Selecting appropriate auto-adjust nasal PAP devices is important as it may impact the efficacy of treatment.

The first commercially available auto-adjust positive airway pressure (PAP) device was introduced in France in the mid 1990s. The goal was to provide a PAP system that would adjust pressure based on the patient’s requirements. Patient compliance was problematic and the developers of the system thought that if the mean pressure were lower, the patients would use the PAP for a longer period of time.1 Simultaneously, other companies were also developing their own autotitration algorithms and conducting clinical trials. Today, it is estimated that approximately 10% of PAP units sold in the United States are auto-adjust systems.

There is debate regarding the use and benefits of an auto-adjust system. The two most prevalent reasons to use an auto-adjust system are to screen patients in need of PAP and to titrate the device to a therapeutic pressure.2-4 Other reasons to use this type of system are to improve patient usage, improve effectiveness of treatment, reduce pressure-related side effects, and, at this time, monitor compliance and efficacy using the memory capability in auto-adjust units. Research has shown that an auto-adjust unit in diagnostic mode can screen patients for the apnea-hypopnea index (AHI), and some institutions use this system as an alternative to full polysomnography (PSG).5 The potential to get a false-negative with this approach is a concern since there is no way of determining if the patient was actually sleeping, although studies have demonstrated a close correlation with AHI during PSG with some auto-adjust devices.6 The ability to titrate PAP pressure is a benefit, and sleep laboratories use auto-adjust units with split night studies to determine a pressure prescription. The ability to respond to varying breathing conditions may be an advantage in the home with patient needs changing both during the night relating to position, alcohol intake, and sleep stage, and from night-to-night with changes in weight and lifestyle.

There are no standards or specifications that auto-adjust PAP units must meet to be introduced to the market. Each manufacturer determines what breathing patterns its product will respond to and what parameters will control the response. Each product is tested prior to the introduction of the unit and some degree of patient improvement is usually claimed. The clinicians who use and prescribe auto-adjust units may believe that products are similar, although these units are all based on proprietary algorithms.

Figure 1. Test setup.

Auto-Adjust Signals
Auto-adjust units respond to several breathing patterns. Not all systems respond to all patterns, and each manufacturer determines which is most important based on the provision of effective patient therapy, improving patient comfort, and, hence, usage and their perception of market requirements. The following are examples of patterns that may be measured and responded to.

• Apnea—complete or almost complete cessation of airflow for 10 seconds or more.7

• Obstructive Hypopnea—abnormal respiratory event lasting at least 10 seconds with at least a 30% reduction in thorocoabdominal movement of airflow as compared to a baseline, and with at least a 4% oxygen desaturation,7 or an arousal.

• Snoring—thought to be an early indicator of an obstruction, yet is not the best stand-alone signal. Snoring, though symptomatic, may not need therapeutic intervention. There has not been a standard defined for the snoring signal. Device response to snoring was not tested in this study.

• Flow Limitation—has been identified by several groups in many different forms.8,9 There is some data that suggest that flow limitation usually precedes more significant upper airway obstruction.9 This would imply that titration based on flow limitation would be preferred.

• Normal Breathing—has no standard definition.

• Combinations—including all of the above.

All units respond to a signal generated by the patient’s breathing pattern. Understanding what causes the auto-adjust unit to change pressure in response to the signal is important. Clinicians should be aware of the following:

  • what auto-adjusting units respond to
  • how they respond with pressure changes
  • limitations of the unit
  • compensatory mechanisms

This study was designed to simulate, as accurately as possible, a patient’s breathing pattern resulting from sleep-related upper airway obstruction. Identical breathing patterns were simulated and applied to each device to determine how each device would respond. This study did not intend to determine the degree of corrective action provided by the device, or represent all possible combinations of breathing patterns that might be encountered with auto-adjust units.

Figure 2. Waveforms that represent common SDB patterns.

Four commercially available auto-adjust PAP products were challenged with a mechanical breathing simulator. Breathing patterns derived from nasal pressure monitoring were digitally recorded from patients undergoing formal polysomnography. Three variations of waveforms, representing common sleep-disordered breathing (SDB) patterns, were selected from the recorded data (Figure 2). The waveforms consisted of flow limited or flattened waveforms, hypopnea, and apnea. Each of these three waveforms was transformed into a stimulus file to be “played back” by the breathing simulator. A normal breathing pattern was also created. For replication by the breathing simulator, patterns were scaled so that the normal tidal volume was approximately 500 mL.

A period of normal breathing was applied for 30 minutes, followed by 30 minutes of the disordered breathing pattern, and then a return to normal breathing. Pressure at the mask was filtered to remove intra-breath variations and recorded every 15 seconds. Each auto-adjust unit was tested in this way in a separate test for each disordered breathing pattern. The unit was turned off for a period of at least 1 minute before starting each test, to reset its algorithm.

Note that the patterns shown are typical of the approximately 8 minutes of data used for each pattern, and that each 8-minute pattern was “looped” to provide 30 minutes of breathing for each trial.

Figure 3. Unit A.

Figure 4. Unit B.

Figure 5. Unit C.

Figure 6. Unit D.

Figures 3 through 6 (Figures 5 and 6 are on page 45) show the response of each unit to the different stimulus files. On each graph is superimposed the results of three tests, each one using “flow limitation,” “hypopnea,” and “apnea” as the disordered breathing component. The maximum pressure delivered for a particular pattern of SDB varied significantly between the units.

It is important to note that the primary objective of this research was to compare each unit’s response to the presence of a variety of abnormal breathing patterns and response to a transition from abnormal to normal breathing; however, this project did not test the units’ ability to correct the patient’s SDB.

