Timing of Nocturnal Ventricular Ectopy in Heart Failure Patients With Sleep Apnea*

Transcription

1 Original Research SLEEP MEDICINE Timing of Nocturnal Ventricular Ectopy in Heart Failure With Sleep Apnea* Clodagh M. Ryan, MD; Stephen Juvet, MD; Richard Leung, MD, PhD; and T. Douglas Bradley, MD Background: Ventricular ectopy is frequent in heart failure (HF) patients with sleep apnea. A previous report indicated that in HF patients, ventricular premature beats (VPB) occurred more frequently during episodes of recurrent central sleep apnea (CSA) than during normal breathing, and their frequency was greater during hyperpnea than during apnea. We hypothesized that, because respiratory stimuli that might provoke ventricular ectopy are stronger during obstructive apneas than during central apneas, in contrast to CSA, VPBs would be more frequent during apnea than hyperpnea in HF patients with obstructive sleep apnea (OSA). Methods: HF patients in sinus rhythm who have OSA or CSA (apnea-hypopnea index, > 15 events per hour) and with > 30 VPBs per hour were matched for severity of cardiac dysfunction and sleep apnea. The frequency of VPBs was then assessed during stage 2 sleep during the apneic and the hyperpneic phases of recurrent obstructive or central apneas. Results: VPBs occurred more frequently during the apneic phase than during the hyperpneic phase in patients with OSA. In contrast, VPBs occurred more frequently during the hyperpneic phase than the apneic phase in patients with CSA. There was no difference in the degree of apnea-related oxygen desaturation between central and obstructive apneas. Conclusions: In patients with HF, nocturnal ventricular ectopy oscillates in time with oscillations in ventilation, with VPBs occurring predominantly during apneas in patients with OSA, but during hyperpneas in patients with CSA. This difference in VPB timing between OSA and CSA may be attributable to the differences in timing of arrhythmic stresses in these patients. (CHEST 2008; 133: ) Key words: congestive heart failure; sleep-disordered breathing; ventricular arrhythmias Abbreviations: AHI apnea-hypopnea index; CPAP continuous positive airway pressure; CSA central sleep apnea; HF heart failure; LECT lung-to-ear circulation time; LV left ventricular; OSA obstructive sleep apnea; Sao 2 arterial oxygen saturation; VPB ventricular premature beat; Vt tidal volume Ventricular arrhythmias are a common cause of sudden death in patients with heart failure (HF). 1 However, the pathogenesis of ventricular ectopy in HF patients is not fully understood. Potential mechanisms include increased sympathetic outflow, myocardial stretch, and a mismatch between myocardial oxygen demand and supply, especially in those patients with ischemic heart disease. 2 Sleep apnea can provoke these stimuli during sleep. Previous studies 3 6 have demonstrated associations between the occurrence of ventricular arrhythmias and both obstructive sleep apnea (OSA) and central sleep apnea (CSA). However, since the pathophysiology of OSA and CSA differ, it is possible that the timing of these cardiovascular stresses during the apnea-hyperpnea cycle may also differ. In obstructive apneas, chemostimulation and consequent sympathetic activation are present from the onset of the apneas, and progressively increase to maximum when the apnea is terminated by an arousal. 7 In addition, the generation of exaggerated negative intrathoracic pressure against the occluded pharynx during obstructive events applies a distending force to the myocardium. 8 Following apnea termination, rapid lung inflation coupled with a rapid increase in Pao 2 and an increase in intrathoracic pressure cause abrupt falls in sympathetic outflow and cardiac distending forces, respectively. 9 In patients with HF, the treatment of coexisting OSA by continuous positive airway pressure (CPAP) reduces the frequency 934 Original Research

