Canadian Journal of Cardiology 31 (2015) 846e852
Clinical Research
Effect of Nocturnal Intermittent Hypoxia on Left Atrial Appendage Flow Velocity in Atrial Fibrillation Takehiro Kimura, MD, PhD, Takashi Kohno, MD, PhD, Kazuaki Nakajima, MD, Shin Kashimura, MD, Yoshinori Katsumata, MD, PhD, Takahiko Nishiyama, MD, PhD, Nobuhiro Nishiyama, MD, PhD, Yoko Tanimoto, MD, PhD, Yoshiyasu Aizawa, MD, PhD, Keiichi Fukuda, MD, PhD, and Seiji Takatsuki, MD, PhD Division of Cardiology, Department of Medicine, Keio University School of Medicine, Tokyo, Japan
ABSTRACT
RESUM E
Background: The mechanism underlying the associations of sleepdisordered breathing (SDB) with stroke and atrial fibrillation (AF) is not well established. We explored the relationship between nocturnal intermittent hypoxia, a marker of SDB, and left atrial (LA)/LA appendage (LAA) function among AF patients. Methods: We evaluated 134 consecutive AF candidates for catheter ablation (age, 59.6 9.4 years; body mass index [BMI], 24.8 3.2; Congestive Heart Failure, Hypertension, Age (75 years), Diabetes, Stroke/Transient Ischemic Attack, Vascular Disease, Age (65-74 years), Sex (Female) (CHA2DS2-VASc) score, 1.2 1.1, paroxysmal AF, n ¼ 83) using nocturnal pulse oximetry, a noninvasive screening method for nocturnal intermittent hypoxia. Based on 3% oxygen desaturation index (3% ODI), patients were divided into nocturnal intermittent hypoxia (3% ODI > 15; n ¼ 32) and control groups (3% ODI 15; n ¼ 102). Results: The nocturnal intermittent hypoxia group demonstrated significantly higher weight, BMI, Congestive Heart Failure, Hypertension, Age, Diabetes, Stroke/Transient Ischemic Attack (CHADS2) and CHA2DS2-VASc scores, serum hemoglobin A1c and plasma brain natriuretic peptide levels, LA size, and prevalence of hypertension, vascular disease, and sick sinus syndrome. Echocardiographically, nocturnal intermittent hypoxia was associated with a higher grade of
canisme qui sous-tend l’association entre les Introduction : Le me troubles respiratoires du sommeil (TRS), et l’accident vasculaire re bral et la fibrillation auriculaire (FA) n’est pas bien e tabli. Nous ce le lien entre l’hypoxie intermittente nocturne, un maravons examine queur des TRS, et le fonctionnement de l’oreillette gauche (OG) et de l’appendice auriculaire gauche (AAG) chez les patients souffrant de FA. thodes : Nous avons e value l’ablation par cathe ter chez 134 candiMe cutifs souffrant de FA (âge, 59,6 9,4 ans; indice de masse dats conse corporelle [IMC], 24,8 3,2; score CHA2DS2-VASc [Congestive Heart Failure, Hypertension, Age, Diabetes, Stroke/Transient Ischemic Attack, Vascular Disease, Age (65-74 years), Sex (Female), soit l’insuffisance cardiaque congestive, l’hypertension, l’âge, le diabète, l’accident vascure bral/l’ische mie ce re brale transitoire, la maladie vasculaire, laire ce minin)], 1,2 1,1, FA paroxystique, n ¼ 83) l’âge (65 à 74 ans), le sexe (fe trie de pouls nocturne, une me thode de de pistage non à l’aide de l’oxyme invasive de l’hypoxie intermittente nocturne. À partir de l’indice de saturation en oxygène de 3 % (IDO de 3 %), les patients ont e te divise s de comme suit : hypoxie intermittente nocturne (IDO de 3 % > 15; n ¼ 32) et moins (IDO de 3 % 15; n ¼ 102). groupes te sultats : Le groupe souffrant d’hypoxie intermittente a de montre un Re poids, une IMC, un score CHADS2 (Congestive Heart Failure, Hypertension, Age, Diabetes, Stroke/Transient Ischemic Attack, soit l’insuffisance
Sleep-disordered breathing (SDB) is a common but largely undiagnosed disease that is associated with increased cardiovascular morbidity and mortality.1 SDB patients tend to be older, with a higher body mass index (BMI), and increased prevalence of hypertension and diabetes.2 These risk factors
for SDB are also common in stroke and atrial fibrillation (AF).3,4 Approximately 50% of AF patients were reported to have SDB,5 and SDB is associated with increased incidence of AF.6 Severe SDB was also observed in 29% of Japanese stroke patients7 and was also associated with stroke.8 AF is also an established major risk factor for stroke and quintuples stroke incidence.9 Thus, SDB, AF, and stroke are closely related, but studies on the association among these pathologies have been limited thus far. Left atrial (LA) appendage (LAA) has been recognized as the embolic source in 90% of cardiogenic stroke cases.10 Because SDB is associated with adverse cardiac remodelling,11-13 we hypothesized that SDB could be associated with reduced LA/LAA function in patients with AF.
Received for publication October 7, 2014. Accepted December 30, 2014. Corresponding author: Dr Takashi Kohno, Division of Cardiology, Department of Medicine, Keio University School of Medicine, 35 Shinanomachi Shinjuku-ku, Tokyo 160-8582, Japan. Tel.: þ81-3-33531211 61421; fax: þ81-3-5363-3875. E-mail: [email protected] See page 851 for disclosure information.
http://dx.doi.org/10.1016/j.cjca.2014.12.032 0828-282X/Ó 2015 Canadian Cardiovascular Society. Published by Elsevier Inc. All rights reserved.
Kimura et al. Nocturnal Intermittent Hypoxia in AF
847
spontaneous echo contrast and low LAA flow velocity. Multiple regression analysis adjusted for type of AF, CHA2DS2-VASc score, BMI, plasma brain natriuretic peptide level, LA size, and rhythm on echocardiography revealed that 3% ODI was a factor independently associated with LAA flow velocity (b ¼ 0.184; 95% confidence interval, 0.818 to 0.006). Conclusions: Nocturnal intermittent hypoxia was an independent determinant for low LAA flow velocity in patients with AF, suggesting that the connection between SDB and LAA function might underlie the association of AF with stroke.
cardiaque congestive, l’hypertension, l’âge, le diabète, l’accident vasre bral/l’ische mie ce re brale transitoire) et un score culaire ce rique d’he moglobine A1c et une CHA2DS2-VASc, une concentration se tique de type B, une concentration plasmatique du peptide natriure valence de l’hypertension, de la maladie taille de l’OG, et une pre vasculaire et de la maladie du nœud sinusal significativement plus leve s. Échocardiographiquement, l’hypoxie intermittente nocturne a e te associe e à un degre plus e leve de contraste e chographique e et à une ve locite du flux de l’AAG. L’analyse de re gression spontane e du type de FA, du score CHA2DS2-VASc, de l’IMC, de la multiple ajuste tique de type B, de la concentration plasmatique du peptide natriure chocardiographie a re ve le que l’IDO de taille et du rythme de l’OG à l’e tait un facteur inde pendamment associe à la ve locite du flux de 3%e l’AAG (b ¼ 0,184; intervalle de confiance à 95 %, 0,818 à 0,006). tait un de terminant Conclusions : L’hypoxie intermittente nocturne e pendant de la faible ve locite du flux de l’AAG chez les patients inde souffrant de FA, ce qui suggère que le lien entre les TRS et le fonctionnement de l’AAG pourrait sous-tendre l’association entre la FA et re bral. l’accident vasculaire ce
Despite the importance of SDB in managing AF,14-19 the prevalence of undiagnosed SDB is surprisingly high, as shown in the Wisconsin Sleep Cohort Study.20 Although polysomnography (PSG) has been used as a standard for the diagnosis of SDB, referral to sleep study is often difficult, inconvenient, and expensive. Performing fully equipped sleep studies in all AF patients is not feasible. Here, we chose to perform nocturnal pulse oximetry to detect nocturnal intermittent hypoxia, a surrogate marker for SDB,21,22 as a simple screening utility to identify undiagnosed SDB in all patients hospitalized for AF catheter ablation in our facility. In the present study we aimed to estimate the prevalence of nocturnal intermittent hypoxia among AF catheter ablation candidates using nocturnal pulse oximetry and to clarify its relationship with background characteristics, and laboratory, echocardiographic, and Holter monitoring data. Additionally, we assessed the association of LA/LAA function and nocturnal intermittent hypoxia.
Nocturnal intermittent hypoxia
Methods The study protocol conformed to the ethical guidelines of the Declaration of Helsinki of 1975 as reflected in a priori approval by our institution’s human research committee. All patients provided written informed consent to participate. Study population One hundred thirty-four consecutive AF patients who were candidates for catheter ablation in our facility were evaluated retrospectively. The indication for AF catheter ablation was determined according to symptoms due to AF attacks, anti-arrhythmic drug resistance, or patient request, regardless of the type of AF. We excluded patients who were previously diagnosed with SDB and treated with continuous positive airway pressure (CPAP). Patients with chronic obstructive pulmonary disease who required home oxygen therapy were not included in this study. Patients who were contraindicated for anticoagulation or with pre-existing thrombi in the LA were also excluded.
