DRUG EXPERIENCE

Drug Safety 7 (2): 152-158. 1992 0114-5916/ 92/0003-0152/$03.50/0 © Adis International Limited. All rights reserved. DRSl

Effects of Estazolam and Flurazepam on Cardiopulmonary Function in Patients with Chronic Obstructive Pulmonary Disease Martin A. Cohn, David D. Morris and David Juan Sleep Disorder Center of Southwest Florida, Naples, Florida, and Clinical Statistics and Systems Development and Clinical Research Department, Abbott Laboratories, Abbott Park, Illinois, USA

Summary

Benzodiazepine drugs have been shown to suppress respiratory function in patients with chronic obstructive pulmonary disease (COPD). We designed a placebo-controlled crossover study to compare the effects of a new benzodiazepine, estazolam ('ProSom'), with those of flurazepam ('Dalmane') on cardiopulmonary function in COPD patients. 29 patients completed all treatment phases (estazolam 2mg, flurazepam 30mg or placebo). Respiratory and cardiovascular function were assessed in awake patients on days I and 5 (acute and cumulative effects). Eight patients were also assessed during sleep in each period. The effects of estazolam and flurazepam on ventilatory response to C02 and mouth occlusion pressure were no different from those of placebo. However, acute administration of flurazepam lowered tidal volume and increased inspiratory flow. Although no clinical signs of respiratory depression were observed with any long term treatment, flurazepam decreased oxygen saturation and inspiratory time and increased respiratory frequency. Neither drug altered breathing control during sleep. Our results indicate that estazolam 2mg is equally as safe a hypnotic agent as flurazepam for patients with mild COPD.

Estazolam ('ProSom'), a new benzodiazepine drug, has been reported to be an effective hypnotic in several clinical trials (Gustavson & Carrigan 1990; Lamphere et al. 1986; Pierce & Shu 1990). No adverse effects on pulmonary function have been observed in healthy subjects receiving 2 to 3 times the normally recommended adult dose of estazolam 2mg (Pierce et al. 1990). Other benzodiazepines, including flurazepam, lorazepam and diazepam, have been shown to suppress respiratory function in patients with chronic obstructive pulmonary disease (COPD) [Denaut et al. 1974; Geddes et al. 1976]. The purpose of this study was to evaluate the effects of a therapeutic dose of estazolam on pulmonary and cardiac func-

tion in patients with COPD during both awake and sleep states. For intrapatient comparison, patients were also given a therapeutic dose of flurazepam as well as placebo in one of 3 possible treatment sequences.

Methods Patients Patients were 60 years old or less and had COPD characterised by cough or breathlessness (American Thoracic Society 1962). Criteria for entry into the study were: (a) COPD of mild to moderate severity, with forced expiratory volume in I sec (FEV d ranging from 50 to 75% of predicted values;

Estazolam and Flurazepam in COPD Patients

(b) end-expiratory C02 (measured by capnography) of 45mm Hg or less; (c) oxygen saturation (measured by pulse oximetry) ~ 85%; (d) baseline mouth occlusion pressure (MOP) response to C02 ~ 0.15 cm H20/mm Hg pC02; and (e) chest roentgenogram compatible with COPD. None of the patients had evidence of recent myocardial infarction or other electrocardiogram (ECG) abnormalities except those consistent with their pulmonary disease. Patients were excluded from the study if they had used opioids, psychotropics, hypnotics, antihistamines or decongestants within 2 weeks prior to the study, or were taking any concurrent medication that would interfere with the absorption or metabolism of flurazepam or estazolam. Patients were told not to drink alcohol during the entire study period and to avoid caffeinated beverages within 4h of reporting to the sleep laboratory. Each patient signed a consent form before entering the study. The study protocol was approved by the Institutional Review Board, Mt Sinai Medical Center, Miami Beach, Rorida. Study Design This randomised, double-blind, placebo-controlled, 3-period, crossover study utilised a Latin square design. The 7-week study consisted of 3 X 5-day treatment periods, each separated by a 2-week washout period. During the treatment phases, patients received an evening dose of either flurazepam 30mg, estazolam 2mg, or placebo; drugs and placebo were administered in identicalappearing capsules supplied in coded bottles. Each patient was randomly assigned to one of 3 treatment group sequences: estazolam-flurazepamplacebo, flurazepam-placebo-estazolam, or placebo-estazolam-flurazepam. One week before the study, a medical history, physical examination, chest x-ray, laboratory tests (blood chemistry, haematology and urinalysis) and 12-lead ECG were obtained. In addition, baseline pulmonary function tests that were repeated during the study were performed on each patient. The bronchodilator response to 2 puffs from a metered

