Adaptive Servo-Ventilation Therapy for Patients With Chronic Heart Failure in a Confirmatory, Multicenter, Randomized, Controlled Study

Abstract

Background: Adaptive servo-ventilation (ASV) therapy is expected to be novel nonpharmacotherapy with hemodynamic effects on patients with chronic heart failure (CHF), but sufficient evidence has not been obtained.

Methods and Results: A 24-week, open-label, randomized, controlled study was performed to confirm the cardiac function-improving effect of ASV therapy on CHF patients. At 39 institutions, 213 outpatients with CHF, whose left ventricular ejection fraction (LVEF) was <40% and who had mild to severe symptoms [New York Heart Association (NYHA) class: ≥II], were enrolled. After excluding 8 patients, 102 and 103 underwent ASV plus guideline-directed medical therapy (GDMT) [ASV group] and GDMT only [control group], respectively. The primary endpoint was LVEF, and the secondary endpoints were HF deterioration, B-type natriuretic peptide (BNP), and clinical composite response (CCR: NYHA class+HF deterioration). LVEF and BNP improved significantly at completion against the baseline values in the 2 groups. However, no significant difference was found between these groups. HF deterioration tended to be suppressed. The ASV group showed a significant improvement in CCR corroborated by significant improvements in NYHA class and ADL against the control group.

Conclusions: Under the present study’s conditions, ASV therapy was not superior to GDMT in the cardiac function-improving effect but showed a clinical status-improving effect, thus indicating a given level of clinical benefit. (Circ J 2015; 79: 981–990)

Subsequent to the implementation of guideline-directed medical therapy (GDMT), mainly consisting of angiotensin-converting enzyme (ACE) inhibitors, angiotensin II receptor blockers (ARBs), and β-blockers, the prognosis of patients with mild to moderate chronic heart failure (CHF) has improved substantially. Nevertheless, the prognosis of patients with severe CHF still remains poor, and the limitations of pharmacology are indicated. In recent years, therapeutic attempts have been made not only for the extension of CHF patients’ life expectancy (a quantitative approach), but also for an improvement in quality of life (QOL) that declines with their reduced exercise tolerance. Hence, novel nonpharmacotherapy that can be added to GDMT is required.

Noninvasive positive pressure ventilation (NPPV) improves pulmonary congestion by reopening collapsed alveoli, by preventing peripheral airway occlusion, by increasing lung volume, by improving oxygenation and lung compliance, and by reducing cardiac preload through reduced venous return.^1–3 In Japan, furthermore, NPPV is recommended (application standard class: I; evidence level: A) for patients with acute heart failure (AHF) who do not respond to oxygen therapy;⁴ the usefulness of NPPV in the respiratory management of AHF patients is thus demonstrated. Based on its cardiac output-increasing effect,^1,5 the therapeutic application of NPPV therapy to improve the hemodynamics of not only inpatients with AHF but also of outpatients with CHF was conceived. Nevertheless, NPPV therapy at home for CHF patients was difficult to conduct because the conventional NPPV devices were inconvenient (eg, large size and operationally cumbersome) and had poor tolerability; these factors hampered the conception of use outside the hospital.

Adaptive servo-ventilation (ASV), which automatically selects the appropriate support pressure based on the patient’s ventilation volume, is estimated to potentially have a hemodynamic-improving effect on CHF patients because of its nature of being a class of NPPV and, therefore, in recent years it has attracted the attention of Japanese cardiologists. A novel ventilator for ASV has realized ASV-therapy at home because of its patient-friendly properties (compact size, simple operability, and synchronization with the respiration patterns of the patient). A number of single-center, nonrandomized studies have suggested that ASV therapy, regardless of the severity of sleep-disordered breathing (SDB), can inhibit sympathetic nerve overactivity,^6–11 improve left ventricle ejection fraction (LVEF), B-type natriuretic peptide (BNP) levels, and cardiovascular events, as well as clinical symptoms and exercise tolerance, and induce LV reverse remodeling.^11–17 Patients with LV reverse remodeling have a very favorable prognosis as compared with patients who do not.¹⁸

The present study was conducted as a multicenter, randomized, controlled study in order to examine the superiority of the ASV group to the control group regarding the cardiac function-improving and LV remodeling-inducing effects of ASV therapy that SAVIOR-R¹⁴ had suggested.

Methods

Subjects and Study Design

The target number of patients for the present study was 200 in total as described in the protocol article of the present study.¹⁹ At 39 medical institutions in Japan (Appendix 1), 213 patients with CHF (age: ≥20 years; GDMT; capable of using the ASV ventilator at home; New York Heart Association (NYHA) class: ≥II; LVEF: <40%; mean age: 63.3 years; 165 males) were enrolled. The exclusion criteria are described in the protocol article.¹⁹

The Confirmatory, multicenter, randomized, controlled Study of Adaptive servo-VentIlation In patients with chrOnic heaRt failure (SAVIOR-C) is a prospective, multicenter, collaborative, open-label, blinded-endpoint, parallel-group, randomized, controlled study. According to the central registration system using the minimization method, eligible patients were randomly assigned at a 1-to-1 ratio to the ASV group (GDMT+ASV) or the control group (GDMT only) based on the assignment adjustment factors (age: <65 years/≥65 years; sex: male/female; and NYHA class: II/III). ASV therapy used a ventilator, AutoSet^TM CS (Teijin Pharma Ltd, Tokyo, Japan). The protocol of SAVIOR-C was approved by the ethics review board at each participating medical institution, and SAVIOR-C was registered in the University Hospital Medical Information Network (UMIN000006549) and conducted in accordance with the Declaration of Helsinki and with the Ethical Guideline for Clinical Research announced by the Ministry of Health, Labor and Welfare. Prior to enrollment, all patients gave written informed consent after receiving verbal and written explanations of the present clinical study and ASV therapy.

