Home Sleep Apnea Testing

Susan Purdy

Richard B. Berry

INTRODUCTION

Obstructive sleep apnea (OSA) is a common disorder associated with significant morbidity and mortality and requires accurate assessment and proper management. Polysomnography (PSG) is the standard test for the diagnosis of suspected sleep-related breathing disorders and for positive airway pressure (PAP) titration to determine an effective level of PAP for treatment (1, 2, 3, 4). Attended PSG requires highly trained personnel for adequate study performance and interpretation and can be costly. Access to PSG in a timely manner may be limited or delayed in some locales, and this has prompted the use of limited channel monitoring outside the sleep center for the diagnosis of OSA. This type of testing has been called “portable monitoring” (PM), out-of-center sleep testing (OCST or OOCST), or home sleep testing (HST) (5, 6, 7, 8, 9). The term HST is used by the Centers for Medicare and Medicaid Services (CMS) to describe unattended home testing (10, 11, 12). The American Academy of Sleep Medicine (AASM) currently recommends the terminology “Home Sleep Apnea Testing” (HSAT) (4, 13). Unattended limited channel testing can be performed in the sleep center or hospital. Additionally, this testing does not usually determine the amount of sleep (no electroencephalogram [EEG] recorded). Therefore, the names PM, HST, OCST, or HSAT are not ideal but are used in much of the literature on this subject. For the sake of simplicity, unattended limited channel sleep monitoring will be termed HSAT in this chapter, except when discussing Medicare coverage, where the term HST will be used.

Although PSG is the standard test to diagnose OSA in the United States (1, 2, 3, 4), in other locales, more limited studies are routinely used. For example, Flemons et al. (14) stated in 2004 that in the United Kingdom PSG comprised approximately 10% of all sleep studies. Approximately two-thirds of the studies used oximetry alone, with the rest being limited channel studies. In the Veterans Administration (VA) Health Care System, where demand exceeds in-lab testing capacity, HSAT is commonly used in place of PSG for uncomplicated patients. However, despite the widespread use of HSAT worldwide, such testing was only approved in the United States by the CMS to qualify patients for continuous positive airway pressure (CPAP) treatment in 2008 (10, 11). Reimbursement for HSAT was subsequently approved in 2009 (12). Since that time, the number of HSAT studies has continued to increase every year. In 2014, 845,569 sleep studies were completed by 1.4% of Medicare beneficiaries for a total cost of $189 million. Since 2010, annual expenditures for sleep studies have declined, whereas the number of studies performed has increased by 9.1%. In 2014, PSG, split-night PSG, and unattended home sleep studies accounted for 40%, 48%, and 12%, respectively, of the total sleep studies. This represents a dramatic growth in the number of unattended sleep studies performed since 2000, when they represented only 0.9% of the total studies (15). Today, some insurance providers mandate the use of HSAT for the diagnosis of OSA in all uncomplicated patients.

Table 44-1 Classification of Sleep Testing

	Level I (Type 1)	Level II (Type 2)	Level III (Type 3)	Level IV (Type 4)
	Attended PSG	Unattended PSG	Cardiorespiratory monitoring	Continuous single or dual bioparameter recording
Parameters	Minimum of 7 parameters: EEG, EOG, chin EMG, ECG, airflow, respiratory effort, and oxygen saturation	Minimum of 7 parameters: EEG, EOG, chin EMG, ECG, airflow, respiratory effort, and oxygen saturation	Minimum of 4: ventilation (at least 2 parameters of respiratory movement or respiratory movement and airflow), heart rate or ECG, and oxygen saturation	Minimum of 1 oxygen saturation: flow or chest movement
Body position	Documented or objectively measured	Possible	Possible	No
Leg movement	EMG or motion sensor desirable but optional	Optional	Optional	No
Personnel interventions	Possible	No	No	No
The terminology of Level I to IV was used, but recent terminology is Type I to IV or Type 1 to 4.
ECG, electrocardiogram; EEG, electroencephalogram; EMG, electromyogram; EOG, electrooculogram; PSG, polysomnography.
From Ferber, R., Millman, R., Coppola, M., et al. (1994). Portable recording in the assessment of obstructive sleep apnea. ASDA standards of practice. Sleep, 17(4), 378-392. Adapted by permission of American Sleep Disorders Association and Sleep Research Society.

