The pioneering work of Moldofsky and colleagues yielded two critically important findings with regard to the relationship of sleep to widespread musculoskeletal pain and fatigue in rheumatological patients without known medical disease:

Manifestations of disturbed sleep typically comprise the experience of light and unrefreshing sleep, nocturnal restlessness with frequent awakenings, and at times, sleep-related loud snoring and breathing problems, or periodic involuntary leg movements. Typically, there is a variation in the diurnal pattern of symptoms. Upon awakening from their poor quality of sleep, patients usually experience generalized muscular stiffness, diffuse pain, and profound fatigue, which are unchanged or even increased compared to the previous evening. They report improvement in symptoms with improved energy from late morning to mid-afternoon; then, thereafter there is a downhill course in symptoms as the day progresses. The narrower the period of improvement, the more disabled is the patient. On the rare occasion, they might achieve a restful night of sleep and improved symptoms on the following day. Indeed, unrefreshing or nonrestorative sleep is closely correlated to the widespread pain and tender points in FM while psychological distress is not [12]. FM subjects with light unrefreshing sleep have diurnal impairment in speed of performance on complex cognitive tasks, which are accompanied by sleepiness, fatigue, and negative mood. Such psychological impairment could account for the functional disabilities that are encountered in a work environment and in social behavior. There may be an associated predisposition to emotional hypersensitivity resulting from distrust and insecure attachment in their personality, which is shown to be associated with the alpha-EEG anomaly during sleep [13].

These disabling behavioral symptoms are commonly associated with disturbances in the physiology of sleep [8, 9]. How these FM symptoms are linked to the subjective poor sleep quality and objective PSG measures continues to be of considerable interest in determining the pathophysiology of this illness. Early on, aspects of anomalous changes in EEG sleep physiology were identified in FM patients that were considered to be consistent with subjective poor quality of sleep. Nocturnal polysomnography (PSG) shows a prominent EEG alpha frequency (8–12 Hz) rhythm in the central EEG regions during stages 2 and slow-wave sleep (SWS) NREM sleep (see Fig. 48.1) but less frequently in patients with severely diminished or absent delta SWS [8]. As Hauri and Hawkins had noticed this alpha EEG anomaly to coincide with the delta frequency (0–<4 Hz) slow waves of SWS, they coined the term alpha–delta sleep, which they described in a small group of not depressed psychiatric patients with poorly understood illness who complained of chronic, somatic malaise, and fatigue [14]. Subsequently, Moldofsky and colleagues reported this anomaly in patients with so-called fibrositis [11], later considered by rheumatologists to be a wastebasket diagnosis because of the lack of replicated evidence of presumed inflammation of fibrous tissue. Subsequently, the symptoms were relabeled as FM or FMS. This anomalous alpha EEG sleep is seen predominantly in the frontal–central brain area in contrast to the alpha EEG frequency occurring in quiet wakefulness that is localized in the occipital region [15, 16]. Subsequently, reports of this anomalous increased alpha EEG activity in NREM sleep were reported in FM patients [8, 17–21]. Patients with FMS, however, are heterogeneous with respect to exhibiting the alpha anomaly [21]. This alpha activity in NREM sleep is not specific to FM, since it occurs in patients with other painful rheumatic diseases. Moreover, it is described to occur in patients with primary insomnia [22]. As EEG alpha-like rhythm is associated with mentation [16, 18], alpha occurring in NREM sleep may represent a vigilant arousal state that interferes with bodily restorative functions [8]. This abnormal EEG sleep feature may account for a common aspect of impaired sleep quality with symptoms of light sleep and sleep mentation with external environmental awareness and internal self-awareness of subjective discomfort. This sense of vigilance is shown to be associated with an underlying faulty personality characteristic in such people with the prominent alpha EEG sleep anomaly and physical illness which stems from a faulty interactional pattern of childhood personality development . In formal personality testing of such people with the alpha EEG sleep anomaly, most of whom have FMS, they show a pattern of overall distrust and insecurity in interpersonal relationships. Consequently, this pervasive sense of insecurity and distrust, which are ingrained in their personalities, is thought to make them difficult to accept any medical therapeutic effort that results in perpetuation of chronic physical illness behavior [13].

A second physiological component of impaired sleep quality is thought to be a specific EEG pattern that contributes to unstable or nonrestorative sleep, a major diagnostic feature of FMS and CFS. This objective physiological feature in the sleep EEG sleep comprises a high frequency of the CAP. The CAP phenomenon occurs variably throughout NREM sleep characterized by a frequency of approximately 20–30 cycles per second (see Fig. 48.1). The specific type of CAP is reported to reflect a measure of relative stable–unstable sleep. In patients with FMS and CFS where unstable or nonrestorative sleep is commonly reported, there is an increase of CAP activity. The sleep EEG dominant rhythm of CAP A1 and no significant ancillary increase in autonomic or peripheral EMG activity, as seen in CAP A2 and increased in CAP A3, is associated with sleep stability that is commonly found in young asymptomatic healthy people. In the subgroup CAP A2 and CAP A3, there is physiological evidence for progressive increases in sleep instability [23–25]. This high amount of CAP correlates with the severity of pain as measured by the number of tender points in FMS [23]. The EEG CAP A2 and especially CAP A3 are associated with pathological sleep characterized by physiological evidence of motor and autonomic nervous system abnormalities, e.g., periodic limb movement (PLM) and sleep apnea disorders [24, 26]. Such clinically significant periodic PSG sleep disorders are found in a large sample of US patients with FMS. In this multicenter PSG study of more than 200 FMS patients, 94 % were female, with a mean body mass index of about 30 who had FMS for more than 5 years. Approximately 15 % had moderate to severe apnea–hypopnea respiratory index and 20 % had PLM disorder [24]. Their sedentary behavior with resultant weight gain and obesity may have promoted the sleep apnea disorder and the coincident CAP A2 and A3 EEG unstable and nonrestorative sleep .

Another feature of sleep EEG disturbance, which is reported in FMS, is a decrease in stage 2 EEG sleep spindles [27]. The presence of EEG sleep spindles, a characteristic of this stage of NREM sleep, is hypothesized to be a physiological feature of sleep stability. This hypothesis is supported by the shorter duration of stage 2 sleep periods, which is normally dominated by CAP A1. In addition, there is increased frequency of EEG sleep stage shifts [28], increased sleep awakenings, and decreased sleep efficiency (ratio of time asleep/time in bed) [23, 29].

Moreover, there is a reduction of SWS in FM patients [11, 19–21, 29]. One such study showed that decreased SWS is correlated with high-phasic EEG alpha rhythm [21]. The finding of reduced SWS in FMS patients also coincides with nocturnal neuroendocrine abnormalities of increased nocturnal cortisol levels and decreased release of growth hormone (GH) [30]. Quantitative studies suggest that the duration of SWS serves to regulate homeostasis and is correlated with the amount of prior waking. That is, SWS increases following extended wakefulness and is decreased during nocturnal sleep following a daytime nap containing SWS [31]. This decreased amount of SWS in FM might indicate an impairment of the SWS homeostatic drive.

Some sleep abnormalities in FM patients have not been consistently reported. In some studies, for example, FM patients are reported to be not significantly different from controls with respect to alpha sleep EEG activity or the duration of SWS [28, 32]. It is unclear whether these discrepancies result from heterogeneity of sleep pathology in FM patients, technical considerations such as differences in scoring of sleep EEG alpha anomalies, or the night-to-night variability in many sleep measures, including SWS [33].