Cardiac resynchronization therapy for patients with cardiac sarcoidosis

Introduction

Sarcoidosis is a global disease, with a prevalence of ∼4.7–64 in 100 000.¹ Although the diagnosis of cardiac involvement is the most challenging in sarcoidosis patients, studies have suggested that symptomatic cardiac involvement occurs in perhaps 5% of the patients with pulmonary or systemic sarcoidosis.^1,2 Since sarcoidosis patients with cardiac involvement have a poorer prognosis than those without,³ its diagnosis and treatment are critically important. Cardiac manifestations include atrioventricular block, ventricular arrhythmias, left ventricular (LV) systolic dysfunction, congestive heart failure, and sudden death. The presence of a fragmented QRS or a bundle branch block pattern is also associated with cardiac involvement.¹ Among patients with cardiac sarcoidosis, those who have LV dysfunction, left bundle branch block (LBBB), and consequent heart failure can receive cardiac resynchronization therapy (CRT).⁴ However, to date, there are few reports on cardiac sarcoidosis requiring CRT, because of its rareness.⁴ We in the present study reported the clinical response to CRT in patients with cardiac sarcoidosis.

Methods

Patients

We retrospectively reviewed the databases of the CRT cases in the cardiovascular centres of Nagoya Daini Red Cross Hospital, Hiroshima University Hospital, and Hiroshima City Hospital. The present study was approved by the institutional review board of each hospital. Cardiac sarcoidosis was diagnosed on the basis of updated guidelines.^2,4 Briefly, cardiac sarcoidosis was diagnosed if a non-caseating granuloma was demonstrated by a myocardial biopsy (histologic diagnosis), or if non-invasive evidence of cardiac involvement was found in a patient with pathologically proven extracardiac sarcoidosis (clinical diagnosis). The patients with cardiac sarcoidosis were considered eligible if they were implanted with a CRT-pacemaker (CRT-P) or CRT-defibrillator (CRT-D) for the treatment of New York Heart Association (NYHA) Class II–IV heart failure despite adequate medical treatment, and if the devices were implanted from April 2004 to March 2012 based on Class I or IIa indications for CRT shown in the current guidelines⁵: drug-resistant heart failure patients who had an LV ejection fraction (LVEF) of ≤35% and LBBB with a QRS duration of ≥120 ms or non-LBBB with a QRS duration of ≥150 ms, or those with an LVEF of ≤35% who required a CRT upgrade from a pacemaker or implantable cardioverter defibrillator because of deterioration of their cardiac function resulting from permanent right ventricular pacing for bradycardia. The patients were excluded from the present study if their follow-up data were unavailable.

Device implantation and follow-up

In principle, a CRT-D was implanted if the patients had a history of sustained ventricular arrhythmias or syncope, or if the patients desired treatment with a CRT-D rather than a CRT-P. The device implantation was carried out with the use of standard transvenous techniques.⁶ In all the patients, an LV lead was inserted through the coronary sinus into either the lateral or posterolateral cardiac vein. In the patients undergoing a de novo CRT implantation, a right ventricular lead or defibrillation lead was usually placed in the right ventricular apex, and a right atrial lead was placed in the right atrial appendage. In patients with longstanding persistent atrial fibrillation, adequate pharmacotherapy or a radiofrequency atrioventricular junctional ablation was performed to guarantee maximal ventricular pacing.

The patients were seen in the clinical follow-up at 1–3-month intervals. A regular clinical assessment, electrocardiogram recording, and device interrogation to ensure biventricular pacing were performed at each clinical visit. The patients implanted with a CRT-D were also followed up with a remote monitoring system to check for any ventricular arrhythmias. Two-dimensional echocardiographic data were obtained at least twice, at baseline and at the 6-month follow-up, to assess any changes in the LV volumes.⁷

Endpoints

The primary endpoint was the composite outcome of cardiovascular death or hospitalization from worsening heart failure within 5 years after the CRT. The secondary endpoints were sustained ventricular arrhythmia events requiring any therapeutic intervention that occurred at any time before or within 5 years after the CRT, and a change in the LV end-systolic volume (LVESV) and LVEF before and 6 months after the CRT. The clinical composite score (CCS) developed by Packer⁸ was also included in the secondary endpoints. Briefly, patients were considered improved if their NYHA class improved, and they survived without any hospitalization from worsening heart failure at the 6-month follow-up. Patients were considered worsened if they had any hospitalization from decompensated heart failure or died from cardiovascular causes within 6 months after the CRT, or reported worsening of their NYHA class at the 6-month follow-up. Patients were considered unchanged if they neither improved nor worsened.