TYPES OF HSAT

The commonly used classification of sleep testing was proposed by Ferber et al. (16) in a 1994 American Sleep Disorders Association review (Table 44-1). This classification used the terminology Level I, II, III, and IV. This terminology was later modified to Type 1, 2, 3, and 4 (or Type I, II, III, and IV) (17). In 2008, the CMS issued a decision to allow HST to qualify a patient for CPAP (10, 11). CMS defined HST types slightly differently and created G-codes (G0398, G0399, and G0400) to describe HST services (Table 44-2) (11, 12). The G-codes are found in the Healthcare Common Procedure Coding System Level II codebook and are maintained and valued by the CMS. G-codes are procedure codes developed by the CMS to identify products, supplies, and services that do not have an assigned Common Procedural Terminology (CPT) code for
which there is a programmatic operating need to separately identify them on a national level. Reimbursement is determined regionally by the specific local coverage determination (LCD). Later HSATs were assigned CPT codes: 95800, 95801, and 95806 (Table 44-3). The method of determination of sleep time (95800) is not specified. CPT codes are copyrighted and maintained by the American Medical Association. Although CPT codes exist for HSAT studies, many insurance providers require the use of G-codes for billing. The AASM uses the HSAT CPT codes in their recommended standards for such testing (7, 13). The CPT classification of HSAT differs somewhat from the G-code classification. However, 95806 describes most Type 3 devices. Sleep testing using peripheral arterial tonometry (PAT), actigraphy, and oximetry was also approved for HST by the CMS, but a G-code was not assigned to this type of testing (11). This type of testing would fall under the CPT classifications 95801 or 95802. An estimate of total sleep time (TST) is determined by PAT devices on the basis of actigraphy, heart rate, and characteristics of the sympathetic tone (PAT signal) in different sleep stages (18). However, unless EEG and electrooculogram (EOG) are recorded, the AASM standards (10) require that monitoring time (MT) rather than TST be reported. In the original classification, Type 4 testing required one or two bioparameter recording (Table 44-1). However, in 2009 when the CMS issued a memo approving reimbursement for HST (Table 44-2), Type IV devices were mandated to record three or more channels, one of which was airflow (10, 11, 12). Most Type 4 devices monitor oximetry (oxygen saturation, SaO₂) as one of the channels being recorded.

Table 44-2 Centers for Medicaid and Medicare Classification of Sleep Testing

Code	Type	Setting	Monitoring
	I	Attended in facility	Minimum of 7 channels including EEG, EOG, EMG, ECG/heart rate, and oxygen saturation
G0398	II	Unattended in or out of a sleep lab facility or attended in a sleep lab facility	Minimum of 7 channels including EEG, EOG, EMG, ECG/heart rate, and oxygen saturation
G0399	III	Unattended in or out of a sleep lab facility or attended in a sleep lab facility	Minimum of 4 channels and must record ventilation, oximetry, and ECG or heart rate
G0400	IV	Unattended in or out of a sleep lab facility or attended in a sleep lab facility	Minimum of 3 channels, one of which is airflow
–		Unattended in or out of a sleep lab facility or attended in a sleep lab facility	PAT, minimum of 3 channels: PAT, actigraphy, and oximetry
ECG, electrocardiogram; EEG, electroencephalogram; EMG, electromyogram; EOG, electrooculogram; PAT, peripheral arterial tonometry.
Based on Centers for Medicare and Medicaid Services. (2009, March 3). Decision memo for sleep testing for obstructive sleep apnea (CAG-00405N). Baltimore, MD: Author; Berry, R. B., Albertario, C. M., Harding, S. M., et al.; for the American Academy of Sleep Medicine. (2018). The AASM manual for the scoring of sleep and associated events: Rules, terminology and technical specifications [Version 2.5]. Darien, IL: American Academy of Sleep Medicine.