Nocturnal pulse oximetry was performed to identify the presence of nocturnal intermittent hypoxia in all patients 1 day before ablation. Arterial oxyhemoglobin saturation was recorded using a finger probe at a 1-Hz sampling frequency and 5-second average time (PULSOX-Me300; Teijin Pharma, Tokyo, Japan). These recordings were scored using specialized software (DS-Me; Teijin Pharma, Tokyo, Japan). Because the measurement time of pulse oximetry is often longer than the true total sleep time, we used a single-night sleep log to exclude waking time from the analysis and minimize the potential for overestimating total sleep time. We used oxygen desaturation index 3% (3% ODI), the frequency of episodes of 3% desaturation per hour, as an indicator of nocturnal intermittent hypoxia. Based on 3% ODI, patients were divided into nocturnal intermittent hypoxia (3% ODI > 15) and control groups (3% ODI 15). The validity of pulse oximetry has been previously reported based on synchronous overnight recording of pulse oximetry and standard PSG, and its sensitivity and specificity were 85% and 100%, respectively, for detecting apnea hypopnea index of 20 determined using PSG and a cutoff threshold of 3% ODI ¼ 15.23-25 Parameters for analysis Patient background data including age, height, weight, BMI, Congestive Heart Failure, Hypertension, Age, Diabetes, Stroke/Transient Ischemic Attack (CHADS2) score,26 Congestive Heart Failure, Hypertension, Age (75 years), Diabetes, Stroke/Transient Ischemic Attack, Vascular Disease, Age (65-74 years), Sex (Female) (CHA2DS2-VASc) score,27 paroxysmal AF duration from initial diagnosis to ablation, duration of persistent AF, number of anti-arrhythmic drugs before ablation, number of ablation sessions, type of AF, history of dyslipidemia, and sick sinus syndrome (SSS) were collected and compared between groups. Type of AF was classified as paroxysmal (persisting < 1 week), persistent (persisting longer than 1 week and < 1 year), or longstanding persistent (persisting longer than 1 year) according to American College of Cardiology/American Heart Association/Heart
848
Canadian Journal of Cardiology Volume 31 2015
Table 1. Comparison of atrial fibrillation type and CHADS2/CHA2DS2-VASc scores 3% ODI 15 (n ¼ 102) Persistent and LS-persistent AF Congestive heart failure Hypertension Age older than 65 years Age older than 75 years Diabetes Stroke Vascular disease Female sex Dyslipidemia Sick sinus syndrome
33.3 4.9 34.3 29.4 2.9 8.8 3.9 0 13.7 24.5 6.9
(34) (5) (35) (30) (3) (9) (4) (0) (14) (25) (7)
3% ODI > 15 (n ¼ 32) 53.1 9.4 71.9 40.6 3.1 18.8 9.4 6.3 3.1 31.3 18.8
(17) (3) (23) (13) (1) (6) (3) (2) (1) (10) (6)
P
OR
95% CI
0.044 0.351 0.000 0.236 0.957 0.120 0.226 0.011 0.097 0.449 0.047
2.267 2.007 4.892 1.642 1.065 2.385 2.534 N/A 0.203 1.400 3.132
1.011-5.081 0.452-8.907 2.045-11.704 0.720-3.744 0.107-10.605 0.777-7.314 0.536-11.98 N/A 0.026-1.606 0.585-3.352 1.012-9.733
Values are % (n). CHADS2, Congestive Heart Failure, Hypertension, Age (75 years), Diabetes, Stroke/Transient Ischemic Attack; CHA2DS2-VASc, Congestive Heart Failure, Hypertension, Age (75 years), Diabetes, Stroke/Transient Ischemic Attack, Vascular Disease, Age (65-74 years), Sex (Female); CI, confidence interval; LSpersistent AF, longstanding persistent atrial fibrillation; N/A, not applicable; ODI, oxygen desaturation index; OR, odds ratio.
Rhythm Society guidelines.28 Blood samples to test serum levels of creatinine, hemoglobin A1c (HbA1c), low-density lipoprotein, uric acid, and plasma brain natriuretic peptide (BNP) were obtained from all patients before ablation. Transthoracic echocardiography (TTE) was performed in all patients, and transesophageal echocardiography (TEE) was performed in 126 patients 1 week before ablation. The systolic (emptying) and diastolic (filling) flow velocity of the LAA (LAA flow) were averaged. Consistent with previous reports, the severity of spontaneous echo contrast (SEC) was graded from 0 to 3: 0 ¼ none; 0.5 ¼ none to mild; 1 ¼ mild; 2 ¼ moderate; and 3 ¼ severe.29,30 All echocardiographic studies were interpreted by the same 2 experienced echocardiologists without the knowledge of the patient’s background. Holter monitoring was performed in 50 of 83 paroxysmal AF patients, and the incidence of nocturnal AF attacks was evaluated. A nocturnal AF attack was defined as AF initiation during sleep, regardless of its symptoms, and patients with persistent AF documented over a whole day were excluded. Statistical methods Parametric data were expressed as mean SD. The MannWhitney U test was used to compare parameters across groups. The correlation between parameters was investigated using Pearson correlation coefficient test. The c2 test was used to compare the number of patients across groups. Odds ratios and 95% confidence intervals [CIs] were also computed. Multivariate analysis was also performed to elucidate independent determinants. Parameters with statistical significance in the Mann-Whitney U test or the c2 test (P < 0.10) were included in the multivariate analysis. P values < 0.05 were considered statistically significant. Analyses were performed using IBM SPSS software (version 22; IBM, Armonk, NY). Results Overall analysis A total of 134 AF patients (age, 59.6 9.4 years; BMI, 24.8 3.2; CHADS2 score, 0.7 0.9; CHA2DS2-VASc score, 1.2 1.1; paroxysmal AF, n ¼ 83; persistent AF, n ¼ 30; longstanding persistent AF, n ¼ 21) were evaluated. A total of 32 patients who demonstrated nocturnal intermittent hypoxia
were in the nocturnal intermittent hypoxia group, with a mean 3% ODI of 23.1 8.5, compared with 102 patients in the control group with a mean 3% ODI of 6.7 3.8 (P < 0.001). The percentage of persistent and longstanding persistent AF patients was significantly greater in the nocturnal intermittent hypoxia group (persistent AF, 28.1% [n ¼ 9] and longstanding persistent AF, 25.0% [n ¼ 8]) compared with the control group (persistent AF, 20.5% [n ¼ 21] and longstanding persistent AF, 12.7% [n ¼ 13]; Table 1). The prevalence of hypertension, vascular disease, SSS, body weight, and BMI was significantly greater in the nocturnal intermittent hypoxia group (Tables 1 and 2). CHADS2 and CHA2DS2-VASc scores were also significantly higher, but age, number of AADs used before ablation, and number of ablation sessions did not differ between the groups. Laboratory analyses indicated that serum HbA1c and plasma BNP levels were significantly higher in the nocturnal intermittent hypoxia group. TTE data revealed that the LA size was significantly larger in the nocturnal intermittent hypoxia group, and most notably among patients with AF rhythm during TTE. The percentage of patients evaluated using TEE did not differ between groups (nocturnal intermittent hypoxia, 96.9% [n ¼ 31] vs control, 93.1% [n ¼ 95]; P ¼ 0.436). TEE data revealed that nocturnal intermittent hypoxia was associated with a higher grade of SEC, although the difference was not significant when the rhythm during TEE was considered. LAA flow velocity was lower in the nocturnal intermittent hypoxia group, which was significant most notably in patients with AF rhythm during TEE (Table 2 and Fig. 1). The value of 3% ODI was significantly correlated with mean saturation data of pulse oximetry (r ¼ 0.604; P < 0.001), minimum saturation data of pulse oximetry (r ¼ 0.565; P < 0.001), BMI (r ¼ 0.387; P < 0.001), LA size (r ¼ 0.366; P < 0.001), plasma BNP level (r ¼ 0.332; P < 0.001), weight (r ¼ 0.323; P < 0.001), LAA flow velocity (r ¼ 0.304; P ¼ 0.001), grade of SEC (r ¼ 0.299; P ¼ 0.001), CHADS2 score (r ¼ 0.286; P ¼ 0.001), paroxysmal AF duration (r ¼ 0.281; P ¼ 0.017), CHA2DS2-VASc score (r ¼ 0.265; P ¼ 0.002), E/e’ (r ¼ 0.260; P ¼ 0.019), and HbA1c level (r ¼ 0.249; P ¼ 0.006). Analysis of paroxysmal and persistent AF Next, we analyzed paroxysmal and persistent AF separately and evaluated the associations between nocturnal intermittent
Kimura et al. Nocturnal Intermittent Hypoxia in AF
849
Table 2. Comparison of patient background and echocardiographic data Total (n ¼ 134) 3% ODI Mean SpO2, % Minimum SpO2, % Age Height, cm Weight, kg BMI CHADS2 CHA2DS2-VASc No. AAD No. ablation Cr, mg/dL HbA1c, % LDL, mg/dL BNP, pg/mL UA, mg/dL LVEDD, cm LVESD, cm LVEF, % LA, cm Total NSR AF E, cm/sec A, cm/sec Dct, msec E/e’ SEC Total NSR AF LAA flow (cm/sec) Total NSR AF
3% ODI 15 (n ¼ 102)
3% ODI > 15 (n ¼ 32)
P
10.6 95.7 82.8 59.6 168.3 70.6 24.8 0.7 1.2 0.6 1.3 0.9 5.4 108 89.6 6.2 4.8 3.1 58.3
8.8 1.3 6.6 9.4 8.2 12.9 3.2 0.9 1.1 0.8 0.5 0.5 0.5 27 89.8 1.4 0.5 0.5 9.1
6.7 96 84.4 59.3 167.8 68.3 24.2 0.6 1 0.7 1.2 1 5.3 108.1 79.9 6.1 4.7 3 58.9
3.8 1.1 5.5 9.7 8 11.2 2.8 0.8 1 0.8 0.5 0.6 0.5 27.5 82.7 1.3 0.5 0.4 7.9
23.1 94.7 77 60.5 170 78.3 26.9 1.2 1.7 0.6 1.3 0.9 5.6 107.4 120.4 6.3 4.9 3.3 56.2
8.5 1.5 7.2 8.4 8.7 15.2 3.6 1 1.3 0.8 0.7 0.2 0.6 25.5 104.9 1.4 0.6 0.7 12.2
0.000 0.000 0.000 0.681 0.188 0.000 0.000 0.001 0.004 0.384 0.816 0.963 0.004 0.942 0.012 0.364 0.068 0.162 0.343
4 3.7 4.2 70.9 60.8 189.2 8.9
0.6 0.6 (66) 0.6 (68) 15.4 15.7 51.3 4
3.9 3.7 4.1 70.9 61.7 191.2 8.4
0.6 0.6 (57) 0.6 (45) 16.2 14.6 49.7 3.8
4.3 4.1 4.4 71 54.7 183.7 10.6
0.6 0.7 (9) 0.5 (23) 13.2 21.9 55.9 4.4
0.000 0.127 0.020 0.807 0.344 0.731 0.030
0.5 0.6 (126) 0.3 0.4 (59) 0.7 0.7 (67)
0.4 0.6 (95) 0.3 0.4 (48) 0.6 0.7 (47)
0.7 0.7 (31) 0.4 0.4 (11) 0.9 0.8 (20)
0.045 0.290 0.151
54.9 18.9 (126) 59.2 16.6 (59) 51.1 20.2 (67)
57.6 18.4 (95) 60.8 15.7 (48) 54.3 20.5 (47)
46.6 18.3 (31) 52.1 19.1 (11) 43.6 17.6 (20)
0.005 0.116 0.045
Values are mean SD or mean SD (n). AF, atrial fibrillation during echocardiography; BMI, body mass index; BNP, plasma brain natriuretic peptide level; CHADS2, Congestive Heart Failure, Hypertension, Age, Diabetes, Stroke/Transient Ischemic Attack; CHA2DS2-VASc, Congestive Heart Failure, Hypertension, Age (75 years), Diabetes, Stroke/ Transient Ischemic Attack, Vascular Disease, Age (65-74 years), Sex (Female); Cr, serum creatinine level; Dct, deceleration time; E/e’, calculated E/e’ of the septum; Hb, hemoglobin; LA, left atrial diameter measured using transthoracic echocardiography; LAA flow, average flow velocity of systolic (emptying) and diastolic (filling) of the left atrial appendage measured using transesophageal echocardiography; LDL, serum low-density lipoprotein level; LVEDD, left ventricular end diastolic diameter; LVEF, left ventricular ejection fraction; LVESD, left ventricular end systolic diameter; No. AAD, number of anti-arrhythmic drugs before ablation; No. ablation, number of ablation sessions; NSR, normal sinus rhythm during echocardiography; 3% ODI, frequency of episodes of 3% desaturation per hour; SEC, grade of spontaneous echo contrast measured using transesophageal echocardiography; SpO2, saturation data of pulse oximetry; UA, serum uric acid level.
