Sleep Disordered Breathing in Older Adults with Heart Failure with Preserved Ejection Fraction

Abstract

Heart failure in older adults is frequently accompanied by sleep disordered breathing (SDB). Treatment of SDB in persons with heart failure with preserved ejection fraction (HFpEF) is unclear because most data is on heart failure with reduced ejection fraction (HFrEF). The purpose of this paper was to evaluate studies that report on the effects of positive airway pressure on patient outcomes in older adults with HFpEF and comorbid SDB. A search of the literature found six data-based studies (N = 36 to 126). Treatment with positive airway pressure reduced nighttime SDB symptoms and improved daytime functional status in persons with HFPEF and SDB (New York Heart Association Functional Class: effect sizes= −.67 to −1.60). Limitations (i.e. only two studies were randomized controlled trials, small sample sizes, and women were under-represented) suggest that additional evidence is needed to guide treatment of SDB in older adults with HFpEF.

Introduction

According to the American Heart Association, in 2014 over 6.5 million adults in the U.S. had heart failure. The prevalence of heart failure is expected to increase by 46% by 2030 to over 8 million.¹ Heart failure is associated with significant decreases to health-related quality of life (HRQoL), daytime dysfunction, fatigue, dyspnea, and increased mortality.² The total cost of heart failure in 2012 was estimated as almost 31 billion dollars; this cost is projected to increase to 70 billion dollars by 2030.³ Although cardiac disease is not part of healthy aging, older adults have a disproportionate risk. The prevalence of heart failure is less than 2% of adults under 60 years old, it increases to 6% in the 60 to 79 year-old age group and is greater than 13% in persons over 80 years old.¹

Heart failure is characterized as either heart failure with reduced ejection fraction (HFrEF; left ventricular ejection fraction [LVEF] ≤ 40%) or as heart failure with preserved ejection fraction (HFpEF; LVEF ≥ 50%).⁴ Half of all patients with heart failure present with a preserved LVEF;¹ these patients have a unique phenotype from HFrEF. Patients with HFpEF are frequently older (often in the 8^th decade of life), more likely to be female, without a myocardial infarction, and have a history of hypertension, coronary artery disease, obesity compared to patients with HFrEF.^1,5–7 The prevalence of HFpEF relative to HFrEF is increasing at a rate of 1% per year and may become the predominant form of heart failure in the developed world.² Unlike HFrEF, there are no established therapies to improve mortality and morbidity in patients with HFpEF.⁸ There are differences in the etiology of morbidity and mortality between HFrEF and HFpEF, with morbidity in HFpEF influenced more by non-heart failure related conditions.^2,9 Thus, therapy in HFpEF is aimed at treating the more prevalent comorbidities that coexist in persons with HFpEF in order to reduce symptom burden and to improve HRQoL.²

Sleep disordered breathing (SDB) is a highly prevalent comorbidity in adults with HFpEF, estimated to range from 50% to over 80% when compared to persons who have no history of heart failure.^10–14 According to the American Academy of Sleep,¹⁵ SDB refers to a number of breathing disruptions that occur during sleep that include:

• obstructive apneas (cessation of airflow for 10 seconds or longer because of an obstruction in airflow in the upper pharynx with continued respiratory effort),

• hypopneas (a decrease in airflow by 50% from baseline accompanied by a decrease of 3% in oxygen saturation),

• central apneas (absence of a central nervous system signal to breath with a cessation in breathing for at least 10 seconds),

• mixed apneas (an apnea that starts as an obstructive event but converts to a central apnea, or vice versa), and

• Cheyne-Stokes respirations (a cyclic pattern of waxing and waning airflow with respirations).

