Long-Term Nasal Continuous Positive Airway Pressure Administration Can Normalize Hypertension in Obstructive Sleep Apnea Patients

Transcription

1 Sleep. 16(6): American Sleep Disorders Association and Sleep Research Society Long-Term Nasal Continuous Positive Airway Pressure Administration Can Normalize Hypertension in Obstructive Sleep Apnea Patients...,; *tmayumi Suzuki, tkuniaki Otsuka and *Christian Guilleminault *Stanford University Sleep Disorders Clinic and Research Center. Stanford, California, U.S.A.; and ttokyo Women's Medical College, Daini Hospital, Tokyo, Japan Summary: We investigated the way in which nasal continuous positive airway pressure (CPAP) affects the circadian profiles of blood pressure (BP) and heart rate (HR) in obstructive sleep apnea syndrome (OSAS) patients. Nine patients with OSAS, confirmed by nocturnal polysomnography, were studied with ambulatory blood pressure monitoring (Colin ABPM-630) during two 48-hour periods, before and during nasal CPAP treatment, at the Stanford University Sleep Disorders Clinic. During each 48-hour monitoring period, blood pressure measurements were taken by the ambulatory device every 30 minutes. During the ambulatory blood pressure recordings, nocturnal sleep time was defined as the period between 0000 hours to 0600 hours and active daytime was defined as the period between 0 hours and 2000 hours. An average systolic blood pressure> 135 mm Hg during the 48-hour baseline recording was defined as hypertensive. Using these criteria, we selected four hypertensive and five normotensive patients. Average BP (systolic/diastolic) and HR during the 48-hour periods decreased significantly from 148.6/88.2 mm Hg to 138.7/81.4 mm Hg, and from 77.9 beats per minute (bpm) to 67.2 bpm in hypertensives during CPAP treatment (p = 0.04), but there were no significant changes observed in normotensives. Average BP, during the day and at night, decreased from 152.3/ 91.8 mm Hg to 141.2/85.1 mm Hg and from mm Hg to mm Hg, respectively, in the hypertensives during CPAP, but such changes were not observed in normotensives. Average HR during the day and at night decreased significantly from 85.2 bpm to 72.6 bpm and from 69.8 bpm to 56.5 bpm in the hypertensives (p = 0.04), but not in normotensives. In conclusion, nasal CPAP is effective in improving circadian BP profiles and in normalizing daytime hypertension in some hypertensive OSAS patients. Key Words: Obstructive sleep apnea syndrome-nasal continuous positive airway pressure-ambulatory blood pressure monitoring. Obstructive sleep apneas (OSA) are known to be frequently accompanied by episodic elevations of systemic blood pressure (BP) during sleep (1-3). Day timesustained elevation of systemic BP is also observed in 40-90% of patients with OSAS (4-10). Many studies monitoring the circadian BP rhythm have been performed [see bibliography of reference (11)], including BP monitoring in bed-restricted subjects. However, few studies have monitored the circadian BP profile offreely active OSAS patients. We questioned whether the nocturnal decrease in BP observed in normal subjects (12) was also seen in OSAS patients. We also questioned what effect nasal continuous positive airway pressure (nasal CPAP), a recognized treatment of OSAS, might have on the circadian BP rhythm and on Accepted for publication May Address correspondence and reprint requests to Christian Guilleminault, M.D., Stanford Sleep Research Center, 701 Welch Road, Suite #2226, Palo Alto, CA 94304, U.S.A. 545 any existing hypertension. To explore these questions we used an ambulatory BP monitoring device (ABPM- 630, Colin, San Antonio, TX) (13,14). Entry criteria POPULATION All patients included were diagnosed with OSAS based upon clinical complaints, clinical symptoms and nocturnal polygraphic monitoring. The following variables were recorded: electroencephalogram (EEG) leads C3/A2 and C4/Al of the international placement system, electrooculogram, chin and leg electromyogram, and electrocardiogram (ECG) (modified V-2 lead). Respiration was monitored by uncalibrated inductive plethysmography, naso-oral airflow and pulse oximetry (Nelcor ). Informed consent was obtained from all subjects for participation in the study protocol, which had been

