Moving toward standardization: physician reporting of sleep studies

Abstract

Detailed primary data collected from sleep studies should lead to specific and clear reports with evidence-based clinical recommendations that, when introduced by sleep medicine specialists, create a window of opportunity to support our non–sleep medicine referring teams and to engage patients and caregivers in their care as recipients of the reports. This is how sleep study reporting differs from other test reports; currently, there is wide variation in how the data collected are presented and summarized. The goal of this document is to offer recommendations for structured reporting of sleep studies. We offer a practical, complete, and relevant document and a structure that can be implemented across sleep centers nationwide and does not burden the interpreter. We anticipate some readers will opine that some of the content is beyond the scope of what the interpreter physician needs to include, while others will propose missing data that they feel should have been included. We feel that the flexibility of the proposal accommodates for this and allows for a “first step” toward standardization of physician reporting of sleep studies. High-quality structured reporting of sleep studies is becoming ever more important for patient care, benefiting patients, caregivers, clinicians, durable medical equipment companies, and payers.

Citation:

Lastra AC, Ingram D, Park J, et al. Moving toward standardization: physician reporting of sleep studies. J Clin Sleep Med. 2023;19(3):595–603.

INTRODUCTION

The prevalence of sleep disorders continues to increase,^1–3 driving an increased demand for sleep medicine services. We face, however, a significant shortage of board-certified sleep medicine physicians and behavioral sleep medicine specialists.^4,5 Even though providers outside sleep medicine recognize the importance of evaluating, diagnosing, and treating sleep disorders, a significant proportion may not feel comfortable doing so.⁶ Detailed primary data collected from sleep studies should lead to specific and clear reports with evidence-based clinical recommendations that, when introduced by sleep medicine specialists, create a window of opportunity to support our non–sleep medicine referring teams and to engage patients and caregivers in their care as recipients of the reports. In our opinion, this is how sleep study reporting differs from other test reports; currently, there is wide variation in how the data collected are presented and summarized. The goal of this document is to offer recommendations for structured reporting of sleep studies.

General principles of structured reporting have been published⁷ and are listed as follows as basis for the framework for this document:

• Clinical relevance

• Completeness

• Clarity

• Consistency

• Reproducibility

• Practicality

• Applicable to all modalities

• Able to evolve over time

• Adequate for billing

• Balanced approach

High-quality structured reporting of sleep studies is becoming ever more important for patient care, benefiting patients, caregivers, clinicians, durable medical equipment companies, and payers.

METHODS

Formulating key aspects

The authors used a “modified-Delphi” method. The process is an “iterative, multistage method to solicit information and synthesize opinion into group consensus.”⁸ The authors started with a round of ideas proposal for sleep study reporting elements; then, anonymous voting rounds were alternated with video-conferencing discussion of the results. Consequently, some responses were changed, a desirable outcome for consensus building from the Delphi method (Figure 1).⁹ When an agreement was not reached, the voting was taken to members of the American Academy of Sleep Medicine (AASM) Education Committee and a final consensus reached.

Figure 1. “Modified-Delphi” method used.

How to use this document

This document is not intended to offer clinical guidelines and is not a substitute to The AASM Manual for the Scoring of Sleep and Associated Events: Rules, Terminology and Technical Specifications (AASM Scoring Manual). Sleep study reporting should include data points previously delineated by the latest version of the AASM Scoring Manual and are not in the scope of this document. These data are delivered in many cases based on the computer program or software used by individuals or institutions. These detailed primary data collected, however, can lead to specific, clear, evidence-based recommendations. Currently there is no standardization or recommended structure in how these data are interpreted, presented, and summarized.

We include considerations for standardization of sleep study interpretation divided by test type and population (adult vs pediatric). This document was designed with the sleep medicine interpreter in mind and at center stage, while also understanding that, inherent to the multidisciplinary nature of the sleep medicine field, the audience is multidisciplinary and may include not only clinicians but also patients, caregivers, durable medical equipment companies providing services, and payers. The authors acknowledge that the clinician ordering the study in conjunction with the patient has the responsibility to make the best recommendations for follow-up care. In certain circumstances and considering the interpreter’s “audience,” a sleep study report may represent an opportunity to provide education, and including additional information may be important regarding prognostic, diagnostic, or treatment considerations of sleep disorders.