Data from previous research suggests that SDB results in sympathetic activation, sleep fragmentation, and intermittent hypoxia that is associated with a variety of adverse consequences in older adults including excessive daytime sleepiness,¹⁶ depression,¹⁶ cognitive decline,^17,18 reduced functional status,¹⁹ cardiovascular disease,^20,21 and increased mortality.²¹

In patients with heart failure, the choice of treatment for SDB depends on the severity and type of breathing disturbance, symptom presentation (i.e. presence of snoring or excessive daytime sleepiness), and the type of cardiovascular disease present.^22–24 Positive airway pressure (e.g., continuous positive airway pressure [CPAP], bi-level positive airway pressure [bi-PAP], or adaptive servo-ventilation [ASV]) is the mainstay treatment of SDB.²² CPAP involves wearing a mask that delivers a steady stream of air to prevent airway collapse and obstructive apneas. Unlike CPAP, Bi-PAP machines have two pressure settings with a higher pressure for inhalation and a lower pressure for exhalation. Bi-PAP is useful for patients who find it difficult to exhale against positive pressure, require a high positive pressure setting to prevent airway obstruction, require oxygen, have a neuro-muscular disorder, or who have HFrEF. ASV is a further refinement as a type of positive airway pressure that uses an algorithm based on the detection of apneas to the level of pressure required to maintain respiration with inspiratory support to treat central apneas.^25,26 ASV is used for patients with more complex SDB consisting with mixed obstructive and central apneas and Cheyne-Stokes respirations.

While the prevalence of SDB is similar between the populations of HFrEF and HFpEF, whether the predominant type of sleep apnea (e.g. obstructive versus central sleep apnea) is similar remains unclear.^14,27 Central sleep apnea is more common in persons with heart failure compared to the general population.¹² The prevalence of central sleep apneas has been shown to increase significantly (p<.05) with increasing heart failure severity.^27,28 In general, patients with HFpEF tend to have more obstructive apneas in comparison with patients with HFrEF.¹⁴ In a prospective study of both HFpEF and HFrEF, OSA was the predominant type of SDB in the HFpEF patients (62%) but not in HFrEF patients.¹⁴ A larger study of 244 persons with only HFpEF also found more OSA (40%) versus central sleep apnea (30%) in those identified with SDB.²⁷ However, several studies suggest that the pattern of central apneas, Cheyne-Stokes respirations, obstructive apneas, and hypopneas changes on a night-to-night basis further complicating the delineation of SDB between persons with HFrEF and HFpEF as well as the management of SDB in each population.^26,29

A growing body of evidence over the past decade supports positive airway pressure (e.g., CPAP, bi-PAP, or ASV) as effective methods to treat SDB with demonstrated improvements to nocturnal respiratory function and daytime sleepiness.^30–32 The impact of positive airway pressure on cardiac function and mortality is less clear.^30–33 For example, the Sleep Apnea Cardiovascular Endpoints (SAVE) study³³ was a randomized control trial to evaluate the effectiveness of CPAP in reducing adverse cardiovascular events among patients with cardiovascular disease and comorbid moderate-to-severe OSA (apnea + hypopnea index [AHI] ≥ 15).³³ Those patients randomized to CPAP (n = 1346) had significant improvements in daytime sleepiness, respiratory function, and quality of life measures but not in mortality or the recurrence of myocardial infarction when compared to the usual-care group (n = 1341).

Additionally, adverse results occurred in a second study the Treatment of Sleep-Disordered Breathing with Predominant Central Sleep Apnea by Adaptive Servo-Ventilation in Patients with Heart Failure (SERVE-HF).²⁵ The SERVE-HF study, a large randomized controlled trial (N= 1,325) of subjects with HFrEF (LVEF ≤ 45%) and comorbid moderate-to-severe central sleep apnea (AHI ≥ 15), prompted healthcare professionals to question the safety of using ASV in persons with heart failure. The SERVE-HF trial was prematurely stopped because of a statistically significant increase in the primary end-point of all-cause mortality and cardiovascular deaths in those participants randomized to ASV (p <.01). Although the results from the SERVE-HF study cannot be generalized to persons with HFpEF, both the SAVE and SERVE-HF trials have rendered the treatment of SDB among heart failure patients a controversial topic regardless of treatment type.

While the negative prognostic impact of HFpEF is similar to that of HFrEF,³⁴ each type of heart failure represents a different clinical syndrome that should be studied and treated separately.³⁵ Important differences exist between HFpEF and HFrEF with respect to the extent of myocardial dysfunction, patterns of cardiac remodeling in heart chambers, and in the response to therapeutic interventions.³⁵ Additionally, OSA in older adults greater than 65 years has a distinctly different physiological phenotype in old versus young individuals.³⁶ As a result of these differences, outcomes during positive airway pressure therapy in patients with HFpEF, typically persons greater than 65 years with comorbid SDB, would be expected to be different from the samples in both the SERVE-HF and SAVE trials.

