Effects of acetazolamide in patients the apnoea syndrome · The effects of acetazolamide on arterial blood gas dividedbytotal timeinbed), andsleep stagedistribu-tensions, resting

Thorax 1988;43:1 13-119

Effects of acetazolamide in patients with the sleepapnoea syndromeHIROKAZU TOJIMA, FUMIO KUNITOMO, HIROSHI KIMURA,KOICHIRO TATSUMI, TAKAYUKI KURIYAMA, YOSHIYUKI HONDAFrom the Department of Chest Medicine, Institute ofPulmonary Cancer Research, and Department ofPhysiology, School ofMedicine, Chiba University, Chiba, Japan

ABSTRACT There is as yet no convincing evidence that acetazolamide, a carbonic anhydraseinhibitor, is effective in obstructive sleep apnoea. A study was therefore designed to examine the effectof acetazolamide (250 mg/day) on sleep events and ventilatory control during wakefulness in ninepatients with the sleep apnoea syndrome. In eight of the nine patients the apnoea index and the totalduration ofapnoea were reduced by acetazolamide, and the mean (SEM) apnoea index of all patientschanged from 25-0 (6 7) to 18 1 (5 8) episodes an hour. Furthermore, the total time ofarterial oxygendesaturation (Sao2-more than 4% depression in Sao2 from the baseline sleeping level-divided bytotal sleep time was also significantly decreased and its mean (SEM) value improved from 24- 1 (7 9) to13-6 (4-8)% of total sleep time. Five of the seven patients with varying degrees of daytimehypersomnolence had their symptoms obviously improved. There was no patient whose predominanttype of apnoea was converted from the obstructive to the central type, or vice versa. In the studies ofwakefulness, metabolic acidosis, an increase of arterial oxygen tension (Pao2) and a decrease ofarterial carbon dioxide tension (PaCo2) were observed. The slopes of the occlusion pressure responseand the ventilatory response to carbon dioxide increased, and the carbon dioxide ventilatory responseline shifted to the left. It is suggested that acetazolamide cannot remove apnoea completely but has abeneficial effect in mild cases of obstructive sleep apnoea through an augmentation of central (CO2,H+) drive and a stabilising effect on ventilatory control.

Owing to insufficient knowledge about the patho-genesis of the sleep apnoea syndrome, effective treat-ment for most such patients has yet to be established.Current treatment is based merely on the physicalfindings of the patients, the type of apnoea, and theseverity ofthe syndrome. Drugs such as protriptyline'2and medroxyprogesterone acetate34 for obstructiveand central sleep apnoea were reported to be effectivein only a few patients.

Acetazolamide induces metabolic acidosis byinhibiting carbonic anhydrase in the renal tubularstructures and stimulates ventilation. Furthermore, itmay increase cerebral carbon dioxide tension (Pco2)by impeding carbon dioxide transport and may sup-press formation of cerebrospinal fluid bicarbonate atthe same time, resulting in a sustained increase inalveolar ventilation.

Address for reprint requests: Dr H Tojima, Department of ChestMedicine, Institute of Pulmonary Cancer Research, School ofMedicine, Chiba University, Chiba 280, Japan.

Accepted 13 October 1987

This agent has been used successfully to improveperiodic breathing during sleep at high altitude,56 andwas also reported to be useful in central sleep apnoea.7On the other hand, Sharp et al8 observed thatacetazolamide induced metabolic acidosis convertedcentral to obstructive apnoea and worsened hypox-aemia in two patients with mixed apnoea. Shore et al9reported a similar case. It remains to be determinedtherefore whether acetazolamide is useful or not forthe treatment of sleep apnoea syndrome, and par-ticularly for the subgroup with obstructive disease.

In the present investigation we have assessed theeffect of acetazolamide in patients with sleep apnoeasyndrome on indices of sleep disorder and ventilatorycontrol during wakefulness.

