Canadian Journal of Cardiology 31 (2015) 853e859
Clinical Research
Predictors of Blood Pressure Fall With Continuous Positive Airway Pressure Treatment in Hypertension With Coronary Artery Disease and Obstructive Sleep Apnea Zhiwei Huang, MD, Zhihong Liu, MD, PhD, Qin Luo, MD, Qing Zhao, MD, Zhihui Zhao, MD, Xiuping Ma, BS, Qunying Xi, MD, and Dan Yang, MD State Key Laboratory of Cardiovascular Disease, Sleep Disorders Research Center, Fuwai Hospital, National Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China
ABSTRACT
RESUM E
Background: The present study aimed to investigate the predictors of changes in blood pressure (BP) with continuous positive airway pressure (CPAP) treatment in hypertensive patients with coronary heart disease (CHD) and obstructive sleep apnea (OSA). Methods: Seventy-one hypertensive patients with CHD and OSA were enrolled in this study. Daytime systolic BP (SBP), diastolic BP (DBP), Epworth Sleepiness Scale (ESS), and anthropometric characteristics were assessed at baseline and follow-up. Results: Sixty-six patients completed the study. The median follow-up period was 36 months (interquartile range, 24-60 months). The mean duration of CPAP application was 4.3 1.2 hours per night. From baseline to follow-up, SBP and DBP were reduced by 5.6 mm Hg (95% confidence interval [CI], 3.0-8.1) and 3.0 mm Hg (95% CI, 0.85.3), respectively. Daytime somnolence was significantly improved (ESS, from 9.5 3.4 at baseline to 3.6 2.0 at follow-up; P < 0.001); the mean improvement in ESS was 6.0 (95% CI, 5.1-6.9). Correlation analysis of the fall in mean BP (MBP) showed that baseline MBP, change in ESS, heart rate, and CPAP compliance
sente e tude avait pour but d’examiner les Introduction : La pre dicteurs de changements dans la pression arte rielle (PA) par le pre e en pression positive expiratoire traitement par ventilation spontane continue (CPAP : continuous positive airway pressure) des patients e obstructive du hypertendus souffrant de coronaropathie et d’apne sommeil (AOS). thodes : Soixante-et-onze (71) patients hypertendus souffrant de Me te recrute s pour cette e tude. La PA coronaropathie et d’AOS ont e chelle systolique (PAS) et la PA diastolique (PAD) diurnes ainsi que l’e ristiques anthrode somnolence d’Epworth (ESE) et les caracte triques ont e te e value es au de but et durant le suivi. pome sultats : Soixante-six (66) patients ont mene à terme l’e tude. La Re e me diane de suivi a e te de 36 mois (intervalle interquartile, dure e moyenne de l’utilisation de la CPAP a e te de 4,3 24-60 mois). La dure but et la pe riode de suivi, la PAS et la 1,2 heures par nuit. Entre le de de 5,6 mm Hg (intervalle de conPAD ont respectivement diminue fiance [IC] à 95 %, 3,0-8,1) et de 3,0 mm Hg (IC à 95 %, 0.8-5,3). La liore e (ESE, de 9,5 3,4 somnolence diurne s’est significativement ame
Obstructive sleep apnea (OSA), an important public health problem, is characterized by recurrent complete (apnea) or partial (hypopnea) upper airway obstruction during sleep.1 Such repeated airway obstructions often expose patients with OSA to intermittent hypoxemia/hypercapnia and sleep fragmentation. There is considerable evidence that OSA carries substantial morbidity and mortality, particularly from cardiovascular complications.2,3 The current prevalence estimates of moderate to severe sleep-disordered breathing (apnea-to-hypopnea index, measured as events per hour 15)
are 10% among 30- to 49-year-old men, 17% among 50- to 70-year-old men, 3% among 30- to 49-year-old women, and 9% among 50- to 70-year-old women.4 One study has shown that 56% of hypertensive patients exhibit OSA.5 Meanwhile, the most common cause of resistant hypertension and poor response to antihypertensive medications is OSA.6,7 This is likely to be a risk factor for hypertension; however, some studies have demonstrated the lack of a relationship between OSA and hypertension.8-11 Most studies indicate that continuous positive airway pressure (CPAP) is effective in reducing blood pressure (BP), although the reported efficacy has been variable. Several randomized controlled trials (RCTs) have shown that CPAP decreases BP in patients with OSA,12-15 although the findings of recent large RCTs suggest that the observed average decrease in BP is relatively small.16-18 Further, a recent systematic review and metaanalysis, which included 16 RCTs with 1000 treated patients with OSA, showed similar results, ie, CPAP treatment promoted significant but small reductions in BP in individuals
Received for publication July 17, 2014. Accepted September 18, 2014. Corresponding author: Dr Zhihong Liu, Sleep Disorders Research Center, Fuwai Hospital, National Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences and Peking Union Medical College, No. 167 Beilishi Road, Xicheng District, Beijing 100037, China. Tel. þ86-01088398196; fax: þ86-010-88396589. E-mail: [email protected] See page 858 for disclosure information.