2 of ventricular premature beats (VPBs) during sleep, 10 indicating a causal relationship between OSA and VPBs. Central apneas are triggered by a fall in chemostimulation, especially Paco 2, below the apnea threshold. 11 Consequently, in contrast to obstructive apneas, chemostimulation and sympathetic activity are at their minimum during central apneas. 12 In addition, since no negative intrathoracic pressure is generated during central apneas, no distending pressure is applied to the heart. Owing to prolonged lung-to-chemoreceptor circulation time, maximum chemostimulation and sympathetic activation occur well after apnea termination during the ventilatory phase, when large tidal volumes (Vts) and negative intrathoracic pressure swings also expose the heart to distending forces. 12 Indeed, in HF patients with CSA, VPBs occur more frequently during hyperpneas than apneas. Furthermore, the abolition of CSA by the inhalation of CO 2 -enriched gas reduces the frequency of VPBs, 4 which also indicates a causal relationship between CSA and VPBs. In view of the above, we hypothesized that VPBs would be more frequent during apneas in HF patients with OSA, but more frequent during hyperpneas in those with CSA. To this end, we compared the frequency of VPBs during apneas and hyperpneas in HF patients with either OSA or CSA who had similar degrees of left ventricular (LV) dysfunction and severity of sleep apnea. Subjects Materials and Methods As part of an ongoing prospective epidemiologic study, all patients with HF who are newly referred to our HF clinic *From the Sleep Research Laboratory (Drs. Ryan, Juvet, and Bradley), Toronto Rehabilitation Institute, Toronto, ON, Canada; and the Sleep Research Laboratory (Dr. Leung), St. Michael s Hospital, Toronto, ON, Canada. Supported by operating grant MOP from the Canadian Institutes of Health Research (CIHR). Dr. Ryan was supported by a research fellowship from the Toronto Rehabilitation Institute, Dr. Leung was supported by a Clinician Scientist Award from the CIHR, and Dr. Bradley was supported by a Senior Scientist Award from the CIHR. The authors have reported to the ACCP that no significant conflicts of interest exist with any companies/organizations whose products or services may be discussed in this article. Manuscript received October 22, 2007; revision accepted December 18, Reproduction of this article is prohibited without written permission from the American College of Chest Physicians ( org/misc/reprints.shtml). Correspondence to: T. Douglas Bradley, MD, Toronto General Hospital/University Health Network, 9N-943, 200 Elizabeth St, Toronto, ON, Canada M5G 2C4; douglas.bradley@ utoronto.ca DOI: /chest undergo overnight polysomnography. This protocol was approved by the University of Toronto Research Ethics Board, and subjects provided written informed consent before study entry. The inclusion criteria were as follows: (1) chronic HF (LV ejection fraction 45%, as assessed by echocardiography) secondary to ischemic or nonischemic cardiomyopathy; (2) sinus rhythm on the ECG; (3) 30 VPBs per hour of sleep on an overnight ECG recording; and (4) moderate-to-severe sleep apnea, defined as an apnea-hypopnea index (AHI) of 15 events per hour of sleep. Subjects were then divided into groups with either OSA predominantly, in which 85% of the events were obstructive, or CSA predominantly, in which 85% of the events were central. The exclusion criteria included patients with cardiac pacemakers or atrial fibrillation. ECG data were analyzed from the 20 most recent consecutive patients with OSA and the 20 most recent consecutive patients with CSA who had not been included in previous research studies. Sleep Studies All patients underwent overnight polysomnography using standard techniques and criteria for scoring sleep stages and arousals. 11,13 Thoracoabdominal movements and Vt were monitored by a calibrated respiratory inductance plethysmograph (Respitrace; Ambulatory Monitoring Inc; White Plains, NY), 14 arterial oxygen saturation (Sao 2 ) was monitored by an ear oximeter (Nellcor N200; Tyco International Healthcare; Pleasanton, CA), 13 and cardiac rhythm was monitored from a one-lead ECG. Transcutaneous Pco 2 was measured continuously with a transcutaneous capnograph (Kontron Medical, Hoffman LaRoche; Basel, Switzerland), which had previously been validated against arterial Pco All signals were recorded on a computerized sleep-scoring system (Sandman; Tyco Ltd; Ottawa, ON, Canada). Apneas were defined as an absence of Vt for 10 s, and were classified as obstructive if there was out-of-phase thoracoabdominal motion, and as central if there was no thoracoabdominal motion. Hypopneas were defined as a 50% reduction in Vt for 10 s. Hypopneas were classified as being obstructive or central, respectively, in the presence or absence of out-of-phase thoracoabdominal motion. 