اثربخشی درمان تمرین تصویرسازی برای بیماران مبتلا به اختلالات کابوس با و بدون همبودی افسردگی یا PTSD

Nightmares are a common and serious problem in psychotherapeutic practice, although they are seldom considered as independent mental disorders. There are some promising approaches to the treatment of nightmares, notably Imagery Rehearsal Therapy, a cognitive-restructuring treatment. The core of this approach is the modification of the nightmare script and repeated imagination of the new script. However, most evaluation surveys have been conducted only with trauma patients, and thus far there is no standardized manual in the German language. 69 participants were examined using self-rating questionnaires. Participants belonged to three groups: 22 primarily nightmare sufferers, 21 patients with major depression and nightmares, 26 with PTSD and nightmares. 12 of the PTSD patients were randomly assigned to a control condition. Primary outcome measures were nightmare frequency and anxiety during nightmares. Overall, nightmare frequency and the anxiety they caused decreased following the treatment. Nightmare frequency and anxiety during the nightmares were highest in the PTSD group initially. Nightmare frequency decreased in all groups. Anxiety scores decreased least in PTSD patients, in depressive patients and primarily nightmare sufferers anxiety scores decreased during intervention. In primarily nightmare sufferers anxiety remained low up to the catamnesis period as well. Thus, those who suffered primarily from nightmares showed the strongest benefit from the nightmare treatment.

Nightmares are a common psychological experience. They are typically defined as repeated awakening from sleep while recalling intensely disturbing dreams, usually involving fear or anxiety, but also other negative emotions such as anger or disgust (Schredl, 2009a). Awakening typically occurs in the second half of the sleep period during Rapid Eye Movement (REM) sleep. The person awakens and is quickly alert and oriented (DSM-IV-TR, American Psychiatric Association, 2000; ICSD-2, American Academy of Sleep Medicine, 2005; ICD-10, WHO, 2005). The criterion of awakening has been disputed. Nightmares that lead to awakening are assumed to be more intense than those that do not lead to awakening, but there is a large overlap (usually referred to as “bad dreams”; Schredl, 2009a; Spoormaker, 2008; Zadra & Donderi, 2000). Nightmares can occur either on their own or with a concurrent psychological disorder, such as depression, anxiety, schizophrenia, borderline personality disorder and, in particular, post-traumatic stress disorder (PTSD; Hartmann, 1984; WHO, 2005). Ten percent of adults report nightmares at least once a month (Belicki & Belicki, 1986; Levin, 1994). Prevalence rates of people who suffer from their nightmares are about 3–5% (Schredl, 2010; Spoormaker, Schredl, & van den Bout, 2006). Methodological aspects differed between the various epidemiological studies, and there is no consistent criterion for frequency or duration, but the aspect of suffering was relevant in most studies. Women report nightmares more frequently than men (Levin & Nielsen, 2007; Schredl & Reinhard, 2011), an effect not found in children, thus gender effects occur up from adolescence and narrows with increasing age (Schredl & Reinhard, 2011). Younger people have more nightmares than older people with the highest prevalence rates found between the age of five and ten (Schredl, 2009b). Hitherto, nightmares have not been a major focus of treatment in behavior therapy. For the most part, they have been regarded as symptoms of an underlying disorder, believed to vanish once this disorder has been treated. Nonetheless, there are several reports of successful treatment of nightmares (overview: Spoormaker, 