http://dx.doi.org/10.1016/j.cjca.2014.09.015 0828-282X/Ó 2015 Canadian Cardiovascular Society. Published by Elsevier Inc. All rights reserved.
854
Canadian Journal of Cardiology Volume 31 2015
showed a positive correlation, whereas the baseline body mass index (BMI) and ESS had an inverse relationship. Stepwise multiple linear regression analysis, however, indicated that only baseline BMI, baseline MBP, and CPAP compliance were independently correlated with the fall in MBP. Conclusions: Long-term CPAP treatment reduces BP in hypertensive patients with CHD and moderate/severe OSA; baseline BMI, baseline MBP, and CPAP compliance are independent predictors of the decrease in BP with CPAP treatment in these patients.
but à 3,6 2,0 durant le suivi; P < 0,001); l’ame lioration au de te de 6,0 (IC à 95 %, 5,1-6,9). L’analyse de moyenne selon l’ESE a e lation de la baisse de la PA moyenne (PAM) a montre que la PAM corre initiale, le changement à l’ESE, le rythme cardiaque et l’observance de lation positive, tandis que l’indice de masse la CPAP avaient une corre but avaient une relation inverse. corporelle (IMC) et l’ESE au de quentielle de re gression line aire multiple a Cependant, l’analyse se que seuls l’IMC initial, la PAM initiale et l’observance de la indique pendamment corre le avec la baisse de la PAM. CPAP avaient inde duit la PA des Conclusions : Le traitement à long terme par CPAP re re e et patients hypertendus souffrant de coronaropathie et d’AOS mode grave; l’IMC initial, la PAM initiale et l’observance de la CPAP sont des dicteurs inde pendants de la diminution de la PA de ces patients pre s par CPAP. traite
with OSA.19 In clinical practice, however, a reduction in BP is not observed in all patients undergoing CPAP. Surprisingly, some patients even demonstrate an increase in BP after CPAP treatment. The exact mechanism underlying this difference in the effects of CPAP therapy remains unclear. Uncontrolled hypertension may exacerbate the conditions of patients with coronary heart disease (CHD) and OSA. There are scarce data regarding the predictors of a decrease in BP with CPAP treatment in hypertensive patients with CHD and OSA. Examining these predictors may provide a deeper insight into the mechanism of hypertension associated with OSA. The aim of this study was thus to evaluate the predictors of changes in BP after CPAP therapy in hypertensive patients with CHD and OSA.
pharyngeal surgery for OSA, or current use of CPAP treatment for OSA. They were also excluded if they declined to participate or were unable to provide informed consent.
