Downloading data from positive airway pressure devices in adults

INTRODUCTION — Positive airway pressure (PAP) devices are primarily used to treat sleep disordered breathing including obstructive sleep apnea (OSA). Adherence and efficacy of PAP is routinely monitored subjectively. However, downloadable data from PAP devices provide objective data that complement subjective assessment to facilitate further management [1-3]. Documentation of adherence (ie, compliance) from these downloads is often mandated by insurers. Although downloading additional data from PAP devices has become commonplace, their accuracy in determining efficacy of treatment compared with in-laboratory polysomnography is uncertain and data demonstrating improved outcomes with their use are few.

Using device data downloads to determine adherence with and efficacy of PAP is discussed in this topic review. Details regarding use of downloadable information for continuous PAP (CPAP) titration following initiation are discussed separately. (See "Mode selection for positive airway pressure titration in adult patients with central sleep apnea syndromes" and "Mode selection for titration of positive airway pressure in adults with obstructive sleep apnea", section on 'Follow-up after initial titration' and "Mode selection for titration of positive airway pressure in adults with obstructive sleep apnea".)

RATIONALE FOR DEVICE DOWNLOADS — Data from PAP devices are downloaded for the following reasons:

●Documentation of adherence (ie, compliance) to PAP devices is often mandated by third party insurers so that reimbursement can be justified. Adherence can be readily identified and quantified from device downloads. (See 'Assessing adherence' below.)

●Downloadable data may also identify PAP efficacy. (See 'Assessing effectiveness' below.)

●Downloads may also be able to provide clues as to the underlying reason for nonadherence or poor efficacy (eg, airway leak or a complex sleep disorder) and in some cases, prompt the need for a formal in-laboratory sleep study. (See 'Assessing leak' below and 'Assessing detailed data to assist in troubleshooting' below.)

Downloadable data from the device itself has many advantages when compared with an in-laboratory study while the patient uses continuous PAP (CPAP) since it is cheaper, an overnight stay is not required, and data from more than one night can be obtained. However, they may be less accurate for the detection of sleep related events, underestimate the frequency of hypopneas, and fail to diagnose complex sleep disorders when present. Thus, the clinician should be aware of the associated caveats so that patients may receive further evaluation when necessary and be treated accordingly. (See "Overview of polysomnography in adults".)

ACCESSING DEVICE DATA — All modern PAP devices are able to be externally accessed, and the stored data downloaded or reviewed. Data can be downloaded using various methods, which depend upon the device brand and model (secure digital cards, modems, and wireless transmission). Thus, the clinician should be aware of what device is being used by the patient and how information can be accessed from it. With increasing use of telemedicine outpatient care, remote access to data with use of modem devices or wireless transmission is increasingly important.

Secure digital card — Secure digital (SD) cards can be used to externally download data using an SD card reader and software provided by the device manufacturer. In general, cards can be removed by the patient or medical supply company, and taken or mailed to the provider's office to be read by a card-reader device.

Modem device — Modems can be externally attached to a device or internal to the device. These can remotely send the data to an online database that can be accessed by the medical supply company and/or the clinician.

Wireless transmission — Wireless transmission technology is incorporated into some of the machines, allowing data to be uploaded daily to an online database as long as the machine is connected to an electrical outlet. Providers should be aware that some older devices that previously transmitted data wirelessly may no longer do so if they used a 3G network as all US carriers shut down their 3G networks in 2022. For clinicians or office staff to access data that are remotely transmitted, they must request to be added as a clinical provider by the medical supply company. Alternatively, some online monitoring systems allow the provider to add themselves as a care provider to a patient if the PAP serial number and personal health information is available.

Occasionally, if the device card or modem are not accessible, the device itself may have a menu that can be accessed to reveal some of the stored data, although these data are considerably less detailed than those accessed from the SD card or modem.

CONTINUOUS PAP DOWNLOADS — Continuous PAP (CPAP) devices are the most commonly used PAP devices. Download reports can be generated with various levels of detail with some older devices providing minimal information only. However, most newer devices provide the following information:

●The basic download report consists of data on adherence (ie, compliance), effectiveness, and leak, usually reported out as a summary of events over a specific period (eg, 30 to 90 days). Most physicians routinely review this information within the first month of therapy as adherence patterns are established early [4]. Long term, it is recommended that patients have adherence documented at least annually or when symptoms develop [5,6].

●A detailed or comprehensive report provides the same basic information as well as information from individual days, and may also provide information on individual respiratory events occurring throughout a night. Most clinicians access this information when further detail on poor adherence and effectiveness are needed or when data is incongruent with the patient symptoms.

