Diagnostic Assessment Methods in Adults

Overview

Thousands of people undergo “sleep studies” every day. These studies involve collecting electrophysiologic data in hospitals, freestanding sleep disorders centers, and patients’ homes. The vast majority of these studies rely on determining the presence, type, and severity of sleep-disordered breathing. A small percentage help diagnose narcolepsy, nocturnal seizures, parasomnias, and other sleep disorders, and a few will be conducted for research purposes. Until very recently, recording parameters, how tests were conducted, techniques applied for record review, and procedures used for data reduction varied substantially from one test facility to another. The increasing standardization largely stems from the American Academy of Sleep Medicine (AASM) developing and publishing a standardized clinical polysomnography (PSG) manual and a trend for third-party insurance carriers (and government programs) to require facility accreditation for procedure reimbursement. This chapter describes, summarizes, and discusses current AASM standards.

The diversity in methods and techniques of sleep research likely arises from the origin of “sleep medicine” as a research endeavor. Discovering recordable bioelectric correlates for suspected sleep problems paved the way for adopting the psychophysiologic research method known as polysomnography as the major procedure in sleep medicine. However, research methods designed to chart new frontiers in knowledge necessarily use diverse methods. Furthermore, researchers continually modify these methods to increase sensitivity, incorporate new technology, and widen the scope of inquiry. Nevertheless, as clinical utility for PSG grew, some de facto standards gained acceptance, and expert panels developed guidelines. The AASM manual was based largely on existing techniques, and its content represents the efforts of many individuals. Previous guidelines included the standardized manual for terminology, techniques, and scoring (see next section); the guidelines developed by the American Sleep Disorders Association (ASDA) task forces on arousal scoring and periodic leg movement scoring; and work by “the Chicago Group” on sleep-related breathing. The AASM manual, developed through the use of evidence-based research techniques, provides a single information source that describes how to record, score, and distill human sleep data for clinical purposes.

Sleep Staging

In 1937, Loomis and colleagues recorded the first known, continuous, all-night electrophysiologic sleep study. It became imperative to distill the massive amount of data collected into a manageable form immediately. Thus they invented sleep staging. During the following three decades, an assortment of sleep stage classification schemes emerged, culminating in the standardized technique described in A Manual of Standardized Terminology, Techniques and Scoring System for Sleep Stages in Human Subjects. Doctors Allan Rechtschaffen and Anthony Kales chaired this project, thus the origin of its nickname, the “R and K manual.” It focused on normal sleep and established rules for sleep staging data recorded from adults. As pointed out in the first AASM scoring manual preface, “the rapidly emerging field of sleep medicine requires a more comprehensive system of standardized metrics that considers events occurring outside of normal brain activity.”

Recording

Standard PSG practice uses three types of electrophysiologic activity to categorize sleep into four different stages and differentiate it from wakefulness. The three electrophysiologic activities are 1) brain activity, 2) eye movements, and 3) skeletal muscle tone.

Electroencephalogram for Brain Activity

For classifying sleep stages, the AASM manual recommends recording frontal, central, and occipital electroencephalograms (EEGs)—specifically, monopolar derivations from F₄, C₄, and O₂ linked to the contralateral mastoid (M). Backup electrodes are placed at homologous sites at left scalp loci. An alternative recording montage allows substitution of midline bipolar recordings from frontal and occipital derivations (Table 18-1, Figs. 18-1 through 18-3).

Table 18-1

Recording Montage for Sleep Staging and Respiratory Monitoring

C_z, central zero (midline); E₁, left eye; E₂, right eye; ECG, electrocardiogram; EEG, electroencephalogram; EMG, electromyogram; EOG, electrooculogram; f, frequency; F_p, frontal pole; Hz, Hertz; M, mastoid; O_z, occipital zero (midline).

^*Place electrode 1 cm below the eye’s outer canthus.

Figure 18-1 Classic Rechtschaffen and Kales derivations for recording the electroencephalogram.
Electrode locations as recommended in 1969. Electrode pairs are indicated by the same colors.