hypoxia and patient characteristics in each population. Among 83 paroxysmal AF patients, 68 patients in the control group with a mean 3% ODI of 6.5 4.0 were compared with 15 patients in the nocturnal intermittent hypoxia group with a mean 3% ODI of 21.8 7.9 (Supplemental Table S1). In the nocturnal intermittent hypoxia group, weight was significantly greater, but no differences were observed in BMI, CHADS2, and CHA2DS2-VASc scores, serum HbA1c level, or plasma BNP level between the groups. The disease duration of paroxysmal AF was significantly longer in the nocturnal intermittent hypoxia group. TTE data demonstrated a significantly larger LA size in the nocturnal intermittent hypoxia group, especially in patients with AF rhythm during TEE. LAA flow was significantly lower in the nocturnal intermittent hypoxia group, especially in patients with AF rhythm during TEE (Fig. 1). Neither the percentage of paroxysmal AF patients evaluated with Holter monitoring nor the incidence of nocturnal AF attacks differed between the groups. Among 51 persistent AF patients (34 persistent and 17 longstanding persistent AF), 17 patients in the nocturnal
intermittent hypoxia group with a mean 3% ODI of 24.2 9.1 were compared with 34 patients in the control group with a mean 3% ODI of 7.1 3.3 (Supplemental Table S2). In the nocturnal intermittent hypoxia group, weight, BMI, CHADS2 score, and serum HbA1c level were significantly higher. The prevalence of hypertension was also significantly greater (nocturnal intermittent hypoxia, 82.4% [n ¼ 14] vs control group, 35.3% [n ¼ 12], P ¼ 0.002), and the duration of persistent AF tended to be longer in the nocturnal intermittent hypoxia group. The effect of nocturnal intermittent hypoxia on TTE and TEE data in persistent AF tended to be similar to that observed overall or in paroxysmal AF; however, differences were not statistically significant. Three percent ODI as an independent determinant of LAA velocity Multiple regression analysis adjusted for type of AF (b ¼ 0.313; P ¼ 0.002; 95% CI, 19.816 to 4.590), CHA2DS2-VASc score (P ¼ 0.423), BMI (P ¼ 0.553),
850
Canadian Journal of Cardiology Volume 31 2015
data revealed larger LA size, higher grade of SEC, and lower LAA flow velocity in AF patients with 3% ODI > 15. Multiple regression analysis adjusted by these confounding factors along with rhythm during TEE revealed that 3% ODI was an independent determinant for reduced LAA flow velocity (b ¼ 0.184; 95% CI, 0.818 to 0.006), suggesting that nocturnal intermittent hypoxia might be associated with reduced LAA function, which is an important determinant of cardiogenic stroke in patients with AF. Pulse oximetry as a screening tool for SDB
Figure 1. The relationship between the flow velocity of the LAA and the 3% ODI. The LAA flow velocity measured using transesophageal echocardiography is significantly reduced in patients with 3% ODI >15 with respect to the type of AF and the rhythm during measurements. * P < 0.05; ** P < 0.001. AF, atrial fibrillation; LAA, left atrial appendage; NSR, normal sinus rhythm; 3% ODI, frequency of episodes of 3% desaturation per hour.
plasma BNP level (P ¼ 0.209), LA size (P ¼ 0.159), and rhythm during TEE (P ¼ 0.510) revealed that 3% ODI was a factor independently associated with LAA flow velocity (b ¼ 0.184; P ¼ 0.047; 95% CI, 0.818 to 0.006) in all patients (Table 3). In paroxysmal AF patients, 3% ODI was also an independent determinant for LAA flow velocity (b ¼ 0.293; P ¼ 0.002; 95% CI, 1.282 to 0.038) after adjustment for multiple potential confounding factors including LA size (b ¼ 0.502; P ¼ 0.002; 95%, CI 24.394 to 6.103), rhythm during TEE (P ¼ 0.146), AF duration (P ¼ 0.382), and weight (P ¼ 0.191). Discussion We established the high prevalence of nocturnal intermittent hypoxia in AF ablation candidates using pulse oximetry as a simple screening method for SDB. AF patients with 3% ODI > 15 had significantly greater weight, BMI, CHADS2 score, CHA2DS2-VASc score, serum HbA1c, and plasma BNP level, and greater prevalence of persistent AF, hypertension, vascular disease, and SSS. Echocardiography Table 3. Multivariate analysis of determinants of LAA flow velocity 3% ODI Persistent AF AF during TEE CHA2DS2-VASc BMI BNP LA size
Standardized b
P
0.184 0.313 0.067 0.07 0.059 0.123 0.147
0.047 0.002 0.51 0.423 0.553 0.209 0.159
95% CI 0.818 19.816 5.094 4.253 0.799 0.067 10.871
to to to to to to to
0.006 4.59 10.197 1.797 1.486 0.015 1.802
AF, atrial fibrillation; BMI, body mass index; BNP, plasma brain natriuretic peptide level; CHA2DS2-VASc, Congestive Heart Failure, Hypertension, Age (75 years), Diabetes, Stroke/Transient Ischemic Attack, Vascular Disease, Age (65-74 years), Sex (Female); CI, confidence interval; LA, left atrium; LAA, left atrial appendage; 3% ODI, frequency of episodes of 3% desaturation per hour; TEE, transesophageal echocardiography.
SDB is highly prevalent and recognized as a potential risk factor for the development of AF.6 Patients with untreated SDB had a greater rate of recurrence of AF after medical therapy or electrical cardioversion.14 In addition, patients with SDB had more recurrence after ablation due to nonpulmonary vein triggers.15,16 Because a recent study demonstrated that CPAP was an additional therapy for improving ablation outcome,15,17-19 the ability to screen undiagnosed SDB patients efficiently among a large number of AF ablation candidates is valuable to clinical practice. We chose pulse oximetry as a simple screening method based on its validity, which was established using synchronous overnight PSG.22,25 In the present study, nocturnal pulse oximetry detected 23.9% of AF ablation candidates to have nocturnal intermittent hypoxia based on 3% ODI > 15. The prevalence of men and women with 3% ODI > 15 in Japanese community-based populations was reported as 7.9%-9.2% and 2.0%-3.1%, respectively, and mean 3% ODI was 2.3 events per hour.24 Our results suggest that its prevalence and severity might be greater among AF patients than in the general population. Because the use of the current screening process using PSG can only identify a limited number of SDB patients among a large cohort of ablation candidates, more patients might benefit from the timely diagnosis and treatment of SDB using nocturnal pulse oximetry as a readily available and inexpensive screening system. Nocturnal intermittent hypoxia, LA remodelling, and nocturnal arrhythmia In patients with obstructive sleep apnea, mechanical and autonomic cardiovascular effects and increased oxidative stress and inflammation are known to exacerbate LA overload and remodelling, leading to the development of structural and electrical substrates for AF.11 We demonstrated that nocturnal intermittent hypoxia was significantly associated with increased LA size and a greater prevalence of SSS. In addition, although the mean age of paroxysmal AF patients did not differ between the nocturnal intermittent hypoxia and control groups, the disease duration of paroxysmal AF was significantly longer in the nocturnal intermittent hypoxia group, which might suggest an earlier onset of AF in the presence of SDB. We also found that the incidence of persistent AF was significantly higher among patients with 3% ODI > 15. SDB itself might increase the likelihood that AF is persistent, or a hemodynamic consequence of persistent AF could contribute to SDB deterioration. Although our study could not clarify the cause-and-effect relationship between nocturnal intermittent hypoxia and AF perpetuation, several unique data demonstrated a direct effect of SDB on AF development in an
Kimura et al. Nocturnal Intermittent Hypoxia in AF
experimental rat model.31,32 Frequent episodes of intermittent hypoxia along with extremes in negative intrathoracic pressure might be associated with the progression of atrial remodelling.31,32 In an experimental pig model, negative tracheal pressure was a strong trigger of AF development due to shortening of the atrial effective refractory period and thus increase susceptibility to AF through vagal activation.33 The possible link between SDB and nocturnal arrhythmias is supported by the finding that 48% of SDB patients who underwent Holter monitoring demonstrated cardiac arrhythmias during the night.1,34 However, another report found that the mean number of apneic events and lowest oxygen saturation during sleep did not significantly differ between patients with and without nocturnal arrhythmias.35 In the present study, the incidence of nocturnal arrhythmia was not significantly greater among paroxysmal AF patients in the nocturnal intermittent hypoxia group. These results might be a result of the small number of patients or the limitation of detecting nocturnal arrhythmia using 24-hour Holter monitoring. Further evaluation should be conducted with concern for the timing of Holter monitoring and pulse oximetry. SDB, AF, and stroke interrelationships Stratifying patients at high risk for stroke is a major consideration in the management of AF. The Stroke Prevention in Atrial Fibrillation III (SPAF-III) study demonstrated that decreased LAA flow velocity, in addition to age, systolic blood pressure, sustained AF, ischemic heart disease, and LA area, was an independent predictor of SEC, thrombi, and cardioembolic events.36 Although SDB has been commonly observed in AF patients, its association with LAA function has not been established. Only 1 previous study involving 79 AF/atrial flutter patients referred for cardioversion found no significant difference in LAA average velocity between patients with and without SDB.14 However, the definition of SDB and type of AF were not described precisely, and ejection fraction was relatively low compared with that in our study. Our results revealed that nocturnal intermittent hypoxia was independently associated with low LAA flow velocity, which is known to be associated with cardiogenic stroke. In patients with paroxysmal AF, LAA flow velocity was significantly lower in patients with 3% ODI > 15 who demonstrated AF rhythm during TEE, but not in normal sinus rhythm. The period after conversion from AF to sinus rhythm was not measurable in this study, and the effect of atrial stunning might be a potential source of heterogeneity. To resolve this issue, long-term Holter monitoring might be needed in the future. In persistent AF, LAA flow velocity was slightly reduced in the nocturnal intermittent hypoxia group, but the difference was not statistically significant. This lack of significance might be because the number of patients was small or because LAA flow velocity was generally reduced by persistent AF itself. The grade of SEC was also significantly greater in the nocturnal intermittent hypoxia group. In addition to the hemodynamic stasis due to decreased LA/LAA function in SDB patients, SDB itself might be associated with hypercoagulation status1,37; therefore, it is plausible that SDB could be associated with SEC. Further large clinical studies are needed to assess this hypothesis using coagulation biomarkers and more precise quantification of SEC.
