Terbutaline and diaphragm function in chronic obstructive pulmonary disease: A double-blind randomized clinical trial

Terbutaline and diaphragm function in chronic obstructive pulmonary disease: A double-blind randomized clinical trial

Br. J. Dis. Chest (1988) 82, 242 TERBUTALINE AND DIAPHRAGM FUNCTION IN CHRONIC OBSTRUCTIVE PULMONARY DISEASE: A DOUBLEBLIND RANDOMIZED CLINICAL TRIAL...

623KB Sizes 0 Downloads 39 Views

Br. J. Dis. Chest (1988) 82, 242

TERBUTALINE AND DIAPHRAGM FUNCTION IN CHRONIC OBSTRUCTIVE PULMONARY DISEASE: A DOUBLEBLIND RANDOMIZED CLINICAL TRIAL JAMES K. STOLLER* 7, HERBERT P. WIEDEMANNt , JACOB LOKE, PETER SNYDER, JAMES VIRGULTO AND RICHARD A. MATTHAYS Departments of Medicine, Pulmonary Section and Robert Wood Johnson Clinical Scholars Program, Yale University School of Medicine, New Haven, CT, USA.

Summary

We conducted a double-blind, randomized crossover trial to evaluate whether oral terbutaline (2.5 mg orally three times daily for a week) increased the force of diaphragmatic contraction in normocapnic patients with chronic obstructive pulmonary disease. Ten patients with moderate to severe airway obstruction completed the trial. Compared with placebo, terbutaline produced a mean increase of 5.8 cmH,O in peak inspiratory mouth pressure and a mean increase of 5.0 cmH,O in transdiaphragmatic pressure during a maximal inspiratory manoeuvre. These small changes with terbutaline failed to achieve statistical significance. Also, terbutaline failed to alter flow rates (FEV,, ?‘,,,ax50) or patients’ dyspnoea ratings using two separate clinical scales (Pneumoconiosis Research Unit Score and the Modified Dyspnoea Index). Because all observed changes in respiratory muscle strength were small and because the trial had power to detect small changes in inspiratory mouth pressures, we suggest that oral terbutaline at the dose administered in this study has little noteworthy effect on respiratory muscle strength in normocapnic patients with chronic obstructive pulmonary disease. INTRODUCTION As the important role of respiratory muscle fatigue in ventilatory failure has become appreciated (l-3)) recent attention has focused both on muscle training (4, 5) and on pharmacological measures to improve respiratory muscle performance (3, 6, 7). Of the available drugs, methylxanthines (e.g. aminophylline) have been most extensively evalu*Supported in part by Grant No. 721P-41-55511 from the Robert Wood Johnson Foundation, and by a grant from fGeigy Pharmaceuticals. tCurrent address: Department of Pulmonary Medicine, The Cleveland Clinic Foundation, 9500 Euclid Avenue, Cleveland, Ohio 44106, USA. Reprint requests to: James K. Stoller, MD, The Cleveland Clinic Foundation, Department of Pulmonary Medicine, 9500 Euclid Avenue, Cleveland, Ohio 44106, USA.