Previous studies on the treatment of SDB in adults with heart failure have historically focused on middle aged adults, featured samples for which the type of heart failure was not identified, or where the majority of the participants had symptoms suggestive of HFrEF. Therefore, the effect of positive airway therapy on patient-centered and clinical outcomes in older adults with HFpEF is poorly understood. The purpose of this paper is to evaluate the studies that report on the effects of positive airway pressure on patient-centered and clinical outcomes in older adults with HFpEF and comorbid SDB.

Methods

Search and data abstraction

In June 2017, a systematic all-field and no-date-limit literature search using PubMed, OVID, and Web of Science was conducted. Two investigators, working independently, used the following combination of search terms: sleep apnea, or sleep apnea—obstructive, or sleep apnea—central, or sleep apnea—mixed; AND continuous positive airway pressure, or CPAP, or adaptive servo ventilation, or ASV, or bi-level positive airway pressure, or BiPAP; AND heart failure, or heart failure—diastolic, or heart failure—preserved ejection fraction, or HFpEF. The identified articles were analyzed using the protocol below.

The search identified 279 publications (i.e., articles in peer-reviewed journals); 71 were excluded from further analysis because they were written in languages other than English or were not data-based research articles. The titles and abstracts of the remaining 208 articles were screened to determine whether or not the associated full-text articles would be analyzed (Figure 1). One-hundred thirty four publications were excluded for further analysis because the abstracts either did not mention heart failure participants, or lacked content relevant to our topic. This yielded 74 publications that qualified for our full-text review. Following this full-text review, 68 articles were excluded because (1) no delineation was made between the HFrEF and the HFpEF participants or the sample comprised a majority of adults with HFrEF, and/or (2) the study reported did not feature a positive airway pressure intervention for the treatment of sleep apnea in the participants with co-existing HFpEF and SDB.

Figure 1.

Results of systematic review process.

As a result of this search protocol, six articles were evaluated. A common outcome in four of the studies, the New York Heart Association Functional Class, was used to determine the effect of SDB treatment. A standardized effect size for each study was calculated by subtracting the average pre-treatment functional class from the average post-intervention functional class and then dividing the difference by the standard deviation of the pretreatment functional class.³⁷ The relative magnitude of change was based on Cohen’s definitions of a small (0.2), moderate (0.5), and large (> 0.8) effect size.³⁸

Results

Design and samples

The six studies that examined the effect of treatment of sleep apnea in adults with HFpEF with positive airway pressure are described in Table 1.^39–43 Four of these studies^39,40,42,43 had a prospective observational design^39,40,42 and two were randomized controlled trials with a control group that received usual care (i.e. cardiac medications).^41,44 Four studies included a homogenous sample of participants with HFpEF^39,41–43 while two studies included a mixed sample of participants with either HFrEF or HFpEF.^41,44 The majority (56%) of the sample in the Oldenburg study⁴¹ were diagnosed with HFpEF while in O’Connor study⁴⁴ only 19% (n=24) of the sample had HFpEF, although subgroup analyses were completed within this population. The studies were conducted by four independent groups of investigators: two groups in Japan,^41–43 one in Germany,^39,40 and one multicenter study that included investigators from both Germany and United States.⁴⁴ The six studies recruited convenience samples from clinical populations that were small-to-modest in size; the smallest study included 36 participants and the largest study included 126.

Table 1.

Key characteristics of studies on treatment of sleep disordered breathing in adults with heart failure with preserved ejection fraction.

