Master Thesis Clinical Psychology
The Influence of Propranolol on Fear Memory
Reconsolidation in Patients with Panic Disorder
Jana Barkhof
Faculty of Clinical Psychology, University of Amsterdam
Student number: 10196099 Supervisor: Jamie Elsey Date: 29-07-2017
Abstract
This study investigates whether the administration of propranolol HCL after reactivation of a panic attack is an effective intervention for patients with panic disorder. Traditional cognitive behavioural therapy supports the idea that fear reduction results from the formation of a new inhibitory memory trace, which competes with the old fear memory. The current study aims to explore whether the old fear memory trace in patients with panic disorder can be changed directly. Fourteen participants, divided into a reactivation+ propranolol (PP_prop) condition and reactivation + placebo (PP_placebo) condition, took part in this pilot study.Results show that participants who were given propranolol after the first session had a larger decrease in panic symptom severity compared to the participants who were given a placebo. Results reveal no significant effects of the propranolol based intervention compared to placebo on the amount of panic attacks per week but show a positive trend on the perceived physical consequences during a panic attack. These results are quite promising, but further research is needed with a focus on replicating this study with a larger group of participants to get more conclusive results.
Index
1. Introduction p. 4
1.1 Propranolol and the Process of Reconsolidation p. 4
1.2 Introduction on Panic Disorder p. 6
2. Problem Definition and Aim of the Current Study p. 8
2.1 Current Models and Theories on Fear Acquisition p. 8
2.2 Current Treatments for Panic Disorder and a New Treatment Approach p. 11
3. Hypotheses p. 13
4. Method p. 14
4.1 Participants p. 14
4.2 Operationalisation p. 15
4.3 Apparatus & Materials p. 16
4.5 Procedure p. 17
5. Results p. 20
6. Discussion p. 30
1 Introduction
1.1 Propranolol and the Process of Reconsolidation
Propranolol has recently been suggested as a possible agent to change fear memories (Steenen et al., 2016). Propranolol is a beta-adrenergic receptor antagonist, which is commonly used in treatments for hypertension and performance anxiety (Vaiva et al. 2003). It was first introduced as a medicine for the treatment of angina pectoris, a heart condition in which people experience chest pain (File &Lister, 1985), but propranolol can also be used for migraine or severe headaches (File & Lister, 1985). In the treatment of performance anxiety propranolol is given before a patient enters a highly anxious state (Lockwood, 1989; Kenny, 2005).
This thesis focuses on the effect of propranolol on memory restabilisation when it is administered after such an anxious state.
To understand in which way propranolol has an effect on the brain and fear memory, it is firstly important to understand the neurobiological processes that take place in developing and maintaining an anxiety disorder. Through the release of adrenergic hormones, anxiety can reinforce emotional memory (Friedman, 2016). Hence, it is much harder to forget being in a scary situation (e.g. getting a panic attack) than to forget where you’ve put your credit card. Noradrenaline plays a central role during the process of memory enhancement. It is released in anxiety provoking situations and can serve to strengthen the learning process and the emotional memory for that event (Kindt, Soeter & Vervliet, 2009).
Reconsolidation is the stabilisation of a memory after it has been reactivated (Kindt, Soeter & Vervliet, 2009). More specifically, after reactivation the memory turns into a temporary unstable state, during which it can be modified or disrupted. After entering such an unstable state, the memory must be restabilised or reconsolidated (Nader, Schafe & le Doux, 2000). However memory reconsolidation does not automatically take place when the memory has been reactivated (Lee, 2009). If retrieval per se was sufficient to destabilise the memory trace, memory would be hyper-adaptable (Sevenster, Beckers & Kindt, 2012). It is suggested that the function of
reconsolidation is to keep the memory updated with new learning (Sevenster, Beckers & Kindt, 2013). Therefore it seems that reconsolidation only happens when one can learn something from that particular situation during the memory retrieval (Morris et al. 2006). This is called prediction error; it is a violation of expectation, which is based on prior learning (Forcato, Rodriguez, Pedreira & Maldonado, 2010). In that sense updating of the fear memory trace should not take place in retrieval conditions in which the outcome is totally predictable (Sevenster, Beckers & Kindt, 2012).
Propranolol passes the blood-brain barrier and blocks the beta-adrenergic receptors in the amygdala, which are crucial for the reconsolidation of the memory (Soeter & Kindt, 2010; 2011; 2012; 2013; 2015). Propranolol can disrupt the process of reconsolidation in such a way that a change can be made in the original memory trace (Soeter & Kindt 2013). Propranolol does not delete the whole memory trace (otherwise people would not remember being afraid of certain things or situations) but only focuses on the emotional component of the memory resulting in a decreased fear response (Lonergan, Olivera-Figueroa, Pitman & Brunet, 2013). However the process of reconsolidation is relatively short; there is a limited time after retrieving a fear memory to be disrupted with propranolol. When the interval is too long, the window of the
reconsolidation process closes causing the drug to have limited or no effect (Soeter & Kindt, 2010; 2011; 2012; 2013; 2015).
In different fear conditioning studies (Kindt & Soeter 2009; 2010; 2011; 2012; 2015; Sevenster, Beckers & Kindt, 2012, 2013, 2015) it has been shown that the process of memory reconsolidation can be interrupted by propranolol. Because reconsolidation seems to change the original fear memory instead of developing a new memory trace as is seen in other therapies such as cognitive behavioural therapy (Davis, Barad, Otto & Southwick, 2006), it is suggested that the results of fear reactivation after propranolol administration would last longer than cognitive behavioural therapy, and that recurrence of anxiety symptoms would be less frequent (Soeter & Kindt, 2015).
This effect is also seen in studies with patients with spider phobia. Disrupting the reconsolidation process causes a change from avoidance to approach behaviour towards tarantulas (Soeter & Kindt, 2015). In this study, participants with great fear of spiders received propranolol after fear memory reactivation after which they showed a decreased fear response and were even able to pet the tarantula (Soeter & Kindt, 2015). Related studies showed that through propranolol administration only the fear and avoidant behaviour decreased, leaving factual or declarative memory intact (Kind, Soeter & Vervliet., 2009; Sevenster, Beckers & Kindt, 2012, 2013, 2015). Hence, participants in this study were able to remember that they were afraid of spiders before, but remarkably didn’t show the emotional and behavioural response to the spider. It is however relevant to mention that in previous studies it has been shown that the absence of fear memory expression only occurred in patients whose fear memory was reactivated accurately and were given propranolol afterwards and not in those who only received propranolol (Kindt, Soeter, & Vervliet, 2009; Soeter & Kindt, 2015). This means that the success of treating patients with anxiety disorders by giving propranolol relies on whether their memory was properly reactivated and that the non-appearance of fear responding is not simply a result of the fear dampening effect of propranolol (Kindt & van Emmerik, 2016).
Different anxiety disorders do not have the same pattern of symptoms, but may share a similar neurobiological basis (Garakani, Mathew & Charney, 2006). It is interesting to examine whether this kind of intervention will be effective for other anxiety disorders, such as panic disorder. This pilot study aims to explore whether administration of propranolol after
reactivation of panic symptoms with the intention of disrupting fear memory reconsolidation, may reduce the symptoms of panic in patients with panic disorder.