243

STOLLER,WIEDEMANN,LOKEETAL.:TERBUTAL~NEANDDIAPHRAGMFUNCTION

ated (6, 7), and studies both in animals and patients have shown that aminophylline can strengthen diaphragmatic contraction (6), and can both prevent and reverse diaphragmatic fatigue (7). However, very little attention has been given to promising drugs other than methylxanthines. Beta-adrenergic agonists such as isoprenaline (isoproterenol), salbutamol and terbutaline increase skeletal muscle contraction in vitro (8), and available studies in dogs (9-12) suggest that these drugs can also improve diaphragmatic performance. Specifically, isoprenaline (9) and terbutaline (12) have been shown to increase the force of contraction of fatigued canine diaphragm without altering resting or non-fatigued diaphragm contraction (12). More recently, Howell et al. (11) have shown that intravenous salbutamol produces small increases in diaphragmatic force in dogs with compensated metabolic acidosis. To date, however, no trial of the effects of oral beta-adrenergic agonists on diaphragm contraction in patients with lung disease has been reported. To address this issue and to extend the observations made in animal studies, we performed a randomized clinical trial to assess whether oral terbutaline increases the force of diaphragmatic contraction in normocapnic patients with chronic obstructive pulmonary disease (COPD). Methods Figure 1 shows the design of this study, which was a double-blind, placebo-controlled randomized clinical trial with a treatment crossover. After a baseline evaluation of pulmonary and respiratory muscle function, eligible patients were randomly assigned to receive either 2.5 mg of oral terbutaline three times daily for a week or a comparably administered placebo. At the end of this period, treatment effects were assessed, and each patient then crossed over to a week on the alternate treatment (i.e. either placebo or terbutaline) followed again by assessmentof treatment effects. Use of this crossover design allowed each patient to serve as his own control when the respiratory muscle effects of terbutaline were assessed. Patients were enlisted from the outpatient pulmonary service of the Yale-New Haven Hospital and were eligible for study if they: 1. had clinically stable chronic obstructive pulmonary disease, characterized by forced expiratory volume in one second (FEV,) <80% predicted, and/or a ratio of FEV, to forced vital capacity (FVC)<60%; 2. were not hypercapnic; and 3. were not taking oral terbutaline currently. ---1 week ---TERBUTALINE 2.5 mg po t.i.d. BASELINE

R

----

1 week ---TERBUTALINE 2.5 mg po t.i.d. *

* x

* G

PLACEBO t.i.d. -m-m 1 week

----

----

PLACEBO t.i.d. 1 week

----------*

assess muscle

pulmonary function

and

respiratory

Fig. 1. Design of current randomized crossover trial

s-s-

244

BR. J. DIS.

CHEST:

VOL.

82 NO.

3

Patients were not excluded if they were taking a theophylline preparation and/or inhaled betaadrenergic agonists. However, to isolate the effects of oral terbutaline on the diaphragm, oral theophylline was withheld for at least 24 hours, and inhaled beta-agonists were withheld for at least 6 hours before all measurements were made. Compliance of the study population was monitored by pill count (92% overall compliance) and by ensuring that all patients had taken the study medication on the three mornings of testing. The final study group was 10 men. Of 12 patients originally referred to the study by pulmonary physicians, one patient declined to participate and another patient withdrew after baseline testing but before taking any study medication. Treatment outcomes included measurements of airflow (FEV,, FEVi/FVC and maximum midflow rate, ri max.O) and lung volumes (residual volume, RV; functional residual capacity, FRC; and total lung capacity, TLC). Dyspnoea ratings, using the Pneumoconiosis Research Unit Score (13) and the Modified Dyspnoea Index (14, 15), were obtained and measurements were made of respiratory muscle strength (maximal inspiratory mouth pressure and transdiaphragmatic pressure). All measurements were obtained at baseline, after terbutaline, and after placebo treatment. Spirometric measurements were made using a Collins 9-litre spirometer. Lung volumes were measured by helium dilution using a Collins DS 560 system. Predicted values for FEV, and FVC were based on data from Morris et al. (16), and for FRC and RV from Bates et al. (17) and Goldman and Becklake (18). Integrated inspiratory muscle strength was assessedby the maximal inspiratory mouth pressure (PI,,,), which was measured using the technique of Black and Hyatt (19). PI,,, was recorded as the greatest pressure generated during a series of at least three maximal inspirations from residual volume against a partially occluded mouthpiece. Diaphragm strength was assessedby transdiaphragmatic pressure (Pdi), which was measured using oesophageal and gastric catheters. After topically anaesthetizing each subject’s nose and oropharynx with 2% lignocaine (lidocaine), two latex balloons (each 5 cm long, inflated with 1 cm3of air and connected to 100 cm polyethylene tubing, PE 205) were passed into the mid-oesophagus and stomach to monitor oesophageal and gastric pressures(20). The catheter positions (at approximately 45 cm and 60 cm from the nares) were marked to enable reproducible replacement in the same patient on subsequent testing sessions. Each catheter was connected to a separate pressure transducer (Gould/Statham, PM 6TCk2.5 psid); and transidaphragmatic pressure, which was transduced separately, was determined as the difference between gastric and oesophageal pressures (Gould/Statham, PM 6TC+5 psid). All pressures were recorded relative to atmosphere. With the oesophageal and gastric catheters in place, transdiaphragmatic pressures were measured during a maximal inspiratory effort from functional residual capacity against a mouthpiece with a partially occluded orifice, generating a pressure we called Pdi,,,. All inspiratory manoeuvres were repeated at least three times during each sitting, and the highest pressure generated was recorded for data analysis. Results were analysed with a two-sided paired Student’s t-test, using P
RESULTS Table I shows the baseline characteristics of the 10 participating patients. In aggregate, participants had moderate to severe airflow obstruction (mean FEV,/FVC 45% and mean FEV, 1.47 litres, 52% of predicted). Mean TLC was 108% of predicted and mean RV was 143% of predicted. Only one of the subjects had baseline evidence of reversible airflow obstruction (i.e . 315% increase in FEV, with isoprenaline inhalation) and no patients had resting hypercapnia.

