The review included original articles and meta-analyses from 2015 to 2021 that addressed the relationship between OSA and sudden death and/or tachyarrhythmias, published in English, Spanish, or Portuguese.

Studies whose central subject was not the relationship between sleep apnea and the development of sudden death and/or tachyarrhythmias were excluded from the process. Studies involving animals and articles whose full version was not available were also excluded.

The MEDLINE® via PubMed® and SciELO databases were searched for articles published until March 4, 2021 using the descriptors “sudden death” and “sleep apnea” and “tachyarrhythmias” and “sleep apnea”. Original articles and meta-analyses published from 2015 to 2021 in the national and international literature were selected. The descriptors used are listed in the Medical Subject Headings terms (MeSH) of the US National Library of Medicine (NLM).

The search strategy used for the MEDLINE® via PubMed® database was as follows:

• 1.

  (“tachyarrhythmias”[All Fields] AND “sleep apnea”[All Fields])

• 2.

  (“tachyarrhythmias”[All Fields] AND “sleep apnea”[All Fields]) AND (2015:2021[pdat])

• 3.

  (“sudden death”[All Fields] AND “sleep apnea”[All Fields])

• 4.

  (“sudden death”[All Fields] AND “sleep apnea”[All Fields]) AND (2015:2021[pdat])

The methodological selection was descriptive, independent, and performed by two reviewers. Initially, the titles and abstracts of the articles were evaluated according to the eligibility and exclusion criteria. Then, the selected articles were read in full for a new evaluation according to the inclusion and exclusion criteria.

The selected articles were analyzed descriptively, and two independent reviewers performed data extraction by listing the main results and conclusions that addressed the relationship between OSA and the development of sudden death and/or tachyarrhythmias.

Sleep apnea is characterized by repetitive ventilation interruptions longer than or equal to 10 seconds during sleep and is associated with an ongoing ventilatory effort by the affected subject.26 In turn, hypopnea is characterized by an airflow reduction for ten or more seconds associated with a drop in oxyhemoglobin saturation or electroencephalographic arousal.27

Apneas can be classified as central (CSA) and obstructive (OSA). OSA occurs due to upper airway occlusion. In contrast, CSA is related to changes in the nervous system responsible for generating the breathing rhythm.27 Sleep apnea can also be classified according to the severity of the disorder using the AHI, representing the number of apnea and hypopnea episodes per hour of sleep.28 OSA is considered mild with an AHI from 5 to 15/h, moderate with an AHI from 15 to 30, and severe with an AHI >30.29

The activity of the genioglossus muscle decreases during sleep, causing the tongue to fall backward and obstruct the airways of susceptible individuals. Obesity is the main predisposing factor for sleep apnea due to direct mechanical effects that result in reduced caudal traction of the upper airway, compromising its patency due to the deposition of fat within the upper airway and lung volume reduction.27

The basic mechanism of OSA pathophysiology is related to several factors, e.g., age and sex, and it has shown to be more prevalent among older men. Posture and gravity also contribute to the development of this disorder, especially due to the decubitus position during sleep. Finally, genetic, functional, and anatomical factors are also important causes of OSA due to morphological abnormalities in the oral cavity and the muscles of this region.30

Sleep-disordered breathing is related to several pathological mechanisms such as changes in intrathoracic pressure, intermittent hypoxia and injury, sympathetic nervous system activation, sleep reduction and fragmentation, cardiac remodeling, metabolic dysregulation, and endothelial dysfunction (Figure 2).12,31,32

Figure 2.

OSA arrhythmogenic mechanisms. ANS: autonomic nervous system; RAAS: renin-angiotensin-aldosterone system; ROS: reactive oxygen species.

(0.23MB).