1.2 Introduction on Panic Disorder
& Barlow, 2001). According to the DSM-5 a patient with panic disorder repeatedly experiences panic attacks (American Psychiatric Association, 2013). A panic attack is characterised by a rush of great fear or uneasiness, which reaches a peak in a period of ten minutes (Kampman, Keijsers & Hendriks, 2004). In this short period of time, patients can show some of the following
symptoms: palpitations, shaking, shortness of breath, feeling suffocated, chest pains, dizziness, nausea, derealisation, depersonalisation, fear to get crazy or lose control, fear of dying,
paraesthesias and cold shivers (van Balkom, Bakker, Visser, 2004). The panic attacks are typically unpredictable and can occur unexpectedly across a range of situations. (Blanco, Nissenson & Liebowitz, 2001). Therefore patients with panic disorder usually suffer from anticipation anxiety for longer periods of time and have enduring concern about the reappearance and the
consequences of possible attacks in the future (Kampman, Keijsers & Hendriks, 2004) or show significant behavioural change (such as avoidance behaviour) due to the attacks (van Balkom, Bakker & Visser, 2004).
If not treated, the symptoms tend to persist because the stress and worries alone can also cause bodily sensations such as palpitations, which in turn patients get scared of (Clark, 1986; Roth, Wilhelm & Pettit, 2005). This vicious circle can further be maintained by avoiding certain events and exhibiting safety behaviours such as continuously paying attention to the exits of a public space or ensuring to be with a protective other at all times (Salkovskis, Clark, Hackman, Wells & Gelder, 1999).
In the Netherlands the lifetime- and year prevalence of panic disorder are respectively 3,8 and 2,2 percent (Bijl, van Zessen & Ravelli, 1997; van Balkom, Bakker & Visser, 2004). The typical age of onset is somewhere between adolescence and early adulthood (Kessler et al., 2005), but patients with panic disorder usually seek treatment much later, around the age of 30
(Biederman et al., 1997). The prognosis of patients with panic disorder is favourable if they seek treatment (Vandereycken, Hoogduin & Emmelkamp, 2008), but panic disorder tends to be a chronic condition, which comes with considerable financial costs and interpersonal problems
(Wittchen et al., 2011). Most patients who have not been treated properly relapse within a few years (Emmelkamp & Wittchen, 2009).
2 Problem Definition and Aim of the Current Study
2.1 Current Models and Theories on Fear Acquisition
One explanation for the fact that some patients acquire symptoms of panic disorder is that acquiring fear in itself is an adaptive response to a threatening situation (Bos, Beckers & Kindt, 2012). The body prepares itself for fight or flight so that harm can be avoided. Under normal circumstances, such actions are required to keep one safe. However, anxiety may become maladaptive if, in the case of a panic attack, the feared events or sensations no longer
proportional to the potential dangers (Bos, Beckers & Kindt, 2012). The fear that patients with panic disorder experience during the first sudden panic attack is usually not validated because of the absence of a detectable or attributable precedent; it represents a ‘false alarm’ (Barlow, 2002).
There is an ongoing debate on how to successfully explain the acquisition of fear or developing an anxiety disorder, such as panic disorder (Bouton, Mineka & Barlow, 2001). Some propose that anxiety disorders can be properly explained by the Pavlovian fear-conditioning paradigm. In this model a neutral conditioned stimulus (CS) is paired with an unconditioned stimulus (UCS) (Bouton, Mineka & Barlow, 2001). The UCS is almost always a naturally aversive stimulus, such as experiencing palpitations, and the CS is a stimulus, which only gets its special meaning after a frightening situation occurs and functions as an alarming precursor for the UCS. The pairing of the CS and UCS results in the acquisition of a new fear memory trace, which will then be activated when confronted with the CS and evoke a conditioned fear response (Bouton, Mineka & Barlow, 2001). In other words, an associative memory trace will be formed (Soeter & Kindt, 2015). Moreover, in recent studies it has been found that the associative memory trace in patients with panic disorder can even foster greater fear generalisation and impair safety learning (Lissek et al., 2009b). As a result, patients with panic disorder have the tendency to generalise
their fear to numerous bodily states (e.g. other UCS) (Lissek et al., 2009a). Associative fear memory can result from direct conditioning experiences but can also be formed through vicarious or indirect fear learning (Soeter & Kindt, 2015).
The cognitive model suggests that physical sensations are interpreted as signs of
catastrophes that could occur; for example, palpitations could be a sign for having a heart attack (Bouton, Mineka & Barlow, 2001). It is suggested that people who are prone to have panic attacks are more sensitive to these bodily sensations than others (Gorman, Kent, Sullivan & Coplan, 2000). They are more inclined to see these sensations as red flags for medical
emergencies than to attribute these sensations to more probable causes, such as feeling bloated after a big lunch (Casey, Oei & Newcombe, 2004). The misinterpretations of the bodily
sensations increase the fear and the fear intensifies the interoceptive signals; in this way a patient with panic disorder gets into a vicious circle (Vandereycken, Hoogduin & Emmelkamp, 2008). For instance, many patients with panic disorder tend to (unintentionally) hyperventilate in
stressful situations (Dratcu, 2000). This makes them think that they could suffocate and therefore get a panic attack, which reinforces hyperventilation. In short, the cognitive model assumes that the catastrophic misinterpretations of the bodily sensations are a sufficient cause for the
development and continuance of panic disorder (Davis, Barad, Otto & Southwick, 2006). Thereby cognitive therapy targets these misinterpretations and tries to correct or change them (Vandereycken, Hoogduin & Emmelkamp, 2008).
According to the anxiety sensitivity (AS) model the fear of anxiety (‘the fear of fear’) is relevant in the development and continuance of anxiety disorders (Barlow, 2002; Clark, 1986, McNally, 2002; Schmidt, Zvolensky & Maner, 2006). Anxiety sensitivity is a cognitive
characteristic that makes individuals vulnerable to developing panic problems (Bouton, Mineka & Barlow, 2001). People with high AS are fearful of mental and physical detectable experiences, which can increase pre-existing anxiety and may lead to a vicious circle as previously described (McNally, 2002). The basis of this theory is that anxiety and its related symptoms can have
detrimental psychological, physical or social effects on patients with an anxiety disorder (Schmidt Zvolensky & Maner, 2006). These effects are far more invalidating than the immediate physical discomfort during a panic attack or episode of anxiety (Bouton, Mineka & Barlow, 2001). Reiss (1991) and McNally (2002) differentiate the AS theory from other theories or perspectives in two ways. Firstly, they see AS as a permanent trait-like tendency as opposed to a situational tendency (McNally, 2002). Secondly, AS theorists presume that patients with panic disorder are usually very conscious of the causes of their sensations and yet feel frightened by them because they still have this deep rooted belief that the sensations could be catastrophic to their body or mental state (Reiss, 1991). This means they do not attribute these bodily signals to a dangerous cause, but believe that these signals are dangerous itself. Moreover, when comparing the AS theory to the cognitive theory, which points out the idea of an instant threatening catastrophe, the AS theory indicates that threats or harm of the symptoms may expand over time (Schmidt,
Zvolensky & Maner, 2006). Studies show that AS can be a risk factor for anxiety disorders. For example one study showed (Hayward, Ahmad & Wardle, 2000; Schmidt, Zvolensky & Maner, 2006) that people with a high AS have a higher chance to develop a panic attack in the future. Furthermore, in the study of Schmidt, Zvolensky and Maner (2006) it was found that AS could predict the occurrence of spontaneous panic attacks in people with no history of panic. These results strongly demonstrate that AS is a risk factor for developing anxiety disorders.