153

dose inhaler of isoprenaline (isoproterenol) [total dose 262 ~g] was measured 15 min after inhalation in order to separate asthmatic patients from those with chronic bronchitis. On days I and 5 of each treatment period, the patients reported to the sleep laboratory at approximately 1800h. Blood and urine specimens were collected, pulmonary function data were obtained, and blood pressure and heart rate were measured. The study medication (estazolam 2mg, flurazepam 30mg or placebo) was administered in the early evening. The acute effects of the medication on cardiopulmonary function were assessed at 2, 4 and 6h after the first dose on the evening of day I; the cumulative effects were evaluated on the evening of day 5 of each treatment period at 2, 4 and 6h after administration. Patients were awake during the assessment. Patients slept in the sleep laboratory during all 5 treatment nights. On days 2, 3 and 4, they reported to the laboratory at approximately 2200h and received the study drug 30 min before bedtime. Assessment of Cardiopulmonary Function During the Awake State On the evenings of day I and day 5, the following tests were performed to assess respiratory and cardiac function: ventilatory response to C02, MOP, respiratory inductive plethysmography (RIP), ear oximetry, spirometry and cardiac measurements. Ventilatory response to C02, a sensitive test of respiratory drive, was performed using the method by Read (1967); patients breathed through a mouthpiece attached to a Rudolph valve, which separated the inspiratory line from the expiratory line. Row was measured by a bag-in-box system connected to a Reisch pneumotachograph. C02 concentration was measured from the expiratory line using a rapid-responding C02 analyser. All signals were digitised and analysed by computer. Minute ventilation (V min), calculated from the tidal volume derived from the measured flow signal using the 3 breaths preceding the occlusion pressure

154

measurement (described below), was plotted against pC02. From the graph, the ventilatory C02 response intercept and slope were derived by linear regression. Ventilatory response to C02 was assessed before study drug administration and 2, 4 and 6h after dosing. Mouth occlusions for the measurement of MOP, a sensitive measure of central respiratory drive, were performed by briefly closing a stopcock on the inspiratory line without warning to the patient. This manoeuvre was done approximately every 5 breaths until the end of the rebreathing period. Pressure was measured at the mouthpiece with a Validyne pressure transducer. The MOP measured 0.1 sec after occlusion was plotted by computer against pC02. The slope by linear regression of this line was considered the value for MOP. During wakefulness, MOP was assessed before dosing and 2, 4, and 6h after dosing. RIP was used to measure ventilation and breathing patterns. Insulated wire coils were placed around the patient's rib cage and abdomen (Cohn et al. 1982). During respiration, changes in the cross-sectional areas of the rib cage and abdomen altered the self-inductances of the coils, modulating the frequencies of the oscillators to which the coils were connected. The signal emitted was then calibrated to spirometric measurements. Mean values of all breaths during a 10 min period of quiet breathing were obtained for the following respiratory parameters: (a) frequency of breathing; (b) tidal volume (VT); (c) Vmin; (d) inspiratory time (Ti); (e) fractional inspiratory time of the total breath period (Ti/Ttot); (t) index of rib cage and abdominal asynchrony based on maximum compartmental amplitude (MCAjVT); and (g) percentage contribution of rib cage to total VT (%RCjVT). Measurements were made before dosing, every 30 min for the first 2h after the study medication was taken, and 4 and 6h after dosing. Arterial oxygen saturation was measured transcutaneously by ear pulse oximetry (BIOX II, Ohmeda). The tests were performed before dosing and 2, 4 and 6h after dosing while subjects were awake. Forced vital capacity (FVC) and FEV 1 were obtained using a rolling-seal spirometer (Ohio Instru-

