Disrupting Reconsolidation of Trauma Memory
A Pilot Study
Marissa van der Sluis
First assessor : Prof. Dr. Merel Kindt Second assessor : Dr. Arnold van Emmerik
Year : 2013-2014
ABSTRACT
Aims
Numerous studies have revealed that disrupting reconsolidation of fear memory after reactivation proved to be
successful in erasing fear responding and preventing the return of fear. The aim of this pilot study was to test the effect of disrupting reconsolidation of trauma memory in patients with a posttraumatic stress disorder (PTSD).
Method
Data of three patients with PTSD were gathered and analyzed following a single case design. Treatment consisted of a maximum of three sessions. During these sessions, the most intrusive parts of trauma memory were reactivated. Directly after trauma reactivation, the noradrenergic beta-blocker propranolol HCI was given.
Results
Although reactivation of trauma memory succeeded in two of the three patients, only one patient showed a significant reduction in STAI scores between baseline and treatment phase. However, when power was enhanced by combining the p-value of the two experiments, a significant decrease in STAI scores was observed. Analysis revealed no significant
differences in negative or positive affect between baseline and treatment phase. Even though we observed a decrease in the overall scores on the PDS, BDI, BAI and PTCI between session 1 and one month after the last session, this effect was not significant.
Conclusion
The present findings demonstrated that disrupting reconsolidation of trauma memory in patients with PTSD reduced symptoms of anxiety, but did not improve affect. Moreover, PTSD symptoms only slightly improved one month after treatment. Despite its exploratory nature, this pilot study extends our knowledge of the effect of disrupting
3
INTRODUCTION
From an evolutionary perspective, fear memory is highly useful. It allows people to protect themselves against harmful events by responding to warning signals (LeDoux, 1995). But fear memory can also become maladaptive and
invalidating, as can be seen in patients suffering from anxiety disorders. Anxiety disorders such as specific phobia, panic disorders and posttraumatic stress disorders (PTSD) are the most prevalent forms of psychiatric disorders with a twelve-month prevalence up to 18 % (Kessler, Chiu, Demler & Walters, 2005). These large numbers underscore that anxiety disorders are a pressing public health issue.
To date, most interventions focus on the maintaining factors of fear reactions, such as cognitions and
behavioral responses. For instance Cognitive Behavioral Therapy (CBT), which is considered one of the most effective and well- researched treatments for anxiety disorders (GGZ-richtlijnen, 2011). Even though prior research indicates that CBT is effective in treating anxiety disorders (Stewart & Chambless, 2009), a substantial amount of patients
experiences a relapse following an apparently successful treatment (Craske, 1999).
Efficacious behavioral components of CBT for anxiety disorders involve exposure-based interventions such as in vivo exposure therapy or Imaginary Exposure (IE) therapy. In exposure-based treatments patients are confronted with a feared situation in order to diminish avoidance behavior and reduce subjective fear (Moscovitch, Antony & Swinson, 2009). However, the effect of fear extinction through exposure does not always persist over time. This might be explained by the fact that exposure does not involve ‘unlearning’ a previously learned association, but leaves the fear memory intact. It solely involves the formation of a novel inhibitory memory trace, that is available alongside the original fear memory (Bouton, 2002). This original fear memory can easily re-emerge, for instance by the presentation of an unpredictable aversive event (i.e., US) (‘reinstatement’), or by changing the context from treatment to test (‘renewal’), or by simply the passage of time (‘spontaneous recovery’) (Bouton, 2002).
For decades it was assumed that, once stored into the long-term memory, fear memory is permanent and indelible (McGaugh, 1966; Nader, 2003). However, several animal studies have shown that fear memory can be modified when retrieved (Lewis, 1969; Nader, Schafe & LeDoux, 2000; Eisenberg, Kobilo, Berman & Dudai, 2003; Duvarci & Nader, 2004). After reactivation, a consolidated memory returns into a labile state that requires protein synthesis. This process, which makes the memory prone to change, is referred to as ‘reconsolidation’ (Przybyslawski & Sara, 1997; Nader, et al., 2000). Nader et al. demonstrated that the process of reconsolidation can be disrupted after reactivation of fear memory. Injection of the protein synthesis inhibitor anisomycin after reactivation appeared to cause amnesia of the original learning in animals. The same treatment with anisomycin without memory reactivation left the original fear memory intact. These overall results showed that protein synthesis inhibitors can disrupt the
reconsolidation process, but only after reactivation (Nader et al.,2000). Because protein inhibitors such as anisomycin are very toxic, this method is not applicable in humans. Therefore, attempts have been made to investigate the effects of other nontoxic drugs that indirectly interfere with the required protein synthesis. Propranolol HCI, which can be administered orally, is a noradrenergic beta-blocker, which supposedly blocks the protein synthesis by acting on the beta-adrenergic receptors in the basolateral amygdala (McGaugh, 2004). Dębiec and LeDoux (2004) showed that systematic administration of propranolol HCI shortly after reactivation, disrupted reconsolidation of fear memory in rats.
More recently, the effects of targeting the process of reconsolidation was demonstrated in humans. Kindt, Soeter and Vervliet (2009) conducted an experiment, using a fear conditioning procedure. By contingently exposing pictures of spiders and electric shocks to healthy participants, a fear response for pictures of spiders was created. Results indicated that oral administration of propranolol HCI, either before or after memory reactivation substantially decreased the behavioral expression (startle reflexes) of fear as well as subjective feelings of fear 24 hours later (Kindt, et al., 2009; Soeter & Kindt 2010, 2011, 2012; Sevenster, Beckers & Kindt, 2013). Most importantly, the effect persisted at one-month follow-up and was not only restricted to the reactivated fear association, but even generalized to category-related cues. Notably, the intervention left the declarative memory for the acquired association untouched. Disrupting reconsolidation also diminished the subjectively experienced fear of an aversive event (i.e., electric stimulus) that was anticipated but was never actually experienced.