As mentioned above, the current investigated intervention with propranolol tries to target the original emotional component of the memory trace, as seen in the studies of Soeter and Kindt (2010; 2011; 2012; 2013; 2015). The intervention with propranolol relates to the cognitive model and AS theory by breaking the link between the feared bodily sensations (and
subsequently the feared catastrophic consequences) and the feelings of anxiety caused by the bodily sensations. In other words, the emotional part of the memory (the feelings of anxiety) can be restructured and therefore cannot be linked to the feared bodily sensations any longer.
However, it could also be argued that the current intervention with propranolol also seems to work in an opposite way compared to the cognitive model used in cognitive therapy. As seen in the spider study (Soeter & Kindt, 2015), propranolol can change the emotional aspect of the memory trace causing the participants to modify their behaviour; the participants seemed to approach the spider rather than to avoid it. Their cognitions about the feared object (in that case the spider) changed afterwards (Soeter & Kindt, 2015). This suggests that the current investigated intervention with propranolol could circumvent the necessity of challenging the cognitions and tackles the cognitions after the behavioural response of the patients has been changed.
2.2 Current Treatments for Panic disorder and a New Treatment Approach
The most commonly used therapy for patients with panic disorder is cognitive
behavioural therapy (CBT), which consists of challenging dysfunctional cognitions, often through behavioural testing and exposure (Kampman, Keijsers & Hendriks, 2004). By means of exposure, patients learn that exposing themselves to bodily sensations isn’t as disastrous as they believe; new interpretations of the feared situation or object are introduced by the therapist. One
explanation for the efficacy (and limitations) of this kind of treatment is that a new memory trace develops to compete with or inhibit the old fear memory trace (Kindt, Soeter & Vervliet, 2009). In previous studies was shown that not every patient with an anxiety disorder benefits from this kind of treatment and that the relapse rate still remains high (Kindt, Soeter & Vervliet, 2009). A possible explanation for this might be that the older memory trace remains intact and can still be activated, even though there is a new alternative memory trace formed (Sevenster, Beckers & Kindt, 2015).
As an alternative or in addition to CBT, antidepressant medication is used in the treatment of panic disorder. Antidepressant medication is known for its decreasing effect on complaints of depression but it can also suppress anxiety (Nutt, 2002). Within the group of antidepressants, SSRI’s (selective serotonin reuptake inhibitors) are most commonly used (Vasile, Bruce,
Goisman, Pagano & Keller, 2005). This type of medication tends to focus on a supposed imbalance of the neurotransmitter serotonin. Taking SSRI’s is intended to restore the balance (Leo & Lacasse, 2008). These medications however do not target the original cause of the imbalance (Leo & Lacasse, 2008). When medication is eventually stopped, it is possible that the achieved balance will probably get disturbed again. Another disadvantage of antidepressant medication are the side effects such as agitation, headache, nausea, sleepiness, tremors and weight change (weight gain or weight loss), which may have a great impact on patients’ well being (Hu et al., 2004).
It seems that patients with panic disorder have to overcome major obstacles in their daily lives with precarious long-term treatment effects, since there still is a considerable relapse rate following cognitive behavioural treatment because of the remaining fear memory trace. As just mentioned, pharmacological interventions, such as antidepressant medication, can have some effect but also have unpleasant side effects and usually have to be taken for a longer period of time. Therefore, there is a need for alternative therapies to raise standards of care for patients with panic disorder. In this thesis a new experimental intervention is reviewed, which might help patients with panic disorder in overcoming their fear in a relatively short period of time.
As SSRIs have become the preferred pharmacological treatment across several anxiety disorders for a few decades, beta-blockers such as propranolol experienced declining attention as a potential treatment for these disorders (Baldwin et al, 2014; Steenen et al, 2016). However the neurobiological effects of propranolol have gained new interest and attention in more recent studies, because of their essential role in development and continuance of anxiety disorders (Steenen et al, 2016). In the earlier studies researchers investigated the general anxiolytic properties of propranolol, but more recently the focus has shifted to the amnesic effect on memory reconsolidation as a more relevant effect to examine (Steenen et al, 2016).
3. Hypotheses
In this pilot study participants were randomly assigned to a propranolol based condition, in which the panic was reactivated and propranolol was given afterwards, or to a placebo
condition, in which the panic was reactivated and a placebo pill was given afterwards. The procedure is explained further in the Method section.
It was expected that participants in the PP_prop (panic provocation + propranolol) condition would show a significant reduction in panic symptomatology after intervention, whereas no change was expected in participants in the PP_placebo (panic provocation + placebo) condition. This may be demonstrated by:
1: Lower scores on self-report measures of panic symptomatology post-intervention relative to pre-intervention in the PP_prop condition (and no change in the PP_placebo condition).
2: A significant reduction in the frequency of panic attacks in the PP_prop condition post-intervention relative to pre-intervention in the PP_prop condition (and a lack of reduction in the PP_placebo condition).
If the participants from the PP_prop condition did not show a significant reduction in panic symptomatology after intervention, the following three explanations could explain why this happened.
Firstly it can be concluded that either the propranolol wasn’t effective in disrupting the reconsolidation process or the memory reactivation was not effective in destabilising the memory trace. A different conclusion might be that propranolol or the memory reactivation was effective in disrupting the process of reconsolidation, but the participants displayed too many avoidant behaviours, which didn’t allow them to notice any change in anxiety in situations where they normally would have panic attacks. Thirdly, it is also possible that this kind of intervention is simply not effective in altering panic symptoms.
As a second measure the effects of the intervention during the three-month follow up (explained in the Procedure section) were examined. Participants from the PP_prop condition were expected to (still) show a significant reduction in panic symptomatology during the three month follow up relative to pre-intervention (for the PP_prop condition) or relative to the second session (for the PP_placebo condition).
If the participants from the PP_prop condition did not show a decline in panic symptomatology during the three-month follow up relative to the first session, either two explanations might be possible. One explanation might be that the intervention was either not successful at all - either the memory reactivation wasn’t effective in destabilising the memory or propranolol didn’t disrupt the process of reconsolidation. Another explanation could be that the intervention would only provide an effect on fear memory for a short period of time. Lastly in this pilot study only 14 participants participated, so it is possible that this pilot study would not have sufficient power to detect significant effects.
4. Method
4.1 Participants
A total of 14 participants took part in the double-blind placebo controlled pilot study. 10 participants were women, ranged in age from 19 - 63 years (mean ± SD age, 36.10 ± 12,79 years), whereas the 4 men ranged in age from 22 – 48 years (mean ± SD age, 31.25 ± 11.64 years)
As an inclusion criterion all participants had to be diagnosed with panic disorder and were screened by an intake interview and the SCID-I questionnaire. Also, a medical screening was performed because propranolol can have somatic effects; heart rate and blood pressure were assessed and the use of medication was assessed. Previous use of an antidepressant, another betablocker or other medication that could counteract the propranolol constituted a
At the start of the study a total of 34 participants were selected, from which 13 participants were excluded during the intake interview. Four of these excluded participants turned out to have social phobia instead of panic disorder, two of them turned out to have posttraumatic stress disorder, two of them had a specific phobia rather than panic disorder, one participant had epilepsy, which can result in symptoms much like panic disorder (Bahn et al., 2011), and four of the participants had treatment elsewhere, which excluded them from the study. During the study seven participants dropped out; one participant did not show up after the advisory meeting, two participants did not want to continue the study because of an alternative treatment offered elsewhere and four participants dropped out because their memory reactivation did not succeed after three trials. This meant that 14 participants completed the study.