Statistical analysis

The continuous variables were summarized as the means ± SD or means ± SE, and categorical variables as proportions. A Kaplan–Meier analysis was used to assess the time required for the primary endpoint to occur. The differences in the categorical variables between groups were examined with the use of a Fisher's exact test. Comparisons of the paired or non-paired continuous variables were done by means of a Wilcoxon signed-rank test or Mann–Whitney U test, respectively. All statistical analyses were performed with the use of JMP software version 11.0 (SAS Institute, Cary, NC, USA). A P-value of <0.05 was considered significant.

Results

Patients

We reviewed a total of 202 patients who were implanted with a CRT-P or CRT-D device on the basis of the guidelines. Of those, we found 20 (9.9%) patients who were diagnosed with cardiac sarcoidosis (Figure 1). Two patients were excluded because of a loss to follow-up. We thus finally analysed 18 (8.9%) patients with cardiac sarcoidosis. Table 1 shows the clinical characteristics of the patients at the time of the CRT implantation. The vast majority of the patients were female [15 (83.3%)], were previously diagnosed with advanced atrioventricular block [14 (77.8%)], and had a clinical diagnosis of cardiac sarcoidosis [15 (83.3%)]. Among 15 patients with a clinical diagnosis, 10 patients were diagnosed with extracardiac sarcoidosis by a lung biopsy, 3 by a lymph node biopsy, 1 by a skin biopsy, and 1 by a conjunctiva biopsy. We found 9 (50%) and 13 (72.2%) patients who received corticosteroids and amiodarone, respectively. Ten (55.6%) patients received a CRT-D, and of those, 9 were implanted with it for the purpose of secondary prevention of arrhythmic mortality (Table 1). Only one patient had longstanding persistent atrial fibrillation with a slow ventricular response. All the patients studied had a biventricular pacing percentage of >90%.

Table 1

Open in new tab

Clinical characteristics of the patients at the CRT implantation

Patient no. .	Age (year) .	Female .	NYHA .	Histologic diagnosisa .	Upgrade or de novo .	QRS morphology .	QRS width (ms) .	LVEF (%) .	LVEDV (mL) .	LVESV (mL) .	Beta-blocker .	ACE-I or ARB .	Corticosteroid .	Amiodarone .
1	66	Yes	Class III	No	Upgrade from a PM to CRT-P	Paced LBBB	200	18	270	220	Yes	Yes	Yes	No
2	67	Yes	Class III	No	Upgrade from a PM to CRT-P	Paced LBBB	200	10	244	218	No	No	No	No
3	61	Yes	Class IV	No	Upgrade from a PM to CRT-D	Paced LBBB	212	23	282	216	Yes	Yes	No	Yes
4	69	Yes	Class III	No	Upgrade from a PM to CRT-P	Paced LBBB	180	30	208	144	No	Yes	Yes	Yes
5	67	Yes	Class III	No	Upgrade from a PM to CRT-D	Paced LBBB	200	25	253	190	Yes	Yes	No	Yes
6	71	Yes	Class III	Yes	Upgrade from a PM to CRT-P	Paced LBBB	206	31	331	227	Yes	Yes	Yes	No
7	74	Yes	Class III	No	De novo CRT-P	Intrinsic LBBB	194	28	210	151	No	Yes	No	Yes
8	63	Yes	Class II	No	Upgrade from a PM to CRT-P	Paced LBBB	138	33	144	96	Yes	Yes	No	Yes
9	54	Yes	Class III	No	Upgrade from an ICD to CRT-D	Paced LBBB	160	23	292	224	No	No	No	Yes
10	72	Yes	Class III	No	Upgrade from a PM to CRT-P	Paced LBBB	262	21	255	201	Yes	Yes	No	No
11	64	No	Class IV	No	De novo CRT-D	Intrinsic LBBB	132	11	180	160	Yes	Yes	No	Yes
12	64	Yes	Class III	No	Upgrade from a PM to CRT-D	Paced LBBB	200	35	167	108	Yes	Yes	Yes	Yes
13	56	Yes	Class III	No	Upgrade from a PM to CRT-D	Intrinsic RBBB	152	34	233	154	Yes	Yes	Yes	No
14	76	Yes	Class III	Yes	Upgrade from a PM to CRT-P	Paced LBBB	248	29	335	239	Yes	Yes	Yes	Yes
15	68	Yes	Class III	No	De novo CRT-D	Intrinsic LBBB	140	30	154	108	Yes	Yes	No	Yes
16	35	No	Class II	Yes	Upgrade from a PM to CRT-D	Paced LBBB	190	19	231	187	Yes	Yes	Yes	Yes
17	70	Yes	Class III	No	De novo CRT-D	Intrinsic RBBB	164	20	135	108	Yes	Yes	Yes	Yes
18	75	No	Class III	No	De novo CRT-D	Intrinsic LBBB	147	23	216	167	Yes	Yes	Yes	Yes
Mean ± SD or %	65 ± 10	83.3	N/A	16.7	N/A	N/A	185 ± 37	24 ± 6	230 ± 60	173 ± 47	77.8	88.9	50	72.2