Even though disruption of reconsolidation may point to a promising strategy to erase fear memory, there are boundary conditions putting constraints on this approach. Sevenster, Beckers and Kindt (2013) found that propranolol HCI only reduced fear when a mismatch between the actual and expected situation (‘prediction error’, PE) took place during memory reactivation. PE was measured by the change in US-expectancy ratings from the end of acquisition to the beginning of extinction. The results indicate that the occurrence of a PE is a necessary condition for reconsolidation and provide a useful instrument for developing and optimizing reconsolidation-based treatments for patients suffering from anxiety disorders.
Since reconsolidation of fear memory can be disrupted, there is a possibility that this strategy could be particularly efficacious in the treatment of strong, intrusive trauma memories in disorders such as PTSD. Brunet et al. (2008) tested the effect of propranolol HCI administered after retrieval of trauma memory on physiologic responding during subsequent retrieval. Patients suffering from chronic PTSD were asked to describe their traumatic event in order to reactivate their memory. Immediately following this memory reactivation procedure, patients received a short- and long-acting dose of propranolol HCI or placebo. Physiologic responses during reactivation a week later were
significantly smaller in patients who had received propranolol HCI compared to those who had received placebo. In three open-label trials, six brief trauma reactivation sessions under the influence of propranolol HCI appeared to reduce
5
PTSD symptoms in patients with chronic PTSD (Brunet et al., 2011). Furthermore, treated patients reported less comorbid depressive symptoms, less negative emotions in daily life and better quality of life (Poundja, Sanche, Tremblay & Brunet, 2012).The main weakness of these studies is that it is unclear whether disruption of reconsolidation was the active therapeutic ingredient that was responsible for the observed symptom reduction. It cannot be ruled out that the
intervention itself might have induced extinction (Brunet et al., 2011; Poundja et al., 2012). In addition, the intervention only focused on the general description of the traumatic event, rather than on the ‘hotspots‘ of trauma memory.
Hotspots refer to memories of detailed moments of high levels of emotional distress during a traumatic event (Grey & Holmes, 2008), which are often re-experienced. Since the script-driven mental imaginary did not focus on the hotspots, it might not have triggered the reconsolidation process sufficiently. In addition, there were no indications of the occurrence of a PE, which is a necessary condition for reconsolidation. Another limitation is the administration of Propranolol HCI prior to reactivation. This might have hampered the later expression of fear, rather than disrupting the reconsolidation of trauma memory.
In this current pilot study we tested the effect of disrupting reconsolidation of trauma memory in patients with PTSD. As opposed to previous studies, we administered propranolol HCI after trauma memory reactivation, thereby targeting the process of reconsolidation rather than the retrieval of trauma memory. In order to trigger the
reconsolidation process, our intervention focused on the hotspots of trauma memory. We followed a procedure that is similar to the procedure used in imaginary exposure, but with the difference that it is only activated once as opposed to prolonged exposure in traditional CBT. The reactivation session ended when the peak level of fear was reached to prevent unwanted extinction effects. Since a mismatch between the expected situation and the actual situation (PE) is required for reconsolidation (Sevenster et al.,2013), PE was measured both before and after reactivation. Patients were asked to rate the expected intensity of a feared emotion, as strong emotions evoked by memory reactivation are often experienced as the aversive event (Foa, Steketee & Rothbaum, 1989). We hypothesized that disrupting reconsolidation of trauma memory would reduce symptoms of PTSD.
METHOD
Participants
Participants were recruited through advertisements in local newspapers (Metro and Het Parool) and by referrals of local general practitioners (Bureau Studentenartsen), psychologists (Bureau Studentenpsychologen) or Slachtofferhulp. Advertisements were posted on online fora and flyers were distributed among several colleges and universities (HvA, UvA, UU).
Inclusion criteria were a diagnosis of PTSD, the condition that the traumatic event happened at least three months ago, the presence of intrusions and a minimal age of 18. Blood pressure had to be at least equal to or higher than 60/90 and heart rate had to be equal to or higher than 60 and had to increase at the exercise test during the screening session.
Exclusion criteria were medical conditions including (former) cardiac problems, (former) cardiac problems in patients family, inability to moderately exercise, (former) asthma, (former) lung diseases, (former) low blood pressure, (former) fainting, (former) diabetes, liver problems, kidney problems, metabolic acidosis, excessive production of thyroid, dysfunctions of blood circulation, (former) epilepsy and/or seizures and use of medication that could (dangerously) interact with propranolol HCI (anti-psychotics, anti-anxiety drugs, antidepressants, heart medication, anti-inflammatory sedatives, medication for lowering blood pressure, medication for migraine, asthma, psoriasis, diabetes, tuberculosis and dizziness). Other medical conditions were (former) aversive responses to beta-blockers, current participation to any form of psychotherapy other than supportive, pregnancy or breastfeeding, severe psychiatric problems including dementia, brain damage, (former) psychotic tendencies, bipolar disorder, severe dissociative tendencies, depression with suicidal ideations, borderline personality disorder, severe physical or sexual abuse in childhood, substance dependency and the use of sedatives. Participants were invited for further screening when their score is higher than 15 on the Posttraumatic Diagnostic Scale (PDS). Based on these criteria, three patients were included. Descriptions of the patients were gathered during the screening session and can be summarized as follows:
P1 is a thirty year old female student who suffered from PTSD. During her past relationship she was abused by her ex-boyfriend, who is also the father of her nine year old child. She left her ex-boyfriend when her son was a couple of months old. P1 primarily suffered from two intrusions: 1. Her ex-boyfriend chases her through the house, bangs at the doors and finally hits her with a shower head. 2. Her ex-boyfriend hits her with a hot-water bottle after she passes out. When she wakes up she realizes that her son may be at risk. Treatment history consisted of psychological support in 2013 for helping her cope with her father’s alcohol abuse.
