Parasomnias are disorders of arousal during sleep or dysfunctions associated with particular sleep stages. These disorders are manifestations of abnormal or excessive activation of the central nervous system. Parasomnias do not necessarily lead to a complaint of insomnia or hypersomnia, although in their most severe forms either one of these difficulties may be present. Although they are often simply undesirable phenomena, some parasomnias may cause physical injuries (sleepwalking, REM behavior disorder) and significant distress (sleep terror).

DSM-IV lists three specific parasomnias, whereas the ICSD lists 24 specific types. In some of the treatment literature, parasomnias are not well differentiated. For example, nightmare disorder and sleep terror disorder have not always been clearly distinguished, and some patients (i.e., those with posttraumatic stress disorder [PTSD]) may have features of both nightmares and sleep terrors.

Nightmare Disorder

Nightmares are frightening dreams occurring during REM sleep. They usually involve vivid imagery of impending danger that is depicted in scenes of chasing, falling, or even killing. Nightmares typically lead to full awakenings during which the dream content is vividly recalled. Residual anxiety may interfere with the ability to return to sleep. Nightmares are associated with limited verbal or motor components. Autonomic arousal is elevated but less intense than in sleep terror, and the individual becomes oriented and alert quickly after awakening. Nightmares are considered a hallmark feature of PTSD (Ross et al. 1989). They can also be associated with the use of medications, including sedative-hypnotics, β-blockers, and amphetamines (Thompson and Pierce 1999). The occasional experience of nightmares is also fairly common in both children and adults, but it is unclear what proportion of these individuals are clinically distressed by nightmares.

Behavioral/psychotherapeutic treatment Although uncontrolled studies have suggested that psychotherapy may be useful, the nonpharmacological management of nightmares has relied mostly on anxiety reduction methods such as relaxation, systematic desensitization, and rehearsal (Halliday 1987). Desensitization and rehearsal are two exposure-based interventions that facilitate emotional processing of the underlying anxiety. These procedures are intended to eliminate nightmares by repeated exposures to the fearful stimuli. Desensitization involves gradual exposure, after the patient has been trained in relaxation, to a hierarchy of increasingly anxiety-provoking scenes. Rehearsal is a procedure that consists of writing down a narrative of the nightmare, changing its course or outcome, and rehearsing the revised version several times a day.

The findings from four group studies (Kellner et al. 1992; Krakow et al. 1995; W. R. Miller and DiPilato 1983; Neidhardt et al. 1992) indicated that the average frequency of nightmares is decreased from 8-12 per month at baseline to fewer than 1-2 per month after treatment. Approximately 80%-90% of treated patients were either free from nightmares or experienced them less than once a month after treatment. The duration of treatment varied between a single therapy session and six weekly sessions. Therapeutic gains were generally well maintained at short-term follow-up evaluations, and in one 18-month follow-up study (Krakow et al. 1996), 68% of the treated patients had decreased their nightmares below criteria for a chronic nightmare disorder. There was no differential treatment effectiveness in decreasing frequency or intensity of nightmares, and all active treatments were more effective than placebo or self-monitoring alone. Exposure-based methods, however, may facilitate emotional processing and produce better outcomes on measures of psychological distress (W. R. Miller and DiPilato 1983; Neidhardt et al. 1992). Although these preliminary results are impressive, future studies need to include longer-term follow-up evaluations, particularly in light of the very persistent nature (10-year duration) of nightmares before treatment.

Pharmacological treatment The pharmacological treatment of nightmare disorder has not been well described in controlled clinical trials. Case series suggest the efficacy of benzodiazepines (Allen 1983). In patients for whom psychotherapy or other psychological treatment has been ineffective in suppressing disturbing nightmares or in circumstances when a rapid response is required, an empirical trial of either a benzodiazepine or an antidepressant is warranted. When benzodiazepines are used, beneficial effects are likely to result more from suppression of arousals than from REM sleep suppression. Conversely, antidepressants are expected to suppress nightmares because of their REM-suppressing activity. Some patients, however, actually report an increase in dream activity during treatment with antidepressants (Strayhorn and Nash 1978). As a further example, the nightmares that often constitute part of PTSD improve but do not resolve completely, even with potent REM-suppressing agents such as MAOIs (Kosten et al. 1991; Shestatzky et al. 1988). Cyproheptadine may reduce the intensity and/or frequency of nightmares associated with PTSD (Gupta et al. 1998).