THE SCIENCE OF DESIGNING ETHICAL CPAP TRIALS

To the Editor :

In their editorial, Karlawish and Pack (1) raised important and difficult issues. We agree that the problems are real but believe their analysis is one-sided, excessively favoring sham CPAP as the placebo of choice. We have been wrestling with these issues for many years and indeed published in this area (2) well before Wright and colleagues (3).

We do not believe there is a perfect placebo for CPAP since a perfect placebo must be absolutely indistinguishable to the patient. However, because offering treatment to patients with the obstructive sleep apnea/hypopnea syndrome (OSAHS) produces such a large placebo effectwhich is seen equally with placebo tablets (4) or sham CPAP (5), both lowering Epworth Sleepiness Score by 2 and significantly improving quality of lifeit is vital that carefully controlled trials are carried out and that patients believe they may be receiving active treatment. Indeed, it is uncontrolled trials that are arguably the least ethical since patients are inconvenienced to produce no interpretable results.

We have adopted the placebo tablet approach, whereas Jenkinson and colleagues (5) used "sham CPAP," which was specifically designed to be ineffective. One could argue that both research groups set out to specifically deceive the treatment population, as we both knew that our "placebo" was ineffective but could not tell the patients this. In this regard, there is no difference between the deception with placebo tablets or sham CPAP. We have been very careful not to mislead patients at any point and have had the full support of our Ethical Advisory Committee. After our two most recent studies (6, 7), we have written to all the patients to inform them that a placebo was used as suggested (1) and none have expressed concern.

Some of the reasons we originally adopted an oral placebo no longer pertain. Specifically, there was an early report that sub-therapeutic CPAP could worsen hypoxemia (8), which has now proven to be unfounded. We thus believe that there is a role for sham CPAP but are concerned that this approach, like the use of placebo tablets, has limitations. Placebos should have no effect on the variable under study, whereas sleeping with an ineffective CPAP machine attached to your face is likely, particularly in mild patients, to perturb sleep and could alter daytime function, artificially impairing the comparators for active treatment below baseline values, thus biasing in favor of active treatment. Our oral placebo will have no effect on the variables under study. The nocturnal discomfort of the mask system presumably explains why Jenkinson and colleagues (5) had lower CPAP use (p = 0.035) in their sham CPAP group than in the active CPAP group. This decreased use occurred despite these patients having very severe OSAHS with a median of thirty 4% desaturations/hr in comparison to a median of seven 4% desaturations/hr in our recent study that stimulated the Editorial (6), and four 4% desaturations/ hr our mild patients (9). We are concerned that less severe patients might show even lower use of sham CPAP compared to active CPAP, indicating that this would not be a true placebo. We designed our original studies to try to determine the threshold of severity above which patients with OSAHS benefit from therapy, and to achieve this it is the study of mild patients that is most important. This was one of the reasons why we chose an oral placebo and remain concerned that sham CPAP might not be ideal in less severe patients. As a side issue, the editorial is surprising in its statement that "data are not as convincing for mild disease" in view of the three controlled trials showing benefit (4, 9, 10), two of which were published in this journal (4, 10).

The ethical issues are difficult and there is no perfect placebo to a CPAP machine; but the important scientific message is that whatever control system is used, all studies show clear benefits in favor of CPAP therapy for symptoms, quality of life, mood, objective sleepiness, driving simulator performance, cognitive function, and nocturnal sleep (2, 4, 5, 7, 9). It is important that skeptics do not lose this message behind a smokescreen of ethical difficulties and dilemmas.

The University of Edinburgh, Scotland, United Kingdom

1. Karlawish JH, Pack AI. Addressing the ethical problems of randomized and placebo-controlled trials of CPAP. Am J Respir Crit Care Med 2001; 163: 809-810 [Free Full Text].

2. Engleman HM, Martin SE, Deary IJ, Douglas NJ. Effect of continuous positive airway pressure treatment on daytime function in sleep apnoea/hypopnoea syndrome. Lancet 1994; 343: 572-575 [Medline].

3. Wright J, Johns R, Watt I, Melville A, Sheldon T. Health effects of obstructive sleep apnoea and the effectiveness of continuous positive airways pressure: a systematic review of the research evidence. BMJ 1997; 314: 851-860 [Abstract/Free Full Text].

4. Engleman HM, Kingshott RN, Wraith PK, Mackay TW, Deary IJ, Douglas NJ. Randomized placebo-controlled crossover trial of continuous positive airway pressure for mild sleep apnea/hypopnea syndrome. Am J Respir Crit Care Med 1999; 159: 461-467 [Abstract/Free Full Text].

5. Jenkinson CR, Daview J, Mullins R, Stradling JR. Comparison of therapeutic and subtherapeutic nasal continuous positive airway pressure for obstructive sleep apnoea: a randomised prospective parallel trial. Lancet 1999; 353: 2100-2105 [Medline].

6. Faccenda JF, Mackay TW, Boon NA, Douglas NJ. Randomized placebo-controlled trial of continuous positive airway pressure on blood pressure in the sleep apnea-hypopnea syndrome. Am J Respir Crit Care Med 2001; 163: 344-348 [Abstract/Free Full Text].