Patient no. .	Age (year) .	Female .	NYHA .	Histologic diagnosisa .	Upgrade or de novo .	QRS morphology .	QRS width (ms) .	LVEF (%) .	LVEDV (mL) .	LVESV (mL) .	Beta-blocker .	ACE-I or ARB .	Corticosteroid .	Amiodarone .
1	66	Yes	Class III	No	Upgrade from a PM to CRT-P	Paced LBBB	200	18	270	220	Yes	Yes	Yes	No
2	67	Yes	Class III	No	Upgrade from a PM to CRT-P	Paced LBBB	200	10	244	218	No	No	No	No
3	61	Yes	Class IV	No	Upgrade from a PM to CRT-D	Paced LBBB	212	23	282	216	Yes	Yes	No	Yes
4	69	Yes	Class III	No	Upgrade from a PM to CRT-P	Paced LBBB	180	30	208	144	No	Yes	Yes	Yes
5	67	Yes	Class III	No	Upgrade from a PM to CRT-D	Paced LBBB	200	25	253	190	Yes	Yes	No	Yes
6	71	Yes	Class III	Yes	Upgrade from a PM to CRT-P	Paced LBBB	206	31	331	227	Yes	Yes	Yes	No
7	74	Yes	Class III	No	De novo CRT-P	Intrinsic LBBB	194	28	210	151	No	Yes	No	Yes
8	63	Yes	Class II	No	Upgrade from a PM to CRT-P	Paced LBBB	138	33	144	96	Yes	Yes	No	Yes
9	54	Yes	Class III	No	Upgrade from an ICD to CRT-D	Paced LBBB	160	23	292	224	No	No	No	Yes
10	72	Yes	Class III	No	Upgrade from a PM to CRT-P	Paced LBBB	262	21	255	201	Yes	Yes	No	No
11	64	No	Class IV	No	De novo CRT-D	Intrinsic LBBB	132	11	180	160	Yes	Yes	No	Yes
12	64	Yes	Class III	No	Upgrade from a PM to CRT-D	Paced LBBB	200	35	167	108	Yes	Yes	Yes	Yes
13	56	Yes	Class III	No	Upgrade from a PM to CRT-D	Intrinsic RBBB	152	34	233	154	Yes	Yes	Yes	No
14	76	Yes	Class III	Yes	Upgrade from a PM to CRT-P	Paced LBBB	248	29	335	239	Yes	Yes	Yes	Yes
15	68	Yes	Class III	No	De novo CRT-D	Intrinsic LBBB	140	30	154	108	Yes	Yes	No	Yes
16	35	No	Class II	Yes	Upgrade from a PM to CRT-D	Paced LBBB	190	19	231	187	Yes	Yes	Yes	Yes
17	70	Yes	Class III	No	De novo CRT-D	Intrinsic RBBB	164	20	135	108	Yes	Yes	Yes	Yes
18	75	No	Class III	No	De novo CRT-D	Intrinsic LBBB	147	23	216	167	Yes	Yes	Yes	Yes
Mean ± SD or %	65 ± 10	83.3	N/A	16.7	N/A	N/A	185 ± 37	24 ± 6	230 ± 60	173 ± 47	77.8	88.9	50	72.2

CRT, cardiac resynchronization therapy; NYHA, New York Heart Association; LBBB, left bundle branch block; LVEF, left ventricular ejection fraction; LVEDV, left ventricular end-diastolic volume; LVESV, left ventricle end-systolic volume; ACE-I, angiotensin converting enzyme inhibitor; ARB, angiotensin II receptor blocker; CRT-P, cardiac resynchronization pacemaker; CRT-D, cardiac resynchronization defibrillator; LBBB, left bundle branch block; RBBB, right bundle branch block.

^aCardiac sarcoidosis was diagnosed if a non-caseating granuloma was demonstrated by myocardial biopsy (histologic diagnosis), or if non-invasive evidence of cardiac involvement was found in a patient with pathologically proven extracardiac sarcoidosis (clinical diagnosis).

