- Email: [email protected]
Severe Obstructive Sleep Apnea Is Associated With Left Ventricular Diastolic Dysfunction
Severe Obstructive Sleep Apnea Is Associated With Left Ventricular Diastolic Dysfunction* Jeffrey W. H. Fung, MBChB; Thomas S. T. Li, MBChB; Dominic K. L. Choy, MBBS; Gabriel W. K. Yip, MBChB; Fanny W. S. Ko, MBChB; John E. Sanderson, MD; and David S. C. Hui, MBBS, FCCP
Introduction: Hypertension is common in patients with obstructive sleep apnea (OSA). However, the effect of OSA on ventricular function, especially diastolic function, is not clear. Therefore, we have assessed the prevalence of diastolic dysfunction in patients with OSA and the relationship between diastolic parameters and severity of OSA. Methods: Sixty-eight consecutive patients with OSA confirmed by polysomnography underwent echocardiography. Diastolic function of the left ventricle was determined by transmitral valve pulse-wave Doppler echocardiography. Various baseline characteristics, severity of OSA, and echocardiographic parameters were compared between patients with and without diastolic dysfunction. Results: There were 61 male and 7 female patients with a mean age of 48.1 ⴞ 11.1 years, body mass index of 28.5 ⴞ 4.3 kg/m2, and apnea/hypopnea index (AHI) of 44.3 ⴞ 23.2/h (mean ⴞ SD). An abnormal relaxation pattern (ARP) in diastole was noted in 25 patients (36.8%). Older age (52.7 ⴞ 8.9 years vs 45.1 ⴞ 11.3 years, p ⴝ 0.005), hypertension (56% vs 20%, p ⴝ 0.002), and a lower minimum pulse oximetric saturation (SpO2) during sleep (70.5 ⴞ 17.9% vs 78.8 ⴞ 12.9%, respectively; p ⴝ 0.049) were more common in patients with ARP. By multivariate analysis, minimum SpO2 < 70% was an independent predictor of ARP (odds ratio, 4.34; 95% confidence interval, 1.23 to 15.25; p ⴝ 0.02) irrespective of age and hypertension. Patients with AHI > 40/h had significantly longer isovolumic relaxation times than those with AHI < 40/h (106 ⴞ 19 ms vs 93 ⴞ 17 ms, respectively; p ⴝ 0.005). Conclusion: Diastolic dysfunction with ARP was common in patients with OSA. More severe sleep apnea was associated with a higher degree of left ventricular diastolic dysfunction in this study.
(CHEST 2002; 121:422– 429) Key words: diastolic dysfunction; echocardiography; obstructive sleep apnea Abbreviations: A ⫽ late peak atrial systolic velocity; AHI ⫽ apnea/hypopnea index; ARd ⫽ duration of pulmonary-atrial reversal signal; ARP ⫽ abnormal relaxation pattern; BMI ⫽ body mass index; CHF ⫽ congestive heart failure; CI ⫽ confidence interval; CPAP ⫽ continuous positive airway pressure; CSR-CSA ⫽ Cheyne-Stokes respiration with central sleep apnea; D ⫽ peak diastolic flow velocity in the pulmonary vein; DHF ⫽ diastolic heart failure; DT ⫽ deceleration time; E ⫽ early peak transmitral flow velocity; E/A ratio ⫽ ratio between the early peak transmitral flow velocity and the late peak atrial systolic velocity; ESS ⫽ Epworth Sleepiness Scale; IVRT ⫽ isovolumic relaxation time; LV ⫽ left ventricular; LVEF ⫽ left ventricular ejection fraction; LVH ⫽ left ventricular hypertraophy; LVM ⫽ left ventricular mass; LVMI ⫽ left ventricular mass index; OR ⫽ odds ratio; OSA ⫽ obstructive sleep apnea; S ⫽ peak systolic flow velocity in the pulmonary vein; SDB ⫽ sleep-disordered breathing; Spo2 ⫽ pulse oximetric saturation
sleep apnea (OSA) syndrome is a O bstructive common disorder affecting 2 to 4% of middle*From the Divisions of Cardiology (Drs. Fung, Yip, and Sanderson) and Respiratory Medicine (Drs. Li, Choy, Ko, and Hui), Department of Medicine and Therapeutics, Chinese University of Hong Kong, Prince of Wales Hospital, Shatin, New Territories, Hong Kong. Supported by Chinese University of Hong Kong direct grant (ref. 2040898). Manuscript received February 26, 2001; revision accepted August 15, 2001. Correspondence to: David S. C. Hui, MBBS, FCCP, Department of Medicine and Therapeutics, Chinese University of Hong Kong, Prince of Wales Hospital, Shatin, New Territories, Hong Kong; e-mail: [email protected] 422
aged adults.1,2 Excessive daytime sleepiness is a major consequence of OSA, due to sleep fragmentation triggered by repetitive episodes of partial or complete upper-airway obstruction.3 In a retrospective study by He et al,4 patients with OSA with an apnea index ⬎ 20/h had a higher morbidity and mortality rates related to vascular events than patients with an apnea index ⬍ 20/h. There is growing evidence that patients with OSA have an increased risk of having cardiovascular complications, such as hypertension, cardiac arrhythmia,5 myocardial infarction,6 pulmonary hypertension,7 and stroke.8 Several epidemiologic studies9 –13 have Clinical Investigations
shown an independent association between sleepdisordered breathing (SDB) and hypertension after controlling for confounding factors such as age, body mass index (BMI), sex, alcohol, and smoking. A recent case control study14 has also shown that patients with OSA have increased ambulatory diastolic BP both day and night and increased systolic BP at night. In a study of 200 consecutive Hong Kong Chinese patients presenting clinically with congestive heart failure (CHF), Yip et al15 showed that diastolic heart failure (DHF) is more common than systolic heart failure, with 66% of these patients having a normal left ventricular (LV) ejection fraction (LVEF). Chan et al16 performed sleep studies on 20 Chinese patients with symptomatic DHF and found 55% of patients to have significant SDB of the obstructive type with an apnea/ hypopnea index (AHI) ⬎ 10/h. Patients with DHF and SDB may be associated with worse diastolic dysfunction than those without SDB. Nevertheless, the relationship between diastolic dysfunction and OSA is not clear. This study investigates the prevalence of diastolic dysfunction in patients with newly diagnosed OSA and assesses whether there is any correlation between the severity of OSA and the degree of diastolic dysfunction. Materials and Methods Patients We recruited 68 consecutive patients with newly diagnosed OSA from our sleep unit at the Prince of Wales Hospital for this study. Sleep Study Significant OSA was defined as an AHI ⱖ 5 events per hour of sleep as shown by overnight polysomnography (Alice 4; Healthdyne; Atlanta, GA) plus self-reported sleepiness. Subjective sleepiness was evaluated by the Epworth sleepiness scale (ESS),17,18 a questionnaire specific to symptoms of daytime sleepiness, and the subjects were asked to score the likelihood of falling asleep in eight different situations with different levels of stimulation, adding up to a total score of 0 to 24. Overnight polysomnography recorded EEG, electro-oculogram, submental electromyogram, bilateral anterior tibial electromyogram, ECG, chest and abdominal wall movement by inductance plethysmography, airflow by a nasal pressure transducer (PTAF; Pro-Tech; Woodinville, WA) and backed up by oronasal airflow measured with a thermistor and finger pulse oximetry as in our previous study.19 Sleep stages were scored according to standard criteria by Rechtshaffen and Kales.20 Apnea was defined as cessation of airflow for ⬎ 10 s, and hypopnea was defined as a reduction of airflow ⱖ 50% for 10 s plus an oxygen desaturation of ⬎ 4% or an arousal. Hypertension was defined if patients were receiving antihypertensive medications without regard to the actual measurement of BP, or having a systolic BP ⬎ 140 mm Hg or a diastolic BP ⬎ 90 mm Hg9,11 on awakening following completion of polysomnography. Patients received their usual cardiac medications, includ-
ing antihypertensive agents. Exclusion criteria included malignant hypertension, unstable angina, renal failure, recent myocardial infarction, and significant valvular heart disease. Echocardiography Following confirmation of significant OSA by polysomnography, echocardiography (System FIVE; GE Vingmed; Horten, Norway) was performed to assess the LV size with systolic function assessed by fractional shortening obtained from Mmode recordings of LV systolic and diastolic dimensions,21 as well as by the LVEF estimated from the two-dimensional Simpson’s method.22 An LVEF ⬎ 50% was regarded as normal. LV diastolic dysfunction was assessed principally by Doppler echocardiography for patients with significant OSA, and normal LV systolic function was assessed as previously described.23 The pulse-wave Doppler echocardiographic sample volume was placed at the mitral valve leaflet in the apical four-chamber view. The diastolic parameters were measured from at least three beats and were defined as follows: E-wave, early maximal transmitral flow velocity; A-wave, peak velocity during atrial contraction in late diastole; and ratio between the early peak transmitral flow velocity (E) and late peak atrial systolic velocity (A) [E/A ratio], expressed in terms of peak velocities; and deceleration time (DT), calculated and expressed as the time for the peak filling velocity (E-wave) to fall to zero. Pulse-wave Doppler echocardiographic sample volume was then placed at the inflow area to measure the isovolumic relaxation time (IVRT), the time from aortic valve closure to the onset of mitral valve inflow. The pulse-wave Doppler echocardiographic sample volume was then placed inside the pulmonary vein in the apical four-chamber view. The pulmonary venous flow parameters were defined as follows: S-wave, peak systolic flow velocity in the pulmonary vein (S); D-wave, peak diastolic flow velocity in the pulmonary vein (D); and duration of pulmonary-atrial reversal signal (ARd). Diastolic function of the left ventricle was divided into four patterns: normal, abnormal relaxation pattern (ARP), pseudonormal pattern, and restrictive filling pattern depending on the abovementioned transmitral and pulmonary venous parameters and IVRT. A schematic drawing for the various diastolic dysfunction patterns is shown in Figure 1. Left ventricular mass (LVM) was estimated by M-mode echocardiography. LVM index (LVMI, grams per meters squared) was calculated from LVM corrected by body surface area. Statistical Analysis All data were presented as mean ⫾ SD unless otherwise stated. Comparisons between patients with and without diastolic dysfunction in relation to age, sex, cardiovascular risk factors, echocardiographic parameters, and severity of OSA were performed by 2 and unpaired t tests. For qualitative data, Fisher’s Exact Test was used when expected p value was ⬍ 5. A p value of ⬍ 0.05 was used to indicate differences between the groups that were statistically significant. Data analysis was performed with a commercially available statistical analysis software package (SPSS 10.0 for Windows; SPSS; Chicago, IL).