P2 is a forty-eight year old woman who works as a purser for a Dutch airline. At the time of the screening session she suffered from PTSD for two years. Thirty years ago she was raped by a stranger after a night out in
7
Amsterdam. After the rape, the man tried to choke her with his hands. P2 particularly suffered from the image of a large, white hand coming towards her. She also reported severe feelings of shame and guilt. The years after the raping she did not pay much attention to the emotional effects of the event. When a female friend of her was murdered two years ago, the PTSD-symptom worsened. Eighteen months ago she tried EMDR to treat her PTDS-symptoms. She reported that this treatment reduced her symptoms slightly, but that she did not feel comfortable during the treatment sessions. Based on information gathered during the screening session, P2 was diagnosed with a comorbid generalized anxiety disorder (GAD), alcohol abuse in the past and light alcohol dependence in partial remission.P3 is a thirty-seven year old woman who experienced a tsunami in 2004 during a holiday on a Malaysian island near Thailand. When the tsunami hit the island, P3 and her family were driving in a minivan. She and her family got away, but were shocked by seeing dead bodies lying in the water. She primarily suffered from the image of her and her ill daughter, being in the hospital trying to get help. P3 reported feelings of guilt for the preferential treatment she and her daughter got in that hospital. P3 suffered from PTSD for ten years, since the moment of the tsunami. The birth of her third daughter two years later seemed to have worsened her posttraumatic stress symptoms. Treatment history consisted of a 20 to 30 EMDR sessions in 2008 and a couple of CBT sessions in 2009. Although these treatments did decrease her symptoms, she still suffered from intrusions at the time of the screening session.
Materials
Posttraumatic Diagnostic Scale (PDS). Presence and severity of PTSD symptoms were measured with the
Posttraumatic Diagnostic Scale (PDS), a self-report measure that consists of 17 items on a 4-point scale. These 17 items are divided into the categories Re-experiencing (5 items), Arousal (7 items) and Avoidance (5 items). Higher PDS-scores reflect greater severity of PTSD. Cronbach’s alpha was .92 for Total Symptom Severity, .78 for Re-experiencing, .84 for Avoidance, and .84 for Arousal. The PDS showed high internal consistency (α = .92) and good test-retest reliability (κ = .74) in its original English version (Foa, Cashman, Jaycox & Perry, 1997).
State-Trait Anxiety Inventory (STAI). Since other measurements of anxiety (such as the PDS) are not adequate for daily measurement, the State-Trait Anxiety Inventory (STAI; Spielberger, Gorsuch, Lushene, Vagg & Jacobs, 1983) was used to assess daily states of anxiety. State anxiety is defined as situational fear, discomfort and nervousness. The original STAI consist of 40 questions based on a 4-point Likert scale and measures both State and Trait anxiety. The two forms of anxiety are measured separately by two scales consisting of 20 items. Examples of State anxiety questions are ‘I am tense’; ‘I am worried’ and ‘If feel calm’; ‘I feel secure’. Scores on this scale range from 20 to 80, with higher scores indicating greater anxiety. The STAI showed good to excellent internal consistency, with alpha ranging from .86 to .95. Test-retest reliability coefficients for the Trait anxiety scale were satisfactory, ranging
from .73 to .86, but relatively low (.16 to .62) for scores on the State Anxiety scale (Spielberger et al., 1983). However, this lack of stability is considered desirable, since emotional states should reflect changes in situational factors at the time of testing.
Positive and Negative Affect Schedule (PANAS). Daily measures of affective mood states were collected with the Positive and Negative Affect Schedule (PANAS; Watson, Clark & Tellegen, 1988). This 20-item self-report measure consists of two mood scales, one assessing negative affect (NA) and the other positive affect (PA). Items are rated on a 5-point scale ranging from 1 to 5. This score represents the extent to which a patient has felt this way today. Internal consistency was good to excellent, with alpha ranging from .86 to .90 for the Positive Affect scale and .84 to .87 for the Negative Affect scale. Test-retest correlations ranged from .47 to .68 for Positive Affect and .39 to .71 for Negative Affect.
Posttraumatic Cognitions Inventory (PTCI). Trauma-related thoughts and beliefs were measured with the Posttraumatic Cognitions Inventory (PTCI; Foa, Tolin, Ehlers, Clark & Orsillo, 1999), consisting of 36 items that can be answered on a 7-point scale. The PTCI is divided into three subscales labeled Negative Cognitions About Self (21 items), Negative Cognitions About the World (7 items) and Self-Blame (5 items). The PTCI demonstrated high internal consistency in a treatment-seeking sample (α = .94) and in a college sample (α = .93). High test–retest reliability is indicated by significant positive correlations between test and retest scores on the total scale (r = .79, p < .01, two-tailed) (Van Emmerik, Schoorl, Emmelkamp & Kamphuis, 2006).
Beck’s Depression Inventory–II (BDI-II). Cognitive and physical symptoms of depression were assessed with Beck’s Depression Inventory–II (BDI-II; Beck, Steer & Brown, 1996), a 21-item self-report scale. Each item offers four descriptions of different intensity (indexed from 0 to 3). Accordingly, the BDI score can range between 0 and 63. Ratings are summed, with higher scores indicating greater severity of symptoms. The cut-off scores for minimal depression are 0-13, mild depression 14-19, moderate depression 20-28 and severe depression 29-63. In a college sample, internal consistency of reliability was found to be good (α = .90). Scores of the BDI-II correlated positively and significantly with measures of anxious and depressive symptomatology, supporting the convergent validity of the scale (Storch, Roberti & Roth, 2004).
Becks’ Anxiety Inventory (BAI). Physiological and cognitive symptoms of anxiety were determined with the Becks’ Anxiety Inventory (BAI; Beck, Epstein, Brown & Steer, 1988). Each of the 21 items of the BAI is descriptive of a symptom of anxiety and is rated on a 4-point scale. Higher scores on the BAI represent greater severity of anxiety symptoms. Among patients diagnosed as having anxiety disorders, the BAI had a high level of internal consistency (α = .94). Test-retest correlation with an interval of 7 weeks was .62, which was viewed as reasonable (Creamer, Foran & Bell, 1995).