Table 1

Open in new tab

Clinical characteristics of the patients at the CRT implantation

Patient no. .	Age (year) .	Female .	NYHA .	Histologic diagnosisa .	Upgrade or de novo .	QRS morphology .	QRS width (ms) .	LVEF (%) .	LVEDV (mL) .	LVESV (mL) .	Beta-blocker .	ACE-I or ARB .	Corticosteroid .	Amiodarone .
1	66	Yes	Class III	No	Upgrade from a PM to CRT-P	Paced LBBB	200	18	270	220	Yes	Yes	Yes	No
2	67	Yes	Class III	No	Upgrade from a PM to CRT-P	Paced LBBB	200	10	244	218	No	No	No	No
3	61	Yes	Class IV	No	Upgrade from a PM to CRT-D	Paced LBBB	212	23	282	216	Yes	Yes	No	Yes
4	69	Yes	Class III	No	Upgrade from a PM to CRT-P	Paced LBBB	180	30	208	144	No	Yes	Yes	Yes
5	67	Yes	Class III	No	Upgrade from a PM to CRT-D	Paced LBBB	200	25	253	190	Yes	Yes	No	Yes
6	71	Yes	Class III	Yes	Upgrade from a PM to CRT-P	Paced LBBB	206	31	331	227	Yes	Yes	Yes	No
7	74	Yes	Class III	No	De novo CRT-P	Intrinsic LBBB	194	28	210	151	No	Yes	No	Yes
8	63	Yes	Class II	No	Upgrade from a PM to CRT-P	Paced LBBB	138	33	144	96	Yes	Yes	No	Yes
9	54	Yes	Class III	No	Upgrade from an ICD to CRT-D	Paced LBBB	160	23	292	224	No	No	No	Yes
10	72	Yes	Class III	No	Upgrade from a PM to CRT-P	Paced LBBB	262	21	255	201	Yes	Yes	No	No
11	64	No	Class IV	No	De novo CRT-D	Intrinsic LBBB	132	11	180	160	Yes	Yes	No	Yes
12	64	Yes	Class III	No	Upgrade from a PM to CRT-D	Paced LBBB	200	35	167	108	Yes	Yes	Yes	Yes
13	56	Yes	Class III	No	Upgrade from a PM to CRT-D	Intrinsic RBBB	152	34	233	154	Yes	Yes	Yes	No
14	76	Yes	Class III	Yes	Upgrade from a PM to CRT-P	Paced LBBB	248	29	335	239	Yes	Yes	Yes	Yes
15	68	Yes	Class III	No	De novo CRT-D	Intrinsic LBBB	140	30	154	108	Yes	Yes	No	Yes
16	35	No	Class II	Yes	Upgrade from a PM to CRT-D	Paced LBBB	190	19	231	187	Yes	Yes	Yes	Yes
17	70	Yes	Class III	No	De novo CRT-D	Intrinsic RBBB	164	20	135	108	Yes	Yes	Yes	Yes
18	75	No	Class III	No	De novo CRT-D	Intrinsic LBBB	147	23	216	167	Yes	Yes	Yes	Yes
Mean ± SD or %	65 ± 10	83.3	N/A	16.7	N/A	N/A	185 ± 37	24 ± 6	230 ± 60	173 ± 47	77.8	88.9	50	72.2