STOLLER,

WIEDEMANN,

LOKE

ET AL.:

TERBUTALINE

AND

DIAPHRAGM

FUNCTION

245

Table I. Baseline patient characteristics of the 10 study

patients Feature

Mean+-SD

Range

Age

63+9 1.47kO.69 51.7k19.6 45.4k8.3 108f21 143+49 75+5 36f5

49-73 0.7-3.0 27-85 35-59 65-151 79-234 65-83 2641

FEV, (litres) % Predicted FEV, FEVJFVC (%) % Predicted TLC % Predicted RV Po2 (mmHg) Pco2 (mmHg)

FEV,, forced expiratory volume in one second; TLC, total lung capacity; RV, residual volume; PO*, arterial oxygen tension breathing room air; Pco,, arterial carbon dioxide tension breathing room air.

The results of serial pulmonary function tests and dyspnoea ratings are shown in Table II. For each measurement, group mean differences (terbutaline vs. placebo) were clinically unimpressive and failed to achieve statistical significance (P~0.05). Mean FRC after placebo was 120 ml higher than after treatment with terbutaline, a difference that also failed to achieve statistical significance (P=O.22). Resting end-expiratory diaphragm position, therefore,

did not vary notably over the course of the trial. Dyspnoea ratings with

Table II. Pulmonary function test and dyspnoea rating results Pulmonary tests

function

Pretreatment baseline

After placebo

After terbutaline

course

course

Statistically significant change? (terbutaline vs. placebo)

FEV, (litres)

1.47f0.69*

1.35k0.66

% Predicted FEVl FVC (litres) % Predicted FVC

51.7k19.6 3.26f 1.26 81.7k28.7 45.4k8.3 0.68f0.42 4.53k1.10

47.7k20.8 2.96f1.31 74.7k32.5 46.3k9.06 0.67f0.41 4.39f1.01

1.47f0.67 52.Ok20.5 3.14f1.33 79.2k31.7 48.1f10.3 0.65kO.41 4.27rt0.87

No No No No No No No

2.821.5 6.0f3.2

3.Ok1.4 6.3k3.7

2.8kl.O 5.3k3.4

No No

FEVI/FVC (%) (litres/min) v maxS0 FRC (litres) Dyspnoea

scores

Pneumoconiosis Research unit score Dyspnoea index

FEV1, forced expiratory volume in one second; FVC, forced vital capacity; vrnaxSO, maximum flow rate at 50% of vital capacity; FRC, functional residual capacity. *Meanfso.

246

BR. J. DIS.

CHEST:

VOL.

82 NO.

3

-7”

130 120 110

100

PI MAX (cm

H,O)

90 80 70 00 50 40

““BASELINE MeantsD (cm H,O)

71.3

f 20

TERBUTALINE 00.0

PLACEBO

f 23

75.0

f 26

5.0 t 14 p*O.24.

paired

t

Fig. 2. Effects of terbutaline vs. placebo on integrated inspiratory muscle strength (PI,,,, N=9)

both scoring indices suggested that patients were moderately impaired functionally and these scores did not change with terbutaline treatment versus placebo. Figure 2 shows the results of integrated inspiratory muscle strength measurements, or PI max,which were obtained in nine patients. (One patient declined the measurement during a testing session.) Mean PI,,, at baseline, on terbutaline, and on placebo were 71.3, 80.8, and 75.0 cmH,O, respectively. Thus, although several patients did have higher inspiratory pressures on terbutaline than on placebo, the mean individual increase with terbutaline, 5.8 cmH,O, was small and failed to achieve statistical significance.