Participants were randomly assigned to either the PP_prop condition (n= 7, 5 women) or the PP_placebo condition (n= 7, 5 women).
4.2 Operationalisation
The independent variable in this study was the intervention condition to which
participants were assigned (PP_prop condition vs. PP_placebo condition, further explained in the procedure section). The dependent variable in this study was panic symptomatology, which was measured in several ways.
The primary dependent variable was the PDSS-SR (Panic Disorder Severity Scale). The secondary variable to measure panic symptomatology was the score on the item of the Mobility Inventory, which measured the frequency of panic attacks during the past week (see below). Other exploratory dependent variables were the ACQ (Agoraphobic Cognitions Questionnaire), the BSQ (Body Sensation Questionnaire) and the SCID I diagnosis during the follow up
assessment. The study design was a randomised double blind placebo-controlled pilot study with condition (panic provocation + propranolol vs. panic provocation + placebo) as a between subject factor and time (pre- and post and follow up assessments) as a within subject factor.
4.3 Apparatus & Materials
Propranolol has a calming effect on heart rate and can cause a decrease in blood pressure (Stumpe et al., 1976). In this study it was therefore important to exclude participants with very low blood pressure or low heart rate (<60 bpm heart rate, or blood pressure below 90/60).
The SCID-I questionnaire was used, which is a structured clinical interview to determine several axis 1 disorders of the DSM-IV; based on the scores in the interview it was determined whether a patient suffered from panic disorder. The reliability of this questionnaire has been found to be moderate to good (Lobbestael, Leurgans & Arntz, 2011). According to Beekman & de Beurs (2004) the criterion validity of the SCID-I is good.
Other questionnaires that were used were the PDSS-SR (Panic Disorder Severity Scale), the ACQ (Agoraphobic Cognitions Questionnaire), the BSQ (Body Sensations Questionnaire) and the Mobility Inventory. The PDSS-SR is a questionnaire that measures the severity of a panic disorder. The questionnaire includes seven items in which the frequency of the attacks, stress during attacks, avoidance and impairment in functioning is assessed (Houck, Spiegel, Shear, Rucci & Stat, 2002). The (test-retest) reliability of this questionnaire was found to be high and validity was found to be good (Houck, Spiegel, Shear, Rucci & Stat, 2002).
The ACQ questionnaire assesses the frequency of certain thoughts or worries about having panic attacks; the so-called ‘fear of the fear’ (Chambless, Caputo, Bright & Gallagher, 1984). The reliability of the ACQ was found to be good and construct validity was found to be sufficient (Chambless, Caputo, Bright & Gallagher, 1984).
The BSQ questionnaire measures anxiety towards bodily sensations, which can occur when one is feeling tense or anxious (Chambless, Caputo, Bright & Gallagher, 1984). It assesses seventeen symptoms, which can arise when someone is having a panic attack and distinguishes somatic symptoms such as palpitations, from more psychological symptoms such as dissociation (Chambless, Caputo, Bright & Gallagher, 1984). The reliability of this test was found to be high and construct validity was good (Chambless, Caputo, Bright & Gallagher, 1984).
The Mobility Inventory measures the level of avoidance of certain places or situations by patients with panic disorder (and agoraphobia) (Chambless, Caputo, Jasin, Gracely & Williams, 1985). The reliability of this questionnaire was found to be good and construct validity was also found to be good (Chambless, Caputo, Jasin, Gracely & Williams, 1985).
During the first and second session of the intervention the participant received a registration form by the therapist. On this form participants were asked to report their
(subjective) anxiety during the panic provocation exercise and what they thought would happen during the exercise.
4.6 Procedure
a. Screening process
Before the intervention started, a screening process was performed. Participants received an online questionnaire to check for any medical or psychological contra indications for the use of propranolol. After this screening process an intake interview was planned.
b. Intake interview
In this structured interview participants received information on the nature of panic disorder and how fear memory traces are formed and can be disrupted or changed. After this participants were asked to sign an informed consent, in which the procedure of the study was explained. Participants were informed that there were two conditions (reactivation + propranolol vs. the reactivation + placebo condition) to which they were allocated randomly. Next, the complete SCID-I questionnaire was assessed, to confirm the diagnosis of panic disorder and to rule out other possible disorders.
Next the participants had to fill in a series of questionnaires: the PDSS-SR (Panic Disorder Severity Scale), the ACQ (Agoraphobic Cognitions Questionnaire), the BSQ (Body Sensation Questionnaire) and the Mobility Inventory (See the Materials section above).
A first session was planned when it was clear that the participants met the criteria for panic disorder diagnosis and didn’t have any other physical or psychological issues that could influence or interfere with the effect of the intervention. The participants were randomly assigned to either the propranolol condition or the placebo condition by an independent staff member; the participants and the therapist were blind to this allocation.
c. The three protocols used during the pilot study
During the pilot study the protocol of the procedure was slightly adapted. In the first protocol, participants came in after the intake for an advisory consultation; a face to face conversation about the diagnosis, which the participants received in the intake interview, and if they could participate in the study. After this the participant was asked to describe their latest panic attack and what he or she thought would reproduce a similar panic attack during the sessions and if he or she was willing to undergo this.
After a while the protocol was adapted (protocol 2) to a procedure without such an advisory consultation, because the therapists, who worked for the pilot study, discovered that such an appointment caused the participants to prepare themselves too much before the sessions, which resulted in fewer success to reactivate the fear memory. In other words, the preparation of the participants caused them to be more in control of their emotions and succeeded less in actually reproducing a panic attack.
A final adaption (protocol 3) was made in which participants were also required to fill in the series of questionnaires weekly in a period of four weeks between the first and second session. This was done to allow for additional analyses of the weekly development of the participants and to observe when the possible effect of the intervention would occur.
d. First session
After the advisory consultation (for the participants in the first protocol) or after the intake interview (for the participants in the second or third protocol) a first session took place. During this session the same series of questionnaires from the intake, were assessed again. Before the memory reactivation started, the expected negative outcomes of a panic attack (e.g. ‘I will lose control’) were made explicit to participants to support the disconfirmation (the prediction error, as explained in the introduction) thereof and to ensure the fear memory could be destabilised. After that, blood pressure and heart rate were measured. Participants were then asked to
stimulate a panic attack. To achieve this, participants were asked to hyperventilate for at least 90 seconds in an upright position, turning around in an office chair, running up the stairs or another reactivation situation thought of by the therapist based on the precise fear of the participants. After this the participants were asked if their anxiety was at a maximum level, which was
registered by the therapist. Scores ranged from 0 (not anxious at all) to 100 (very anxious). If the participant hadn’t reached a high level of anxiety (at least a score of 75), another round of
hyperventilation or another reactivation exercise would start. If that failed also, a new session was planned after a week to try to reactivate the fear memory once again. Propranolol was not given when the fear didn’t reach at least a 75-100 score.