When initiating the use of auto-adjust PAP technology, in either a diagnostic or therapeutic application, it is logical to assume that the clinician will anticipate that the auto-adjust will perform any or all of the following functions:

  • The unit will deliver pressures above the minimum setting only when necessary.
  • Changes in the delivery of pressure will take place gradually to minimize potential sleep disruption.
  • The unit will effectively compensate for mask leaks that occur.
  • The unit will detect variations of sleep-related upper airway obstruction.
  • The unit will detect the transition from abnormal to normal breathing and reduce pressures accordingly.

Only Unit A responded to “flow limited” breathing. Units B and D are reported to primarily respond to the presence of snoring. It was not surprising, then, that these units did not respond to flow limitation or hypopnea. Unit B, however, did respond aggressively to the apnea pattern (Unit D did not respond).

Unit C seems to have a “testing” algorithm, whereby the pressure is intermittently increased to see if a change in breathing occurs. These pressure changes were seen during both the normal and flow limited breathing patterns. This device is reported to respond to the presence of flow limitation; however, it did not respond to the particular “flow limited” breathing pattern tested here.

The units varied in their response to a return to normal breathing. Unit A returned to the baseline pressure in approximately 1 hour. Unit C dropped pressure faster, returning in approximately 20 minutes, but the fall in pressure did not occur until normal breathing had been present for 10 to 15 minutes. Unit B dropped pressure more gradually. After 1 hour, it had dropped from 18 cm H2O to 12 cm H2O (when the test was ended). Extrapolating, it would take more than 2 hours to drop all the way to 5 cm H2O.

Figure 7. Maximum pressure response.

Of the myriad of factors evaluated by auto-adjusting devices, only three abnormal breathing patterns were tested, single examples of flow limitation, apnea, and hypopnea (Figure 7). Specifically, response to snoring was not evaluated. In addition, patient adaptation was not allowed for. A real patient would hopefully correct their abnormal breathing pattern as the pressure increased. The auto-adjusting devices in this study did not have the opportunity to see any correction, so they may have responded differently if the abnormality was corrected. Nonetheless, it is still significant to note the differences in their performance.

Devices varied markedly in their response to the four (including normal) simulated patient breathing patterns. Unit A was the only device to respond to all three abnormal breathing patterns. This unit also responded favorably to the transition from abnormal to normal breathing with a gradual pressure decrease.

In diagnostic use, a statistical analysis of an auto-adjusting pressure response is sometimes used to determine a patient’s appropriate fixed CPAP pressure. Either a mean pressure or a pressure at which the unit spends 90% to 95% of the total time below is commonly used. The differences in the response of the various units could be expected to produce different values in this analysis. In a therapeutic application, it is expected that different patients will have varied experience and outcome. This may depend on the device they are treated with and the type or variation of SDB pattern, which is not always a static condition.

Clinicians should be aware that each manufacturer’s device has a different algorithm, which affects its response to patient breathing patterns. Device selection may impact the efficacy of the treatment and should be conducted with the performance capabilities of each device in mind.

Peter Bliss, BME, is technical director, and Robert McCoy, BSM, RRT, is managing director, both at Valley Inspired Products LLC, Burnsville, Minn;; and Todd Eiken, RPSGT, is director of the Metropolitan Sleep Disorders Center, St Paul, Minn; he is also a member of Sleep Review’s Editorial Advisory Board.

1. Series F. Is treatment of obstructive sleep apnea syndrome with auto-NCPAP useful? Sleep. 2000;23:S161-165.
2. Plywaczewski R, Zgierska A, Bednarek M, Zielinski J. Comparison of automatic (auto-NCPAP) and “manual” NCPAP pressure titration in patients with obstructive sleep apnea. Pneumonol Alergol Pol. 2000;68:232-237.
3. Hudgel DW, Fung C. A long-term randomized, cross-over comparison of auto-titrating and standard nasal continuous airway pressure. Sleep. 2000;23:645-648.
4. Hosselet JJ. Auto-controlled continuous positive pressure in the titration and treatment of obstructive sleep disorders. Rev Mal Respir. 2000;17:S81-S89.
5. Teschler H, Wessendorf TE, Farhat AA, Konietzko N, Berthon-Jones M. Two months auto-adjusting vs conventional NCPAP for obstructive sleep apnea syndrome. Eur Respir J. 2000;15:990-995.
6. Fletcher EC, Stich J, Yang KL. Unattended home diagnosis and treatment of obstructive sleep apnea without polysomnography. Arch Fam Med. 2000;9:168-174.
7. Clinical Practice Review Committee. Hypopnea in sleep-disordered breathing in adults. Sleep. 2002;24:469-470.
8. Hosselet JJ, Norman RG, Ayappa I, Rapoport DM. Detection of flow limitation with a nasal cannula/pressure transducer system. Am J Respir Crit Care Med. 1998;157:1461-1467.
9. Berthon-Jones M, Lawrence S, Sullivan CE, Grunstein R. Nasal continuous positive airway pressure treatment: current realities and future. Sleep. 1996;19(9 suppl):S131-S135.

Additional Reading
d’Ortho MP, Grillier-Lanoir V, Levy P, et al. Constant vs automatic continuous positive airway pressure therapy: home evaluation. Chest. 2000;118:1010-1017.

Series F. Accuracy of an unattended home NCPAP titration in the treatment of obstructive sleep apnea. Am J Respir Crit Care Med. 2000;162:94-97.

Series F, Marc I. Importance of sleep stage- and body position-dependence of sleep apnea in determining benefits to auto-NCPAP therapy. Eur Respir J. 2001;18:170-175.