Methods Design and setting From January 2009-December 2013, we performed a prospective cohort study including 71 patients attending the Sleep and CHD Trials Unit, Fuwai Hospital, National Centre for Cardiovascular Medicine. This study was approved by the Ethics Committee of Fuwai Hospital. All participants provided written informed consent. Selection of patients Consecutive patients of both sexes from outpatient and inpatient departments of Fuwai Hospital were recruited into the long-term follow-up study. Patients were eligible for the trial if they were newly diagnosed with moderate to severe OSA, hypertension, and CHD. Further, the following criteria were also fulfilled: (1) both men and women were included and (2) patients who had received at least 3 months of previous standardized treatment for hypertension according to the current guideline were included.20 The participants were excluded if they had secondary hypertension (including primary aldosteronism, renal artery stenosis, or chronic renal disease), central sleep apnea, New York Heart Association class III/IV, diagnosis of malignant cancer with a life expectancy of < 2 years, severe hepatic failure or pulmonary disease, long-term use of drugs known to have an impact on BP (including corticosteroids or sedative drugs), significant psychiatric disease, chronic alcohol use or addiction, history of
Procedures Initial visit Hypertensive patients with CHD suspected of having OSA were screened after obtaining written informed consent. Hypertension was defined as (1) systolic BP (SBP) 140 mm Hg or diastolic BP (DBP) 90 mm Hg (or both) at rest or (2) as ongoing treatment with antihypertensive drugs. CHD was diagnosed based on the results of selective coronary angiography showing at least 1 major epicardial coronary artery luminal stenosis segment 70% or stenosis in the left main coronary artery 50% or based on a history of myocardial infarction or coronary artery bypass grafting documented in medical records. Moderate OSA was defined as an apnea-hypopnea index (AHI) of 15-29 episodes per hour and severe OSA as an AHI of at least 30 episodes per hour. When participants were diagnosed with moderate to severe OSA, demographic data including age, sex, medical history, current medicine, lifestyle habits, height (cm), weight (kg), waist circumference (cm), hip circumference (cm), and neck circumference (cm) were recorded, and body mass index (BMI) was calculated as weight divided by height squared (kg/m2). The Epworth Sleepiness Scale (ESS) was used to quantify daytime somnolence.21 Patients underwent regular clinical examinations and laboratory testing to exclude secondary hypertension. Participants were assigned antihypertensive and CHD drug treatment according to current guidelines in a 3-month run-in period that allowed for modifications in the therapeutic regimen. Patients were asked to bring the empty blister packs of their pills to evaluate compliance for drug therapy at each visit. Sleep studies All patients underwent overnight polysomnography in the Sleep Center of Fuwai Hospital. The Embletta X100 (Medcare Flaga, Reykjavik, Iceland) was used, as described previously.22 Apneic episodes were defined as airflow reduction to 10% or less of the baseline value for 10 seconds. Hypopnea was defined as a 30%-90% reduction in the oronasal airflow for > 10 seconds, associated with an oxygen desaturation of 4%. Central sleep
Huang et al. Predictors of BP Fall in Patients With OSA
apnea was defined as at least 50% of respiratory events having a pattern of apnea or hypopnea without thoracic and abdominal movement.18 The AHI was calculated by dividing the total number of apneic and hypopneic episodes by the total sleep time. Subsequently, participants underwent a full-night CPAP titration study and received fixed-level CPAP titration. CPAP application The CPAP group received fixed-level CPAP titration using an automated pressure setting device for 1 night in a sleep laboratory. The optimal CPAP pressure for each patient in the CPAP group was set at the minimum pressure required to abolish snoring, obstructive respiratory events, and airflow limitation for 95% of the night, according to a previous validation study by our group.22 Fixed pressure was then maintained throughout the study for patients when using the CPAP machine. CPAP compliance was objectively measured by the built-in compliance software of CPAP devices at regular examination at follow-up. Each patient received standardized instructions by 1 investigator specializing in OSA and by a home health care provider at the start of the CPAP treatment. A specialist OSA team assisted patients with telephone or outpatient advice for any difficulties with CPAP during the study, and masks were adjusted as necessary. BP measurement The BP in the morning was measured between 8 and 10 AM and determined by the average results of 3 readings of SBP and DBP obtained on the same arm of the patient at 5-minute intervals using a mercury sphygmomanometer with an appropriately sized cuff after participants had been seated in a chair with feet on the floor and arm supported at heart level for at least 10 