2008), for instance with various cognitive-behavioral techniques such as exposure (Burgess, Gill, & Marks, 1998; Grandi, Fabbri, Panattoni, Gonnella, & Marks, 2006), systematic desensitization (Cellucci & Lawrence, 1978; Miller & DiPilato, 1983), or imagery rehearsal therapy (Krakow, Kellner, Pathak, & Lambert, 1995; Krakow & Zadra, 2006), as well as hypnosis (Kennedy, 2002; Seif, 1985). Imagery rehearsal therapy (IRT) is based on earlier approaches which used the rehearsal of the nightmare with a modified ending (Bishay, 1985) and is comparable to the transformation technique in hypnotherapy (Kennedy, 2002). IRT consists in the rehearsal of the modified dream using imagination techniques instructing patients to create a new dream script. Several studies have found a positive effect of IRT on nightmare frequency and nightmare distress (Forbes et al., 2001, 2003; Krakow, Hollifield, et al., 2001; Krakow, Johnston, et al., 2001; Krakow, Kellner, Neidhardt, Pathak, & Lambert, 1993; Krakow et al., 1995; Lu, Wagner, Van Male, Whitehead, & Boehnlein, 2009). However, studies describing a successful application of IRT in nightmare sufferers were mainly undertaken in a group-therapy setting. In earlier studies, participants received information on how to imagine a new dream script and practiced the imagination technique in only one session. Subsequently, the participants were instructed to practice the technique at home on their own (Krakow et al., 1993, 1995). Most of the more recent studies examined the nightmare treatment of PTSD patients (Forbes et al., 2003; Krakow, Hollifield, et al., 2001; Krakow, Johnston, et al., 2001). Our goal was to adapt this approach for an individual therapy setting and patients suffering from nightmares only, as well as patients suffering from other mental disorders such as depression and PTSD as well as nightmares. As there was no standardized therapy for nightmares in German, we developed a nightmare treatment for an individual therapy setting and standardized the instructions and exercises according to a manualized therapy based on IRT (Thünker & Pietrowsky, 2011). Moreover, we added specific adaptations for traumatized patients, namely an additional imagery exercise (“the safe place”) and a technique designed to minimize the nightmare – and therefore trauma – confrontation during nightmare reconstruction. The present study was designed to test the effectiveness of this standardized nightmare therapy (Thünker & Pietrowsky, 2011) in patients suffering from nightmares only (“primarily nightmare sufferers”), as well as patients suffering from nightmares associated with major depression and PTSD. We expected all patient groups to benefit from the standardized nightmare therapy with a reduced nightmare frequency and lower nightmare intensity (i.e. less anxiety during the nightmare). We also expected a reduction of the number of awakenings due to the nightmares and a lower level of daytime distress on the day after the nightmare. Since the number of nightmares and the distress on the day after a nightmare are only moderately intercorrelated, daytime distress following a nightmare is more likely to be associated with personality variables like psychopathology or personality traits (Blagrove, Farmer, & Williams, 2004; Levin & Fireman, 2002; Schredl, Landgraf, & Zeiler, 2003). Thus, daytime distress following a nightmare was assessed as an important dependent variable within the context of nightmare treatment. Therapy effects were expected to last up to the follow-up measurement ten weeks after the end of treatment. To control for unspecific treatment effects, effects in PTSD patients were compared with a randomized control group receiving treatment as usual which was expected to have inferior effects on the examined outcome measures.