11,15 The scoring of apneas and hypopneas was performed by a polysomnographic technician who was blinded to the ECG findings. The AHI was calculated as the number of apneas and hypopneas per hour of sleep. We confined our data analysis to stage 2 non-rapid eye movement sleep for several reasons. First, this was the dominant stage in all subjects. Second, apnea-hyperpnea cycles were most commonly present during this stage. Third, the cardiovascular and respiratory systems were under predominantly metabolic regulation during this stage, and therefore were not subject to behavioral influences. Finally, by analyzing all data from a single sleep state, we were able to control for the potential effects of sleep state on apnea-hyperpnea characteristics. During episodes of recurrent obstructive apneas in stage 2 sleep, the respiratory cycle was divided into the following two phases: the apneic phase; and the hyperpneic phase. The apnea duration was defined as the time between the end of inspiration of the breath preceding the onset of apnea and the onset of inspiration during the breath that terminated the apnea. The hyperpnea duration was defined as the time between the onset of inspiration of the first breath terminating the apnea and the end of the inspiration of the breath preceding the next apnea. Cycle duration was calculated as the sum of the apnea and the hyperpnea durations. 16 The time to peak Vt was defined as the interval from the onset of the breath terminating the apnea to the largest Vt. 17 Lung-to-ear circulation time (LECT), taken as the time from the end of an apnea until the subsequent nadir in Sao 2 detected at the ear, was used as an estimate of lung-to-carotid chemoreceptor CHEST / 133 / 4/ APRIL,

3 circulation time. This was performed in a subset of patients who had Sao 2 measured by an oximeter placed on the ear (Nellcor N200; Tyco International Healthcare). We have previously validated 17 this technique against cardiac output as a measure of circulation time in patients with sleep apnea, with and without HF. The changes in Sao 2 from both the peak Sao 2 and the Sao 2 at the end of the apnea to the nadir in Sao 2 following apnea termination for each apnea-hyperpnea cycle were calculated. The desaturation time, the rate of change in Sao 2 per second from the end of the apnea to the nadir was calculated. Ten consecutive apnea-hyperpnea cycles during the first and last episode of stage 2 sleep were analyzed in each subject. Timing of VPBs The ECG was sampled at a frequency of 1000 Hz. The ECG tracing during all stages of sleep was inspected for the presence of VPBs. The QRS complexes were scored as VPBs if they were (1) premature, (2) not preceded by a premature P wave, (3) 0.12 s in duration, and (4) of different morphology from those arising from sinus beats. VPBs were identified as occurring during the apneic or hyperpneic phase during episodes of recurrent central or obstructive apneas for total sleep time and for stage 2 sleep. The frequency of VPBs during apneas and hypopneas was compared with that during hyperpneas for patients with OSA and CSA. Statistical Analysis Data were expressed as the mean SEM. Statistical analysis was performed using a statistical software package (SPSS, version 13.0; SPSS Inc; Chicago, IL). Continuous variables for the patients with OSA or CSA were compared using two-tailed unpaired t tests for variables with normally distributed data, the Mann-Whitney rank sum test for variables with nonnormally distributed data, and the Fisher exact test for nominal variables. VPB frequency and timing were assessed for within-group differences using paired t tests. Comparisons between the OSA and CSA groups were performed using the analysis of covariance to adjust for baseline differences in VPB frequency. A p value of 0.05 was considered to be statistically significant. Results Characteristics of the Subjects Forty patients (20 subjects in each group) with HF met our inclusion criteria. The clinical and polysomnographic data are summarized in Tables 1 and 2, respectively. had a moderate-to-severe degree of sleep apnea, as indicated by their AHI. with OSA were younger and more obese than those with CSA. In all subjects, LV ejection fraction was moderately to severely depressed, and all subjects were receiving appropriate medical therapy for HF. Medical therapy for HF was similar in both patient groups (Table 1). Sleep architecture was similar in both groups, although those patients with CSA had shorter total sleep times and fewer arousals during