minutes. The first and last Korotkoff sounds were used to determine SBP and DBP, respectively. The mean BP (MBP) was calculated by one-third SBP and two-thirds DBP. Follow-up Patients visited the sleep research laboratory at 1 month and 3 months after CPAP application and then every 6 months. Follow-up was carried out in all participants by the same investigator. The following were recorded at every medical appointment: the morning office BP, heart rate (HR) by electrocardiography, adherence to CPAP, medical treatment, lifestyle habits, height, weight, waist circumference, hip circumference, neck circumference, ESS, and cardiovascular or cerebrovascular events, or both. CPAP compliance was objectively measured by the built-in compliance software of CPAP devices. We considered adherence as adequate if the mean CPAP use was at least 4 hours per night. Statistical analysis Continuous variables with normal distribution were expressed as mean SD and as median and interquartile range without normal distribution, whereas categorical variables were presented as absolute numbers and percentages. A 2-tailed paired Student t test or a c2 test was used to compare variables between baseline and the end of follow-up. The 2sample t test was used for normally distributed variables, and a Mann-Whitney test was used for non-normally
855
distributed variables in subgroup analysis. Altman recommended that there be 10 participants per predictive variable in a multiple linear regression model.23 Further, bivariate analysis suggests that it requires at least 60 participants to explore a maximum of 6 variables. The change in MBP between baseline and follow-up was a dependent variable, whereas the other measurements (including baseline BMI, ESS, AHI, MBP, change in ESS and HR, and CPAP compliance) were independent variables. These were assessed with linear modeling analysis (stepwise multiple linear regression). PASW Statistics, version 18 (SPSS, Chicago, IL) was used for statistical analysis. Appropriate 95% CIs were calculated, and a 2-sided P value < 0.05 was considered statistically significant. Results We initially recruited 238 hypertensive patients with newly diagnosed CHD and clinical suspicion of OSA. Of these, 89 (37.4%) were diagnosed with OSA; 6 patients declined to take part in the study, and 12 patients were excluded, including 9 patients with AHI < 15, 1 patient with severe pulmonary disease, and 2 patients with systolic heart failure and poor cardiac function (ejection fraction < 35%). Thus, 71 patients were enrolled in the study. Of these participants, 3 patients withdrew from the study before the 1-month visit because of intolerance to the CPAP treatment, and 2 participants lacked complete data. Eventually, 66 patients completed the study. The median duration of follow-up was 36 months (24-60 months). The baseline and follow-up characteristics of the included patients are summarized in Table 1. The mean age was 62.6 6.7 years. There were 58 (87.9%) male patients in this study. The mean BMI was 28.3 3.2 kg/m2 and 20 (30.3%) participants were obese. Further, 30 (45.5%) participants had an ESS score > 10 and 25 (37.9%) participants had severe OSA. With respect to cardiovascular events, 3 patients experienced unstable angina, and 1 patient was hospitalized for heart failure during the course of follow-up. The mean time of CPAP application was 4.3 1.2 hours per night. The mean AHI derived from the CPAP machine at the end of follow-up was 2.9 1.4 events per hour, indicating good control of OSA. Compared with baseline, values for SBP and DBP showed a reduction (SBP, 141.1 12.4 mm Hg vs 135.5 11.0 mm Hg; DBP, 83.8 7.6 mm Hg vs 80.7 7.3 mm Hg; both P < 0.05) (Fig. 1). From baseline to followup, SBP and DBP were reduced by 5.6 mm Hg (95% CI, 3.0-8.1) and 3.0 mm Hg (95% CI, 0.8-5.3), respectively. Daytime somnolence was significantly improved (ESS, from 9.5 3.4 at baseline to 3.6 2.0 at follow-up; P < 0.001). The mean improvement in ESS was 6.0 (95% CI, 5.1-6.9) at follow-up. Post-treatment HR was lower than that at baseline (64.7 7.2 bpm vs 67.4 7.8 bpm; P ¼ 0.003). The results of the correlation analysis of the fall in MBP between baseline and follow-up are shown in Table 2. The change in mean MBP at follow-up correlated closely with the baseline MBP (r ¼ 5.6; P < 0.001). Using an illustrative post hoc analysis, we found that patients with a baseline MBP value 100 mm Hg had a significant reduction of 6.1 mm Hg in MBP compared with a mean fall of 2.0 mm Hg in those with a baseline MBP < 100 mm Hg (P < 0.001). The change in mean MBP at follow-up was also significantly correlated with CPAP compliance (r ¼ 0.37; P ¼ 0.002). An
856
Canadian Journal of Cardiology Volume 31 2015
Table 1. Baseline and follow-up characteristics of study patients (n ¼ 66) Variable Age (y) Male sex (n, %) BMI (kg/m2) Neck circumference (cm) Heart rate (bpm) ESS (points) AHI (events/h) SBP (mm Hg) DBP (mm Hg) MBP (mm Hg) Antihypertensive drug use b-blocker use (n, %) ACEI use (n, %) ARB use (n, %) CCB use (n, %) Diuretic use (n, %) CPAP compliance (h/night)