REPORTS

All sleep study types

The standard components of the report include the following:

Site administrative information: sleep center site, including name, location, address, phone number and other contact information, and accreditation status.

Patient demographics: the personal information and unique identifiers linking the patient to the report, should be reported including the following:

• a. First and last name

• b. Date of birth

• c. Age at the time of the study

• d. Date when the study was performed

• e. Sex at birth (include gender-affirming therapy when applicable if clinically relevant)

• f. Weight

• g. Body mass index

• h. Current medications and pertinent medical/surgical history

• i. Reason for referral describing the clinical situation, question, study indication, and referring provider identification when available

Technical adequacy: Last, the technical adequacy of the study should be clearly detailed, as defined by sleep center policy and procedures, including limitations for the study or recommendations for repeat testing, if applicable.

Adult polysomnograms

AASM Scoring Manual¹⁰ recommendations will not be relisted here.

Type of study performed

The type of study should be clearly stated. A typical study may be a split-night polysomnography (PSG) vs a full diagnostic or therapeutic study. Other PSG options may include a full electroencephalogram montage, a complex therapeutic study involving advanced positive airway pressure (PAP) devices (eg, adaptive servoventilation or average volume-assured pressure support system), or supplemental oxygen use.

Criteria used to determine respiratory events

Scoring criteria used to define respiratory events should be clearly stated. More specifically and given changes in hypopnea definition over time, hypopnea definition criteria should be included in the report to account for variations in payor policies, when applicable, and their determination for PAP therapy coverage.^10,11

Primary diagnosis and its severity, if applicable

The report should state the most appropriate primary diagnosis and any additional diagnoses based on current diagnostic criteria.

Additional clinically relevant findings during PSG

Periodic limb movements in sleep (PLMS), rapid eye movement (REM) sleep without atonia, any additional behavioral observations from video-monitoring (eg, sleep talking, sleep eating or other parasomnias or abnormal movements, frequent bruxism, relevant respiratory or other findings), electroencephalogram abnormalities, or electrocardiogram abnormalities it should be reported as per the recommendations specified in the AASM Scoring Manual.

Remarkable features of sleep architecture

A summative statement about sleep quality should be included in the report. Although specific proportions of sleep stages, sleep efficiency, arousal index, and REM sleep latency varies by age,^12–15 if the patient’s PSG parameters fall outside the general normative range, it should be stated.

Details of the device mode(s) used and the most effective settings when applicable

For PAP titrations, include the details of the mask (ie, size, type, and name), the type of the PAP device, the range of the settings used, and the most effective PAP setting. If it was an unsuccessful titration study, then the report should state as such and provide an explanation if known. This reporting would allow appropriate preparation for a repeat titration study if applicable. If an oral appliance therapy (OAT) was used, optimum turns or the setting should be reported along with device type. If supplemental oxygen was used, the range and the optimum setting should be reported. The report should also state when the supplemental oxygen was started (eg, either at the beginning of the test or after optimal PAP settings). The Centers for Medicare and Medicaid Services publishes reimbursement requirements for nocturnal supplemental oxygen,¹⁶ and we recommend that, whenever possible as long as the patient’s care is not compromised, Centers for Medicare and Medicaid Services reimbursement guidelines be followed when starting supplemental oxygen as it could impact the coverage of this therapy.

Summary

Refer to Table 1 for a summary of recommendations and additional considerations.

Table 1.

Adult polysomnograms.