9
Subjective Units of Distress Scale (SUDS). Intensity of currently experienced disturbance or distress was measured with the Subjective Units of Distress Scale (SUDS; Wolpe, 1969). The patient was asked to tell the therapist how distressed he/she felt on a scale of 0 (piece, total relief) to 100 (severely upset, unbearably bad). The purpose of the SUDS was the evaluation of the treatment process.Prediction Error (PE). The occurrence of a prediction error (PE) was assessed by asking patients to rate the expected intensity of the feared emotion before and after trauma reactivation on a scale of 0 (expected emotion is not feared at all) to 100 (expected emotion is extremely feared). Before trauma reactivation patients answered the question: ‘Which feeling do you fear the most during the treatment session when we try to focus your attention on the traumatic memory and you don’t avoid it as you normally do?’ and ‘How strongly do you fear that feeling?’. Directly after trauma reactivation patients answered the question ‘When we try to focus your attention on the traumatic memory again (during the second treatment session next week) and you don’t avoid it as you normally do, how strongly do you fear the feeling that you feared before treatment?’.
Procedure
Internet screening. In order to determine whether patients were invited to the screening session, they were checked for in- and exclusion criteria online. Patients completed the PDS and the McLean Screening Instrument for Borderline Personality Disorder (MSI-BPD).
Screening session. During the screening session, patients were informed about various study procedures and
signed informed consent. Furthermore, patients were asked to give permission to inform their general practitioner. To examine the presence of contra-indications, a medical screening was carried out. Blood pressure and heart rate were measured using an electronic sphygmomanometer (OMRON M6 Comfort) and patients were asked to perform an exercise test, which involved running up and down the stairs for one minute. In order to be certain that patients met the criteria of PTSD and suffered from intrusive memories about the traumatic event, the Clinician-Administered PTSD Scale (CAPS) was administered. During the CAPS interview, patients were asked to decide on the most intrusive memory (hotspot). This memory was targeted during the treatment session(s). Finally, the Structured Clinical Interview for DSM-IV Axis I Disorders (SCID-I) was administered. The borderline section of the Structured Clinical Interview for DSM-IV Axis II Disorders (SCID-II) was administered only when patients scored 7 or higher on the McLean Screening Instrument for Borderline Personality Disorder (MSI-BPD).
In order to obtain a baseline measure of affective mood states, patients who were invited to participate in the study were asked to complete the STAI and PANAS on a daily basis until the first treatment session. Furthermore,
patients were asked not to use alcohol or medication and not heavily exercise the night before the first session and not to eat, drink coffee or tea or use chewing gum at least an hour before the treatment session.
Treatment session 1. During the first treatment session, patients completed the PDS, the Posttraumatic Cognitions Inventory (PTCI), Becks’ Anxiety Inventory (BAI) and Beck’s Depression Inventory–II (BDI-II). After that, a medical screening was conducted by measuring blood pressure and heart rate and obtaining saliva samples. The treatment started with the reactivation of the hotspot of trauma memory by an imaginary exposure procedure, mentioned in the treatment protocol for PTSD of Minnen and Arntz (2007). Imaginary exposure continued until the observed fear was maximal. Directly after the reactivation session, patients answered the following questions: ‘If you think about the way you felt the moment you focused your attention on the traumatic memory, which feeling was the worst?’ and on scale of a 0 (not at all) to 100 (extremely strong) ‘How strong was that feeling?’ Reactivation was considered successful when scores were 50 or higher. After a successful reactivation, patients weighing less than 70 kg received 40 mg of propranolol HCI, patients between 70-90 kg received 60 mg and patients weighing more than 90 kg received 80 mg. During the ninety-minute break that followed, patients were asked not to eat or drink anything, except water. After the break, blood pressure, heart rate, and saliva samples were obtained again in order to determine the effects of propranolol HCI on the sympatic nervous system. Lower heart rate and lower levels of sAA and blood pressure (especially systolic blood pressure) were expected (Van Stegeren, Rohleder, Everaerd & Wolf, 2006). In order to observe changes in affective mood states, patients were asked to complete the STAI and PANAS on a daily basis until the end of the treatment phase. If patients reported that the intensity of the experienced emotion was less than 40, reactivation of trauma memory did not succeed. In this case, patients did not receive propranolol HCI. Instead, an appointment was made for the next week. During this session, a second attempt was made to reactivate fear maximally.
Treatment session 2 & 3. During the second session a week later, patients who did receive propranolol HCI during the first treatment session were asked to complete the PDS, the PTCI and the BAI. When the patient showed a substantial decrease of fear after the first treatment session, an appointment was made for a follow-up session one month later. Patients who did not receive propranolol HCI during the first treatment session took part in the medical screening again. Hotspots of trauma memory were reactivated using the same imaginary exposure procedure as the first treatment session. Again, propranolol HCI was administered only when fear was reactivated maximally. This treatment procedure continued for a maximum of two extra treatment sessions. If reactivation of fear did not succeed during the last treatment session, he or she was informed about other treatment options. After the last treatment session, patients were asked to complete the STAI and PANAS until the end of the treatment phase.
Follow-up. During the follow-up session (one month after the last treatment session), the PDS, PTCI, BAI and
BDI were completed. When a patient did not show a substantial decrease of fear, he or she was informed about other treatment options.
11
Analysis
In order to analyze the effect of disrupting reconsolidation of trauma memory on affective mood states (STAI and PANAS scores), a single-case design was carried out for each PTSD patient. During the baseline phase and the intervention phase STAI and PANAS scores were observed on a daily basis. Length of the baseline (A) and treatment phase (B) was determined randomly between 14 and 33 days. The total duration of the measurement was 47
consecutive days. To attain a minimal p-value of .05, a number of 20 (1/20 = 0.5) permutations was needed. A p-value of .05 or smaller was considered as significant. Daily STAI and PANAS scores in both phases were visually analyzed following single case visual analysis (SCVA; Bulté & Onghena, 2012). To evaluate the effect of the intervention, a randomization test was performed following a phase (AB) design (Bulté & Onghena, 2008) using Single Case Randomization Tests (SCRT) software (Onghena & Van Damme, 1994). Subsequently, p-values of all patients were aggregated in a quantitative and systematic way, following a sequential replication strategy. The combined p-value was determined with the additive method (Edgington, 1972) in SCRT. Effect sizes of all single case experiments were estimated in order to measure the amount of change in affective mood states between baseline and treatment phase. Wilcoxon tests were used for the total group of patients, exploring differences in scores between pretreatment and one month follow-up on the secondary outcome measures (PDS, PTCI, BDI-II and BAI).
