Evidence is mounting that automatic CPAP can simplify the diagnosis process and regulate optimal air pressure by adjusting to each OSA patient’s specific needs.

Automatic continuous positive airway pressure (CPAP) therapy is most likely the next-generation answer to the question of why fixed pressure CPAP for the treatment of obstructive sleep apnea is not uniformly successful in all patients. In fixed CPAP therapy, the titration pressure is chosen to cover the worst situation on a single night of observation. It assumes that overtreatment has no ill effects and that an observation on a single night can be extrapolated to a prescription for every night for several years. This is inconsistent with what we have learned in human biology and feedback control mechanisms of human homeostasis. Although clinical experience and published studies confirm that fixed pressure CPAP is effective treatment for many patients with obstructive sleep apnea, those who work in the field recognize that many challenges remain. Could the failure to raise or lower CPAP pressure to maintain airway patency contribute to unanswered questions in CPAP treatment of obstructive sleep apnea? Why don’t all hypertensive patients show improvement in their hypertension? Why is compliance with CPAP use not uniformly good? When should CPAP be discontinued? When should CPAP begin? All of these issues lack complete answers, although opinions abound.

Perhaps a historical perspective is important. Just as our understanding of the spectrum of sleep-disordered breathing has grown over the past 2 decades, so has our understanding of the mechanisms of the disease process. Questions remain as to what physiologic parameters best reflect the pathophysiology of sleep-disordered breathing. Are there cardiovascular parameters, or neuropsychiatric parameters, or some combination of both that best predict successful outcomes? Without well-defined outcome markers, it is difficult to assess what is optimal or adequate treatment. If we think back about the history of hypertension and diabetes, we see parallels that can probably teach us in our evolving understanding of sleep-disordered breathing. Decades of work were required to establish appropriate treatment guidelines in these diseases and those guidelines continue to evolve. So it is likely to be in sleep-disordered breathing.1

For example, any night technician who does CPAP titration studies for the treatment of obstructive sleep apnea patients will be quick to tell you that they observe change in CPAP pressure requirements based on body position and stage of sleep. This is well documented as is the influence of upper airway infections and thyroid function. These latter parameters, however, are likely to change, not over the course of a single night, but over the course of days, weeks, and months. Clinically, little is done to track changes in CPAP pressure requirements as they occur in our patients. What studies that do exist suggest that CPAP pressure requirements vary from night to night in many patients, though not all. There are studies that reflect stability of CPAP pressure needs over seasons and years. Thus, from a single night CPAP titration study, treatment is initiated, and initial clinical improvement may be seen. Subsequently, we may end up overtreating or undertreating our patients in the months and years ahead. It is not common practice in most communities to repeat CPAP titration studies on a regular basis. There really are no data to guide us.2-5

Again, looking to the lessons from hypertension and diabetes, what was once thought to be acceptable therapy was recognized as being better than in the past but not as good when new information and outcome markers became available, such as hemoglobin A1c in the assessment of the adequacy of treatment of diabetes. We may have some patients who faithfully use their CPAP every night but are being undertreated and are continuing to have some degree of pathophysiologic adverse effect. There may be others who are thwarted in successful CPAP use because of unnecessarily high CPAP pressures. Lower pressures might improve long-term compliance, thereby improving clinical outcomes.6,7

With this perspective, a study group at Sleep Consultants Inc undertook the initial development and evaluation of automatic CPAP in the late 1980s. We began having clinical experience with automatic CPAP in the early 1990s and were fortunate to learn much from early prototype models of automatic CPAP. Later, we had the opportunity to assess automatic CPAP in the home and compare it to fixed pressure CPAP. The early experiences will be summarized, and more detailed information regarding automatic CPAP in home use at the time of diagnosis of new sleep apnea patients will be discussed.

Our development of automatic CPAP was based on the observation that snoring frequently preceded many obstructive events and was usually present at the termination of all events. A laryngeal microphone had been part of nocturnal polysomnography for many years and thus detecting associated arousal was easily done. Many times we recognized that the laryngeal microphone detected pressure changes (sound) that was below the level of hearing of our technicians. We translated this observation to sensing airway sound at the mask and found again that pharyngeal wall vibration (snoring) could be identified. The airway pressure could be filtered and amplified in such a way as to maximize the identification of the pharyngeal wall vibration. This energy signature was sufficiently consistent between patients to allow development of an algorithm control to command CPAP pressure adjustments so as to minimize the degree of pharyngeal wall vibration. This algorithm had several parameters, which could be varied to improve the sensitivity and specificity of pressure adjustments both up and down as appropriate through the night.8,9

The initial bench prototype of automatic CPAP was then used as part of CPAP titration with direct observation of outcomes by the night technician. After hundreds of study nights, adjustments in the detection and treatment algorithms improved the outcome of the single night such that, in most nights, the technician only had to observe the patient while automatic CPAP provided pressure adjustment. An algorithm required a 2-cm rise in CPAP pressure with repeated detection of pharyngeal wall vibration (snoring ), and after a defined period of time, a 1-cm lowering of CPAP pressure was allowed if no indication of snoring was detected.10-12