Item Proposed	P	C
Test information
Type of study (eg, PSG, split PSG, PAP titration, OAT titration)	X
For titration studies indicate device(s) used (eg, mask, type of PAP/OAT)	X
AHI scoring criteria used	X
Interpretation
List primary diagnoses (include severity and positional component when applicable)	X
List other abnormal findings (eg, PLMS, bruxism, RWA, arrhythmias, hypoxia burden)	X
Sleep architecture: sleep efficiency, arousal index, sleep onset latency, stage R latency	X
Recommendations by diagnoses
OSA
If applicable, indicate recommendation for titration study with rationale vs APAP with close follow up if no known contraindications		X
For titrations: indicate evaluated and optimum settings	X
Treatment-emergent CSA
When applicable, include wording to indicate possible need for an RAD		X
If CPAP recommended, state AHI may trend down /improve with consistent use		X
If RADs recommended, recommend titration device/settings		X
CSA
Include possible causes		X
Sleep-related hypoxia
Include sleep time with oxygen saturation ≤ 88% and AHI		X
If oxygen recommended, include flow and mode (via PAP, nasal cannula)		X
Include assessment of causes/diagnostic workup		X
Sleep-related hypoventilation
Include recommendations for causes/diagnostic workup		X
Recommendations for treatment (eg, PAP titration study, CPAP vs RAD)		X
Other general recommendations
Sleepiness safety precautions (eg, avoidance of drowsy driving)		X
Weight management (goal BMI 18–25 kg/m2)		X
When to consider repeat testing (eg, weight changes, therapy evaluation, symptoms)		X
When to consider consultation with sleep medicine specialist		X
Obtain adequate sleep with regular sleep-wake schedule
Education for patients/caregivers/non–sleep providers by diagnosis
Consequences of untreated OSA		X
PLMS evaluation and management		X
Safety precautions and evaluation for parasomnias/RWA		X
Avoidance of tobacco and respiratory depressants including alcohol		X
Management suggestions of primary snoring		X

AHI = apnea-hypopnea index, APAP = autotitrating positive airway pressure, BMI = body mass index, C = consideration, CPAP = continuous positive airway pressure, CSA = central sleep apnea, OAT = oral appliance therapy, OSA = obstructive sleep apnea, P = primary, PAP = positive airway pressure, PLMS = periodic limb movements in sleep, PSG = polysomnography, RAD = respiratory assist device, RWA = rapid eye movement sleep without atonia.

Home sleep apnea tests

Detailed recommendations for home sleep apnea test (HSAT) reporting parameters have been listed in the AASM Scoring Manual¹⁰ and will not be relisted here. Additional considerations are listed below.

Scoring criteria

Scoring criteria should be clearly stated. The respiratory event index (REI) is currently reported for flow-based devices without electroencephalogram sleep data monitoring. The apnea-hypopnea index (AHI or pAHI) is reported in peripheral arterial tonometry devices that provide estimates of sleep time and architecture.

Does the test support the diagnosis of obstructive sleep apnea (OSA)? Attention to possible underestimation of severity if supportive of OSA

The AASM Scoring Manual recommends that the interpretation of the REI/AHI be reported, indicating whether or not the study supports the diagnosis of OSA.¹⁰ The AASM Scoring Manual also recommends classification of severity. The interpreter may indicate possible underestimation of severity followed by recommendation with clinical correlation.

Additional abnormal findings, if applicable

An HSAT is indicated for the diagnosis of OSA in uncomplicated patients and in conjunction with comprehensive evaluation and follow-up.¹⁷ Occasionally, the interpreter may encounter other abnormal findings, such as significant hypoxia, central sleep apnea (CSA), or periodic breathing. When observed, the interpreter should verify the accuracy of the findings by accessing the raw data and highlight findings in the report, listing them within the diagnoses. Including considerations for further evaluation with in-laboratory PSG is encouraged, given limited data on accuracy of HSAT devices on diagnoses other than OSA.¹⁷

Nondiagnostic studies

The AASM Scoring Manual recommends that, if an HSAT is not diagnostic, then recommend in-laboratory PSG if clinically indicated.¹⁰ As noted previously, HSATs commonly lead to underestimation of the REI/AHI. To avoid false reassurance after a nondiagnostic HSAT, the interpreter should note that a “negative” HSAT does not rule out sleep apnea and in-laboratory PSG should be performed, especially if the clinical suspicion for OSA remains given the negative impact on quality of life and potentially dangerous consequences on health.^17,18

Additional considerations

Management options for testing supportive of OSA diagnosis based on findings and available clinical history.

Any management recommendations should follow AASM practice guidelines and parameters.

Summary

Refer to Table 2 for a summary of recommendations.

Table 2.

Home sleep apnea test.