sleep than those with OSA. with OSA also had a higher mean Sao 2, but there was no significant difference in minimum Sao 2 values between the two groups. Table 1 Baseline Characteristics of the Subjects* Characteristics Timing of VPBs Figure 1 shows a representative polysomnographic recording from a patient with OSA (Fig 1, left, A) and CSA (Fig 1, right, B). Note that VPBs occur predominantly during apnea in the patient with OSA, and predominantly during hyperpnea in the patient with CSA. The frequency of VPBs during sleep (during apneic phases, hyperpneic phases, and normal breathing) was similar in patients with OSA and CSA ( vs events per hour, respectively; p 0.655). In subjects with OSA, VPB frequency was significantly higher during the apneic phase than during the hyperpneic phase both during total sleep ( vs events per hour, respectively; p 0.001) and in Table 2 Overnight Sleep Study Data* Variables OSA (n 20) OSA (n 20) CSA (n 20) p Value Age, yr Sex 1.0 Male Female 1 0 BMI, kg/m Cause of HF Ischemic Idiopathic LVEF, % NYHA class Blockers ACEI Diuretics Amiodarone Digoxin *Values are given as the mean SEM or No. of patients, unless otherwise indicated. BMI body mass index; LVEF left ventricular ejection fraction; NYHA New York Heart Association; ACEI angiotensin-converting enzyme inhibitor. CSA (n 20) p Value AHI, events/h of sleep TST, min Stage 1 sleep, % Stage 2 sleep, % SWS, % REM sleep, % Arousals, No./h of sleep Sao 2,% Minimum Sao 2,% Ptcco 2,mmHg *Values are given as the mean SEM, unless otherwise indicated. TST total sleep time; SWS slow-wave sleep; Ptcco 2 transcutaneous Pco Original Research

4 Figure 1. Left, A: a representative polysomnographic tracing from a subject with HF and OSA during stage 2 sleep. It demonstrates the occurrence of VPBs during apneas. Right, B: a representative polysomnographic tracing from a subject with HF and CSA. It demonstrates the occurrence of VPBs during hyperpneas. Note that LECT (A to D), the time to peak Vt (A to B), and hyperpnea duration (A to C) are all longer in the subject with CSA than in the subject with OSA (30 vs 24 s, 28 vs 16 s, and 51 vs 17 s, respectively). SUM sum of the ribcage and abdominal excursions measured by the plethysmograph (Respitrace; Ambulatory Monitoring Inc). stage 2 sleep ( vs events per hour, respectively; p 0.015). Conversely in the 20 subjects with CSA, VPB frequency was significantly greater during the hyperpneic phase than the apneic phase for total sleep ( vs events per hour, respectively; p 0.001) and stage 2 sleep ( vs events per hour, respectively; p 0.001). When the VPB frequency was expressed per 100 heart beats, the differences between the apneic and hyperpneic phase in patients with OSA and CSA remained significant (Fig 2). The mean change in the VPB frequency and the number of VPBs per 100 heart beats between the apneic and hyperpneic phases of events were significantly different between patients with OSA and those with CSA during stage 2 sleep (Table 3). The total cycle and apnea durations were similar in the patients with OSA and CSA (Table 4). However, the hyperpnea duration, the time to peak ventilation, and the number of breaths per hyperpnea were significantly greater in the subjects with CSA. The change in Sao 2 during each cycle did not differ significantly between the CSA or OSA subjects either from the peak or from the end of the apnea to the nadir. In the subset of patients whose Sao 2 was measured by an ear oximeter (OSA, 12 patients; CSA, 8 patients), the LECT was significantly greater in patients with CSA than in those with OSA ( vs s, respectively; p 0.028). Discussion Figure 2. VPBs per 100 heart beats during apneas and hyperpneas in stage 2 sleep in patients with OSA and CSA. Note that the data are plotted on a log scale. The frequency of VPBs is greater during apneas in patients with OSA, but greater during hyperneas in patients with CSA. In a previous study, 4 we observed that the frequency of VPBs was higher during hyperpnea than apnea in HF patients with CSA. The present findings confirm that observation. However, the timing of ventricular ectopy during sleep in HF patients with CHEST / 133 / 4/ APRIL,