Methods

Five men and four women patients with sleep apnoeasyndrome were studied (table 1). In most of thempolysomnography was performed because of theirsymptoms-for example, heavy snoring, excessive

113

on June 21, 2021 by guest. Protected by copyright.

http://thorax.bmj.com

/T

horax: first published as 10.1136/thx.43.2.113 on 1 February 1988. D

ownloaded from

http://thorax.bmj.com/

Table 1 Anthropometric andpulmonaryfunction data of the patients with sleep apnoea syndrome

Resting arterial blood gas

Height Weight Weight Paco2 Pao2 [HCO31 VC FEV, %Age, sex (cm) (kg) (% ofideal) pH (kPa) (kPa) (mmol/l) (% pred)

1 62, F 147 76 153 7 40 6-3 7-5 28-8 69 492 33, M 162 90 150 7-40 5-0 10-5 23-4 109 853 58, F 147 85 171 7-35 6-9 7-8 27-9 70 704 65, M 172 76 113 7 38 5-9 10-4 25 7 71 425 47, M 163 76 134 7-40 5-2 8-8 24-7 97 706 72, F 159 65 123 7-42 5 9 9-0 28-5 81 707 72, F 144 36 74 7-41 5-3 11-4 25-2 72 608 53, M 167 92 153 7 35 5-9 10-0 25 0 81 799 61, M 163 78 138 7-40 5-3 10-2 24-7 88 84Mean 58 158 75 134 7 39 5-7 9 5 26-0 82 68SEM 4 3 6 10 0 01 0-2 0-4 0-6 5 5

Conversion: SI to traditional units-Blood gas data: I kPa = 75 mm Hg.

daytime sleepiness, and insomniac sensation. Thepredominant apnoea was the obstructive type, whichwas present in eight obese patients; one patient (case 7)who had purely central apnoea accompanied byCheyne-Stokes respiration was not obese. Patients 1and 3 were mildly hypercapnic, and patients 1 and 4had chronic obstructive pulmonary disease. Some ofthe patients were being treated with several drugs thatcontinued to be administered throughout the study. Atthe time of the study all subjects were clinically stable.

Before administering acetazolamide we performedpolysomnography for two consecutive nights, studiesof ventilatory control during wakefulness, pulmonaryfunction tests, and arterial blood gas analysis. In theacetazolamide treatment run, 250 mg was given orallyonce a day and the tests were repeated on the seventhor eighth day of administration.

During the sleep studies surface electrodes wereapplied to obtain an electroencephalogram (EEG), anelectrooculogram (EOG) and a submental electro-myogram [EMG). Arterial oxygen saturation (Sao2)was measured continuously with an ear oximeter (BioxIIA). Movement of the chest wall and abdomen wasmeasured with an inductance plethysmograph (Res-pitrace) and nasal flow with a thermistor. A polygraphwas used to record all the variables.

Sleep stages were determined by the criteria ofRechtschaffen and Kales.'° Apnoea was defined ascessation of flow at the nose for at least 10 seconds.Central apnoea was thought to occur when respiratoryeffort and airflow were absent, and obstructive apnoeawhen respiratory effort continued without airflow.Hypopnoea was defined when airflow and respiratorymovement were reduced in amplitude by one third,with a depression in Sao2."

Resting ventilatory indices, inspiratory minute ven-tilation (Vi), tidal volume (VT), respiratory frequency(f), mean inspiratory flow (VT/TI) and duty cycle (Ti/TT) were determined in seven patients while they were

breathing through a mouthpiece. VT, inspiratoryduration (TI), and expiratory duration (TE) wereelectrically displayed by an analogue calculator fromthe flow signal detected with a hot wire flow meter(Minato RF-H).