851
Limitations First, the number of patients enrolled in this study was small, which might reduce the power of the statistics. Second, PSG evaluation was not performed, and apnea hypopnea index and 3% ODI were not directly compared. Pulse oximetry cannot distinguish obstructive apnea from central apnea. The previous study reported that most SDB patients treated with ablation for AF had apnea that was predominantly obstructive,19 which is consistent with our unpublished data. Pulse oximetry could be a useful screening tool in a population that is unlikely to have central sleep apnea. Furthermore, PSG will be needed to elucidate the important question of which parameters of SDB are associated with AF development and remodelling of the LA and LAA. Third, blood pressure and underlying rhythm might affect the value of pulse oximetry. Fourth, it remains unknown whether nocturnal intermittent hypoxia could be a therapeutic target for CPAP because of a lack of randomized control trials. Based on our data, SDB treatment might be beneficial for stroke prevention by improving LAA function. These issues should be investigated in a further study. Conclusions Nocturnal intermittent hypoxia is an independent determinant for low LAA flow velocity in AF ablation candidates. The association of SDB with reduced LAA function might explain the relationship among AF, SDB, and stroke. Disclosures The authors have no conflicts of interest to disclose. References 1. Somers VK, White DP, Amin R, et al. Sleep apnea and cardiovascular disease: an American Heart Association/American College of Cardiology Foundation Scientific Statement from the American Heart Association Council for High Blood Pressure Research Professional Education Committee, Council on Clinical Cardiology, Stroke Council, and Council on Cardiovascular Nursing. In collaboration with the National Heart, Lung, and Blood Institute National Center on Sleep Disorders Research (National Institutes of Health). Circulation 2008;118: 1080-111. 2. Young T, Shahar E, Nieto FJ, et al. Predictors of sleep-disordered breathing in community-dwelling adults: the Sleep Heart Health Study. Arch Intern Med 2002;162:893-900. 3. Valenza MC, Baranchuk A, Valenza-Demet G, et al. Prevalence of risk factors for atrial fibrillation and stroke among 1210 patients with sleep disordered breathing. Int J Cardiol 2014;174:73-6. 4. Bassetti C, Aldrich MS. Sleep apnea in acute cerebrovascular diseases: final report on 128 patients. Sleep 1999;22:217-23. 5. Gami AS, Pressman G, Caples SM, et al. Association of atrial fibrillation and obstructive sleep apnea. Circulation 2004;110:364-7. 6. Gami AS, Hodge DO, Herges RM, et al. Obstructive sleep apnea, obesity, and the risk of incident atrial fibrillation. J Am Coll Cardiol 2007;49:565-71. 7. Shibazaki K, Kimura K, Uemura J, et al. Atrial fibrillation is associated with severe sleep-disordered breathing in patients with ischaemic stroke and transient ischaemic attack. Eur J Neurol 2013;20:266-70.
852
8. Muñoz R, Duran-Cantolla J, Martinez-Vila E, et al. Severe sleep apnea and risk of ischemic stroke in the elderly. Stroke 2006;37:2317-21. 9. Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation as an independent risk factor for stroke: the Framingham Study. Stroke 1991;22:983-8. 10. Watson T, Shantsila E, Lip GY. Mechanisms of thrombogenesis in atrial fibrillation: Virchow’s triad revisited. Lancet 2009;373:155-66. 11. Dimitri H, Ng M, Brooks AG, et al. Atrial remodeling in obstructive sleep apnea: implications for atrial fibrillation. Heart Rhythm 2012;9: 321-7. 12. Drager LF, Bortolotto LA, Pedrosa RP, Krieger EM, Lorenzi-Filho G. Left atrial diameter is independently associated with arterial stiffness in patients with obstructive sleep apnea: potential implications for atrial fibrillation. Int J Cardiol 2010;144:257-9. 13. Kim SM, Cho KI, Kwon JH, Lee HG, Kim TI. Impact of obstructive sleep apnea on left atrial functional and structural remodeling beyond obesity. J Cardiol 2012;60:475-83. 14. Kanagala R, Murali NS, Friedman PA, et al. Obstructive sleep apnea and the recurrence of atrial fibrillation. Circulation 2003;107:2589-94. 15. Patel D, Mohanty P, Di Biase L, et al. Safety and efficacy of pulmonary vein antral isolation in patients with obstructive sleep apnea: the impact of continuous positive airway pressure. Circ Arrhythm Electrophysiol 2010;3:445-51. 16. Ng CY, Liu T, Shehata M, et al. Meta-analysis of obstructive sleep apnea as predictor of atrial fibrillation recurrence after catheter ablation. Am J Cardiol 2011;108:47-51. 17. Li L, Wang ZW, Li J, et al. Efficacy of catheter ablation of atrial fibrillation in patients with obstructive sleep apnoea with and without continuous positive airway pressure treatment: a meta-analysis of observational studies. Europace 2014;16:1309-14. 18. Fein AS, Shvilkin A, Shah D, et al. Treatment of obstructive sleep apnea reduces the risk of atrial fibrillation recurrence after catheter ablation. J Am Coll Cardiol 2013;62:300-5. 19. Naruse Y, Tada H, Satoh M, et al. Concomitant obstructive sleep apnea increases the recurrence of atrial fibrillation following radiofrequency catheter ablation of atrial fibrillation: clinical impact of continuous positive airway pressure therapy. Heart Rhythm 2013;10:331-7. 20. Young T, Palta M, Dempsey J, et al. The occurrence of sleep-disordered breathing among middle-aged adults. N Engl J Med 1993;328:1230-5. 21. Tanigawa T. Obstructive sleep apnea: its prevention and screening may contribute to the prevention of hypertension, diabetes and cardiovascular diseases. EPMA J 2011;2:83-9. 22. Vazquez JC, Tsai WH, Flemons WW, et al. Automated analysis of digital oximetry in the diagnosis of obstructive sleep apnoea. Thorax 2000;55: 302-7. 23. Tanigawa T, Tachibana N, Yamagishi K, et al. Usual alcohol consumption and arterial oxygen desaturation during sleep. JAMA 2004;292:923-5. 24. Muraki I, Tanigawa T, Yamagishi K, et al. Nocturnal intermittent hypoxia and the development of type 2 diabetes: the Circulatory Risk in Communities Study (CIRCS). Diabetologia 2010;53:481-8.
Canadian Journal of Cardiology Volume 31 2015 25. Nakamata MKY, Sakai K, Kinebuchi S, et al. The limitation of screening test for patients with sleep apnea syndrome using pulse oximetry. J Jpn Soc Respir Care 2003;12:401-6. 26. Gage BF, Waterman AD, Shannon W, et al. Validation of clinical classification schemes for predicting stroke: results from the National Registry of Atrial Fibrillation. JAMA 2001;285:2864-70. 27. Lip GY, Nieuwlaat R, Pisters R, Lane DA, Crijns HJ. Refining clinical risk stratification for predicting stroke and thromboembolism in atrial fibrillation using a novel risk factor-based approach: the Euro Heart Survey on Atrial Fibrillation. Chest 2010;137:263-72. 28. January CT, Wann LS, Alpert JS, et al. 2014 AHA/ACC/HRS guideline for the management of patients with atrial fibrillation: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines and the Heart Rhythm Society. J Am Coll Cardiol 2014;64:e1-76. 29. Daniel WG, Nellessen U, Schroder E, et al. Left atrial spontaneous echo contrast in mitral valve disease: an indicator for an increased thromboembolic risk. J Am Coll Cardiol 1988;11:1204-11. 30. Fatkin D, Kelly RP, Feneley MP. Relations between left atrial appendage blood flow velocity, spontaneous echocardiographic contrast and thromboembolic risk in vivo. J Am Coll Cardiol 1994;23:961-9. 31. Iwasaki YK, Kato T, Xiong F, et al. Atrial fibrillation promotion with long-term repetitive obstructive sleep apnea in a rat model. J Am Coll Cardiol 2014;64:2013-23. 32. Iwasaki YK, Shi Y, Benito B, et al. Determinants of atrial fibrillation in an animal model of obesity and acute obstructive sleep apnea. Heart Rhythm 2012;9:1409-1416.e1. 33. Linz D, Schotten U, Neuberger HR, Bohm M, Wirth K. Negative tracheal pressure during obstructive respiratory events promotes atrial fibrillation by vagal activation. Heart Rhythm 2011;8:1436-43. 34. 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;173:910-6. 35. Guilleminault C, Connolly SJ, Winkle RA. Cardiac arrhythmia and conduction disturbances during sleep in 400 patients with sleep apnea syndrome. Am J Cardiol 1983;52:490-4. 36. Goldman ME, Pearce LA, Hart RG, et al. Pathophysiologic correlates of thromboembolism in nonvalvular atrial fibrillation: I. Reduced flow velocity in the left atrial appendage (The Stroke Prevention in Atrial Fibrillation [SPAF-III] study). J Am Soc Echocardiogr 1999;12:1080-7. 37. Shamsuzzaman AS, Gersh BJ, Somers VK. Obstructive sleep apnea: implications for cardiac and vascular disease. JAMA 2003;290:1906-14.
Supplementary Material To access the supplementary material accompanying this article, visit the online version of the Canadian Journal of Cardiology at www.onlinecjc.ca and at http://dx.doi.org/10. 1016/j.cjca.2014.12.032.