Drug Safety 7 (2) 1992

ments, Inc.). Spirometry was performed before and 2, 4 and 6h after administration. Blood pressure, heart rate and systolic time interval were also monitored. The systolic time interval, a measure of left ventricular function, was obtained using a modification of RIP, in which a small insulated coil was applied around the neck to obtain a clear carotid pulse signal. The signal was analysed using standard techniques to assess left ventricular function (Spodick & Lance 1976). All cardiac measurements were taken before dosing and 1, 4 and 6h after administration of the study medication. Assessment of Pulmonary Function During Sleep The first 8 patients who were willing to undergo additional observations of sleep and respiration participated in the assessment of cardiopulmonary function during sleep. On 1 night (excluding nights 1 and 5) of each treatment period, the following were monitored during sleep: sleep stages, RIP and respiratory events. Electroencephalography, electro-oculograph y and digastric electromyography were continuously monitored on a polygraph recorder (Grass Instruments) according to standard techniques (Rechtschaffen & Kales 1968). Sleep stages were scored using 1 min epochs. Arterial oxygen saturation was continuously measured according to the technique described during the wake phase. Using the technique previously described, data were collected for approximately 10 min during wakefulness, REM sleep, sleep stages 1 and 2, and sleep stages 3 and 4. Data were sampled from both the first and second halves of the night and average values of the previously described respiratory data were compared. Data were collected on respiratory events, defined as a 4% or greater reduction in oxygen saturation from baseline. Such respiratory events were usually accompanied by either apnoea (VT less than 100ml) or hypopnoea (VT reduced from baseline).

Estazolam and Aurazepam in COPD Patients

Statistical Analysis All analyses were performed with the Statistical Analysis System, and analysis of variance (ANOVA) comparisons were conducted with the Type III sums of squares of the general linear model (GLM) procedure. Values of p < 0.05 were considered statistically significant. Acute effects on cardiopulmonary function were evaluated on days 1 and 5 by the change from predose value to the average postdose value. Cumulative effects were evaluated by the change in the predose measurement from day 1 to day 5. Given the lack of carryover effects, acute and cumulative effects were examined for treatment group differences within the framework of ANOV A with patient, study period and treatment effects. Pairwise treatment group comparisons were made between the least-squares means.

155

¥

Mouth occlusion pressure

!0'6l~"'~ ~0.4

................ .

8~ 0.2 Q.~

45

¥

Ventilatory C02 response (intercept) ",'

40

.. '

E

.s

~

35

,'.

.,.'

",

30

."",.",

15-

t:.

¥ 2.51

Ventilatory C02 response (slope)

_"_"._"._"_"._.~

l_-_CS::_·

o

•._,,.__,,_,,._._ .. _......_',._ .. ,_',._.. _..._,.._.,'_..•, -

2-3 Time (h)

4-5

6-7

Acute effects of estazolam (e), flurazepam (_) and placebo (0) on mouth occlusion pressure response and on the intercept and slope of the ventilatory response to C02 on day I. No significant changes were noted. Fig. 1.

Results 29 of the 34 patients who enrolled in the study completed all 3 treatment periods. Their mean age was 48.2 ± 9.0 years. In addition to the clinical symptoms of COPD, all patients had mild obstructive changes on baseline pulmonary function tests but were neither hypoxic nor hypercapnic. Serum drug levels verified that 2 weeks was an adequate washout between treatment phases. Cardiopulmonary Function While Awake Changes in respiratory and cardiovascular function during wakefulness were assessed after both short and long term treatment. There were no significant differences between estazolam and flurazepam recipients on tests of pulmonary function on day 1. On day 1, MOP and ventilatory response to C02 were similar in the estazolam, flurazepam and placebo groups (fig. 1). Similar findings were seen with ear oximetry and spirometry measurements, which revealed no significant treatment group differences in the average change from predose to postdose on day 1. Compared with the placebo group, 2 signifi-

cantly greater changes in the RIP measurements from predose to postdose were seen with flurazepam treatment on day 1. Mean VT fell from 506.1 to 452.4mI (11 %) in the flurazepam group, whereas mean VT increased from 454.0 to 461.8ml (2%) in the placebo group (p = 0.013). Ti/Ttot increased from 0.39 to 0.69 (8%) in the flurazepam group but remained essentially unchanged at 0.40 in the placebo group (p = 0.026). A significant difference in the change in heart rate was seen between the estazolam and placebo groups on day 1. Heart rate increased from 69.5 to 73.6 beats/min (6%) in the estazolam group and decreased from 74.9 to 73.0 beats/min (3%) in the placebo group (p = 0.037). None of the systolic time interval variables yielded any statistically significant treatment group differences on day I. On day 5, the only significant treatment group difference was in the change from predose to postdose in TJTtot. The increase from 0.40 to 0.43 (8%) in the estazolam group was significantly higher (p = 0.048) than the increase from 0.41 to 0.42