Hypoxic and hypercapnic ventilatory responsesduring wakefulness were measured in seven and eightpatients respectively. Hypoxic ventilatory responsewas determined by an isocapnic progressive hypoxiatest. During acetazolamide administration, as the endtidal PCo2 (PET CO2) decreased owing to drug inducedhyperventilation, measurement of the hypoxic venti-latory response was made at a PET CO2 lower than thepretreatment level. The hypoxic ventilatory responsewas evaluated from the linear regression between Viand Sao2 as well as between mouth occlusion pressureat 0-2 seconds (P02)'2 and Sao2, and their responseslopes were termed as AVi/ASao2 and APO.2/ASao2respectively. The hypercapnic ventilatory responsewas measured by Read's rebreathing method, and wasevaluated by the slopes of linear regression between Viand PET CO2 as well as between P0O2 and PET Co2 (AVi/APco2 and AP0 2/APcO2 respectively). Details aboutthe sleep study and the ventilatory response test havebeen reported previously.'3 14

Statistical analysis was performed by paired t testafter confirmation that the variability within eachgroup was the same.

Results

SYMPTOMS AND SIDE EFFECTSFive of the seven patients who had varying degrees ofdaytime sleepiness, the four who felt difficulty inachieving full arousal on awakening, and three of thefour who had an insomniac sensation had theirsymptoms improved to some degree duringacetazolamide administration. Though two patientscomplained of dysaesthesia of the extremities and

Tojima, Kunitomo, Kimura, Tatsumi, Kuriyama, Honda114

on June 21, 2021 by guest. Protected by copyright.

http://thorax.bmj.com

/T

horax: first published as 10.1136/thx.43.2.113 on 1 February 1988. D

ownloaded from

http://thorax.bmj.com/

Effects ofacetazolamide in patients with sleep apnoea syndrome 115

Table 2 Effects ofacetazolamide on arterial blood gases, resting ventilation, and hypoxic amd hypercapnic ventilatoryresponses (mean (SEM) values)

n Control Acetazolamide p

pH 9 7 39 (0-01) 7-34 (0-01)

Tojima, Kunitomo, Kimura, Tatsumi, Kuriyama, Honda

Apnoea index

EPi0ods/h60-

50

40-

30-

20-

10-

0

P

Effects ofacetazolamide in patients with sleep apnoea syndromeThe changes in the apnoea index, total apnoea time/

total sleep time, and desaturation time/total sleep timewere not significantly related to changes in theventilatory and occlusion pressure responses tocarbon dioxide during wakefulness. Moreover, theywere not correlated with changes in Paco2 and Pao2

Discussion

We found in this study that acetazolamide reduced butdid not remove the episodes of obstructive apnoea andoxygen desaturation during sleep and that itameliorated clinical symptoms associated with dis-ordered breathing. Acetazolamide has been shown tobe effective in two sorts of sleep disordered breathing.Firstly, during acute high altitude exposureacetazolamide was reported to reduce the number ofepisodes of periodic breathing and improve arterialoxygen saturation.56 Acetazolamide is known to beuseful for the prevention of acute mountain sickness.'6Secondly, White and colleagues7 reported thatacetazolamide substantially reduced episodes of cen-tral apnoea in all their six patients. Findley et al'7noted that acetazolamide calmed down the oscilla-tions of ventilatory movement and abolished therecurrent episodes of apnoea in Cheyne-Stokesbreathing. In contrast, Sharp and colleagues8 showedthat metabolic acidosis produced by acetazolamideconverted central apnoea to obstructive or mixedapnoea. The reason for this remains uncertain, but itmay very well be due to the generation of greaternegative inspiratory pressure, facilitating upperairway closure. No further reports, however,have appeared to substantiate the evidence of eitherbeneficial or adverse effects of acetazolamide onobstructive sleep apnoea.The number of apnoeic episodes per hour of total

sleep time during the control period varied from 5 4 to571; in two mild cases it was less than 10. Theconventional definition for sleep apnoea syndrome-more than five apnoeic episodes an hour-may in factbe an inappropriately small index number for predict-ing increased health risk or somnolence in subjectsover 60 years of age.'8 The two patients with the mildsyndrome, however, were a 33 year old man and a 58year old woman. We therefore diagnosed them ashaving sleep apnoea syndrome. The mean apnoeaindex of our subjects was 25 0, indicating a relativelymild sleep apnoea syndrome. The one patient withsevere obstructive apnoea treated with acetazolamidebut failing to show improvement in oxygen desatura-tion had the so called "saw tooth" configuration inher expiratory flow-volume curve."' In cases whereobstructing lesion or abnormality is present in theupper airway it may be difficult to improve thedisordered breathing.

