INTRODUCTION — Portable monitoring (PM) is a diagnostic test for obstructive sleep apnea (OSA), a disorder characterized by repetitive episodes of apnea or reduced inspiratory airflow due to upper airway obstruction during sleep. It has evolved as a relatively accurate alternative to overnight, attended, in-laboratory polysomnography (PSG) in selected patients.

Advantages of PM include its convenience (it can be performed in the patient's home or in a hospital room) and its lower costs (PM devices are less costly than complete polysomnography systems and the attendance of a technologist is not required). However, a disadvantage is that fewer physiologic variables are measured during PM than PSG, which can lead to misinterpretation of the results. Other advantages and disadvantages are listed in the figure (table 1).

The United States Centers for Medicare and Medicaid Services (CMS) have released guidelines that state that results from PM can be used to support a prescription for positive airway pressure therapy [1]. The American Academy of Sleep Medicine (AASM) has also released clinical practice guidelines to guide clinicians in the use of PM [2,3].

The use of PM in the diagnostic evaluation of suspected OSA is reviewed here. The diagnostic approach to a patient with suspected OSA is described separately. (See "Clinical presentation and diagnosis of obstructive sleep apnea in adults".)

TYPES OF MONITORING DEVICES — Four types of sleep study monitoring devices have been defined [2,4].

Type 1 devices — Type 1 monitoring devices are used for in-laboratory, technician-attended, overnight polysomnography (PSG). These devices are not used for PM and are discussed separately. (See "Polysomnography in obstructive sleep apnea in adults".)

Type 2 devices — Type 2 monitoring devices can record the same variables as type 1 devices. The major difference from type 1 devices is that they can be used outside of the sleep laboratory and a technologist is not present during the recording.

Type 3 devices — Type 3 monitoring devices typically measure four physiologic variables, including two respiratory variables (eg, respiratory movement and airflow), a cardiac variable (eg, heart rate or an electrocardiogram), and arterial oxyhemoglobin saturation via pulse oximetry. Some devices have additional signals that can detect snoring, determine body position, or detect movement. Sleep variables (eg, sleep stages, sleep continuity) cannot be measured by a type 3 device and a technologist is not present during the recording.

Recordings from a typical type 3 portable monitoring device are shown in the figure (figure 1). In this example, respiratory effort is detected by a chest wall impedance monitor, airflow is detected by a heat-sensitive thermistor at the nose and mouth, and oxyhemoglobin saturation is measured by a pulse oximeter. The heart rate is also recorded. Measurement of these variables is generally sufficient to detect most apneas (obstructive, central, mixed) and hypopneas.

Type 3 devices have important limitations:

Type 3 devices cannot detect arousals from sleep because they cannot monitor electroencephalography (EEG). As a result, respiratory effort related arousals (RERAs) cannot be detected and hypopneas characterized by arousals may remain undetected. The inability to detect RERAs may lead to underestimation of the respiratory disturbance index (RDI) and under-recognition of upper airway resistance syndrome (UARS). (See "Sleep related breathing disorders in adults: Definitions", section on 'Respiratory effort related arousals' and "Sleep related breathing disorders in adults: Definitions", section on upper airways resistance syndrome and respiratory disturbance index.)

• The apnea hypopnea index (AHI) is calculated by dividing the number of apneas and hypopneas by the study duration when determined by a type 3 device. In contrast, the AHI is calculated by dividing the number of apneas and hypopneas by the duration of sleep in hours when derived from conventional PSG. The effect is that an AHI derived by PM is usually lower than an AHI derived by PSG. This decreases the diagnostic sensitivity of type 3 devices, compared to type 1 and 2 devices. (See "Sleep related breathing disorders in adults: Definitions", section on 'Summary measures'.)