Definitions and Endpoints

The definitions of and treatments for CHF followed the Guidelines for Treatment of Chronic Heart Failure (JCS 2010).²⁰ The primary endpoint for efficacy was LVEF measured according to the biplane Simpson’s method,²¹ and the Central Adjudication Committee (Appendix 2) determined, in a blinded manner, the acceptability of LVEF measurements obtained at each participating medical institution in accordance with the specified procedures for echocardiography.

The secondary endpoints for efficacy were: (1) cardiac events (death from heart failure, sudden death, admission to the hospital because of disease progression, treatment modification for HF, and long-term (≥12 weeks) deterioration of symptoms) that the Central Adjudication Committee finally determined in a blinded manner; (2) plasma BNP levels measured at each participating medical institution; and a clinical composite response (CCR: categorized as worsened, unchanged, and improved) consisting of NYHA classes and clinical events.²² The endpoint for safety was an adverse event. The endpoints, which were based on the case report form, were as follows with respect to the ASV group and the control group: patient background, QOL, activities of daily living (ADL), vital signs, physical findings, echocardiograms, chest X-ray, hematology results, sleep study outcomes (eg, apnea-hypopnea index (AHI) at baseline only), and ventilator usage. At each participating institution, AHI was determined by polysomnography or a portable monitoring device type 2 or 3 as per the published methodology²³ and in accordance with the published definition.²⁴ Electronic media were used to collect (1) echocardiograms, ventilator usage, and air flow of the sleep study in the ASV group, as well as (2) echocardiograms and air flow of the sleep study in the control group.

Daytime sleepiness was self-assessed according to the Epworth sleepiness scale.²⁵ The attending physician assessed the ADL based on a 21-item questionnaire²⁶ entered by the patient. Furthermore, the patient assessed QOL by him/herself via a 21-item questionnaire to evaluate the physical and mental dimensions.²⁷

Statistical Analysis

LVEF, the primary endpoint, was analyzed using analysis of covariance. The cumulative incidences of cardiac events were analyzed using the Kaplan-Meier method and log-rank test. Intra- and intergroup differences for continuous variables were evaluated using paired t-test and unpaired t-test, respectively. Intra- and intergroup differences for BNP and categorical variables were analyzed using 1- and 2-sample Wilcoxon tests, respectively. Binary variables were analyzed using the Fisher’s exact probability test. Missing values were imputed using the Last-Observation-Carried-Forward approach. All analyses were conducted with the SAS Analysis System, version 9.2 (SAS Institute Inc, Cary, NC, USA), and a 2-tailed P value <0.05 was considered statistically significant.

Results

Patients

Patients were enrolled in the present study between December 2011 and May 2013 and underwent the predetermined examinations and assessments at baseline (8 weeks before enrollment to day of enrollment), as well as at weeks 12 and 24 of the study.

A total of 213 patients with CHF were enrolled and then randomly assigned to either the ASV or control group. After excluding 8 patients, 102 patients in the ASV group and 103 patients in the control group received their allocated interventions (Figure 1). Patients were followed up for 24 weeks after the baseline.

Figure 1.

Flow diagram of patient eligibility and allocation in a confirmatory, multicenter, randomized, controlled study of adaptive servo-ventilation (ASV) therapy for patients with chronic heart failure.

Patient characteristics at baseline are shown in Table 1. The mean ages of the patients in the ASV group and the control group were 63.3±13.0 years and 63.3±13.9 years, respectively. Male patients accounted for 80.4% and 80.6% in the ASV group and the control group, respectively. The proportions of ischemic heart disease among the underlying heart diseases were 40.2% and 35.0% in the ASV group and the control group, respectively. The proportions of patients with NYHA classes II and III/IV heart failure were 55.9% and 44.1%, and 61.2% and 38.8% in the ASV group and the control group, respectively. A significant difference (P=0.013) was found between the groups regarding LVEF at baseline: 29.7±7.7% in the ASV group and 27.2±7.1% in the control group. However, no statistically significant difference was found between the groups for any of the other variables examined.

Table 1. Baseline Characteristics of Patients Who Received the Allocated Interventions

	ASV group	Control group	P value†
No. of patients	102	103	–
Sex, M/F, n (%)	82/20 (80.4/19.6)	83/20 (80.6./19.4)	1.000
Age (years), mean±SD	63.3±13.0	63.3±13.9	0.986
Body mass index, mean±SD	23.2±3.8	23.3±5.3	0.910
Underlying heart disease	38/41/9/3/11	48/36/9/0/10	0.386
DCM/IHD/HHD/VHD/others, n (%)	(37.3/40.2/8.8/2.9/10.8)	(46.6/35.0/8.7/0/9.7)
Duration of CHF, <1 year/≥1 year, n (%)	49/48 (50.5/49.5)	52/51 (50.5/49.5)	1.000
Admission to hospital in the past year, n (%)	71 (69.6)	66 (64.1)	0.459
Comorbidities, n (%)
Diabetes mellitus	37 (36.3)	25 (24.3)	0.069
Hypertension	40 (39.2)	44 (42.7)	0.671
Chronic kidney disease	40 (39.2)	38 (36.9)	0.775
Chronic obstructive pulmonary disease	4 (3.9)	1 (1.0)	0.212
Atrial fibrillation	27 (57.4)	22 (51.2)	0.672
Medications, n (%)
Diuretics	79 (77.5)	85 (82.5)	0.387
ACE inhibitors	45 (44.1)	46 (44.7)	1.000
ARBs	38 (37.3)	33 (32.0)	0.465
β-blockers	94 (92.2)	96 (93.2)	1.000
Aldosterone antagonists	64 (62.8)	66 (64.1)	0.885
Amiodarone	21 (20.6)	19 (18.5)	0.727
Nonpharmacological therapies, n (%)
Hemodialysis	3 (7.0)	2 (5.3)	1.000
Pacemaker	4 (3.9)	5 (4.9)	1.000
Implantable cardioverter defibrillator	5 (4.9)	3 (2.9)	0.498
Cardiac resynchronization therapy	17 (16.7)	15 (14.6)	0.705
NYHA classes, II/III/IV, n (%)	54/44/1 (55.9/43.1/1.0)	63/38/2 (61.2/36.9/1.9)	0.488
LVEF, %, mean±SD	29.7±7.7	27.2±7.1	0.013
CTR, %, mean±SD	55.0±6.6	55.6±6.9	0.552
BNP, pg/ml, median [25–75 percentile range]	238 [102–502]	299 [108–640]	0.389
AHI, events/h, mean±SD	29.1±24.4	25.6±18.4	0.262