Assessing adherence — CPAP devices provide information on duration and frequency of use. Although the definition of adherence is ill-defined, most medical insurance companies and professional organizations consider usage for ≥4 hours per night for ≥70 percent of nights to define CPAP adherence. However, in general, most experts agree that the greater the number of hours spent receiving CPAP, the more likely a patient is to benefit. (See "Assessing and managing nonadherence with continuous positive airway pressure (CPAP) for adults with obstructive sleep apnea", section on 'Sequalae' and "Assessing and managing nonadherence with continuous positive airway pressure (CPAP) for adults with obstructive sleep apnea", section on 'Definition (hours of CPAP use)'.)

The optimal period for adherence analysis (eg, days or weeks) is unknown. However, we agree with the American Thoracic Society (ATS) recommendations that suggest selecting a period between 7 and 90 days for the adequate assessment of adherence [7]. For the period that is preselected by the physician, total CPAP usage is typically displayed both as total percentage of days used as well as percentage of days used ≥4 hours (figure 1). In addition to providing information on CPAP usage, this section of the download will display the mode of use (ie, CPAP or auto-adjusting CPAP [APAP]) and whether or not expiratory pressure relief is being used. Expiratory pressure relief is a technology that allows for expiratory pressure reduction. This mechanism may have different names based on the manufacturer. The reduction of pressure during exhalation can help patients who are expiratory pressure intolerant (ie, patients who have difficulty exhaling against PAP). The amount of pressure reduction can be adjusted between levels of 1 and 3, but the amount of pressure reduction based on these levels are not equivalent between manufacturers.

The clinician should be aware of whether they are reviewing the "mask-on" or "machine-on" time, or both. Downloads report the cumulative hours used during the assessed time period (mask-on) as well as the cumulative hours during which the device was turned on (blower time; machine-on). An estimated 10 percent difference between these measurements may be associated with normal use [8]. A large discrepancy in these numbers may suggest excessive leak (eg, mouth breathing with a nasal mask in place) or a patient leaving the machine on when not actually in use (see 'Assessing detailed data to assist in troubleshooting' below).While most CPAP machines can differentiate between mask-on and machine-on time, some older machines may only contain meters that record machine-on time, making them less accurate for the assessment of adherence since they only record total hours or days used.

If it is determined that a patient is nonadherent, then measures can be taken to investigate the underlying reason(s) (eg, mask intolerance, dry nares, embarrassment) and to fix them (eg, mask adjustment, humidification, education). Reasons and interventions for nonadherence are discussed separately. (See "Assessing and managing nonadherence with continuous positive airway pressure (CPAP) for adults with obstructive sleep apnea".)

Assessing effectiveness — Most newer CPAP device downloads provide information on effectiveness of therapy. Devices usually determine effectiveness by measuring the average apnea hypopnea index_flow (AHI_flow) during CPAP therapy averaged over the date range requested (figure 1), but other types of qualitative information can also be obtained in some devices:

●The AHI_flow – An apnea hypopnea index (AHI) is defined as apneas plus hypopneas divided by total sleep time ([apneas + hypopneas]/total sleep time) based on polysomnography. Under optimal circumstances, apneas and hypopneas during in-laboratory polysomnography are identified by changes in oximetry, arousals, and airflow [9]. However, since CPAP devices cannot measure oximetry or sleep including arousals from sleep, only airflow can be used to calculate a surrogate value for the AHI known as AHI_flow. Although an AHI_flow of <10 events per hour is sometimes considered adequate therapy [10,11], validated outcomes studies to determine a clinically significant residual AHI_flow have not yet been done. While this AHI_flow can be used as an indicator of CPAP efficacy, it should be noted that it is an estimate only. Respiratory effort-related arousals (RERAs) are also reported here. However, RERAs reported on CPAP downloads are not the same as those scored during a polysomnogram. Since there is no measure of sleep or arousal on a CPAP device, RERAs are estimated based on progressive flow limitation in an event that does not meet criteria for apnea or hypopnea.

Several studies have compared device-calculated AHI_flow with polysomnographic measurement of AHI (ie, the gold standard). The findings generally show good correlation for apneas [12-14], although some studies have suggested that the detection of hypopneas and other sleep-related events are underestimated by devices, which may lead to overestimation of CPAP efficacy [10,15]. In addition, it is important for the clinician to understand that definitions of apnea and hypopnea event detection are not uniform among manufacturers. Each device manufacturer has a unique algorithm to determine AHI_flow. For example, one commercial entity uses a 50 percent decrease from baseline ventilation as a hypopnea and a 75 percent decrease from baseline ventilation as an apnea. Conversely, a different commercial entity uses a 40 to 80 percent reduction in flow to determine hypopneas and a greater than 80 percent reduction in flow to determine apneas [7].

