Role of Protriptyline and Acetazolamide in the Sleep Apnea/Hypopnea Syndrome

Transcription

1 Sleep 11(5): , Raven Press, Ltd., New York 1988 Association of Professional Sleep Societies Role of Protriptyline and Acetazolamide in the Sleep Apnea/Hypopnea Syndrome K. F. Whyte, G. A. Gould, M. A. A. Airlie, C. M. Shapiro, and N. J. Douglas Rayne Laboratory, Department of Respiratory Medicine, University of Edinburgh, City Hospital, Edinburgh, Scotland Summary: The role of drug therapy in the treatment of the sleep apnea! hypopnea syndrome is unclear. In a randomised, double-blind, placebocontrolled study, we investigated the value of 14-day therapy with protriptyline (20 mg daily) or acetazolamide (250 mg 4 times per day) on symptoms and on the frequency of apneas, hypopneas, arousals, and 4% de saturations in 10 patients with obstructive sleep apnea!hypopnea syndrome. Overall, protriptyline did not have a significant effect either on symptoms or on any of the above polysomnographic criteria. Acetazolamide reduced the apnea!hypopnea frequency [placebo 50 ± 26 (SD); acetazolamide 26 ± 20/h of sleep, p < 0.03] and tended to decrease the frequency of 4% desaturations (placebo 29 ± 20; acetazolamide 19 ± 16/h of sleep, p = 0.06). Despite these physiological improvements, acetazolamide did not significantly improve symptoms and paraesthesiae were common. Contrary to earlier studies, we conclude that protriptyline may have a limited role in the treatment of the sleep apnea syndrome. The reason why acetazolamide produced a physiological, but not a symptomatic, response requires further investigation. Key Words: Protriptyline Acetazolamide-Apnea!hypopnea syndrome. The role of drug therapy in the treatment of the sleep apnealhypopnea syndrome is unclear. Many drugs have been tried, but protriptyline is probably the most wiqely used (1-3). Protriptyline is believed to act both by suppressing rapid eye movement (REM) sleep (4), during which apneas are longest (5), and by specifically increasing tone in the upper airway opening muscles (6). Previous studies (7-10) on protriptyline are difficult to relate to current clinical practice. First, most patients in these studies had relatively severe sleep apnealhypopnea syndrome and would usually now be treated by continuous positive airway pressure (CPAP) or surgery (7,9,10). Second, most of the investigations were performed with a diagnostic study followed some weeks later by a study on protriptyline and, thus, weight loss could have influenced results (8-10). Third, most studies were not blinded (8-10), thereby making symptomatic benefit difficult to assess. Accepted for publication June Address correspondence to Dr. K. F. Whyte at Department of Respiratory Medicine, Rayne Laboratory, City Hospital, Edinburgh, EHIO 5S8, Scotland. 463