140 130 120

Pdi max

(cm H,O)

110 100 90 90 70 00 50 ,

40

BASELINE Momi km H,W

105.7 9 30

,

TERBUTALH’JE 102.7 t

PLACEBO

35

97.7

t 35

5.0 t 14 pao.41. peeed t

Fig. 3. Effects of terbutaline vs. placebo on diaphragm strength (Pdi,,,, N=6)

STOLLER,

WIEDEMANN,

LOKE

ET

AL.:

TERBUTALINE

AND

DIAPHRAGM

FUNCTION

247

Results of Pdi,,, measurements are shown in Fig. 3. Complete Pdi,,, data were available in six patients, as three patients declined serial oesophageal intubation, and catheters could not be passed in a fourth patient who had been treated for oesophageal carcinoma. Mean Pdi,,, at baseline, on terbutaline, and on placebo were 105.7,102.7 and 97.7 cmH,O, respectively. Compared with placebo, the mean increase in Pdi,,, with terbutaline was 5 cmH,O and this change in transdiaphragmatic pressure with terbutaline was both small and statistically non-significant. Thus, terbutaline did not appear to strengthen diaphragm contraction in this study. Because a failure to detect a difference between terbutaline and placebo does not, by itself, prove that no difference exists, we calculated the statistical power of this trial (21), which is its ability to detect and ascribe statistical significance to noteworthy changes in treatment outcome. This trial had 80% power to detect an increase in PI,,, as small as 12 cmH,O and at least 90% power to detect a 20 cm pressure increase. The trial had less power to detect comparable changes in Pdi,,,, in part because complete measurements were available in fewer patients. However, assuming a crossover trial with 10 patients and the observed variance of Pdi,,, measurements, the trial had 80% power to detect a Pdi,,, change of 20 cmH,O, and 22% power to detect a change of 5 cmH,O. Thus, although the Pdi,,, measurements may have been subject to a type II error (i.e. failure to detect a true difference between measurements on terbutaline versus placebo), a type II error was unlikely to affect the conclusion that terbutaline does not increase PI,,,. DISCUSSION The main finding of this study is that terbutaline administered orally as 2.5 mg three times daily had little effect on the force of diaphragm contraction in normocapnic patients with COPD. Although this is the first trial to examine the effect of a sympathomimetic drug on diaphragm contractility in COPD patients, the results are consistent with findings in animal studies (9,10,12) and with results of a preliminary report in normal human subjects (22). Aubier et al. (12) administered terbutaline (0.5 mg i.v.) to dogs and observed no change in transdiaphragmatic pressure (Pdi) before diaphragm fatigue was induced by supramaximal phrenic nerve stimulation. However, these investigators noted up to a 37% rise in Pdi when terbutaline was given after diaphragmatic fatigue had developed in these dogs. Furthermore, this inotropic effect of terbutaline on fatigued diaphragm was blocked by pretreating the dogs with propranolol(1 mg i.v.), suggesting the Pdi enhancement was at least partly mediated by beta-adrenergic receptors, which are found in abundance in the diaphragm (12). Separate studies in dogs by Howell et al. (9,23) showed that isoprenaline (up to 20pl/min i.v.) did strengthen diaphragm contraction following fatigue induction by supramaximal phrenic nerve stimulation (9) but not in dogs with decreased diaphragm contractility due to hypercapnia (23). Thus, despite demonstrable electrophysiological effects that make beta-adrenergic agonists attractive diaphragmatic inotropes (8,23,24), improved contractility with these drugs has been observed inconsistently and has not been observed in animals without diaphragmatic fatigue. Also, in a preliminary report on four normal human subjects, Lanigan et al. (22) noted no increase in maximal inspiratory or expiratory mouth pressures after single terbutaline doses of 7.5-15 mg. Our findings extend these results to a common clinical setting, namely patients with stable COPD. Reasons for these varying effects of beta-adrenergic agonists on diaphragm contraction

248

BR.

