The Effects of Continuous Positive Airway Pressure on Premature Ventricular Contractions and Ventricular Wall Stress in Patients with Heart Failure and Sleep Apnea

Background We aimed to investigate the effects of continuous positive airway pressure (CPAP) treatment on electrocardiography (ECG), premature ventricular contraction load on 24-hour Holter recordings, and implantable cardioverter defibrillator (ICD) shocks in patients with obstructive sleep apnea syndrome (OSAS) and heart failure. Methods Patients with heart failure and ICD and patients with newly diagnosed OSAS were divided into two groups according to CPAP treatment. To compare the impact of CPAP on ECG parameters, both baseline and 6-month ECG, 24-hour Holter ECG, ambulatory blood pressure monitoring, echocardiography, polysomnography, and laboratory parameters were collected. Results CPAP treatment significantly reduced the frequency of premature ventricular contractions, T-peak to T-end, corrected QT, corrected QT dispersion, and T-peak to T-end/corrected QT ratio in the study group (p < 0.001 for all). Although the baseline NT-pro-BNP levels were similar between study and control groups, after six months, the NT-pro-BNP levels of the study group were significantly lower than that of the control group (39.18 ± 7.57 versus 46.11 ± 7.65; p < 0.001). Conclusions CPAP treatment in patients with heart failure and ICD and in patients with newly diagnosed OSAS may have beneficial effects on premature ventricular contractions and electrocardiographic arrhythmia indices and NT-pro-BNP levels. However, these results are needed to be clarified with further studies.


Introduction
Obstructive sleep apnea syndrome (OSAS) is characterized by temporary airway collapse during sleep and a ects more than 4% of men and 2% of women [1]. Many cardiovascular diseases have been found to be associated with OSAS including arterial hypertension, ischemic heart disease, heart failure, arrhythmias, and stroke. Cardiovascular morbidity and mortality have been shown to be high in patients with OSAS [2,3]. Sympathetic overactivity, activation of in ammatory cascades, vascular endothelial dysfunction, abnormalities in the coagulation pathway, and metabolic dysregulation are likely to be involved in the pathogenesis of the cardiovascular complications of OSAS. e relation between the congestive heart failure and OSAS is somewhat complex. OSAS may a ect cardiac functions and contribute to the development of heart failure in the long-term. On the other hand, heart failure itself also plays a role in the pathogenesis of OSAS [4][5][6][7]. Several studies reported that continuous positive airway pressure (CPAP) treatment in patients with congestive heart failure and OSAS reduced sympathetic activation and improved left ventricular ejection fraction [8][9][10]. However, the positive e ect of CPAP treatment on the cardiovascular system is unclear in terms of the prognosis of patients with congestive heart failure and OSAS [11].
ere are some arrhythmic changes on electrocardiography (ECG) in patients with OSAS, and some improvement with CPAP therapy has been reported [12][13][14]. e association between arrhythmias and OSAS is well known [15,16]. OSAS associated with complex ventricular ectopia and ventricular tachycardia has been documented in a previous study [17]. CPAP treatment also resulted in a 58% reduction in premature ventricular beats [18]. ose ndings may help to explain the pathogenesis of sudden death in OSAS patients [19]. It has also been shown that OSAS patients have a 4-fold higher frequency of atrial brillation if apnoeahypopnoea index (AHI) index is greater than 30 [17].
Heart failure is a life-threatening disease with a growing incidence in developed countries due to better treatment options for heart failure, improvement of survival after myocardial infarction, and increased life expectancy. e introduction of implantable devices such as implantable cardioverter de brillators (ICD) has improved the overall survival of patients with heart failure [20]. ICD treatment reduces mortality in patients with heart failure. However, certain conditions such as drug-resistant ventricular arrhythmias and atrial brillation may increase morbidity and mortality due to inappropriate and frequent shocks. e aim of the present study is to investigate whether the regular CPAP treatment improves ECG ndings and reduces premature ventricular contractions (PVC) and ICD shocks in patients with OSAS, cardiac failure, PVC, and ICD implant.