In addition to endothelial dysfunction, the inflammatory mechanisms unleashed by OSA can also increase atherosclerosis, contributing to the progression of cardiovascular diseases. It has also been described that OSA patients without other risk factors for arteriosclerosis showed increased intima-media thickness compared to patients without OSA, which is related to nocturnal hypoxia.33

Sleep-stage dependent changes in respiration, heart rate, and blood pressure normally occur during sleep. However, in OSA and CSA cases, these variations are more correlated with the severity of the disorder and the duration of airflow restriction.34 Physiologically, non-REM sleep (rapid eye movement) occupies most of the sleep architecture and is characterized by the stability of the autonomic nervous system, with heightened parasympathetic tone and lower activity of the sympathetic tone,32 keeping the heart at a low rate and maintaining low blood pressure levels.27 However, in individuals with OSA, chemoreceptors and baroreceptors are activated in response to the collapse of the upper airways and reduced partial pressure of oxygen. These receptors stimulate catecholamine release by the sympathetic autonomic nervous system, resulting in increased blood pressure and heart overload,12 implying OSA as a possible factor in the genesis of arterial hypertension. Therefore, this apnea-induced sympathetic response to hypoxia and hypercapnia could explain the increased risk of cardiovascular diseases and death among OSA patients.34

Patients with OSA have an increased risk of developing cardiac arrhythmias, the most prevalent of which is atrial fibrillation (AF) arrhythmia.12,35 The European Society of Cardiology highlights OSA as a risk factor for AF.36,37 Several pathophysiological mechanisms unleashed by OSA have been related to arrhythmogenesis, e.g., catecholamine release in the ischemic myocardium.11 In the long run, the hypertension generated, as well as endothelial dysfunction and ventricular remodeling caused by exposure to the sympathetic tone, would also be implied.32

Changes in intrathoracic pressure are also important pre-arrhythmogenic effects. Increased intrathoracic pressure and respiratory effort stimulate cardiac mechanoreceptors that promote changes in cardiac geometry. By themselves, these changes can generate arrhythmias by electrical feedback mechanisms.12 It is worth noting that cardiac remodeling represents the formation of fibrous tracts that act as obstacles to electrical conduction. Therefore, this constant electrical and anatomical remodeling is a perpetuating factor of arrhythmias.38

Heart rate variability could represent the sympathetic activity resulting from apnea or hypopnea events. However, an observational study conducted by Xie et al. under the hypothesis that the prevalence of cardiac arrhythmias is associated with OSA and that heart rate variability parameters could be effective markers for OSA screening, found no difference in the parameters between OSA patients and the control group, suggesting that the ECG alone is not an effective screening method, even though severe obstructive sleep apnea is an important risk factor for arrhythmias.39

With regard to the differences between the types of sleep apnea in the assessment of the circadian distribution of arrhythmias in patients with sleep-disordered breathing, CSA predominated in the studied population compared to OSA, and the association of heart failure with reduced ejection fraction (HFrEF) and ventricular arrhythmias was more frequent among CSA patients. With regard to the differences in the underlying mechanisms that drive and promote arrhythmias caused by OSA in relation to CSA, arrhythmias in OSA patients occur during apnea moments due to oxygen desaturation, increased intrathoracic pressure, and myocardial wall stress. Conversely, in CSA, hypopneas cause intrathoracic pressure swings, pH displacements, changes in oxygen and carbon dioxide partial pressures, and increased circulatory delay. Moreover, differences in the time profiles of these arrhythmias may occur because OSA patients experience such events during sleep, whereas CSA patients might experience them while awake.25

Atrial fibrillation, a type of supraventricular arrhythmia, is more common among CSA patients, corroborating the hypothesis that different arrhythmias may be linked to different types of sleep apnea-related stresses (OSA or CSA).34 In experiments with pig models to assess the differences between OSA and CSA on the origin of ventricular repolarization changes, it was observed that intrathoracic pressure swings during obstructive apneas are associated with prolonged RT and Tp-e intervals, which was not significantly observed in central apneas.40

Although more common in individuals with CSA, the relationship between OSA and atrial fibrillation is well known in the scientific community, with OSA being an independent risk factor for atrial fibrillation. Furthermore, the increased risk was directly correlated with the severity of apnea and the recurrence and worsening of the arrhythmia prognosis.41