5 Table 3 Change in VPBs Between Apneas and Hyperpneas* OSA (n 20) CSA (n 20) Variables Mean 95% CI p Value Mean 95% CI p Value p Value VPBs/h (A H) TST 97.9 (43.4 to 152.3) ( 223 to 64.6) VPB/h (A H) S (21.45 to 226.7) ( to 21.8) VPB/100 hb (A H) S2 2.7 (0.15 to 5.12) ( 7.6 to 3.4) *Values are given as the mean (95% confidence interval), unless otherwise indicated. VPB/h (A H) difference in VPBs per hour between apneas and hyperpneas of a given event; S2 stage 2 sleep; VPB/100 hb difference in VPBs per 100 heart beats between apneas and hyperpneas of a given event; CI confidence interval. See Table 2 for abbreviation not used in the text. Paired t tests were used for within-group comparisons. Analysis of covariance was used for between-group comparisons. Characteristics Table 4 Cycle Characteristics* OSA (n 20) CSA (n 20) p Value Total cycle duration, s Apnea duration, s Hyperpnea duration, s Time to peak Vt, s Breaths/hyperpnea ratio Sao 2 (a n), % Sao 2 (p n), % Rate of Sao 2,% *Values are given as the mean SEM, unless otherwise indicated. TCL total cycle length; Sao 2 (a n) change in Sao 2 from the end of the apnea to the nadir following apnea; Sao 2 (p n) change in Sao 2 from the peak Sao 2 prior to apnea to the nadir following apnea; Rate of Sao 2 change in Sao 2 from the end of the apnea to the nadir per second. OSA had not been examined previously. In this study, we compared the timing of the occurrence of ventricular ectopy in HF patients with either CSA or OSA matched for LV ejection fraction, sex, and AHI. During stage 2 sleep in HF patients with OSA, we demonstrated that, in contrast to those with CSA, there was a significantly greater frequency of ventricular ectopy during apnea than during hyperpnea. Nevertheless, in both CSA and OSA, VPBs occur in a characteristic cyclic fashion that is synchronous with respiratory oscillations during apneas and hyperpneas. with HF often die suddenly, and ventricular arrhythmias have been implicated in many such deaths. 1 The pathogenesis of ventricular arrhythmias in patients with HF has yet to be fully elucidated. The factors thought to be involved include the following: transient or sustained myocardial stretch that causes mechanoelectrical dissociation; elevated sympathetic nervous system activity; and a mismatch between myocardial oxygen demand and supply that facilitates arrhythmogenic abnormalities of repolarization and conduction. The superimposition of OSA or CSA in patients with HF is associated with an increase in the frequency of ventricular arrhythmias. 5,6,18 Moreover, in patients with HF, the alleviation of OSA by CPAP, and of CSA by the inhalation of a CO 2 -enriched gas or CPAP markedly attenuates the frequency of VPBs, 4,10,19 indicating that the breathing disorder contributes to the causation of ventricular ectopy. Differences in the timing of VPBs between patients with OSA and those with CSA may be due to several mechanisms. First, nocturnal ventricular ectopy in both obese subjects and in those with acute coronary syndrome has been attributed to nocturnal episodic hypoxia, presumably due to sleep-disordered breathing. 20,21 In patients with OSA, Shepard et al 22 reported that VPBs occurred in association with severe O 2 desaturation in patients free of overt cardiac disease. However, Leung and colleagues 4 reported that while the administration of oxygen to subjects with CSA prevented intermittent hypoxia, it had no effect on CSA or VPB frequency. Both of our subject groups had similar cardiovascular profiles and severity of sleep-disordered breathing. While in the present study the mean Sao 2 was lower in those with CSA than in those with OSA, it was within the normal range in both groups; the minimum Sao 2 was approximately 80% and did not differ between the groups. Therefore, it is highly unlikely that the difference in VPB timing between patients with OSA and those with CSA is due to a difference in the severity of hypoxia. Second, the timing of VPBs both in patients with CSA and those with OSA may be related to the generation of negative intrathoracic pressures that can cause myocardial stretch. During obstructive apneas, negative intrathoracic pressure is generated by inspiratory efforts against the occluded pharynx, which results in increased venous return and right ventricular distension during the apneic phase. 8 At apnea termination, the opening of the pharynx causes an abrupt resumption of airflow accompanied by an increase in intrathoracic pressure, which re- 938 Original Research