1st author, county, year	Sampl e size	Study Design	Sample	Mean age (years)	% Women, % HFpEF	PAP Device/ Comparison Group	Study Duration	Therapeutic Benefits
Bitter, Germany 2010	N= 60	Observational Prospective	Type of HF: HFpEF class II – III, LVEF ≥ 50% Type of SDB: Moderate to Severe, AHI > 15/h with CSR	69 ± 8	15, 100%	ASV (n = 39)/Comparison (rejected treatment, stopped treatment, or noncompliant; n = 21)	11.6 ± 3 months	Improved respiratory function (AHI longest apnea and hypopnea length, maximum and mean oxygen desaturation, percentage of study time with an oxygen saturation of < 90%), functional capacity (cardiopulmonary exercise testing), cardiac function (echocardiography measure), and NYHA functional class (ES = −0.67); all p- values < 0.05.
Oldenburg, Germany, 2013	N= 45	Observational Prospective	Type of HF: Both HFpEF and HFrEF, LVEF ≥ 50% Type of SDB: Moderate to Severe, AHI ≥ 15/h with Central and Mixed apneas	71 ± 10	9, 56%	ASV (N=38)/No comparison	3.6 ± 1.2 months	Improved NYHA functional class (ES = −0.80), exercise tolerance and oxygen uptake via cardiopulmonary exercise test, and AHI.
Yoshihisa, Japan, 2013	N= 36	Randomized control trial	Type of HF: HFpEF (LVEF > 50%) Type of SDB: Moderate to Severe, AHI > 15/h with OSA and CSR-CSA	64 ± 14	19, 100%	ASV (n = 18)/Comparison (Usual care; n = 18)	6 months, long term follow up	Improved cardiac diastolic function, arterial stiffness, NYHA Functional class (ES = −1.60), and decrease in cardiac death and hospitalizations due to cardiac disease, all p-values < 0.05.
Yoshihisa, Japan, 2015	N= 109	Observational Prospective	Type of HF: HFpEF (LVEF > 50%) Type of SDB: AHI ≥ 15/h with OSA, CSA, or Mixed	68 ± 13	37, 100%	CPAP & ASV (n = 31)/Comparison (Usual care; n = 78)	6 months, long term follow up	Improved, systolic and diastolic blood pressure, % vital capacity, peak oxygen uptake, NYHA Functional class (ES = −1.60), and a reduced cardiac and all-cause mortality, all p- values < 0.05.
Arikawa, Japan, 2016	N = 58	Observational Prospective	Type of HF: HFpEF (LVEF ≥ 50%) Type of SDB: Mild to severe, AHI ≥ 5, OSA	65 ± 15	41%, 100%	CPAP (n=39)/ Comparison (non-OSA; n = 19)	36 months	BNP levels were higher in those with OSA + CPAP vs comparison group at: 6 months [221 (137–324) vs 76 (38–96) pg/ml, p<0.05], 12 months [123 (98–197) vs 52 (38–76) pg/ml, p<0.05] and 36 months [115 (64–174) vs 56 (25–74) pg/ml, p<0.05]
O’Connor, Germany and United States 2017	N = 126	Randomized, control trial, prospective, multicenter	Type of HF: Both HFrEF (LVEF ≤ 45%) and HFpEF (LVEF > 45%) Type of SDB: Moderate to Severe, AHI ≥ 15, CSA, OSA, and mixed	62 ± 14	32, 19%	ASV (n = 65)/Comparison (usual care; n = 61)	6 months	Enrollment stopped early (126 of 215 planned) following the release of the SERVE-HF results, limiting statistical power. Improvements to AHI only. Subgroup analysis with HFpEF suggested significant positive effect on a composite global rank score (hierarchy of death, cardiovascular hospitalizations, and percent changes in 6-min walk distance) at 6 months (p = 0.036).

Note: AHI = apnea/hypopnea index, ASV = adaptive servo-ventilation, BNP = Plasma Brain Natriuretic Peptide, CPAP = continuous positive airway pressure, CSA = central sleep apnea, CSR = Cheyne-Stokes Respirations, ES = Effect Size, HF = Heart Failure, HFpEF = heart failure with preserved ejection fraction, HFrEF = heart failure with reduced ejection fraction, LVEF = left ventricular ejection fraction, OSA = obstructive sleep apnea, PAP = positive airway pressure, SDB = sleep disordered breathing.

Effect of positive airway pressure on outcomes in persons with HFpEF and comorbid SDB

Bitter et al. examined persons with stable HFpEF (NYHA classes II–III) and primarily central sleep apnea and Cheyne-Stokes respiration (i.e., a ratio of central to obstructive apneas > 50%).³⁹ All participants (N=60) were offered ASV. Assessments were made between the treatment group (n = 39) and a non-equivalent comparison group (n = 21) of participants who either refused or could not tolerate treatment. Significant improvements were observed in cardiopulmonary exercise testing, NYHA class, and cardiac function among the participants of the ASV treatment group (mean follow up 11.6 ± 3 months; all p-values < .01) compared to the untreated comparison group.