RESULTS
Visual Analysis
Participant 1. The patterns of daily STAI ratings and PANAS ratings of Negative Affect and Positive Affect of P1 during 47 consecutive days are shown in respectively Figure 1, 2 and 3. The first two plots (a) show the trend and course with missing values and the two latter plots (b) show the trend and course where missing values were replaced by either the mean of phase A or the mean of phase B. The first reactivation session of P1 took place at day 19. The second reactivation session took place at day 28. In both sessions, propranolol HCI was administered. A total of 9 values were missing.
13
Figure 1.a.
b.
Figure 1. The trend (left plots) and course (right plots) of the scores on the STAI-S of P1 in the Baseline Phase (A) and Treatment
Phase (B).The first two plots represent the scores with missing values (a) and the second two plots represent the scores with missing
Figure 2.
a
.
b.
Figure 2. The trend (left plots) and course (right plots) of the scores on the Negative Affect scale of the PANAS of P1 in the Baseline
Phase (A) and Treatment Phase (B).The first two plots represent the scores with missing values (a) and the second two plots
15
Figure 3.a.
b.
Figure 3. The trend (left plots) and course (right plots) of the scores on the Positive Affect scale of the PANAS of P1 in the Baseline
Phase (A) and Treatment Phase (B).The first two plots represent the scores with missing values (a) and the second two plots
represent the scores with missing values replaced by the mean of phase A or B (b).
Participant 2. The patterns of daily STAI ratings and PANAS ratings of Negative Affect and Positive Affect of P2 during 47 consecutive days are shown in respectively Figure 4, 5 and 6. The first reactivation session of P2 took place at day 28. However, the reactivation of trauma memory did not succeed. During a second session on day 35, trauma memory could not be reactivated once again. Since reactivation of trauma memory failed during both sessions, P2 never received propranolol HCI. Data of day 20 were missing.
Figure 4.
a
.
b
.
Figure 4. The trend (left plots) and course (right plots) of the scores on the STAI-S of P2 in the Baseline Phase (A) and Treatment
Phase (B).The first two plots represent the scores with missing values (a) and the second two plots represent the scores with missing
17
Figure 5.a.
b.
Figure 5. The trend (left plots) and course (right plots) of the scores on the Negative Affect scale of the PANAS of P2 in the Baseline
Phase (A) and Treatment Phase (B).The first two plots represent the scores with missing values (a) and the second two plots
Figure 6.
a.
b.
Figure 6. The trend (left plots) and course (right plots) of the scores on the Positive Affect scale of the PANAS of P2 in the Baseline
Phase (A) and Treatment Phase (B).The first two plots represent the scores with missing values (a) and the second two plots
represent the scores with missing values replaced by the mean of phase A or B (b).
Participant 3. The patterns of daily STAI ratings and PANAS ratings of Negative Affect and Positive Affect of P3 during 47 consecutive days are shown in respectively Figure 7, 8 and 9. The first reactivation session of P3 took place at day 19. The second session took place at day 28. During the first session reactivation of trauma memory succeeded. P3 therefore received propranolol HCI. No reactivation took place during the second session. Hence, no propranolol HCI was administered during session 2.
19
Figure 7.a.
b
.
Figure 7. The trend (left plots) and course (right plots) of the scores on the STAI-S of P3 in the Baseline Phase (A) and Treatment
Phase (B).The first two plots represent the scores with missing values (a) and the second two plots represent the scores with missing
Figure 8.
a.
b.
Figure 8. The trend (left plots) and course (right plots) of the scores on the Negative Affect scale of the PANAS of P3 in the Baseline
Phase (A) and Treatment Phase (B).The first two plots represent the scores with missing values (a) and the second two plots
21
Figure 9.a.
b.
Figure 9. The trend (left plots) and course (right plots) of the scores on the Positive Affect scale of the PANAS of P3 in the Baseline
Phase (A) and Treatment Phase (B).The first two plots represent the scores with missing values (a) and the second two plots
represent the scores with missing values replaced by the mean of phase A or B (b).
Randomization Tests
For both STAI and PANAS scores p-values were calculated. In order to obtain these p-values, missing values had to be replaced by the mean of phase A (baseline) and phase B (treatment).
State-Trait Anxiety Inventory (STAI). Analysis of the effect of treatment on STAI scores revealed an expected decrease in STAI scores of P1 in phase B, compared to phase A. This effect was significant (observed test statistic = 6.27, p = .05) with a medium effect size of -0.62 (Faraone, 2008). As expected, we observed no significant
effect of treatment on STAI scores in P2 (observed test statistic = 3.69, p = .30) and, unexpectedly, no significant differences between phase A and B in P3 (observed test statistic = 5, p = .20). Effect sizes were small (-0.34) for P2 and medium (-0.59) for P3 (Faraone, 2008). When the p-values of p1 en p3 (in which actual memory reactivation took place) were combined, the effect of treatment on STAI scores was significant (p = .031).
Positive and Negative Affect Schedule (PANAS). The analysis of the Negative Affect (NA) scale of the PANAS showed no differences in scores between phase A and phase B. Although the overall scores seem to have decreased, this effect was not significant for P1 (observed test statistic = 8.35, p = .10), P2 (observed test statistic = 3.21 , p = .10) and P3 (observed test statistic = 0.89, p = .35). However, the effect size was large (-1.21) for P1. Effect sizes were medium (-0.51) for P2 and small (-0.16) for P3 (Faraone, 2008). No significant effect (p = 0.10) was found on the NA scores when separate p-values of P1 and P3 were combined.
Analysis of the Positive Affect (PA) scale of the PANAS showed no effect of treatment in P1, nor in P2 and P3 (observed test statistic = 1.74, p = .45; observed test statistic = -0.74, p = .65; observed test statistic = -3.17, p = .25, respectively. Effect sizes in P1, P2 and P3 were small, -0.30, 0.12 and 0.45 respectively (Faraone, 2008). When separate p-values of P1 and P3 were combined, the effect of treatment on PA scores was not significant (p = .25).