Table 44-3 Common Procedural Terminology (CPT) Codes* for Home Sleep Apnea Testing

95800	Sleep study, unattended, simultaneous recording; heart rate, oxygen saturation, respiratory analysis (e.g., by airflow or peripheral arterial tone), and sleep time.
95801	Sleep study, unattended, simultaneous recording; minimum of heart rate, oxygen saturation, and respiratory analysis (e.g., by airflow or peripheral arterial tone).
95806	Sleep study, unattended, simultaneous recording of heart rate, oxygen saturation, respiratory airflow, and respiratory effort (e.g., thoracoabdominal movement).
Used with permission of the American Medical Association. Copyright © American Medical Association. All rights reserved.

Type 1 testing is an attended PSG in a facility. Type 2 testing (unattended or ambulatory PSG) records similar parameters, although often a reduced number of parameters are recorded. Both Type 1 and 2 testing allow the scoring of arousals and sleep staging. Type 2 studies allow recording of either heart rate (usually from oximetry) or electrocardiogram (ECG). Most Type 2 devices have the capability of recording ECG. Type 2 studies have been used in the Sleep Heart Health Study, testing a large population for the effect of sleep apnea on cardiovascular morbidity (19, 20). However, Type 2 studies are rarely used outside the research setting because of the expertise required and limited reimbursement when compared with Type 1 studies.

A Type 3 study, also called “cardiorespiratory testing,” consists of at least two channels of respiratory monitoring, oximetry, and ECG or heart rate. The G0399 classification requires a minimum of four channels and must record ventilation, ECG/heart rate, and SaO₂. Typically, airflow and respiratory effort (e.g., chest and abdominal respiratory inductance plethysmography [RIP]) are used for respiratory monitoring. The CPT 95806 classification specifies monitoring of airflow and respiratory effort. Some of the simpler HSAT devices use only one respiratory effort belt. In most Type 3 devices, the heart rate is derived from the oximetry data. An example of a Type 3 study from a comprehensive HSAT device is shown in Figure 44-1. When reading CMS documents, it is important to note that the respiratory disturbance index (RDI) is the number of apneas and hypopneas per hour of MT. This RDI definition differs from the one used in the AASM scoring manual that specifies the RDI to equal the number of apneas, hypopneas, and respiratory effort-related arousals per hour of TST (10). The AASM also recommends that the term “respiratory event index” (REI) be used to describe the number of apneas and hypopneas per hour of MT in HSAT reports (9, 13).

HISTORY OF HSAT IN THE UNITED STATES

The history of HSAT in the United States has been characterized by a slow acceptance of this type of testing as data supporting the use of the devices accumulated. CMS considered HSAT several times before finally approving the method for the diagnosis of sleep apnea. A detailed step-by-step description of the developments in evaluation and subsequent approval of HSAT for the diagnosis of OSA is provided in two of the references (21, 22). An editorial discussing the rationale for the CMS decision to approve HSAT describes the final stages of the process (23). Early studies of the utility of HSAT for
the diagnosis of OSA focused on a comparison of the ability of devices to rule in or rule out sleep apnea with that of PSG (17, 24, 25, 26, 27, 28, 29, 30), and AASM practice parameters recommended narrow indications for HSAT (31, 32). However, in 2007 before the approval of HSAT by the CMS, the AASM did publish important clinical guidelines for the use of HSAT in a wider spectrum of indications (8) if certain evaluation and testing procedures were followed. The guidelines influenced the requirements for reimbursement of HSAT studies by insurance providers. The ultimate decision of CMS to allow HSAT for the diagnosis of OSA in patients to be treated with CPAP was heavily influenced by studies showing that clinical pathways using HSAT devices resulted in equivalent outcomes from CPAP treatment, including improvement in daytime sleepiness and the amount of adherence to treatment (33, 34). These and later studies documenting the effective use of HSAT (35, 36, 37, 38, 39, 40) will be discussed in another section.

Figure 44-1 Tracings (120-second window) from a Type 3 home sleep apnea testing device (PDX, Philips Respironics) showing a hypopnea. In this study, the signal from the oronasal device was suboptimal. RIP, respiratory inductance plethysmography.