In our study one patient who had purely centralapnoea did not develop obstructive apnoea withacetazolamide treatment. Onal and colleagues20 foundthat occlusive and mixed apnoeic episodes occurred atthe nadir of periodic fluctuation in diaphragmatic andgenioglossal activities, suggesting that both types ofapnoea resulted from an instability of ventilatorycontrol during sleep. Their recent report2' showed thatocclusive sleep apnoea resulted from hypoxia inducedperiodic breathing in the presence of inspiratoryresistive loading in normal volunteers. Remmers andcolleagues,22 having studied patients with the Pick-wickian syndrome, indicated that the genioglossalelectromyogram fluctuated systematically in relationto the periodic breathing cycle: low level activity at theonset of occlusion and prominent discharge at theinstant of pharyngeal opening. From these investiga-tions it is suggested that the pathogenesis of obstruc-tive apnoea is related to the instability of ventilatorycontrol in the presence of structural encroachment inthe oropharyngeal lumen. Thus administration of anagent capable of producing a stabilising effect onrespiratory control may be useful for patients withobstructive sleep apnoea, unless the patient has anobstructive lesion in the upper airway.

In the present study distinctive findings weremetabolic acidosis and an augmentation of hypercap-nic chemosensitivity during wakefulness afteracetazolamide administration. Furthermore, weobserved augmentation ofminute ventilation at a Pco2of 8 0 kPa (60mm Hg), which indicated left hand shiftin the hypercapnic ventilatory response line. Wepreviously confirmed that metabolic acidosisproduced by acetazolamide increased ventilation andchemosensitivity to carbon dioxide in healthy men.23In this previous study, however, the augmentation ofthe ventilatory response to carbon dioxide byacetazolamide was larger than that by ammoniumchloride induced acidosis and therefore the effect ofimpeding carbon dioxide transport with a resultantincrease in cerebral Pco2 and [H+] might affect carbondioxide chemosensitivity. If acetazolamide preservedventilatory stability during sleep and functioned toreduce periodic apnoea, the following factors that mayaccount for improving sleep disorders can be con-sidered. Firstly, a left hand shift of the carbon dioxideresponse line reduces the tendency for periodic breath-ing owing to a stabilising effect.24 Secondly, theenhanced ventilation induced by metabolic acidosisreduces hypoxia, and the augmented central (CO2, H +)drive and the relief of hypoxia decrease the relativeinfluence of the hypoxic drive that makes the ven-tilatory control more unstable than the (CO2, H+)driving system. Several studies suggest that oxygenadministration may favourably influence central sleepapnoea.25

117

on June 21, 2021 by guest. Protected by copyright.

http://thorax.bmj.com

/T

horax: first published as 10.1136/thx.43.2.113 on 1 February 1988. D

ownloaded from

http://thorax.bmj.com/

118During acetazolamide treatment comparison of the

hypoxic ventilatory response with the control responsewas difficult because Pco2 was depressed by the drugadministration. Whether there is an interactive effectbetween metabolic acidosis and hypoxic chemosen-sitivity remains to be determined,2627 but our previousstudy23 and that of Powles et al8 indicated thatacetazolamide did not enhance the hypoxic ventilatoryresponse in eucapnic conditions. Another possiblemechanism for reducing episodes of apnoea byacetazolamide is alteration of the distribution profileof sleep stages, as was seen after protriptyline.2 In ourresults, however, there was no remarkable change insleep stages, and acetazolamide reduced sleep apnoeasimilarly in both non-REM and REM sleep, except inone patient, whose apnoea index increased only innon-REM sleep. A significant increase in lowest Sao2was seen in non-REM sleep, whereas a significantimprovement in apnoea index was observed in eachsleep stage. From these findings it is difficult to believethat acetazolamide had different effects on non-REMand on REM sleep.To evaluate the effect of a chemical agent on sleep