6 verses these right ventricular distending forces during hyperpnea. In contrast, during central apneas no negative intrathoracic pressure is generated, so that no distending force is applied to the right ventricle. During hyperpnea, however, the increased respiratory drive that is characteristic of patients with CSA 12 causes the generation of markedly negative intrathoracic pressure compared to apnea, which corresponds to the peak Vt in the middle of hyperpnea and applies a distending force to the ventricles. This can cause subsequent ventricular irritability. 23,24 Thus, differences in the timing of ventricular distending forces related to the generation of negative intrathoracic pressure may, in part, account for differences in the timing of VPBs between OSA and CSA. Another mechanism that may contribute to differences in the timing of VPBs is the differing timing of chemical and autonomic stimuli between patients with OSA and those with CSA. In OSA patients, hypoxia, hypercapnia, and recurrent arousals elicited by obstructive apneas induce repetitive surges in sympathetic neural outflow, heart rate, and BP that converge with peak ventricular distending forces due to negative intrathoracic pressure at or just after the termination of apneas. 7 This convergence of mechanical, chemical, and autonomic stimuli during obstructive apneas could increase ventricular irritability and provoke VPBs during apneas. In CSA patients, however, the nadir of Sao 2 and the peak in Vt occur later in the hyperpnea than they do in OSA patients, probably owing to the longer lung-tochemoreceptor transit time. 25 Similarly, in contrast to OSA patients, in CSA patients sympathetic nerve traffic peaks during hyperpnea rather than at apnea termination. 9,12 This convergence of peak respiratory drive and sympathetic outflow in concert with myocardial stretch due to the generation of negative intrathoracic pressure may act as arrhythmogenic stimuli that provoke VPBs during hyperpnea. The timing of arousals might also contribute to the timing of VPBs. Unlike OSA, in which arousals occur at apnea termination, in CSA, arousals often occur several breaths after apnea termination, which could further promote sympathetic activation during hyperpnea. 9,12 However, studies 26,27 in both OSA and CSA patients have suggested that peak increases in BP and heart rate are more closely linked to peak respiratory drive and ventilation than to arousals. A limitation of our study was that we could not identify, with certainty, the factors responsible for the differences in VPB timing between the OSA and CSA patients that we observed. Nonetheless, the difference in timing of VPBs was consistent in all subjects, indicating differences in the timing of factors that provoke ventricular ectopy between OSA and CSA in patients with HF. A recent study 28 demonstrated that 59% of patients receiving long-term cardiac pacing had sleep apnea. Furthermore, the classic risk factors for sleep apnea (older age, higher body mass index, and excessive daytime sleepiness) were not present in these patients so that the suspicion of sleep apnea would be unlikely to have been aroused. Potentially, the detection and treatment of sleep apnea could avert pacemaker or cardioverter-defibrillator implantation in a subset of such patients. 28 In this context, a nocturnal cyclic pattern of VPBs detected by Holter monitoring, as described in this study, would suggest underlying sleep apnea, and would alert physicians to the need for an overnight sleep study and therapy for CSA or OSA, if present. In summary, we have demonstrated in patients with HF that the timing of VPBs during sleep differs between patients with OSA and those with CSA. VPBs occur more frequently during apneas in OSA patients, but more frequently during hyperpneas in CSA patients. The most likely explanation for this difference in the timing of VPBs is a greater circulatory delay in patients with CSA, causing arrhythmogenic stimuli to peak later following apnea than in OSA patients, in whom such stimuli peak prior to or coincident with apnea termination. Further studies will need to be performed to determine the cause of these observations as they may have important implications for the treatment of nocturnal ventricular arrhythmias in HF patients with sleep-disordered breathing. References 1 Chakko CS, Gheorghiade M. Ventricular arrhythmias in severe heart failure: incidence, significance, and effectiveness of antiarrhythmic therapy. Am Heart J 1985; 109: Sweeney MO. Sudden death in heart failure associated with reduced left ventricular function: substrates, mechanisms, and