Clinical Research
Effect of Nocturnal Intermittent Hypoxia on Left Atrial Appendage Flow Velocity in Atrial Fibrillation Takehiro Kimura, MD, PhD, Takashi Kohno, MD, PhD, Kazuaki Nakajima, MD, Shin Kashimura, MD, Yoshinori Katsumata, MD, PhD, Takahiko Nishiyama, MD, PhD, Nobuhiro Nishiyama, MD, PhD, Yoko Tanimoto, MD, PhD, Yoshiyasu Aizawa, MD, PhD, Keiichi Fukuda, MD, PhD, and Seiji Takatsuki, MD, PhD Division of Cardiology, Department of Medicine, Keio University School of Medicine, Tokyo, Japan
ABSTRACT
RESUM E
Background: The mechanism underlying the associations of sleepdisordered breathing (SDB) with stroke and atrial fibrillation (AF) is not well established. We explored the relationship between nocturnal intermittent hypoxia, a marker of SDB, and left atrial (LA)/LA appendage (LAA) function among AF patients. Methods: We evaluated 134 consecutive AF candidates for catheter ablation (age, 59.6 9.4 years; body mass index [BMI], 24.8 3.2; Congestive Heart Failure, Hypertension, Age (75 years), Diabetes, Stroke/Transient Ischemic Attack, Vascular Disease, Age (65-74 years), Sex (Female) (CHA2DS2-VASc) score, 1.2 1.1, paroxysmal AF, n ¼ 83) using nocturnal pulse oximetry, a noninvasive screening method for nocturnal intermittent hypoxia. Based on 3% oxygen desaturation index (3% ODI), patients were divided into nocturnal intermittent hypoxia (3% ODI > 15; n ¼ 32) and control groups (3% ODI 15; n ¼ 102). Results: The nocturnal intermittent hypoxia group demonstrated significantly higher weight, BMI, Congestive Heart Failure, Hypertension, Age, Diabetes, Stroke/Transient Ischemic Attack (CHADS2) and CHA2DS2-VASc scores, serum hemoglobin A1c and plasma brain natriuretic peptide levels, LA size, and prevalence of hypertension, vascular disease, and sick sinus syndrome. Echocardiographically, nocturnal intermittent hypoxia was associated with a higher grade of
canisme qui sous-tend l’association entre les Introduction : Le me troubles respiratoires du sommeil (TRS), et l’accident vasculaire re bral et la fibrillation auriculaire (FA) n’est pas bien e tabli. Nous ce le lien entre l’hypoxie intermittente nocturne, un maravons examine queur des TRS, et le fonctionnement de l’oreillette gauche (OG) et de l’appendice auriculaire gauche (AAG) chez les patients souffrant de FA. thodes : Nous avons e value l’ablation par cathe ter chez 134 candiMe cutifs souffrant de FA (âge, 59,6 9,4 ans; indice de masse dats conse corporelle [IMC], 24,8 3,2; score CHA2DS2-VASc [Congestive Heart Failure, Hypertension, Age, Diabetes, Stroke/Transient Ischemic Attack, Vascular Disease, Age (65-74 years), Sex (Female), soit l’insuffisance cardiaque congestive, l’hypertension, l’âge, le diabète, l’accident vascure bral/l’ische mie ce re brale transitoire, la maladie vasculaire, laire ce minin)], 1,2 1,1, FA paroxystique, n ¼ 83) l’âge (65 à 74 ans), le sexe (fe trie de pouls nocturne, une me thode de de pistage non à l’aide de l’oxyme invasive de l’hypoxie intermittente nocturne. À partir de l’indice de saturation en oxygène de 3 % (IDO de 3 %), les patients ont e te divise s de comme suit : hypoxie intermittente nocturne (IDO de 3 % > 15; n ¼ 32) et moins (IDO de 3 % 15; n ¼ 102). groupes te sultats : Le groupe souffrant d’hypoxie intermittente a de montre un Re poids, une IMC, un score CHADS2 (Congestive Heart Failure, Hypertension, Age, Diabetes, Stroke/Transient Ischemic Attack, soit l’insuffisance
Sleep-disordered breathing (SDB) is a common but largely undiagnosed disease that is associated with increased cardiovascular morbidity and mortality.1 SDB patients tend to be older, with a higher body mass index (BMI), and increased prevalence of hypertension and diabetes.2 These risk factors
for SDB are also common in stroke and atrial fibrillation (AF).3,4 Approximately 50% of AF patients were reported to have SDB,5 and SDB is associated with increased incidence of AF.6 Severe SDB was also observed in 29% of Japanese stroke patients7 and was also associated with stroke.8 AF is also an established major risk factor for stroke and quintuples stroke incidence.9 Thus, SDB, AF, and stroke are closely related, but studies on the association among these pathologies have been limited thus far. Left atrial (LA) appendage (LAA) has been recognized as the embolic source in 90% of cardiogenic stroke cases.10 Because SDB is associated with adverse cardiac remodelling,11-13 we hypothesized that SDB could be associated with reduced LA/LAA function in patients with AF.
Received for publication October 7, 2014. Accepted December 30, 2014. Corresponding author: Dr Takashi Kohno, Division of Cardiology, Department of Medicine, Keio University School of Medicine, 35 Shinanomachi Shinjuku-ku, Tokyo 160-8582, Japan. Tel.: þ81-3-33531211 61421; fax: þ81-3-5363-3875. E-mail: [email protected] See page 851 for disclosure information.
http://dx.doi.org/10.1016/j.cjca.2014.12.032 0828-282X/Ó 2015 Canadian Cardiovascular Society. Published by Elsevier Inc. All rights reserved.
Kimura et al. Nocturnal Intermittent Hypoxia in AF
847
spontaneous echo contrast and low LAA flow velocity. Multiple regression analysis adjusted for type of AF, CHA2DS2-VASc score, BMI, plasma brain natriuretic peptide level, LA size, and rhythm on echocardiography revealed that 3% ODI was a factor independently associated with LAA flow velocity (b ¼ 0.184; 95% confidence interval, 0.818 to 0.006). Conclusions: Nocturnal intermittent hypoxia was an independent determinant for low LAA flow velocity in patients with AF, suggesting that the connection between SDB and LAA function might underlie the association of AF with stroke.
cardiaque congestive, l’hypertension, l’âge, le diabète, l’accident vasre bral/l’ische mie ce re brale transitoire) et un score culaire ce rique d’he moglobine A1c et une CHA2DS2-VASc, une concentration se tique de type B, une concentration plasmatique du peptide natriure valence de l’hypertension, de la maladie taille de l’OG, et une pre vasculaire et de la maladie du nœud sinusal significativement plus leve s. Échocardiographiquement, l’hypoxie intermittente nocturne a e te associe e à un degre plus e leve de contraste e chographique e et à une ve locite du flux de l’AAG. L’analyse de re gression spontane e du type de FA, du score CHA2DS2-VASc, de l’IMC, de la multiple ajuste tique de type B, de la concentration plasmatique du peptide natriure chocardiographie a re ve le que l’IDO de taille et du rythme de l’OG à l’e tait un facteur inde pendamment associe à la ve locite du flux de 3%e l’AAG (b ¼ 0,184; intervalle de confiance à 95 %, 0,818 à 0,006). tait un de terminant Conclusions : L’hypoxie intermittente nocturne e pendant de la faible ve locite du flux de l’AAG chez les patients inde souffrant de FA, ce qui suggère que le lien entre les TRS et le fonctionnement de l’AAG pourrait sous-tendre l’association entre la FA et re bral. l’accident vasculaire ce
Despite the importance of SDB in managing AF,14-19 the prevalence of undiagnosed SDB is surprisingly high, as shown in the Wisconsin Sleep Cohort Study.20 Although polysomnography (PSG) has been used as a standard for the diagnosis of SDB, referral to sleep study is often difficult, inconvenient, and expensive. Performing fully equipped sleep studies in all AF patients is not feasible. Here, we chose to perform nocturnal pulse oximetry to detect nocturnal intermittent hypoxia, a surrogate marker for SDB,21,22 as a simple screening utility to identify undiagnosed SDB in all patients hospitalized for AF catheter ablation in our facility. In the present study we aimed to estimate the prevalence of nocturnal intermittent hypoxia among AF catheter ablation candidates using nocturnal pulse oximetry and to clarify its relationship with background characteristics, and laboratory, echocardiographic, and Holter monitoring data. Additionally, we assessed the association of LA/LAA function and nocturnal intermittent hypoxia.
Nocturnal intermittent hypoxia
Methods The study protocol conformed to the ethical guidelines of the Declaration of Helsinki of 1975 as reflected in a priori approval by our institution’s human research committee. All patients provided written informed consent to participate. Study population One hundred thirty-four consecutive AF patients who were candidates for catheter ablation in our facility were evaluated retrospectively. The indication for AF catheter ablation was determined according to symptoms due to AF attacks, anti-arrhythmic drug resistance, or patient request, regardless of the type of AF. We excluded patients who were previously diagnosed with SDB and treated with continuous positive airway pressure (CPAP). Patients with chronic obstructive pulmonary disease who required home oxygen therapy were not included in this study. Patients who were contraindicated for anticoagulation or with pre-existing thrombi in the LA were also excluded.
