Use of Continuous Positive Airway Pressure for Sleep Apnea in the Treatment of Hypertension

Abstract

Purpose of review

Obstructive sleep apnea (OSA) and hypertension are highly prevalent and treatable conditions that often coexist and both contribute to increased cardiovascular risk. The ability of continuous positive airway pressure (CPAP) to improve blood pressure in hypertensive patients with OSA is debated. This review highlights findings from recent studies that have investigated the impact of CPAP on blood pressure in patients with OSA.

Recent findings

Comparing the results of various studies is complicated by important methodological differences among them. In hypertensive patients with OSA, treatment with CPAP improves blood pressure to a smaller degree than that derived from antihypertensive medication. Patients with more severe OSA and with greater adherence to CPAP are likely to gain the most benefit from the therapy.

Summary

CPAP should be used in combination with antihypertensive medications in hypertensive patients with OSA. CPAP has additional benefits of restoring nocturnal dipping and improving arterial stiffness, thus potentially influencing cardiovascular morbidity in these high-risk patients.

Introduction

Obstructive sleep apnea (OSA) is common among patients with hypertension and especially resistant hypertension, yet many remain undiagnosed [1, 2]. OSA is characterized by intermittent collapse of the upper airway during sleep, resulting in repeated episodes of hypopnea or apnea. These episodes can be prevented with the use of continuous positive airway pressure (CPAP) during sleep. OSA represents a major public health burden due to an increasing prevalence of the disorder [3] and an increased risk of hypertension [4], depression [5], motor vehicle accidents [6], cardiovascular (CV) events [7, 8], and all-cause mortality [9] associated with OSA. In this review, we describe the mechanisms for hypertension in OSA and the impact of OSA on cardiovascular disease (CVD) as well as discuss salient findings from recent studies that have examined the effect of CPAP on blood pressure (BP).

Epidemiology and Clinical Features

OSA is estimated to affect 2-4% of the general adult population, typically occurring in middle-aged, overweight men [10]. Symptoms of OSA may be subtle, such as fatigue or problems with memory and concentration. OSA should be suspected in an obese patient who snores as snoring is almost universal among patients with OSA, but only half of patients who snore for the majority of the night have OSA [11]. A detailed sleep history provided by the patient should be supplemented with information from the patient's sleep partner. Additional clinical characteristics that may be indicative of OSA include headaches, excessive daytime sleepiness, and apnea or choking spells during sleep. While any of these signs or symptoms may be clues of OSA, definitive diagnosis requires nocturnal polysomnography, which also grades the severity of OSA with an apnea-hypopnea index (AHI).

Mechanisms of hypertension and CV morbidity

There is an overwhelming body of evidence establishing a clear link between hypertension and OSA [4, 12-14]. Approximately one-third to one-half of patients with hypertension have OSA [1, 2], which represents a greater proportion than that of normotensive controls and is independent of body weight [1]. However, among hypertensive patients who subsequently undergo polysomnography testing as part of a study protocol, many have evidence of OSA that was previously undiagnosed [1, 15, 16]. Although causal inference is limited in cross-sectional studies, a causal link between OSA and hypertension is supported by the “doseresponse” relationship such that the severity of OSA is independently associated with the prevalence and degree of hypertension [4, 13, 14], and the prevalence of OSA in resistant hypertension may exceed 80% [15, 16]. This link is strengthened by the temporal relationship between the two entities that is observed in longitudinal studies. Patients with mild OSA have a two-fold increased odds of subsequently developing hypertension compared to those with AHI of 0 [17]. In fact, the Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure lists OSA as an identifiable cause of hypertension [18]. However, the potential effect of OSA on BP is relatively small and dependent on OSA severity. For a patient with AHI of 15, BP is predicted to be 3.6/1.8 mmHg higher than a patient with the same body mass index (BMI) and AHI of 0 [14].