Study Population.
is study is an observational study conducted with patients followed at the Department of Cardiology with diagnosis of chronic heart failure. Our inclusion criteria for the study population were ejection fraction < 35%, stable NYHA class ≥ II, the presence of ICD, sinus rhythm, and >30 PVC/hour on 24-hour Holter monitorization. In addition, patients had to be on stable heart failure medication for at least 4 weeks prior to enrolment and during follow-up. Patients lled Epworth Sleepiness Scale (ESS) form. Clinically suspected patients were referred to Department of Chest Diseases for possible OSAS diagnosis. After polysomnographic study in the sleep laboratory, CPAP treatment was recommended for moderate to severe OSAS patients. Patients only with OSAS were included, and patients with central sleep apnea (CSA) and mixed apnea were not included. Protocol of the study was approved by the local ethics committee, and subjects involved in the study signed informed consent approved by the institution.
Baseline ECG, 24-hour Holter ECG, ambulatory blood pressure monitoring, echocardiography, laboratory investigations, and ICD measurements were recorded. Age, sex, body mass index (BMI), the cause of heart failure, heart rate, arterial blood pressure, left ventricular end diastolic diameter (LVEDD), left atrium (LA) diameter, polysomnography ndings, levels of creatinine, and hemoglobin of the participants were also recorded. Class of drugs used by the patient was also recorded. Study group (n � 40) consisted of ICD implanted heart failure and OSAS patients who accepted CPAP treatment, and control group (n � 40) consisted of ICD implanted heart failure and OSAS patients who did not accept CPAP treatment. Chronic atrial brillation, bundle branch block, application of cardiac resynchronization therapy (CRT), the presence of pacemaker rhythm, estimated glomerular ltration rate of less than 60 mL/min/1.73 m 2 (CKD stage ≥ III), signi cant lung disease, thyroid dysfunction, and New York Heart Association (NYHA) grade IV heart failure were the exclusion criteria. Six months after the initiation of CPAP therapy, ECG, 24-hour Holter monitorization, ambulatory blood pressure monitoring, and echocardiography were repeated in both groups.

Polysomnography.
All referred individuals with heart failure and suspected OSAS were subjected to 1 full-night polysomnography at the sleep laboratory using the Nox A1 PSG System (Nox Medical, Iceland). Physiological variables evaluated during polysomnography included electroencephalogram, electrooculogram, electromyogram (submentonian and tibialis), electrocardiogram, oral/nasal air ow measured by oronasal thermistors, respiratory e ort (thorax and abdomen), body position, and blood oxygen saturation (SpO 2 ) (mean and minimum SpO 2 ). e exam was performed according to speci c criteria for the de nition of sleep stages [21] by trained technicians who were unaware of the patient characteristics.
Sleep-related respiratory events and arousals were scored according to the American Academy of Sleep Medicine Manual for Scoring Sleep and Associated Events [22]. Central sleep apnea was de ned as the absence of thoracic and abdominal movements with absence of air ow. Obstructive sleep apnea was de ned as the absence of air ow in the presence of thoracic and abdominal movements. Mixed apnea was de ned as apnea without any thoracoabdominal movements in whom thoracoabdominal movements reoccurred prior to reoccurrence of nasal ow. Apnea was dened as complete cessation of the air ow ≥ 10 s. Hypopnea was de ned as ≥30% decrease in air ow from baseline for at least 10 seconds with a≥3% oxygen desaturation from preevent baseline, and/or the event is associated with an arousal. Apnea/hypopnea index was determined via careful calculations. AHI score between 15 and 29 was considered as OSAS with moderate severity. Severe OSAS was de ned as AHI ≥ 30.