Nocturnal pulse oximetry was performed to identify the presence of nocturnal intermittent hypoxia in all patients 1 day before ablation. Arterial oxyhemoglobin saturation was recorded using a finger probe at a 1-Hz sampling frequency and 5-second average time (PULSOX-Me300; Teijin Pharma, Tokyo, Japan). These recordings were scored using specialized software (DS-Me; Teijin Pharma, Tokyo, Japan). Because the measurement time of pulse oximetry is often longer than the true total sleep time, we used a single-night sleep log to exclude waking time from the analysis and minimize the potential for overestimating total sleep time. We used oxygen desaturation index 3% (3% ODI), the frequency of episodes of 3% desaturation per hour, as an indicator of nocturnal intermittent hypoxia. Based on 3% ODI, patients were divided into nocturnal intermittent hypoxia (3% ODI > 15) and control groups (3% ODI 15). The validity of pulse oximetry has been previously reported based on synchronous overnight recording of pulse oximetry and standard PSG, and its sensitivity and specificity were 85% and 100%, respectively, for detecting apnea hypopnea index of 20 determined using PSG and a cutoff threshold of 3% ODI ¼ 15.23-25 Parameters for analysis Patient background data including age, height, weight, BMI, Congestive Heart Failure, Hypertension, Age, Diabetes, Stroke/Transient Ischemic Attack (CHADS2) score,26 Congestive Heart Failure, Hypertension, Age (75 years), Diabetes, Stroke/Transient Ischemic Attack, Vascular Disease, Age (65-74 years), Sex (Female) (CHA2DS2-VASc) score,27 paroxysmal AF duration from initial diagnosis to ablation, duration of persistent AF, number of anti-arrhythmic drugs before ablation, number of ablation sessions, type of AF, history of dyslipidemia, and sick sinus syndrome (SSS) were collected and compared between groups. Type of AF was classified as paroxysmal (persisting < 1 week), persistent (persisting longer than 1 week and < 1 year), or longstanding persistent (persisting longer than 1 year) according to American College of Cardiology/American Heart Association/Heart
848
Canadian Journal of Cardiology Volume 31 2015
Table 1. Comparison of atrial fibrillation type and CHADS2/CHA2DS2-VASc scores 3% ODI 15 (n ¼ 102) Persistent and LS-persistent AF Congestive heart failure Hypertension Age older than 65 years Age older than 75 years Diabetes Stroke Vascular disease Female sex Dyslipidemia Sick sinus syndrome
33.3 4.9 34.3 29.4 2.9 8.8 3.9 0 13.7 24.5 6.9
(34) (5) (35) (30) (3) (9) (4) (0) (14) (25) (7)
3% ODI > 15 (n ¼ 32) 53.1 9.4 71.9 40.6 3.1 18.8 9.4 6.3 3.1 31.3 18.8
(17) (3) (23) (13) (1) (6) (3) (2) (1) (10) (6)
P
OR
95% CI
0.044 0.351 0.000 0.236 0.957 0.120 0.226 0.011 0.097 0.449 0.047
2.267 2.007 4.892 1.642 1.065 2.385 2.534 N/A 0.203 1.400 3.132
1.011-5.081 0.452-8.907 2.045-11.704 0.720-3.744 0.107-10.605 0.777-7.314 0.536-11.98 N/A 0.026-1.606 0.585-3.352 1.012-9.733
Values are % (n). CHADS2, Congestive Heart Failure, Hypertension, Age (75 years), Diabetes, Stroke/Transient Ischemic Attack; CHA2DS2-VASc, Congestive Heart Failure, Hypertension, Age (75 years), Diabetes, Stroke/Transient Ischemic Attack, Vascular Disease, Age (65-74 years), Sex (Female); CI, confidence interval; LSpersistent AF, longstanding persistent atrial fibrillation; N/A, not applicable; ODI, oxygen desaturation index; OR, odds ratio.
Rhythm Society guidelines.28 Blood samples to test serum levels of creatinine, hemoglobin A1c (HbA1c), low-density lipoprotein, uric acid, and plasma brain natriuretic peptide (BNP) were obtained from all patients before ablation. Transthoracic echocardiography (TTE) was performed in all patients, and transesophageal echocardiography (TEE) was performed in 126 patients 1 week before ablation. The systolic (emptying) and diastolic (filling) flow velocity of the LAA (LAA flow) were averaged. Consistent with previous reports, the severity of spontaneous echo contrast (SEC) was graded from 0 to 3: 0 ¼ none; 0.5 ¼ none to mild; 1 ¼ mild; 2 ¼ moderate; and 3 ¼ severe.29,30 All echocardiographic studies were interpreted by the same 2 experienced echocardiologists without the knowledge of the patient’s background. Holter monitoring was performed in 50 of 83 paroxysmal AF patients, and the incidence of nocturnal AF attacks was evaluated. A nocturnal AF attack was defined as AF initiation during sleep, regardless of its symptoms, and patients with persistent AF documented over a whole day were excluded. Statistical methods Parametric data were expressed as mean SD. The MannWhitney U test was used to compare parameters across groups. The correlation between parameters was investigated using Pearson correlation coefficient test. The c2 test was used to compare the number of patients across groups. Odds ratios and 95% confidence intervals [CIs] were also computed. Multivariate analysis was also performed to elucidate independent determinants. Parameters with statistical significance in the Mann-Whitney U test or the c2 test (P < 0.10) were included in the multivariate analysis. P values < 0.05 were considered statistically significant. Analyses were performed using IBM SPSS software (version 22; IBM, Armonk, NY). Results Overall analysis A total of 134 AF patients (age, 59.6 9.4 years; BMI, 24.8 3.2; CHADS2 score, 0.7 0.9; CHA2DS2-VASc score, 1.2 1.1; paroxysmal AF, n ¼ 83; persistent AF, n ¼ 30; longstanding persistent AF, n ¼ 21) were evaluated. A total of 32 patients who demonstrated nocturnal intermittent hypoxia
were in the nocturnal intermittent hypoxia group, with a mean 3% ODI of 23.1 8.5, compared with 102 patients in the control group with a mean 3% ODI of 6.7 3.8 (P < 0.001). The percentage of persistent and longstanding persistent AF patients was significantly greater in the nocturnal intermittent hypoxia group (persistent AF, 28.1% [n ¼ 9] and longstanding persistent AF, 25.0% [n ¼ 8]) compared with the control group (persistent AF, 20.5% [n ¼ 21] and longstanding persistent AF, 12.7% [n ¼ 13]; Table 1). The prevalence of hypertension, vascular disease, SSS, body weight, and BMI was significantly greater in the nocturnal intermittent hypoxia group (Tables 1 and 2). CHADS2 and CHA2DS2-VASc scores were also significantly higher, but age, number of AADs used before ablation, and number of ablation sessions did not differ between the groups. Laboratory analyses indicated that serum HbA1c and plasma BNP levels were significantly higher in the nocturnal intermittent hypoxia group. TTE data revealed that the LA size was significantly larger in the nocturnal intermittent hypoxia group, and most notably among patients with AF rhythm during TTE. The percentage of patients evaluated using TEE did not differ between groups (nocturnal intermittent hypoxia, 96.9% [n ¼ 31] vs control, 93.1% [n ¼ 95]; P ¼ 0.436). TEE data revealed that nocturnal intermittent hypoxia was associated with a higher grade of SEC, although the difference was not significant when the rhythm during TEE was considered. LAA flow velocity was lower in the nocturnal intermittent hypoxia group, which was significant most notably in patients with AF rhythm during TEE (Table 2 and Fig. 1). The value of 3% ODI was significantly correlated with mean saturation data of pulse oximetry (r ¼ 0.604; P < 0.001), minimum saturation data of pulse oximetry (r ¼ 0.565; P < 0.001), BMI (r ¼ 0.387; P < 0.001), LA size (r ¼ 0.366; P < 0.001), plasma BNP level (r ¼ 0.332; P < 0.001), weight (r ¼ 0.323; P < 0.001), LAA flow velocity (r ¼ 0.304; P ¼ 0.001), grade of SEC (r ¼ 0.299; P ¼ 0.001), CHADS2 score (r ¼ 0.286; P ¼ 0.001), paroxysmal AF duration (r ¼ 0.281; P ¼ 0.017), CHA2DS2-VASc score (r ¼ 0.265; P ¼ 0.002), E/e’ (r ¼ 0.260; P ¼ 0.019), and HbA1c level (r ¼ 0.249; P ¼ 0.006). Analysis of paroxysmal and persistent AF Next, we analyzed paroxysmal and persistent AF separately and evaluated the associations between nocturnal intermittent
Kimura et al. Nocturnal Intermittent Hypoxia in AF
849
Table 2. Comparison of patient background and echocardiographic data Total (n ¼ 134) 3% ODI Mean SpO2, % Minimum SpO2, % Age Height, cm Weight, kg BMI CHADS2 CHA2DS2-VASc No. AAD No. ablation Cr, mg/dL HbA1c, % LDL, mg/dL BNP, pg/mL UA, mg/dL LVEDD, cm LVESD, cm LVEF, % LA, cm Total NSR AF E, cm/sec A, cm/sec Dct, msec E/e’ SEC Total NSR AF LAA flow (cm/sec) Total NSR AF
3% ODI 15 (n ¼ 102)
3% ODI > 15 (n ¼ 32)
P
10.6 95.7 82.8 59.6 168.3 70.6 24.8 0.7 1.2 0.6 1.3 0.9 5.4 108 89.6 6.2 4.8 3.1 58.3
8.8 1.3 6.6 9.4 8.2 12.9 3.2 0.9 1.1 0.8 0.5 0.5 0.5 27 89.8 1.4 0.5 0.5 9.1
6.7 96 84.4 59.3 167.8 68.3 24.2 0.6 1 0.7 1.2 1 5.3 108.1 79.9 6.1 4.7 3 58.9
3.8 1.1 5.5 9.7 8 11.2 2.8 0.8 1 0.8 0.5 0.6 0.5 27.5 82.7 1.3 0.5 0.4 7.9
23.1 94.7 77 60.5 170 78.3 26.9 1.2 1.7 0.6 1.3 0.9 5.6 107.4 120.4 6.3 4.9 3.3 56.2
8.5 1.5 7.2 8.4 8.7 15.2 3.6 1 1.3 0.8 0.7 0.2 0.6 25.5 104.9 1.4 0.6 0.7 12.2
0.000 0.000 0.000 0.681 0.188 0.000 0.000 0.001 0.004 0.384 0.816 0.963 0.004 0.942 0.012 0.364 0.068 0.162 0.343
4 3.7 4.2 70.9 60.8 189.2 8.9
0.6 0.6 (66) 0.6 (68) 15.4 15.7 51.3 4
3.9 3.7 4.1 70.9 61.7 191.2 8.4
0.6 0.6 (57) 0.6 (45) 16.2 14.6 49.7 3.8
4.3 4.1 4.4 71 54.7 183.7 10.6
0.6 0.7 (9) 0.5 (23) 13.2 21.9 55.9 4.4
0.000 0.127 0.020 0.807 0.344 0.731 0.030
0.5 0.6 (126) 0.3 0.4 (59) 0.7 0.7 (67)
0.4 0.6 (95) 0.3 0.4 (48) 0.6 0.7 (47)
0.7 0.7 (31) 0.4 0.4 (11) 0.9 0.8 (20)
0.045 0.290 0.151
54.9 18.9 (126) 59.2 16.6 (59) 51.1 20.2 (67)
57.6 18.4 (95) 60.8 15.7 (48) 54.3 20.5 (47)
46.6 18.3 (31) 52.1 19.1 (11) 43.6 17.6 (20)
0.005 0.116 0.045
Values are mean SD or mean SD (n). AF, atrial fibrillation during echocardiography; BMI, body mass index; BNP, plasma brain natriuretic peptide level; CHADS2, Congestive Heart Failure, Hypertension, Age, Diabetes, Stroke/Transient Ischemic Attack; CHA2DS2-VASc, Congestive Heart Failure, Hypertension, Age (75 years), Diabetes, Stroke/ Transient Ischemic Attack, Vascular Disease, Age (65-74 years), Sex (Female); Cr, serum creatinine level; Dct, deceleration time; E/e’, calculated E/e’ of the septum; Hb, hemoglobin; LA, left atrial diameter measured using transthoracic echocardiography; LAA flow, average flow velocity of systolic (emptying) and diastolic (filling) of the left atrial appendage measured using transesophageal echocardiography; LDL, serum low-density lipoprotein level; LVEDD, left ventricular end diastolic diameter; LVEF, left ventricular ejection fraction; LVESD, left ventricular end systolic diameter; No. AAD, number of anti-arrhythmic drugs before ablation; No. ablation, number of ablation sessions; NSR, normal sinus rhythm during echocardiography; 3% ODI, frequency of episodes of 3% desaturation per hour; SEC, grade of spontaneous echo contrast measured using transesophageal echocardiography; SpO2, saturation data of pulse oximetry; UA, serum uric acid level.