2 464 K. F. WHYTE ET AL. Acetazolamide has been advocated as therapy for central sleep apneas, acting as a respiratory stimulant by indu(;ing metabolic acidosis (11,12). However, acetazolamide has not been given in a therapeutic trial in patients with obstructive sleep apnea. As we believe the pathogenesis of obstructive and central apneas may be similar (13), we have investigated the effect of acetazolamide on patients with the sleep apnea!hypopnea syndrome. Thus, in a double-blind, placebo-controlled, crossover study, we have examined the effects of 2-week therapy with protriptyline or acetazolamide on symptoms, apnea! hypopnea index, and the incidence of side effects in a group of 10 patients with mild to moderately severe obstructive sleep apneaihypopnea syndrome. METHODS We studied 10 subjects (8 men, 2 women; age range years) (Table 1) with the obstructive sleep apnea/hypopnea syndrome, whose symptoms had failed to respond to weight reduction and alcohol avoidance. For admission to the study, each patient must have had more than 15 apneas + hypopneas/h of sleep in a diagnostic sleep study, with at least two of the following supporting symptoms, daytime somnolence, loud snoring, unsatisfying nocturnal sleep, or recurrent nocturnal awakening. No patient had fallen asleep driving or during other active tasks, and none had congestive cardiac failure. All patients gave informed written consent and the protocol of the study was approved by the local ethical advisory committee. All staff members involved in patient care and data scoring were unaware of the therapy administered each study night. Using a balanced Latin-square design, we randomly allocated patients to consecutive 2-week periods of treatment with placebo (two tablets at night), protriptyline (20 mg at night), and acetazolamide (250 mg twice daily for 1 week, followed by 250 mg 4 times daily for the second week). Subjects attended the sleep laboratory on the 14th day of each treatment period, and after answering a standard visual analog questionnaire that enquired about both the symptoms of the sleep apnea!hypopnea syndrome and side effects of the various treatments, they received that.evening's dose of medication under supervision. Arterial blood gases were obtained while the patient was awake and were analysed on a Radiometer ABL-2 blood TABLE 1. Individual patient details and patient weights (as percentage of ideal body weight) on each study night on each therapy Percent of ideal body weight Patient Age Height no. Sex (years) Smoker (cm) Placebo Acetazolamide Protriptyline 1 M 61 Yes M 59 Yes M 37 Yes F 54 Yes M 46 No M 44 Yes M 50 Yes M 63 No M 67 No F 34 Yes Mean SD Sleep, Vol. 11, No.5, 1988

3 DRUG THERAPY OF SLEEP APNEA 465 gas analyser (Radiometer, Copenhagen, Denmark). On completion of the study, the patients were asked which treatment they preferred. Electroencephalogram (EEG), electrooculogram (EOG), and electromyogram (EMG) were recorded using our standard electrode placement (14). Thoracoabdominal movement was assessed by inductive plethysmography (Respitrace, Respitrace Corp., Ardsley, NY, U.S.A.), air flow at the nose and mouth by thermocouples (14), and ear oxygen saturation by Biox IIa ear oximeter (15) (Ohmeda Biox, Boulder, CO, U.S.A.). Electrocardiogram was recorded continuously. All physiological data were recorded simultaneously onto a multichannel recorder (S.L.E., Croydon, Surrey, U.K.) running at 15 mm s -1, and sleep stage was scored in 20-s epochs by standard criteria (16). Apnea was defined as a greater than lo-s cessation of air flow (17) and hypopnea as a 50% or greater reduction in thoracoabdominal movement (18). Arousal was defined as an increase in EMG tone for greater than 1.5 s associated with a or 6 EEG activity. Frequencies for apneas, hypopneas, arousals, and de saturations were scored per hour of sleep. Results are expressed as mean ± standard deviation and were analysed by Student's t test for paired observations with the Bonferroni correction for multiple comparisons (19). RESULTS All patients completed the study, but two did not tolerate protriptyline at 20 mg/day, which was reduced to 10 mg/day on days 3 and 7, respectively. Compliance Arterial blood gas tensions were obtained in 9 subjects, and in each, acetazolamide induced a significant metabolic acidosis (mean ph 7.42 ± 0.04 on placebo, 7.36 ± 0.04 on acetazolamide; p < 0.01). In all patients, heart rate was higher on the protriptyline study night both awake (84 ± 13 protriptyline, 72 ± 6 beats/min placebo, p < 0.01) and during stage II sleep (80 ± 8 protriptyline, 67 ± 6 beats/min placebo, p < 0.01). Polysomnograpby Breathing pattern. There was a significant reduction in the apnea hypopnea index following acetazolamide (50 ± 26 placebo, 26 ± 20/h of sleep acetazolamide, n = 10, p < 0.03) (Fig. 1). There was no change in apnea/hypopnea index following protriptyline (46 ± 23). There was wide interindividual variability in the change in apnea + hypopnea index with both treatments. No physical characteristics of the patients (weight, age, sex, apnealhypopnea index, or symptoms) predicted response to either drug. Mean apnea duration was unaltered by either active treatment (16 ± 2 placebo, 19 ± 8 protriptyline, 20 ± 6 s acetazolamide). Oxygenation. Arterial saturation when awake was not different on either active treatment compared to placebo. The number of 4% desaturations per hour of sleep tended to be lower on acetazolamide (19 ± 16/h of sleep) compared to placebo (29 ± 20/h of sleep, p = 0.06). Protriptyline did not significantly alter the number of desaturations (26 ± 15/h of sleep, p > 0.4) (Fig. 2). The lowest oxygen saturation (Sa02) asleep was unaltered by either treatment (70 ± 24% placebo, 72 ± 13% acetazolamide, 77 ± 14% protriptyline). Sleep, Vol. /1, No.5, 1988