Results There were 61 male and 7 female consecutive patients with significant OSA. Mean age was 48.1 ⫾ 11.1 years; mean BMI was 28.5 ⫾ 4.3 kg/m2. Thirty-two patients (47%) had hypertension, but CHEST / 121 / 2 / FEBRUARY, 2002
423
Figure 1. Schematic drawing illustrating the four transmitral and pulmonary venous flow patterns for LV diastolic function. The upper tracing in each panel illustrates transmitral flow, while lower tracing illustrates pulmonary flow. Left upper panel: normal transmitral flow pattern: E/A ratio ⬎ 1; DT, 180 to 240 ms; and IVRT, 70 to 110 ms. Normal pulmonary flow pattern: S/D ratio ⬎ 1; pulmonary atrial reversal (AR) velocity ⬍ 25 cm/s; and duration (d) of mitral A-wave/ARd ⬎ 1. Right upper panel: abnormal relaxation transmitral flow pattern: E/A ratio ⬍ 1; DT ⬎ 240 ms; and IVRT ⬎ 110 ms. Abnormal relaxation pulmonary flow pattern: S/D ratio ⬎ 1, AR velocity ⬍ 25 cm/s, and duration of mitral A-wave/ARd ⬎ 1. Left lower panel: pseudonormalized transmitral flow pattern: E/A ratio ⬍ 1; DT, 180 to 240 ms; and IVRT, 70 to 110 ms. Pseudonormalized pulmonary flow pattern: S/D ratio ⬍ 1; AR velocity ⬍ 25 cm/s; and duration of mitral A-wave/ARd ⬍ 1. Right lower panel: restrictive transmitral flow pattern: E/A ratio ⬎ 1; DT ⬍ 180 ms; and IVRT ⬍ 70 ms. Restrictive pulmonary flow pattern: S/D ratio ⬍ 1; AR velocity ⬍ 25 cm/s; and duration of mitral A-wave/ARd⬍ 1.
none had a clinical history of heart failure. Mean ESS score was 12.9 ⫾ 6.0. From the polysomnography, the total sleep time was 6.6 ⫾ 1.2 h, arousal index was 24.2 ⫾ 12.5/h, AHI was 44.4 ⫾ 23.2/h, minimum pulse oximetric saturation (Spo2) was 75.8 ⫾ 15.3%, and mean Spo2 was 89.2 ⫾ 3.9%. From the echocardiography assessment, ARP was the only diastolic dysfunction pattern detected in our OSA patient cohort and was noted in 25 patients (36.8%; 1 female patient). Representative Doppler echocardiographic flow signals of the normal vs ARP patterns are shown in Figure 2. Baseline characteristics of patients with normal diastolic function vs those with ARP diastolic dysfunction are shown in Table 1. OSA patients with ARP diastolic dysfunc424
tion were older and had a higher prevalence of hypertension than those with normal diastolic function. Polysomnographic and echocardiographic data in patients with normal diastolic vs those with ARP diastolic dysfunction are shown in Table 2. Diastolic dysfunction with ARP was found to be related only to the minimum Spo2. AHI and mean Spo2 did not correlate with ARP diastolic dysfunction, while there was no significant difference between the two groups in other polysomnographic data. Both groups had comparable and normal LVEF. By multivariate analysis, minimum Spo2 ⬍ 70% during sleep (odds ratio [OR], 4.34; 95% confidence interval [CI], 1.23 to 15.25; p ⫽ 0.02), hypertension (OR, 4.05; 95% CI, Clinical Investigations
Figure 2. Top: Doppler echocardiographic tracings illustrating normal LV diastolic function (left, transmitral pulse-wave tracing; right, pulmonary venous pulse-wave tracing). Bottom: Doppler echocardiographic ARP of LV diastolic dysfunction (left, transmitral pulse-wave tracing; right, pulmonary venous pulse-wave tracing). V ⫽ position of transducer.
Table 1—Comparison of Baseline Characteristics of OSA Patients With and Without Diastolic Dysfunction of ARP*
Table 2—Comparison of Polysomnographic and Echocardiographic Data in OSA Patients With Normal and ARP Diastolic Dysfunction
Characteristics
ARP
Normal
p Value
Variables
ARP
Normal
p Value
Male/female patients, No. Age, yr Systolic BP, mm Hg† Diastolic BP, mm Hg† BMI, kg/m2 Hypertension Cerebrovascular accident Ischemic heart disease Diabetes mellitus
24/1
37/6
0.248
53.2 ⫾ 8.7 142.5 ⫾ 16.9 84.4 ⫾ 13.8 29.1 ⫾ 4.8 18 (72.0) 2 (8.0)
45.1 ⫾ 11.3 137.1 ⫾ 18.8 80.6 ⫾ 11.0 28.2 ⫾ 4.1 14 (32.6) 2 (4.7)
0.003‡ 0.305 0.288 0.448 0.002‡ 0.621
Patients, No. ESS Total sleep time, h Sleep efficiency, % Arousal index, No./h AHI Minimum Spo2, %† Mean Spo2, % LVEF, % E/A ratio DT, ms IVRT, ms
25 12.8 ⫾ 6.5 6.4 ⫾ 1.0 62.7 ⫾ 11.0 24.9 ⫾ 12.6 46.2 ⫾ 24.8 70.5 ⫾ 17.9 88.4 ⫾ 4.0 66.6 ⫾ 10.6 0.75 ⫾ 0.12 262.8 ⫾ 64.6 102.2 ⫾ 24.9
43 12.9 ⫾ 5.8 6.7 ⫾ 1.3 87.1 ⫾ 6.9 23.9 ⫾ 12.7 43.3 ⫾ 22.4 78.8 ⫾ 12.9 89.7 ⫾ 3.8 68.2 ⫾ 9.5 1.18 ⫾ 0.25 224.8 ⫾ 60.0 100.0 ⫾ 15.3
0.932 0.411 0.076 0.751 0.634 0.049‡ 0.228 0.526 ⬍ 0.001‡ 0.017‡ 0.674
6 (24.0) 7 (28.0)
5 (11.6) 4 (9.3)
0.305 0.084
*Data are presented as mean ⫾ SD or No. (%) unless otherwise indicated. †Measured on completion of sleep study. ‡Of statistical significance.
1.20 to 13.68; p ⫽ 0.02), and age ⬎ 55 years (OR, 5.17; 95% CI, 1.39 to 19.28; p ⫽ 0.01) were independent predictors of ARP diastolic dysfunction. To assess any relationship between severity of OSA and diastolic parameters, patients were classified into two groups according to the AHI. Patients with more severe OSA (AHI ⱖ 40/h) had a longer IVRT than patients with AHI ⬍ 40/h. There were 18
*Data are presented as mean ⫾ SD unless otherwise indicated. †During sleep. ‡Of statistical significance.
patients and 14 patients with hypertension in the two groups, respectively (p ⫽ 0.77). There was no significant difference in E, A, E/A ratio, and DT between these two groups, as shown in Table 3. The relationship between diastolic parameters, diastolic BP, mean BP, minimum Spo2 during sleep, and AHI vs LVMI was analyzed by the Pearson correlation. LVMI had a significant negative correCHEST / 121 / 2 / FEBRUARY, 2002
425
Table 3—Diastolic Parameters Between Two Groups of OSA Patients With AHI > 40/h vs AHI < 40/h* Parameters
AHI ⱖ 40/h (n ⫽ 39)
AHI ⬍ 40/h (n ⫽ 29)
p Value
E, m/s A, m/s E/A ratio DT, ms IVRT, ms
0.68 ⫾ 0.12 0.70 ⫾ 0.12 0.98 ⫾ 0.22 248.7 ⫾ 61.6 106.4 ⫾ 19.1
0.72 ⫾ 0.17 0.70 ⫾ 0.27 1.06 ⫾ 0.39 225.4 ⫾ 66.0 92.7 ⫾ 16.6
0.211 0.991 0.299 0.142 0.005†
*Data are presented as mean ⫾ SD. †Of statistical significance.