Baseline 62.6 6.7 58 (87.9) 28.3 3.2 41.0 3.2 67.4 7.8 9.5 3.4 28.3 12.6 141.1 12.4 83.8 7.6 102.8 6.5 3.0 0.8 56 (84.8) 30 (45.5) 25 (37.9) 37 (56.1) 60 (90.9)
Follow-up
P value
27.8 3.1 40.9 3.5 64.7 7.2 3.6 2.0 2.9 1.4 135.5 11.0 80.7 7.3 98.8 6.7 3.2 0.7 57 (86.3) 32 (48.5) 27 (40.9) 35 (53.0) 61 (92.4) 4.3 1.2
0.137 0.541 0.003
Clinical Research
Predictors of Blood Pressure Fall With Continuous Positive Airway Pressure Treatment in Hypertension With Coronary Artery Disease and Obstructive Sleep Apnea Zhiwei Huang, MD, Zhihong Liu, MD, PhD, Qin Luo, MD, Qing Zhao, MD, Zhihui Zhao, MD, Xiuping Ma, BS, Qunying Xi, MD, and Dan Yang, MD State Key Laboratory of Cardiovascular Disease, Sleep Disorders Research Center, Fuwai Hospital, National Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China
ABSTRACT
RESUM E
Background: The present study aimed to investigate the predictors of changes in blood pressure (BP) with continuous positive airway pressure (CPAP) treatment in hypertensive patients with coronary heart disease (CHD) and obstructive sleep apnea (OSA). Methods: Seventy-one hypertensive patients with CHD and OSA were enrolled in this study. Daytime systolic BP (SBP), diastolic BP (DBP), Epworth Sleepiness Scale (ESS), and anthropometric characteristics were assessed at baseline and follow-up. Results: Sixty-six patients completed the study. The median follow-up period was 36 months (interquartile range, 24-60 months). The mean duration of CPAP application was 4.3 1.2 hours per night. From baseline to follow-up, SBP and DBP were reduced by 5.6 mm Hg (95% confidence interval [CI], 3.0-8.1) and 3.0 mm Hg (95% CI, 0.85.3), respectively. Daytime somnolence was significantly improved (ESS, from 9.5 3.4 at baseline to 3.6 2.0 at follow-up; P < 0.001); the mean improvement in ESS was 6.0 (95% CI, 5.1-6.9). Correlation analysis of the fall in mean BP (MBP) showed that baseline MBP, change in ESS, heart rate, and CPAP compliance
sente e tude avait pour but d’examiner les Introduction : La pre dicteurs de changements dans la pression arte rielle (PA) par le pre e en pression positive expiratoire traitement par ventilation spontane continue (CPAP : continuous positive airway pressure) des patients e obstructive du hypertendus souffrant de coronaropathie et d’apne sommeil (AOS). thodes : Soixante-et-onze (71) patients hypertendus souffrant de Me te recrute s pour cette e tude. La PA coronaropathie et d’AOS ont e chelle systolique (PAS) et la PA diastolique (PAD) diurnes ainsi que l’e ristiques anthrode somnolence d’Epworth (ESE) et les caracte triques ont e te e value es au de but et durant le suivi. pome sultats : Soixante-six (66) patients ont mene à terme l’e tude. La Re e me diane de suivi a e te de 36 mois (intervalle interquartile, dure e moyenne de l’utilisation de la CPAP a e te de 4,3 24-60 mois). La dure but et la pe riode de suivi, la PAS et la 1,2 heures par nuit. Entre le de de 5,6 mm Hg (intervalle de conPAD ont respectivement diminue fiance [IC] à 95 %, 3,0-8,1) et de 3,0 mm Hg (IC à 95 %, 0.8-5,3). La liore e (ESE, de 9,5 3,4 somnolence diurne s’est significativement ame
Obstructive sleep apnea (OSA), an important public health problem, is characterized by recurrent complete (apnea) or partial (hypopnea) upper airway obstruction during sleep.1 Such repeated airway obstructions often expose patients with OSA to intermittent hypoxemia/hypercapnia and sleep fragmentation. There is considerable evidence that OSA carries substantial morbidity and mortality, particularly from cardiovascular complications.2,3 The current prevalence estimates of moderate to severe sleep-disordered breathing (apnea-to-hypopnea index, measured as events per hour 15)
are 10% among 30- to 49-year-old men, 17% among 50- to 70-year-old men, 3% among 30- to 49-year-old women, and 9% among 50- to 70-year-old women.4 One study has shown that 56% of hypertensive patients exhibit OSA.5 Meanwhile, the most common cause of resistant hypertension and poor response to antihypertensive medications is OSA.6,7 This is likely to be a risk factor for hypertension; however, some studies have demonstrated the lack of a relationship between OSA and hypertension.8-11 Most studies indicate that continuous positive airway pressure (CPAP) is effective in reducing blood pressure (BP), although the reported efficacy has been variable. Several randomized controlled trials (RCTs) have shown that CPAP decreases BP in patients with OSA,12-15 although the findings of recent large RCTs suggest that the observed average decrease in BP is relatively small.16-18 Further, a recent systematic review and metaanalysis, which included 16 RCTs with 1000 treated patients with OSA, showed similar results, ie, CPAP treatment promoted significant but small reductions in BP in individuals
Received for publication July 17, 2014. Accepted September 18, 2014. Corresponding author: Dr Zhihong Liu, Sleep Disorders Research Center, Fuwai Hospital, National Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences and Peking Union Medical College, No. 167 Beilishi Road, Xicheng District, Beijing 100037, China. Tel. þ86-01088398196; fax: þ86-010-88396589. E-mail: [email protected] See page 858 for disclosure information.