4 466 K. F. WHYTE ET AL. 100 I- 80 a:: => o :r: a:: 60 i 40 z ««L.W o 20 * ** FIG. 1. Apnealhypopnea index for individual patients (n = 10) on each therapy (mean is shown as the open circle). *p < 0.03; **p > 0.4 compared to placebo. O r- PLACEBO ACETAZOLAMIDE PROTRIPTYLINE Sleep. The number of arousals per hour of sleep was unchanged on acetazolamide therapy (26 ± 26 placebo, 16 ± 10/h of sleep acetazolamide, p > 0.2) or protriptyline (21 ± Il1h, P > 0.5) (Fig. 3). Neither acetazolamide nor protriptyline significantly altered sleep quality as measured by total sleep time or sleep efficiency index (Fig. 4). There was no change in the distribution of sleep stages, either in absolute time or in the percentage of total sleep time (TST) (Fig. 4). Protriptyline did not significantly reduce REM sleep (placebo 19 ± 8%, protriptyline 18 ± 12%, acetazolamide 23 ± 8% TST). There was no significant change in the number of sleep stage shifts on any treatment (8.0 ± 1.7/h of sleep on placebo, 7.4 ± 2.6 on acetazolamide, and 8.6 ± 3.0 on protriptyline). Physiological improvements. Defining a physiological improvement as a > 35% fall in all three conditions of apnea!hypopnea index, 4% desaturations/h of sleep and arousal index, only one patient (no. 3) responded to protriptyline, and he also improved in all three indices following acetazolamide. Three other patients also had falls in the apnea! hypopnea index with protriptyline (nos. 6, 7, 8) but the de saturation or the arousal index failed to improve. Four patients had >35% improvement in apnea!hypopnea index, desaturations, and arousals (nos. 3, 6, 9, and 10) following acetazolamide. Symptoms Daytime somnolence, morning headache, and sleep disturbance, as assessed by a visual analog scale, were not significantly altered by either treatment compared to placebo for the whole group (Fig. 5). Patients felt no more rested in the morning with Sleep, Vol. 11, No.5, 1988

5 DRUG THERAPY OF SLEEP APNEA * ** FIG. 2. Number of 4% desaturations per hour of sleep in individual patients (n = 10) (mean is shown as the open circle). *p = 0.06; **p > 0.4 compared to placebo. OL r_ Placebo Acetazolamide Protriptyline either active treatment (Fig. 5). Patient 3, who had responded to both treatments in terms of apnea!hypopnea index, desaturations, and arousals, had improved sleep quality and less somnolence with protriptyline and, to a lesser extent, with acetazolamide. Of the four patients who had responded physiologically to acetazolamide, patient 3 reported some improvement in both sleep quality and somnolence; patient 6 had no change in symptoms; patient 9 reported poorer sleep quality, but some reduction in somnolence; and patient lo's relatively mild symptoms regressed completely. Side Effects Two patients could not tolerate the 20-mg protriptyline dose because of severe dry mouth and visual upset in one and urinary symptoms and a dry mouth in the other. A third patient complained of altered sexual potency and testicular discomfort, but finished the trial period. Dry mouth was the most common and most troublesome complaint with protriptyline therapy, but was also common after acetazolamide therapy. Acetazolamide produced paraesthesia in eight patients, which one subject found intolerable. On the last study night, patients were asked to express a preference for one treatment. Seven subjects expressed no preference and three subjects prefered acetazolamide (nos. 3, 4, and 8). Follow-up Following analysis of the results and based on the change in the number of apnea! hypopnea events, 4% desaturations, or on patient preference, we suggested acetazol- Sleep. Vol. 11. No

