I propose changes in the definition of EEG arousals so that the number of patients who have obstructive sleep apnea/hypopnea will not be reduced based upon an edict of the steering committee. Specifically, I propose the recognition of two types of arousals: spontaneous and caused. It is suggested that the restrictive parts of the arousal definitions such as requiring 10 seconds of stable sleep prior to scoring an EEG arousal or requiring a one-second increase in the amplitude of the chin EMG for an arousal during REM sleep be applied only to spontaneous arousals and not to the various types of caused arousals. It is recommended that caused arousals be subdivided into several subtypes, such as hypopnea-caused arousals and periodic limb movement-caused arousals among others, and that these caused arousals be exempt from the “10 seconds of stable sleep” preceding the arousal rule.
A Practice Guidelines and Scoring Manual FAQ section of the AASM website has a question designated as M.3. The answer to this question was so troubling that it prompted this letter to the editor. The answer provided by the Steering Committee stated that “a minimum of 10 seconds (of stable sleep) is necessary to reliably determine that that patient has returned to sleep” answer is wrong, contradicts other aspects of the scoring manual, and if implemented for hypopneas as well as for leg jerks, this edict would reduce the number of patients who can receive a diagnosis of obstructive sleep apnea/hypopnea. The reason is that if all arousals are required to have 10 seconds of “stable sleep” preceding their identification, then a significant number of potential EEG arousals that follow a hypopnea would no longer be able to be scored as arousals because of the absence of “10 seconds of stable sleep.”
We fully agree with the definition of an EEG arousal as defined in the ASDA Report EEG Arousals: Scoring Rules and Examples (Sleep 1992, pages 173–184), but only if it applied to spontaneous arousals. This is also noted on page 33 of the scoring manual, which states:
“Score an EEG arousal when there is an abrupt shift in frequencies, alpha, theta, or frequencies 16 Hz and above (but not spindles). The duration must be 3 seconds or longer and there must be 10 seconds of stable sleep preceding the change.”
The two key words to focus on in that definition are “stable sleep.” What is stable sleep? Stable sleep can be defined by any combination of low-amplitude, mixed-frequency 4 to 7 Hz EEG characteristic of stage N1, vertex sharp waves, sleep spindles not followed by alpha, K-complex also not followed by alpha, sawtooth waves, and delta waves with no visible alpha intrusions.
What are the signs of transition from wakefulness to sleep? They are alpha attenuation, alpha slowing, a decrease in the clarity of the alpha rhythm, the appearance of vertex sharp waves, insipient sleep spindles, slow eye movements, and/or brief periods of microsleep defined as any recognizable low-amplitude, 4–7 Hz mixed frequency EEG that is visibly different from the recognizable EEG patterns of alpha or beta activity.
In contrast, what are the EEG signs of transition from sleep to wakefulness? They are the appearance of at any recognizable alpha rhythm, alpha augmentation, an increase in alpha clarity or density, an increase in alpha frequency, the appearance of beta activity (16–30 Hz), and/or the presence of muscle or movement artifact in the EEG.
When scoring a spontaneous EEG arousal, ensure that there is 10 seconds of really stable sleep preceding the change in EEG that contains background EEG activity of low-amplitude, 4–7 Hz-mixed-frequency EEG, with no hint at all of alpha, beta, or any synchronized fast activity.
Apply the 10 seconds of stable sleep rule to only the spontaneous arousals, but not to any of the caused arousal types. When scoring a hypopnea-caused arousal, be attentive to even the most subtle changes that could be interpreted as a transition from sleep to wakefulness. If the EEG during a hypopnea shows any perceptible increase in alpha amplitude, alpha clarity, alpha frequency, even the slightest hint of beta or muscle activity from the immediately preceding EEG in the central or occipital regions, score a hypopnea-caused arousal regardless of whether or not there are 10 seconds of stable sleep preceding the arousal.
The author has indicated no financial conflicts of interest.
Zimmerman J. Stability versus transitional changes in the EEG: from sleep to wakefulness. J Clin Sleep Med 2015;11(4):495.