If the participant did reach that level and subsequently experienced a panic attack, the reactivation exercise was stopped. The participant was asked to report the anxiety level on a scale from 0 to 100 and what he or she had expected to happen during the reactivation exercise. After this either a propranolol or a placebo pill was given to the participant. At this point both the therapist as well as the participant were unaware of the condition the participant was in and which pill was being given. The participant then had to wait for 90 minutes during which the blood pressure and heart rate were measured again to examine whether the propranolol did have a somatic effect (e.g. lowered blood pressure and heart rate). Another session was planned after four weeks.
e. Second session
In the second session after four weeks, the questionnaires and medical screening - as described in the first session - were assessed again. After this, another reactivation exercise was started to see if the same panic provocation procedure as used in the first session would still induce a panic attack. After this, the condition to which the participant had been assigned was revealed. If the participant was assigned to the propranolol based condition, the participant did not get another pill. If the participant had been assigned to the placebo based condition, and if they experienced another panic attack during this session, he or she received the propranolol pill at this point. This was done for ethical reasons to ensure that all participants would eventually get the same proposed effective intervention. For the participants in the placebo condition, who were given propranolol after the second session, the follow up session would function as a check whether the propranolol worked effectively in decreasing their anxiety.
f. Follow up session
The follow up sessions were planned after two months (three months after the first session). In this session the participants were asked how they were doing since the intervention and the same questionnaires were assessed again. After that the SCID-I was taken again to see whether the patient still qualified for a panic disorder diagnosis.
All assessments and interventions were performed by the team at the UvA PsyPoli, by psychologists with at least a master’s degree in clinical psychology and training in cognitive behavioural therapy.
5. Results
5.1 PDSS
The differences between the conditions and within the conditions were analysed for the PDSS scores, using a repeated measures ANOVA (intervention condition [PP_prop or
PP_placebo] x time [first session and second session]). This analysis could show if PDSS scores changed over time and if these scores differed between conditions.
There was a significant main effect of intervention condition on the PDSS scores,
F(1,11)=12.40, p=.01; meaning that the differences between the PDSS scores depended on which condition a participant was in. PP_prop condition showed a greater decrease in their scores from M=11.67 (SD=4.80) to M=7.33 (SD=5.68), whereas PP_placebo condition changed somewhat less, from M=11.14 (SD=6.15) to M=8.71 (SD=4.46), as seen in figure 1.
Figure 1. Mean PDSS scores of the different conditions at the first and second session.
There was also a significant effect of time, F(1)=48.13, p<.01, which means that overall, scores on the PDSS were lower in the second session relative to the first session. There was however no significant interaction effect between the intervention conditions and the scores of the PDSS, F(1,11)=.98, p=.34; meaning that changes in time of PDSS scores did not significantly differ between the two conditions. Additional independent t-tests and paired t-tests were
performed to further detect differences between the conditions. There were no significant differences between the two groups during the first session and second session (all p>.1). The
0 2 4 6 8 10 12 14 16 Session 1 Session 2 P DSS s co re s PP_placebo PP_prop
pairwise comparisons revealed that there was no difference in scores in the placebo condition between the first and second session, t(6)=1.74, p=.13, but there was a significant difference between the first and second session for the propranolol condition, t(5)=3.38, p=.02.
Another repeated measures ANOVA (intervention condition [PP_prop or PP_placebo] x time [second session and follow up session]) was used to analyse the differences between the conditions and within the conditions. The differences between the conditions and within the conditions for the PDSS scores during the second session and follow up measures are shown in figure 2.
Figure 2. Mean PDSS scores of the different conditions at the second session and follow up session.
There was no significant main effect, F(1, 8)=1.98, p=.20, meaning that the differences between the PDSS scores did not depend on which condition a participant was in. There was also no significant interaction effect, F(1, 8)=2.95, p=.12, showing no significant differences in time between the two conditions in PDSS scores. Since the PP_placebo condition also got
propranolol after the second session for ethical reasons as discussed in the method section, the conditions could not be meaningfully compared against each other at follow-up as a comparison between propranolol and placebo. Therefore two paired T-tests were also used, to compare the
0 2 4 6 8 10 12 Session 2 Follow up P DSS sco res PP_placebo PP_prop
results within the groups, as the PP_placebo turned into a new propranolol condition
(PP_newprop) comparable to the original PP_prop condition. Pairwise comparisons revealed no significant difference between the second and follow up measures for the PP_prop condition,
t(4)=-.17, p=.88, but revealed a significant difference in the PP_newprop condition between
second and follow up measures, t(4)=4,47, p=.01, meaning that the scores of the PP_newprop condition changed and were lower in the follow up session compared to the second session before they were given the propranolol.
5.2 Mobility inventory (frequency of panic attacks)
Scores on item 3 of the Mobility Inventory was measured by another repeated measures ANOVA, as this item referred to the frequency of the panic attacks during the past seven days. The differences between the conditions and within the conditions for the frequency of panic attacks during the first and second session measures are shown in figure 3.
There was a significant effect of time on the frequency of panic attacks, F(1)=11.82,
p=.01, revealing that overall the scores on this item of the Mobility Inventory were lower during
the second session than during the first session. There was no a significant main effect of the intervention condition on the amount of panic attacks during the past 7 days, F(1,11)=.59, p=.46. Participants from the PP_prop condition didn’t have a greater decrease in the amount of panic attacks during the past 7 days compared to the PP_placebo condition; there was also no interaction found between condition and the amount of panic attacks, F(1,11)=0.04, p=.84.
Figure 3. Mean scores on the Mobility Inventory item 20 (frequency of panic attacks during the past week) of the
different conditions at the first and second session.
5.3 Explorative analyses 5.4 BSQ
For the BSQ a repeated measures ANOVA was used. The differences between the conditions and within the conditions for the scores on the BSQ during the first and second session measures are shown in figure 4.
Figure 4. Mean BSQ scores of the different conditions at the first and second session.
0 0,5 1 1,5 2 2,5 Session 1 Session 2 Pa nic a tt ac ks du rin g o ne w ee k PP_placebo PP_prop 0 0,5 1 1,5 2 2,5 3 3,5 Session 1 Session 2 B SQ s co re s PP_placebo PP_prop
There was a main effect of time, F(1)=90.98, p<.01, indicating that both conditions displayed changes in scores on the BSQ over time. There was also a main effect of condition on the scores of the BSQ, F(1,11)=5.02, p=.05. This indicates that the differences between the BSQ scores depended on which condition a participant was in. There was however no interaction found between the condition and the BSQ scores of the different conditions, F(1,11)=.01, p=.75; the participants from the PP_prop condition didn’t have a significantly greater decrease in BSQ scores compared to the PP_placebo condition.
5.5 ACQ
Scores for the ACQ were measured during the two sessions. The section accounting for the physical consequences during a panic attack was measured separately from the section accounting for the social consequences during a panic attack. There was a main effect of time, F(1)=126.30, p<.01, indicating that the scores differed over time, regardless of the conditions. There was a main effect of condition on the scores on the ACQ, F(1,11)=41.67, p<.01. This indicated that the differences between ACQ scores depended on which group a participant was in. However no interaction was found between the condition and the scores on the physical consequences part of the ACQ, F(1,11)=3.59, p=.09, suggesting that participants in the PP_prop condition didn’t have a significantly greater decrease in their scores compared to the PP_placebo condition. Although this difference was not statistically significant, this was a trend-significant effect, showing that the average change in scores over time is greater in the PP_prop condition than in the PP_placebo condition, as seen in figure 5.
Figure 5. Mean ACQ (physical consequences) scores of the different conditions at the first and second session.
For the social consequences part of the ACQ, another repeated measures ANOVA was used. The mean scores and standard deviations thereof are found in table 1.
Table 1. Mean and standard deviations for the social consequences experienced during panic attacks measured by the
ACQ at the first and second session.