hypoxia and patient characteristics in each population. Among 83 paroxysmal AF patients, 68 patients in the control group with a mean 3% ODI of 6.5 4.0 were compared with 15 patients in the nocturnal intermittent hypoxia group with a mean 3% ODI of 21.8 7.9 (Supplemental Table S1). In the nocturnal intermittent hypoxia group, weight was significantly greater, but no differences were observed in BMI, CHADS2, and CHA2DS2-VASc scores, serum HbA1c level, or plasma BNP level between the groups. The disease duration of paroxysmal AF was significantly longer in the nocturnal intermittent hypoxia group. TTE data demonstrated a significantly larger LA size in the nocturnal intermittent hypoxia group, especially in patients with AF rhythm during TEE. LAA flow was significantly lower in the nocturnal intermittent hypoxia group, especially in patients with AF rhythm during TEE (Fig. 1). Neither the percentage of paroxysmal AF patients evaluated with Holter monitoring nor the incidence of nocturnal AF attacks differed between the groups. Among 51 persistent AF patients (34 persistent and 17 longstanding persistent AF), 17 patients in the nocturnal
intermittent hypoxia group with a mean 3% ODI of 24.2 9.1 were compared with 34 patients in the control group with a mean 3% ODI of 7.1 3.3 (Supplemental Table S2). In the nocturnal intermittent hypoxia group, weight, BMI, CHADS2 score, and serum HbA1c level were significantly higher. The prevalence of hypertension was also significantly greater (nocturnal intermittent hypoxia, 82.4% [n ¼ 14] vs control group, 35.3% [n ¼ 12], P ¼ 0.002), and the duration of persistent AF tended to be longer in the nocturnal intermittent hypoxia group. The effect of nocturnal intermittent hypoxia on TTE and TEE data in persistent AF tended to be similar to that observed overall or in paroxysmal AF; however, differences were not statistically significant. Three percent ODI as an independent determinant of LAA velocity Multiple regression analysis adjusted for type of AF (b ¼ 0.313; P ¼ 0.002; 95% CI, 19.816 to 4.590), CHA2DS2-VASc score (P ¼ 0.423), BMI (P ¼ 0.553),
850
Canadian Journal of Cardiology Volume 31 2015
data revealed larger LA size, higher grade of SEC, and lower LAA flow velocity in AF patients with 3% ODI > 15. Multiple regression analysis adjusted by these confounding factors along with rhythm during TEE revealed that 3% ODI was an independent determinant for reduced LAA flow velocity (b ¼ 0.184; 95% CI, 0.818 to 0.006), suggesting that nocturnal intermittent hypoxia might be associated with reduced LAA function, which is an important determinant of cardiogenic stroke in patients with AF. Pulse oximetry as a screening tool for SDB
Figure 1. The relationship between the flow velocity of the LAA and the 3% ODI. The LAA flow velocity measured using transesophageal echocardiography is significantly reduced in patients with 3% ODI >15 with respect to the type of AF and the rhythm during measurements. * P < 0.05; ** P < 0.001. AF, atrial fibrillation; LAA, left atrial appendage; NSR, normal sinus rhythm; 3% ODI, frequency of episodes of 3% desaturation per hour.
plasma BNP level (P ¼ 0.209), LA size (P ¼ 0.159), and rhythm during TEE (P ¼ 0.510) revealed that 3% ODI was a factor independently associated with LAA flow velocity (b ¼ 0.184; P ¼ 0.047; 95% CI, 0.818 to 0.006) in all patients (Table 3). In paroxysmal AF patients, 3% ODI was also an independent determinant for LAA flow velocity (b ¼ 0.293; P ¼ 0.002; 95% CI, 1.282 to 0.038) after adjustment for multiple potential confounding factors including LA size (b ¼ 0.502; P ¼ 0.002; 95%, CI 24.394 to 6.103), rhythm during TEE (P ¼ 0.146), AF duration (P ¼ 0.382), and weight (P ¼ 0.191). Discussion We established the high prevalence of nocturnal intermittent hypoxia in AF ablation candidates using pulse oximetry as a simple screening method for SDB. AF patients with 3% ODI > 15 had significantly greater weight, BMI, CHADS2 score, CHA2DS2-VASc score, serum HbA1c, and plasma BNP level, and greater prevalence of persistent AF, hypertension, vascular disease, and SSS. Echocardiography Table 3. Multivariate analysis of determinants of LAA flow velocity 3% ODI Persistent AF AF during TEE CHA2DS2-VASc BMI BNP LA size
Standardized b
P
0.184 0.313 0.067 0.07 0.059 0.123 0.147
0.047 0.002 0.51 0.423 0.553 0.209 0.159
95% CI 0.818 19.816 5.094 4.253 0.799 0.067 10.871
to to to to to to to
0.006 4.59 10.197 1.797 1.486 0.015 1.802
AF, atrial fibrillation; BMI, body mass index; BNP, plasma brain natriuretic peptide level; CHA2DS2-VASc, Congestive Heart Failure, Hypertension, Age (75 years), Diabetes, Stroke/Transient Ischemic Attack, Vascular Disease, Age (65-74 years), Sex (Female); CI, confidence interval; LA, left atrium; LAA, left atrial appendage; 3% ODI, frequency of episodes of 3% desaturation per hour; TEE, transesophageal echocardiography.
SDB is highly prevalent and recognized as a potential risk factor for the development of AF.6 Patients with untreated SDB had a greater rate of recurrence of AF after medical therapy or electrical cardioversion.14 In addition, patients with SDB had more recurrence after ablation due to nonpulmonary vein triggers.15,16 Because a recent study demonstrated that CPAP was an additional therapy for improving ablation outcome,15,17-19 the ability to screen undiagnosed SDB patients efficiently among a large number of AF ablation candidates is valuable to clinical practice. We chose pulse oximetry as a simple screening method based on its validity, which was established using synchronous overnight PSG.22,25 In the present study, nocturnal pulse oximetry detected 23.9% of AF ablation candidates to have nocturnal intermittent hypoxia based on 3% ODI > 15. The prevalence of men and women with 3% ODI > 15 in Japanese community-based populations was reported as 7.9%-9.2% and 2.0%-3.1%, respectively, and mean 3% ODI was 2.3 events per hour.24 Our results suggest that its prevalence and severity might be greater among AF patients than in the general population. Because the use of the current screening process using PSG can only identify a limited number of SDB patients among a large cohort of ablation candidates, more patients might benefit from the timely diagnosis and treatment of SDB using nocturnal pulse oximetry as a readily available and inexpensive screening system. Nocturnal intermittent hypoxia, LA remodelling, and nocturnal arrhythmia In patients with obstructive sleep apnea, mechanical and autonomic cardiovascular effects and increased oxidative stress and inflammation are known to exacerbate LA overload and remodelling, leading to the development of structural and electrical substrates for AF.11 We demonstrated that nocturnal intermittent hypoxia was significantly associated with increased LA size and a greater prevalence of SSS. In addition, although the mean age of paroxysmal AF patients did not differ between the nocturnal intermittent hypoxia and control groups, the disease duration of paroxysmal AF was significantly longer in the nocturnal intermittent hypoxia group, which might suggest an earlier onset of AF in the presence of SDB. We also found that the incidence of persistent AF was significantly higher among patients with 3% ODI > 15. SDB itself might increase the likelihood that AF is persistent, or a hemodynamic consequence of persistent AF could contribute to SDB deterioration. Although our study could not clarify the cause-and-effect relationship between nocturnal intermittent hypoxia and AF perpetuation, several unique data demonstrated a direct effect of SDB on AF development in an
Kimura et al. Nocturnal Intermittent Hypoxia in AF
experimental rat model.31,32 Frequent episodes of intermittent hypoxia along with extremes in negative intrathoracic pressure might be associated with the progression of atrial remodelling.31,32 In an experimental pig model, negative tracheal pressure was a strong trigger of AF development due to shortening of the atrial effective refractory period and thus increase susceptibility to AF through vagal activation.33 The possible link between SDB and nocturnal arrhythmias is supported by the finding that 48% of SDB patients who underwent Holter monitoring demonstrated cardiac arrhythmias during the night.1,34 However, another report found that the mean number of apneic events and lowest oxygen saturation during sleep did not significantly differ between patients with and without nocturnal arrhythmias.35 In the present study, the incidence of nocturnal arrhythmia was not significantly greater among paroxysmal AF patients in the nocturnal intermittent hypoxia group. These results might be a result of the small number of patients or the limitation of detecting nocturnal arrhythmia using 24-hour Holter monitoring. Further evaluation should be conducted with concern for the timing of Holter monitoring and pulse oximetry. SDB, AF, and stroke interrelationships Stratifying patients at high risk for stroke is a major consideration in the management of AF. The Stroke Prevention in Atrial Fibrillation III (SPAF-III) study demonstrated that decreased LAA flow velocity, in addition to age, systolic blood pressure, sustained AF, ischemic heart disease, and LA area, was an independent predictor of SEC, thrombi, and cardioembolic events.36 Although SDB has been commonly observed in AF patients, its association with LAA function has not been established. Only 1 previous study involving 79 AF/atrial flutter patients referred for cardioversion found no significant difference in LAA average velocity between patients with and without SDB.14 However, the definition of SDB and type of AF were not described precisely, and ejection fraction was relatively low compared with that in our study. Our results revealed that nocturnal intermittent hypoxia was independently associated with low LAA flow velocity, which is known to be associated with cardiogenic stroke. In patients with paroxysmal AF, LAA flow velocity was significantly lower in patients with 3% ODI > 15 who demonstrated AF rhythm during TEE, but not in normal sinus rhythm. The period after conversion from AF to sinus rhythm was not measurable in this study, and the effect of atrial stunning might be a potential source of heterogeneity. To resolve this issue, long-term Holter monitoring might be needed in the future. In persistent AF, LAA flow velocity was slightly reduced in the nocturnal intermittent hypoxia group, but the difference was not statistically significant. This lack of significance might be because the number of patients was small or because LAA flow velocity was generally reduced by persistent AF itself. The grade of SEC was also significantly greater in the nocturnal intermittent hypoxia group. In addition to the hemodynamic stasis due to decreased LA/LAA function in SDB patients, SDB itself might be associated with hypercoagulation status1,37; therefore, it is plausible that SDB could be associated with SEC. Further large clinical studies are needed to assess this hypothesis using coagulation biomarkers and more precise quantification of SEC.