Figure 18-2 Updated American Association of Sleep Medicine (AASM) recommended derivations for recording the electroencephalogram.
Electrode locations as recommended in 2007 by AASM.

Figure 18-3 Updated American Association of Sleep Medicine alternative derivations for recording the electroencephalogram.
Alternative electrode locations as recommended in 2007. Backup electrode substitutions: O₁ for O_z, F_pz for F_z, and C₃ for C_z.

Eye Movements

To detect eye movements, electrodes are placed near the eyes’ right and left outer canthi (E₂ and E₁, respectively). The manual recommends a monopolar electrooculographic (EOG) montage, with E₂ placed 1 cm above and E₁ placed 1 cm below the outer canthus. When a lateral eye movement occurs, the positive corneal potential moves toward one electrode and away from the other. Therefore this electrode arrangement produces robust right-versus-left EOG out-of-phase activity when horizontal eye movements occur. Consequently, eye movements are easily differentiated by frontal EEG activity proximal to the EOG electrodes, which appears as an in-phase signal. Staggering the electrode placements slightly above and slightly below each eye’s horizontal plane allows some limited appreciation of vertical eye movements on the PSG. An alternative recording montage that allows better visualization of vertical eye movements (albeit sacrificing the in-phase/out-of-phase differentiation by frontal EEG activity) has both EOG placements 1 cm below the outer canthi and referenced to the middle of the forehead (see Table 18-1).

Electromyogram for Skeletal Muscle Tone

Finally, to record a submentalis electromyogram (EMG), one electrode is placed 1 cm above the mandible’s inferior edge (on the midline); the other two are placed 2 cm below and 2 cm to the right and left of midline, respectively. Recordings should reference one of the two electrodes below the mandible to the one above, with the remaining electrode for backup. Because of this placement, submentalis EMG is commonly referred to as “chin EMG.”

Staging

Classification

Sleep-stage classification involves characterizing the predominant electrocortical activity over a set length of time. This time-domain categorization proceeds for each specified time interval, called an epoch. The AASM manual standardized the epoch length at 30 seconds, which corresponds to what had been one page when paper recordings were made, with the chart drive set to 10 mm/sec. Overall, the sleep process is classified into three distinct conceptual organizational states of the central nervous system (CNS): wakefulness, rapid eye movement (REM) sleep, and non-REM (NREM) sleep. On the basis of EEG, EOG, and EMG activity, each recorded epoch is classified as either stage N1 (NREM 1), N2 (NREM 2), N3 (NREM 3), R (REM), or W (wakefulness).*

Waves

Six EEG waveforms are commonly used to differentiate states and classify sleep stages: 1) alpha activity, 2) theta activity, 3) vertex sharp waves, 4) sleep spindles, 5) K-complexes, and 6) slow-wave activity (Table 18-2).

Table 18-2

Sleep Electroencephalogram Waveforms

AASM, American Association of Sleep Medicine; EEG, electroencephalogram; SEM, slow eye movement.

Stages

Stage W.

In patients with well-defined alpha activity, wakefulness and sleep are easily differentiated. Stage W is scored when alpha activity occupies 15 or more seconds of an epoch (Fig. 18-4). By contrast, when clear alpha activity is absent, differentiating wakefulness from sleep can, at times, be difficult.

Figure 18-4 Epoch of stage W.
This figure depicts 30 seconds of polysomnographic activity characterizing stage W (wakefulness) according to AASM scoring criteria. C₃, left central scalp derivation; E₁, left outer canthus eye electrode location; E₂, right outer canthus eye electrode location; EMG_SM, surface electrode electromyogram (EMG) from submentalis muscle; F₃, left frontal scalp derivation; M₂, right mastoid electrode location; O₃, left occipital scalp derivation.

Stage N1.