2 546 M. SUZUKI ET AL. approved by an institutional review board. Subjects using hypertensive medication(s) agreed to stay on the same drug at the current dosage during the duration of the study and to use their nasal CPAP equipment for at least 6 days a week and for a minimum of 6 hours per night. (A counter on the machine verifi(~d the number of hours the equipment had been turned on, although it could not verify whether the mask had been placed appropriately.) Finally, all subjects were required to return to the clinic within a prearrang(~d follow-up time. Criteria for data rejection Subjects who did not keep a regular day/night cycle, who did not tum on their equipment for at least 5 days per week, who increased or decreased weight signhicantly during the nasal CPAP treatment or who modified their hypertensive treatment during the study were discontinued. These criteria led to a large number of subject rejections at the time offollow-up monitoring. Of the 37 subjects who underwent baseline recordings, only nine were included for analysis at follow-up. Blood pressure monitoring METHODS Blood pressure and heart rate profiles were obtained for up to 72 hours, using an ambulatory BP monitoring system, the ABPM-630 (Colin, San Antonio, TX), which has been used in several studies on normal subjects. The Colin Medical Instruments ABPM-630 used in this study provides oscillometric and auscultatory measurements (13,14). The presence of both measurements allows for rejection of abnormal values. The device is lightweight (830 g) and stores data and events in solid state memory, which can be transferred to a personal computer for analysis and printed. The device recorded and stored a data-point every 30 minutes. Each patient underwent an in-depth interview evaluating cardiovascular status, cardiovascular family history, sleep/wake schedule, eating habits, drug intake and drug history. At physical evaluation, the upper arm circumference was measured to sekct appropriate cuff size. Patients were requested to perform their regular activities and to record their behavior in log books during the entire period of BP monitoring. BP monitoring was started early in the day to ensure habituation of the subject to the equipment and cuff inflation by the first night. After the baseline recording, all of the patients were titrated with nasal CPAP and followed prospectively. Every 15 days, each patient was contacted to provide Sleep, Vol. 16, No, 6, 1993 encouragement to continue following the protocol. At the clinic follow-up, patients were interviewed to assess any changes in daily activities, sleep habits and drug intake. The patients' private physicians were contacted to confirm drug intake information. Ambulatory BP monitoring was performed again after a minimum of 5 weeks of nasal CPAP treatment (mean = 3 months). DATA ANALYSES "Nocturnal sleep" was defined as the time between 0000 and 0600 hours. "Active daytime" was defined as the period between 0 and 2000 hours. When a patient's average BP during his or her 48-hour recording was> 135 mm Hg, the patient was defined as hypertensive. The average BP and HR during "nocturnal sleep" and "active daytime" before and during nasal CPAP were tabulated and compared. We used the cosine fitting method to derive the midline estimatory statistics of rhythm (MESOR), amplitude and acrophase values for systolic and diastolic blood pressure and heart rate on data obtained before and during treatment with nasal CPAP (15). Finally, we investigated whether certain subjects lacked the drop in blood pressure and heart rate observed during nocturnal sleep in normal subjects (subjects without this "dip" were called "non-dippers") and whether this abnormal pattern disappeared with nasal CPAP treatment. A subject was called a "non-dipper" ifhis or her systolic and diastolic blood pressures did not decrease