Treatment.
Autotitrating CPAP was used in the sleep laboratory within a period of less than 7 days after the diagnostic study to get a xed CPAP pressure value. e xed CPAP pressure was used for the rest of the study. All patients were scheduled for follow-up 2 weeks after randomization and, subsequently, at 4, 8, 12, 18, and 24 weeks. During the follow-up visits, adherence to CPAP and antihypertensive medications was documented. Compliance with CPAP therapy was based on the self-report of patients at the end of the study period. e use ratio of CPAP (%) during the study period, which was the average time with CPAP to total time spent in bed, was 91% (range 72-100%).

Electrocardiography.
e 12-lead ECG (CARDIOVIT AT-102 Plus, Schiller, Switzerland) was recorded at a paper speed of 25 mm/s at 10 mm/mV amplitude in the supine position. ECGs were performed while the patient was at rest and at the morning hours (between 8:00 and 10:00 AM). All of the ECGs were scanned with a resolution of 800 dpi and transferred to a personal computer. ECG measurements of QT and Tp-e intervals at V5 were performed by experienced cardiologists who were blinded to the patient data. If V5 was not suitable, V4 or V6 were measured, respectively. A mean value of three readings was calculated for each lead. e QT interval was measured from the beginning of the QRS complex to the end of the T wave. To adjust QTfor heart rate, we calculated QTc according to Bazett's formula QTc � QT/√R ̅ R ̅ , where RR is the RR interval in seconds. e Tp-e interval was de ned as the interval from the peak of T wave to the end of T wave. e Tp-e/corrected QT ratio was calculated from these measurements. QTc dispersion is determined as the di erence between the maximum and minimum QTc in all considered leads. All ECG measurements were performed by a cardiologist who was blinded to other patient information.

Echocardiographic Evaluation.
We used a 2-dimensional, M-mode cardiovascular ultrasound system (Vingmed Vivid 7 system GE, Germany) for echocardiographic evaluation. e examinations were performed in the left lateral decubitus position. Parasternal long-and short-axis views and apical views were used as standard imaging windows. We measured left atrial and left ventricular end systolic diameters from parasternal long-axis view. Ejection fraction was calculated by using modi ed Simpson method. All echocardiographic examinations were performed by an experienced cardiologist who was blinded to patient data.

Holter Assessment. Holter ECG (medilog ® AR4 plus
Holter system, Schiller, Switzerland) recordings were analyzed using a MARS 8000 Holter scanner (GE Medical Systems, Milwaukee, Wisconsin). We determined all PVC, atrial brillation (AF), and ventricular tachycardia (VT) episodes. e percentage of PVCs was calculated as the total number of ventricular ectopic beats/the total number of beats recorded during 24 hour Holter ECG monitoring.

Ambulatory Blood Pressure
Monitoring. Ambulatory blood pressure monitoring was carried for 24 h using an automated sphygmomanometer (BR-102 PLUS PWA, Schiller, Switzerland). e blood pressure was recorded every 20 min during the 24 h period. Mean values were recorded before and after CPAP treatment.

Blood Sampling and Analysis.
Fasting blood samples were collected from an antecubital vein in the morning and were centrifuged within one hour. Serum samples were stored at −80°C pending later measurements. N-terminal pro-brain natriuretic peptide (NT-proBNP) was measured by electrochemiluminescent immunoassay (Roche Diagnostics, Basel, Switzerland). Creatinine and hemoglobin levels were measured by standard laboratory methods. Serum NT-proBNP level was measured at baseline and at the end of 6-month follow-up in both groups. e estimated glomerular ltration rate (mL/min per 1.73 m 2 ) was calculated to exclude renal dysfunction as a cause for increased NT-proBNP concentrations [23].
2.9. Statistical Analysis. All data were entered into a spreadsheet, and statistical analyses were performed by using R 3.3.2v (open source). Data were shown as mean ± standard deviation for continuous variables, median (minimum-maximum) for ordinal ones, and frequency with percent for categorical ones. Comparisons for continuous variables were made by using independent samples t-test and for categorical variables Fisher's exact test. Normally distributed variables were compared with paired t-test and nonnormally distributed variables were compared with Mann-Whitney U test for intragroup and intergroup di erences. Two-Way Repeated Measures of ANOVA was also used for intragroup comparisons. A p value less than 0.05 was considered statistically signi cant.