Table 44-4 Typical Requirements for Reimbursement for Home Sleep Apnea Testing

Treating physician who orders the study must perform a face-to-face evaluation. Evaluation must include the following:
1. Sleep history and symptoms including, but not limited to, snoring, daytime sleepiness, observed apneas, choking or gasping during sleep, morning headaches
2. Epworth Sleepiness Scale
3. Physical examination documents body mass index, neck circumference, and a focused cardiopulmonary and upper airway evaluation
Sleep center performing home sleep apnea testing study must be accredited by the American Academy of Sleep Medicine (AASM), The Joint commission (TJC), or Accreditation Commission for Health Care (ACHC).
Raw data must be reviewed by a sleep physician who is a board-certified, Diplomate in Sleep Medicine by a member board of the American Board of Medical Specialities (ABMS), or an active physician staff of sleep center AASM, TJC, or ACHC accredited.

The AASM clinical guidelines for the use of unattended portable monitors in the diagnosis of OSA in adults (8) provide important recommendations on how HSAT is to be performed and used. The guidelines provided a recommendation that HSAT be performed under the auspices of AASM-accredited sleep centers. Subsequent to the national carrier determination of CMS, a number of LCDs by regional durable medical equipment Medicare administrative contractors or private insurance providers were published, further defining requirements for the performance of HST (HSAT) (Table 44-4). The requirements vary by region and funding entity but in general require HSAT following
a clinical evaluation of the patient and should be performed under the auspices of an accredited sleep center (AASM, Joint Commission, or Accreditation Commission for Health Care). HSAT must be reviewed and interpreted by a physician who is either a board-certified/eligible sleep physician or a staff physician associated with an accredited sleep center. An AASM review of HSAT technology, recent AASM practice guidelines, and the AASM scoring manual also provide guidance on the technology to be used for HSAT (4, 9, 13).

ACCURACY OF HSAT

AHI PSG versus REI by HSAT

There are a number of potential reasons that the apnea + hypopnea index (AHI) by HSAT and PSG might differ. The AHI by PSG divides the number of respiratory events by the TST in hours. The AHI by HSAT divides the number of events by the hours of monitoring. The TST is always less than or equal to the MT and is often shorter than the MT by an hour or more. Thus, even if the same number of events is detected by PSG and HSAT devices, the AHI will be greater by PSG as the TST is less than the total MT. For example, if 100 events are recorded and the TST is 5 hours and the MT is 6 hours, the AHI for PSG is 20 per hour, but the AHI by HSAT will be 16.6 per hour.

The AHI values may vary from HSAT and PSG studies if the respiratory sensors used in the studies are not the same. For example, a study by Dingli et al. used a thermal sensor to detect apnea and hypopneas during PSG and nasal pressure during HSAT (25). If PSG and HSAT studies are done on different nights, night-to-night variability may also contribute to differences in AHI. Many patients have a much higher AHI in the supine position or during rapid eye movement (REM) sleep. In these patients, the relative amounts of supine and REM sleep contribute to night-to-night variability in the AHI. Of interest, a study by Levendowski et al. (41) found greater night-to-night variability with PSG compared with HSAT. Another study by Smith et al. comparing HSAT with PSG found that patients had less supine sleep at home compared with the sleep center (42). Having less supine sleep at home would often result in a lower HSAT AHI compared with a PSG AHI. In addition, if some events are missed by HSAT because of sensor dislodgement for a portion of the night, this will also reduce the AHI. On the contrary, patients may sleep better at home and have a greater amount of REM sleep. Individuals may consume more alcohol at home, therefore changing the HSAT REI (AHI).

Exact agreement between AHI values by PSG and REI by HSAT is less important than correct classification of a patient as having OSA or not having OSA. Although AHI (REI) values of 30 and 50 per hour are quite different, both would clearly support a diagnosis of OSA. On the contrary, values of 3 and 8 per hour are very similar, but only one would support a diagnosis of OSA. It is not surprising that HSATs often have their greatest utility in patients likely to have a moderate-to-high AHI. Differences in AHI by HSAT and PSG usually have minimal impact on the diagnosis of OSA in patients with moderate-to-severe OSA who have a relatively high AHI by either technique.

Only gold members can continue reading. Log In or Register to continue