Later prototype models of automatic CPAP became available and allowed comparison of a CPAP titration night with a hands-off automatic titration night. Patients were randomized as to which they received first and the results were published in Sleep in 1998. Both the technician CPAP titration and automatic CPAP titration were effective treatment according to the parameters of the diagnostic night of study. The titration CPAP pressure and the peak automatic CPAP pressure were not significantly different, but automatic CPAP did provide an approximate 30 percent reduction in the mean CPAP pressure required to maintain airway patency through the night for the patient.13

These prototype automatic CPAP units were then utilized in the home over an extended period of time in a limited number of patients. The patients reported treatment as being effective, and data collected from the automatic CPAP taught us that the peak CPAP pressure requirement differed from night to night and the mean pressure of CPAP also varied from night to night. Clinically, the patients were considered to have had a good clinical response to CPAP as assessed in the usual clinical fashion. Interesting additional information was learned. The time and hours of use from night to night over weeks, months, and seasons led to additional insight as to patients’ use, and lack thereof, of their nasal CPAP treatment. No adverse effects were seen in these knowledgeable CPAP-using patients with obstructive sleep apnea.14

With funding from the National Institutes of Health, automatic CPAP was compared to traditional CPAP in newly diagnosed obstructive sleep apnea patients. This comparison was made in the patients’ homes using a randomized crossover design. Subjects were entered into the study once they had successfully used CPAP for approximately 3 weeks after initial diagnosis of clinically significant obstructive sleep apnea. The 6 weeks of therapy using either fixed pressure CPAP or automatic CPAP was compared. No significant difference was seen in machine use or in the clinical response to treatment as subjectively measured by Epworth Sleepiness Score or objectively analyzed by nocturnal oximetry. Although these results were most encouraging and group data consistently showed no significant difference between fixed pressure and automatic pressure, there was always an occasional individual patient who needed additional attention. A question was raised as to whether choosing only successful CPAP patients with obstructive sleep apnea might bias the experience of automatic CPAP if used in a more general population.15,16

At the 1998 American Thoracic Society annual meeting in Chicago, the comparison of continuous positive airway pressure to automatic CPAP in the diagnosis and treatment of sleep apnea patients was presented. Automatic CPAP units had been used to study patients in their homes prior to their nocturnal polysomnographic study and CPAP titration study at the sleep laboratory. Twenty-seven untreated patients who had been referred to Sleep Consultants Inc for symptoms of obstructive sleep apnea were studied by nocturnal home oximetry and placed on automatic CPAP for 1 to 2 weeks prior to being studied by nocturnal polysomnography in the sleep laboratory. The patients’ orientation to automatic CPAP was similar to that provided for all patients being introduced to nasal CPAP. Nocturnal oximetry was repeated while the patient used automatic CPAP prior to the nocturnal polysomnogram, and experience with automatic CPAP prior to diagnosis could then be compared with both the nocturnal polysomnogram and the subsequent CPAP titration study. As expected, statistically significant differences were seen in several parameters from pretreatment to 1 to 2 weeks on automatic CPAP. Epworth Sleepiness Score, low oxygen saturation, percent time spent with less than 90 percent oxygen saturation, and the 3 percent desaturation event index all experienced statistically significant decreases.

In this study group, a comparison of the approximate 1 week of automatic CPAP data with the CPAP titration study showed no significant difference between the fixed titration pressure found in the laboratory and the peak automatic pressure found in the home. There were again statistically significant differences in the fixed CPAP pressure in the laboratory and the mean automatic CPAP pressure in the home. This allowed, again, approximately 30 percent reduction in mean CPAP pressure. The automatic CPAP pressure results demonstrated change throughout each night, with variability between the nights, in the patients. In 19 of the 27 patients with clinically significant obstructive sleep apnea, the nocturnal oximetry was normalized with automatic CPAP showing no drop below 90 percent and a 3 percent desaturation event index of less than 5.17

The suspicion exists that these patients might well have been adequately treated with automatic CPAP and could avoid the time spent in the sleep laboratory. To implement this, we would be assuming the reliability of a normalized Epworth Sleepiness Scale and nocturnal oximetry as adequate outcome parameters of obstructive sleep apnea patients. Unfor-tunately, though important outcome parameters, they are not yet adequately validated to implement the strategy of patients avoiding laboratory studies. The additional compliance information made available by an automatic CPAP unit may, however, help guide the clinician toward successful treatment, as many models of CPAP do not record pressure and time of use.18

Further work with automatic CPAP and the treatment of obstructive sleep apnea is ongoing around the world. As experience builds and results are reported, it is clear that automatic CPAP will find its place in the treatment of obstructive sleep apnea. Exactly what that role will be is actively being defined. In the sleep laboratory, automatic CPAP will frequently simplify the CPAP titration study of the obstructive sleep apnea patient. In the home, automatic CPAP can now be used in treating some obstructive sleep apnea patients. For most of these patients successfully treated with automatic CPAP, an approximate 30 percent reduction in mean CPAP pressure will be seen. The advanced technology in automatic CPAP will allow sleep clinicians to monitor the time and hours of use in their patients over weeks, months, or years. This increased documentation of therapy may have a further beneficial effect for sleep apnea patients whose jobs require a high level of alertness and for whom licensing agents and employers require documentation of adequate therapy. Before automatic CPAP is likely to find a place in a diagnostic mode, we will need better parameters of successful patient outcome. With continued understanding of the impact of sleep disordered breathing and the beneficial effect of therapy, automatic CPAP is surely to be the answer to the clinical challenge of many obstructive sleep apnea patients.