Time Condition ACQ (social consequences)
Mean (SD) Session 1 PP_placebo (n=7) PP_prop (n=6) 3.10 (0.99) 2.98 (1.19) Session 2 PP_placebo (n=7) PP_prop (n=6) 2.63 (1.13) 2.55 (1.50)
There was a significant effect of time on the scores on the social consequences part from the ACQ, F(1)=82.27, p<.01. This means that every participant had less concerns about the social consequences during a panic attack after intervention. There was however no main effect
0 0,5 1 1,5 2 2,5 3 Session 1 Session 2 A C Q ( ph ys ic al c on se qu en ce s) s co re s PP_placebo PP_prop
of condition on the scores of the ACQ, F(1,11)=3.22, p=.10, and also no interaction found between the condition and the decrease on the scores on the ACQ, F(1)=0.01, p=.93.
5.6 SCID-II diagnosis
A chi-square test was used to see how many participants still received a panic disorder diagnosis at the follow up appointment, as seen in table 2. The results show no significant
differences between the two conditions, χ2(1)=.31, p=.58. This was not surprising, as participants
in both conditions received propranolol at this point (participants of the PP_placebo condition received propranolol after the second session, turning into the PP_newprop condition).
Therefore the differences between the conditions could not be meaningfully compared. Nevertheless it was interesting to look at the amount of panic disorder diagnosis during follow up, to see whether the propranolol (either being given the first or second time) could lead to a remission of panic disorder diagnosis.
Table 2: The amount of participants with and without panic disorder diagnosis measured at the follow up session.
Condition No diagnosis Diagnosis still qualifies Total PP_newprop PP_prop Total 5 4 9 2 3 5 7 7 14
As seen in table 2, five participants from the PP_newprop condition no longer met the criteria of a diagnosis of panic disorder after receiving intervention in the second session. Still two of them qualified for a panic disorder diagnosis. For the PP_prop condition four participants no longer met criteria for the diagnosis, while three of them still did. In total more than half of the participants (9 out of 14) of this pilot study ended up without a panic disorder diagnosis.
5.7 PDSS weekly measures
As an exploratory qualitative analysis the scores on the PDSS were measured over 6 six time points, regarding the participants who participated during the third protocol, who were given weekly questionnaires. In figure 6 the scores per participant are presented to see whether both groups differed in scores on the PDSS over time and if one of the groups had a more rapid decrease in the scores.
Figure 6. Weekly PDSS scores for the participants in the PP_prop condition (marked as red) and the PP_placebo
condition (marked as blue) during the first session, the four weekly questionnaires and the second session.
Firstly, the PDSS scores for four of these participants dropped noticeably. It is possible that the intervention had an effective period during these two time points, however this is debatable considering the small number of participants and the alternative score pattern of participant 3 of the PP_placebo condition.
0 2 4 6 8 10 12 14 16
Session 1 Questionnaire 1 Questionnaire 2 Questionnaire 3 Questionnaire 4 Session 2
P
DSS s
co
re
s
PP_prop participant 1 PP_prop participant 2
PP_placebo participant 1 PP_placebo participant 2
Participant 1 of the PP_placebo condition showed a small decline in scores from session one to session two; the small drop in the scores might have been due to a slight time effect but the intervention did not seem to have any significant effect on the panic disorder symptoms. The score pattern for participant 2 of the PP_placebo condition is the most striking in figure 6. This participant started off with a high score of 15 at session one, dropped to a score of 3 at the third weekly questionnaire and then rose again at session two. This could mean a lot of things. One explanation for this score could be that the participant filled in their questionnaire
differently online during the weekly measures than during the session with the therapist present. Another explanation might be that in this case there might have been a exposure or placebo effect where this participant might have been convinced of a change in panic severity symptoms at first, but ultimately these symptoms and the fear returns after a while, because the fear memory is not actually tackled.
Participant 3 of the PP_placebo condition did not seem to have any high scores on the PDSS questionnaire during the whole trial. The scores of this participant didn’t seem to change much, suggesting this participant did not benefit all that much from the PP_placebo intervention.
Participant 1 of the PP_prop condition showed a preferable and expected score pattern during the whole trial. Dropping down from a score of 11 during session one to a six at the third weekly questionnaire and then to a 3 at session two, this participant did seem to display less panic disorder complaints suggesting a possible effective result of the PP_prop intervention.
Participant 2 of the PP_prop condition started off with an expected drop in scores from session one until the last weekly questionnaire but then rises again to a score of 7 at session two. However this last score was not as high as the score during session one. This participant could have either benefitted a bit from the intervention or time had an effect on the lowering of the scores, suggesting these participants still displayed some mild panic symptoms during the follow up session.
6. Discussion
Based on the present findings, it is hard to draw solid conclusions about the influence of propranolol on fear memory reconsolidation in patients with panic disorder. However, as
expected, the results showed that the intervention combining panic provocation with propranolol led to a decrease in panic symptom severity in patients with panic disorder, measured by the PDSS. Participants who received placebo during the first panic provocation did not show a significant decrease in panic symptom severity. There was however no significant interaction between the condition and the PDSS scores, meaning that the change in panic symptom severity did not significantly differ between the participants who received propranolol during the memory reactivation and those who received placebo.
The intervention with propranolol did not have a significant effect on the amount of panic attacks per week either, which was not expected. This could represent a floor effect. The number of panic attacks were quite low to begin with in both conditions. In some cases
participants only had one panic attack per week before the intervention started, so therefore the frequency of panic attacks for these participants could not change much. Overall this resulted in a considerable small change in the number of panic attacks, causing the difference between post and pre-intervention panic attacks to not be significant.
The long-term effect of propranolol on the panic symptom severity seemed stable, which was as expected. It is however interesting that the participants, who were in the placebo based condition, had a larger decrease in panic symptom severity after receiving propranolol during the second session. This effect may suggest that the interference with memory reconsolidation does not have to be a one-time opportunity, in which it is only possible to reactivate the memory once to get the propranolol to target the reconsolidation process. In other words the participants in the placebo condition seemed to benefit from the intervention with propranolol after the second session, even though they had already gone through a similar procedure with placebo. This may
suggest that propranolol is also effective the second time, which can be reassuring for those cases not succeeding during the first try.
The participants from the third protocol, who received propranolol after the memory reactivation, showed a pattern of decrease in symptom severity whereas the participants from the placebo based condition showed either no visible change in symptom severity or a smaller change in severity than the participants who received propranolol. Surprisingly between the first session and first online assessment there was a decrease in scores for almost every participant. This may be explained by the fact that the effects of propranolol on memory reconsolidation act quickly and it may also reflect that the subjective feeling of panic severity (the fear cognitions) is changed quickly as well after the reconsolidation. As only two participants from the propranolol condition were monitored at this point, this drop in scores could also be due to a time effect. Further research could focus on monitoring a larger group of participants during their weekly assessments, to see whether this decrease in scores take place during the same time points in different participants.
The influence of panic provocation in combination with propranolol does seem to have a favourable influence on panic disorder diagnosis, as measured by the SCID-I, as more than half of the participants did not have enough panic symptoms to qualify for a panic disorder diagnosis at the follow up session. Unfortunately the SCID-I was only assessed during the intake interview and follow up session, meaning that the results from both conditions could not be compared to each other as both the conditions eventually received propranolol. Hence, we cannot reasonably attribute these effects to the active intervention, rather than nonspecific effects such as the passage of time or a placebo effect. However, the trend towards less symptom reporting in the propranolol based condition relative to placebo indicates that the changes in symptoms might indicate a legitimate intervention effect. It would be strongly recommended for the upcoming larger randomised control study to also assess the SCID-I during the second session, before the
placebo based condition gets the propranolol, to examine the quantitative differences between the conditions.