As shown in Figure 1, there are numerous proposed mechanisms for the development of hypertension in OSA including increased sympathetic nervous system activity [19], inflammation [20], oxidative stress [21], endothelial dysfunction [22], vascular stiffness [23], hyperaldosteronism [24], and volume expansion [25, 26]. In addition, some of these mechanisms may be partially mediated or potentiated by obesity [27].

Figure 1. Proposed mechanisms contributing to blood pressure elevation in OSA.

Abbreviations: HTN, hypertension; SNS, sympathetic nervous system.

Despite OSA frequently occurring in combination with traditional CV risk factors, untreated severe OSA represents an independent risk factor for CV events [7, 8]. Twenty-four hour ambulatory BP monitoring of patients with OSA commonly demonstrates the absence of nocturnal dipping [28], which is an independent CV risk factor [29]. Also, sleep-related breathing disorders are associated with both a trial and ventricular arrhythmias [30], likely mediating an increased risk of sudden cardiac death observed in these patients [31]. Finally, OSA confers a 6-fold increased risk of all-cause mortality, independent of traditional CV risk factors [9]. These risks underscore the importance of identifying and treating patients with OSA.

Treatment

Given the suboptimal control rates of patients with hypertension [32], considerable CV morbidity and mortality associated with uncontrolled hypertension [33, 34], and an extremely high prevalence of OSA in those with resistant hypertension, the identification and treatment of patients with OSA represents a critical aspect of hypertension management. Lifestyle modifications, including dietary sodium restriction, weight loss, aerobic exercise, and limiting alcohol intake, should be recommended to all hypertensive patients with OSA. Diuretics seem to be a logical agent of choice to ameliorate fluid redistribution during sleep [35^■]; and a randomized trial comparing diuretics to calcium channel blocker in hypertensive patients with OSA is planned [36].

CPAP represents an OSA-targeted therapy to potentially reverse the pathophysiologic mechanisms responsible for hypertension. The acute and chronic effects of CPAP on sleep characteristics and BP were demonstrated in an uncontrolled, single-arm study of 11 patients with resistant hypertension and OSA [37]. Aside from significantly reducing AHI, 2 months of CPAP significantly reduced 24-hour BP by 10.5/5.7 mmHg (P<0.05) in the absence of medication changes or weight loss. This small case series provided proof of concept and laid important groundwork for future studies.

Numerous randomized controlled trials (RCTs) have demonstrated BP improvement with the use of CPAP [38-40], but this benefit has not been universal [41-43]. In 2007, a meta-analysis was published that pooled results from 12 RCTs conducted in patients with OSA [44]. Investigators reported a 1.69 mmHg incremental improvement in 24-hour mean BP attributed to CPAP. However, many of these studies were limited by short follow-up and small sample sizes, and the percentage of hypertensive patients that comprised the study cohorts ranged from 0-100%. Other potential explanations for conflicting results include differences in study design (e.g. how investigators handled patients with sub-optimal adherence, which patients served as controls), study populations (e.g. baseline BP, OSA severity), number of hours of CPAP use, co-interventions (e.g. use of antihypertensive medications), or outcome measures (e.g. ambulatory or office BP). A subsequent meta-analysis was published in 2012 [45], which pooled 28 RCTs (including 11 of the 12 trials included in the aforementioned study). This updated meta-analysis was also hampered by study heterogeneity. Nevertheless, investigators reported a statistically significant BP reduction with CPAP with weighted mean reductions in daytime and nocturnal BP of 2.58/2.01 mmHg and 4.09/1.85 mmHg, respectively. Since the latter meta-analysis was published, additional studies have been published and are described below.