Item proposed	P	C
Test information
REI/AHI scoring criteria used	X
Interpretation
Indicate if testing supports or does not support diagnosis of OSA based on REI/AHI	X
Include severity of OSA (mild, moderate, severe) if applicable and highlight severity/REI/AHI may be underestimated	X
List other abnormal findings (eg, hypoxia, CSA)	X
Recommendations
For positive test: list possible treatment options. If applicable, indicate recommendation for titration study with rationale vs APAP with close follow up if no known contraindications.		X
For negative test: specify that a negative HSAT does not rule out OSA, recommend PSG for high-pretest probability for OSA	X
If OAT recommended or further adjustments needed based on residual OSA, indicate repeat testing with OAT in place		X

AHI = apnea-hypopnea index, APAP = autotitrating positive airway pressure, C = consideration, CSA = central sleep apnea, HSAT = home sleep apnea test, OAT = oral appliance therapy, OSA = obstructive sleep apnea, P = primary, REI = respiratory event index.

Pediatric polysomnograms

Primary diagnosis and its severity, if applicable

Clinical concern for sleep apnea is a frequent indication for obtaining a PSG in children. The International Classification of Sleep Disorders, third edition (ICSD-3), clearly delineates PSG criteria required, in conjunction with appropriate clinical features, for diagnosis of OSA or CSA in children.¹² Between ages 13 to 18 years, respiratory rules used to score events (pediatrics vs. adults) should be stated. The severity of observed OSA should be noted.

Oxygen distribution and gas exchange

Sleep-related hypoxemia and hypoventilation, as defined by the ICSD-3 should be highlighted if present.¹² These findings may or may not be due to coexisting sleep apnea, a distinction that the interpreting physician may sometimes be able to render. Nonapneic hypoxemia or gas exchange abnormalities noted during wake may be of clinical import and may help inform additional evaluation and management decisions.

Capnography is part of the standard montage in pediatric studies.¹² Both end tidal and transcutaneous capnographs may be used; the modality used should be noted on the report. In addition, sleep-related hypoventilation may be either obstructive (associated with snoring or increased work of breathing) or nonobstructive in nature, and, when available, supportive findings should be reported to make this determination. The finding of significant obstructive hypoventilation is of particular importance in pediatrics, as it represents a supportive criterion for the diagnosis of OSA even in the absence of an elevated AHI. Nonobstructive hypoventilation may have different etiologies and treatment considerations.

Remarkable features of sleep architecture

Although not typically the main indication for obtaining a pediatric PSG, a sleep study represents an opportunity to explore sleep staging and progression in a child. Overall sleep architecture may be influenced by a variety of factors. Sleep stage progression, distribution, and arousals develop in a typical fashion during childhood from infancy through adolescence, and normative data have been previously published.^15,19–23

The point of emphasis in this discussion is that the experienced pediatric sleep physician will recognize and report overall features of sleep architecture that are likely of clinical importance, rather than any one standalone metric that may fall outside of previously published norms.

Presence or absence of sleep-related limb or movement disorders

Several possible sleep-related movement disorders can be diagnosed during nocturnal PSG in children. Excessive PLMS (≥ 5/hour) are seen in approximately 5% of sleep studies performed in children and can be a result of many different possible causes, including periodic limb movement disorder, restless legs syndrome, certain medications, narcolepsy, sleep apnea, and selected chronic medical conditions.²⁴ Episodes of REM sleep behavior disorder or REM sleep without atonia should be noted and reported. Other movements that may be observed and reported include hypnic jerks, hypnagogic foot tremor, bruxism, or sleep-related rhythmic movement disorder.

Of note, clinicians may observe frequent large body movements during sleep that contribute to awakenings and sleep disruption. Recently, a new sleep-related movement disorder, restless sleep disorder, has been described and formal diagnostic criteria codified.^25,26 Interpreting pediatric sleep physicians should report if supportive features of this diagnosis are found during PSG.

Remarkable features observed on video clips in the report

Parasomnias, such as confusional arousals, sleep terrors, bruxism, and REM sleep behavior disorder may be observed and reported. In addition, parent–child interactions that occur during the sleep study may inform potential contributors to sleep challenges and may sometimes suggest a component of behavioral insomnia.

Observed work of breathing on video clips is an important feature to be included in the report, as discussed in Recommendation 1. OSA in children is best characterized by incorporating signs of increased work of breathing from video clips, such as snoring, snorting arousals, mouth breathing, neck hyperextension, and paradoxical respirations.

Presence or absence of arrhythmias

Limited electrocardiogram data are recorded during PSG, and any abnormalities of possible clinical import should be reported. Heart rate normally slows from wake to sleep, and definitions of bradycardia are highly dependent on child age.²⁷

Additional considerations

Next steps in management.