disordered breathing in patients with sleep apnoeasyndrome, quantitative measurement appears to be asimportant for hypopnoeic episodes as for apnoea. Infact, in our study there was one patient whose apnoeicepisodes were diminished but hypopnoeic episodeswere increased, so that the total desaturation time wasnot improved. Oxygen desaturation time depends onlung volume (functional residual capacity), baselineSao2, and the number and duration of episodes ofapnoea and hypopnoea. Thus the beneficial effects ofacetazolamide on sleep apnoea syndrome seem to bereflected by the reduction of desaturation time.On the other hand, care must be taken regarding the

negative effects of acetazolamide. Since it is knownthat metabolic acidosis affects pulmonary arterialresistance, cardiac contractility, and oxygen transport,this drug should be used only in conjunction withcareful control of arterial pH. We used only 250 mg ofacetazolamide a day to avoid the development ofsevere metabolic acidosis. Since mean arterial pHdecreased from 7 39 to 7-34 with this dose, theadministration of more than 250 mg per day wouldappear to be unwise. If the depression of bicarbonateconcentration develops without concomitant loweringof Paco2, the resulting severe metabolic acidosis mayendanger some vital organs, and might finally becomelife threatening.2930

We thank Drs S Okita, Y Yuguchi, S Masuyama, SKouchiyama, T Shinozaki, S Tazawa, and T Miyagifor cooperation and valuable discussion. This workwas supported in part by a grant from the researchcommittee for intractable respiratory failure of theMinistry of Health and Welfare of Japan.

Tojima, Kunitomo, Kimura, Tatsumi, Kuriyama, HondaReferences

1 Clark RW, Schmidt HS, Schaal SF, Boudoulas H,Schuller DE. Sleep apnea: treatment with protriptyline.Neurology 1979;29: 1287-92.

2 Brownell LG, West P, Sweatman P, Acres JC,Kryger MH. Protriptyline in obstructive sleep apnea. Adouble-blind trial. N Engl J Med 1982;307:1037-42.

3 Sutton FD, Zwillich CW, Creagh CE, Pierson DJ,Weil JV. Progesterone for outpatient treatment ofPickwickian syndrome. Ann Intern Med 1975;83:476-9.

4 Strohl KP, Hensley MJ, Saunders NA, Scharf SM,Brown R, Ingram RH. Progesterone administrationand progressive sleep apneas. JAMA 198 1;245:1230-2.

5 Weil JV, Kryger MH, Scoggin CH. Sleep and breathing athigh altitude. In: Guilleminault C, Dement WC, eds.Sleep apnea syndrome. New York: Alan R Liss,1978:119-36.

6 Sutton JR, Gray GW, Houston CS, Powles ACP. Effectsof duration at altitude and acetazolamide on ventila-tion and oxygenation during sleep. Sleep 1980;3:455-64.

7 White DP, Zwillich CW, Pickett CK, Douglas NJ,Findley LJ, Weil JV. Central sleep apnea. Improve-ment with acetazolamide therapy. Arch Intern Med1982;142: 1816.

8 Sharp JT, Druz WS, D'Souza V, Diamond E. Effect ofmetabolic acidosis upon sleep apnea. Chest 1985;87:619-24.

9 Shore ET, Millman RP. Central sleep apnea and aceta-zolamide therapy. Arch Intern Med 1983;143:1278-80.

10 Rechtaschaffen A, Kales A, eds. A manual ofstandardizedterminology, techniques and scoring system for sleepstages of human subjects. Bethesda, Maryland:National Institutes of Health, 1968. (NIH publicationNo 2041.)

11 Wynne JW, Block AJ, Hemenway J, Hunt LA, Flick MR.Disordered breathing and oxygen desaturation duringsleep in patients with chronic obstructive lung disease(COLD). Am J Med 1979;66:573-9.

12 Yoshida A, Hayashi F, Sasaki K, Masuda Y, Honda Y.Analysis of pressure profile in the occluded airwayobtained at the beginning of inspiration in steady statehypercapnia. Am Rev Respir Dis 1981;124:252-6.

