Shahrokh Javaheri

Impact of Cardiovascular Diseases

An estimated 81 million Americans have one or more cardiovascular diseases. The prevalence of various cardiovascular diseases, the associated mortality, and the economic burden are presented in Table 14.1-1.

Since 1900, with the exception of the 1918 influenza pandemic, cardiovascular diseases have been the No. 1 cause of death in the United States. In 2007, cardiovascular disease was responsible for 34% of all deaths, or 1 of every 2.9 deaths in the United States. On the basis of 2007 mortality rate data, an estimated 2200 adult Americans die of cardiovascular disease each day, or an average of 1 death every 39 seconds.

Effect of Sleep Disorders on Cardiovascular Physiology

One of the most significant advances in clinical medicine has been our understanding that OSA and/or hypopnea, hitherto referred to collectively as OSA, may be either a cause of or may contribute to the progression, morbidity, and mortality of various cardiovascular diseases. OSA is characterized by recurrent upper airway occlusion that is either complete (apnea) or partial (hypopnea). The acute consequences of OSA include recurrent episodes of hypoxemia/reoxygenation; increases and decreases in Pco₂; and large, negative swings in intrathoracic pressure. Figure 14.1-1 shows an example of OSA resulting in desaturation followed by reoxygenation with resumption of breathing. An arousal occurs at the termination of the apnea. In this patient, an esophageal balloon was inserted to measure intrathoracic pressure. Note the progressively increasing negative swings in intrathoracic pressure as a result of diaphragmatic contraction against the closed upper airway. The negative pressure swings are also reflected in the surrounding cardiac chambers, pulmonary vascular bed and interstitium, and the intrathoracic aorta, increasing the transmural pressure of the chambers and blood vessels. Increased transmural pressure of the atria may contribute to atrial premature excitation and other arrhythmias, such as atrial fibrillation. Increased left ventricular transmural pressure increases its afterload and oxygen consumption. Increased aortic transmural pressure increases the propensity for aortic dilation and dissection, whereas increased negative interstitial pressure increases the propensity for developing pulmonary edema.

Increasing evidence suggests that through these acute mechanisms, OSA leads to activation of redox-sensitive genes, oxidative stress, inflammatory processes, sympathetic overactivity, and hypercoagulability. All of these can contribute to endothelial dysfunction (Fig. 14.1-2), which is the underlying pathophysiologic mechanism that mediates a variety of cardiovascular disorders, including hypertension, atherosclerosis (coronary artery disease as well as cerebrovascular disease), pulmonary hypertension, right and left ventricular systolic and diastolic dysfunction, stroke, and even sudden death (Fig. 14.1-3).

At times apneas can be very prolonged. This results in severe desaturation, as depicted in Figure 14.1-4.

Obstructive Sleep Apnea as a Cause of Cardiac Arrhythmias

Disorders of cardiac rhythms can occur in patients with OSA. These dysrhythmias may include bradycardia, sinus pause (Fig. 14.1-5), heart block, ventricular ectopy and tachycardia (Fig. 14.1-6), and atrial fibrillation. Arrhythmias are apt to occur in patients with severe OSA and in those with cardiac disease. The mechanisms that mediate these dysrhythmias relate to the three major consequences of OSA: 1) altered blood gases that result in hypoxemia, hypercapnic acidosis, and hypocapnic alkalosis; 2) changes in autonomic tone; and 3) negative swings in intrathoracic pressure, which may distend the atria and ventricles. Particularly in the presence of coronary artery disease, the threshold for developing arrhythmias may be low.

In patients with OSA, nocturnal arrhythmias may contribute to the higher nighttime prevalence of sudden death. In the general population, sudden cardiac death, presumably in part the result of ventricular arrhythmias and myocardial infarction, shows a day-night variation with preponderance in the early morning hours, from 6:00 am to noon. In contrast, in patients with sleep apnea, the peak in sudden death occurs during sleeping hours between midnight and 6:00 am, and the severity of sleep apnea correlates with the risk of nocturnal sudden death. The clinical implication is that the recognition and treatment of OSA may eliminate arrhythmias and improve morbidity and mortality in this population.

Obstructive Sleep Apnea as a Cause of Coronary Artery Disease

Through the mechanisms of endothelial dysfunction, which were reviewed earlier (see Figs. 14.1-2 and 14.1-3), OSA may either be a cause of or may contribute to coronary artery disease or its progression. Evidence is mounting, from both population and sleep clinic cohort studies, that OSA causes or contributes to the progression of coronary artery disease and eventually to elevated cardiovascular death and all-cause mortality.

OSA has been implicated as a cause of nocturnal ST-segment depression and angina pectoris. These conditions may be more prevalent in patients with established coronary artery disease. We recommend polysomnography for the diagnosis of OSA and continuous positive airway pressure (CPAP) titration, if appropriate, for all patients with nocturnal angina, which should be eliminated with therapy.

In a retrospective sleep laboratory cohort study, Peker and colleagues¹ reported incident coronary artery disease in almost 25% of untreated OSA patients during a 7-year follow-up. The corresponding numbers in treated sleep apnea patients and nonapneic snorers were 4% and 6%, respectively. In an observational sleep laboratory cohort study of 1300 OSA patients and healthy study participants from Spain, Marin and colleagues² observed that during a mean follow-up period of approximately 10 years, a 3 to 4 times higher incidence of fatal and nonfatal cardiovascular events was observed in the patients with severe OSA, defined as an apnea-hypopnea index (AHI) of 30 or more per hour, compared with simple snorers. Furthermore, in 372 patients with severe OSA who were compliant with CPAP, cardiovascular event rates were similar to those in nonapneic snorers.