●Respiratory event type (open or closed airway) – In addition to AHI_flow, some CPAP devices use pressure pulse or forced oscillation technology to determine if the airway is open or closed during a respiratory event. Thus, when a respiratory event is detected by the device, it may be recorded as an open airway (a possible central event) or a closed airway (a possible obstructive event) (figure 2). However, compared with polysomnography, data from these devices are less accurate. One study of 45 patients with OSA and complex sleep apnea reported that the device correctly detected airway obstruction in most cases (87 percent), while only 16 percent of apneic events associated with an open airway were not correctly identified [11].

The AHI_flow should not be interpreted in isolation but rather, always be done in conjunction with assessing symptoms, airflow leak, and the quality of the event (eg, open airway or obstructive), if feasible. As examples:

●If the events are closed airway events with an AHI_flow greater than 20 events per hour (ie, supporting obstructive events only) or greater than 10 in a symptomatic patient and there is no significant leak detected by the machine, an increase in delivered PAP may be indicated [7].

●A high number of persistent open airway events on a download suggests a concern for residual central apneas (ie, treatment-emergent central sleep apnea) and further testing with polysomnography may be warranted to better determine the ideal treatment course.

●If clinical symptoms do not correlate with the AHI_flow, an assessment utilizing in-laboratory CPAP titration should be performed.

Respiratory parameters should be interpreted with caution if an inline filter or airflow resistor is used as these adaptations can alter the PAP device algorithm accuracy by altering flow.

Assessing leak — Excessive airflow leak, usually via the mask (eg, ineffective seal from an ill-fitting mask) or mouth (eg, mouth breathing with nasal mask in place), can alter the effectiveness of PAP therapy by failure to deliver the desired airway pressure. Leak data definitions may differ among device manufacturers; thus, when evaluating leak data, it is important to know how leak is defined by each device. In addition, a certain amount of mask leak is intentional ("allowable" leak; ie, masks are purposely designed to have fenestrations that allow a portion of exhaled air to escape to wash out exhaled carbon dioxide and prevent rebreathing). Thus, the volume of airflow leak varies based upon each mask type (eg, higher with nasal mask than full face mask). The mask type may be programmed into some devices to improve the accuracy of the calculated leak volume. The threshold for determining excessive leak should be dependent on the device being used.

Devices also vary in how they report leak:

●Some devices report unintentional leak only (ie, that above the intentional amount)

●Other devices report "percent time in large leak." This is typically describing the percent of time the patient’s leak is greater than two times the intentional leak

The threshold for determining excessive leak also varies among devices. Examples include:

●A leak level that exceeds a preset "flow versus pressure" curve (the averaged leak through all mask exhalation ports at various pressure)

●A patient spends >1 hour during therapy with a large leak

●The 95th percentile unintentional leak >24 L/minute (with correct mask interface programmed into device)

If a clinician determines that the device download demonstrates an excessive leak that is affecting OSA control, an intervention to improve the leak may be made, such as changing the mask type, ensuring the mask size is correct, refitting the mask, and occasionally use of a chin strap. Leak through the mouth using a nasal mask may be reduced by use of a chin strap or a full face mask. (See "Titration of positive airway pressure therapy for adults with obstructive sleep apnea", section on 'Choosing the correct patient-device interface'.)

Assessing detailed data to assist in troubleshooting — In addition to a summary page that averages data over the date range specified, many device downloads will also include detailed data. These data may be particularly helpful for troubleshooting patients with poor adherence, poor tolerance, or an elevated AHI_flow (that suggests CPAP is ineffective). In addition, to clinical evaluation and assessment for mask leaks, the clinician should review data from individual nights on these patients. Such data may provide clues as to the underlying issues and guide the clinician to the next appropriate intervention. As examples:

●Detailed data may guide the clinicians intervention when there is an elevated AHI_flow on the summary page. As an example, respiratory events associated with vibratory snoring may suggest an increased pressure is needed. As another example, respiratory events that occur in high frequency and then abruptly stop may suggest changes in sleep stage or body position and prompt a repeat titration or an increased pressure (figure 3).

●The detailed data can also provide insight when there is a discrepancy between the patient's reported use and documented adherence. The download can show when the machine is on but not detecting breathing (from mask leak or the patient's failing to wear it). (See 'Assessing adherence' above.)

●Some devices provide breath-by-breath waveform graphics which can further characterize a patient's respiratory pattern during the events scored by the device. Additionally, one study showed that direct visualization of flow data may improve detection of residual respiratory events not scored by the device [16].