^†Intergroup differences were calculated using Student’s t-test, two-sample Wilcoxon test, or Fisher’s exact probability test. ACE, angiotensin-converting enzyme; AHI, apnea-hypopnea index; ARBs, angiotensin receptor blockers; ASV, adaptive servo-ventilation; BNP, B-type natriuretic peptide; CHF, chronic heart failure; CTR, cardiothoracic ratio; DCM, dilated cardiomyopathy; HHD, hypertensive heart disease; IHD, ischemic heart disease; LVEF, left ventricular ejection fraction; NYHA, New York Heart Association; VHD, valvular heart disease.

Echocardiography

Baseline-completion changes in LVEF in the ASV and control groups are shown in Figure 2. LVEF in the ASV group and the control group increased significantly (P<0.001 each) from 29.6±7.5% and 27.1±6.9% to 35.2±11.0% and 33.5±10.7%, respectively. Namely, LVEF improved by 5.6±9.0% and 6.4±11.7% in the ASV group and the control group, respectively. Analysis of covariance revealed no significant difference between the groups (P=0.659), which did not demonstrate the additive effect of ASV therapy on LVEF.

Figure 2.

Baseline-completion changes in left ventricular ejection fraction (LVEF) in the adaptive servo-ventilation (A) and control (B) groups. Black line: individual data; red and blue lines: mean±SD. Intragroup P values (paired t-test); intergroup P values (analysis of covariance).

Echocardiographic variables other than LVEF at baseline and completion are shown in Table 2. LV end-diastolic volume (LVEDV) at completion decreased in the ASV group and the control group, with a tendency (P=0.053) and a significant difference (P<0.001), respectively; hence, no statistically significant difference (P=0.087) was found between the groups with respect to LVEDV change. LV end-systolic volume (LVESV) at completion decreased significantly in the ASV group (P=0.002) and the control group (P<0.001), without an intergroup significance (P=0.174). At completion, stroke volume increased significantly (P<0.001) and variables for left ventricular diastolic function ⎯ E/A, deceleration time, and E/e’ also improved significantly (P=0.004, 0.007, and 0.031, respectively) only in the ASV group; however, no intergroup significant difference was found with respect to these echocardiographic variables.

Table 2. Clinical, Echocardiographic, and Radiographic Variables at Baseline and Completion of the Study

Items	Group	n	Baseline	Completion	Intragroup P value†	Difference or improvement	Intergroup P value‡
Vital signs
SBP, mmHg	ASV	95	109.09±17.07	117.03±19.13	<0.001	7.94±17.34	0.647
Mean±SD	Control	87	108.61±18.08	115.32±19.33	0.001	6.71±18.61
DBP, mmHg	ASV	95	65.41±10.99	67.61±11.02	0.112	2.20±13.36	0.131
Mean±SD	Control	87	65.49±13.36	70.92±13.17	0.001	5.43±15.29
Pulse rate, beats/min	ASV	95	70.06±11.95	70.07±11.79	0.994	0.01±13.52	0.827
Mean±SD	Control	86	71.03±13.05	71.51±11.95	0.772	0.48±15.21
Symptoms, ADL, and QOL
NYHA class, I/II/III/IV	ASV	93	0/53/39/1	21/51/21/0	<0.001	45 (48.4%)	0.009
n	Control	85	0/54/30/1	14/41/29/1	0.044	24 (28.2%)
Specific activity scale, Mets	ASV	91	4.14±1.46	5.34±1.72	<0.001	1.20±1.52	0.042
Mean±SD	Control	84	4.13±1.35	4.82±1.77	<0.001	0.69±1.75
MLHFQ	ASV	87	44.14±26.22	25.92±21.37	<0.001	−18.22±24.65	0.612
Mean±SD	Control	80	40.88±24.11	24.71±22.72	<0.001	−16.16±27.67
Echocardiogram
LVEDV, ml	ASV	66	176.79±69.60	166.71±64.22	0.053	−10.08±41.60	0.087
Mean±SD	Control	58	177.69±62.30	152.46±55.80	<0.001	−25.24±54.17
LVESV, ml	ASV	66	127.85±64.07	112.29±59.87	0.002	−15.56±39.27	0.174
Mean±SD	Control	58	131.05±51.39	104.09±48.88	<0.001	−26.96±51.64
Stroke volume, ml	ASV	66	48.94±13.55	54.42±14.91	<0.001	5.48±11.50	0.155
Mean±SD	Control	58	46.65±17.41	48.37±16.25	0.437	1.73±16.78
E/A	ASV	52	1.40±1.00	1.03±0.60	0.004	−0.36±0.87	0.788
Mean±SD	Control	41	1.46±1.09	1.16±1.05	0.123	−0.30±1.22
Deceleration time, ms	ASV	66	183.05±70.42	207.15±64.56	0.007	24.10±70.77	0.393
Mean±SD	Control	55	195.61±82.99	207.97±67.34	0.255	12.37±79.63
E/e’	ASV	59	18.34±7.97	16.33±7.13	0.031	−2.02±7.02	0.833
Mean±SD	Control	52	17.25±9.71	14.88±7.55	0.098	−2.38±10.17
TRPG, mmHg	ASV	50	29.21±13.91	23.48±10.20	0.006	−5.73±13.99	0.053
Mean±SD	Control	48	24.93±10.82	24.54±10.81	0.836	−0.39±12.90
Chest X-ray
CTR,%	ASV	87	54.88±6.40	53.33±6.29	0.005	−1.55±4.97	0.856
Mean±SD	Control	87	55.01±6.86	53.32±6.92	0.004	−1.69±5.39

