3 5
New OSA Patient Simulator for Testing PAP Devices
correct such a disturbance. For instance, it is interesting to note that, as we explained previously [21], the simulated hypopneas in our model were defined according to specific values of a flow- limitation pattern index initially introduced by Teschler et al [33]. Therefore, it could be possi- ble that automatic CPAP devices set to detect hypopneas using this index, or something simi- lar, could be more suitable for detecting our simulated events than other devices that use other metrics to define and detect hypopneas. Another reason for the observed different response in the automatic CPAP devices tested is that the optimal rate of pressure increase after detection of obstructive events has not been clinically defined. In fact, APAP devices are designed to nor- malize breathing at a rate which treats actual SDB, avoiding any response to false events, thereby unnecessarily modifying pressure. The results of this bench test have shown that, under well-controlled conditions, there are marked variations in response by different APAP devices, and that there may be high residual AHI or uncontrolled flow limitation in some female patients on some APAP devices. Therefore, all APAP devices should not be considered equal, and efficacy and patient comfort should be carefully examined following APAP initiation. It must be noted that our results are restricted to the specific patterns of disturbed breathing used in this bench test to simulate a specific OSA patient. It is possible that the response of the tested devices would have been different from the ones reported here if SBD was simulated using different patterns or patient phenotypes. In addition, a limitation of this study is that one device of each type was used. Hence, a more complete assessment would require testing of a larger number of each type of device randomly obtained from those available in the market. Finally, it should be stressed that although bench testing is a useful way to investigate the behaviour of different devices, testing outcomes may vary in clinical practice due to the almost unlimited spectrum of events and phenotypes found in real life. Indeed, crucial factors such as changes in loop gain, and upper airway compliance and pharyngeal critical pressure are not considered in our model. Accordingly, bench testing should be considered as a preliminary assessment before clinical evaluation in patients. In conclusion, this study showed that a dynamic bench model tailored to represent specific OSA patient phenotypes, incorporating a variety of disturbed breathing events within the same simulated night, including different degrees of severity along sleep stages, and a period of wake- fulness, can be useful to characterize treatment responses of commercially-available APAP devices. This demonstrates that bench testing can be modified to better represent a “ real ” patient, and that APAP devices can show markedly different responses to the same simulated breathing patterns. Realistically mimicking OSA patients during bench testing is useful as a first step to aid in the understanding of actual APAP device responses observed in the clinical setting, and can be helpful in selecting the device that best meets the individual needs of each patient, thereby improving comfort and increasing adherence to therapy, which is essential for effective treatment and reducing the consequences of OSA [34]. Acknowledgments Authors wish to thank Mr. Miguel Angel Rodriguez for his technical support. English language medical writing assistance was provided by Nicola Ryan, independent medical writer, funded by ResMed. Author Contributions Conceived and designed the experiments: VI JMM RS AJW DR HW DN RF. Performed the experiments: VI RS RF. Analyzed the data: VI RS RF. Contributed reagents/materials/analysis tools: JMM AJW HW DN RF. Wrote the paper: VI JMM AJW HW DN RF.
PLOS ONE | DOI:10.1371/journal.pone.0151530 March 15, 2016
9 / 11
next generation sourcing
Made with FlippingBook Online newsletter