Results
e mean age of the study group was 58.4 ± 8.14 years and the control group was 59.18 ± 7.97 years. e female/male ratio in the control group was 8/32 and in the control group was 9/31. ere was no signi cant di erence between the groups in terms of age and sex (p � 0.67 and p � 0.79). Baseline characteristics and other demographics of the study and control groups are summarized in Table 1.
Other baseline characteristics including BMI, ejection fraction, AHI, oxygen saturation, heart rate, and blood pressure were also similar between both groups (p > 0.05 for all). Drugs used by the patients in both groups are summarized in Table 2. Drug usage characteristics of the study and control groups were also similar. No medicational change was done during follow-up especially in terms of antiarrhythmic drugs.
At the end of 6 months of CPAP treatment, frequency of PVC, Tp-e, QTc, QTc dispersion, and Tp-e/QTc ratio decreased signi cantly in the study group (p < 0.001 for all). ese parameters did not change signi cantly in the control group after 6 months (p > 0.05 for all). Table 3 summarizes the chances of frequency of PVC, Tp-e, QTc, QTc dispersion, and Tp-e/QTc ratio during the follow-up period in both groups.
CPAP treatment signi cantly reduced the heart rate and systolic and diastolic blood pressure and increased the Canadian Respiratory Journal ejection fraction in the study group (p < 0.05 for all). BMI and LVEDD did not change signi cantly after 6 months of CPAP treatment. Table 4 summarizes the BMI, ejection fraction, heart rate, blood pressure, and LVEDD measures of the groups in detail.
Appropriate ICD shocks during follow-up in study and control groups due to VT/VF were seen in 5 patients and 9 patients, respectively (%12.5 versus %22.5, p � 0.24). Inappropriate ICD shocks were much more in control group (%2.5 versus %7.5, p � 0.62) but did not reach statistical signi cance. Also atrial brillation rate detected by ICD devices was higher in control group but again did not have statistical signi cance.
Baseline NT-Pro BNP levels were 43.88 ± 7.79 pg/mL in the study group and 44.65 ± 7.79 pg/mL in the control group.
e NT-Pro BNP levels at 6th month were 39.18 ± 7.57 pg/mL in the study group and 46.11 ± 7.65 pg/mL in the control group. ere was a statistically signi cant di erence between baseline and 6th-month measurements in the study group in terms of NT-Pro BNP (p < 0.001). ere was also a statistically signi cant di erence between the study and control groups at 6th month (p < 0.001), but not  Figure 1 shows the baseline and 6th-month NT-pro-BNP levels in the groups.