6 468 K. F. WHYTE ET AL W W -' I.L 0 0::60 => 0 :r: 0:: w 0- -' j\40 => 0 0:: <! I.L 0 0:: :::520 L => :z * * * FIG. 3. Arousal index for individual patients (n = 10) with each therapy (mean is shown as the open circle). *p > 0.2; **p > 0.5 compared to placebo. 0 Placebo Acetazolamide Protriptyline amide therapy for 3 patients, including patient 10 who had a dramatic fall in apnea! hypopnea index from 48/h to 7/h with acetazolamide and also had a decreased number of arousals and improved oxygenation. This patient was unable to tolerate acetazolamide at 1,000 mg/day because of parasthesae, and this was reduced to 500 mg/day. However, she continued to be troubled by intolerable paraesthesia during physical exertion and intercourse and her symptoms were unchanged. Acetazolamide was therefore withdrawn. One patient (no. 3) who had responded to both protriptyline and acetazolamide, but suffered intolerable side effects with protriptyline, expressed a preference for acetazolamide and was begun on therapy; however, it was withdrawn when the patient developed nocturnal enuresis. The third patient (no. 8), who noted little change in symptoms but did have a significant improvement in apnea!hypopnea events, continues to take acetazolamide 10 months later and reports some symptomatic improvement. DISCUSSION Neither CPAP (20) nor surgery (1) are universally acceptable or ideal in patients with the sleep apnea!hypopnea syndrome, particularly in those with less severe symptoms. Thus, a successful drug therapy would be desirable for these patients. Protriptyline has been the most extensively studied drug, but the applicability of the results of these studies in the treatment of patients with milder symptoms is unclear. Many of the patients studied had severe sleep apnea (7,9,10) with either life-threatening symptoms or complications and would now be preferentially treated with either CPAP or surgery. In addition, interpretation is further complicated because only one study (7) was placebo-controlled and randomised. Thus, in the other studies, a diagnostic sleep study was followed by a repeat sleep study after protriptyline, and no allowance for an Sleep. Vol. 11. No.5, 1988

7 DRUG THERAPY OF SLEEP APNEA 469 x 10 o > LJ Z LJ u: L.L ' w ;'! 0... w -' :r: w Wakefulness 2 FIG. 4. Effect of each therapy on total sleep time, sleep efficiency index, and the percentage of total sleep time spent in each sleep stage for all patients (n = 10). (D) Placebo; (1'21) acetazolamide; (_) protriptyline. There were no significant differences between active treatments in any sleep stage compared to placebo study night. 3/4 REM order effect was made (8-10). These studies (8-10) have suggested that protriptyline is capable of variably decreasing the number of apneas and improving oxygenation in some patients with sleep apnea. In addition, in the controlled study (7), four of five patients reported a reduction in daytime somnolence with protriptyline, yet there was no reduction in the number of arousals, and the number of apneas was only reduced because of the reduction in the amount of REM sleep, the number of non-rem apneas being unchanged (7). The doses used in these studies varied between 5 and 30 mg protriptyline, and side effects of varying severity were common, necessitating cessation of therapy in some individuals (8,9). In one study (9), though symptomatic improvement was maintained for 12 months, repeat sleep studies demonstrated increased apnea frequency, indicating a possible decrease in efficacy with time. Previous studies have assessed symptomatic response simply by direct questioning of patients. In this study, we have used visual analog scales to assess the symptomatic response to therapy quantitatively. Visual analog scales do have limitations; however, we attempted to minimise these by using the same staff member to explain the scales to the patients and using repeated testing for each patient. This method also has the inherent problem that it is a subjective, rather than objective, measurement of somnolence, but we were unable to preform multiple sleep latency testing (21) in this study. However, the subjective impression is important, and we doubt whether patients will comply well with a medication they feel is not helping their symptoms. In this placebo-controlled, randomised study, we were unable to demonstrate that protriptyline caused a significant reduction in apnealhypopnea index or in the number Sleep, Vol. 11. No.5, 1988