Vertex sharp waves often herald sleep onset and thereby provide a valuable landmark for recognizing stage N1 sleep. Wakefulness usually transitions to either stage N1 or N2. Stage N1 is marked by a general slowing of background EEG activity with the appearance of theta and vertex sharp waves. N1 may also be marked by the cessation of blinking and saccadic eye movements and by the appearance of slow eye movements (Fig. 18-5). Although not required, N1 can be discerned by its low-voltage, mixed-frequency background EEG containing theta activity in the absence of slow waves, sleep spindles, K-complexes, prominent alpha activity, or REM. Sleep stage scoring reliability for stage N1 is substantially lower than for any other stage; this holds true for both interscoring and intrascoring agreement. This problem arises from stage N1 being largely defined by rules of exclusion rather than by clearly recognizable electrophysiologic events. Luckily, stage N1 is transitory and accounts for only a small percentage of the night’s sleep complement.

Figure 18-5 Epoch of stage N1.
This figure depicts 30 seconds of PSG activity characterizing stage N1 sleep according to AASM scoring criteria. C₃, left central scalp derivation; E₁, left outer canthus eye electrode location; E₂, right outer canthus eye electrode location; EMG_SM, surface electrode EMG from submentalis muscle; F₃, left frontal scalp derivation; M₂, right mastoid electrode location; O₃, left occipital scalp derivation.

Stage N2.

By contrast, stage N2 is easily recognized by sleep spindles or K-complexes (Fig. 18-6) that occur on a low-voltage, mixed-frequency background EEG in the absence of significant slow-wave activity (i.e., less than 6 seconds of 75 µV or more).

Figure 18-6 Epoch of stage N2.
This figure depicts 30 seconds of PSG activity characterizing stage N2 sleep according to AASM scoring criteria. C₃, left central scalp derivation; E₁, left outer canthus eye electrode location; E₂, right outer canthus eye electrode location; EMG_SM, surface electrode EMG from submentalis muscle; F₃, left frontal scalp derivation; M₂, right mastoid electrode location; O₃, left occipital scalp derivation.

When 6 or more seconds of slow-wave activity are present in an epoch, it is scored as stage N3 (Fig. 18-7). The designation N3 generally conforms to activity previously designated as slow-wave sleep because of the highly synchronized low-frequency activity. Slow-wave activity scoring requires 75 µV or greater peak-to-peak amplitude from a frontal derivation. However, it is important to realize that the recording montage dramatically affects EEG amplitude. Thus if a frontal EEG is recorded using the AASM alternate bipolar montage (F_z-C_z), amplitudes will be lower, and N3 sleep will be reduced. The AASM manual does not provide a guideline for amplitude adjustment; therefore stage N3 scoring should derive from the monopolar central lead included in the alternate montage.

Figure 18-7 Epoch of stage N3.
This figure depicts 30 seconds of PSG activity characterizing stage N3 sleep according to AASM scoring criteria. C₃, left central scalp derivation; E₁, left outer canthus eye electrode location; E₂, right outer canthus eye electrode location; EMG_SM, surface electrode EMG from submentalis muscle; F₃, left frontal scalp derivation; M₂, right mastoid electrode location; O₃, left occipital scalp derivation.

Stage R.

Criteria for scoring REM sleep—or stage R sleep, as it has been renamed, although most will likely continue to call it REM sleep—remained essentially unchanged. Stage R is characterized by low-voltage, mixed-frequency EEG, very low chin EMG levels, and REM (Fig. 18-8). Sawtooth activity represents a unique variant of theta activity, containing waveforms with a notched or sawtooth-shaped appearance, frequently observed during stage R (Table 18-3).

Table 18-3

Electrophysiologic Characteristics of Each Sleep Stage

EEG, electroencephalogram; EMG, electromyogram; s, seconds.

^*With eyes closed.

^†Special form of theta wave called sawtooth owing to its appearance.

^‡For purposes of sleep staging, monopolar slow-wave activity from frontal or central derivations must be ≥75 µV.

