed to collect clinical data, including information on genetic variants, to compare age of onset and DCM severity by generation.

Although pedigree analysis does not have enough sensitivity for hereditary DCM detection, this simple approach is efficient for identifying patients with a high risk of future cardiac events.

Study Limitations

First, this study was performed in a single, high-volume center, which introduces the possibility of referral bias. Second, the prevalence of FHx in our study (7.4%) was lower than in previous studies based on pedigree analysis,^4,7,9 although we used the same criteria to diagnose FHx as a previous study.⁹ Haas et al showed statistically significant difference of DCM mutation even among 8 European countries.²⁸ Thus, the difference of prevalence between our study and the other studies may be related to the genetic heterogeneity of DCM between Japanese and Western individuals. Third, it is possible that late onset may have masked the disease phenotype in asymptomatic relatives. This problem is difficult to resolve until all genetic etiologies of DCM have been discovered and assessed. Fourth, we did not perform any genetic examinations. According to a recent study with next-generation sequencing of 639 patients with DCM, genetic mutations were identified in 203 of 265 patients with hereditary DCM.²⁸ Unfortunately, however, it is difficult to routinely offer genetic testing to patients with DCM even in a referral center in Japan. Future studies should address the clinical importance of genetic mutations in patients with FHx. Fifth, although lower eGFR has been reported as a prognostic factor for cardiac events in patients with chronic HF, patients with chronic renal insufficiency (eGFR <30 mL/min/1.73 m²) were excluded because of the risk of nephrogenic systemic fibrosis associated with gadolinium exposure. Finally, we did not perform further evaluation of diffuse fibrosis using T1 mapping to quantify fibrosis.²⁹ Puntmann et al³⁰ showed that native T1 is significantly associated with all-cause death and HF events in patients with DCM. Future studies with T1 mapping should investigate the relationship between diffuse fibrosis and FHx.

Conclusions

Patients with FHx of DCM had a higher prevalence of severe myocardial fibrosis detected through LGE and worse prognosis than patients without FHx.

Acknowledgements

We give special thanks to Ms. Yuko Yoshioka and Ms. Emi Kanai for their excellent assistance as a Clinical Research Coordinator.

Data Availability

The deidentified participant data will not be shared.

Disclosures

The present work was supported in part by the Takeda Science Foundation and the Japan Cardiovascular Research Foundation. H.O. and S.Y. are members of Circulation Journal ’ Editorial Team.

The authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.

IRB Information

The National Cerebral and Cardiovascular Center (M24-081).

Supplementary Files

Please find supplementary file(s);