Martínez-García et al. [46^■■] recently conducted a large, multicenter RCT among 194 patients in Spain with resistant hypertension and moderate to severe OSA. All antihypertensive medications were continued at their entry dosages; and after 12 weeks, those randomized to CPAP demonstrated a 3 mmHg greater reduction in 24-hour mean BP compared to those in the control group (P=0.02). CPAP had the additional benefit of restoring nocturnal dipping. The CPAP group had a significantly greater percentage of dippers at the end of the study (35.9% CPAP vs. 21.6% control; adjusted odds ratio 2.4; P=0.02); and CPAP reduced the percentage of risers (i.e. patients whose BP increases during sleep—a group at particular high risk for CV events [29]) compared to baseline, while the percentage of risers remained unchanged in the control group. The number of hours of CPAP use was correlated with the degree of BP improvement, which explains why the above differences were further amplified when CPAP adherence was considered in a per-protocol analysis.

Pedrosa et al. [47^■■] conducted a similar but longer RCT to evaluate the effect of CPAP on BP control among patients in Brazil with resistant hypertension and moderate to severe OSA. Patients were randomized to 6 months of CPAP plus medical therapy vs. the control arm of medical therapy alone, and medications were not adjusted during the trial. Of 40 patients who underwent randomization, 35 were followed for the full 6 months. As opposed to a BP rise in the control group, patients in the CPAP group demonstrated a daytime BP reduction of 6.5/4.5 mmHg during the study, while nocturnal BP increased slightly in both groups. Remarkably, this daytime BP reduction occurred despite higher baseline BMI and waist circumference in the intervention arm, and BMI did not change in either group during the course of the study. The lack of a beneficial effect of CPAP on nocturnal BP is both counterintuitive and contrary to findings from most other studies. Nevertheless, the studies by Martínez-García et al. and Pedrosa et al. support the use of CPAP even in patients with resistant hypertension and moderate-severe OSA.

In a prospective observational study that reinforced the BP-lowering effect of CPAP, Kartali et al. [48^■■] recruited 38 hypertensive patients with severe OSA to investigate the effect of CPAP on arterial stiffness, which is associated with OSA [23] and represents an independent CV risk factor [49]. Patients in the study were relatively free of other comorbidities, were not receiving antihypertensive medication, and were adherent to CPAP (≥5 hours per night of use). After 3 months of CPAP, 24-hour mean BP was significantly lowered from 141.5/87.8 to 133.5/83.0 mmHg without antihypertensive medication. Arterial stiffness was measured by pulse wave velocity (PWV) at baseline and at 3 months in patients with OSA and in normotensive controls without OSA. At baseline, PWV was higher in those with OSA than in controls (8.8 vs 7.2 m/s, P=0.003). However, at 3 months, PWV was lowered from 8.8 to 7.4 m/s in the CPAP group, such that PWV did not significantly differ between groups. In the absence of a suitable comparator group, it is unclear whether the improvement in PWV was derived from BP reduction, vascular effects of OSA and its treatment, or a combination of these. Supporting a BP-independent mechanism are that reduction in BP was not correlated with reduction in PWV in this study and that a prior study demonstrated that CPAP reduced PWV without affecting BP [50]. As the study by Kartali et al. was not an RCT and generalizability may be limited by the unique study cohort, its findings should be considered hypothesis-generating, and additional studies are needed to fully appreciate the effects of CPAP on arterial stiffness. A recent double-blind RCT was performed in 43 patients with OSA and compared CPAP to sham CPAP [51]. Though CPAP had no effect on PWV or endothelial function, participants in the trial were predominantly normotensive, and CPAP was used for a mean of only 3 hours nightly. Thus, these results are likely not generalizable to hypertensive patients with OSA as the latter patients would presumably gain more benefit and show greater change in vascular parameters from CPAP than their normotensive counterparts; and in fact, CPAP improved endothelial function and measures of vascular stiffness in prior studies that included patients with more severe OSA or hypertension [23, 52, 53]. These studies highlight the potential direct role that OSA exerts on vascular structure and function in promoting hypertension and increased CV risk as well as the amelioration of these effects with CPAP.