Figure 18-8 Phasic and tonic rapid eye movement (REM) sleep.
This figure depicts PSG activity characterizing stage R sleep (REM sleep) according to AASM scoring criteria. A, Stage R, with phasic bursts of REM (phasic REM). B, A quiescent period of stage R sleep (tonic REM). C₃, left central scalp derivation; E₁, left outer canthus eye electrode location; E₂, right outer canthus eye electrode location; EMG_SM, surface electrode EMG from submentalis muscle; F₃, left frontal scalp derivation; M₂, right mastoid electrode location; O₃, left occipital scalp derivation.

Smoothing Rules

When an epoch contains characteristic features for more than one sleep stage, it is scored according to the characteristics that make up its majority. One source of scoring difficulty arises when epochs without actual eye movements are contiguous with REM sleep. The AASM manual does not address the concept of differentiation between phasic and tonic REM sleep (see Fig. 18-8). Nevertheless, this concept is widely used in research and can provide useful instruction in this matter. The term phasic REM sleep refers to epochs scored as stage R clearly accompanied by eye movements and sometimes by other phasic events. By contrast, an epoch designated as tonic REM sleep has the same background EEG activity, nearly completely diminished chin EMG, but lacks eye movements. If such an epoch occurred in isolation, it would likely be scored as N1; however, when surrounded by epochs of phasic REM sleep, it is considered stage R. In fact, N1-like epochs without eye movement contiguous with REM sleep continue to be scored as stage R until some indication occurs that another sleep process has emerged. Indicators can be a spindle, a K-complex, a CNS arousal, an increase in chin EMG, a body movement followed by slow eye movements, or the appearance of another scorable sleep stage in the first half of the epoch. The AASM manual provides examples to guide scoring decisions at stage transitions.

Arousal Scoring

Electroencephalogram Speeding and Central Nervous System Arousals

Sleep staging forms what may be called the “building blocks” of sleep macroarchitecture. Staging provides information useful for forming a general impression of sleep’s overall continuity and integrity. Staging also serves as a biomarker for detecting gross disturbances and for correlating pathophysiology with the organizational state of the CNS (i.e., REM, NREM, or wakefulness). Staging parameters, however, fail to capture transient sleep disturbances, which last less than 15 seconds. Therefore techniques are needed to identify brief CNS arousals that fail to meet the epoch scoring criteria for wakefulness.

CNS arousals involve an abrupt shift during sleep to faster EEG activities—including theta, alpha, and beta but not sleep spindles—for 3 seconds or longer. CNS arousal scoring rules largely derive from a previous concept called EEG speeding. Usually, the frequency shift manifests as alpha activity and is best visualized in occipital leads. For the shift to qualify as an arousal, the person must have been asleep for at least 10 seconds. Finally, during REM sleep, the EEG shift must be accompanied by at least 1 second of increased chin EMG tone (Fig. 18-9) because alpha bursts routinely appear during REM sleep and do not intrinsically represent pathophysiology. The 3-second duration was not arbitrary; it was the minimum arousal duration reliably scored by hand among the task force members. Undoubtedly, digital systems could reliably score shorter arousals, but the clinical significance of such events is not known.

Figure 18-9 Central nervous system arousals.
Central nervous system (CNS) arousals from non-REM (A) and REM sleep (B). C₃, left central scalp derivation; E₁, left outer canthus eye electrode location; E₂, right outer canthus eye electrode location; EMG_SM, surface electrode EMG from submentalis muscle; F₃, left frontal scalp derivation; M₂, right mastoid electrode location; O₃, left occipital scalp derivation.

Cyclic Alternating Pattern

The cyclic alternating pattern (CAP) refers to an EEG pattern seen during sleep (Fig. 18-10) in which EEG activity bursts—usually a high-amplitude slow, sharp, or polymorphic wave burst—alternate with periods of quiescence. The burst constitutes the A phase, and the quiescence is labeled the B phase. The Parma Groups, led by Terzano and Parrino, organized an international workshop during which recording and scoring techniques were codified. In essence, the A phase was subcategorized into three types based on the presence and duration of intermixed or intermingled EEG alpha activity. Type A1 contains little or no alpha EEG activity and consequently does not meet AASM arousal scoring criteria. Type A2 often meets AASM arousal criteria, and A3 meets AASM criteria 95% of the time. Sleep associated with A1 bursts is conceptualized as unstable but is preserved by a descending cortical response that helps reinforce a protective thalamic gating mechanism. By contrast, if the attempt to reinforce this gate fails, an arousal or awakening occurs, and an A3 burst is observed. A2 responses fall somewhere in between (Fig. 18-11).