lation with peak E-wave velocity (r ⫽ 0.419, p ⫽ 0.001) and minimum Spo2 during sleep (r ⫽ 0.257, p ⫽ 0.040), and a positive correlation with the diastolic BP (r ⫽ 0.325, p ⫽ 0.023) and mean BP (r ⫽ 0.315, p ⫽ 0.027). LVMI had no significant correlation with either AHI (r ⫽ 0.109, p ⫽ 0.393) or mean Spo2 during sleep (r ⫽ 0.143, p ⫽ 0.269). DT was related to the minimum Spo2 during sleep (r ⫽ 0.27, p ⫽ 0.029), but peak E-wave velocity was not (r ⫽ 0.14, p ⫽ 0.29). Discussion Diastolic dysfunction is a condition with increased resistance to filling of the left ventricle, leading to an inappropriate rise in the diastolic pressure-volume relationship and causing symptoms of pulmonary congestion during exercise. In contrast, DHF indicates that all these changes occur when the patient is at rest.24 This study with echocardiography has shown that diastolic dysfunction occurred in 36.8% of our patients with OSA. An ARP was the only diastolic dysfunction pattern observed. All these OSA patients had normal LV systolic function and no history of overt heart failure. The OSA patients with diastolic dysfunction were older with a higher prevalence of hypertension, and diastolic dysfunction was associated with more severe minimum Spo2 during sleep, irrespective of age or hypertension. In patients with more severe OSA, as reflected by AHI ⱖ 40/h, the diastolic parameter, IVRT, was significantly longer than those with AHI ⬍ 40/h. IVRT is the time between the closure of the aortic valve and opening of the mitral valve, and reflects the compliance of the left ventricle independent of the effect of age.25 The potential mechanisms leading to changes in cardiac structure and function in patients with OSA have been studied in animal models. Fletcher et al26 demonstrated ventricular hypertrophy in rats exposed to short bursts of repetitive hypoxia over an extended period and that intermittent severe hypoxia can lead to a sustained rise in BP within 35 days. 426
Brooks et al,27 by inducing OSA via tracheostomy in the canine model, showed that OSA can lead to the development of sustained hypertension over approximately 100 days. Parker et al,28 using the same canine model, showed that in patients with chronic OSA, acute airway occlusion during sleep is associated with increase in LV afterload and decrease in fractional shortening, whereas chronic OSA also leads to sustained decrease in LV systolic performance that can be caused by the development of systemic hypertension and/or transient increase in LV afterload during episodes of airway obstruction. Nevertheless, the observation that severe OSA, induced in the canine model over 3 months, leading to LV systolic dysfunction, may not be the same in the human model, as the clinical syndrome severity in humans is highly variable and typically evolves over many years.29 There are conflicting data on the effect of OSA on the cardiac structure and function in human subjects. Increased LV wall thickness independent of daytime BP has been observed in OSA patients compared to age-matched and BMI-matched control subjects, suggesting a direct effect of nocturnal hypertension.30 A study by Davies et al,31 investigating a small group of OSA patients, snorers, and control subjects matched for age, sex, obesity, smoking, and alcohol consumption, found no difference between groups in LV diameter or thickness, or LVM. Hanly et al32 reported normal indexes of LV function between a group of typical OSA patients and a group of habitual snorers without OSA. Noda et al33 reported LV hypertrophy (LVH) in 41% of 51 OSA patients. Nocturnal hypoxia and apnea index were significantly correlated with LVH and 24-h BP. However, obesity was also commonly seen in the group with LVH, and the relation to preexisting hypertension was not clear in this uncontrolled study. However, in a separate study, Noda et al34 showed that the survival rate was significantly lower among untreated middle-aged Japanese hypertensive patients with OSA than normotensive patients (57.9% vs 90.4%, respectively). Recently Alchanatis et al35 reported that LV diastolic function was impaired in 15 OSA patients with neither history of nor current systemic hypertension, compared to 11 subjects matched for age and BMI. In addition, treatment with nasal continuous positive airway pressure (CPAP) over 3 months resulted in significant improvement in both LV diastolic function and diastolic BP, while systolic function and wall thickness remained within normal limits. In our study, LVMI correlated negatively with the minimum Spo2 during sleep and positively with the mean BP and the diastolic BP, suggesting that both hypoxia related to OSA per se and hypertension may cause LVH and Clinical Investigations
impair the diastolic function of the left ventricle. Nevertheless, OSA may have other unfavorable effects on LV diastolic function independent of LVH. There is strong epidemiologic evidence showing an independent association between OSA and hypertension.9 –13 High levels of AHI9 –13 and sleep time below 90% oxygen saturation11,12 were associated with greater odds of hypertension in a dose-response fashion. LV afterload is increased by peripheral vasoconstriction as a result of recurrent arousals terminating the obstructive respiratory events and activating the sympathetic nervous system,36 – 40 and activation of the arterial chemoreceptors by hypoxia and hypercapnia. Plasma levels of nitric oxide, a powerful vasodilator released from the endothelium, have been shown to be decreased in OSA patients but can be promptly reversed by nasal CPAP treatment.41,42 More recently Kraiczi et al43 showed in 20 subjects with OSA that worsening nocturnal hypoxemia (measured as minimum Spo2 or percentage of sleep time with Spo2 below 90%) was associated with a gradual deterioration of LV diastolic function (increased interventricular septum thickness, prolonged IVRT, and decreased E/A ratio), as well as reduced endothelium-dependent dilatory capacity of the brachial artery. Other mechanisms that may result in LV dysfunction include increased preload by intermittent negative intrathoracic pressure during apnea, which may also increase the LV transmural pressure gradient and impair diastolic relaxation and LV filling.44 – 46 Increase in right ventricular volume, together with hypoxia-induced pulmonary hypertension, may displace the interventricular septum leftward during diastole and impair LV filling.47,48 Hypoxia and hypercapnia may also decrease myocardial contractility.49,50 DHF is more common than systolic heart failure in our Chinese population.15 SDB of the obstructive type was noted in 55% of our patients with DHF16; a lower minimum Spo2 during sleep, but not AHI, was associated with more severe diastolic dysfunction.16,43 In this study, asymptomatic diastolic dysfunction was prevalent among our OSA patients; more severe OSA, as reflected by the minimum Spo2 and AHI ⱖ 40/h, was associated with worse diastolic parameters. It is possible that OSA, through hypoxia, hypertension, and other mechanisms discussed earlier, causes LV diastolic dysfunction that, in the long run, may lead to symptomatic DHF. It is important to keep a high index of suspicion for OSA when assessing patients with CHF. Not only can nasal CPAP effectively relieve disabling symptoms such as sleepiness,51 it may potentially improve LV systolic52 and diastolic function,35 decrease activation of the sympathetic nervous system activity,53
and increase nitric oxide levels41,42 in patients with both OSA and heart failure. In addition OSA and Cheyne-Stokes respiration with central sleep apnea (CSR-CSA) can coexist in patients with CHF; however, in contrast to OSA, CSR-CSA is likely a consequence rather than a cause of CHF.54 CSRCSA is associated with increased mortality in CHF, probably because of sympathetic nervous system activation caused by recurrent apnea-induced hypoxia and arousals from sleep.54 Nasal CPAP can reduce ventricular irritability,55 improve LVEF, and reduce the combined mortality-cardiac transplantation rate in such patients.56 There were several limitations in our study. This study was uncontrolled, and we did not perform 24-h BP recording and therefore could not demonstrate any diurnal pattern of changes in BP profile in our patients. Echocardiography is a well-known operator-dependent investigation, but all our studies were performed by the same experienced technician to avoid any individual variation in the assessment. Clear recordings of diastolic parameters were obtained in all our OSA patients via the transthoracic approach. Had there been poor transthoracic echocardiographic images in patients with severe obesity, the transesophageal approach would have been required to assess systolic and diastolic functions. A longitudinal study with serial echocardiography will be of great interest to assess whether long-term nasal CPAP treatment can reverse the diastolic dysfunction in a larger sample size of OSA patients. In summary, this study has shown that an ARP in diastole on echocardiography is common in patients with OSA, and patients with more severe OSA have a higher degree of diastolic dysfunction of the left ventricle. ACKNOWLEDGMENT: The authors thank Ms. Mabel Tong, M.Y. Leung, and Fanny Chan for their technical support with the sleep studies, and Ms. Doris Chan, Dr. Anthony T. Chan. and Mr. K.K. Wong for performing statistical analysis of the data. We are most grateful to Miss Pearl Ho for performing all of the echocardiography studies for this project.