8 470 K. F. WHYTE ET AL. VERY FREQUENTLY SOMNOLENCE VERY SEVERE HEADACHE IN MffiNING NO., Placebo Acetazolamide Protrlpty line NO Placebo Acetazolamide Frotrlptyline RESTED IN MORNING SLEEP QUALITY NOT RESIEO IN THE HORNING r LJ Z Z o '" L I 0- o 0- '" 8 u.. => VERY RESTED 1r r- f- I \2 z < J.:J Z i'[ t:::i vi z I m Placebo Aceluzolamlde Protr iptyline Placebo Acetazolamide Frotn ptyllne FIG. 5. Effect of each therapy on individual patient's (n 10) symptoms of the sleep apnea/hypopnea syndrome (daytime somnolence, sleep quality, rested in the morning, and morning headache). of arousals or resulted in an improvement in nocturnal oxygenation, despite improvements in sleep quality. Furthermore protriptyline did not cause significant REM suppression. However, REM comprised 50% of the increased sleep duration (TST) on acetazolamide (26 of 52 min), but only 18% of the increased TST on protriptyline (9 of 50 min) (Fig. 4). Thus, protriptyline did produce a degree of relative REM suppression, even though absolute REM suppression did not occur. The patient (no. 3) who had both an objective and symptomatic response to protriptyline had a fall in REM sleep from Sleep, Vol. 11, No.5, 1988

9 DRUG THERAPY OF SLEEP APNEA % of TST with placebo to 4.8% with protriptyline, and two of the other three responders (nos. 7 and 8) with objective evidence of improvement had falls in the amount of REM sleep (from 22% to 17.5% and from 19.8% to 8.9%, respectively). We do not think our failure to suppress REM sleep resulted from noncompliance, as not only were heart rates higher on protriptyline, but also, staff members watched the patients taking their medication on each study night. The amount of REM suppression with tricyclic antidepressants has been demonstrated to vary widely between individual patients (22) and to variably decrease with time (23), and this may partly explain our observations. We have no explanation for this lack of clinical success of protriptyline. The mild symptoms in relation to the high frequency of side effects may have contrib uted. In an uncontrolled study, White et al. (11) reported that five of six patients with central sleep apnea reported symptomatic improvement with acetazolamide and that all six had a reduction in the total number of apneas. Sharp et al. (12) treated a group of patients who had different types of sleep apnea with acetazolamide. Acetazolamide produced no effect in patients with pure obstructive or pure central sleep apnea in comparison to a study performed when these patients were alkalotic. However, this group did show an increase in the frequency and severity of obstructive episodes in two of their patients with mixed sleep apnea. These authors postulated that the acidosis stimulates lower inspiratory muscles to contract, resulting in an increased negative pressure and more airway collapse and longer periods of obstruction. We found no prolongation of apneas with acetazolamide in this study, nor any consistent conversion of the few central and mixed apneas (always less than 20%) that our patients demonstrated into obstructive apneas. Our results would suggest that the role of either protriptyline or acetazolamide in the treatment of mild to moderate sleep apnea is limited. We failed to confirm previous reports on the efficacy of protriptyline, perhaps because our patients were less symptomatic (7,9,10) or because our study was randomised and blinded (8-10). The efficacy of acetazolamide in reducing apneas and hypopneas and improving sleep without affecting symptoms suggests that this agent merits further trial in patients with more severe forms of the sleep apnea/hypopnea syndrome. Acknowledgment: We wish to thank Helen Biernacki, Carol Hoy, and MaJjorie Venelles for technical assistance in the completion of this study. REFERENCES l. Strohl KP, Chemiack NS, Gothe B. Physiologic basis of therapy for sleep apnea. Am Rev Respir Dis 1986;134: Bradley D, Phillipson EA. The treatment of obstructive sleep apnea. Am Rev Respir Dis 1983;128: Fletcher EC. Sleep apnea treatment algorithm. In: Fletcher EC, ed. Abnormalities of respiration during sleep. Orland, FL: Grune & Stratton. 1986: Hollister LE. Tricyclic antidepressants. N Eng J Med 1978;299: 1l06-9, Phillipson EA. Regulation of breathing during sleep. Am Rev Respir Dis 1977;1l5: Bonora M, St. John WM, Bledsoe TA. Differential elevation by protriptyline and depression by diazepam of upper airway respiratory motor activity. Am Rev Respir Dis 1985;131: Brownell LG, West P, Sweatman P, Acres JC, Kryger MH. Protriptyline in obstructive sleep apnea. N Engl J Med 1982;307: Smith PL, Haponik EF, Allen RP, Bleecker ER. The effects of protriptyline in sleep disordered breathing. Am Rev Respir Dis 1983;127: Conway WA, Zorick F, Piccione P, Roth T. Protriptyline in the treatment of sleep apnea. Thorax 1982;37: Sleep, Vol. 11, No.5, 1988