Another study that deserves mention is a recent prospective observational study of 91 hypertensive patients with moderate-severe OSA, which failed to demonstrate an incremental improvement in BP when CPAP-treated patients were compared to controls [54^■]. Patients who were offered but refused CPAP were allocated, not randomized, to the control group—a common method utilized in similar observational studies. After a mean follow-up of 3.1 years, both groups had a marked and statistically significant reduction in BP compared to baseline; but BP reduction did not differ between groups. BP decreased from 145/95 to 133/85 mmHg in the control group and from 148/96 to 133/84 mmHg in the CPAP group. CPAP did not reduce the number of antihypertensive medications used; and on the contrary, there was a small but statistically significant increase in the number of antihypertensive medications used that did not differ between groups. Medication doses were also increased during the follow-up period; but again, there was no difference between groups. Though patients tended to be adherent to CPAP with mean use of 5.9 hours nightly, adherence was assessed by patient self-report, which tends to be an overestimate [55]. Though this long-term study failed to show additional benefit of CPAP in lowering BP beyond the use of antihypertensive medications, the remarkable BP reduction in the control group warrants further attention. This BP improvement of 12/10 mmHg in the control group greatly exceeds that seen in other studies and occurred in the absence of weight loss or clinically meaningful increase in the number of antihypertensive medications.

A common theme of studies in patients with OSA is that many either refuse CPAP or have suboptimal adherence, thus limiting clinical effectiveness. An alternative to CPAP is a mandibular advancement device (MAD), which is designed to reposition the tongue and/or lower jaw to increase the airway lumen. The efficacy of MAD was tested in a short-term randomized crossover trial that compared the two therapies, each being used for 1 month, in a group of Australian patients with OSA [56^■■]. Patients preferred and were more adherent to MAD, but CPAP-treated patients had greater improvement in AHI. In the 42% of patients who were hypertensive at baseline, there was a statistically significant reduction from baseline in 24-hour BP of ∼3/2 mmHg that did not differ between treatment groups. This study demonstrated the non-inferiority of MAD in terms of BP change as well as certain advantages of this therapy. While potentially promising, studies of longer duration with larger cohorts of hypertensive patients are necessary to fully understand the role of MAD in the treatment of OSA.

In summary, the BP reduction seen with CPAP is modest, necessitating its use adjunctively with antihypertensive medications. Baseline OSA severity, baseline BP, and duration of CPAP use are likely important contributing factors that influence the therapy's clinical effectiveness such that a more robust BP response to CPAP is expected in adherent patients with more severe hypertension. Given that mean CPAP use in clinical trials tends to be only 4-5 hours nightly, leaving a considerable portion of sleep untreated, we believe the full impact of CPAP may be underappreciated; and the lack of substantial BP reduction with CPAP should not temper enthusiasm for its use as there are many potential benefits to CPAP (see Table 1). Most importantly, multiple studies demonstrate a reduction of CV risk with CPAP use [7, 62].

Table 1. The potential effects of CPAP.

Decreased:

Daytime sleepiness [57]

Blood pressure

Sympathetic nervous system activity [58]

Inflammation [59]

Plasma aldosterone [60]

Vascular stiffness

Structural cardiac abnormalities [61]

Cardiovascular risk [62]

Increased:

Quality of life [57]

Endothelial function [63]

Nocturnal dipping

Conclusion

OSA has a strong link with hypertension and likely has a causal influence on BP. Treatment with CPAP reverses some of the pathophysiologic mechanisms responsible for hypertension and has a modest effect on lowering BP. While a small change in BP may be relatively unimpressive at the individual patient level, this change could have substantial implications for health outcomes at the population level. CPAP has additional advantages of restoring nocturnal dipping, improving arterial stiffness, and reducing CV risk. As patients with OSA report decreased quality of life and are at increased risk of CV events and death, an OSA targeted therapy with CPAP should be an integral part of treatment to reduce the morbidity burden associated with OSA.

Key Points.

• The overall BP response to treatment with CPAP is modest.

• The more severe the hypertension and OSA, the more robust the BP response to CPAP can be expected.

• The potential of CPAP to improve arterial stiffness and restore normal dipping patterns may have far-reaching consequences on reduction of CV risk.