Figure 18-10 Cyclic alternating pattern (CAP).
This figure depicts polysomnographic activity characterizing phase A (burst) and phase B (quiescent) activity. This period of CAP activity appears not to be associated with either periodic leg movements or breathing abnormalities. AB_Mvmt, abdominal movement measured with inductive plethysmography; C₃, left central scalp derivation; E₁, left outer canthus eye electrode location; E₂, right outer canthus eye electrode location; EMG_Leg, surface electrode EMG from left and right anterior tibialis; EMG_SM, surface electrode EMG from submentalis muscle; F₃, left frontal scalp derivation; Flow, airflow measured with nasal-oral thermistors; M₂, right mastoid electrode location; O₃, left occipital scalp derivation; RC_Mvmt, rib cage movement measured with inductive plethysmography.

Figure 18-11 Examples of cyclic alternating pattern A phase types 1, 2, and 3 activity.
Type 1 does not appear to have EEG activity suggestive of central nervous system (CNS) arousal. By contrast, type 2 activity suggests possible arousal, whereas type 3 meets AASM criteria for CNS arousal from NREM sleep.

Sleep-Related Breathing Disorders

It has taken many years and substantial educational effort, but the medical community now largely understands that sleep-related breathing disorders do not exclusively affect hypoventilating, morbidly obese patients. An estimated 90% or more of all PSG studies performed on any given night are conducted to diagnose or assess treatment for sleep-disordered breathing (SDB). Ironically, recording and scoring techniques for evaluating SDB were among the last to be standardized.

Recording Technique

Key features needed to assess SDB include airflow, respiratory effort, blood oxygenation, and sleep disturbance. The AASM manual recommends recording the four data channels when clinically evaluating adult patients with overnight PSG (see Table 18-1 for sampling rates and filter settings). This is done with 1) a thermal sensor at the nose and mouth to detect apnea; 2) a nasal pressure transducer to detect hypopnea; 3) either an esophageal manometer or chest/abdominal inductance plethysmograph—or, alternatively, intercostal EMG—to detect respiratory effort; and 4) a pulse oximeter with its signal averaged over 3 seconds or less.

Although end-tidal CO₂ measurement is not recommended for adults by the AASM manual, it can be useful for quantifying hypoventilation. This volume contains illustrations of end-tidal CO₂ recordings. Snoring sounds can help differentiate different types of SDB events, even though the means to use this parameter is not specified in the manual.

SDB scoring is event based, not time-domain based. Consequently, the 30-second fixed epoch length is not necessarily recommended. Scoring mainly involves identifying, counting, and characterizing discrete sequences of SDB events. Sleep-related respiratory signals are considerably slower than variations in EEG-EOG-EMG activity. Therefore viewing tracings over longer time frames can facilitate breathing pattern and event recognition. The viewing flexibility provided by digital PSG reflects a great advantage when scoring and interpreting sleep-related respiratory impairments. Nonetheless, the 30-second fixed epoch length specified by the AASM manual applied to sleep staging does affect SDB scoring. Sleep onset can occur during an epoch scored as stage W if it begins in the latter portion of an epoch. When apnea occurs in such a case, some tabulation programs do not count it because it is not recognized as sleep related. Similarly, when a sleep-related respiratory event occurs early enough during an epoch to provoke an awakening that renders the epoch to be staged as wakeful, again, some software fails to count the episode. Clinicians must determine how their digital systems handle such disconnects and must adjust summaries accordingly.

< div class='tao-gold-member'>

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