Patient no. .	Age (year) .	Female .	NYHA .	Histologic diagnosisa .	Upgrade or de novo .	QRS morphology .	QRS width (ms) .	LVEF (%) .	LVEDV (mL) .	LVESV (mL) .	Beta-blocker .	ACE-I or ARB .	Corticosteroid .	Amiodarone .
1	66	Yes	Class III	No	Upgrade from a PM to CRT-P	Paced LBBB	200	18	270	220	Yes	Yes	Yes	No
2	67	Yes	Class III	No	Upgrade from a PM to CRT-P	Paced LBBB	200	10	244	218	No	No	No	No
3	61	Yes	Class IV	No	Upgrade from a PM to CRT-D	Paced LBBB	212	23	282	216	Yes	Yes	No	Yes
4	69	Yes	Class III	No	Upgrade from a PM to CRT-P	Paced LBBB	180	30	208	144	No	Yes	Yes	Yes
5	67	Yes	Class III	No	Upgrade from a PM to CRT-D	Paced LBBB	200	25	253	190	Yes	Yes	No	Yes
6	71	Yes	Class III	Yes	Upgrade from a PM to CRT-P	Paced LBBB	206	31	331	227	Yes	Yes	Yes	No
7	74	Yes	Class III	No	De novo CRT-P	Intrinsic LBBB	194	28	210	151	No	Yes	No	Yes
8	63	Yes	Class II	No	Upgrade from a PM to CRT-P	Paced LBBB	138	33	144	96	Yes	Yes	No	Yes
9	54	Yes	Class III	No	Upgrade from an ICD to CRT-D	Paced LBBB	160	23	292	224	No	No	No	Yes
10	72	Yes	Class III	No	Upgrade from a PM to CRT-P	Paced LBBB	262	21	255	201	Yes	Yes	No	No
11	64	No	Class IV	No	De novo CRT-D	Intrinsic LBBB	132	11	180	160	Yes	Yes	No	Yes
12	64	Yes	Class III	No	Upgrade from a PM to CRT-D	Paced LBBB	200	35	167	108	Yes	Yes	Yes	Yes
13	56	Yes	Class III	No	Upgrade from a PM to CRT-D	Intrinsic RBBB	152	34	233	154	Yes	Yes	Yes	No
14	76	Yes	Class III	Yes	Upgrade from a PM to CRT-P	Paced LBBB	248	29	335	239	Yes	Yes	Yes	Yes
15	68	Yes	Class III	No	De novo CRT-D	Intrinsic LBBB	140	30	154	108	Yes	Yes	No	Yes
16	35	No	Class II	Yes	Upgrade from a PM to CRT-D	Paced LBBB	190	19	231	187	Yes	Yes	Yes	Yes
17	70	Yes	Class III	No	De novo CRT-D	Intrinsic RBBB	164	20	135	108	Yes	Yes	Yes	Yes
18	75	No	Class III	No	De novo CRT-D	Intrinsic LBBB	147	23	216	167	Yes	Yes	Yes	Yes
Mean ± SD or %	65 ± 10	83.3	N/A	16.7	N/A	N/A	185 ± 37	24 ± 6	230 ± 60	173 ± 47	77.8	88.9	50	72.2

CRT, cardiac resynchronization therapy; NYHA, New York Heart Association; LBBB, left bundle branch block; LVEF, left ventricular ejection fraction; LVEDV, left ventricular end-diastolic volume; LVESV, left ventricle end-systolic volume; ACE-I, angiotensin converting enzyme inhibitor; ARB, angiotensin II receptor blocker; CRT-P, cardiac resynchronization pacemaker; CRT-D, cardiac resynchronization defibrillator; LBBB, left bundle branch block; RBBB, right bundle branch block.

^aCardiac sarcoidosis was diagnosed if a non-caseating granuloma was demonstrated by myocardial biopsy (histologic diagnosis), or if non-invasive evidence of cardiac involvement was found in a patient with pathologically proven extracardiac sarcoidosis (clinical diagnosis).

Figure 1

Representative images prior to the CRT in patient No. 18. An electrocardiogram showing sinus rhythm with LBBB and a QRS interval of 147 ms (A), echocardiogram exhibiting a severely decreased LV contraction with an ejection fraction of 26% (B), magnetic resonance imaging with a late gadolinium enhancement in the anterior and inferior LV regions (C), and F-18-fluorodeoxyglucos positron emission tomography image that shows diffuse inhomogeneous uptakes in the LV wall (D).

Open in new tabDownload slide

Outcomes

The echocardiographic data assessed 6 months after the CRT and details of the clinical endpoints in each patient are shown in Table 2. We found 1 (5.6%) cardiovascular death, a sudden death, during the follow-up period (mean ± SD, 4.7 ± 3.0 years). A composite outcome occurred in 7 (38.9%) patients during the follow-up period (Figure 2).