Drug Safety 7 (2) 1992

156

(2%) and the increase from 0.39 to 0.40 (3%) in the flurazepam and placebo groups (p = 0.018), respectively. Cumulatively, estazolam produced only I statistically significant difference from placebo between day I predose and day 5 predose (table I). The average decrease from 105.5 to 101.0 msec (4%) in the pre-ejection period for systolic time interval was significantly different from the increase from 99.8 to 108.1 msec (8%) for the placebo group (p = 0.038). The remaining significant treatment group differences involved flurazepam and included increases in respiratory frequency and heart rate, as well as decreases in oxygen saturation, inspiratory time, and systolic blood pressure. However, no clinical signs of respiratory depression were observed in any patient during the study. Pulmonary Function During Sleep In the 8 patients who were assessed while asleep, breathing patterns were similar during the estazolam, flurazepam and placebo periods (fig. 2). Regardless of treatment, Ti/Ttot was lowest and VTI Ti was highest while patients were awake. During REM stages, decreases were observed in Vmin, VT and VT/Ti. The %RC/VT during the awake state and during REM sleep was lower than that during non-REM sleep.

No significant drug effects were observed reo garding the number or duration of adverse respi· ratory events or the lowest oxygen saturation.

Discussion Any central nervous system depressant, includ· ing benzodiazepines, may suppress respiration if given in sufficient dosage (Widdicombe 1981). Patients with CO PO may be more sensitive to res· piratory depressant effects. In this study, estazolam and flurazepam administered in therapeutic doses to patients with mild CO PO produced no signifi. cant effects on ventilatory response to C02 or MOP when compared with placebo; this held true fol· lowing either short or long term administration of the drugs. We used MOP as a measure of respiratory function because it is unaffected by pulmonary function or responses of respiratory muscles, and therefore is considered a better test of respiratory drive than ventilatory response to C02 in patients with lung disease (Zackon et al. 1976). While the subjects in our study were not insomniacs, it was assumed that the study drugs would have produced similar effects in insomniacs with mild CO PD. Monitoring of breathing patterns with RIP disclosed some acute effects on day I with flurazepam but none with estazolam. A small but significant

Table I. Cardiopulmonary measurements with significant treatment group differences after long term benzodiazepine administration Change from day 1 to day 5 predose

Oxygen saturation (%) Respiratory frequency (beats/min) Inspiratory time (msec) Systolic blood pressure (mm Hg) Heart rate (beats/min) Systolic time interval: preejection period (msec)

placebo

estazolam

flurazepam

0.09 «1%) 0.01 «1%) -0.02 ( 2%) -0.84 ( 1%) -7.02 ( 9%) 8.27 ( 8%)

-0.33 «1%) 0.92 ( 4%) 0.00 «1%) -0.47 «1%) 0.30 «1%) -4.47 ( 4%)""'

-1.64 ( 2%)" 3.15 (15%)" -0.18 (13%)"··· -7.17 ( 6%)"··· 5.87 ( 8%)" 1.59 ( 2%)

• Flurazepam significantly different from placebo: p = 0.011 for oxygen saturation; p = 0.021 for respiratory frequency; p = 0.032 for inspiratory time; p = 0.010 for systolic blood pressure; and p = 0.003 for heart rate . •• Flurazepam significantly different from estazolam: p = 0.017 for inspiratory time and p = 0.006 for systolic blood pressure . ... Estazolam significantly different from placebo: p = 0.038 for systolic time interval.

Estazolam and Aurazepam in COPD Patients

157

90~~

·e080"'~ ~ ...... .

> 7.0

······· ...... 0

6.0

~··*'·········7 ~ ~r

"400'

>

....

···0

300

. . . 300 250

75~_

~:: ~~ .................... ~

()

~

•

.