Management of sleep disorders in children is often challenging, and treatment decisions should be made within a shared decision-making paradigm involving the treating clinician and family. If the interpreting physician is not otherwise caring for the child, a range of treatment options should be included. As an example, while adenotonsillectomy is generally an initial consideration for many children with OSA, other options may be more appropriate in a given child based on age, airway and orthodontic characteristics, clinical symptoms, severity of disease, comorbidities, weight status, and family preferences²⁸; in addition, surgery may be higher risk in children with certain conditions, such as bleeding disorders. Furthermore, if surgery were pursued, a decision about the most appropriate operative setting would be needed. These decisions are complex and likely best made by the clinician who is physically seeing the child rather than a physician who is providing an interpretation of the sleep study without other clinical contact with the child or family.

Incorporation of prior polysomnographic data.

If prior PSGs have been obtained and are available for review, a brief notation of these results may be included, and any substantial changes on the current study may be noted. For example, follow-up studies after surgery for moderate or severe OSA are recommended.²⁹ Improvement post-treatment (or lack thereof) should be noted.

Summary

Refer to Table 3 for summary of recommendations.

Table 3.

Pediatric polysomnograms.

Item proposed	P	C
Test information
Type of study (eg, PSG, PAP titration)	X
For titration studies indicate device mode(s) used, settings, masks	X
AHI scoring criteria used: pediatrics vs adults	X
Interpretation
Describe the presence or absence of sleep apnea, its nature (obstructive vs central), and severity	X
Report any remarkable features of sleep architecture	X
Report the oxygen distribution and gas exchange	X
Report the presence or absence of sleep-related limb or movement disorders	X
Describe any remarkable features observed on video clips in the report	X
Report the presence or absence of arrhythmias	X
Recommendations
Diagnosis-specific recommendations for next steps in management		X
Miscellaneous
Incorporation of prior polysomnographic data		X

C = consideration, P = primary, PAP = positive airway titration, PSG = polysomnography.

Hypoglossal nerve stimulator titrations

Hypoglossal nerve stimulator surgery date

Reporting the implant date will allow the interpreter and the report reviewer to assess whether sufficient time for therapy adaptation was available.

Pertinent incoming hypoglossal nerve stimulator settings: amplitude and patient control range

Reporting the incoming settings, including the amplitude and patient control range should be included.

Hypoglossal nerve stimulator settings trialed during the study

When not included automatically by the computer program used for hypoglossal nerve stimulator (HNS) titration reporting, the amplitude range trialed during the titration should be documented. Any advanced setting changes made during the study should also be reported. Respiratory parameters to report in addition to AHI include oxygen desaturation index of 3% or 4% and percentage of time and minutes with oxygen level < 89%.

Most effective settings, supine and/or stage R sleep data

An optimal setting should be one that effectively reduces the number of obstructive events, resolves sleep apnea symptoms, and does not exceed arousal threshold.

Additional considerations

Positional residual sleep-disordered breathing events.

Persistence of sleep-disordered breathing (SDB) may observed with HNS in the supine position. When this finding is observed, adjunctive positional therapy may be of benefit for optimal control of SDB or allow for lower amplitudes if discomfort is present at higher settings.

Summary

Refer to Table 4 for summary of recommendations.

Table 4.

Hypoglossal nerve stimulator titrations.

Item proposed	P	C
Test information
Type of study (ie, HNS titration) and device	X
AHI scoring criteria used	X
HNS device information
Date device implanted	X
Incoming settings (ie, amplitude, patient control range)	X
Average use/week		X
Other pertinent history if available		X
Settings trialed during the study	X
Interpretation
Indicate whether or not SDB was well controlled with HNS during this study, including during supine and stage R sleep	X
List other abnormal findings (eg, PLMS, RWA, arrhythmias, hypoxia burden)	X
Recommendations
If optimal setting observed: indicate recommended setting.	X
If suboptimal titration: indicate if additional evaluation or adjustments are needed.	X
If residual SDB events were predominantly in the supine position, positional therapy may be recommended		X

Additional findings and management recommendations per adult PSG section. AHI = apnea-hypopnea index, C = consideration, HNS = hypoglossal nerve stimulator, P = primary, PLMS = periodic limb movements in sleep, RWA = rapid eye movement sleep without atonia, SDB = sleep-disordered breathing.