7. McArdle N, Douglas NJ. The effect of continuous positive airway pressure on sleep architecture in the sleep apnea/hypopnea syndrome: a randomised controlled trial (Submitted for publication).

8. Krieger J, Weitzenblum E, Monassier JP, Stoeckel C, Kurtz D. Dangerous hypoxaemia during continuous positive airway pressure treatment of obstructive sleep apnoea. Lancet 1983; 2: 1429-1430 [Medline].

9. Engleman HM, Martin SE, Deary IJ, Douglas NJ. Effect of CPAP therapy on daytime function in patients with mild sleep apnoea/hypopnoea syndrome. Thorax 1997; 52: 114-119 [Abstract].

10. Redline S, Adams N, Strauss ME, Roebuck T, Winters M, Rosenberg C. Improvement of mild sleep-disordered breathing with CPAP compared with conservative therapy. Am J Respir Crit Care Med 1998; 157: 858-865 [Abstract/Free Full Text].

11. Douglas NJ. Systematic review of the efficacy of nasal CPAP. Thorax 1998; 53: 414-415 [Free Full Text].

12. Ballester E, Badia JR, Hernández L, Carrasco E, de Pablo J, Fornas C, Rodriguez-Roisin R, Montserrat JM. Evidence of the effectiveness of continuous positive airway pressure in the treatment of sleep apnea/ hypopnea syndrome. Am J Respir Crit Care Med 1999; 159: 495-501 [Abstract/Free Full Text].

13. Hack M, Davies RJ, Mullins R, Choi SJ, Ramdassingh-Dow S, Jenkinson C, Stradling JR. Randomised prospective parallel trial of therapeutic versus subtherapeutic nasal continuous positive airway pressure on simulated steering performance in patients with obstructive sleep apnoea. Thorax 2000; 55: 224-231 [Abstract/Free Full Text].

From the Authors:

In response to our editorial (1) about the design of a CPAP clinical trial by Faccenda and colleagues (2), Douglas and colleagues suggest that a discussion of the ethics of sham controls in CPAP research obscures the message that all studies show clear benefits in favor of CPAP therapy for a variety of clinical endpoints. To the contrary, we believe that this discussion illuminates important issues.

A major issue is what control should we test CPAP against: a placebo pill or sham CPAP? Douglas and colleagues claim a placebo pill is best because it has "no effect on the variable under study." But what are we really trying to measure when we use a control? The absolute effect of treatment is equal to the effect of active agent minus the effect of control. Arguably, a sham CPAP informs us of the true absolute effect of treatment because both active and control groups experience the same discomforts of the mask and its gadgetry. Clearly, however, data are needed. We repeat our editorial's conclusion (1): the study that needs to be done is a comparison of CPAP, sham CPAP, sham placebo pills, and other viable "controls." Even more useful would be to randomize subjects to true versus false disclosure that they are receiving active CPAP. Of course, after these studies, the investigators should debrief subjects individually about the design and its results.

A second major issue is whether the data for efficacy of treatment of mild to moderate sleep apnea are less well-established than for severe apnea. A recent Pro-Con Editorial in the American Journal of Respiratory and Critical Care Medicine discussed this (3, 4). Dr. Douglas and colleagues cite studies that raise an important question: what is the desired outcome of CPAP therapy? Their own studies (5, 6) show significant improvement using a cross-over design in self-reported sleepiness, a depression score, some aspects of quality of life, but no difference in objectively measured sleepiness. It is unlike the results of sham CPAP-controlled trials in patients with more severe disease where improvements in subjective sleepiness concur with improvements in objective sleepiness (3, 4).

These results raise an important ethical and scientific question: what defines benefit of CPAP when one finds improvements in only subjective but not objective assessment of sleepiness?

University of Pennsylvania, Philadelphia, Pennsylvania

1. Karlawish JHT, Pack AI. Addressing the ethical problems of randomized and placebo-controlled trials of CPAP. Am J Respir Crit Care Med 2001; 163: 809-810 .

2. Faccenda JF, Mackay TW, Boon NA, Douglas NJ. Randomized placebo-controlled trial of continuous positive airway pressure on blood pressure in the sleep apnea-hypopnea syndrome. Am J Respir Crit Care Med 2001; 163: 344-348 .

3. Davies RJO, Stradling JR. The efficacy of nasal continuous positive airway pressure in the treatment of obstructive sleep apnea syndrome is proven. Am J Respir Crit Care Med 2000; 161: 1775-1778 [Free Full Text].

4. Wright J, Sheldon T. The efficacy of nasal continuous positive airway pressure in the treatment of obstructive sleep apnea syndrome is not proven. Am J Respir Crit Care Med 2000; 161: 1776-1778 [Free Full Text].

5. Engelman HM, Martin SE, Deary IJ, Douglas NJ. Effect of CPAP therapy on daytime function in patients with mild sleep apnoea/hypopnoea syndrome. Thorax 1997; 52: 114-119 .

6. Engelman HM, Kingshott RN, Wraith PK, Mackay TW, Deary IJ, Douglas NJ. Randomized placebo-controlled crossover trial of continuous positive airway pressure for mild sleep apnea/hypopnea syndrome. Am J Respir Crit Care Med 1999; 159: 461-467 .