Secondary outcome measures (PDS, BDI and BAI and PTCI). The overall scores on the PDS, BDI, BAI and PTCI between session 1 and one month after the last session (Table 1.) seemed to have decreased. However, Wilcoxon tests showed that this effect was not significant (z = -1.826, p = .068; z = -1.604, p = .109; z = -1.826, p = .068 and z = -1.604, p = .109, respectively).
Table 1. Differences in scores on the PDS, BDI, BAI and PTCI
Measure P1 P2 P5 P6 P13 (P1) P14 (P3) p-value PDS S1 PDS FU1 BDI S1 28 8 29 43 25 54 22 14 32 14 3 8 40 - 35 24 - 31 .068, ns BDI FU1 14 - 7 0 - - .109, ns BAI S1 51 64 40 40 64 33 BAI FU1 28 54 26 21 - - .068, ns PTCI S1 90 194 150 103 198 150 PTCI FU1 42 - 98 45 - - .109, ns
Note. -, missing value; PDS, Posttraumatic Diagnostic Scale; BDI, Becks’ Depression Inventory-II; BAI, Becks’ Anxiety Inventory; PTCI, Posttraumatic Cognitions Inventory; P, participant; S1, session 1; FU1, one month follow-up. P1, P2, P5 and P6 represent participants who were not included in this pilot study, but completed the treatment. P13 and P14 represent P1 and P3 in this current study. P12 (P2 in the current pilot study) did not complete the treatment and was therefore not included in the analysis.
23
Prediction Error. Trauma memory of P1 was reactivated successfully during session 1. Before trauma reactivation, she reported that she primarily feared feelings of panic and grief. To the question how strongly she feared those feelings on a scale from 0 to 100, she answered 80. Directly after trauma reactivation she reported that the strongest emotion she felt during the session was fear (of being in danger). In answer to how strong the emotion was on a scale from 0 to 100, she reported 100. P1 reported that she feared feelings of panic and grief even more (100 on a scale from 0 to 100) than at the beginning of the session (80 on a scale from 0 to 100). Since trauma reactivation appeared successful, propranolol HCI was administered.At session 2, posttraumatic symptoms were decreased, but some images remained intrusive. Therefore, the second session focused on the remaining hotspots. Before trauma reactivation, P1 reported that she particularly feared feelings of panic, grief and shame (100 on a scale from 0 to 100). Directly after trauma reactivation she reported that the strongest emotions she felt during the session were shame and fear. In answer to how strong these emotions were on a scale from 0 to 100, she reported 100. In answer to the question how strongly she would fear the feelings that she dreaded before treatment, she answered 70 (on a scale from 0 to 100). Again, propranolol HCI was administered.
Trauma memory of P2 was not successfully reactivated during session 1, nor during session 2. During the first session before reactivation, she reported that she primarily feared the feeling of total powerlessness (80-90 on a scale from 0 to 100). Although the therapist tried to reactivate trauma memory, P2 only experienced mild feelings of guilt and shame. She told the therapist that she normally avoids feelings related to her trauma. Since trauma memory was not reactivated during this first session, P2 did not receive propranolol HCI. During session 2, the therapist spoke with her about future options to reactivate trauma memory. No reactivation took place and no propranolol HCI was administered during that session.
Trauma memory of P3 seemed to have been reactivated fully during the first session. Before trauma reactivation she reported she particularly feared the feeling of weakness and loss of control. To the question how strongly she feared those feelings on a scale from 0 to 100, she answered 80. To the question how strongly she would fear the feelings that she did before treatment, she answered 75 (on a scale from 0 to 100). Propranolol HCI was administered after reactivation. At session 2, P3 reported that she did not feel different and that she felt as if the
reactivation of trauma memory during session 1 only partially succeeded. She told her therapist that she could only fully reactivate her trauma memory by watching a movie about the tsunami. During session 2, this was discussed and a new appointment was made for session 3. Results of this session could not be included in this pilot study. No propranolol HCI was administered during the session 2.
DISCUSSION
In the present pilot study, we investigated the effect of disrupting reconsolidation of trauma memory in patients with PTSD following a single case design. As opposed to previous studies (Brunet et al., 2008, 2011) we tested this effect under clear boundary conditions. Our hypothesis that disrupting reconsolidation would reduce symptoms of PTSD was partially confirmed. Only one of the two patients in which trauma memory was successfully reactivated showed a significant reduction in symptoms of anxiety between baseline and treatment phase. However, the most interesting finding was that when power was enhanced by the combination of the two experiments, a significant decrease in symptoms of anxiety was observed. This seems to be consistent with other studies (Brunet et al., 2008, 2011) and suggests that disrupting reconsolidation of trauma memory in patients with PTSD reduces symptoms of anxiety. Contrary to our expectations and previous research (Poundja, Sanche, Tremblay & Brunet, 2012), this pilot study did not find a significant difference in negative or positive affect between baseline and treatment phase. Moreover, PTSD symptoms only slightly improved one month after treatment.
The fact that two of the three patients did not show a significant improvement in state anxiety between baseline and treatment phase may be explained by the following. Firstly, P2 did not receive propranolol HCI and was therefore not actively treated in this pilot study. Since disrupting reconsolidation is considered to be the active therapeutic ingredient that is responsible for symptom reduction, no improvement was expected in P2. Secondly, the lack of effect between the baseline and treatment phase in P3 might be explained by the fact that the single reactivation of trauma memory during session 1 was not sufficient to fully reactivate the large and complex associative fear networks that underlie trauma memory (Foa & Kozak, 1986). Possibly, solely one particular hotspot of trauma memory was disrupted from reconsolidation, leaving otherhotspots intact.
The lack of differences in positive or negative effect between the baseline and treatment phase might be
explained by the length of the treatment phase wherein affect was measured. The intervention is likely to be effective in immediately reducing symptoms of anxiety by eliminating intrusive memories. However, changes in affect might set in later. Possibly, both positive and negative affect will improve after the patient’s realization that they no longer, or to a lesser extent, suffer from their PTSD symptoms. Therefore, the (treatment) phase in which affect after the intervention was observed, might have been too short to observe a significant improvement. However, the therapeutic effects of disrupting reconsolidation of trauma memory may be limited to symptoms of PTSD, leaving other affective states unchanged. Possibly, additional treatment such as CBT is needed to achieve more improvement in affect.