• Type 3 devices do not record when the patient is lying in the supine or lateral decubitus position. Thus, OSA that develops or worsens in the supine position may not be detected.
• Type 3 devices cannot distinguish rapid eye movement (REM) sleep from non-REM (NREM) sleep because they do not monitor EEG. As a result, the interpreting clinician cannot be certain that a period of REM sleep was captured during the study. This is an important limitation because severe perturbations are most common during REM sleep.

Type 4 devices — Type 4 monitoring devices are defined differently by different organizations. The American Academy of Sleep Medicine (AASM) defines type 4 monitoring devices as devices that record one or two variables (eg, arterial oxyhemoglobin saturation and airflow) and can be used without a technician. These devices are called continuous single or dual bioparameter devices. In contrast, the United States Centers for Medicare and Medicaid Services (CMS) guidelines include devices that measure three variables as type 4 monitoring devices. A technologist is not present during the recording.

Type 4 devices that record one or two variables provide limited information. Pulse oximetry and airflow are the physiological variables that are most commonly measured. As a result, derived information typically includes the frequency of apneas, frequency of hypopneas, AHI, baseline oxyhemoglobin saturation (SpO2), mean SpO2, frequency of oxyhemoglobin desaturation, duration of oxyhemoglobin desaturation, degree of oxyhemoglobin desaturation, and nadirs of SpO2. All of the limitations of type 3 devices described above also apply to type 4 devices.

Pulse oximetry alone is also considered a type 4 device, which is discussed further below. (See 'Pulse oximetry' below.)

INDICATIONS — The most common indications for PM include the diagnostic evaluation of suspected OSA and titration of positive airway pressure therapy.

Diagnosis — Many PM devices have been validated against standard PSG, typically by testing the same patient with both modalities in the sleep laboratory [5-8]. A wide range of diagnostic characteristics have been reported using PM, but the sensitivity and specificity seem to be high in populations considered by sleep specialists to be at high risk of OSA on the basis of clinical symptoms, assuming there are no comorbid medical disorders or sleep disorders [2].

We endorse the most recent clinical practice guidelines from the American Academy of Sleep Medicine (AASM) [2,3]:

• PM for diagnostic evaluation of suspected OSA should be performed only in conjunction with a comprehensive sleep evaluation, preferably by a sleep medicine specialist.
• PM may be used as an alternative to PSG for the diagnosis of OSA in patients with a high pre-test probability of moderate to severe OSA (see "Clinical presentation and diagnosis of obstructive sleep apnea in adults".
• PM should not be used in patients who have comorbid medical conditions that predispose to sleep related breathing disorders or in whom another sleep disorder is suspected. As an example, PM should not be used in patients with heart failure because they are predisposed to Cheyne-Stokes breathing (see "Cheyne-Stokes breathing and obstructive sleep apnea in heart failure".
• PM must record airflow (preferably with both a thermistor and a nasal cannula pressure transducer), respiratory effort, and blood oxygenation.
• An experienced sleep technician, sleep technologist, or appropriately trained provider must apply the PM sensors or directly educate the patient about the correct application of sensors. In addition, there should be methodology in place to monitor the quality of recordings.
• PM devices must be capable of displaying the raw data for review by the clinician, in order to allow assessment of the quality of the data.
• All patients undergoing PM should receive a follow-up visit with an appropriately trained provider to discuss the results of the test.

The CMS guidelines do not make any specific recommendations about which type of PM device should be used, nor do they specify the appropriate population to undergo PM. However, they do indicate that positive airway pressure therapy prescriptions will be covered by Medicare and Medicaid only if OSA is diagnosed using a type 1, 2, or 3 device [1]. In addition, a Type 4 device that measures at least three variables is acceptable.