^†Intragroup differences were calculated using paired t-test or one-sample Wilcoxon test; ^‡intergroup differences were calculated using Student’s t-test or two-sample Wilcoxon test. ADL, activities of daily living; DBP, diastolic blood pressure; E/A, ratio of the E wave velocity to the A wave velocity; E/e’, ratio of the E wave velocity to the early velocity of mitral annulus maximal expansion; LVEDV, left ventricular end-diastolic volume; LVESV, left ventricular end-systolic volume; MLHFQ, Minnesota Living with Heart Failure Questionnaire; QOL, quality of life; SBP, systolic blood pressure; TRPG, tricuspid regurgitation pressure gradient. Other abbreviations as in Table 1.

BNP

Baseline-completion changes in plasma BNP levels in the ASV and control groups are shown in Figure 3. Plasma BNP levels at baseline and completion in the ASV group and the control group were 818.6±3,876.1 pg/dl and 289.2±415.3 pg/dl, and 642.6±2,031.2 pg/dl and 431.2±1,136.5 pg/dl, respectively, showing statistically significant decreases in the 2 groups (P<0.001 each). However, no statistically significant difference was found between the groups (P=0.575).

Figure 3.

Baseline-completion changes in plasma level of B-type natriuretic peptide (BNP) in the adaptive servo-ventilation (A) and control (B) groups. Black line: individual data; red and blue lines: mean±SD. Intragroup P values (one-sample Wilcoxon test); intergroup P values (two-sample Wilcoxon test).

Cardiac Events

The Kaplan-Meier curves for event-free survival exhibited time-course declines in the 2 groups, and the extent of dissociation between the 2 curves tended to increase at approximately 2 months or more of follow-up (Figure 4). The total numbers of cardiac events in the ASV group and the control group were 14 and 21, respectively, with the following breakdowns: death from HF: 1 and 2; sudden death: 0 and 0; admission to hospital because of disease progression: 10 and 16; treatment modification for HF: 2 and 3; and long-term deterioration (≥12 weeks) of symptoms: 1 and 0. No statistically significant difference (P=0.283) was found between the groups with respect to these events. Furthermore, the numbers of noncardiac deaths in the ASV and control groups were 1 (hepatocyte carcinoma) and 1 (pneumonia), respectively.

Figure 4.

Kaplan-Meier curves for event-free survival in the ASV (red) and control (blue) groups. P value (log-rank test). ASV, adaptive servo-ventilation.

Symptoms, ADL, and QOL

The proportions of CCR at completion in the ASV group and the control group are shown in Figure 5. The proportion of patients with improved CCR was significantly greater (P=0.002) in the ASV group than in the control group. At completion, NYHA class (a component of CCR) had improved significantly (P<0.001 and P=0.044) in the ASV group and the control group, respectively, with a significant intergroup difference (P=0.009).

Figure 5.

Proportions of clinical composite response in the ASV and control groups at completion of the study. P value (two-sample Wilcoxon test). ASV, adaptive servo-ventilation.

ADL assessed with the specific activity scale improved significantly (P<0.001) in both the ASV group and the control group at completion in conjunction with improvements in NYHA class, with a significant intergroup difference (P=0.042). In contrast, QOL assessed with the Minnesota Living with Heart Failure Questionnaire improved significantly (P<0.001) in the 2 groups at completion; however, QOL in the ASV group did not improve to such an extent to generate a significant intergroup difference (P=0.612).

Miscellaneous

Vital signs (systolic blood pressure (SBP), diastolic blood pressure (DBP), and pulse rate), and X-ray (cardiothoracic ratio (CTR) and intensity of pulmonary congestion) exhibited no intergroup differences at completion, although significant changes at completion were found with respect to some variables (eg, BMI, SBP, DBP, and CTR).

Safety Assessment

The adverse events were as follows: in the ASV group, 29 episodes (diarrhea, worsened HF, ventricular tachycardia, conjunctivitis, anemia, hemorrhagic stroke, alcohol-induced hepatopathy, arthralgia, complete total occlusion, and influenza infection) in 18 patients, including 16 serious (ie, hepatocyte carcinoma, colon cancer, worsened HF, ventricular tachycardia, urinary infection, fever, fracture, and decreased renal function); in the control group, 41 episodes (ie, worsened HF, atrial fibrillation, ventricular tachycardia, cardiac resynchronization therapy, suspected thrombotic phlebitis of the legs, anemia, coronary atherosclerosis, dyspnea, fatigability, weight gain, lightheadedness after standing, interstitial pneumonia, skin ulcer, and hematuria) in 29 patients, including 17 serious (ie, worsened HF, stomach cancer, bladder cancer, complete total occlusion, pneumonia, choledocholithiasis, increased BNP, and sick sinus syndrome). None of these adverse events was considered by the attending physician to have causality with ASV therapy.