during sleep by 10 and 5 mm Hg, respectively (16). We also used Verdecchia et al.'s definition of "dipping" (daytime BP - night-time BPl24-hour BP x ) (17). Stored data were analyzed using commercially available software (Colin PC-630, San Antonio, TX). Wilcoxon's signed rank test was used for BP and HR analysis using the Statview 2 statistical package for the Macintosh. A probability of p < 0.05 was considered to be statistically significant. Subjects RESULTS Eight male and one female obese OSAS patient [body mass index (BMI) 31.2 ± 5.4 kglmz, mean ± SD], aged years (mean 55.7 ± 11.7 years), completed the study. Three male subjects and the female subject were classified as normotensive. Five male subjects were hypertensive. The average ages ofnormotensives and hypertensives were 58.6 and 52 years, respectively. There was no statistical difference in BMI between the two groups. Their mean respiratory disturbance index (RDI = number of apneas and hypopneas per hour of sleep) was 39.2 ± 24.1 and 48.6 ± 39.1 events per

3 Night time (0: : 00) NASAL CPAP AND HYPERTENSION IN OSAS 547 ', '.; ~" Systolic BP Diastolic BP Heart Rate mmhg mmhg /min Normotension Hypertension 110 Normotension Hypertension I~I ~1 --- I~I I~ I I Normotension Hypertension L-NS~ LNS--.J Lp=O.04...J 60 L p=o.oaj 50 I;:::: I I~I r I I I I l' l' I I! before during before during before during before during before during before during FIG. 1. Changes in systolic and diastolic blood pressures and heart rate in individual subjects during "active daytime". Legend: Before = Before treatment with nasal CPAP; During = While being treated with nasal CPAP. hour, respectively, with a nonsignificant trend (p = 0.07) toward a higher index in the hypertensive group. Four hypertensives and three normotensives had a family history of hypertension (first degree relative). One subject in each group took hypertensive medication. Effect of nasal CP AP The average systolic/diastolic BP and HR obtained during the two 48-hour monitoring periods decreased significantly from 148.6/88.2 mm Hg to /8l.4 mm Hg and from beats per minute (bpm) in the hypertensive group during CPAP (p = 0.04 for all variables), but there were no significant changes in the normotensive group. The "active daytime" average systolic BP decreased from mm Hg to mm Hg (p = 0.08), the diastolic BP decreased from 9l.8 mm Hg to 85.1 mm Hg (p = 0.14), and HR decreased significantly from 85.2 bpm to 72.6 bpm (p = 0.04) in the hypertensive group during nasal CPAP, but there were no significant changes in the normotensive group in these parameters (Fig. 1). The "nocturnal sleep" average systolic BP decreased from mm Hg to mm Hg (p = 0.04), the diastolic BP decreased from 76.8 mm Hg to 73.7 mm Hg (p = 0.08), and the HR decreased significantly from 69.8 bpm to 56.5 bpm in the hypertensive group during nasal CPAP. Again, there were no significant changes in the normotensive group in these measurements (Fig. 2). Calculations of amplitude and acrophase for systolic and diastolic BP and HR on data collected for 48 hours before and during nasal CPAP treatment showed no statistically significant changes in any of the groups. There were, however, significant changes in systolic (p = 0.04) and diastolic (p = 0.04) blood pressure and HR (p = 0.001) MESORs with nasal CPAP in the hypertensive group, as expected from the above results. Finally, only one ofthe nine subjects (a hypertensive subject) could be classified as a "non-dipper" during baseline monitoring (daytime BP = vs at night). He evidenced clear nocturnal sleep variation with nasal CPAP treatment (daytime = 132.2/ 75.7 vs /68 at night). COMMENTS The relationship between hypertension and obstructive sleep apnea syndrome is controversial. A relationship between increases in blood pressure and single apneas has been well documented (1-3), but the 24- hour hypertension frequently observed in OSAS patients is still unexplained. Many of the subjects in these studies have had other coexisting clinical problems, with obesity being the most common, and it is difficult to affirm the role of OSAS in the development of daytime hypertension (10,18,19), except in children treated with tracheostomy (20). Our study is a prospective follow-up investigation with one intervention: treatment of obstructive sleep apnea with nasal CPAP. Great efforts were made to include only subjects in whom compliance with the nasal CP AP trial had been ascertained as well as possible and in whom none ofthe other selected variables Sleep. Vol. 16. No