.I. DIS.

CHEST:

VOL.

82 NO.

3

are not well understood, but may concern the differential effects of sympathomimetic drugs on slow versus fast twitch muscle fibres (23, 24), both of which are present in diaphragm muscle. Specifically, in non-fatigued slow twitch muscles, sympathomimetics decrease twitch duration and amplitude, whereas non-fatigued fast twitch fibres behave in an opposite fashion (24). It is possible that competing muscle fibre effects of these drugs in non-fatigued diaphragm may produce no net increase in diaphragmatic contraction. Further research will be needed to establish whether the diaphragmatic effects of betaadrenergic agonists in hypercapnic patients or in patients with induced diaphragm fatigue differ from the effects of these agents in human non-fatigued diaphragm. Whenever a study fails to detect a difference between compared treatments, it is important to consider both statistical and methodological reasons that true differences may have been overlooked. Statistically, the current research had at least an 80% chance to ascribe significance to increases in PI,,, (terbutaline versus placebo) as small as 12 cmH,O. Because observed rises in PI,,, were <6 cmH,O with terbutaline and other changes in respiratory muscle strength measurements were even smaller, it is unlikely that a true effect of terbutaline was overlooked. Changes in resting end-expiratory diaphragm position could mask a true inotropic effect of terbutaline if the subjects’ diaphragms consistently were lower and flatter after terbutaline treatment than after placebo. However, FRC was measured serially in this study and mean changes in FRC (terbutaline versus placebo) were both clinically unimpressive and statistically non-significant. Thus, altered diaphragm position was unlikely to account for the absence of observed change in diaphragm strength with terbutaline. ‘Compliance bias’ (25) could aso mask a true effect of terbutaline if patients failed to take the prescribed study medication and if this non-compliance was not appreciated. However, we defended against compliance bias by assuring that overall compliance with study medications was high (92%) and that each patient took his study medication on each morning of testing. It is also possible that co-therapy with methylxanthines could mask an inotropic effect of terbutaline, because aminophylline has been shown to increase the strength of both resting and fatigued diaphragm contraction in patients with COPD (6, 7). To ensure that other diaphragm inotropes were ‘washed out’ before study measurements were made, subjects were instructed not to take their methylxanthine preparation for at least 24 hours before each testing session. Each patient acknowledged withholding aminophylline before each testing session, although we cannot exclude inaccurate reporting or incomplete washout, because theophylline levels were not measured in this study. We also cannot rule out the possibility that patients were aware of taking terbutaline (vs. placebo) because of its sympathomimetic activity. However, even this possibility would not be expected to bias this study because examiners remained blind to the study drug and consistently assured patients’ maximal inspiratory efforts during each testing session. Finally, it is possible that the administered dose of terbutaline (2.5 mg by mouth three times daily for a week) was too low and that a higher dose would have improved respiratory muscle function, as was seen when dogs were given 0.5 mg of intravenous terbutaline (12). Although the issue requires further study, preliminary experience in human subjects (22) suggests that even higher doses would not be effective. Also, the likelihood of intolerance to higher doses of terbutaline might diminish the clinical impact of such a finding.

STOLLER,

WIEDEMANN,

LOKE

ET AL.:

TERBUTALINE

AND

DIAPHRAGM

FUNCTION

249

We believe that these points provide compelling evidence that terbutaline is unlikely to be useful as a diaphragm inotrope in normocapnic patients with COPD. However, further investigation will be required to define whether terbutaline improves contractility in patients with diaphragmatic fatigue, as in animals (12), or whether terbutaline given in higher doses or by other routes will show positive inotropic effects on the diaphragm. ADDENDUM Since submitting this manuscript, Javaheri et al. (Am Rev Resp Dis 1988;137:197-201) have reported on the diaphragmatic effects of oral albuterol in five normal subjects. The study shows that a dose of albuterol (4 mg three times daily for 3 days) produces no significant increase in the strength of fatigued or non-fatigued diaphragm in normal human subjects. REFERENCES 1. Grassino A, Macklem PT. Respiratory muscle fatigue and ventilatory failure. Ann Rev Med 1984;35:625-47. 2. Roussos CH. Function and fatigue of respiratory muscles. Chest 1986;88(suppl):S24-32.