The relationship between OSA and complex ventricular ectopy has also been described, tending to increase with OSA severity. High OSA prevalence was found among patients with Brugada syndrome, a condition known to be potentially fatal, raising the possibility that the association of the two disorders could lead to increased mortality in this group of patients.34

Prolonged QT, QTc, and Tp-e intervals could imply changes in ventricular repolarization, causing ventricular arrhythmias. A significant prolongation of the QTc interval was reported by Sökmen et al. during apnea, which was maintained in the post-apnea hyperventilation period. The Tp-e interval, whose increase was observed in the apnea periods of this study, is a recent parameter of cardiac repolarization whose prolongation could be associated with ventricular arrhythmias and SCD. On the other hand, Tp-e/QT is a ventricular repolarization index that could be a more stable and, therefore, more accurate predictor of ventricular arrhythmias.20

Experiments that studied the association between OSA and Long QT syndrome (LQTS) (Figure 3) evidenced that the association of the two diseases increased respiratory disturbances, AHI, and the arousal index. Furthermore, QTc increased both during sleep and while awake among these patients, with an association between apnea severity and QT prolongation corrected for heart rate, constituting an important biomarker of sudden death. The study suggests that OSA treatment in LQTS patients could reduce the QTc and thus reduce the risk of sudden death.42

Figure 3.

Predisposing factors for arrhythmias associated with OSA. Based on electrocardiographic tracings of Hampton, 2013.43

(0.49MB).

Regarding ventricular arrhythmias, Salama et al. studied the relationship between OSA and increased arrhythmias and concluded that ventricular tachycardia and ventricular fibrillation are more prevalent among OSA patients than among patients without an OSA diagnosis.17

Considering the relationship between OSA and the increased risk of ventricular arrhythmias and conduction disturbances, it has been reported that OSA patients showed a combination of changes in the QRS complex, e.g., a leftward shift of the electrical axis and low QRS voltage, suggesting a deterioration of the depolarization sequence possibly related to electrical remodeling. Furthermore, these changes become more important with the severity of obstructive apnea.5

In experiments with rats to investigate the effects of intermittent hypoxia on the development and severity of myocardial ischemia-related ventricular arrhythmias, it was possible to observe an increase in the incidence of ischemic arrhythmias, particularly ventricular fibrillation, in addition to QTc prolongation. These findings highlight the role of intermittent hypoxia as an important risk factor for SCD through mechanisms such as myocardial ischemia, sympathetic activation, and alterations in ventricular repolarization.11

In addition to the already reported ventricular arrhythmias and atrial fibrillation, apnea and hypoxia induced by OSA could lead to vagal activation and cause bradyarrhythmia (Figure 4)43 due to the cardiac vagal activation resulting from apneic and hypoxic episodes even if the increase in the sympathetic tone, also resulting from hypoxia, is implied in the origin of tachyarrhythmias. There would be a large distribution of the R-R interval with the presence of contrary stimuli resulting from vagal and sympathetic activation. Yajima et al. reported the case of a patient with sinus arrest and atrioventricular block accompanied by OSA. Since the patient could not be treated with pacemaker therapy, conventional OSA treatment by continuous positive airway pressure (CPAP) was shown to be effective, decreasing the plasma brain natriuretic peptide levels and highlighting the reduction in cardiac overload caused by bradyarrhythmia and the change in the distribution of the R-R interval, which was close to normal, demonstrating the balance of the autonomic nervous system and suggesting that CPAP therapy could effectively control cardiac arrhythmias in OSA patients.44

Figure 4.

Arrhythmias caused by OSA. Based on electrocardiographic tracings of Hampton, 2013.43

(0.49MB).