Table 2

Open in new tab

Echocardiographic findings 6 months after the CRT and clinical endpoints

Patient no. .	LVEDV (mL) .	LVESV (mL) .	LVEF (%) .	% Change of LVESVa (%) .	% Change of LVEFa (%) .	Primary endpointsb .	Ventricular arrhythmia .	CCS .
1	348	295	15	34.0	−15.6	Death	No	Unchanged
2	231	187	19	−14.3	92.0	Hospitalization	No	Improved
3	326	264	19	22.1	−17.3	No	Before and after CRT	Improved
4	239	209	13	44.8	−57.2	No	After CRT	Unchanged
5	281	224	20	17.7	−18.6	Hospitalization	Before and after CRT	Unchanged
6	133	76	43	−66.7	39.4	No	No	Improved
7	216	163	25	8.0	−12.4	Hospitalization	After CRT	Unchanged
8	148	94	36	−1.7	9.8	No	No	Improved
9	264	194	26	−13.4	15.0	No	Before and after CRT	Improved
10	201	97	52	−51.6	145.8	No	No	Improved
11	239	207	13	29.5	21.8	Hospitalization	Before CRT	Worsened
12	141	83	41	−23.1	17.8	No	Before CRT	Improved
13	224	167	26	8.2	−25.0	No	No	Worsened
14	264	216	18	−9.6	−37.6	No	Before CRT	Improved
15	124	74	40	−31.3	33.5	No	Before CRT	Improved
16	264	209	21	11.5	10.2	No	Before CRT	Improved
17	167	113	32	4.5	61.3	Hospitalization	Before CRT	Worsened
18	239	173	28	3.7	20.0	Hospitalization	Before CRT	Worsened
Mean ± SD	225 ± 64	169 ± 66	27 ± 11	−1.5 ± 28.7	15.7 ± 48.3	N/A	N/A	N/A

Patient no. .	LVEDV (mL) .	LVESV (mL) .	LVEF (%) .	% Change of LVESVa (%) .	% Change of LVEFa (%) .	Primary endpointsb .	Ventricular arrhythmia .	CCS .
1	348	295	15	34.0	−15.6	Death	No	Unchanged
2	231	187	19	−14.3	92.0	Hospitalization	No	Improved
3	326	264	19	22.1	−17.3	No	Before and after CRT	Improved
4	239	209	13	44.8	−57.2	No	After CRT	Unchanged
5	281	224	20	17.7	−18.6	Hospitalization	Before and after CRT	Unchanged
6	133	76	43	−66.7	39.4	No	No	Improved
7	216	163	25	8.0	−12.4	Hospitalization	After CRT	Unchanged
8	148	94	36	−1.7	9.8	No	No	Improved
9	264	194	26	−13.4	15.0	No	Before and after CRT	Improved
10	201	97	52	−51.6	145.8	No	No	Improved
11	239	207	13	29.5	21.8	Hospitalization	Before CRT	Worsened
12	141	83	41	−23.1	17.8	No	Before CRT	Improved
13	224	167	26	8.2	−25.0	No	No	Worsened
14	264	216	18	−9.6	−37.6	No	Before CRT	Improved
15	124	74	40	−31.3	33.5	No	Before CRT	Improved
16	264	209	21	11.5	10.2	No	Before CRT	Improved
17	167	113	32	4.5	61.3	Hospitalization	Before CRT	Worsened
18	239	173	28	3.7	20.0	Hospitalization	Before CRT	Worsened
Mean ± SD	225 ± 64	169 ± 66	27 ± 11	−1.5 ± 28.7	15.7 ± 48.3	N/A	N/A	N/A

CCS, clinical composite score. The other abbreviations are as in Table 1.

^aThe % change in the LVESV or LVEF was calculated with the formula, ([LVESV or LVEF after the CRT] − [LVESV or LVEF before CRT]) × 100/LVESV or LVEF before the CRT.

^bCardiovascular death or hospitalization from worsening heart failure within 5 years after the CRT.

Table 2

Open in new tab

Echocardiographic findings 6 months after the CRT and clinical endpoints

Patient no. .	LVEDV (mL) .	LVESV (mL) .	LVEF (%) .	% Change of LVESVa (%) .	% Change of LVEFa (%) .	Primary endpointsb .	Ventricular arrhythmia .	CCS .
1	348	295	15	34.0	−15.6	Death	No	Unchanged
2	231	187	19	−14.3	92.0	Hospitalization	No	Improved
3	326	264	19	22.1	−17.3	No	Before and after CRT	Improved
4	239	209	13	44.8	−57.2	No	After CRT	Unchanged
5	281	224	20	17.7	−18.6	Hospitalization	Before and after CRT	Unchanged
6	133	76	43	−66.7	39.4	No	No	Improved
7	216	163	25	8.0	−12.4	Hospitalization	After CRT	Unchanged
8	148	94	36	−1.7	9.8	No	No	Improved
9	264	194	26	−13.4	15.0	No	Before and after CRT	Improved
10	201	97	52	−51.6	145.8	No	No	Improved
11	239	207	13	29.5	21.8	Hospitalization	Before CRT	Worsened
12	141	83	41	−23.1	17.8	No	Before CRT	Improved
13	224	167	26	8.2	−25.0	No	No	Worsened
14	264	216	18	−9.6	−37.6	No	Before CRT	Improved
15	124	74	40	−31.3	33.5	No	Before CRT	Improved
16	264	209	21	11.5	10.2	No	Before CRT	Improved
17	167	113	32	4.5	61.3	Hospitalization	Before CRT	Worsened
18	239	173	28	3.7	20.0	Hospitalization	Before CRT	Worsened
Mean ± SD	225 ± 64	169 ± 66	27 ± 11	−1.5 ± 28.7	15.7 ± 48.3	N/A	N/A	N/A