For nightmare frequency no main effects of gender or age could be found. Nightmare frequency depended on age and time of measurement (time of measurement × age, F(2,51) = 3.29, p < .05, η2 = .11), thus no significant main effect of time of measurement could be found (F(2,51) = 1.46, n.s.). Mainly younger people showed a reduction of nightmare frequency during time (correlation between pre–post difference of frequency and age r = −.32, p < .05). There were no effects of gender for this parameter (all data are described in Table 2). Nightmare frequency differed significantly between the three groups of patients (F(2,54) = 7.98, p ≤ .001, η2 = .23, Fig. 2). PTSD patients had more nightmares per month (15.6 ± 2.5) than primarily nightmare sufferers (5.8 ± 1.9; t(34) = −3.15, p < .05) and depressive patients (6.9 ± 1.2; t(33) = −3.58, p ≤ .001). Nightmare frequency decreased from 11.3 (±1.2) per month at the onset of intervention to 7.4 (±1.2) after treatment, and 7.1 (±1.1) at follow-up. The main effect of time of measurement was affected by age of the participants (time × age F(2,51) = 3.29, p < .05, η2 = .11), showing a reduction of the frequency of nightmares over the intervention period mainly for younger participants. Nightmare frequency decreased during the intervention period (t(56) = 5.36, p < .001), an effect that persisted at follow-up (post – follow up: t(56) = 1.03, n.s.). There was no significant interaction between group and time of measurement (F(4,108) = .55, n.s.), indicating that nightmare reduction was independent of group membership. Table 2. Means and standard errors of mean of nightmare frequency, anxiety during nightmares, and depression in the three patient groups and PTSD control group. Primarily nightmare sufferers Depression PTSD intervention PTSD control-group Pre Post FU Pre Post FU Pre Post FU Pre Post Frequency (mean, SEM) 8.4 (2.1) 4.9 (1.8) 4.2 (1.8) 9.6 (1.4) 5.4 (1.1) 5.7 (1.2) 18.4 (2.2) 14.5 (2.8) 13.8 (2.5) 17.3 (3.0) 16.7 (3.2) Anxiety (mean, SEM) 5.8 (.2) 4.2 (.4) 3.1 (.4) 6.2 (.2) 3.8 (.3) 3.2 (.3) 6.4 (.2) 5.7 (.3) 5.7 (.2) 6.4 (.2) 5.9 (.5) Awakening (yes/no) 15/4 9/13 10/12 15/6 10/11 12/9 7/7 7/7 7/7 10/2 11/1 Daytime distress (yes/no) 19/1 11/11 8/14 21/0 16/5 16/5 14/0 14/0 14/0 12/0 11/1 BDI (mean, SEM) 8.3 (1.2) 5.3 (1.2) 5.8 (1.2) 24.5 (1.6) 15.7 (1.9) 14.4 (2.0) 32.7 (2.8) 28.0 (2.3) 28.4 (2.9) 29.1 (2.8) 26.8 (3.7) FU = follow-up, BDI = Beck's Depression Inventory. Table options Full-size image (21 K) Fig. 2. Means (and standard errors of mean) of nightmare frequency (nightmares per month) in the three groups of patients (primary nightmare sufferers, depressive patients, PTSD patients) across the treatment and a 10-week follow-up. Figure options For anxiety during the nightmare neither effects of age nor of gender could be found. Concerning anxiety during the nightmare, there were also significant differences between the groups of patients (primarily nightmare sufferers: 4.4 ± .3, depression: 4.4 ± .3, PTSD: 5.9 ± .2), with anxiety being highest in PTSD patients (main effect group: F(2,53) = 13.84, p < .001, η2 = .34). The anxiety score during nightmares decreased from 6.1 (±.1) at pre-measurement to 4.4 (±.2) at post, and 3.8 (±.2) at follow-up. This is reflected in a highly significant main effect of time of measurement (F(2,52) = 62.85, p < .001, η2 = .71). The reduction of anxiety took place mainly during the intervention period (pre – post: t(55) = 8.13, p < .001), but there was also a further significant reduction during the catamnesis period (post – follow up: t(55) = 2.99, p < .01). Anxiety was highest in the PTSD group at pre-measurement and decreased less than in the other groups across treatment (interaction group × time of measurement: F(4,106) = 5.92, p < .001, η2 = .18, Fig. 3). In detail, anxiety during the nightmare decreased markedly during the intervention in primarily nightmare sufferers (t(21) = 5.36, p < .001) and depressive patients (t(19) = 7.12, p < .001), while only a marginal decrease occurred in PTSD patients (t(13) = 2.11, p < .1). A further decline during