Multiple Sleep Latency Tests

Preceding PSG: total sleep time, AHI, and stage R latency

Mean sleep latency and the presence of stage R sleep during Multiple Sleep Latency Test (MSLT) naps are influenced by many factors, including the prior night’s sleep. For this reason, the practice parameters require that PSG be performed immediately before the MSLT during the patient’s usual major sleep period.³⁰

Urine drug screen results (or lack of results)

Many drugs can affect the validity of the MSLT data; therefore, reporting the urine drug screen data is important for the MSLT report reviewer to assess the validity of the results. The ICSD-3 notes that, in order for correct interpretation of MSLT results, a urine drug screen must be performed.¹²

Nap opportunities

Start and end times, latency from lights out to the first epoch of sleep, mean sleep latency (arithmetic mean of all naps or nap opportunities), and number of sleep-onset REM periods (defined as > 15 seconds of stage R sleep in a 30-second epoch) should be included.

Additional considerations

Actigraphy and/or sleep diary information.

Objective average total sleep time per day by actigraphy and/or sleep diary prior to the MSLT may be included in the report. The ICSD-3 strongly recommends that at least 1 week of actigraphy with sleep logs be performed prior to the MSLT to determine if the results are affected by another cause, such as insufficient sleep, shift work, or other circadian rhythm disorder.¹²

Is the study normal? Is the validity in question? Repeat testing or further evaluation recommendations may be included in the report.

The current MSLT practice parameters note that a repeat MSLT may be indicated when the MSLT is normal but narcolepsy is suspected. Other considerations of repeat testing include if the test is affected by factors that could disturb its validity or when ambiguous findings are present.³⁰

Summary

Refer to Table 5 for summary of recommendations.

Table 5.

Multiple Sleep Latency Test.

Item proposed	P	C
Test information
Type of study	X
Summary of data
Summary of findings from PSG (ie, TST, AHI, stage R latency)	X
Actigraphy data		X
Sleep diary data		X
UDS results	X
Nap data (start and end times, sleep latency, and R latency of each nap)	X
Number of SOREMPs	X
MSL	X
Diagnoses
Indicate if study is normal or abnormal and if findings are consistent with a specific diagnosis (eg, narcolepsy)	X
Document the presence of any known factors that could affect the validity of the results.	X
Recommendations based on diagnosis
Normal—if applicable, indicate consideration to retest based on possible false-negative results		X
Narcolepsy		X
Hypersomnia		X
Not diagnostic—indicate possible reasons/propose recommendations		X

AHI = apnea-hypopnea index, C = consideration, MSL = mean sleep latency, P = primary, PSG = polysomnography, SOREMP = sleep-onset rapid eye movement period, TST = total sleep time, UDS = urine drug screen.

DISCUSSION AND FUTURE DIRECTIONS

We offer a practical, complete, and relevant document and a structure that can be implemented across sleep centers and does not burden the interpreter. We anticipate that some readers will opine that some of the content is beyond the scope of what the interpreter physician needs to include, while others will propose missing data that they feel should have been included. We feel that the flexibility of the proposal accommodates for this and allows for a “first step” toward standardization of physicians reporting of sleep studies. Some interpreters may consider including information that, if mentioned in the report, may improve adherence to treatment and follow-up; this may be especially useful in the care of patients with comorbidities and particular vulnerabilities, may help with limiting inequities in health care, especially related to mental health and developmental disabilities, and supports personalized medicine being more inclusive as the “audience” or recipient of the report may vary with the area of practice and aspirational in regard to the “Open Notes” standard.

DISCLOSURE STATEMENT

All authors have seen and approved the manuscript. The authors report no conflicts of interest.

ABBREVIATIONS

AASM

American Academy of Sleep Medicine

AASM Scoring Manual

The AASM Manual for the Scoring of Sleep and Associated Events: Rules, Terminology and Technical Specifications

AHI

apnea-hypopnea index

CSA

central sleep apnea

HNS

hypoglossal nerve stimulator

HSAT

home sleep apnea test

MSLT

Multiple Sleep Latency Test

OAT

oral appliance therapy

OSA

obstructive sleep apnea

PAP

positive airway pressure

PLMS

periodic limb movements in sleep

PSG

polysomnography/polysomnogram

REI

respiratory event index

SDB

sleep-disordered breathing