Stratified Analysis

The results from the analysis of variables (LVEF, cardiac events without HF deterioration, BNP, and CCR) that were stratified by AHI to assess the severity of SDB are shown in Table 3. The stratified analysis revealed no enhancement of the cardiac function-improving effect of ASV therapy.

Table 3. Analysis of Variables Stratified by the AHI (<15/≥15/h), an Index of the Severity of Sleep-Disordered Breathing

Variable	Stratified threshold	Group	n	Change from baseline to completion		Intragroup P value†	Intergroup P value‡	P value§ on stratification
				Mean±SD
LVEF (%)	AHI <15/h	ASV	20	3.32±4.99		0.008	0.385	0.098
		Control	21	5.18±8.23		0.009
	AHI ≥15/h	ASV	41	6.52±9.89		<0.001	0.447
		Control	31	8.76±13.86		0.001
	Stratified threshold	Group	n	No. of patients with events	Hazard ratio [95% CI]		Intergroup P value	P value on stratification
Cardiac events with HF deterioration	AHI <15/h	ASV	33	3	0.506 [0.127–2.025]		0.327	0.467
		Control	33	6
	AHI ≥15/h	ASV	62	9	0.744 [0.313–1.766]		0.501
		Control	60	12
	Stratified threshold	Group	n	Change from baseline to completion		Intragroup P value	Intergroup P value	P value on stratification
				Median [25–75 percentile rang]
BNP (pg/ml)	AHI <15/h	ASV	30	−61.6 [−149.8 to 72.9]		0.204	0.699	0.348
		Control	26	−21.9 [−94.7 to 49.5]		0.235
	AHI ≥15/h	ASV	56	−62.7 [−269.0 to 3.9]		<0.001	0.632
		Control	48	−60.0 [−285.8 to 42.4]		0.022
	Stratified threshold	Group	n	n (%) at completion			Intergroup P value	P value on stratification
				Improved	Unchanged	Worsened
Clinical composite response	AHI <15/h	ASV	29	17 (58.6%)	5 (17.2%)	7 (24.1%)	0.194	0.996
		Control	26	8 (30.8%)	13 (50.0%)	5 (19.2%)
	AHI ≥15/h	ASV	58	33 (56.9%)	13 (22.4%)	12 (20.7%)	0.014
		Control	55	18 (32.7%)	18 (32.7%)	19 (34.5%)

^†Intragroup differences were calculated using paired t-test or one-sample Wilcoxon test; ^‡intergroup differences were calculated using Student’s t-test, log-rank test, or two-sample Wilcoxon test; ^§differences between two stratified ASV groups were calculated using Student’s t-test, log-rank test, or two-sample Wilcoxon test. CI, confidence interval; HF, heart failure. Other abbreviations as in Table 1.

Discussion

Along with increments in clinicopathological and biochemical knowledge about CHF, pharmacotherapeutic strategies for the disease have changed evolutionally: digitalis and diuretics to treat congestion and edema since the late 1950 s; vasodilators to treat hemodynamic abnormalities since the 1970 s; and ACE inhibitors, ARBs, and β-blockers to treat neurohumoral factor abnormalities since the late 1980 s.²⁸ In particular, ACE inhibitors, and β-blockers have the proven effect of improving prognosis, the true endpoint for patients with CHF.²⁹ Despite these therapeutic efforts and advances in pharmacotherapy, HF severe enough to require hospitalization is more “malignant” with regard to 5-year survival than many of the common types of cancer, with the exception of lung cancer.³⁰ This alarming fact clearly indicates the therapeutic limitations of the current GDMT and how critically important are the early diagnosis and treatment of CHF and the prevention of disease progression, which drove us to seek a novel, noninvasive, nonpharmacological modality that could be added to the current GDMT.

To examine whether ASV therapy could become such a modality, we established LVEF as the primary endpoint and a surrogate variable for the assessment of the true endpoint. Subsequently, we conducted the present study on the following assumption based on prior studies that had suggested increases in stroke volume and cardiac output due to NPPV^1,5 and LV reverse remodeling¹⁴ due to ASV: the addition of ASV therapy to GDMT might additively improve the cardiac function of CHF patients in conjunction with LV reverse remodeling. In the present study, however, a statistically significant difference between the ASV group and the control group was not found with respect to LVEF and BNP, although these variables at completion improved significantly as compared with the baseline values in both groups. One conceivable reason for these findings for LVEF and BNP is that the stabilization of clinical status (GDMT and the absence of medication initiation/modification and medication dose increment within 3 months prior to study enrollment) was not established as an inclusion criterion in our consideration of the actual usage of the ventilator in the real-world clinical setting. Consequently, patients for whom medication with β-blockers was initiated or their dose was increased within 30 days and 90 days prior to the study onset accounted for 35% and 55% of the study population, respectively. Beta-blockers have an established suppressive effect on LV remodeling through their inhibitory effect on sympathetic nerve overactivity, which leads to reductions in LV oxygen consumption and LV filling pressure.³¹ On the other hand, ASV therapy was suggested to stabilize respiration, resulting in suppression of sympathetic nerve overactivity. Therefore, we speculate that ASV therapy and β-blockers probably exert their cardiac function-improving effect through these common pathways within the putative mechanisms of action¹⁴ to such an extent that no statistically significant difference could be detected between the ASV group and the control group in the present study.

Symptoms improved significantly after intervention as observed in SAVIOR-R;¹⁴ furthermore, symptoms and CCR (ie, clinical status²²), as well as ADL improved significantly in the ASV group compared with the control group. On the other hand, QOL improved after intervention but showed no intergroup difference; this result is probably related to the fact that the QOL of patients in the control group improved more than we expected, which resulted in no demonstration of superiority of ASV therapy in improving QOL over the control group. Nevertheless, the ASV group was superior to the control group in the symptom-improving effect. Furthermore, adherence is a consequence of subjective benefits that the patient experiences in therapeutic interventions.³² Therefore, the improved clinical status of the patient through awareness of alleviated symptoms will contribute to an improvement in QOL.