Discussion
Individuals with OSAS have frequent cessation or reduction of air ow during sleep that results in hypoxemia and arousals from sleep. Continuous positive airway pressure (CPAP) is the standard rst-line treatment for OSAS [24].
One metaanalysis from consistent evidence from good-and fair-quality randomized controlled trials reported that CPAP e ectively reduces AHI and excessive sleepiness in patients with OSAS [25]. Several studies showed the bene cial e ects of CPAP on cardiac functions in OSAS patients [26,27]. e possible mechanism of action of CPAP may include improved myocardial oxygen delivery, decreased sympathetic activity, left ventricular transmural pressure, and afterload [28]. In the present study, we have found bene cial e ects of CPAP treatment in patients with OSAS   and heart failure in terms of reduction in premature ventricular contractions and ventricular wall stress. e issue of whether OSAS is an independent risk factor for patients with heart failure is controversial. e results of published studies on this subject vary signi cantly [29,30]. Lanfranchi et al. [29] reported approximately 4-fold increased mortality rates in patients with OSAS and heart failure compared to heart failure patients without OSAS. Fries et al. [30] investigated the prognostic value of OSAS in congestive heart failure patients with ICD implants and found that mortality rates were 44% in central sleep apnea and 13% in the absence of apnea at the 2 years of follow-up (p < 0.05), but this study found no association between sleep apnea and appropriate ICD therapy for VT or VF. Adlakha and Shepard [31] speculated that hypoxia in the obstructive sleep apnea leads to cardiac arrhythmias. Sudden deaths, particularly at night, in OSAS may be due to these arrhythmias [19,32]. Clay et al. [33] reported that OSAS patients with ICD had higher rates of ventricular ectopia and nonsustained ventricular tachycardia. Javaheri [34] found an increased frequency of VT and VES in patients with congestive heart failure and OSAS. Bitter et al. [35] speculated that OSAS is an independent risk factor for malignant ventricular arrhythmias that require ICD treatment. We found reduced premature ventricular contraction rates after 6 months of CPAP treatment in patients with OSAS and heart failure compared to untreated controls. We did not make major changes in the heart failure treatment of patients. So, favorable results would be attributable to CPAP treatment. Ventricular repolarization and arrhythmogenesis can be assessed by several electrocardiographic parameters including QTc interval, QT dispersion, and T wave measurements [36]. Tp-e is used as an indicator of transmural dispersion of ventricular repolarization [37]. Prolonged Tp-e interval may predict ventricular arrhythmias and mortality [38,39].
us, Tp-e/QTc ratio has been proposed to be a more speci c index of ventricular arrhythmogenesis [40,41]. Kilicaslan et al. [36] reported prolonged Tp-e interval and increased Tp-e/QTc ratio in patients with moderate and severe OSAS as compared to healthy controls. eir nding of Tp-e and QTc interval in OSAS patients were comparable with the present study results. We found that, CPAP treatment signi cantly reduced the frequency of PVC, Tp-e interval, QTc interval, QTc dispersion, and Tp-e/QTc ratio in the study group, whereas the control group did not show any signi cant di erence for the mentioned parameters.
Myocytes of cardiac ventricle constitute the major source of BNP-related neurohormones. B-type natriuretic peptide (BNP) is a 32 amino acid peptide primarily synthesized, stored, and secreted from left ventricle. NT proBNP is derived from the BNP and secreted when the ventricular wall stress increased [42,43]. It has been used as an indicator of heart failure and symptomatic idiopathic PVC [42,44,45]. NT proBNP suggested as a sensitive indicator of PVCinduced increased ventricular wall stress [44,45]. Skranes et al. [46] reported that the higher levels of NT-proBNP levels are independently associated with the incidence of frequent ventricular ectopy and complex ventricular ectopy in a moderately large, community-based population. Another study [47] found that fatigue was associated with higher baseline NT-proBNP in patients with frequent PVCs and preserved LV function. Tasci et al. [48] reported that nasal CPAP treatment signi cantly reduced NT-proBNP in patients with normotensive and hypertensive OSAS. Similarly, Strehmel et al. reported reduced NT-proBNP levels after CPAP treatment in patients with OSAS and coronary artery disease [49]. In the present study, we have found that 6 months CPAP treatment for OSAS signi cantly reduced the NT-proBNP levels which means reduced ventricular wall stress. e present study has some limitations. We did not evaluate the AHI index using polysomnography at the 6-month followup. erefore, an association between AHI and cardiac functions remained unclari ed. e randomization in this study depended on the patient preferences (whether or not CPAP treatment is desired). However, real randomization in this context carries ethical considerations. e relatively small number of patients in our study is another limitation. A larger patient population with a longer follow-up would provide more precise results.

Conclusions
We have found that the frequency of PVC and NT-proBNP levels can be reduced by CPAP treatment in patients with heart failure and OSAS. CPAP therapy has been shown to signi cantly improve cardiac functions. Further studies are needed to clarify the possible relationships between CPAP and cardiac functions in detail.   Canadian Respiratory Journal