The Wisconsin Sleep Cohort Study³ is an 18-year mortality follow-up study (mean observation, approximately 14 years) conducted on a cohort sample of 1522 participants aged 30 to 60 years at baseline. The adjusted hazard ratio (HR) for all-cause mortality with severe versus no sleep-disordered breathing (SDB) was 3 (95% confidence interval [CI] = 1.4 to 6.3; Fig. 14.1-7). Importantly, after excluding the 126 participants who had used CPAP therapy, the association for all-cause mortality with severe versus no SDB became stronger, and the adjusted heart HR increased to 3.8 (CI = 1.6 to 9) for all-cause mortality and to 5.2 (CI = 1.4 to 19.2) for cardiovascular mortality (see Fig. 14.1-7).

Thus severe OSA is a major risk factor for all-cause mortality and cardiovascular disease, and effective treatment of OSA with CPAP improves survival. As reviewed in the next section, the improvement in survival is due at least in part to improvement in hypertension, which has been shown to decrease incident stroke and MI.

Obstructive Sleep Apnea as a Cause of Systemic Hypertension

OSA is an independent risk factor for systemic arterial hypertension. Based on biologic plausibility (see Figs. 14.1-2 and 14.1-3), animal studies, longitudinal epidemiologic human studies that account for confounding factors, and therapeutic trials, evidence has evolved that indicates OSA is an important cause of hypertension. In the Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure, OSA is recognized as the first identifiable cause of hypertension.⁴

In a longitudinal study, Peppard and associates⁵ reported that the incidence of newly diagnosed arterial hypertension in a 4-year follow-up period increased as the AHI increased—a dose-dependent relationship. The increase in the risk for developing hypertension was doubled in those with an AHI of 5 to 15 events per hour, and a threefold increase in risk was seen with an AHI greater than 15 events per hour compared with study participants without a sleep disorder at baseline.

From a clinical point of view, it is important to recognize that approximately half of the 70 million Americans who have systemic hypertension may have OSA. Furthermore, it is important to note that effective treatment of OSA has been shown to decrease systemic blood pressure. These reports are cited (Fig. 14.1-8) because of the power of the studies, the presence of a placebo control group that received sham CPAP, and the fact that the patients had hypertension to begin with. Collectively these studies show a small but significant decrease in hypertension in CPAP-treated patients (see Fig. 14.1-8), and hypertension improved the most in patients with the most severe sleep apnea and desaturation index and in those who used CPAP the most. Therefore it is safe to conclude that adequate therapy of OSA with CPAP is critical for reduction in blood pressure.

In summary, the results of the aforementioned studies indicate that both adequate control of OSA and adherence to CPAP are required elements for OSA-induced hypertension to improve with CPAP therapy. Improvement in hypertension plays a key role in decreasing incident stroke and myocardial infarction, and this contributes to increased survival.

Heart Failure

Epidemiology

More than 5 million Americans have HF. The prevalence of sleep apnea, both central and obstructive, remains high in the era of β-blockers. Several studies have shown a high prevalence of both OSA and CSA in patients with HF, both with left ventricular systolic function that is reduced (HF with reduced ejection fraction [HFrEF]) or preserved (HF with preserved ejection fraction [HFpEF]).

Table 14.1-2 shows the prevalence of CSA and OSA in recent studies of patients with HFrEF. In the prevalence of these two forms of sleep apnea, reported variations have been considerable; this is due to a number of issues, such as the differentiation of obstructive from central hypopneas and the enrollment of obese patients, which increases the prevalence of OSA. Combining the results of the most recent series in the literature, as shown in Table 14.1-2, and using an AHI of 15 events per hour of sleep as the threshold, 31% of 1250 patients have CSA and 21% have OSA (Fig. 14.1-9).

The largest study of patients with HFpEF involved 244 consecutive patients who were well characterized by right heart catheterization and echocardiography. Bitter and colleagues¹⁴ reported that 48% had an AHI of 15 or more events per hour, 25% with OSA and 23% with CSA (see Fig. 14.1-9).

The hallmark of OSA in patients with HF is similar to that in patients with OSA in the general population, in that such patients are obese and have habitual snoring. In contrast, patients with CSA are normally thinner than patients with OSA, and the prevalence of habitual snoring is much less (Fig. 14.1-10). For these reasons, the suspicion for the presence of CSA in cardiology and in primary care clinics is low, so these patients often go undiagnosed. The risk factors for CSA in patients with HFrEF include a high New York Heart Association class, the presence of atrial fibrillation, more than 30 premature ventricular excitations per hour of sleep, the presence of one or more couplets per hour of sleep, the presence of one or more ventricular tachycardias per hour of sleep, a low steady-state awake Pco₂ of less than 36 mm Hg, and a left ventricular ejection fraction of less than 20% (Fig. 14.1-11). Meanwhile, it is emphasized that in our studies⁶ the prevalence of excessive daytime sleepiness was similar in patients with systolic heart failure (SHF) with or without CSA and/or OSA (see Fig. 14.1-10). This further contributes to the underrecognition of sleep apnea in patients with SHF.