A prior single-center observational study described the significantly decreased incidence of event-free survival, which serves as the assessment of 1 of the objectives of CHF treatment: prevention of disease progression at 1 year of follow-up.¹³ Therefore, the tendency for dissociation observed with the Kaplan-Meier curves for event-free survival in the present study might generate a statistically significant difference if patients were followed up for 1 year.

In other countries, the AutoSet^TM CS was originally developed for the treatment of SDB in patients with HF; therefore, ASV therapy using the device is conducted exclusively for patients who are complicated by SDB. In Japan, on the other hand, ASV therapy is used to treat patients with HF, regardless of SDB severity, because of its hemodynamic effects (eg, reduction in preload³ and an increase in intrathoracic pressure^1,2). In fact, the analysis of LVEF, cardiac events without HF deterioration, BNP, and CCR stratified by AHI in the present study revealed no enhancement of the effects of ASV therapy. Namely, the hemodynamic effects of ASV therapy were not influenced by SDB severity as previously suggested by SAVIOR-R.¹⁴

Study Limitations

First, the effects of ASV therapy on the primary endpoint could not be duly assessed because the present study was conducted in real-world settings where the clinical status of the patient undergoing GDMT was not necessarily stable. Second, the present study could not be blinded because of allowing ASV therapy at home (ie, AutoSet^TM CS). Therefore, the possibility of biases (eg, selection bias, evaluator bias, and participant expectation bias) cannot be ruled out. However, physicians, technicians, and statisticians were all blinded to treatment assignment. Third, the true endpoint of ASV therapy (patient prognosis) was assessed indirectly by surrogate markers (eg, LV remodeling and BNP). The endpoint needs to be assessed as the primary endpoint in a future clinical study.

Conclusions

Under the conditions of the present study, we could not demonstrate the superiority of ASV therapy at home to GDMT in the cardiac function-improving effect on patients with CHF. Nevertheless, ASV therapy showed improving effects on symptoms, ADL, and CCR, thus providing a given level of clinical benefit for this outpatient population.

Acknowledgments

The authors are grateful to the members of the Central Adjudication Committee for their great contribution to the present study and to Satoshi Sakima, MD, for the linguistic review of the manuscript.

The research fund for SAVIOR-C was provided to the Comprehensive Support Project for Clinical Research of Lifestyle-Related Disease of the Public Health Research Foundation by Teijin Pharma Limited.

Conflict of Interest

Shin-ichi Momomura, lecture fee from Teijin Pharma Limited; Yasuki Kihara, lecture fee from Teijin Pharma Limited.

Appendix 1

Lists of Steering Committee, Central Adjudication Committee, Study Promotion Committee, and Adviser

Steering Committee

Shin-ichi Momomura (Principal Investigator, Saitama Medical Center, Jichi Medical University), Yoshihiko Seino (Nippon Medical School Chiba Hokusoh Hospital), Yasuki Kihara (Hiroshima University Graduate School of Biomedical & Health Sciences), Hitoshi Adachi (Gunma Prefectural Cardiovascular Center), Yoshio Yasumura (Osaka National Hospital), and Hiroyuki Yokoyama (National Cerebral and Cardiovascular Center).

Central Adjudication Committee (Independent Data and Safety Monitoring Committee)

Kiyoshi Yoshida (Kawasaki Medical School), Akihiro Hayashida (Kawasaki Medical School), Naoki Sato (Nippon Medical School Musashi-Kosugi Hospital), Takeshi Hozumi (Osaka City University), Hiroshi Yamashita (the University of Tokyo), Kazuhiro Yamamoto (Tottori University Faculty of Medicine) Study Promotion Committee: Hidetsugu Asanoi (Imizu Municipal Hospital), Shinichi Ando (Kyushu University Graduate School of Medicine), Masaaki Ito (Mie University Graduate School of Medicine), Hiroshi Inoue (University of Toyama), Hisao Ogawa (Kumamoto University Graduate School of Medical Science), Masataka Sata (University of Tokushima), Hiroaki Shimokawa (Tohoku University Graduate School of Medicine), Hiroyuki Tsutsui (Hokkaido University Graduate School of Medicine), Hitonobu Tomoike (Sakakibara Heart Institute), Tohru Masuyama (Hyogo College of Medicine), Toyoaki Murohara (Nagoya University Graduate School of Medicine), Tsutomu Yoshikawa (Sakakibara Heart Institute).

Adviser

Kenji Ueshima (Department of EBM Research, Institute for Advancement of Clinical and Translational Science, Kyoto University Hospital).