Sudden death is a nontraumatic, unexpected, fatal event that occurs within one hour of the onset of the first symptoms in previously healthy individuals. The conditions established by the European Society of Cardiology for sudden death characterization are the presence of a congenital or acquired potentially fatal cardiac condition, autopsy identifying a cardiac or vascular anomaly as the probable cause of the event, and no obvious extracardiac causes, suggesting an arrhythmic event as the likely cause of death. These parameters highlight the lethal potential of cardiac arrhythmias and the consequent implication of OSA.12

Several OSA-stimulated biological pathways lead to cardiac electrical abnormalities and increase the susceptibility to arrhythmias, including ventricular arrhythmias. Ventricular fibrillation is among ventricular arrhythmias and is the main arrhythmia associated with sudden death. It is also known that sudden death occurs predominantly during sleeping hours in individuals with OSA.45 Increased sympathetic tone, intermittent hypoxia, and intrathoracic pressure swings are described as the main changes that explain the association between OSA and increased arrhythmogenesis propensity, also explaining the occurrence of nocturnal arrhythmias in patients with OSA.45

In addition to the apnea/hypopnea index, nocturnal hypoxia constitutes an independent risk factor for SCD.46 Furthermore, there is a relationship between the oxygen saturation threshold <93% and a 2.9 times higher risk of SCD.12 Chronic exposure to hypoxia alters the left ventricular transmural action potential gradient. This alteration is responsible for QT prolongation in the electrocardiographic tracing and constitutes a risk factor for arrhythmias and sudden death.11

Organic changes due to hypoxemia, e.g., oxidative stress, production of reactive oxygen species, endothelial dysfunction, and activation of the renin-angiotensin-aldosterone system, lead to ventricular remodeling due to arrhythmogenesis.3 However, myocardial ischemia resulting from hypoxemia can also lead to sudden death and is one of its primary causes.11 Studies also suggest that, in addition to myocardial ischemia, fibrosis and wall stress due to hypertension after sympathetic stimulation favor modifications in the depolarization waves and reentry mechanisms, increasing the risk of ventricular arrhythmias and sudden death.47

Although including the QRS complex (a component of ventricular depolarization), the prolongation of the QT interval, which mainly represents ventricular repolarization, is described as a risk factor often associated with sudden death.24 It has also been reported that, in OSA patients, the QT interval increases with the severity of the disorder. Furthermore, QT prolongation reflects the increased heterogeneity of ventricular repolarization and the presence of ventricular tachycardia.38

In the analysis of 619 death certificates of adults who underwent polysomnography, of which 160 resulted in cardiovascular-related deaths, patients without OSA and the general population showed lower mortality rates during the night, while OSA patients showed a uniform distribution of cardiovascular death during daytime hours. Furthermore, in the general population and patients without OSA, there was a peak of sudden death in the early morning hours and a nocturnal decrease, which was not seen among patients with obstructive apnea. In this study, 23% of deaths among OSA patients occurred in the morning and 23% at night.48 In turn, a previous study observed 46% of sudden deaths occurring at night and 20% in the morning among patients with obstructive apnea. However, this study observed that the AHI of people with SCD during sleeping hours was higher than among those who died at other times of the day, demonstrating the direct relationship between AHI and the risk of SCD during sleep, differing from the general population, with a nocturnal nadir and a morning peak.49

A meta-analysis of 13 studies highlighted the association between sleep apnea and fatal events, cardiovascular outcomes, and hospitalization time, suggesting that the treatment of sleep apnea could mitigate these indicators.50 However, a study that compared patients with severe obstructive sleep apnea (AHI >30) and a control group of patients with AHI <5 showed that the QTc interval of patients with severe apnea was more dispersed than in the control group, highlighting a significant improvement among these patients after CPAP therapy, which could reduce the risk of arrhythmias and cardiovascular events.51

Finally, a study conducted for six months with 303 patients with minimally symptomatic obstructive sleep apnea using ECG risk markers analyzed the role of CPAP therapy on the behavior of atrial fibrillation and the remaining factors that could lead to sudden death, observing no effect of CPAP therapy on the risk markers for arrhythmias and sudden death in this group.24