Results of this pilot study indicate that disrupting reconsolidation might be an effective strategy in significantly reducing symptoms of anxiety. However, the current procedure is far from optimal. Also, alternative explanations, such as placebo and extinction effects, could be responsible for the observed fear-reducing effects. Decreases in anxiety
25
symptoms may be due to the therapeutic effects of completing daily questionnaires and the personal attention during the treatment sessions. Yet, the patient who went through the same process as the other patients, but who’s trauma memory was not reactivated, showed no improvement in anxiety symptoms. Furthermore, it is unlikely that the amount of change in anxiety symptoms are the result of extinction effects. Extinction effects should have been observable at the day of the session, while the effects of disrupting reconsolidation start the day after the session, as this process is supposed to occur during sleep (Stickgold & Walker, 2005, 2007; Soeter & Kindt, in press). As in our pilot study effects showed up the day after the session, it can be assumed that disruption of reconsolidation was the active therapeutic ingredient that was responsible for the observed symptom reduction.It is interesting to note that in both successful reactivation sessions a prediction error (PE) occurred. This PE might have triggered the reconsolidation process of the trauma memory, which makes it prone to change.
Reconsolidation may be viewed as an updating process that maintains a memory’s relevance in guiding future
behavior (Dudai, 2009; Lee; 2009). As a result, reconsolidation does not necessary occur when fear memory is being reactivated, but only when something new canbe learned (PE) from the memory reactivation session (Sevenster, Beckers & Kindt, 2013). After a successful reactivation of trauma memory, P3 reported that she would fear feelings of weakness and loss of control less than she feared those feelings before treatment. This statement might indicate that a mismatch between the actual and expected situation (PE) took place during memory reactivation. After the first successful reactivation of trauma memory in P3, she, somewhat surprisingly, reported that she
would fear feelings of panic and grief and even more than before reactivation.However, the nature of the discrepancy between the expected and the actual situation does not make a difference inthe way it triggers the reconsolidation process. Both negative PE’s (actual situation better than expected) and positive PE’s (actual situation worse than expected) can cause destabilization of the original memory trace (Sevenster et al., 2013). In sum, attempts to create a PE in patients, which is a necessary condition for reconsolidation, appeared successful.
Finally, a number of important limitations to this pilot study need to be considered. Firstly, evoking strong emotional responses during trauma reactivation has been problematic with one of the three patients. This difficulty might be related to the patients need to avoid strong emotions. This (often automatic) tendency could be an aspect of PTSD or could be the expression of a personality trait (i.e. the need for control). Despite of the origin of this tendency, this problem should be addressed in future studies through optimizing the reactivation procedure. A second limitation is the power of the study, as only three patients were analyzed. The few amount of data points might have reduced power. In addition, the replacement of missing data by means made it less probable to observe a significant improvement. Also, the lack of data made it difficult to analyze the effect on the secondary outcome measures after one month. Future research should focus on enhancing power, for instance by including more patients or by using a multiple baseline
design (Bulté & Onghena, 2009). In order to investigate whether disruption of the reconsolidation process is the active ingredient responsible for improvement in PTSD symptoms, a double-blind, randomized, placebo-controlled trial (RCT) must be conducted.
Despite its exploratory nature, this pilot study extends our knowledge of the effect of disrupting
reconsolidation of trauma memory in patients with PTSD. If it turns out that this novel intervention is effective in reducing PTSD symptoms, disrupting reconsolidation of trauma memory could be a promising low-cost strategy for the treatment of patients suffering from PTSD and other anxiety disorders.
27
REFERENCESBeck, A. T., Epstein, N., Brown, G., & Steer, R. A. (1988). An inventory for measuring clinical anxiety:
Psychometric properties. Journal of Consulting and Clinical Psychology, 56 (6), 893-897. Beck, A. T., Steer R. A., & Brown, G. K. (1996). Manual for the Beck Depression Inventory–II. San Antonio, TX:
Psychological Corporation.
Bouton, M. E. (2002). Context, ambiguity, and unlearning: sources of relapse after behavioral extiction. Biological Psychiatry, 52, 976-986.
Brunet, A., Orr, S. P., Tremblay, J., Robertson, K., Nader, K., & Pitman, R. K. (2008). Effect of post-retrieval propranolol on psychophysiologic responding during subsequent script-driven traumatic imagery in post-traumatic stress disorder. Journal of Psychiatric Research, 42 (6), 503-506.
Brunet, A., Poundja, J., Tremblay, J., Bui, E., Thomas, E., Orr, S. P., Pitman, R. K. (2011). Trauma reactivation under the influence of propranolol decreases posttraumatic stress symptoms and disorder: 3 open-label trials. Journal of Clinical Psychopharmacology, 31 (4), 547-550.
Bulté, I., & Onghena, P. (2008). An R package for single-case randomization tests. Behavior Research Methods, 40 (2), 467-478.
Bulté, I., & Onghena, P. (2009). Randomization tests for multiple-baseline designs: An extension of the SCRT-R package. Behavior Research Methods, 41 (2), 477-485.
Bulté, I., & Onghena, P. (2012). When the truth hits you between the eyes. A software tool for the visual analysis of single-case experimental data. Methodology, 8 (3), 104–114.
Craske, M. G. (1999). Anxiety disorders: psychological approaches to theory and treatment. Boulder: Westview
Press.
Creamer, M., Foran. J., & Bell, R. (1995). The Beck Anxiety Inventory in a non-clinical sample. Behaviour Research and Therapy, 33, 477–485.
Dębiec, J., & LeDoux, J. E. (2004). Disruption of reconsolidation but not consolidation of auditory fear conditioning by noradrenergic blockade in the amygdala. Neuroscience, 129, 267-272.
Dudai, Y. (2009). Predicting not to predict too much: how the cellular machinery of memory anticipates the uncertain future. Philosophical Transactions of the Royal Society B: Biological Sciences, 364, 1255-1262.
Duvarci, S., & Nader, K. (2004). Characterization of fear memory reconsolidation. Journal of Neuroscience, 24, 9269-9275.