Appendix 2

List of Investigators at the Participating Medical Institutions of SAVIOR-C

Saitama Medical Center, Jichi Medical University: S. Momomura, H. Wada, Y. Sugawara, N. Ikeda, Y. Tsurumaki; University of Tokushima: M. Sata, T. Ise, S. Yagi; Miyoshi Central Hospital: K. Tanaka, K. Kobayashi, Y. Orita; Hyogo Brain and Heart Center: Y. Taniguchi, S. Oishi, K. Yokoi; Hiroshima University Graduate School of Biomedical & Health Sciences: Y. Kihara, N. Oda, T. Hidaka, M. Kajikawa, Y. Watanabe; National Cerebral and Cardiovascular Center: H. Yokoyama and T. Anzai, K. Nakamura, Y. Sugano, M. Yasuda, M. Fukumitsu, T. Nakayama; Juntendo University School of Medicine: T. Kasai; Mie University Graduate School of Medicine: M. Ito, N. Kumagai, Y. Sato, K. Dohi, M. Tanimura, K. Watanabe; University of Occupational and Environmental Health: N. Haruki; Nippon Medical School Chiba Hokusoh Hospital: Y. Seino, T. Inami; Nagoya Ekisaikai Hospital: T. Kato, H. Hiraiwa, T. Kondo, T. Ishihara; National Hospital Organization Higashihiroshima Medical Center: K. Yanagihara, M. Hara, H. Jo; Matsuyama Red Cross Hospital: T. Ashihara and T. Kubo; Saitama International Medical Center, Saitama Medical University: T. Muramatsu, S. Nakano; Sakakibara Heart Institute: T. Yoshikawa, S. Sasaki, N. Iguch, M. Nagayama, Y. Izumi; Gunma Prefectural Cardiovascular Center: H. Adachi, H. Takamatsu; University of Toyama: H. Inoue, S. Joho; Imizu Municipal Hospital: H. Asanoi, D. Harada, H. Ishise, K. Oshitani, H. Ueno; Kyushu University Graduate School of Medicine: K. Sunagawa, T. Higo, M. Nishizaka; Kumamoto University Graduate School of Medical Science: H. Ogawa, S. Kojima; Nippon Medical School: K. Mizuno and N. Fukuma, Y. Kubota, H. Takahashi; National Defense Medical College: T. Adachi, T. Kimura, D. Hakuno, S. Isobe, K. Akita, N. Masaki; University of the Ryukus Graduate School of Medicine: T. Yasu and Y. Ohya, N; Aizawa; Tenri Hospital: Y. Nakagawa, H. Kondo; Tohoku University Graduate School of Medicine: H. Shimokawa, K. Nochioka, Y. Fukumoto; Nagoya University Graduate School of Medicine: T. Murohara, A. Hirashiki, T. Okumura; Anjo Kosei Hospital: M. Watarai, M. Koyasu, A. Miura, S. Ishikawa; Higashisumiyoshi Morimoto Hospital: Y. Sakanoue, T. Kanemori, Y. Nakamura; Kansai Rosai Hospital: M. Fujita; Kokura Memorial Hospital: M. Iwabuchi and K. Ando, T. Mazaki; Nagano Chuo Hospital: T. Kono, C. Itamoto, H. Miura. Kitano Hospital: T. Izumi; Osaka National Hospital: Y. Yasumura, G. Hamano; Keio University School of Medicine: K. Fukuda, T. Kohno, S. Mogi; National Hospital Organization Nagoya Medical Center: T. Yamada; Saiseikai Futsukaichi Hospital: T. Kadokami, K. Hosokawa, T. Touyama; Ehime Prefectural Central Hospital: H. Okayama, G. Hiasa; Hokkaido University Graduate School of Medicine: H. Tsutsui, S. Yamada; Osaka Red Cross Hospital: T. Inada, H. Ito.