The absence of significant results in most measures between the panic provocation with propranolol condition and the panic provocation with placebo condition seems first of all due to the small number of participants, as the small sample size provided low statistical power to detect differences between the two conditions. Secondly the small sample size might also have resulted in the study group not being fully representative for the whole population of people suffering from panic attacks. During the intake it was specifically noticeable that several participants already had different forms of therapy prior to the study, including CBT or antidepressant medication. These participants considered this study as their last resort to get rid of their panic attacks. It is possible that prior treatment has caused them to form a new inhibitory trace competing against their old fear memory. Perhaps for these participants, it is more difficult to tackle the old fear memory because of the interfering new inhibitory trace formed by prior therapy. It also possible that these participants were more difficult to treat or even resistant to treatment. It would be interesting to replicate this pilot study in a bigger randomised controlled trial with more participants to more easily detect changes between the conditions, as in that case there are more differences between the participants and their symptoms. This helps to see whether the propranolol condition does significantly differ from the placebo condition in the severity of panic symptoms. It would also be interesting to compare results between participants who did not have had a prior treatment against those who have had treatment, to see whether prior treatment perhaps causes an unfavourable prognosis for participants undergoing the current intervention.
When looking at the procedure of this study, it may be possible that procedural difficulties caused the scores of the PDSS and other questionnaires to not turn out as significantly different between the conditions as expected. After the sessions a lot of the times the therapists reported that they did not know exactly when the fear memory of the participant was actually activated the
most during the panic provocation and that they had trouble deciding when to give the (propranolol or placebo) pill. All participants were asked how high their fear was a couple of times during the fear reactivation exercise on a scale of 0 (not scared at all) to a 100 (completely terrified). The fear has to be at a maximum for propranolol to have the best effect, as otherwise extinction learning can happen (Kindt & van Emmerik, 2016) causing a new inhibitory trace to be formed, which competes with the effects of the propranolol trying to tackle the original memory. It is imaginable that some participants could not subjectively pinpoint whether their fear was at a maximum level or could have risen higher. This may have resulted in the therapists giving the pills anyway during a certain amount of time, hoping that the reactivation was
successful and propranolol could initiate access the reconsolidation process. Considering that the window of reconsolidation is relatively short (Sevenster, Beckers & Kindt, 2013, 2015; Kindt & van Emmerik, 2016), the propranolol had to be given quickly after the memory reactivation to disrupt the process. Therefore it could have been that for a couple of participants, who were given the propranolol, the actual memory reactivation did not fully succeed, causing the propranolol to not fully intervene on the memory reconsolidation.
Currently it is unknown which component of the session provides for a successful reactivation. The subjective anxiety measure used in the protocols did not always help the therapists figure out when to give the propranolol. Therefore further research has to focus on which features of the reactivation session can predict later improvement in symptoms.
Another explanation for the lack of significant change between the conditions might be that in some cases the participants did not have a prediction error during the panic provocation. The participant had to fill in their expectations of what would happen during the panic attack before the panic provocation and afterwards these expectations were checked whether they came true. If the participant did not have any expectations of what would happen or if these
expectations were not falsified during the panic provocation, there was nothing to be learned from that situation (the prediction error). When there is no prediction error the memory
reconsolidation will not happen, as there is no new information to be stored in the fear memory, causing the propranolol to not (fully) disrupt this process (Morris et al. 2006; Forcato, Rodriguez, Pedreira & Maldonado, 2010; Sevenster, Beckers & Kindt, 2012).
Contrary to the impressive results in the studies by Soeter and Kindt (2015) on the effect of propranolol on memory reconsolidation in patients with spider phobia, this pilot study shows somewhat less convincing results. One explanation for this might be the difference in
symptomatology between panic disorder and spider phobia. Spider phobia is relatively
straightforward as it solely focuses on one object (the spider). Patients with panic disorder show a more complex pattern of symptoms as they can fear different things or unexpected situations: one patient might fear their heartbeat as they think it could be a sign for having a heart attack, the other patient might be more anxious about passing out or tingling sensations in their body. Moreover patients with panic disorder easily generalise their fears and anxiety (Lissek et al., 2009a), so that a patient can start off with one fear symptom (palpitations) and end up with several (sweating, passing out etc.). Consequently it seems that the aspects a patient with panic disorder can fear are more heterogeneous than for patients with spider phobia. Therefore it is arguable that the intervention with propranolol tackles all facets of the fear memory more easily in patients with spider phobia than in patients with panic disorder, resulting in less convincing results.
Another factor, which might have had an influence on the fear reduction and memory reconsolidation process, is anxiety sensitivity. As previously described, anxiety sensitivity is a cognitive characteristic that makes individuals vulnerable to develop panic problems (Bouton, Mineka & Barlow, 2001). Soeter and Kindt (2013) found in their study that anxiety sensitivity predicted how much fear was reduced by disrupting the memory reconsolidation: the higher the anxiety sensitivity was, the less fear reduction (Soeter & Kindt, 2013). They suggested that individuals who have a higher level of anxiety sensitivity might need more memory retrieval sessions to target the fear memory (Soeter & Kindt, 2013). Considering that anxiety sensitivity is
a risk and maintaining factor in patients with panic disorder, it would be interesting to see in a future research whether patients with panic disorder would benefit more from extra memory retrieval sessions. It would also be advisable to include the Anxiety Sensitivity Index (Peterson & Reiss, 1992), a measure of AS, to compare the results from participants with higher levels of AS to participants with lower levels.
Currently the results of propranolol based intervention seem quite promising, but further research is needed to get more convincing results. What is learned from this pilot study is that patients with panic disorder do seem to experience a decrease in the intensity of their panic attacks, their fear of bodily sensations and feared physical and social consequences of the panic attacks. It is also clear that the reactivation process can be done at least twice with participants who received placebo the first time; the second reactivation did not seem to disturb the
decreasing effect of propranolol on panic symptom severity, it made it seem even better. This is encouraging for the patients who are more resistant to treatment. Therefore future research needs to focus on comparing the effects of propranolol during several reactivation time points.
Moreover future research may also focus on the long term effects of propranolol on the memory reconsolidation in patients with panic disorder by incorporating a one year follow up measure. It is known that a certain amount of patients relapse with CBT (Kindt, Soeter &
Vervliet, 2009). If the outcomes would remain stable between the 3 month follow up session and one year follow up for the current intervention with propranolol, clinical treatment using
propranolol could become a suitable and acceptable alternative for the (currently preferred) CBT. This may contribute to raise standards of care for patients with panic disorder.
7. References
American Psychiatric Association (2013). Diagnostic and Statistic Manual of mental disorders, fifth edition
(DSM-V). Arlington, VA: American Psychiatric Association Publishing.
Bahn, R. S., Burch, H. B., Cooper, D. S., Garber, J. R., Greenlee, M. C., Klein, I., Laurberg, P., McDougal, I. R., Montori, V. M., Rivkees, S. A., Ross, D.S., Sosa, J. A., & Stan, M. N. (2011). Hyperthyroidism and other causes of thyrotoxicosis: management guidelines of the American thyroid association and American association of clinical endocrinologists.