http://dx.doi.org/10.1016/j.cjca.2014.09.015 0828-282X/Ó 2015 Canadian Cardiovascular Society. Published by Elsevier Inc. All rights reserved.
854
Canadian Journal of Cardiology Volume 31 2015
showed a positive correlation, whereas the baseline body mass index (BMI) and ESS had an inverse relationship. Stepwise multiple linear regression analysis, however, indicated that only baseline BMI, baseline MBP, and CPAP compliance were independently correlated with the fall in MBP. Conclusions: Long-term CPAP treatment reduces BP in hypertensive patients with CHD and moderate/severe OSA; baseline BMI, baseline MBP, and CPAP compliance are independent predictors of the decrease in BP with CPAP treatment in these patients.
but à 3,6 2,0 durant le suivi; P < 0,001); l’ame lioration au de te de 6,0 (IC à 95 %, 5,1-6,9). L’analyse de moyenne selon l’ESE a e lation de la baisse de la PA moyenne (PAM) a montre que la PAM corre initiale, le changement à l’ESE, le rythme cardiaque et l’observance de lation positive, tandis que l’indice de masse la CPAP avaient une corre but avaient une relation inverse. corporelle (IMC) et l’ESE au de quentielle de re gression line aire multiple a Cependant, l’analyse se que seuls l’IMC initial, la PAM initiale et l’observance de la indique pendamment corre le avec la baisse de la PAM. CPAP avaient inde duit la PA des Conclusions : Le traitement à long terme par CPAP re re e et patients hypertendus souffrant de coronaropathie et d’AOS mode grave; l’IMC initial, la PAM initiale et l’observance de la CPAP sont des dicteurs inde pendants de la diminution de la PA de ces patients pre s par CPAP. traite
with OSA.19 In clinical practice, however, a reduction in BP is not observed in all patients undergoing CPAP. Surprisingly, some patients even demonstrate an increase in BP after CPAP treatment. The exact mechanism underlying this difference in the effects of CPAP therapy remains unclear. Uncontrolled hypertension may exacerbate the conditions of patients with coronary heart disease (CHD) and OSA. There are scarce data regarding the predictors of a decrease in BP with CPAP treatment in hypertensive patients with CHD and OSA. Examining these predictors may provide a deeper insight into the mechanism of hypertension associated with OSA. The aim of this study was thus to evaluate the predictors of changes in BP after CPAP therapy in hypertensive patients with CHD and OSA.
pharyngeal surgery for OSA, or current use of CPAP treatment for OSA. They were also excluded if they declined to participate or were unable to provide informed consent.