References 1 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 –1235 2 Ip MS, Lam B, Lauder I, et al. A community study of sleep-disordered breathing in middle-aged Chinese men in Hong Kong. Chest 2001; 119:62– 69 3 McNamara SG, Grunstein RR, Sullivan CE. Obstructive sleep apnoea. Thorax 1993; 48:754 –764 4 He J, Kryger M, Zorick F, et al. Mortality and apnea index in obstructive sleep apnea: experience in 385 male patients. Chest 1988; 94:9 –14 5 Shepard JW Jr. Hypertension, cardiac arrhythmias, myocardial infarction and stroke in relation to obstructive sleep apnoea. Clin Chest Med 1992; 13:437– 458 CHEST / 121 / 2 / FEBRUARY, 2002
427
6 Hung J, Whitford E, Parsons R, et al. Association of sleep apnoea with myocardial infarction in men. Lancet 1990; 336:261–264 7 Bady E, Achkar A, Pascal S, et al. Pulmonary arterial hypertension in patients with sleep apnoea syndrome. Thorax 2000; 55:934 –939 8 Wessendorf TE, Teschler H, Wang YM, et al. Sleep-disordered breathing among patients with first-ever stroke. J Neurol 2000; 247:41– 47 9 Young T, Peppard P, Palta M, et al. Population-based study of sleep-disordered breathing as a risk factor for hypertension. Arch Intern Med 1997; 157:1746 –1752 10 Grote L, Ploch T, Heitmann J, et al. Sleep-related breathing disorder is an independent risk factor for systemic hypertension. Am J Respir Crit Care Med 1999; 160:1875–1882 11 Lavie P, Herer P, Hoffstein V. Obstructive sleep apnoea as a risk factor for hypertension: population study. BMJ 2000; 320:479 – 482 12 Nieto FJ, Young TB, Lind BK, et al. Association of sleepdisordered breathing, sleep apnea and hypertension in a large community-based study. JAMA 2000; 283:1829 –1836 13 Peppard P, Young T, Palta M, et al. Prospective study of the association between sleep-disordered breathing and hypertension. N Engl J Med 2000; 342:1378 –1384 14 Davies CW, Crosby JH, Mullins RL, et al. Case-control study of 24 hour ambulatory blood pressure in patients with obstructive sleep apnoea and normal matched control subjects. Thorax 2000; 55:736 –740 15 Yip G, Ho P, Woo KS, et al. Comparison of frequencies of left ventricular systolic and diastolic heart failure in Chinese living in Hong Kong. Am J Cardiol 1999; 84:563–567 16 Chan J, Sanderson J, Chan W, et al. Prevalence of sleepdisordered breathing in diastolic heart failure. Chest 1997; 111:1488 –1493 17 Johns MW. A new method for measuring daytime sleepiness: the Epworth sleepiness scale. Sleep 1991; 14:540 –545 18 Johns MW. Sensitivity and specificity of the multiple sleep latency test (MSLT), the maintenance of wakefulness test and the Epworth sleepiness scale: failure of the MSLT as a gold standard. J Sleep Res 2000; 9:5–11 19 Hui DS, Chan JK, Choy DK, et al. Effects of augmented continuous positive airway pressure education and support on compliance and outcome in a Chinese population. Chest 2000; 117:1410 –1416 20 Rechtschaffen A, Kales A. A manual of standardized terminology, techniques and scoring system for sleep stages of human subjects. Los Angeles, CA: Brain Information Institute, 1968; 1–12 21 Sahn DJ, DeMaria A, Kisslo J, et al. Recommendations regarding quantitation in M-mode echocardiography: results of a survey of echocardiographic measurements. Circulation 1978; 58:1072–1083 22 Folland ED, Parisi AF, Moynihan PF, et al. Assessment of left ventricular ejection fraction and volumes by real-time, two-dimensional echocardiography: a comparison of cineangiographic and radionuclide techniques. Circulation 1979; 60:760 –766 23 Cohen GJ, Pietrolungo JF, Thomas JD, et al. A practical guide to assessment of ventricular diastolic function using Doppler echocardiography. J Am Coll Cardiol 1996; 27:1753– 1760 24 Brutsaert DL. Diagnosing primary diastolic heart failure. Eur Heart J 2000; 21:94 –96 25 Yu CM, Sanderson JE. Right and left ventricular diastolic function in heart failure and normal-effect of age, gender, heart rate and respiration on Doppler-derived measurements. Am Heart J 1997; 134:426 – 434 428
26 Fletcher E, Lesske J, Qian W, et al. Repetitive episodic hypoxia causes diurnal elevation of blood pressure in rats. Hypertension 1992; 19:555–561 27 Brooks D, Horner L, Kozar C, et al. Obstructive sleep apnea as a cause of systemic hypertension: evidence from a canine model. J Clin Invest 1997; 99:106 –109 28 Parker J, Brooks D, Kozar L, et al. Acute and chronic effects of airway obstruction on canine left ventricular performance. Am J Respir Crit Care Med 1999; 160:1888 –1896 29 Stradling J, Davies RJ. Sleep apnea and hypertension: what a mess! Sleep 1997; 20:789 –793 30 Hedner J, Ejnell H, Caidahl K. Left ventricular hypertrophy independent of hypertension in patients with obstructive sleep apnoea. J Hypertens 1990; 8:941–946 31 Davies RJ, Crosby J, Prothero A, et al. Ambulatory blood pressure and left ventricular hypertrophy in subjects with untreated obstructive sleep apnoea and snoring compared with matched control subjects and their response to treatment. Clin Sci 1994; 86:417– 424 32 Hanly P, Sasson Z, Zubri N, et al. Ventricular function in snorers and patients with obstructive sleep apnea. Chest 1992; 102:100 –105 33 Noda A, Okada T, Yasuma F, et al. Cardiac hypertrophy in obstructive sleep apnea syndrome. Chest 1995; 107:1538 – 1544 34 Noda A, Okada T, Yasuma F, et al. Prognosis of the middleaged and aged patients with obstructive sleep apnea syndrome. Psychiatr Clin Neurosci 1998; 52:79 – 85 35 Alchanatis M, Paradellis G, Pini H, et al. Left ventricular function in patients with obstructive sleep apnoea syndrome before and after treatment with nasal continuous positive airway pressure. Respiration 2000; 67:367–371 36 Fletcher E, Miller J, Schaaf J, et al. Urinary catecholamines before and after tracheostomy in obstructive sleep apnoea. Sleep Res 1985; 14:154 37 Carlson J, Hedner J, Elam M, et al. Augmented resting sympathetic activity in awake patients with obstructive sleep apnea. Chest 1993; 103:1763–1773 38 Sullivan C, Zamel N, Kozar L, et al. Regulation of airway smooth muscle tone in sleeping dogs. Am Rev Respir Dis 1979; 119:87–99 39 Berthon-Jones M, Sullivan C. Ventilatory and arousal responses to hypoxia in sleeping humans. Am Rev Respir Dis 1982; 125:632– 639 40 Leuenberger U, Jacob E, Sweer L, et al. Surges of muscle sympathetic nerve activity during obstructive apnea are linked to hypoxemia. J Appl Physiol 1995; 79:581–588 41 Schulz R, Schmidt D, Blum A, et al. Decreased plasma levels of nitric oxide derivatives in obstructive sleep apnoea: response to CPAP therapy. Thorax 2000; 55:1046 –1051 42 Ip MS, Lam B, Chan LY, et al. Circulating nitric oxide is suppressed in obstructive sleep apnea and is reversed by nasal continuous positive airway pressure. Am J Respir Crit Care Med 2000; 162:2166 –2171 43 Kraiczi H, Caidahl K, Samuelsson A, et al. Impairment of vascular endothelial function and left ventricular filling: association with the severity of apnea-induced hypoxemia during sleep. Chest 2001; 119:1085–1091 44 Buda AJ, Pinsky MR, Ingels NB, et al. Effect of intrathoracic pressure on left ventricular performance. N Engl J Med 1979; 301:453– 459 45 Virolainen J, Ventila M, Turto H, et al. Effect of negative intrathoracic pressure on left ventricular pressure dynamics and relaxation. J Appl Physiol 1995; 79:455– 460 46 Hall MJ, Ando SI, Floras JS, et al. Magnitude and time course of hemodynamic responses to Mueller maneuvers in patients Clinical Investigations
47 48 49 50 51
with congestive heart failure. J Appl Physiol 1998; 85:1476 – 1484 Robotham JL, Rabson J, Permutt S, et al. Left ventricular hemodynamics during respiration. J Appl Physiol 1979; 47: 1295–1303 Scharf SM, Graver LM, Balaban K. Cardiovascular effects of periodic occlusions of the upper airways in dogs. Am Rev Respir Dis 1992; 146:321–329 Cargill RI, Kiely DG, Lipworth BJ. Adverse effects of hypoxaemia on diastolic filling in humans. Clin Sci 1995; 89:165–169 Somers VK, Mark AL, Zavala DC, et al. Contrasting effects of hypoxia and hypercapnia on ventilation and sympathetic activity in humans. J Appl Physiol 1989; 67:2101–2106 Jenkinson C, Davies RJ, Mullins R, et al. Comparison of therapeutic and subtherapeutic nasal continuous positive airway pressure for obstructive sleep apnoea: a randomised
prospective parallel trial. Lancet 1999; 353:2100 –2105 52 Malone S, Liu PP, Holloway R, et al. Obstructive sleep apnoea in patients with dilated cardiomyopathy: effects of continuous positive airway pressure. Lancet 1991; 338:1480 – 1484 53 Hedner J, Darpo B, Ejnell H, et al. Reduction in sympathetic activity after long-term CPAP treatment in sleep apnoea: cardiovascular complications. Eur Respir J 1995; 8:222–229 54 Naughton MT, Bradley TD. Sleep apnea in congestive heart failure. Clin Chest Med 1998; 19:99 –113 55 Javaheri S. Effects of continuous positive airway pressure on sleep apnea and ventricular irritability in patients with heart failure. Circulation 2000; 101:392–397 56 Sin DD, Logan AG, Fitzgerald FS, et al. Effects of continuous positive airway pressure on cardiovascular outcomes in heart failure with and without Cheyne-Stokes respiration. Circulation 2000; 102:61– 66
CHEST / 121 / 2 / FEBRUARY, 2002
429
Introduction: Hypertension is common in patients with obstructive sleep apnea (OSA). However, the effect of OSA on ventricular function, especially diastolic function, is not clear. Therefore, we have assessed the prevalence of diastolic dysfunction in patients with OSA and the relationship between diastolic parameters and severity of OSA. Methods: Sixty-eight consecutive patients with OSA confirmed by polysomnography underwent echocardiography. Diastolic function of the left ventricle was determined by transmitral valve pulse-wave Doppler echocardiography. Various baseline characteristics, severity of OSA, and echocardiographic parameters were compared between patients with and without diastolic dysfunction. Results: There were 61 male and 7 female patients with a mean age of 48.1 ⴞ 11.1 years, body mass index of 28.5 ⴞ 4.3 kg/m2, and apnea/hypopnea index (AHI) of 44.3 ⴞ 23.2/h (mean ⴞ SD). An abnormal relaxation pattern (ARP) in diastole was noted in 25 patients (36.8%). Older age (52.7 ⴞ 8.9 years vs 45.1 ⴞ 11.3 years, p ⴝ 0.005), hypertension (56% vs 20%, p ⴝ 0.002), and a lower minimum pulse oximetric saturation (SpO2) during sleep (70.5 ⴞ 17.9% vs 78.8 ⴞ 12.9%, respectively; p ⴝ 0.049) were more common in patients with ARP. By multivariate analysis, minimum SpO2 < 70% was an independent predictor of ARP (odds ratio, 4.34; 95% confidence interval, 1.23 to 15.25; p ⴝ 0.02) irrespective of age and hypertension. Patients with AHI > 40/h had significantly longer isovolumic relaxation times than those with AHI < 40/h (106 ⴞ 19 ms vs 93 ⴞ 17 ms, respectively; p ⴝ 0.005). Conclusion: Diastolic dysfunction with ARP was common in patients with OSA. More severe sleep apnea was associated with a higher degree of left ventricular diastolic dysfunction in this study.