851
Limitations First, the number of patients enrolled in this study was small, which might reduce the power of the statistics. Second, PSG evaluation was not performed, and apnea hypopnea index and 3% ODI were not directly compared. Pulse oximetry cannot distinguish obstructive apnea from central apnea. The previous study reported that most SDB patients treated with ablation for AF had apnea that was predominantly obstructive,19 which is consistent with our unpublished data. Pulse oximetry could be a useful screening tool in a population that is unlikely to have central sleep apnea. Furthermore, PSG will be needed to elucidate the important question of which parameters of SDB are associated with AF development and remodelling of the LA and LAA. Third, blood pressure and underlying rhythm might affect the value of pulse oximetry. Fourth, it remains unknown whether nocturnal intermittent hypoxia could be a therapeutic target for CPAP because of a lack of randomized control trials. Based on our data, SDB treatment might be beneficial for stroke prevention by improving LAA function. These issues should be investigated in a further study. Conclusions Nocturnal intermittent hypoxia is an independent determinant for low LAA flow velocity in AF ablation candidates. The association of SDB with reduced LAA function might explain the relationship among AF, SDB, and stroke. Disclosures The authors have no conflicts of interest to disclose. References 1. Somers VK, White DP, Amin R, et al. Sleep apnea and cardiovascular disease: an American Heart Association/American College of Cardiology Foundation Scientific Statement from the American Heart Association Council for High Blood Pressure Research Professional Education Committee, Council on Clinical Cardiology, Stroke Council, and Council on Cardiovascular Nursing. In collaboration with the National Heart, Lung, and Blood Institute National Center on Sleep Disorders Research (National Institutes of Health). Circulation 2008;118: 1080-111. 2. Young T, Shahar E, Nieto FJ, et al. Predictors of sleep-disordered breathing in community-dwelling adults: the Sleep Heart Health Study. Arch Intern Med 2002;162:893-900. 3. Valenza MC, Baranchuk A, Valenza-Demet G, et al. Prevalence of risk factors for atrial fibrillation and stroke among 1210 patients with sleep disordered breathing. Int J Cardiol 2014;174:73-6. 4. Bassetti C, Aldrich MS. Sleep apnea in acute cerebrovascular diseases: final report on 128 patients. Sleep 1999;22:217-23. 5. Gami AS, Pressman G, Caples SM, et al. Association of atrial fibrillation and obstructive sleep apnea. Circulation 2004;110:364-7. 6. Gami AS, Hodge DO, Herges RM, et al. Obstructive sleep apnea, obesity, and the risk of incident atrial fibrillation. J Am Coll Cardiol 2007;49:565-71. 7. Shibazaki K, Kimura K, Uemura J, et al. Atrial fibrillation is associated with severe sleep-disordered breathing in patients with ischaemic stroke and transient ischaemic attack. Eur J Neurol 2013;20:266-70.
852
8. Muñoz R, Duran-Cantolla J, Martinez-Vila E, et al. Severe sleep apnea and risk of ischemic stroke in the elderly. Stroke 2006;37:2317-21. 9. Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation as an independent risk factor for stroke: the Framingham Study. Stroke 1991;22:983-8. 10. Watson T, Shantsila E, Lip GY. Mechanisms of thrombogenesis in atrial fibrillation: Virchow’s triad revisited. Lancet 2009;373:155-66. 11. Dimitri H, Ng M, Brooks AG, et al. Atrial remodeling in obstructive sleep apnea: implications for atrial fibrillation. Heart Rhythm 2012;9: 321-7. 12. Drager LF, Bortolotto LA, Pedrosa RP, Krieger EM, Lorenzi-Filho G. Left atrial diameter is independently associated with arterial stiffness in patients with obstructive sleep apnea: potential implications for atrial fibrillation. Int J Cardiol 2010;144:257-9. 13. Kim SM, Cho KI, Kwon JH, Lee HG, Kim TI. Impact of obstructive sleep apnea on left atrial functional and structural remodeling beyond obesity. J Cardiol 2012;60:475-83. 14. Kanagala R, Murali NS, Friedman PA, et al. Obstructive sleep apnea and the recurrence of atrial fibrillation. Circulation 2003;107:2589-94. 15. Patel D, Mohanty P, Di Biase L, et al. Safety and efficacy of pulmonary vein antral isolation in patients with obstructive sleep apnea: the impact of continuous positive airway pressure. Circ Arrhythm Electrophysiol 2010;3:445-51. 16. Ng CY, Liu T, Shehata M, et al. Meta-analysis of obstructive sleep apnea as predictor of atrial fibrillation recurrence after catheter ablation. Am J Cardiol 2011;108:47-51. 17. Li L, Wang ZW, Li J, et al. Efficacy of catheter ablation of atrial fibrillation in patients with obstructive sleep apnoea with and without continuous positive airway pressure treatment: a meta-analysis of observational studies. Europace 2014;16:1309-14. 18. Fein AS, Shvilkin A, Shah D, et al. Treatment of obstructive sleep apnea reduces the risk of atrial fibrillation recurrence after catheter ablation. J Am Coll Cardiol 2013;62:300-5. 19. Naruse Y, Tada H, Satoh M, et al. Concomitant obstructive sleep apnea increases the recurrence of atrial fibrillation following radiofrequency catheter ablation of atrial fibrillation: clinical impact of continuous positive airway pressure therapy. Heart Rhythm 2013;10:331-7. 20. Young T, Palta M, Dempsey J, et al. The occurrence of sleep-disordered breathing among middle-aged adults. N Engl J Med 1993;328:1230-5. 21. Tanigawa T. Obstructive sleep apnea: its prevention and screening may contribute to the prevention of hypertension, diabetes and cardiovascular diseases. EPMA J 2011;2:83-9. 22. Vazquez JC, Tsai WH, Flemons WW, et al. Automated analysis of digital oximetry in the diagnosis of obstructive sleep apnoea. Thorax 2000;55: 302-7. 23. Tanigawa T, Tachibana N, Yamagishi K, et al. Usual alcohol consumption and arterial oxygen desaturation during sleep. JAMA 2004;292:923-5. 24. Muraki I, Tanigawa T, Yamagishi K, et al. Nocturnal intermittent hypoxia and the development of type 2 diabetes: the Circulatory Risk in Communities Study (CIRCS). Diabetologia 2010;53:481-8.
Canadian Journal of Cardiology Volume 31 2015 25. Nakamata MKY, Sakai K, Kinebuchi S, et al. The limitation of screening test for patients with sleep apnea syndrome using pulse oximetry. J Jpn Soc Respir Care 2003;12:401-6. 26. Gage BF, Waterman AD, Shannon W, et al. Validation of clinical classification schemes for predicting stroke: results from the National Registry of Atrial Fibrillation. JAMA 2001;285:2864-70. 27. Lip GY, Nieuwlaat R, Pisters R, Lane DA, Crijns HJ. Refining clinical risk stratification for predicting stroke and thromboembolism in atrial fibrillation using a novel risk factor-based approach: the Euro Heart Survey on Atrial Fibrillation. Chest 2010;137:263-72. 28. January CT, Wann LS, Alpert JS, et al. 2014 AHA/ACC/HRS guideline for the management of patients with atrial fibrillation: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines and the Heart Rhythm Society. J Am Coll Cardiol 2014;64:e1-76. 29. Daniel WG, Nellessen U, Schroder E, et al. Left atrial spontaneous echo contrast in mitral valve disease: an indicator for an increased thromboembolic risk. J Am Coll Cardiol 1988;11:1204-11. 30. Fatkin D, Kelly RP, Feneley MP. Relations between left atrial appendage blood flow velocity, spontaneous echocardiographic contrast and thromboembolic risk in vivo. J Am Coll Cardiol 1994;23:961-9. 31. Iwasaki YK, Kato T, Xiong F, et al. Atrial fibrillation promotion with long-term repetitive obstructive sleep apnea in a rat model. J Am Coll Cardiol 2014;64:2013-23. 32. Iwasaki YK, Shi Y, Benito B, et al. Determinants of atrial fibrillation in an animal model of obesity and acute obstructive sleep apnea. Heart Rhythm 2012;9:1409-1416.e1. 33. Linz D, Schotten U, Neuberger HR, Bohm M, Wirth K. Negative tracheal pressure during obstructive respiratory events promotes atrial fibrillation by vagal activation. Heart Rhythm 2011;8:1436-43. 34. 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;173:910-6. 35. Guilleminault C, Connolly SJ, Winkle RA. Cardiac arrhythmia and conduction disturbances during sleep in 400 patients with sleep apnea syndrome. Am J Cardiol 1983;52:490-4. 36. Goldman ME, Pearce LA, Hart RG, et al. Pathophysiologic correlates of thromboembolism in nonvalvular atrial fibrillation: I. Reduced flow velocity in the left atrial appendage (The Stroke Prevention in Atrial Fibrillation [SPAF-III] study). J Am Soc Echocardiogr 1999;12:1080-7. 37. Shamsuzzaman AS, Gersh BJ, Somers VK. Obstructive sleep apnea: implications for cardiac and vascular disease. JAMA 2003;290:1906-14.
Supplementary Material To access the supplementary material accompanying this article, visit the online version of the Canadian Journal of Cardiology at www.onlinecjc.ca and at http://dx.doi.org/10. 1016/j.cjca.2014.12.032.