http://dx.doi.org/10.1253/circj.CJ-19-1176

References

• 1.   Petretta M, Pirozzi F, Sasso L, Paglia A, Bonaduce D. Review and metaanalysis of the frequency of familial dilated cardiomyopathy. Am J Cardiol 2011; 108: 1171–1176.
• 2.   Baig MK, Goldman JH, Caforio AL, Coonar AS, Keeling PJ, McKenna WJ. Familial dilated cardiomyopathy: Cardiac abnormalities are common in asymptomatic relatives and may represent early disease. J Am Coll Cardiol 1998; 31: 195–201.
• 3.   Michels VV, Moll PP, Miller FA, Tajik AJ, Chu JS, Driscoll DJ, et al. The frequency of familial dilated cardiomyopathy in a series of patients with idiopathic dilated cardiomyopathy. N Engl J Med 1992; 326: 77–82.
• 4.   Grunig E, Tasman JA, Kucherer H, Franz W, Kubler W, Katus HA. Frequency and phenotypes of familial dilated cardiomyopathy. J Am Coll Cardiol 1998; 31: 186–194.
• 5.   Hershberger RE, Siegfried JD. Update 2011: Clinical and genetic issues in familial dilated cardiomyopathy. J Am Coll Cardiol 2011; 57: 1641–1649.
• 6.   Moretti M, Merlo M, Barbati G, Di Lenarda A, Brun F, Pinamonti B, et al. Prognostic impact of familial screening in dilated cardiomyopathy. Eur J Heart Fail 2010; 12: 922–927.
• 7.   Csanady M, Hogye M, Kallai A, Forster T, Szarazajtai T. Familial dilated cardiomyopathy: A worse prognosis compared with sporadic forms. Br Heart J 1995; 74: 171–173.
• 8.   Michels VV, Driscoll DJ, Miller FA, Olson TM, Atkinson EJ, Olswold CL, et al. Progression of familial and non-familial dilated cardiomyopathy: Long term follow up. Heart 2003; 89: 757–761.
• 9.   Seidelmann SB, Laur O, Hwa J, Depasquale E, Bellumkonda L, Sugeng L, et al. Familial dilated cardiomyopathy diagnosis is commonly overlooked at the time of transplant listing. J Heart Lung Transplant 2016; 35: 474–480.
• 10.   Marume K, Noguchi T, Tateishi E, Morita Y, Kamakura T, Ishibashi K, et al. Mortality and sudden cardiac death risk stratification using the noninvasive combination of wide QRS duration and late gadolinium enhancement in idiopathic dilated cardiomyopathy. Circ Arrhythm Electrophysiol 2018; 11: e006233.
• 11.   Richardson P, McKenna W, Bristow M, Maisch B, Mautner B, O’Connell J, et al. Report of the 1995 World Health Organization/International Society and Federation of Cardiology Task Force on the definition and classification of cardiomyopathies. Circulation 1996; 93: 841–842.
• 12.   Felker GM, Shaw LK, O’Connor CM. A standardized definition of ischemic cardiomyopathy for use in clinical research. J Am Coll Cardiol 2002; 39: 210–218.
• 13.   Tateishi E, Noguchi T, Goto Y, Morita Y, Ishibashi-Ueda H, Yamada N, et al. Prognostic impact of blood pressure response plus gadolinium enhancement in dilated cardiomyopathy. Heart 2015; 101: 774–780.
• 14.   Yoshida A, Ishibashi-Ueda H, Yamada N, Kanzaki H, Hasegawa T, Takahama H, et al. Direct comparison of the diagnostic capability of cardiac magnetic resonance and endomyocardial biopsy in patients with heart failure. Eur J Heart Fail 2013; 15: 166–175.
• 15.   Golwala H, Pandey A, Ju C, Butler J, Yancy C, Bhatt DL, et al. Temporal trends and factors associated with cardiac rehabilitation referral among patients hospitalized with heart failure: Findings from Get With The Guidelines-Heart Failure Registry. J Am Coll Cardiol 2015; 66: 917–926.
• 16.   Klem I, Weinsaft JW, Bahnson TD, Hegland D, Kim HW, Hayes B, et al. Assessment of myocardial scarring improves risk stratification in patients evaluated for cardiac defibrillator implantation. J Am Coll Cardiol 2012; 60: 408–420.
• 17.   Gulati A, Jabbour A, Ismail TF, Guha K, Khwaja J, Raza S, et al. Association of fibrosis with mortality and sudden cardiac death in patients with nonischemic dilated cardiomyopathy. JAMA 2013; 309: 896–908.
• 18.   Hicks KA, Tcheng JE, Bozkurt B, Chaitman BR, Cutlip DE, Farb A, et al. 2014 ACC/AHA key data elements and definitions for cardiovascular endpoint events in clinical trials: A report of the American College of Cardiology/American Heart Association Task Force on Clinical Data Standards (Writing Committee to Develop Cardiovascular Endpoints Data Standards). Circulation 2015; 132: 302–361.
• 19.   Buxton AE, Calkins H, Callans DJ, DiMarco JP, Fisher JD, Greene HL, et al. ACC/AHA/HRS 2006 key data elements and definitions for electrophysiological studies and procedures: A report of the American College of Cardiology/American Heart Association Task Force on Clinical Data Standards (ACC/AHA/HRS Writing Committee to Develop Data Standards on Electrophysiology). J Am Coll Cardiol 2006; 48: 2360–2396.
• 20.   Neilan TG, Coelho-Filho OR, Danik SB, Shah RV, Dodson JA, Verdini DJ, et al. CMR quantification of myocardial scar provides additive prognostic information in nonischemic cardiomyopathy. JACC Cardiovasc Imaging 2013; 6: 944–954.
• 21.   Lehrke S, Lossnitzer D, Schob M, Steen H, Merten C, Kemmling H, et al. Use of cardiovascular magnetic resonance for risk stratification in chronic heart failure: Prognostic value of late gadolinium enhancement in patients with non-ischaemic dilated cardiomyopathy. Heart 2011; 97: 727–732.
• 22.   Bezzina CR, Lahrouchi N, Priori SG. Genetics of sudden cardiac death. Circ Res 2015; 116: 1919–1936.
• 23.   Cho KW, Lee J, Kim Y. Genetic variations leading to familial dilated cardiomyopathy. Mol Cells 2016; 39: 722–727.
• 24.   McNally EM, Mestroni L. Dilated cardiomyopathy: Genetic determinants and mechanisms. Circ Res 2017; 121: 731–748.
• 25.   Hershberger RE, Hedges DJ, Morales A. Dilated cardiomyopathy: The complexity of a diverse genetic architecture. Nat Rev Cardiol 2013; 10: 531–547.
• 26.   Herman DS, Lam L, Taylor MR, Wang L, Teekakirikul P, Christodoulou D, et al. Truncations of titin causing dilated cardiomyopathy. N Engl J Med 2012; 366: 619–628.
• 27.   Tobita T, Nomura S, Fujita T, Morita H, Asano Y, Onoue K, et al. Genetic basis of cardiomyopathy and the genotypes involved in prognosis and left ventricular reverse remodeling. Sci Rep 2018; 8: 1998.
• 28.   Haas J, Frese KS, Peil B, Kloos W, Keller A, Nietsch R, et al. Atlas of the clinical genetics of human dilated cardiomyopathy. Eur Heart J 2015; 36: 1123–1135a.
• 29.   Iles L, Pfluger H, Lefkovits L, Butler MJ, Kistler PM, Kaye DM, et al. Myocardial fibrosis predicts appropriate device therapy in patients with implantable cardioverter-defibrillators for primary prevention of sudden cardiac death. J Am Coll Cardiol 2011; 57: 821–828.
• 30.   Puntmann VO, Carr-White G, Jabbour A, Yu CY, Gebker R, Kelle S, et al. T1-mapping and outcome in nonischemic cardiomyopathy: All-cause mortality and heart failure. JACC Cardiovasc imaging 2016; 9: 40–50.