10 472 K. F. WHYTE ET AL. 10. Clark RW, Schmidt HS, Schaal SF, Boudoulas H, Schuller DE. Sleep apnea: treatment with protriptyline. lveurology 1979;29: White DP, Zwillich CW, Pickett CK, Douglas NJ, Findley LJ, WeilJV. Central sleep apnea. Improvement with acetazolamide therapy. Arch Intern Med 1982;142: Sharp JT, Druz WS, D'Souza V, Diamond E. Effect of metabolic acidosis upon sleep apnea. Chest 1985;87: Douglas NJ. Control of breathing during sleep. Clin Sci 1984;67: Catterall JR, Douglas NJ, Calverley PMA, Shapiro CM, Brezinova V, Brash HM, Flenley DC. Transient hypoxemia during sleep in chronic obstructive pulmonary disease is not a sleep apnea syndrome. Am Rev Respir Dis 1983;128: Tweedale PM, Douglas NJ. Evaluation of Biox IIa ear oximeter. Thorax 1985;40: Rechstaffen A, Kales A. A manual of standardized terminology, techniques and scoring system for sleep stages of human subjects. Bethseda: National Institute of Neurological Disease and Blindness, 1968 (NIH publication no. 204). 17. Guilleminault e, Tilkian A, Dement we. The sleep apnea syndromes. Annu Rev Med 1976;27: Gould GA, Whyte KF, Airlie MAA, Rhind GB, Catterall JR, Shapiro CM, Douglas NJ. The sleep hypopnea syndrome. Am Rev Respir Dis 1988;137: Miller RG. Simultaneous statistical inference. New York: Springer Verlag, 1981: Sullivan CE, Issa FG. Obstructive sleep apnea. In: Kryger MH, ed. Clinics in chest medicine, vol 6: Sleep disorders. Philadelphia: WB Saunders, 1985: Carskadon MA. Guidelines for the Multiple Sleep Latency Test. Sleep 1986;9: Nakazawa Y, Kotorii M, Kotorii T, Ohshima M, Hasuzawa H. Individual variations in response of human REM sleep to amitriptyline and haloperidol. Electroencephalogn Clin NeurophysioI1977;42: Chen e. Sleep, depression and antidepressants. Br J Psychiat 1979;135: Sleep, Vol. 11, No.5, 1988