the following ten weeks (follow-up measurement) was only observed in primarily nightmare sufferers (t(21) = 3.34, p < .01), with no changes in depressive patients (t(19) = 1.37, n.s.) or PTSD patients (t(13) = .00, n.s.). Full-size image (26 K) Fig. 3. Means (and standard errors of mean) of anxiety during the nightmare (1 = no anxiety to 7 = extreme anxiety) in the three groups of patients (primarily nightmare sufferers, depressive patients, PTSD patients) across the treatment and a 10-week follow-up. Figure options Regarding the number of awakenings due to nightmares, there were no significant differences between the three patient groups at any time of measurement (Kruskal–Wallis test; pre: χ2 = .47, df = 2, n.s., post: χ2 = .38, df = 2, n.s., follow-up: χ2 = .28, df = 2, n.s.). Before therapy, 68.5% of the patients were awoken always or most times by their nightmares; this figure dropped to 45.6% after the intervention, and remained lower at follow-up (50.9%). The reduction across all patient groups was significant (Friedman-test, χ2 = 8.49, df = 2, p < .05). Considering the groups separately using Friedman tests, the primarily nightmare sufferers woke up significantly less often from their nightmares during time (χ2 = 7.09, df = 2, p < .05), whereas only marginal effects could be found in depressive patients (χ2 = 4.92, df = 2, p < .1), and no effects were found in PTSD patients (χ2 = .29, df = 2, n.s.). At pre-measurement, 98.2% of the patients declared feeling affected by the nightmares on the following day (daytime distress), reporting feeling tired, having difficulties concentrating, being nervous and ruminating about the nightmare and other things. After the intervention period, only 71.9% of the patients felt affected by a nightmare, at follow-up only 66.7% (Cochran's Q = 27.90, df = 2, p < .001). Groups did not differ at pre-measurement with regard to level of distress on the following day (χ2 = 1.75, df = 2, n.s.), but did so both at post-measurement (χ2 = 10.70, df = 2, p < .01) and follow-up (χ2 = 16.65, df = 2, p < .001). These group differences were due to a decline of distress in primarily nightmare sufferers (Cochran's Q = 21.39, df = 2, p < .001) and depressive patients (Cochran's Q = 7.14, df = 2, p < .05), but not in PTSD patients (Cochran's Q = .00, df = 2, n.s.). See Table 2 for a detailed description. Depression ratings differed significantly between the three groups of patients (F(2,53) = 44.47, p < .001, η2 = .63). The BDI sum score was highest in PTSD patients (29.7 ± 2.7) and differed significantly from that the scores of both primarily nightmare sufferers (6.5 ± 1.2; t(33) = −8.89, p < .001) and depressive patients (18.2 ± 1.8; t(33) = −2.72, p ≤ .01). Depression ratings (BDI) decreased significantly across time of measurement (F(2,52) = 18.88, p < .001, η2 = .42), from 20.5 (±1.7) at pre-measurement to 14.9 (±1.6) after treatment (t(55) = 5.73, p < .001), and remained unchanged at follow-up (14.7 ± 1.6; t(55) = .40, n.s.). In addition, a significant interaction between group and time of measurement (F(4,106) = 3.12, p < .05, η2 = .11) indicates that depressive patients benefited most from the intervention regarding their degree of depression. Subsequent t-tests revealed that depression ratings declined in each group during treatment (primarily nightmare sufferers: t(20) = 3.19, p < .01; depressive patients: t(20) = 4.86, p < .001; PTSD patients: t(13) = 2.22, p ≤ .05), but did not change further during the follow-up period. The comparison of patients taking medication and those who did not revealed no differences in either nightmare frequency (comparison of pre–post difference using independent t-tests, t(45) = −1.50, n.s.) or anxiety during nightmares (t(45) = .258, n.s.). Concerning awakening, there was a tentative improvement in those patients who took no medication (Mann–Whitney test; χ2 = 5.65, df = 2, p < .01), but no change in the other group (χ2 = 2.39, df = 2, n.s.). Daytime distress improved in both groups (medication: χ2 = 6.00, df = 2, p ≤ .05; no medication: χ2 = 19.86, df = 2, p < .001).