4 548 M. SUZUKI ET AL. Day time (10: : 00) Systolic BP Diastolic BP Heart Rate mmhg mmhg /min Normotension Hypertension Normotension Hypertension Normotension Hypertension 1,,! 1!,! 1 before during before during before during before during before during before during FIG. 2. Changes in systolic and diastolic blood pressures and heart rate in individual subjects during nocturnal sleep known to influence blood pressure was modified during the prospective follow-up period. These pre-established criteria for inclusion led to a large number of drop-outs. Subject elimination was performed on the scheduled follow-up day prior to the second BP monitoring period according to the strict application of the protocol. The technology used in our study is well known and has been frequently tested and used at Stanford as well as in Japan (15,21). The monitoring was performed over a period oftime long enough to obtain more than one 24-hour cycle, allowing us to investigate MESOR, amplitude and acrophase. Also, prior investigations with simultaneous polygraphic monitoring have demonstrated the phenomenon of habituation to intermittent cuff inflation and absence of associated alpha EEG arousals during sleep (21). Interestingly, our normotensive and hypertensive subjects were both overweight (elevated BMI compared to U.S. population normative data). This is in contrast to a previous study reported by our group (12). There was no statistical difference in respiratory disturbance index between normotensive and hypertensive subjects. However, we obtained no clear information from any of the subjects regarding the onset of their breathing disorder during sleep. The prior duration of the syndrome is an important consideration, since it may be argued that it is this parameter that determines the development ofhy- TABLE 1. Measurements of amplitude, acrophase and J'vfESOR in normotensive and hypertensive OSAS patients before and during treatment with nasal CPAP Condition Group Variables Amplitude Acrophase MESOR Baseline SBP 14.7 ± ± ± 6.5 Normotensive DBP 9.2 ± ± ± 7.0 HR 10.7 ± ± ± 8.1 Hypertensive SBP 11.3 ± ± ± 10.7 DBP 8.0 ± ± ± 9.3 HR 10.9 ± ± ± 13.1 Nasal CPAP SBP 13.0 ± ± ± 6.9 Normotensive DBP 8.7 ± ± ± 3.9 HR 10.6 ± ± ± 13.2 Hypertensive SBP 10.0 ± ± ± 14.5 DBP 6.5 ± ± ±89.8 HR 9.8 ± ± ± 7.5 Abbreviations: SBP = systolic blood pressure in mm Hg, DBP =, diastolic blood pressure in mm Hg, HR = heart rate in beats per minute, MESOR = midline estimatory statistics of rhythm (15). Sleep. Vol. 16. No ~,

5 NASAL CPAP AND HYPERTENSION IN OSAS 549 pertension in OSAS subjects. Our patients, however, usually associated the onset of health problems with their own recognition of daytime sleepiness, a very subjective symptom often unrelated to the onset of abnormal breathing during sleep. Only one of our nine subjects did not present the normal circadian dip in BP during nocturnal sleep. Interestingly, this subject was hypertensive. Absence of this nocturnal sleep drop was reported in OSAS subjects as early as 1976 by Tilkian et al. (2), with reappearance of normal nocturnal "dipping" following tracheostomy (3) or nasal CPAP treatment (22). However, daytime hypertension may occur without a disappearance of the normal circadian nocturnal decrease. Guilleminault et al. (23) have previously emphasized that the distribution of apneas and hypopneas during the night (i.e. continuously repetitive or intermittent) and the frequency and duration of complete awakenings from sleep are at least two of the factors playing a role in the presence or absence of normal circadian nocturnal dip of