(CHEST 2002; 121:422– 429) Key words: diastolic dysfunction; echocardiography; obstructive sleep apnea Abbreviations: A ⫽ late peak atrial systolic velocity; AHI ⫽ apnea/hypopnea index; ARd ⫽ duration of pulmonary-atrial reversal signal; ARP ⫽ abnormal relaxation pattern; BMI ⫽ body mass index; CHF ⫽ congestive heart failure; CI ⫽ confidence interval; CPAP ⫽ continuous positive airway pressure; CSR-CSA ⫽ Cheyne-Stokes respiration with central sleep apnea; D ⫽ peak diastolic flow velocity in the pulmonary vein; DHF ⫽ diastolic heart failure; DT ⫽ deceleration time; E ⫽ early peak transmitral flow velocity; E/A ratio ⫽ ratio between the early peak transmitral flow velocity and the late peak atrial systolic velocity; ESS ⫽ Epworth Sleepiness Scale; IVRT ⫽ isovolumic relaxation time; LV ⫽ left ventricular; LVEF ⫽ left ventricular ejection fraction; LVH ⫽ left ventricular hypertraophy; LVM ⫽ left ventricular mass; LVMI ⫽ left ventricular mass index; OR ⫽ odds ratio; OSA ⫽ obstructive sleep apnea; S ⫽ peak systolic flow velocity in the pulmonary vein; SDB ⫽ sleep-disordered breathing; Spo2 ⫽ pulse oximetric saturation
sleep apnea (OSA) syndrome is a O bstructive common disorder affecting 2 to 4% of middle*From the Divisions of Cardiology (Drs. Fung, Yip, and Sanderson) and Respiratory Medicine (Drs. Li, Choy, Ko, and Hui), Department of Medicine and Therapeutics, Chinese University of Hong Kong, Prince of Wales Hospital, Shatin, New Territories, Hong Kong. Supported by Chinese University of Hong Kong direct grant (ref. 2040898). Manuscript received February 26, 2001; revision accepted August 15, 2001. Correspondence to: David S. C. Hui, MBBS, FCCP, Department of Medicine and Therapeutics, Chinese University of Hong Kong, Prince of Wales Hospital, Shatin, New Territories, Hong Kong; e-mail: [email protected] 422
aged adults.1,2 Excessive daytime sleepiness is a major consequence of OSA, due to sleep fragmentation triggered by repetitive episodes of partial or complete upper-airway obstruction.3 In a retrospective study by He et al,4 patients with OSA with an apnea index ⬎ 20/h had a higher morbidity and mortality rates related to vascular events than patients with an apnea index ⬍ 20/h. There is growing evidence that patients with OSA have an increased risk of having cardiovascular complications, such as hypertension, cardiac arrhythmia,5 myocardial infarction,6 pulmonary hypertension,7 and stroke.8 Several epidemiologic studies9 –13 have Clinical Investigations
shown an independent association between sleepdisordered breathing (SDB) and hypertension after controlling for confounding factors such as age, body mass index (BMI), sex, alcohol, and smoking. A recent case control study14 has also shown that patients with OSA have increased ambulatory diastolic BP both day and night and increased systolic BP at night. In a study of 200 consecutive Hong Kong Chinese patients presenting clinically with congestive heart failure (CHF), Yip et al15 showed that diastolic heart failure (DHF) is more common than systolic heart failure, with 66% of these patients having a normal left ventricular (LV) ejection fraction (LVEF). Chan et al16 performed sleep studies on 20 Chinese patients with symptomatic DHF and found 55% of patients to have significant SDB of the obstructive type with an apnea/ hypopnea index (AHI) ⬎ 10/h. Patients with DHF and SDB may be associated with worse diastolic dysfunction than those without SDB. Nevertheless, the relationship between diastolic dysfunction and OSA is not clear. This study investigates the prevalence of diastolic dysfunction in patients with newly diagnosed OSA and assesses whether there is any correlation between the severity of OSA and the degree of diastolic dysfunction. Materials and Methods Patients We recruited 68 consecutive patients with newly diagnosed OSA from our sleep unit at the Prince of Wales Hospital for this study. Sleep Study Significant OSA was defined as an AHI ⱖ 5 events per hour of sleep as shown by overnight polysomnography (Alice 4; Healthdyne; Atlanta, GA) plus self-reported sleepiness. Subjective sleepiness was evaluated by the Epworth sleepiness scale (ESS),17,18 a questionnaire specific to symptoms of daytime sleepiness, and the subjects were asked to score the likelihood of falling asleep in eight different situations with different levels of stimulation, adding up to a total score of 0 to 24. Overnight polysomnography recorded EEG, electro-oculogram, submental electromyogram, bilateral anterior tibial electromyogram, ECG, chest and abdominal wall movement by inductance plethysmography, airflow by a nasal pressure transducer (PTAF; Pro-Tech; Woodinville, WA) and backed up by oronasal airflow measured with a thermistor and finger pulse oximetry as in our previous study.19 Sleep stages were scored according to standard criteria by Rechtshaffen and Kales.20 Apnea was defined as cessation of airflow for ⬎ 10 s, and hypopnea was defined as a reduction of airflow ⱖ 50% for 10 s plus an oxygen desaturation of ⬎ 4% or an arousal. Hypertension was defined if patients were receiving antihypertensive medications without regard to the actual measurement of BP, or having a systolic BP ⬎ 140 mm Hg or a diastolic BP ⬎ 90 mm Hg9,11 on awakening following completion of polysomnography. Patients received their usual cardiac medications, includ-
ing antihypertensive agents. Exclusion criteria included malignant hypertension, unstable angina, renal failure, recent myocardial infarction, and significant valvular heart disease. Echocardiography Following confirmation of significant OSA by polysomnography, echocardiography (System FIVE; GE Vingmed; Horten, Norway) was performed to assess the LV size with systolic function assessed by fractional shortening obtained from Mmode recordings of LV systolic and diastolic dimensions,21 as well as by the LVEF estimated from the two-dimensional Simpson’s method.22 An LVEF ⬎ 50% was regarded as normal. LV diastolic dysfunction was assessed principally by Doppler echocardiography for patients with significant OSA, and normal LV systolic function was assessed as previously described.23 The pulse-wave Doppler echocardiographic sample volume was placed at the mitral valve leaflet in the apical four-chamber view. The diastolic parameters were measured from at least three beats and were defined as follows: E-wave, early maximal transmitral flow velocity; A-wave, peak velocity during atrial contraction in late diastole; and ratio between the early peak transmitral flow velocity (E) and late peak atrial systolic velocity (A) [E/A ratio], expressed in terms of peak velocities; and deceleration time (DT), calculated and expressed as the time for the peak filling velocity (E-wave) to fall to zero. Pulse-wave Doppler echocardiographic sample volume was then placed at the inflow area to measure the isovolumic relaxation time (IVRT), the time from aortic valve closure to the onset of mitral valve inflow. The pulse-wave Doppler echocardiographic sample volume was then placed inside the pulmonary vein in the apical four-chamber view. The pulmonary venous flow parameters were defined as follows: S-wave, peak systolic flow velocity in the pulmonary vein (S); D-wave, peak diastolic flow velocity in the pulmonary vein (D); and duration of pulmonary-atrial reversal signal (ARd). Diastolic function of the left ventricle was divided into four patterns: normal, abnormal relaxation pattern (ARP), pseudonormal pattern, and restrictive filling pattern depending on the abovementioned transmitral and pulmonary venous parameters and IVRT. A schematic drawing for the various diastolic dysfunction patterns is shown in Figure 1. Left ventricular mass (LVM) was estimated by M-mode echocardiography. LVM index (LVMI, grams per meters squared) was calculated from LVM corrected by body surface area. Statistical Analysis All data were presented as mean ⫾ SD unless otherwise stated. Comparisons between patients with and without diastolic dysfunction in relation to age, sex, cardiovascular risk factors, echocardiographic parameters, and severity of OSA were performed by 2 and unpaired t tests. For qualitative data, Fisher’s Exact Test was used when expected p value was ⬍ 5. A p value of ⬍ 0.05 was used to indicate differences between the groups that were statistically significant. Data analysis was performed with a commercially available statistical analysis software package (SPSS 10.0 for Windows; SPSS; Chicago, IL).