Thyroid, 21 (6), 593-646.
Baldwin, D. S., Anderson, I. M., Nutt, D. J., Allgulander, C., Bandelow, B., den Boer, J. A., Christmas, D. M., Davies, S., Fineberg, N., Lidbetter, N., Malizia, A., McCrone, P., Nabarro, D., O’Neill, C., Scott, J., van der Wee, N., & Wittchen, H.U. (2014). Evidence-based pharmacological treatment of anxiety disorders, posttraumatic stress disorder and obsessive-compulsive disorder: a revision of the 2005 guidelines from the British association for psychopharmacology. Journal of Psychopharmacology, 28 (5), 1-37.
Barlow, D. H. (2002). Anxiety and its disorders: the nature of treatment and anxiety and panic. New York, NY: The Guilford Press.
Beekman, A. T. F., & de Beurs, E. (2004). Meetinstrumenten bij aanmelding in de psychiatrie.
Biederman, J., Faraone, S. V., Marrs, A., Moore, P., Garcia, J., Ablon, S., Mick, E., Gershon, J., & Kearns, M. E. (1997). Panic disorder and agoraphobia in consecutively referred children and adolescents. Journal of American Academy of Child & Adolescent Psychiatry. 36 (2), 214-223.
Bijl, R. V., van Zessen, G., & Ravelli, A. (1997). Psychiatrische comorbiditeit onder volwassenen in Nederland: het NEMESIS onderzoek. Nederlands Tijdschrift voor Geneeskunde, 418, 2453-2460.
Blanco, C., Nissenson, K., & Liebowitz. M. R. (2001). Social anxiety disorder: recent findings in the areas of epidemiology, etiology and treatment. Current Psychiatry Reports, 3, 273-280.
Bos, M. G. N., Beckers, T., & Kindt, M. (2012). The effects of noradrenergic blockade on extinction in humans. Biological Psychology, 89, 598-605.
Bouton, M. E., Mineka, S., & Barlow, D. H. (2001). A modern learning theory perspective on the etiology of panic disorder. Psychological Review, 108 (1), 4-32.
Casey, L. M., Oei, T. P. S., & Newcombe, P. A. (2004). An integrated cognitive model of panic disorder: the role of positive and negative cognitions. Clinical Psychology Review, 24 (5), 529-555.
Chambless, D. L., Caputo, G. C., Bright, P., & Gallagher, P. (1984). Assessment of fear of fear in agoraphobics: the body sensations questionnaire and the agoraphobic cognitions
Chambless, D. L., Caputo, G. C., Jasin, S. E., Gracely, E. J., & Williams, C. (1985). The mobility inventory for agoraphobia. Behaviour Research and Therapy, 23, 35-44.
Clark, D. M. (1986). A cognitive approach to panic. Behaviour Research and Therapy, 24 (4), 461-470.
Davis. M., Barad, M., Otto, M., & Southwick, S. (2006). Combining pharmacotherapy with cognitive behavioural therapy: traditional and new approaches. Journal of Traumatic Stress,
19(5), 571-581.
Dratcu, L. (2000). Panic, hyperventilation and perpetuation of anxiety. Progress in
Neuro-Psychopharmacology and Biological Psychiatry, 24 (7), 1069-1089.
Emmelkamp, P. M., & Wittchen, H. U. (2009). Specific phobias. In G. Andrews (Ed.)
Stress-induced and fear circuitry disorders: advancing the research agenda for DSM-V (p. 77-104), Arlington,
VA: American Psychiatric Association.
File, S. E., & Lister, G. (1985). A comparison of the effects of lorazepam with those of
propranolol on experimentally induced anxiety and performance. British Journal of Clinical
Pharmacology, 19, 445-451.
Forcato, C., Rodriguez, M. L. C., Pedreira, M. E., & Maldonado H. (2010). Reconsolidation in humans opens up declarative memory to the entrance of new information. Neurobiology of
Learning and Memory, 93 (1), 77-84.
Garakani, A., Mathew, S. J., & Charney, D. S. (2006). Neurobiology of anxiety disorders and implications for Treatment. The Mount Sinai Journal of Medicine, 73 (7), 941-949.
Gorman, J. M., Kent, J. M., Sullivan, G. M., & Coplan, J. D. (2000). Neuroanatomical hypothesis of panic disorder, revised. The American Journal of Psychiatry, 157 (4), 493-505.
Hayward, P., Ahmad, T., & Wardle, J. (2000). Attention to bodily sensations: a test of the cognitive-attentional model of panic. Depression and Anxiety, 12 (4), 203-208.
Houck, P. R., Spiegel, D. A., Shear, M. K., Rucci, P., & Stat, D. (2002). Reliability of the self report version of the panic disorder severity scale. Depression & Anxiety, 15, 183-185.
Hu, X. H., Bull, S. A., Hunkeler, E. N., Ming, E., Lee, J. Y., Fireman, B., & Markson, L. E. (2004). Incidence and duration of side effects and those rated as bothersome with selective serotonin reuptake inhibitor treatment for depression: patient report versus physician estimate. Journal of Clinical Psychiatry, 65 (7), 959-965.
Kampman, M., Keijsers, G., & Hendriks, G. (2004). Protocollaire behandeling van patiënten met een paniekstoornis met of zonder agorafobie: Interoceptieve exposure, cognitieve
gedragstherapie en exposure in vivo. In Protocollaire behandelingen in de ambulante geestelijke
gezondheidszorg I. (pp. 32-62). Houten: Boom Stafleu van Loghum.
Kenny, D. T. (2005). A systematic review of treatments for music performance anxiety. Anxiety,
Stress & Coping, 18 (3), 183-208.
Kessler, R. C., Berglund, P., Demler, O., Jin, R., Merikangas, K. R., Walters, E. E. (2005). Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the national comorbidity survey replication. Archives of General Psychiatry, 62, 593-603.
Kindt, M., Soeter, M., & Vervliet, B. (2009). Beyond extinction: erasing human feat responses and preventing the return of fear. Nature Neuroscience, 12 (3), 256-258.
Kindt, M., & van Emmerik, A. (2016). New avenues for treating emotional memory disorders: towards a reconsolidation intervention for posttraumatic stress disorder. Therapeutic
Advances in Psychopharmacology, 6(4), 283-295.
Lee, J. L. C. (2009). Reconsolidation: maintaining memory relevance. Trends in Neuroscience, 32 (8), 413-420.
Leo, J., & Lacasse, J. R. (2008). The media and the chemical imbalance theory of depression.
Society, 45, 35-45.
Lissek, S., Rabin, S. J., McDowell, D. J., Dvir, S., Bradford, D. E., Geraci, M., Pine, D. S., & Grillon, C. (2009a). Impaired discriminative fear-conditioning resulting from elevated fear responding to learned safety cues among individuals with panic disorder. Behaviour Research
and Therapy, 47 (2), 111-118.
Lissek, S., Rabin, S., Heller, R. E., Lukenbaugh, D., Geraci, M., Pine, D. S., & Grillon, C. (2009b). Overgeneralization of conditioned fear as a pathogenic marker of panic disorder. The
American Journal of Psychiatry, 167 (1), 47-55.
Lobbestael, J., Leurgans, M., & Arntz, A. (2011). Inter-rater reliability of the structured clinical interview for DSM-IV axis I disorders (SCID I) and axis II disorders (SCID II). Clinical