John R. Burk, MD, is co-medical director of Sleep Consultants Inc, and a staff physician at Pulmonary Consultants of Texas, PA, both in Fort Worth, Tex.


Special thanks to the members of the study group on whose work this report is based: Edgar A. Lucas, PhD, ACP; Khosrow Behbehani, PhD; and Sandra L. Knaur, RN, C-ANP.


1. American Thoracic Society Board of Directors. Statement on health outcomes research in sleep apnea. Am J Respir Crit Care Med. 1998;157:335-341.

2. Lofaso F, Lorino AM, Duizabo D, et al. Evaluation of an auto-CPAP device based on snoring detection. Eur Respir J. 1996;9:1795-1800.

3. Konerman M, Sanner BM, Vyleta M, et al. Use of conventional and self-adjusting nasal continuous positive airway pressure for treatment of severe obstructive sleep apnea syndrome. A comparative study. Chest. 1998;113:714-718.

4. Boudewyns A, Grilher-Lanoir V, Willemen MJ, De Cock WA, Van de Heyning PH, De Backer WA. Two months follow up of auto-CPAP treatment in patients with obstructive sleep apnea. Thorax. 1999;54:147-149.

5. Lloberes P, Ballester E, Montserrat JM, et al. Comparison of manual and automatic CPAP titration in patients with sleep apnea/hypopnea syndrome. Am J Respir Crit Care Med. 1996;154:1755-1758.

6. Indications and standards for use of nasal continuous positive airway pressure (CPAP) in sleep apnea syndromes. Am J Respir Crit Care Med. 1994;150:1738-1745.

7. Loube DI, Gay PC, Strohl KP, Pack AI, White DP, Collop NA. Indications for positive airway pressure treatment of adult obstructive sleep apnea patients. A consensus statement. Chest. 1999;115:863-866.

8. Yen FC, Behbehani K, Lucas EA, Burk JR, Axe JR. A noninvasive technique for detecting obstructive and central sleep apnea. IEEE Trans Biomed Eng. 1997;44:1262-1268.

9. Behbehani KR, Yen FC, Burk JR, Lucas EA, Axe JR. Automatic control of airway pressure for treatment of obstructive sleep apnea. IEEE Trans Biomed Eng. 1995; 42:1007-1016.

10. Burk JR, Lucas EA, Axe JR, Behbehani K, Yen FC. Auto-CPAP in the initial treatment of obstructive sleep apnea (OSA) patients [abstract]. Presented at: ALA/ATS International Conference; May 16-19, 1993; San Francisco.

11. Burk JR, Lucas EA, Axe JR, Behbehani K, Yen FC. Comparison of traditional and automatic CPAP in the treatment of obstructive sleep apnea [abstract]. Presented at: World Congress on Sleep Apnea; October 1-4, 1994; San Francisco.

12. Burk JR, Lucas EA, Axe JR, Behbehani K, Yen FC. Auto-CPAP in the treatment of CPAP-using obstructive sleep apnea patients [poster]. Presented at: 61st Annual International Scientific Assembly, American College of Chest Physicians; October 29-November 2, 1995; New York.

13. Behbehani K, Yen FC, Lucas EA, Burk JR. A sleep laboratory evaluation of an automatic positive airway pressure system for treatment of obstructive sleep apnea. Sleep. 1998;21:485-491.

14. Lucas EA, Burk JR, Axe JR, Behbehani K, Yen FC. Auto-CPAP in the treatment of CPAP using obstructive sleep apnea patients. Presented at: World Federation of Sleep Research Societies; September 1995; Nassau, The Bahamas.

15. Behbehani KR, Becker PM, Burk JR, et al. Hours of use is increased and mean therapeutic pressure is reduced by auto CPAP (APAP) therapy in CPAP treated obstructive sleep apnea patients [abstract]. In: Chase MH, Rosenthal LD, O’Connor C, eds. Sleep Research. 1997;26:326.

16. Smith ML, Becker PM, Behbehani K, et al. Comparison of auto CPAP (APAP) and fixed CPAP therapy on nocturnal blood pressure changes in patients with obstructive sleep apnea [abstract]. In: Chase MH, Rosenthal LD, O’Connor C, eds. Sleep Research, 1997;26:511.

17. Burk JR, Bradshaw CM, Knaur SL, Lucas EA, Wexler D. Comparison of continuous positive airway pressure (CPAP) to auto CPAP (APAP) in the diagnosis and treatment of obstructive sleep apnea (OSA) [abstract]. Presented at: ALA/ATS International Conference; April 24-29, 1998; Chicago.

19. Levy P, Pepin JL. Auto-CPAP: an effective and low-cost procedure in the management of OSAS? Eur Respir J. 1998;12:753-755.