Limitations — PM has important limitations that should be routinely considered by the clinician who is interpreting the results of the study:

• Type 3 and type 4 devices often underestimate the AHI (described above), increasing the likelihood of a false-negative result. As a result, the pretest clinical suspicion for OSA and the results of PM studies should always be considered together. Additional testing is indicated whenever there is a reasonable clinical suspicion for OSA, but the PM study is negative.
• Type 4 devices that record only one or two physiologic variables cannot distinguish obstructive and central events, in part because respiratory effort is not measured. Patients at risk for central apneas, hypopneas, or hypoventilation should not be tested with such devices.
• Different manufacturers use different sensors and potentially different algorithms to identify respiratory events. Many devices do not allow the raw data to be viewed by the provider who ordered the test because the scoring algorithm is proprietary. This is problematic according to numerous studies that showed that manual scoring or manual editing of automated scoring improves diagnostic accuracy compared to automated scoring alone [9-14]. Clinical practice guidelines recommend manual scoring or manual editing of automated scoring.

Titration — The role of portable monitoring in determining the initial level of positive airway pressure therapy is described in detail elsewhere. (See "Initiation of positive airway pressure therapy for obstructive sleep apnea in adults", section on 'Determining the amount of positive airway pressure'.)

Reassessment — PM can be used to assess the adequacy of the prescribed therapy within a patient's usual environment and to guide adjustment of the therapy. The approach is similar to that for diagnosis and titration, respectively.

PULSE OXIMETRY — Pulse oximetry is a widely accepted and important component of both polysomnography (PSG) and PM. However, when it is measured alone or with only one other variable, it is not recommended for the diagnostic evaluation of suspected OSA [1,2,15]. In this section, we briefly describe the reason that overnight oximetry alone is not considered an acceptable diagnostic modality.

The characteristics of overnight pulse oximetry alone as a diagnostic test are highly dependent upon whether the criteria used to define a positive test are quantitative or qualitative:

• Overnight pulse oximetry tends to have a high specificity but a low sensitivity when quantitative criteria are used [16-18]. As an example, a study of 200 patients with suspected OSA compared overnight pulse oximetry versus polysomnography [17]. Positive overnight pulse oximetry was defined as a decrease in the oxyhemoglobin saturation (SpO2) of 4 percent or greater from baseline, to a value 90 percent or lower. Overnight pulse oximetry had a sensitivity of 41 percent and a specificity of 97 percent.
• In contrast, overnight pulse oximetry tends to have a high sensitivity but a low specificity when qualitative criteria are used [7]. Qualitative criteria emphasize pattern recognition. The overnight pulse oximetry strip is printed and visually analyzed for a pattern of multiple, short-duration fluctuations of the SpO2. No specific level of desaturation is required for a positive study. As an example, a study of 240 patients with suspected OSA compared overnight pulse oximetry versus polysomnography. Positive overnight pulse oximetry was defined as greater than 10 fluctuations of SpO2 per hour of sleep. The sensitivity and specificity of overnight pulse oximetry were 98 and 48 percent, respectively.

Overnight pulse oximetry alone can be either a sensitive or a specific test for OSA, but not both. As a result, either false-positive or false-negative tests will be common, depending on the criteria chosen to define a positive test.

SUMMARY AND RECOMMENDATIONS

• Portable monitoring (PM) is an acceptable approach to the diagnostic evaluation of suspected obstructive sleep apnea (OSA). However, it should only be used for the diagnostic evaluation of suspected OSA in patients with a high pre-test probability of having moderate to severe OSA and no comorbid medical or sleep disorders. (See 'Diagnosis' above.)

• PM can also be used to guide the titration of positive airway pressure therapy if the chosen mode of positive airway pressure is either continuous positive airway pressure (CPAP) or autotitrating positive airway pressure. (See 'Titration' above.)

• Four types of PM devices have been defined on the basis of the amount of data that they can measure. Type 3 and 4 devices have highly variable diagnostic performance and cannot measure sleep, which has many drawbacks. (See 'Types of monitoring devices' above.)

• Pulse oximetry is a widely accepted and important component of both polysomnography and PM. However, it should NOT be used alone for the diagnostic evaluation of suspected OSA. (See 'Pulse oximetry' above.)