I

Awake

I

Stage 2

..~

I

Stages 3/4

REM

Effects of estazolam (e), flurazepam (_) and placebo on control of breathing during the night. The breathing parameters are minute ventilation (V min), tidal volume (VT), inspiratory time (TifTtot), mean inspiratory flow (VTfTi), and percent rib cage contribution to VT (%RC). Measurements were taken by respiratory inductive plethysmography during wakefulness, stages 1 and 2 non-REM sleep, stages 3 and 4 non-REM sleep, and REM sleep. No significant drug effects were noted. Fig. 2.

(0)

reduction in VT and faster mean inspiratory flow (Ti/T tot) followed short term administration of flurazepam. Oxygen saturation was reduced with long term administration offlurazepam, suggesting mild respiratory depression. However, the clinical importance of these small changes is uncertain, and may be related to differences in the respiratory effects of the doses for each drug that were used. These tests were performed in awake subjects after administration of a hypnotic. Keeping the patients awake sometimes required stimulation by

conversation, which may have influenced the results. More meaningful information on the effects of these drugs on control of breathing may come from the small group that was studied while asleep. Because the drugs are given to induce sleep, their effects compared with the effect of natural sleep alone (i.e. the placebo period) are especially relevant. Only recently have studies been performed on the control of breathing during sleep and the effects of central nervous system depressants such as hypnotics and alcohol. Ventilatory response to C02 decreases during natural sleep, especially in REM sleep (Douglas et al. 1982a, 1982b; Phillipson 1977). The hypoxic ventilatory response is more variable. During sleep, RIP reveals a progressive reduction in VT, Vmin and mean inspiratory flow (respiratory drive) [Cohn et al. 1983; Tabachnik et al. 1981 a, 1981 b). Sleep-related breathing disturbances resulting in oxygen desaturation are sometimes observed, especially in patients with capo (Tobin et al. 1983). In healthy subjects, the administration of therapeutic doses of flurazepam before sleep has been shown to reduce the arousal threshold but not to suppress respiration beyond sleep itself (Gothe et al. 1986; Hedemark & Kronenberg 1983). Although flurazepam previously was found to increase the frequency of sleep-disordered breathing in patients with capo, the magnitude of these changes was small (i.e. an increase in respiratory events from 4 to 8 per night after drug administration), and they were not felt to be clinically significant (Block et al. 1984). Monitoring by RIP in the present study revealed that neither estazolam nor flurazepam altered the control of breathing during sleep. Neither drug significantly increased the occurrence of sleeprelated breathing disturbances such as apnoea or caused a decline in oxygen saturation beyond that seen with placebo in the 8 subjects studied during sleep. These data suggest that estazolam may be administered to nonelderly patients with mild nonhypercapnic capo with as great a degree of safety as flurazepam. Because of the high prevalence of sleep-related

158

respiratory disturbances in the elderly, the safety of estazolam in geriatric patients is an important issue. To date, estazolam has been studied in several geriatric trials (Vogel 1991; Abbott Laboratories 1991). Daytime psychomotor tests revealed no deficits in patients receiving estazolam I mg; in I study, minimal deficits were observed in patients receiving 2mg. In all but I study, no other clinically significant adverse effects were observed, the exception being a trial in which 4 patients had adverse events that warranted discontinuation of therapy. However, the events may not have been linked with drug administration, as they occurred in equal numbers of estazolam and placebo recipients (2 in each group). The recommended initial dose of estazolam for healthy elderly patients is Img; for debilitated elderly patients, it is O.5mg, with the caution that dosage increases be made with particular care. From our findings and those reported in estazolam trials in geriatric populations, we speculate that this drug may be safely administered to elderly patients with mild COPD. However, clinical trials are needed to establish drug safety in these patients with certainty.