References

• 1.    Naughton  MT,  Rahman  MA,  Hara  K,  Floras  JS,  Bradley  TD. Effect of continuous positive airway pressure on intrathoracic and left ventricular transmural pressures in patients with congestive heart failure. Circulation 1995; 91: 1725–1731.
• 2.    Mehta  S,  Hill  NS. Noninvasive ventilation. Am J Respir Crit Care Med 2001; 163: 540–577.
• 3.    Takano  T,  Endo  T,  Tanaka  K,  Seino  Y,  Nitta  T,  Matsuyama  Y, et al. Effects of positive end-expiratory pressure ventilation and extracorporeal ultrafiltration method in patients with refractory heart failure. Jpn Circ J 1986; 50: 359–367.
• 4.   JCS Joint Working Group. Guidelines for treatment of acute heart failure (JCS 2011): Digest version. Circ J 2013; 77: 2157–2201.
• 5.    Gray  A,  Goodacre  S,  Newby  DE,  Masson  M,  Sampson  F,  Nicholl  J; 3CPO Trialists. Noninvasive ventilation in acute cardiogenic pulmonary edema. N Engl J Med 2008; 359: 142–151.
• 6.    Harada  D,  Joho  S,  Oda  Y,  Hirai  T,  Asanoi  H,  Inoue  H. Short term effect of adaptive servo-ventilation on muscle sympathetic nerve activity in patients with heart failure. Auton Neurosci 2011; 161: 95–102.
• 7.    Joho  S,  Oda  Y,  Ushijima  R,  Hirai  T,  Inoue  H. Effect of adaptive servoventilation on muscle sympathetic nerve activity in patients with chronic heart failure and central sleep apnea. J Card Fail 2012; 18: 769–775.
• 8.    Ushijima  R,  Joho  S,  Akabane  T,  Oda  Y,  Inoue  H. Differing effects of adaptive servoventilation and continuous positive airway pressure on muscle sympathetic nerve activity in patients with heart failure. Circ J 2014; 78: 1387–1395.
• 9.    Tamura  Y,  Koyama  T,  Watanabe  H,  Hosoya  T,  Ito  H. Beneficial effects of adaptive servo-ventilation therapy on albuminuria in patients with heart failure. J Cardiol 2014 February 19, doi:org/10.1016/j.jjcc.2014.07.008.
• 10.    Ushijima  R,  Joho  S,  Akabane  T,  Oda  Y,  Inoue  H. Differing effects of adaptive servoventilation and continuous positive airway pressure on muscle sympathetic nerve activity in patients with heart failure. Circ J 2014; 786: 1387–1395.
• 11.    Koyama  T,  Watanabe  H,  Tamura  Y,  Oguma  Y,  Kosaka  T,  Ito  H. Adaptive servo-ventilation therapy improves cardiac sympathetic nerve activity in patients with heart failure. Eur J Heart Fail 2013; 15: 902–909.
• 12.    Takama  N,  Kurabayashi  M. Effectiveness of adaptive servo-ventilation for treating heart failure regardless of the severity of sleep-disordered breathing. Circ J 2011; 75: 1164–1169.
• 13.    Koyama  T,  Watanabe  H,  Igarashi  G,  Terada  S,  Makabe  S,  Ito  H. Short-term prognosis of adaptive servo-ventilation therapy in patients with heart failure. Circ J 2011; 75: 710–712.
• 14.    Momomura  S,  Seino  Y,  Kihara  Y,  Adachi  H,  Yasumura  Y,  Yokoyama  H; the SAVIOR-R investigators. Adaptive servo-ventilation therapy using an innovative ventilator for patients with chronic heart failure: A real-world, multicenter, retrospective, observational study (SAVIOR-R). Heart Vessels 2014 August 8, doi:10.1007/s00380-014-0558-8.
• 15.    Haruki  N,  Takeuchi  M,  Kaku  K,  Yoshitani  H,  Kuwaki  H,  Tamura  M, et al. Comparison of acute and chronic impact of adaptive servo-ventilation on left chamber geometry and function in patients with chronic heart failure. Eur J Heart Fail 2011; 13: 1140–1146.
• 16.    Yamada  S,  Sakakibara  M,  Yokota  T,  Kamiya  K,  Asakawa  N,  Iwano  H, et al. Acute hemodynamic effects of adaptive servo-ventilation in patients with heart failure. Circ J 2013; 77: 1214–1220.
• 17.    Takama  N,  Kurabayashi  M. Safety and efficacy of adaptive servo-ventilation in patients with severe systolic heart failure. J Cardiol 2014; 63: 302–307.
• 18.    Merlo  M,  Pyxaras  SA,  Pinamonti  B,  Barbati  G,  Di Lenarda  A,  Sinagra  G. Prevalence and prognostic significance of left ventricular reverse remodeling in dilated cardiomyopathy receiving tailored medical treatment. J Am Coll Cardiol 2011; 57: 1468–1476.
• 19.    Seino  Y,  Momomura  S,  Kihara  Y,  Adachi  H,  Yasumura  Y,  Yokoyama  H; for the SAVIOR-C investigators. Effects of adaptive servo-ventilation therapy on cardiac function and remodeling in patients with chronic heart failure (SAVIOR-C): Study protocol for a randomized controlled trial. Trials 2015; 16: 14, doi:10.1186/s13063-014-0530-z.
• 20.    Matsuzaki  M,  Ishii  M,  Izumi  T,  Ito  H,  Ito  H,  Ueda  Y, et al. Guidelines for Treatment of Chronic Heart Failure (JCS 2010). http://www.j-circ.or.jp/guideline/pdf/JCS2010_matsuzaki_h.pdf (accessed February 10, 2015).
• 21.    Lang  RM,  Bierig  M,  Devereux  RB,  Flachskampf  FA,  Foster  E,  Pellikka  PA, et al; Chamber Quantification Writing Group; American Society of Echocardiography’s Guidelines and Standards Committee; European Association of Echocardiography. Recommendations for chamber quantification: A report from the American Society of Echocardiography’s Guidelines and Standards Committee and the Chamber Quantification Writing Group, developed in conjunction with the European Association of Echocardiography, a branch of the European Society of Cardiology. J Am Soc Echocardiogr 2005; 18: 1440–1463.
• 22.    Packer  M. Proposal for a new clinical end point to evaluate the efficacy of drugs and devices in the treatment of chronic heart failure. J Card Fail 2001; 7: 176–182.
• 23.    Chesson  AL Jr,  Berry  RB,  Pack  A. Practice parameters for the use of portable monitoring devices in the investigation of suspected obstructive sleep apnea in adults. Sleep 2003; 26: 907–913.
• 24.   American Academy of Sleep Medicine Task Force. Sleep-related breathing disorders in adults: Recommendations for syndrome definition and measurement techniques in clinical research [Report of the Taskforce]. Sleep 1999; 22: 667–689.
• 25.    Johns  MW. A new method for measuring daytime sleepiness: The Epworth sleepiness scale. Sleep 1991; 14: 540–545.
• 26.    Sasayama  S,  Asanoi  H,  Ishizaka  S,  Miyagi  K. Evaluation of functional capacity of patients with congestive heart failure. In: Yasuda H, Kawaguchi H, editors. New aspects in the treatment of failing heart. Tokyo: Springer-Verlag, 1992; 113–117.
• 27.    Rector  TS. A conceptual model of quality of life in relation to heart failure. J Card Fail 2005; 11: 173–176.
• 28.    Katz  AM. Changing strategies in the management of heart failure. J Am Coll Cardiol 1989; 13: 513–523.
• 29.    Packer  M. Treatment of chronic heart failure. Lancet 1992; 340: 92–95.
• 30.    Stewart  S,  MacIntyre  K,  Hole  DJ,  Capewell  S,  McMurray  JJ. More “malignant” than cancer?: Five-year survival following a first admission for heart failure. Eur J Heart Fail 2001; 3: 315–322.
• 31.    Doughty  RN,  Whalley  GA,  Walsh  HA,  Gamble  GD,  López-Sendón  J,  Sharpe  N; CAPRICORN Echo Substudy Investigators. Effects of carvedilol on left ventricular remodeling after acute myocardial infarction: The CAPRICORN Echo Substudy. Circulation 2004; 109: 201–206.
• 32.    Philippe  C,  Stoïca-Herman  M,  Drouot  X,  Raffestin  B,  Escourrou  P,  Hittinger  L, et al. Compliance with and effectiveness of adaptive servoventilation versus continuous positive airway pressure in the treatment of Cheyne-Stokes respiration in heart failure over a six month period. Heart 2006; 92: 337–342.