Methods Design and setting From January 2009-December 2013, we performed a prospective cohort study including 71 patients attending the Sleep and CHD Trials Unit, Fuwai Hospital, National Centre for Cardiovascular Medicine. This study was approved by the Ethics Committee of Fuwai Hospital. All participants provided written informed consent. Selection of patients Consecutive patients of both sexes from outpatient and inpatient departments of Fuwai Hospital were recruited into the long-term follow-up study. Patients were eligible for the trial if they were newly diagnosed with moderate to severe OSA, hypertension, and CHD. Further, the following criteria were also fulfilled: (1) both men and women were included and (2) patients who had received at least 3 months of previous standardized treatment for hypertension according to the current guideline were included.20 The participants were excluded if they had secondary hypertension (including primary aldosteronism, renal artery stenosis, or chronic renal disease), central sleep apnea, New York Heart Association class III/IV, diagnosis of malignant cancer with a life expectancy of < 2 years, severe hepatic failure or pulmonary disease, long-term use of drugs known to have an impact on BP (including corticosteroids or sedative drugs), significant psychiatric disease, chronic alcohol use or addiction, history of
Procedures Initial visit Hypertensive patients with CHD suspected of having OSA were screened after obtaining written informed consent. Hypertension was defined as (1) systolic BP (SBP) 140 mm Hg or diastolic BP (DBP) 90 mm Hg (or both) at rest or (2) as ongoing treatment with antihypertensive drugs. CHD was diagnosed based on the results of selective coronary angiography showing at least 1 major epicardial coronary artery luminal stenosis segment 70% or stenosis in the left main coronary artery 50% or based on a history of myocardial infarction or coronary artery bypass grafting documented in medical records. Moderate OSA was defined as an apnea-hypopnea index (AHI) of 15-29 episodes per hour and severe OSA as an AHI of at least 30 episodes per hour. When participants were diagnosed with moderate to severe OSA, demographic data including age, sex, medical history, current medicine, lifestyle habits, height (cm), weight (kg), waist circumference (cm), hip circumference (cm), and neck circumference (cm) were recorded, and body mass index (BMI) was calculated as weight divided by height squared (kg/m2). The Epworth Sleepiness Scale (ESS) was used to quantify daytime somnolence.21 Patients underwent regular clinical examinations and laboratory testing to exclude secondary hypertension. Participants were assigned antihypertensive and CHD drug treatment according to current guidelines in a 3-month run-in period that allowed for modifications in the therapeutic regimen. Patients were asked to bring the empty blister packs of their pills to evaluate compliance for drug therapy at each visit. Sleep studies All patients underwent overnight polysomnography in the Sleep Center of Fuwai Hospital. The Embletta X100 (Medcare Flaga, Reykjavik, Iceland) was used, as described previously.22 Apneic episodes were defined as airflow reduction to 10% or less of the baseline value for 10 seconds. Hypopnea was defined as a 30%-90% reduction in the oronasal airflow for > 10 seconds, associated with an oxygen desaturation of 4%. Central sleep
Huang et al. Predictors of BP Fall in Patients With OSA
apnea was defined as at least 50% of respiratory events having a pattern of apnea or hypopnea without thoracic and abdominal movement.18 The AHI was calculated by dividing the total number of apneic and hypopneic episodes by the total sleep time. Subsequently, participants underwent a full-night CPAP titration study and received fixed-level CPAP titration. CPAP application The CPAP group received fixed-level CPAP titration using an automated pressure setting device for 1 night in a sleep laboratory. The optimal CPAP pressure for each patient in the CPAP group was set at the minimum pressure required to abolish snoring, obstructive respiratory events, and airflow limitation for 95% of the night, according to a previous validation study by our group.22 Fixed pressure was then maintained throughout the study for patients when using the CPAP machine. CPAP compliance was objectively measured by the built-in compliance software of CPAP devices at regular examination at follow-up. Each patient received standardized instructions by 1 investigator specializing in OSA and by a home health care provider at the start of the CPAP treatment. A specialist OSA team assisted patients with telephone or outpatient advice for any difficulties with CPAP during the study, and masks were adjusted as necessary. BP measurement The BP in the morning was measured between 8 and 10 AM and determined by the average results of 3 readings of SBP and DBP obtained on the same arm of the patient at 5-minute intervals using a mercury sphygmomanometer with an appropriately sized cuff after participants had been seated in a chair with feet on the floor and arm supported at heart level for at least 10 minutes. The first and last Korotkoff sounds were used to determine SBP and DBP, respectively. The mean BP (MBP) was calculated by