AUTO-ADJUSTING DEVICE DOWNLOADS — Auto-adjusting continuous PAP (autoCPAP or APAP) devices are sometimes used to diagnose and treat obstructive sleep apnea (OSA). APAP devices have proprietary modules for adjustment of CPAP during sleep to either set a fixed level of CPAP that can be administered through a regular CPAP device, or to treat obstructive sleep apnea with variable levels of autoCPAP throughout the night. Similar summary information to that provided by regular CPAP devices can be downloaded from APAP devices (eg, indices of efficacy including apnea hypopnea index_flow [AHI_flow], adherence, leak, and graphics over a set period) (figure 4). Detailed data from these devices can provide additional information on individual nights. Download data from APAP devices vary among manufacturers but some reports suggest that AHI_flow from APAP devices may also overestimate the effectiveness of autoCPAP (since AHI_flow may underestimate hypopneas) [15-17]. Further details regarding APAP are provided separately. (See "Home sleep apnea testing for obstructive sleep apnea in adults" and "Mode selection for titration of positive airway pressure in adults with obstructive sleep apnea", section on 'Follow-up after initial titration' and "Mode selection for titration of positive airway pressure in adults with obstructive sleep apnea".)

BILEVEL DEVICE AND ADAPTIVE SERVO VENTILATOR DOWNLOADS — More advanced PAP machines such as bilevel PAP used in the spontaneous timed mode, adaptive servo ventilators, and volume-assured pressure support devices that are used for central sleep apnea and hypoventilation can also generate downloadable data. However, there are no data and guidelines on the use of this information for clinical purposes. Thus, data from these devices should be used with caution. (See "Mode selection for positive airway pressure titration in adult patients with central sleep apnea syndromes" and "Central sleep apnea: Treatment".)

There is significant variation between different device manufacturers with these types of downloads. They provide data on minute ventilation, tidal volume, and respiratory rate in addition to the information provided for CPAP machines (eg, apnea hypopnea index_flow [AHI_flow], compliance, leak, and graphics) (see 'Continuous PAP downloads' above). When used in combination with data from the patient (eg, symptoms and arterial blood gas analysis), this information can help guide pressure adjustments by indicating the need for increased ventilation or for improved control of mask leak to optimize ventilation. (See "Mode selection for positive airway pressure titration in adult patients with central sleep apnea syndromes".)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Sleep-related breathing disorders in adults".)

SUMMARY AND RECOMMENDATIONS

●Rationale – Positive airway pressure (PAP) devices are primarily used to treat sleep disordered breathing including obstructive sleep apnea. In conjunction with subjective monitoring, downloadable data from PAP devices can be used to provide objective data that may facilitate further management. The accuracy of downloaded data compared with in-laboratory polysomnography is uncertain and outcome data demonstrating improved outcomes with their use are few. (See 'Introduction' above and 'Rationale for device downloads' above.)

●Accessing data – Stored data can be downloaded or reviewed from most newer devices using various methods, which depend upon the brand and model (secure digital cards, modems, and wireless transmission). Thus, the clinician should be aware of what device is being used by the patient and how information can be accessed from it. (See 'Accessing device data' above.)

●CPAP downloads – Continuous PAP (CPAP) devices are the most commonly used PAP devices. While downloads from some older devices provide minimal basic data, most newer devices provide information on adherence (ie, compliance), effectiveness, and airway leak, and others provide further details on respiratory events during sleep. Clinicians should be aware of the parameter definitions used by manufacturer since they vary considerably. Data can be used to modify therapy, thereby improving adherence or effectiveness. The clinician should maintain a low threshold to perform in-laboratory polysomnography, particularly when data is incongruent with symptoms. (See 'Continuous PAP downloads' above.)

●Auto-adjusting device downloads – Similar information to that provided by regular CPAP devices can be downloaded from auto-adjusting PAP (APAP) devices to either set a fixed level of CPAP that can be administered through a regular CPAP device, or to treat obstructive sleep apnea with variable levels of APAP throughout the night. (See 'Auto-adjusting device downloads' above and "Mode selection for titration of positive airway pressure in adults with obstructive sleep apnea", section on 'Initial mode and setting options' and "Mode selection for titration of positive airway pressure in adults with obstructive sleep apnea".)

●Other devices – Bilevel PAP used in the spontaneous timed mode, adaptive servo ventilators, and volume-assured pressure support devices that are used for central sleep apnea and hypoventilation can also generate downloadable data. However, there are limited data and guidelines on the use of this information for clinical purposes. (See 'Bilevel device and adaptive servo ventilator downloads' above.)