13 Tatsumi K, Kimura H, Kunitomo F, et al. Effect ofchlormadinone acetate on ventilatory control inpatients with chronic obstructive pulmonary disease.Am Rev Respir Dis 1986;133:552-7.

14 Tatsumi K, Kimura H, Kunitomo F, Kuriyama T,Watanabe S, Honda Y. Sleep arterial oxygendesaturation and chemical control of breathing duringwakefulness in COPD. Chest 1986;89:68-73.

15 Tammelin BR, Wilson AF, Borowiecki BDB, Sassin JF.Flow-volume curves reflect pharyngeal airway abnor-malities in sleep apea syndrome. Am Rev Respir Dis1983;128:712-5.

16 Forwand SA, Landowne M, Follansbee MJN, HansenCJE. Effect of acetazolamide on acute mountainsickness. N Engl J Med 1968;279:839-45.

17 Findley LJ, Blackburn MR, Goldberger AL, Mandell A.Apneas and oscillation of cardiac ectopy in Cheyne-Stokes breathing during sleep. Am Rev Respir Dis1984;130:937-9.

on June 21, 2021 by guest. Protected by copyright.

http://thorax.bmj.com

/T

horax: first published as 10.1136/thx.43.2.113 on 1 February 1988. D

ownloaded from

http://thorax.bmj.com/

Effects ofacetazolamide in patients with sleep apnoea syndrome18 Berry DTR, Webb WB, Block AJ. Sleep apnea syndrome.

A critical review of the apnea index as a diagnosticcriterion. Chest 1984;86:529-31.

19 Haponik EF, Smith PL, Kaplan J, Bleecker ER.Flow-volume curves and sleep-disordered breathing:therapeutic implications. Thorax 1983;38:609-15.

20 Onal E, Lopata M, O'Connor M. Pathogenesis ofapneasin hypersomnia-sleep apnea syndrome. Am Rev RespirDis 1982;125:167-74.

21 Onal E, Burrows DL, Hart RH, Lopata M. Induction ofperiodic breathing during sleep causes upper airwayobstruction in humans. J Appl Physiol 1986;61:1438-43.

22 Remmers JE, deGroot WJ, Sauerland EK, Anch AM.Pathogenesis of upper airway occlusion during sleep.J Appl Physiol 1978;44:931-8.

23 Tojima H, Kunitomo F, Okita S, et al. Difference in theeffects of acetazolamide and ammonium chlorideacidosis on ventilatory responses to CO2 and hypoxiain humans. Jpn J Physiol 1986;36:511-21.

24 Cherniack NS, von Euler C, Homma I, Kao FF.Experimentally induced Cheyne-Stokes breathing.Respir Physiol 1979;37:185-200.

25 Martin RJ, Saunders MH, Gray BA, Pennock BE. Acuteand long-term ventilatory effects of hyperoxia in theadult sleep apnea syndrome. Am Rev Respir Dis1982;125:175-80.

26 Irsigler GB, Stafford MJ, Severinghaus JW. Relationshipof CSF pH, 02, CO2 responses in metabolic acidosisand alkalosis in humans. J Appl Physiol 1980;48:355-61.

27 Knill RL, Clement JL. Ventilatory responses to acutemetabolic acidemia in humans awake, sedated, andanesthetized with halothane. Anesthesiology 1985;62:745-53.

28 Powles ACP, Sutton JR, Rigg JRA. Acetazolamide andthe ventilatory response to hypoxia: comparison ofeucapnic and hypercapnic conditions [abstract]. AmRev Respir Dis 1980;121(suppl):392.

29 Coudon WL, Block AJ. Acute respiratory failureprecipitated by a carbonic anhydrase inhibitor. Chest1976;69:1 12-3.

30 Heller I, Halevy J, Cohen S, Theodor E. Significantmetabolic acidosis induced by acetazolamide not a rarecomplication. Arch Intern Med 1985;145:1815-7.

119

on June 21, 2021 by guest. Protected by copyright.

http://thorax.bmj.com

/T

horax: first published as 10.1136/thx.43.2.113 on 1 February 1988. D

ownloaded from