evidence-based management: part I. Pacing Clin Electrophysiol 2001; 24: Mehra R, Benjamin EJ, Shahar E, et al. Association of nocturnal arrhythmias with sleep-disordered breathing: the Sleep Heart Health Study. Am J Respir Crit Care Med 2006; 173: Leung RS, Diep TM, Bowman ME, et al. Provocation of ventricular ectopy by Cheyne-Stokes respiration in patients with heart failure. Sleep 2004; 27: Javaheri S, Parker TJ, Liming JD, et al. Sleep apnea in 81 ambulatory male patients with stable heart failure: types and their prevalences, consequences, and presentations. Circulation 1998; 97: Alonso-Fernandez A, Garcia-Rio F, Racionero MA, et al. Cardiac rhythm disturbances and ST-segment depression episodes in patients with obstructive sleep apnea-hypopnea syndrome and its mechanisms. Chest 2005; 127: Somers VK, Dyken ME, Clary MP, et al. Sympathetic neural mechanisms in obstructive sleep apnea. J Clin Invest 1995; 96: Bradley TD, Hall MJ, Ando S, et al. Hemodynamic effects of CHEST / 133 / 4/ APRIL,

7 simulated obstructive apneas in humans with and without heart failure. Chest 2001; 119: Bradley TD, Floras JS. Sleep apnea and heart failure: part I. Obstructive sleep apnea. Circulation 2003; 107: Ryan CM, Usui K, Floras JS, et al. Effect of continuous positive airway pressure on ventricular ectopy in heart failure patients with obstructive sleep apnoea. Thorax 2005; 60: Naughton M, Benard D, Tam A, et al. Role of hyperventilation in the pathogenesis of central sleep apneas in patients with congestive heart failure. Am Rev Respir Dis 1993; 148: Bradley TD, Floras JS. Sleep apnea and heart failure: part II. Central sleep apnea. Circulation 2003; 107: Rechtschaffen A, Kales A. A manual of standardized terminology, techniques and scoring system for sleep stages of human subjects. Washington DC: National Institutes of Health, Chadha TS, Watson H, Birch S, et al. Validation of respiratory inductive plethysmography using different calibration procedures. Am Rev Respir Dis 1982; 125: Bradley TD, Rutherford R, Grossman RF, et al. Role of daytime hypoxemia in the pathogenesis of right heart failure in the obstructive sleep apnea syndrome. Am Rev Respir Dis 1985; 131: Ryan CM, Bradley TD. Periodicity of obstructive sleep apnea in patients with and without heart failure. Chest 2005; 127: Hall MJ, Xie A, Rutherford R, et al. Cycle length of periodic breathing in patients with and without heart failure. Am J Respir Crit Care Med 1996; 154: Lanfranchi PA, Somers VK, Braghiroli A, et al. Central sleep apnea in left ventricular dysfunction: prevalence and implications for arrhythmic risk. Circulation 2003; 107: Javaheri S. Effects of continuous positive airway pressure on sleep apnea and ventricular irritability in patients with heart failure. Circulation 2000; 101: Koehler U, Glaremin T, Cassel W, et al. Nocturnal ventricular arrhythmia in patients with sleep apnea and suspected coronary heart disease. Med Klin (Munich) 1993; 88: Galatius-Jensen S, Hansen J, Rasmussen V, et al. Nocturnal hypoxaemia after myocardial infarction: association with nocturnal myocardial ischaemia and arrhythmias. Br Heart J 1994; 72: Shepard JW Jr, Garrison MW, Grither DA, et al. Relationship of ventricular ectopy to oxyhemoglobin desaturation in patients with obstructive sleep apnea. Chest 1985; 88: Franz MR, Cima R, Wang D, et al. Electrophysiological effects of myocardial stretch and mechanical determinants of stretch-activated arrhythmias. Circulation 1992; 86: Chen RL, Penny DJ, Greve G, et al. Stretch-induced regional mechanoelectric dispersion and arrhythmia in the right ventricle of anesthetized lambs. Am J Physiol Heart Circ Physiol 2004; 286:H1008 H Tkacova R, Niroumand M, Lorenzi-Filho G, et al. Overnight shift from obstructive to central apneas in patients with heart failure: role of Pco 2 and circulatory delay. Circulation 2001; 103: O Donnell CP, Ayuse T, King ED, et al. Airway obstruction during sleep increases blood pressure without arousal. J Appl Physiol 1996; 80: Trinder J, Merson R, Rosenberg JI, et al. Pathophysiological interactions of ventilation, arousals, and blood pressure oscillations during Cheyne-Stokes respiration in patients with heart failure. Am J Respir Crit Care Med 2000; 162: Garrigue S, Pepin JL, Defaye P, et al. High prevalence of sleep apnea syndrome in patients with long-term pacing: the European Multicenter Polysomnographic Study. Circulation 2007; 115: Original Research