3. Sharp JT. Therapeutic considerations in respiratory muscle function. Chest 1986;88(suppl):S118-23. 4. Aldrich TK, Karpel JP. Inspiratory muscle resistive training in respiratory failure. Am Rev Resp Dis 1985;131:461-2.

5. Belman MJ. Respiratory failure treated by ventilatory muscle training (VMT): a report of two cases. Eur J Resp Dis 1981;62:391-5. 6. Aubier M, DeTroyer A, Sampson M, Macklem PT, Roussos CH. Aminophylline improves diaphragmatic contractility. New Engl J Med 1981;305:249-52. 7. Aubier M, Roussos CH. Effect of theophylline on respiratory muscle function. Chest 1985;88(suppl):S91-7. 8. Goffart M, Ritchie JM. the effect of adrenaline on the contractility of mammalian skeletal muscle. J Physiol (Lond) 1952;116:357-71. 9. Howell S, Roussos CH. Isoproterenol and aminophylline improve contractility of fatigued canine diaphragm. Am Rev Resp Dis 1984;129:11&24. 10. Howell S, Fitzgerald RS, Roussos CH. Effect of neostigmine and salbutamol on diaphragmatic fatigue. Resp Physiol 1985;62:15-29. 11. Howell S, Fitzgerald RS, Roussos CH. Effects of aminophylline and salbutamol on diaphragmatic force during compensated metabolic acidosis. Am Rev Resp Dis 1986;133:407-13. 12. Aubier M, Viires N, Murciano D, Medrano G, Lecocguic Y, Pariente R. Effects and mechanisms of action of terbutaline on diaphragmatic contractility and fatigue. J Appl Physiol: Resp Environ Exercise Physiol1984;56:922-9. 13. Schilling RSF, Hughes JPW, Dingwall-Fordyce I. Disagreement between observers in an epidemiological study of respiratory disease. Br Med J 1955$:65-g. 14. Mahler DA, Weinberg DH, Wells CK, Feinstein AR. The measurement of dyspnea: contents, interobserver agreement, and physiologic correlates of new clinical indexes. Chest 1984;8.5:751-8.

15. Stoller JK, Ferranti R, Feinstein AR. Further specification of a new clinical index for dyspnea. Am Rev Resp Dis 1986;134:1129-34.

16. Morris JF, Koski A, Johnson LC. Spirometric standards for healthy non-smoking adults. Am Rev Resp Dis 1971;103:57-67.

250

BR.

.I. DIS.

CHEST:

VOL.

82 NO.

3

17. Bates DV, Macklem PT, Christie RV. Respiratory function in disease. 2nd ed. Philadelphia: W. B. Saunders, 1971. 18. Goldman HI, Becklake MR. Respiratory function tests: normal values at median altitudes and the prediction of normal results. Am Rev Tuberc 1959;79:457-67. 19. Black LF, Hyatt RE. Maximal respiratory pressures: normal values and relationship to age and sex. Am Rev Resp Dis 1969;99:696--702.

20. Loke J, Mahler D, Virgulto JA. Respiratory muscle fatigue after marathon running. J Appl Physiol1982;52:821-4.

21. Cohen J. Statistical power analysis for the behavioural sciences. New York: Academic Press, 1969. 22. Lanigan C, Borzone G, Brophy C, Moxham J. The effects of terbutaline on respiratory and limb muscle function in man. Bull Eur Physiopath Resp 1986;22(suppl8):1638. 23. Howell S, Fitzgerald RS, Roussos CH. Effects of aminophylline, isoproterenol, and neostigmine on hypercapnic depression of diaphragmatic contractility. Am Rev Resp Dis 1985;132:241-7. 24. Boloman WC, Mott MW. Actions of sympathomimetic amines and their antagonists on skeletal muscle. Pharmacol Rev 196921~27-72. 25. Feinstein AR. Clinical epidemiology. Philadelphia: W. B. Saunders Corp., 1985:302-3.

Date accepted 4 August

I987