Results There were 61 male and 7 female consecutive patients with significant OSA. Mean age was 48.1 ⫾ 11.1 years; mean BMI was 28.5 ⫾ 4.3 kg/m2. Thirty-two patients (47%) had hypertension, but CHEST / 121 / 2 / FEBRUARY, 2002
423
Figure 1. Schematic drawing illustrating the four transmitral and pulmonary venous flow patterns for LV diastolic function. The upper tracing in each panel illustrates transmitral flow, while lower tracing illustrates pulmonary flow. Left upper panel: normal transmitral flow pattern: E/A ratio ⬎ 1; DT, 180 to 240 ms; and IVRT, 70 to 110 ms. Normal pulmonary flow pattern: S/D ratio ⬎ 1; pulmonary atrial reversal (AR) velocity ⬍ 25 cm/s; and duration (d) of mitral A-wave/ARd ⬎ 1. Right upper panel: abnormal relaxation transmitral flow pattern: E/A ratio ⬍ 1; DT ⬎ 240 ms; and IVRT ⬎ 110 ms. Abnormal relaxation pulmonary flow pattern: S/D ratio ⬎ 1, AR velocity ⬍ 25 cm/s, and duration of mitral A-wave/ARd ⬎ 1. Left lower panel: pseudonormalized transmitral flow pattern: E/A ratio ⬍ 1; DT, 180 to 240 ms; and IVRT, 70 to 110 ms. Pseudonormalized pulmonary flow pattern: S/D ratio ⬍ 1; AR velocity ⬍ 25 cm/s; and duration of mitral A-wave/ARd ⬍ 1. Right lower panel: restrictive transmitral flow pattern: E/A ratio ⬎ 1; DT ⬍ 180 ms; and IVRT ⬍ 70 ms. Restrictive pulmonary flow pattern: S/D ratio ⬍ 1; AR velocity ⬍ 25 cm/s; and duration of mitral A-wave/ARd⬍ 1.
none had a clinical history of heart failure. Mean ESS score was 12.9 ⫾ 6.0. From the polysomnography, the total sleep time was 6.6 ⫾ 1.2 h, arousal index was 24.2 ⫾ 12.5/h, AHI was 44.4 ⫾ 23.2/h, minimum pulse oximetric saturation (Spo2) was 75.8 ⫾ 15.3%, and mean Spo2 was 89.2 ⫾ 3.9%. From the echocardiography assessment, ARP was the only diastolic dysfunction pattern detected in our OSA patient cohort and was noted in 25 patients (36.8%; 1 female patient). Representative Doppler echocardiographic flow signals of the normal vs ARP patterns are shown in Figure 2. Baseline characteristics of patients with normal diastolic function vs those with ARP diastolic dysfunction are shown in Table 1. OSA patients with ARP diastolic dysfunc424
tion were older and had a higher prevalence of hypertension than those with normal diastolic function. Polysomnographic and echocardiographic data in patients with normal diastolic vs those with ARP diastolic dysfunction are shown in Table 2. Diastolic dysfunction with ARP was found to be related only to the minimum Spo2. AHI and mean Spo2 did not correlate with ARP diastolic dysfunction, while there was no significant difference between the two groups in other polysomnographic data. Both groups had comparable and normal LVEF. By multivariate analysis, minimum Spo2 ⬍ 70% during sleep (odds ratio [OR], 4.34; 95% confidence interval [CI], 1.23 to 15.25; p ⫽ 0.02), hypertension (OR, 4.05; 95% CI, Clinical Investigations
Figure 2. Top: Doppler echocardiographic tracings illustrating normal LV diastolic function (left, transmitral pulse-wave tracing; right, pulmonary venous pulse-wave tracing). Bottom: Doppler echocardiographic ARP of LV diastolic dysfunction (left, transmitral pulse-wave tracing; right, pulmonary venous pulse-wave tracing). V ⫽ position of transducer.
Table 1—Comparison of Baseline Characteristics of OSA Patients With and Without Diastolic Dysfunction of ARP*
Table 2—Comparison of Polysomnographic and Echocardiographic Data in OSA Patients With Normal and ARP Diastolic Dysfunction
Characteristics
ARP
Normal
p Value
Variables
ARP
Normal
p Value
Male/female patients, No. Age, yr Systolic BP, mm Hg† Diastolic BP, mm Hg† BMI, kg/m2 Hypertension Cerebrovascular accident Ischemic heart disease Diabetes mellitus
24/1
37/6
0.248
53.2 ⫾ 8.7 142.5 ⫾ 16.9 84.4 ⫾ 13.8 29.1 ⫾ 4.8 18 (72.0) 2 (8.0)
45.1 ⫾ 11.3 137.1 ⫾ 18.8 80.6 ⫾ 11.0 28.2 ⫾ 4.1 14 (32.6) 2 (4.7)
0.003‡ 0.305 0.288 0.448 0.002‡ 0.621
Patients, No. ESS Total sleep time, h Sleep efficiency, % Arousal index, No./h AHI Minimum Spo2, %† Mean Spo2, % LVEF, % E/A ratio DT, ms IVRT, ms
25 12.8 ⫾ 6.5 6.4 ⫾ 1.0 62.7 ⫾ 11.0 24.9 ⫾ 12.6 46.2 ⫾ 24.8 70.5 ⫾ 17.9 88.4 ⫾ 4.0 66.6 ⫾ 10.6 0.75 ⫾ 0.12 262.8 ⫾ 64.6 102.2 ⫾ 24.9
43 12.9 ⫾ 5.8 6.7 ⫾ 1.3 87.1 ⫾ 6.9 23.9 ⫾ 12.7 43.3 ⫾ 22.4 78.8 ⫾ 12.9 89.7 ⫾ 3.8 68.2 ⫾ 9.5 1.18 ⫾ 0.25 224.8 ⫾ 60.0 100.0 ⫾ 15.3
0.932 0.411 0.076 0.751 0.634 0.049‡ 0.228 0.526 ⬍ 0.001‡ 0.017‡ 0.674
6 (24.0) 7 (28.0)
5 (11.6) 4 (9.3)
0.305 0.084
*Data are presented as mean ⫾ SD or No. (%) unless otherwise indicated. †Measured on completion of sleep study. ‡Of statistical significance.
1.20 to 13.68; p ⫽ 0.02), and age ⬎ 55 years (OR, 5.17; 95% CI, 1.39 to 19.28; p ⫽ 0.01) were independent predictors of ARP diastolic dysfunction. To assess any relationship between severity of OSA and diastolic parameters, patients were classified into two groups according to the AHI. Patients with more severe OSA (AHI ⱖ 40/h) had a longer IVRT than patients with AHI ⬍ 40/h. There were 18
*Data are presented as mean ⫾ SD unless otherwise indicated. †During sleep. ‡Of statistical significance.
patients and 14 patients with hypertension in the two groups, respectively (p ⫽ 0.77). There was no significant difference in E, A, E/A ratio, and DT between these two groups, as shown in Table 3. The relationship between diastolic parameters, diastolic BP, mean BP, minimum Spo2 during sleep, and AHI vs LVMI was analyzed by the Pearson correlation. LVMI had a significant negative correCHEST / 121 / 2 / FEBRUARY, 2002
425
Table 3—Diastolic Parameters Between Two Groups of OSA Patients With AHI > 40/h vs AHI < 40/h* Parameters
AHI ⱖ 40/h (n ⫽ 39)
AHI ⬍ 40/h (n ⫽ 29)
p Value
E, m/s A, m/s E/A ratio DT, ms IVRT, ms
0.68 ⫾ 0.12 0.70 ⫾ 0.12 0.98 ⫾ 0.22 248.7 ⫾ 61.6 106.4 ⫾ 19.1
0.72 ⫾ 0.17 0.70 ⫾ 0.27 1.06 ⫾ 0.39 225.4 ⫾ 66.0 92.7 ⫾ 16.6
0.211 0.991 0.299 0.142 0.005†
*Data are presented as mean ⫾ SD. †Of statistical significance.