A study by Oldenburg et al.⁴⁰ screened 45 consecutive patients with heart failure (HFrEF or HFpEF) and moderate-to-severe SDB (either central or mixed sleep apneas) to reveal the long-term (i.e., 3.6 ± 1.2 months) effects of a tri-level ASV device on cardiac function.⁴⁰ Only 38 patients were successfully titrated to tri-level ASV. On follow-up (3.6 ± 1.2 months) those patients (n=23) adherent to ASV (mean ASV use of ≥ 4 hours per night for ≥ 5 days a week) exhibited significant improvements in AHI, NYHA class functional class, and in cardiopulmonary exercise testing (all p-values < .01).

Yoshihisa et al. (2013) conducted a small (N = 36) experimental study of adults with HFpEF and SDB (AHI > 15) randomly assigned to either an ASV group or a medication only group.⁴¹ Compared to the medication only group, the ASV group had significant improvements (all p values < .05) to AHI, NYHA functional class, and cardiac function after 6 months.⁴¹ A Kaplan-Meier analysis found that persons in the ASV group had a lower event-free rate for cardiac events (i.e., cardiac deaths or hospitalizations; p < .05) after one year. The Cox proportional hazard analysis revealed that only ASV was an independent predictor of cardiac events in HFpEF and SDB (HF = 0.576, 95% CI 0.183, 0.799; p = .0164).

Yoshihisa et al. (2015) reported a larger observational study (N = 109) of persons with HFpEF and moderate to severe SDB.⁴² Treatment devices were distributed according to the dominant type of sleep apnea: CPAP for OSA- patients and ASV to central sleep apnea-Cheyne Stokes Respiration each representing approximately 50% of the sample. While all participants were offered treatment, comparisons were made between those who accepted the positive airway pressure treatment (n=31) and those participants who refused or were intolerant of the device (n=78). The intervention group exhibited statistically significant improvements (i.e., all p values ≤ 0.05) in NYHA functional class, cardiac function, and cardiopulmonary exercise tests after 6 months, compared to the control group.⁴² Long-term cardiac and all-cause mortality were lower in the positive airway pressure group than in the control group, as demonstrated by Kaplan-Meier analysis (p < 0.05).

Arikawa et al. (2016) was a small observational study (N = 58) of patients with new-onset HFpEF (66 ± 15 years) to examine the effect of treatment of SDB on plasma brain natriuretic peptide (BNP) levels, a marker for heart failure severity, during a 36 month follow-up.⁴³ Presence of SDB was determined after enrollment into the study and all patients with OSA (n = 39) were treated with CPAP, as well as with lifestyle modifications. Comparisons were made between those HFpEF patients without OSA (n = 19) and those HFpEF patients with OSA. Both groups received usual care for heart failure management. BNP levels reduced in both groups over time, however, the reductions were less evident in the OSA group at each time point. The OSA group had higher BNP levels at six months (221 vs 76 pg/ml, 12 months (123 vs 52 pg/ml, and at 36 months (115 vs 56 pg/ml); all p-values < 0.05.⁴³

Most recently, O’Connor et al. (2017) described results from the Cardiovascular Improvements with Minute Ventilation-targeted ASV Therapy in Heart Failure (CAT-HF) study,⁴⁴ a randomized controlled trial (RCT) designed to evaluate the effect of ASV in hospitalized patients with heart failure. The primary endpoint was a composite global rank endpoint of survival, CV hospitalization, and functional capacity measured by 6-minute walk distance. The study proposed randomized 215 subjects with heart failure; however only 126 participants were randomized in this study before it was the trial was terminated because of release of data suggesting higher mortality in the SERVE-HF study. Preliminary data from the CAT-HF study suggested that SDB was effectively and safely treated by ASV (AHI decreased from 35.7/h to 2.1/h) in a HF population that included both HFrEF and HFpEF. Data collected prior to study termination did not find that use of ASV resulted in an improvement in cardiac status. A limitation of this study was that average adherence to ASV was very low (2.7h/night), limiting the ability to determine the effect of ASV on cardiac function. Despite this, a subgroup analysis of participants with HFpEF (n = 24) showed a positive effect of ASV on the composite global rank score (hierarchy of death, cardiovascular hospitalizations, and percent changes in 6-min walk distance) at 6 months (p = 0.036).⁴⁴

Effect of positive airway pressure treatment pre-and post-intervention on NYHA functional class

Although the type of intervention device varied, the NYHA functional class was a common metric used in four of the studies.^39–42 Overall, NYHA functional class improved from pre-intervention to post-intervention with the use of positive airway pressure treatment (see Table 2). More specifically, treatment of SDB with positive airway pressure was associated with moderate to large improvements in the NYHA functional class with effect sizes that ranged from −.67 to −1.6.