one-third SBP and two-thirds DBP. Follow-up Patients visited the sleep research laboratory at 1 month and 3 months after CPAP application and then every 6 months. Follow-up was carried out in all participants by the same investigator. The following were recorded at every medical appointment: the morning office BP, heart rate (HR) by electrocardiography, adherence to CPAP, medical treatment, lifestyle habits, height, weight, waist circumference, hip circumference, neck circumference, ESS, and cardiovascular or cerebrovascular events, or both. CPAP compliance was objectively measured by the built-in compliance software of CPAP devices. We considered adherence as adequate if the mean CPAP use was at least 4 hours per night. Statistical analysis Continuous variables with normal distribution were expressed as mean SD and as median and interquartile range without normal distribution, whereas categorical variables were presented as absolute numbers and percentages. A 2-tailed paired Student t test or a c2 test was used to compare variables between baseline and the end of follow-up. The 2sample t test was used for normally distributed variables, and a Mann-Whitney test was used for non-normally
855
distributed variables in subgroup analysis. Altman recommended that there be 10 participants per predictive variable in a multiple linear regression model.23 Further, bivariate analysis suggests that it requires at least 60 participants to explore a maximum of 6 variables. The change in MBP between baseline and follow-up was a dependent variable, whereas the other measurements (including baseline BMI, ESS, AHI, MBP, change in ESS and HR, and CPAP compliance) were independent variables. These were assessed with linear modeling analysis (stepwise multiple linear regression). PASW Statistics, version 18 (SPSS, Chicago, IL) was used for statistical analysis. Appropriate 95% CIs were calculated, and a 2-sided P value < 0.05 was considered statistically significant. Results We initially recruited 238 hypertensive patients with newly diagnosed CHD and clinical suspicion of OSA. Of these, 89 (37.4%) were diagnosed with OSA; 6 patients declined to take part in the study, and 12 patients were excluded, including 9 patients with AHI < 15, 1 patient with severe pulmonary disease, and 2 patients with systolic heart failure and poor cardiac function (ejection fraction < 35%). Thus, 71 patients were enrolled in the study. Of these participants, 3 patients withdrew from the study before the 1-month visit because of intolerance to the CPAP treatment, and 2 participants lacked complete data. Eventually, 66 patients completed the study. The median duration of follow-up was 36 months (24-60 months). The baseline and follow-up characteristics of the included patients are summarized in Table 1. The mean age was 62.6 6.7 years. There were 58 (87.9%) male patients in this study. The mean BMI was 28.3 3.2 kg/m2 and 20 (30.3%) participants were obese. Further, 30 (45.5%) participants had an ESS score > 10 and 25 (37.9%) participants had severe OSA. With respect to cardiovascular events, 3 patients experienced unstable angina, and 1 patient was hospitalized for heart failure during the course of follow-up. The mean time of CPAP application was 4.3 1.2 hours per night. The mean AHI derived from the CPAP machine at the end of follow-up was 2.9 1.4 events per hour, indicating good control of OSA. Compared with baseline, values for SBP and DBP showed a reduction (SBP, 141.1 12.4 mm Hg vs 135.5 11.0 mm Hg; DBP, 83.8 7.6 mm Hg vs 80.7 7.3 mm Hg; both P < 0.05) (Fig. 1). From baseline to followup, SBP and DBP were reduced by 5.6 mm Hg (95% CI, 3.0-8.1) and 3.0 mm Hg (95% CI, 0.8-5.3), respectively. Daytime somnolence was significantly improved (ESS, from 9.5 3.4 at baseline to 3.6 2.0 at follow-up; P < 0.001). The mean improvement in ESS was 6.0 (95% CI, 5.1-6.9) at follow-up. Post-treatment HR was lower than that at baseline (64.7 7.2 bpm vs 67.4 7.8 bpm; P ¼ 0.003). The results of the correlation analysis of the fall in MBP between baseline and follow-up are shown in Table 2. The change in mean MBP at follow-up correlated closely with the baseline MBP (r ¼ 5.6; P < 0.001). Using an illustrative post hoc analysis, we found that patients with a baseline MBP value 100 mm Hg had a significant reduction of 6.1 mm Hg in MBP compared with a mean fall of 2.0 mm Hg in those with a baseline MBP < 100 mm Hg (P < 0.001). The change in mean MBP at follow-up was also significantly correlated with CPAP compliance (r ¼ 0.37; P ¼ 0.002). An
856
Canadian Journal of Cardiology Volume 31 2015
Table 1. Baseline and follow-up characteristics of study patients (n ¼ 66) Variable Age (y) Male sex (n, %) BMI (kg/m2) Neck circumference (cm) Heart rate (bpm) ESS (points) AHI (events/h) SBP (mm Hg) DBP (mm Hg) MBP (mm Hg) Antihypertensive drug use b-blocker use (n, %) ACEI use (n, %) ARB use (n, %) CCB use (n, %) Diuretic use (n, %) CPAP compliance (h/night)
Baseline 62.6 6.7 58 (87.9) 28.3 3.2 41.0 3.2 67.4 7.8 9.5 3.4 28.3 12.6 141.1 12.4 83.8 7.6 102.8 6.5 3.0 0.8 56 (84.8) 30 (45.5) 25 (37.9) 37 (56.1) 60 (90.9)
Follow-up
P value
27.8 3.1 40.9 3.5 64.7 7.2 3.6 2.0 2.9 1.4 135.5 11.0 80.7 7.3 98.8 6.7 3.2 0.7 57 (86.3) 32 (48.5) 27 (40.9) 35 (53.0) 61 (92.4) 4.3 1.2
0.137 0.541 0.003