lation with peak E-wave velocity (r ⫽ 0.419, p ⫽ 0.001) and minimum Spo2 during sleep (r ⫽ 0.257, p ⫽ 0.040), and a positive correlation with the diastolic BP (r ⫽ 0.325, p ⫽ 0.023) and mean BP (r ⫽ 0.315, p ⫽ 0.027). LVMI had no significant correlation with either AHI (r ⫽ 0.109, p ⫽ 0.393) or mean Spo2 during sleep (r ⫽ 0.143, p ⫽ 0.269). DT was related to the minimum Spo2 during sleep (r ⫽ 0.27, p ⫽ 0.029), but peak E-wave velocity was not (r ⫽ 0.14, p ⫽ 0.29). Discussion Diastolic dysfunction is a condition with increased resistance to filling of the left ventricle, leading to an inappropriate rise in the diastolic pressure-volume relationship and causing symptoms of pulmonary congestion during exercise. In contrast, DHF indicates that all these changes occur when the patient is at rest.24 This study with echocardiography has shown that diastolic dysfunction occurred in 36.8% of our patients with OSA. An ARP was the only diastolic dysfunction pattern observed. All these OSA patients had normal LV systolic function and no history of overt heart failure. The OSA patients with diastolic dysfunction were older with a higher prevalence of hypertension, and diastolic dysfunction was associated with more severe minimum Spo2 during sleep, irrespective of age or hypertension. In patients with more severe OSA, as reflected by AHI ⱖ 40/h, the diastolic parameter, IVRT, was significantly longer than those with AHI ⬍ 40/h. IVRT is the time between the closure of the aortic valve and opening of the mitral valve, and reflects the compliance of the left ventricle independent of the effect of age.25 The potential mechanisms leading to changes in cardiac structure and function in patients with OSA have been studied in animal models. Fletcher et al26 demonstrated ventricular hypertrophy in rats exposed to short bursts of repetitive hypoxia over an extended period and that intermittent severe hypoxia can lead to a sustained rise in BP within 35 days. 426
Brooks et al,27 by inducing OSA via tracheostomy in the canine model, showed that OSA can lead to the development of sustained hypertension over approximately 100 days. Parker et al,28 using the same canine model, showed that in patients with chronic OSA, acute airway occlusion during sleep is associated with increase in LV afterload and decrease in fractional shortening, whereas chronic OSA also leads to sustained decrease in LV systolic performance that can be caused by the development of systemic hypertension and/or transient increase in LV afterload during episodes of airway obstruction. Nevertheless, the observation that severe OSA, induced in the canine model over 3 months, leading to LV systolic dysfunction, may not be the same in the human model, as the clinical syndrome severity in humans is highly variable and typically evolves over many years.29 There are conflicting data on the effect of OSA on the cardiac structure and function in human subjects. Increased LV wall thickness independent of daytime BP has been observed in OSA patients compared to age-matched and BMI-matched control subjects, suggesting a direct effect of nocturnal hypertension.30 A study by Davies et al,31 investigating a small group of OSA patients, snorers, and control subjects matched for age, sex, obesity, smoking, and alcohol consumption, found no difference between groups in LV diameter or thickness, or LVM. Hanly et al32 reported normal indexes of LV function between a group of typical OSA patients and a group of habitual snorers without OSA. Noda et al33 reported LV hypertrophy (LVH) in 41% of 51 OSA patients. Nocturnal hypoxia and apnea index were significantly correlated with LVH and 24-h BP. However, obesity was also commonly seen in the group with LVH, and the relation to preexisting hypertension was not clear in this uncontrolled study. However, in a separate study, Noda et al34 showed that the survival rate was significantly lower among untreated middle-aged Japanese hypertensive patients with OSA than normotensive patients (57.9% vs 90.4%, respectively). Recently Alchanatis et al35 reported that LV diastolic function was impaired in 15 OSA patients with neither history of nor current systemic hypertension, compared to 11 subjects matched for age and BMI. In addition, treatment with nasal continuous positive airway pressure (CPAP) over 3 months resulted in significant improvement in both LV diastolic function and diastolic BP, while systolic function and wall thickness remained within normal limits. In our study, LVMI correlated negatively with the minimum Spo2 during sleep and positively with the mean BP and the diastolic BP, suggesting that both hypoxia related to OSA per se and hypertension may cause LVH and Clinical Investigations
impair the diastolic function of the left ventricle. Nevertheless, OSA may have other unfavorable effects on LV diastolic function independent of LVH. There is strong epidemiologic evidence showing an independent association between OSA and hypertension.9 –13 High levels of AHI9 –13 and sleep time below 90% oxygen saturation11,12 were associated with greater odds of hypertension in a dose-response fashion. LV afterload is increased by peripheral vasoconstriction as a result of recurrent arousals terminating the obstructive respiratory events and activating the sympathetic nervous system,36 – 40 and activation of the arterial chemoreceptors by hypoxia and hypercapnia. Plasma levels of nitric oxide, a powerful vasodilator released from the endothelium, have been shown to be decreased in OSA patients but can be promptly reversed by nasal CPAP treatment.41,42 More recently Kraiczi et al43 showed in 20 subjects with OSA that worsening nocturnal hypoxemia (measured as minimum Spo2 or percentage of sleep time with Spo2 below 90%) was associated with a gradual deterioration of LV diastolic function (increased interventricular septum thickness, prolonged IVRT, and decreased E/A ratio), as well as reduced endothelium-dependent dilatory capacity of the brachial artery. Other mechanisms that may result in LV dysfunction include increased preload by intermittent negative intrathoracic pressure during apnea, which may also increase the LV transmural pressure gradient and impair diastolic relaxation and LV filling.44 – 46 Increase in right ventricular volume, together with hypoxia-induced pulmonary hypertension, may displace the interventricular septum leftward during diastole and impair LV filling.47,48 Hypoxia and hypercapnia may also decrease myocardial contractility.49,50 DHF is more common than systolic heart failure in our Chinese population.15 SDB of the obstructive type was noted in 55% of our patients with DHF16; a lower minimum Spo2 during sleep, but not AHI, was associated with more severe diastolic dysfunction.16,43 In this study, asymptomatic diastolic dysfunction was prevalent among our OSA patients; more severe OSA, as reflected by the minimum Spo2 and AHI ⱖ 40/h, was associated with worse diastolic parameters. It is possible that OSA, through hypoxia, hypertension, and other mechanisms discussed earlier, causes LV diastolic dysfunction that, in the long run, may lead to symptomatic DHF. It is important to keep a high index of suspicion for OSA when assessing patients with CHF. Not only can nasal CPAP effectively relieve disabling symptoms such as sleepiness,51 it may potentially improve LV systolic52 and diastolic function,35 decrease activation of the sympathetic nervous system activity,53
and increase nitric oxide levels41,42 in patients with both OSA and heart failure. In addition OSA and Cheyne-Stokes respiration with central sleep apnea (CSR-CSA) can coexist in patients with CHF; however, in contrast to OSA, CSR-CSA is likely a consequence rather than a cause of CHF.54 CSRCSA is associated with increased mortality in CHF, probably because of sympathetic nervous system activation caused by recurrent apnea-induced hypoxia and arousals from sleep.54 Nasal CPAP can reduce ventricular irritability,55 improve LVEF, and reduce the combined mortality-cardiac transplantation rate in such patients.56 There were several limitations in our study. This study was uncontrolled, and we did not perform 24-h BP recording and therefore could not demonstrate any diurnal pattern of changes in BP profile in our patients. Echocardiography is a well-known operator-dependent investigation, but all our studies were performed by the same experienced technician to avoid any individual variation in the assessment. Clear recordings of diastolic parameters were obtained in all our OSA patients via the transthoracic approach. Had there been poor transthoracic echocardiographic images in patients with severe obesity, the transesophageal approach would have been required to assess systolic and diastolic functions. A longitudinal study with serial echocardiography will be of great interest to assess whether long-term nasal CPAP treatment can reverse the diastolic dysfunction in a larger sample size of OSA patients. In summary, this study has shown that an ARP in diastole on echocardiography is common in patients with OSA, and patients with more severe OSA have a higher degree of diastolic dysfunction of the left ventricle. ACKNOWLEDGMENT: The authors thank Ms. Mabel Tong, M.Y. Leung, and Fanny Chan for their technical support with the sleep studies, and Ms. Doris Chan, Dr. Anthony T. Chan. and Mr. K.K. Wong for performing statistical analysis of the data. We are most grateful to Miss Pearl Ho for performing all of the echocardiography studies for this project.