Patient no. .	LVEDV (mL) .	LVESV (mL) .	LVEF (%) .	% Change of LVESVa (%) .	% Change of LVEFa (%) .	Primary endpointsb .	Ventricular arrhythmia .	CCS .
1	348	295	15	34.0	−15.6	Death	No	Unchanged
2	231	187	19	−14.3	92.0	Hospitalization	No	Improved
3	326	264	19	22.1	−17.3	No	Before and after CRT	Improved
4	239	209	13	44.8	−57.2	No	After CRT	Unchanged
5	281	224	20	17.7	−18.6	Hospitalization	Before and after CRT	Unchanged
6	133	76	43	−66.7	39.4	No	No	Improved
7	216	163	25	8.0	−12.4	Hospitalization	After CRT	Unchanged
8	148	94	36	−1.7	9.8	No	No	Improved
9	264	194	26	−13.4	15.0	No	Before and after CRT	Improved
10	201	97	52	−51.6	145.8	No	No	Improved
11	239	207	13	29.5	21.8	Hospitalization	Before CRT	Worsened
12	141	83	41	−23.1	17.8	No	Before CRT	Improved
13	224	167	26	8.2	−25.0	No	No	Worsened
14	264	216	18	−9.6	−37.6	No	Before CRT	Improved
15	124	74	40	−31.3	33.5	No	Before CRT	Improved
16	264	209	21	11.5	10.2	No	Before CRT	Improved
17	167	113	32	4.5	61.3	Hospitalization	Before CRT	Worsened
18	239	173	28	3.7	20.0	Hospitalization	Before CRT	Worsened
Mean ± SD	225 ± 64	169 ± 66	27 ± 11	−1.5 ± 28.7	15.7 ± 48.3	N/A	N/A	N/A

CCS, clinical composite score. The other abbreviations are as in Table 1.

^aThe % change in the LVESV or LVEF was calculated with the formula, ([LVESV or LVEF after the CRT] − [LVESV or LVEF before CRT]) × 100/LVESV or LVEF before the CRT.

^bCardiovascular death or hospitalization from worsening heart failure within 5 years after the CRT.

Figure 2

The time to a composite outcome of cardiovascular death or hospitalization from worsening heart failure within 5 years after the CRT.

Open in new tabDownload slide

Twelve (66.7%) patients had a history of sustained ventricular arrhythmias or those occurring after the CRT (Table 2). Among the 5 patients with ventricular arrhythmias during the follow-up, 3 had at least 1 episode of ventricular tachycardia requiring a defibrillator shock. Of the 3 patients, 1 experienced an electrical storm and therefore needed intensive care. The remaining 2 patients received a CRT-P rather than a CRT-D, and their ventricular tachycardias were successful treated with external electrical cardioversion or intravenous antiarrhythmic agents. No statistically significant change was found in both the LVESV and LVEF between the baseline and 6 months after the CRT (Table 2 and Figure 3). An improved CCS was observed in 10 (55.6%) patients (Table 2). Only 1 out of 5 patients with de novo CRT had an improved CCS, while it was seen in 9 out of 13 patients needing upgrade (P = 0.12).

Figure 3

Change in the LVESV and LVEF between the baseline and 6 months after the CRT. LVESV, left ventricular end-systolic volume; LVEF, left ventricular ejection fraction. The mean and standard error are presented.

Open in new tabDownload slide

We additionally compared the patients with ventricular arrhythmias occurring even after the CRT (N = 5) to those without (N = 13). We found weak tendencies for the former group to have been prescribed amiodarone more frequently (100% vs. 61.5%; P = 0.25) and to have less commonly received corticosteroids (20% vs. 61.5%; P = 0.29). The former group was more likely to have unfavourable echocardiographic changes 6 months after the CRT (Table 3).