blood pressure. However, pathological significance of the absence of this drop is unclear. This study demonstrates that when many other factors are controlled, appropriate treatment of upper airway collapse during sleep by mechanical means is associated with a significant improvement in daytime hypertension in adults with OSAS. This finding gives credence to epidemiologic reports that have linked hypertension to OSAS (4-8), despite contrasting views by others (9,10,18,19). Acknowledgement: This study was supported by grant #NA of the National Institute on Aging. We thank Laura Wilkinson for editing the manuscript. REFERENCES 1. Coccagna G, Mantovani M, Brignani F, Parchi C, Lugaresi E. Continuous recording of the pulmonary and systemic arterial pressure during sleep in syndromes of hypersomnia and periodic breathing. Bull Eur Physiopathol Resp 1972;8: Tilkian AG, Guilleminault C, Schroeder JS, Lehrman KL, Simmons FB, Dement We. Hemodynamics in sleep-induced apnea: studies during wakefulness and sleep. Am Int Med 1976;85: Motta J, Guilleminault C, Schroeder JS, Dement We. Tracheostomy and hemodynamic changes in sleep-induced apnea. Ann Int Med 1978;89: Lavie P, Rachamim B Y, Rubin A-H-E. Prevalence of sleep apnea syndrome among patients with essential hypertension. Am Heart J 1984;108: Kales A, Bixler EO, Cadieux RJ, et a!. Sleep apnea in a hypertensive population. Lancet 1984;ii: I Fletcher EC, DeBehnke RD, Lovoi MS, Gorin AB. Undiagnosed sleep apnea in patients with essential hypertension. Ann Int Med 1985; 103: Millman RP, Redline S, Carlisle CC, Assaf A, Levinson PD. Daytime hypertension in obstructive sleep apnea. Prevalence and contributing risk factors. Chest 1991;99: Williams AJ, Houston D, Finberg S, Lam C, Kinney JL, Santiago S. Sleep apnea syndrome and essential hypertension. Am J CardioI1985;55: Stradling JR. Sleep apnea and systemic hypertension. Thorax 1989;44: Jeong D-U, Dinsdale JE. Sleep apnea and essential hypertension: a critical review of the epidemiological evidence for co-morbidity. Clin Exper Hypertens Theory Pract 1989;AII: II. Staessen JA, Fagard RH, Lijnen PL, Thijs L, Van Hoof R, Amery AK. Mean and range of the ambulatory pressure in normotensive subjects from a met-analysis of 23 studies. Am J CardioI1991;67: Staessen J, Bulpitt CJ, O'Brien E, et al. The diurnal blood pressure profile. A population study. Am J Hypertens 1992;5: Otsuka K, Kitazumi T, Watanabe H, et al. Ambulatory blood pressure monitoring device using the auscultatory and oscillometric method. J Jap Coli AngioI1989;29: White WB, Lund-Johansen P, McCabe EJ. Clinical evaluation of the Colin ABPM-630 at rest and during exercise: an ambulatory blood pressure monitor with gas-powered cuff inflation. J Hypertens 1989;7: Halberg F, Watanabe H, eds. Workshop on computer methods on chronobiology and chronomedicine. Presented in the Medical Review (Pub!.), Tokyo, pp. 16. O'Brien E, Sheridan J, O'Malley K. Dippers and non-dippers. Lancet 1988; II :397. I 7. Verdecchia P, Schillaci G, Guerrieri M. Circadian blood pressure changes and left ventricular hypertrophy in essential hypertension. Circulation 1990;81 : Shiner RJ, Carroll N, Sawicka EH, Simonds AK, Branthwaite MA. Role of nocturnal hypoxemia in the genesis of systemic hypertension. Cardiology 1990;77: Ali NJ, Davies RJO, Fleetham JA, Stradling JR. The acute effect of continuous positive airway pressure and oxygen administration on blood pressure during obstructive apnea. Chest 1992; 101: Guilleminault C, Simmons FB, Motta J, Cumminskey J, Schroeder JS, Dement We. Obstructive sleep apnea syndrome and tracheostomy. Long-term follow-up experience. Ann Int Med 1981;141: Rubin P, Blaschke TF, Guilleminault e. Effect of naloxone-a specific opioid inhibitor-on blood pressure fall during sleep. Circulation 1980;63: Guilleminault C, Suzuki M. Sleep-related hemodynamics and hypertension with partial or complete upper airway obstruction during sleep. Sleep 1992; 15:S Guilleminault C, Tilkian AG, Dement We. The sleep apnea syndromes. Ann Rev Med 1976;27: Sleep, Vol. 16, No.6, 1993