Table 2.

Effect of positive airway pressure on the New York Heart Association functional class.

		M	SD	Mean Difference	Effect Size	95% CI
	Group					Lower	Upper
Bitter	Pre	2.4	0.6	−0.40	−0.67	−1.12	−0.21
	Post	2.0	0.8
Oldenburg	Pre	2.4	0.5	−0.40	−0.80	−1.40	−0.20
	Post	2.0	0.4
Yoshihisa, 2013	Pre	2.3	0.5	−0.80	−1.60	−2.35	−0.85
	Post	1.5	0.5
Yoshihisa, 2015	Pre	2.4	0.5	−0.80	−1.60	−2.07	−1.13
	Post	1.6	0.5

Note. Standardized effect size of treatment on the New York Health Association (NYHA) functional class for each study was based on the change score between pre-intervention and post-intervention divided by the standard deviation at pre-intervention (Becker, 1988). M = mean; SD = standard deviation; CI = confidence interval.

Discussion

This review of studies suggests that treatment of SDB in patients with HFpEF may result in improved nighttime sleep and daytime function that included statistically significant improvements in cardiac function, especially among those participants who were adherent to positive airway pressure treatment (e.g. CPAP or ASV) for more than 4 hours each night.^39,41,42 In addition, treatment of SDB with positive airway pressure resulted in improved exercise tolerance^39,40,42 and reduced cardiac complications resulting in hospitalizations and/or mortality.^41,42,44 The significant improvement in NYHA class, an outcome in four of the studies, suggests treatment of SDB results in better overall functional status in persons with HFpEF. However, these results should be interpreted with caution because of the conflicting evidence between previous studies in patients with HFrEF and because of the quality of the studies reviewed.

Quality Appraisal of Each Study

All of the studies had risks of bias that threatened their validity including non-experimental designs,^39,40,42,43 small sample sizes,⁴¹ use of non-equivalent comparison groups,^39,42,43 and attrition (40%).⁴⁰ The use of a one group pretest/posttest analysis in the study by Oldenburg et. al (2013) raises the possibility of rival hypotheses that account for the improvement in functional and cardiac status. None of the studies qualified as a superiority RCT that featured a comparative analysis between different types of positive airway pressure devices. Moreover, none of the studies reported use of a power analysis to determine sample size except for the study by O’Conner⁴⁴ however, this study was terminated when only half the needed sample was recruited.

Five studies had similar eligibility criteria: a diagnosis of HFpEF and moderate-to-severe SDB (mean apnea + hypopnea per hour sleep [AHI] ≥ 15/hour)^39–42,44 while one study⁴³ included patients with HFpEF without SDB as an part of the inclusion criteria. A standard HFpEF diagnosis according to European Society of Cardiology guidelines was used in the two studies that were conducted in Germany^39,40, while the three studies from Japan^41–43 utilized the Framingham Criteria for Congestive Heart Failure with a LVEF ≥ 50%. The study by O’Connor used an LVEF > 45% to identify HFpEF in hospitalized patients admitted for acute decompensated heart failure. In all studies, SDB was diagnosed by an objective sleep study device using modified American Academy of Sleep Medicine guidelines. In-laboratory polysomnography was used in all but two studies; the study by O’Connor and colleagues⁴⁴ utilized a four-channel type III home sleep study device while the study by Arikawa and colleagues used a two stage screening process with first with evaluation of nocturnal desaturations with pulse oximetry and then use of portable polysomnography with EEG monitoring to assess for OSA while the patient was hospitalized with new-onset heart failure.