References Abbott Laboratories. Data on file: M80-060, M81-023, M82-OO3, and M79-062, 1991 American Thoracic Society. Chronic bronchitis, asthma, and pulmonary emphysema. American Review of Respiratory Disease 85: 762-768, 1962 Block AJ, Dolly FR, Slayton Pc. Does flurazepam ingestion affect breathing and oxygenation during sleep in patients with chronic obstructive lung disease? American Review of Respiratory Disease 129: 230-233, 1984 Cohn MA, Nay KN, Kiel M, et al. Pattern of breathing during sleep in insomniac adults. Sleep Research 12: 232, 1983 Cohn MA, Rao ASV, Broudy M, Birch S, Watson H, et al. The respiratory inductive plethysmograph: a new non·invasive monitor of respiration. Bulletin European de Physiopathologie Respiratoire 18: 643-658, 1982 Denaut M, Yernault JC, Coster A. Double-blind comparison of the respiratory effects of parenteral lorazepam and diazepam in patients with chronic obstructive lung disease. Current Medical Research and Opinion 2: 611·615, 1974 Douglas NJ, White DP, Weil JV, Pickett CK, Martin RJ, et al. Hypoxic ventilatory response decreases during sleep in normal

Drug Safety 7 (2) 1992

men. American Review of Respiratory Disease 125: 286·289, 1982a Douglas NJ, White DP, Weil JV, Pickett CK, Zwillich CW, et al. Hypercapnic ventilatory response in sleeping adults. American Review of Respiratory Disease 126: 758-762, 1982b Geddes DM, Rudolf M, Saunders KB. Effect of nitrazepam and flurazepam on the ventilatory response to carbon dioxide. Thorax 31: 548-551, 1976 Gothe B, Cherniack NS, William L. Effect of hypoxia on ventilatory and arousal responses to C02 during NREM sleep with and without flurazepam in young adults. Sleep 9: 24-37, 1986 Gustavson LE, Carrigan PJ. The clinical pharmacokinetics of single doses of estazolam. American Journal of Medicine 88 (Suppl. 3A): 2S-5S, 1990 Hedemark LL, Kronenberg RS. Flurazepam attenuates the arousal response to C02 during sleep in normal subjects. American Review of Respiratory Disease 128: 980·983, 1983 Lamphere J, Roehrs T, Zorick F, Koshorek G, Roth T. Chronic hypnotic efficacy of estazolam. Drugs Under Experimental and Clinical Research 12: 687-691, 1986 Phillipson EA. Regulation of breathing during sleep. American Review of Respiratory Disease 115: 217-224, 1977 Pierce MW, Shu VS. Efficacy of estazolam: the United States clinical experience. American Journal of Medicine 88 (Suppl. 3A): 6S-IIS, 1990 Pierce MW, Shu VS, Groves U. Safety ofestazolam: the United States clinical experience. American Journal of Medicine 88 (Suppl. 3A): 12S-17S, 1990 Read DJC. A clinical method for assessing the ventilatory reo sponse to carbon dioxide. Australasian Annals of Medicine 16: 20-32, 1967 Rechtschaffen A, Kales A (Eds). A manual of standardized terminology, techniques and scoring systems for sleep stages of human subjects. National Institutes of Health, pub. no. 204, 1968 Spodick DH, Lance VQ. Noninvasive stress testing: methodology for elimination of the phonocardiogram. Circulation 53: 673676, 1976 Tabachnik E, Muller NL, Bryan AC, Levison H. Changes in ventilation and chest wall mechanics during sleep in normal adolescents. Journal of Applied Physiology: Respiratory, Environmental and Exercise Physiology 51: 557-564, 1981a Tabachnik E, Muller NL, Levison H, Bryan AC. Chest wall mechanics and pattern of breathing during sleep in asthmatic adolescents. American Review of Respiratory Disease 124: 269273, 1981b Tobin MJ, Cohn MA, Sackner MA. Breathing abnormalities during sleep. Archives of Internal Medicine 143: 1221-1228, 1983 Vogel GW. Effects ofestazolam on sleep, performance, and memory: a long-term sleep laboratory study of elderly insomniacs. Journal of Clinical Pharmacology, in press, 1992 Widdicombe J. The effects of psychotropic drugs on respiration. In International encyclopedia of pharmacology and therapeutics; section 104: respiratory pharmacology, pp. 129-163, Pergamon Press, Oxford, 1981 Zackon H, Despas PJ, Anthonisen NR. Occlusion pressure responses in asthma and chronic obstructive pulmonary disease. American Review of Respiratory Disease 114: 917-927, 1976

Correspondence and reprints: Dr Martin A. Cohn, Sleep Disorder Center of Southwest Florida, 848 First Avenue North, Naples, Florida 33940, USA.