Edgington, E. E. (1972). An additive method for combining probability values from independent experiments. Journal of Psychology, 80, 351 – 364.
Eisenberg, M., Kobilo, T., Berman, D. E., & Dudai, Y. (2003). Stability of retrieved memory: Inverse correlation with trace dominance. Science, 301, 1102-1104.
Faraone, S. V. (2008). Interpreting estimates of treatment effects: implications for managed care. Pharmacy and Therapeutics, 33 (12), 700.
Foa, E. B., Cashman, L., Jaycox, L., & Perry, K. (1997). The validation of a self-report measure of posttraumatic stress disorder: The Posttraumatic Diagnostic Scale. Psychological Assessment, 9 (4), 445-451.
Foa, E.B., & Kozak, M. J. (1986). Emotional processing of fear: Exposure to corrective information. Psychological Bulletin, 99, 20-35.
Foa, E. B., Tolin, D. F., Ehlers, A., Clark, D. M., & Orsillo, S. M. (1999). The Posttraumatic Cognitions Inventory (PTCI): Development and validation. Psychological Assessment, 11, 303-314.
Foa, E. B., Steketee, G., & Rothbaum, B. (1989). Behavioral/cognitive conceptualizations of posttraumatic stress disorder. Behavior Therapy, 20, 155–176.
Grey, N., & Holmes, E. A. (2008). "Hotspots" in trauma memories in the treatment of posttraumatic stress disorder: A replication. Memory, 16 (7), 788-796.
Kessler, R. C., Chiu, W. T., Demler, O., & Walters, E. E. (2005). Prevalence, severity, and comorbidity of 12-month DSM-IV disorders in the national comorbidity survey replication. Arch Gen Psychiatry. 62, 617-627. Kindt, M., Soeter, M., & Vervliet, B. (2009). Beyond extinction: Erasing human fear responses and preventing the
return of fear. Nature Neuroscience, 12, 256-258.
LeDoux, J.E. (1995). Emotion: clues from the brain. Annual Review of Psychology, 46, 209-235.
Lee, J. L. (2009). Reconsolidation: Maintaining memory relevance. Trends in Neurosciences, 32, 413–420. Lewis, D. J. (1969). Sources of experimental amnesia. Pychological Review, 76 (5), 461-472.
McGaugh, J. L. (2004). The amygdala modulates the consolidation of memories of emotionally arousing experiences. Annual Review of Neuroscience, 27, 1-28.
Moscovitch, D. A., Antony, M. M., & Swinson, R. P. (2009). Exposure-based treatments for anxiety disorders: Theory and process. In M. M. Antony, M. Martin & M. B. Stein (Eds.), Oxford handbook of anxiety and related disorders (pp. 461-475). New York, NY: Oxford University Press.
Multidisciplinaire Richtlijnen Angststoornissen (2nd ed.) In GGZ-richtlijnen (2011). Retrieved on may 7th, 2013, from http://ggzrichtlijnen.nl
29
Nader, K., Schafe, G. E., & LeDoux, J. E. (2000). Fear memories require protein synthesis in the amygdala forreconsolidation after retrieval. Nature, 406, 722-726.
Onghena, P., & Van Damme, G. (1994). SRCT 1.1: Single-case randomization tests. Behavior Research Methods, Instruments, & Computers, 26, p. 369.
Poundja, J., Sanche, S., Tremblay, J., & Brunet, A. (2012). Trauma reactivation under the influence of propranolol: an examination of clinical predictors. European Journal of Psychotraumatology, 3, doi: 10.3402/ejpt.v3i0.15470 Przybyslawski, J., & Sara, S. J. (1997). Reconsolidation of memory afte its reactivation. Behavioral Brain Research,
84, 241-246.
Sevenster, D., Beckers, T., & Kindt, M. (2013). Prediction error governs pharmacologically induced amnesie for learned fear. Science, 339, 830-833.
Soeter, M. & Kindt, M. (2010). Dissociating response systems: Erasing fear from memory. Neurobiology of Learning and Memory, 94, 30-41.
Soeter, M., & Kindt, M. (2011). Disrupting reconsolidation: Pharmalogical and behavioral manipulations. Learning & Memory, 18, 357-366.
Soeter, M., & Kindt, M. (2012). Erasing fear for an imagined threat event. Psychoneuroendocrinology, 37 (11), 1769- 177.
Spielberger, C. D., Gorsuch, R. L., Lushene, R., Vagg, P. R., & Jacobs, G. A. (1983). Manual for the State-Trait Anxiety Inventory. Palo Alto, CA: Consulting Psychologists Press.
Stewart, R. E., & Chambless, D. L. (2009). Cognitive-behavioral therapy for adult anxiety disorders in clinical practice: a meta-analysis of effectiveness studies. Journal of Consulting and Clinical Psychology, 77 (4), 595-606.
Stickgold, R., & Walker, M. P. (2005). Memory consolidation and reconsolidation: what is the role of sleep? Trends in neurosciences, 28 (8), 408-415.
Stickgold, R., & Walker, M. P. (2007). Sleep-dependent memory consolidation and reconsolidation. Sleep medicine, 8 (4), 331-343.
Storch, E. A., Roberti, J. W., & Roth, D. A. (2004). Factor structure, concurrent validity, and internal consistency of the Beck Depression Inventory- Second Edition in a sample of college students. Depression and Anxiety, 19, 187-189.
Van Emmerik, A. A. P., Schoorl, M., Emmelkamp, P. M. G., & Kamphuis, J. H. (2006). Psychometric evaluation of the Dutch version of the posttraumatic cognitions inventory (PTCI). Behaviour Research and Therapy, 44, 1053– 106.
Van Minnen, A., & Arntz, A. (2007). Behandelprotocol posttraumatische stressstoonis (3rd ed.) Amsterdam: Boom Cure & Care.
Van Stegeren, A. H., Rohleder, B., Everaerd, W., & Wolf, O. T., (2006). Saliva alpha amylase as marker for adrenergic activity during stress: effect of betablockade. Psychopsysiology, 69, 33-40.
Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology, 54(6), 1063-1070.