Additional threats to validity include the low number of female participants and the lack of data on positive airway pressure device adherence. Positive airway pressure device adherence is especially relevant in Arikawa et al. who concluded that “appropriately” treated OSA may worsen long-term cardiac function in persons with HFpEF, although CPAP adherence was not reported.⁴³ There was no one consistent intervention (CPAP, ASV and tri-level ASV) for all of the studies precluding direct comparison of the impact of treatment on outcomes. Even though all of the studies used cardiac measures as their main outcomes, none had patient-oriented outcomes as endpoints other than the New York Heart Association (NYHA) functional class in four of the studies.^39–42 The NYHA functional class is a useful clinical tool for determining the eligibility of patients for certain medical services and informing therapeutic management. However, the NYHA classification does not fully explain patient-oriented outcomes or health-related quality of life.

Limitations of the review

While there are numerous studies that examine treatment of SDB in unspecified heart failure or in samples with HFrEF, a limitation of this review was the limited number of databased studies on the effect of treatment of SDB in persons with HFpEF and that only two studies had an experimental design. In addition, studies that were non-English, in the grey-literature (i.e. abstracts to conferences), or not published because of publication bias to non-statistically significant findings were not accessed.

Conclusions and Recommendations

In conclusion, the five out of six studies reviewed suggest that treatment of SDB in older adults with HFpEF may have positive effects on cardiac function that may result in decreased cardiac events and hospitalizations. However, minimal evidence exists from randomized clinical trials with HFpEF samples to guide practice regarding appropriate treatment of SDB within this population. Effective management of SDB is especially important because HFpEF is rapidly becoming the predominant form of heart failure among older adults, and complex SDB is highly prevalent in this population.²⁶

Effective treatment of SDB is difficult concerning the choice of which type of positive airway pressure device to use. The different type of device appears to have different effectiveness and consequences depending on type of heart failure and type of SDB. According to the American Academy of Sleep Medicine (AASM), CPAP is indicated for treatment of obstructive apneas in patients with either HFrEF or HFpEF. Treatment of central apneas or Cheyne-Stokes respirations requires use of Bi-PAP or ASV. The updated 2016 AASM guidelines do not recommend ASV therapy in persons with HFrEF (LVEF ≤ 45%) and moderate to severe SDB with predominantly central sleep apneas. However, the guidelines state that ASV can be used in those persons with HFpEF, mild sleep-disordered breathing, or those with OSA predominant SDB (including both HFrEF and HFpEF).²⁴

Until the mechanism behind the increased mortality observed in the SERVE-HF trial is determined, treatment with ASV in patients with HFpEF and predominant central sleep apnea needs to be closely monitored. Additional well designed clinical trials are needed with samples that accurately reflect this demographic group and that have adequate sample sizes to determine the effectiveness of treatment of SDB in persons with HFpEF. Particularly, this can help determine whether or not treatment decisions need to be made according to the type of heart failure (i.e., HFrEF vs. HFpEF) or on the mix of obstructive to central events in the individual patient. Finally, further study is needed among older adults with HFpEF to examine more holistically the impact of treatment of SDB on patient-oriented outcomes such as self-management of this chronic disease and quality of life.

Nursing Implications

Sleep disordered breathing is commonly observed in older patients with HFpEF but is not part of the routine clinical evaluation and management of heart failure.⁴⁵ Therefore, SDB remains untreated in many of these patients.⁴⁵ The adverse physiological and functional effects from SDB on health-related quality of life illustrates the importance of evaluation not only for the presence of but also the risk of SDB in the heart failure population. To date, there is no approved therapy to reduce hospitalization for HFpEF⁴⁶ and evidence based interventions with proven efficacy failed to demonstrate mortality benefits.^2,47 If SDB in patients with HFpEF is an important “hidden variable” for impaired symptom management, it may be a crucial therapeutic target and may improve treatment regimens and health-related quality of life. In addressing SDB in this older adult population with HFpEF, nurses can play a vital role in cardiovascular risk reduction and symptom management.

Note

AHI

apnea/hypopnea index

ASV

adaptive servo-ventilation

CPAP

continuous positive airway pressure

HFpEF

heart failure with preserved ejection fraction

HFrEF

heart failure with reduced ejection fraction

HRQoL

health related quality of life

LVEF

left ventricular ejection fraction

OSA

obstructive sleep apnea

PAP

positive airway pressure

SDB

sleep disordered breathing