References 1 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 –1235 2 Ip MS, Lam B, Lauder I, et al. A community study of sleep-disordered breathing in middle-aged Chinese men in Hong Kong. Chest 2001; 119:62– 69 3 McNamara SG, Grunstein RR, Sullivan CE. Obstructive sleep apnoea. Thorax 1993; 48:754 –764 4 He J, Kryger M, Zorick F, et al. Mortality and apnea index in obstructive sleep apnea: experience in 385 male patients. Chest 1988; 94:9 –14 5 Shepard JW Jr. Hypertension, cardiac arrhythmias, myocardial infarction and stroke in relation to obstructive sleep apnoea. Clin Chest Med 1992; 13:437– 458 CHEST / 121 / 2 / FEBRUARY, 2002
427
6 Hung J, Whitford E, Parsons R, et al. Association of sleep apnoea with myocardial infarction in men. Lancet 1990; 336:261–264 7 Bady E, Achkar A, Pascal S, et al. Pulmonary arterial hypertension in patients with sleep apnoea syndrome. Thorax 2000; 55:934 –939 8 Wessendorf TE, Teschler H, Wang YM, et al. Sleep-disordered breathing among patients with first-ever stroke. J Neurol 2000; 247:41– 47 9 Young T, Peppard P, Palta M, et al. Population-based study of sleep-disordered breathing as a risk factor for hypertension. Arch Intern Med 1997; 157:1746 –1752 10 Grote L, Ploch T, Heitmann J, et al. Sleep-related breathing disorder is an independent risk factor for systemic hypertension. Am J Respir Crit Care Med 1999; 160:1875–1882 11 Lavie P, Herer P, Hoffstein V. Obstructive sleep apnoea as a risk factor for hypertension: population study. BMJ 2000; 320:479 – 482 12 Nieto FJ, Young TB, Lind BK, et al. Association of sleepdisordered breathing, sleep apnea and hypertension in a large community-based study. JAMA 2000; 283:1829 –1836 13 Peppard P, Young T, Palta M, et al. Prospective study of the association between sleep-disordered breathing and hypertension. N Engl J Med 2000; 342:1378 –1384 14 Davies CW, Crosby JH, Mullins RL, et al. Case-control study of 24 hour ambulatory blood pressure in patients with obstructive sleep apnoea and normal matched control subjects. Thorax 2000; 55:736 –740 15 Yip G, Ho P, Woo KS, et al. Comparison of frequencies of left ventricular systolic and diastolic heart failure in Chinese living in Hong Kong. Am J Cardiol 1999; 84:563–567 16 Chan J, Sanderson J, Chan W, et al. Prevalence of sleepdisordered breathing in diastolic heart failure. Chest 1997; 111:1488 –1493 17 Johns MW. A new method for measuring daytime sleepiness: the Epworth sleepiness scale. Sleep 1991; 14:540 –545 18 Johns MW. Sensitivity and specificity of the multiple sleep latency test (MSLT), the maintenance of wakefulness test and the Epworth sleepiness scale: failure of the MSLT as a gold standard. J Sleep Res 2000; 9:5–11 19 Hui DS, Chan JK, Choy DK, et al. Effects of augmented continuous positive airway pressure education and support on compliance and outcome in a Chinese population. Chest 2000; 117:1410 –1416 20 Rechtschaffen A, Kales A. A manual of standardized terminology, techniques and scoring system for sleep stages of human subjects. Los Angeles, CA: Brain Information Institute, 1968; 1–12 21 Sahn DJ, DeMaria A, Kisslo J, et al. Recommendations regarding quantitation in M-mode echocardiography: results of a survey of echocardiographic measurements. Circulation 1978; 58:1072–1083 22 Folland ED, Parisi AF, Moynihan PF, et al. Assessment of left ventricular ejection fraction and volumes by real-time, two-dimensional echocardiography: a comparison of cineangiographic and radionuclide techniques. Circulation 1979; 60:760 –766 23 Cohen GJ, Pietrolungo JF, Thomas JD, et al. A practical guide to assessment of ventricular diastolic function using Doppler echocardiography. J Am Coll Cardiol 1996; 27:1753– 1760 24 Brutsaert DL. Diagnosing primary diastolic heart failure. Eur Heart J 2000; 21:94 –96 25 Yu CM, Sanderson JE. Right and left ventricular diastolic function in heart failure and normal-effect of age, gender, heart rate and respiration on Doppler-derived measurements. Am Heart J 1997; 134:426 – 434 428
26 Fletcher E, Lesske J, Qian W, et al. Repetitive episodic hypoxia causes diurnal elevation of blood pressure in rats. Hypertension 1992; 19:555–561 27 Brooks D, Horner L, Kozar C, et al. Obstructive sleep apnea as a cause of systemic hypertension: evidence from a canine model. J Clin Invest 1997; 99:106 –109 28 Parker J, Brooks D, Kozar L, et al. Acute and chronic effects of airway obstruction on canine left ventricular performance. Am J Respir Crit Care Med 1999; 160:1888 –1896 29 Stradling J, Davies RJ. Sleep apnea and hypertension: what a mess! Sleep 1997; 20:789 –793 30 Hedner J, Ejnell H, Caidahl K. Left ventricular hypertrophy independent of hypertension in patients with obstructive sleep apnoea. J Hypertens 1990; 8:941–946 31 Davies RJ, Crosby J, Prothero A, et al. Ambulatory blood pressure and left ventricular hypertrophy in subjects with untreated obstructive sleep apnoea and snoring compared with matched control subjects and their response to treatment. Clin Sci 1994; 86:417– 424 32 Hanly P, Sasson Z, Zubri N, et al. Ventricular function in snorers and patients with obstructive sleep apnea. Chest 1992; 102:100 –105 33 Noda A, Okada T, Yasuma F, et al. Cardiac hypertrophy in obstructive sleep apnea syndrome. Chest 1995; 107:1538 – 1544 34 Noda A, Okada T, Yasuma F, et al. Prognosis of the middleaged and aged patients with obstructive sleep apnea syndrome. Psychiatr Clin Neurosci 1998; 52:79 – 85 35 Alchanatis M, Paradellis G, Pini H, et al. Left ventricular function in patients with obstructive sleep apnoea syndrome before and after treatment with nasal continuous positive airway pressure. Respiration 2000; 67:367–371 36 Fletcher E, Miller J, Schaaf J, et al. Urinary catecholamines before and after tracheostomy in obstructive sleep apnoea. Sleep Res 1985; 14:154 37 Carlson J, Hedner J, Elam M, et al. Augmented resting sympathetic activity in awake patients with obstructive sleep apnea. Chest 1993; 103:1763–1773 38 Sullivan C, Zamel N, Kozar L, et al. Regulation of airway smooth muscle tone in sleeping dogs. Am Rev Respir Dis 1979; 119:87–99 39 Berthon-Jones M, Sullivan C. Ventilatory and arousal responses to hypoxia in sleeping humans. Am Rev Respir Dis 1982; 125:632– 639 40 Leuenberger U, Jacob E, Sweer L, et al. Surges of muscle sympathetic nerve activity during obstructive apnea are linked to hypoxemia. J Appl Physiol 1995; 79:581–588 41 Schulz R, Schmidt D, Blum A, et al. Decreased plasma levels of nitric oxide derivatives in obstructive sleep apnoea: response to CPAP therapy. Thorax 2000; 55:1046 –1051 42 Ip MS, Lam B, Chan LY, et al. Circulating nitric oxide is suppressed in obstructive sleep apnea and is reversed by nasal continuous positive airway pressure. Am J Respir Crit Care Med 2000; 162:2166 –2171 43 Kraiczi H, Caidahl K, Samuelsson A, et al. Impairment of vascular endothelial function and left ventricular filling: association with the severity of apnea-induced hypoxemia during sleep. Chest 2001; 119:1085–1091 44 Buda AJ, Pinsky MR, Ingels NB, et al. Effect of intrathoracic pressure on left ventricular performance. N Engl J Med 1979; 301:453– 459 45 Virolainen J, Ventila M, Turto H, et al. Effect of negative intrathoracic pressure on left ventricular pressure dynamics and relaxation. J Appl Physiol 1995; 79:455– 460 46 Hall MJ, Ando SI, Floras JS, et al. Magnitude and time course of hemodynamic responses to Mueller maneuvers in patients Clinical Investigations
47 48 49 50 51
with congestive heart failure. J Appl Physiol 1998; 85:1476 – 1484 Robotham JL, Rabson J, Permutt S, et al. Left ventricular hemodynamics during respiration. J Appl Physiol 1979; 47: 1295–1303 Scharf SM, Graver LM, Balaban K. Cardiovascular effects of periodic occlusions of the upper airways in dogs. Am Rev Respir Dis 1992; 146:321–329 Cargill RI, Kiely DG, Lipworth BJ. Adverse effects of hypoxaemia on diastolic filling in humans. Clin Sci 1995; 89:165–169 Somers VK, Mark AL, Zavala DC, et al. Contrasting effects of hypoxia and hypercapnia on ventilation and sympathetic activity in humans. J Appl Physiol 1989; 67:2101–2106 Jenkinson C, Davies RJ, Mullins R, et al. Comparison of therapeutic and subtherapeutic nasal continuous positive airway pressure for obstructive sleep apnoea: a randomised
prospective parallel trial. Lancet 1999; 353:2100 –2105 52 Malone S, Liu PP, Holloway R, et al. Obstructive sleep apnoea in patients with dilated cardiomyopathy: effects of continuous positive airway pressure. Lancet 1991; 338:1480 – 1484 53 Hedner J, Darpo B, Ejnell H, et al. Reduction in sympathetic activity after long-term CPAP treatment in sleep apnoea: cardiovascular complications. Eur Respir J 1995; 8:222–229 54 Naughton MT, Bradley TD. Sleep apnea in congestive heart failure. Clin Chest Med 1998; 19:99 –113 55 Javaheri S. Effects of continuous positive airway pressure on sleep apnea and ventricular irritability in patients with heart failure. Circulation 2000; 101:392–397 56 Sin DD, Logan AG, Fitzgerald FS, et al. Effects of continuous positive airway pressure on cardiovascular outcomes in heart failure with and without Cheyne-Stokes respiration. Circulation 2000; 102:61– 66
CHEST / 121 / 2 / FEBRUARY, 2002
429