Discussion

Unique response to cardiac resynchronization therapy in patients with cardiac sarcoidosis

To the best of our knowledge, we for the first time reported the outcomes after CRT in patients with cardiac sarcoidosis. The hard endpoint in the present study was identical to that of the other large-scaled trials.⁹ Furthermore, we did not find a miserable mid-term survival. Also, the proportion of the patients with an improved CCS was not unacceptably low in the present study.¹⁰ Given the progressive nature of cardiac sarcoidosis,³ these findings may seem illogical. The following are the potential explanations for the finding. (i) Ventricular arrhythmias are known to be one of the major causes of death in patients with cardiac sarcoidosis.^3,4 In the present study, the majority of the patients with cardiac sarcoidosis were treated with amiodarone or CRT-D, or both, possibly leading to an improvement in the prognosis. (ii) Although not reaching statistical significance, sarcoidosis patients needing upgrade were more likely than those with de novo CRT to have a favourable CCS. This suggests that the sarcoidosis patients in whom a conduction disturbance preceded the LV dysfunction might have had a better prognosis, and which possibly drove up the mid-term outcome in sarcoidosis patients because the majority of them received a CRT upgrade. (iii) Corticosteroid therapy is the mainstay of the treatment for cardiac sarcoidosis.^3,4 Half of the patients included in our study received corticosteroids, and further a mild trend towards a lower frequency of recurrent ventricular arrhythmias was noted in the patients receiving corticosteroids. This might have contributed to the reduction in the mortality. (iv) A subanalysis of MADIT-CRT¹¹ showed that women had a more favourable outcome after CRT than men. A female preponderance in prevalence of sarcoidosis is well known,¹ and actually, most of the patients with cardiac sarcoidosis in our study were female.

Yu et al.¹² advanced an argument that a greater reduction in the LV volume consistently translates into a better clinical outcome after CRT, and this concept is widely accepted.¹³ We, however, found that the LVESV in the patients with cardiac sarcoidosis remained unchanged even 6 months after the CRT, while they had acceptable hard outcomes. This discrepancy was hard to explain; however, it perhaps might be attributed to the ‘progressive nature of cardiac sarcoidosis’.³ Namely, a continuous disease progression of cardiac sarcoidosis possibly could have cancelled out the reverse remodelling brought about by the CRT. In agreement with our previous work,¹⁴ the majority of our patients with cardiac sarcoidosis were implanted with a pacemaker for the treatment of advanced atrioventricular block before the upgrade to CRT, probably meaning that their myocardium on the ventricular septum was seriously invaded.^4,14 This particularity seen in our series could also explain their unique response to the CRT because bi-ventricular pacing generally corrects the late activation of the lateral regions of the LV chamber vs. activation of the interventricular septum.¹³ Chakir et al.¹⁵ showed in their dog model that myocardial apoptosis was suppressed by CRT before the reduction in the LV volume was seen. Considering again the ‘progressive nature of cardiac sarcoidosis’, it might be possible that this potential action of CRT slowed the disease progression and consequently resulted in a tolerable outcome in the patients with cardiac sarcoidosis even when they did not have any reverse LV remodelling.

Ventricular arrhythmias

It is known that ventricular arrhythmias are common in patients with cardiac sarcoidosis.^3,4,5 This was confirmed by our finding that sustained ventricular arrhythmias occurred in two-thirds of the patients with cardiac sarcoidosis who received CRT. Considering again that ventricular arrhythmias are a major killer of patients with cardiac sarcoidosis,^3,4 this finding indicates that patients with CRT may require extra attention to the occurrence of ventricular arrhythmias if they have cardiac sarcoidosis. The sarcoidosis patients who developed ventricular arrhythmias even after the CRT were less likely to have LV reverse remodelling in our series. Therefore, a careful follow-up may also be needed in those patients.

Clinical implications

The report on CRT for cardiac sarcoidosis patients may itself be variable because of their rareness. Their features provided important information regarding the management after CRT in patients with cardiac sarcoidosis. In particular, it may worthwhile to remember that (i) a considerable number of candidates for CRT may require upgrade rather than de novo CRT, (ii) reverse remodelling may not be necessarily observed even though they enjoy a favourable outcome, and (iii) controlling ventricular arrhythmias may be critically important to improve the outcome after the CRT implantation in patients with cardiac sarcoidosis.

Limitations

We cannot deny the possibility that we unintentionally included sarcoidosis patients who had a better prognosis because cardiac sarcoidosis was often diagnosed by autopsy after sudden cardiac death.³ A rareness of cardiac sarcoidosis prevented us from including a sufficient number of study subjects, leading to a limited statistical power. A numerical predominance of upgraded CRT over de novo CRT in our series might raise concern about how much the cardiac involvement itself could explain the clinical features of patients with cardiac sarcoidosis requiring CRT because chronic right ventricular pacing is known to contribute to the development of heart failure. Finally, an impact on the clinical practice of the present study is also limited because of its retrospective nature and ethnic homogeneity of the study subjects.

Conflict of interest: none declared.

References

Author notes