Hypotheses

The investigators hypothesize that among individuals who suffer from panic disorder there is higher incidence of co-morbid balance impairment than in the healthy population.

The investigators hypothesize that the treatment of panic disorder, through the treatment of co-morbid balance impairment using virtual reality (VR) exposure therapy environment, is more effective than the exposure to still pictures from the same scene in VR without balance challenge or comparing to standard cognitive behavioral therapy (CBT) for the treatment of panic disorder.

Rationale

This research relies on previous studies, which have shown mutuality between anxiety and balance impairment, even if only sub-clinical. The VR-based training environment enables multi-sensorial stimulus in a dynamic interactively changing setting. With the addition of a cognitive task (dual task distracting the fear), the investigators can add cognitive load and therefore challenge the control of balance even more. Individuals who suffer from balance impairment avoid their exposure to many balance-challenging situations - a fact that may increase their anxiety. The investigators assume that a considerable number of PD individuals also experience balance control impairments - mostly subclinical ones. Moreover, balance impairment accompanies other psychiatric disorders, though not enough literature exists on the subject.

Introduction

Anxiety disorders are a group of ailments, which are characterized by excessive and abnormal vague senses of apprehension followed by stress and restlessness (Berrios 1999). As a group, anxiety disorders were found to have the highest lifetime prevalence, i.e. they are the most commonly occurring class of mental disorders (Kessler et al. 2010) with 17.7% prevalence in a year (Kessler et al. 2005). This group of disorders consists of five main mental disorders: specific phobia, obsessive-compulsive, post-traumatic stress, generalized anxiety and panic disorder - which the investigators will be focusing on.

Panic disorder (PD) is a chronic ailment characterized by spontaneous, unexpected occurrence of panic attacks, i.e. periods of intense fear, which vary dramatically in the recurrence of panic attacks between patients and within a patient, ranging from a few per day to a few per year (American Psychiatric Association 2000). PD often occurs in conjunction with agoraphobia, i.e. the fear and avoidance from situations and settings they perceive they have little control on, such as public places and 'suffocating' surroundings (Kessler et al. 2005). PD can reach a lifetime prevalence of 5% (Kessler et al. 2010) and above (Serretti et al. 2011) and women are two to three times more likely to be affected than men (Schumacher et al. 2011). PD mostly develops in young adulthood (25y mean age), however this is not mandatory as onset during childhood or early adolescence most likely remains undiagnosed (Schumacher et al. 2011). Individuals suffering from PD are also said to have a deregulated autonomic nervous system at rest and during orthostatic challenge (i.e. tilt test), displayed by heightened heart rate and diastolic blood pressure compared to those of healthy controls (Martinez et al. 2010).

Many years of research have been invested in seek of an association between anxiety or PD and impaired physical balance (Perna et al. 2001; Sklare et al. 2001; Staab and Ruckenstein 2003; Staab 2006; Redfern et al. 2007). Many times the treatment of anxiety disorders in general, and PD among them, which target limbic areas, fail to succeed; thus it is recommended to treat the comorbid balance impairment, which its treatment alone can ameliorate anxiety (Shefer et al. 2010). Literature on the subject has indicated high prevalence of comorbid balance disorders and anxiety in general (Alvord 1991; Balaban and Jacob 2001; Kalueff et al. 2008) as well as anxiety disorders specific ranging from agoraphobia, panic attack and generalized anxiety (Alvord 1991; Balaban and Jacob 2001; Balaban and Thayer 2001; Jacob and Furman 2001). Anxiety in general and PDs are a common co-morbidity in acute and chronic vestibular disorders (Eagger et al. 1992; Matheson et al. 1999; Jacob and Furman 2001; Pollak et al. 2003), evolving from vestibular deficiency or sensorimotor disintegration. In spite of abundant literature supporting anxiety-imbalance relationship, it is still uncertain whether they are causality related.

Further investigation on the subject revealed that individuals who suffer from PD may also display altered cognitive patterns (Casey et al. 2004; Starcevic and Berle 2006). In addition, last decade research in the cognitive field has been dedicated to evaluating cerebellar involvement (Andreasen and Pierson 2008). Studies in animal models (Brodal et al. 1991; Dum and Strick 2003; Kelly and Strick 2003) have shown cerebellar connection to the frontal cortex through the pons and thalamus, suggesting its participation in neural circuits mediating higher cognitive functions. While cerebellar lesions in animals have indicated motor coordination and motor learning impairments (Thach 1996; Bastian et al. 1998), studies in humans have assigned the cerebellum a role in time intervals perception (Akshoomoff and Courchesne 1992), rhythm sequences imitation (Fiez et al. 1995) and associative learning (Attwell et al. 2002). Hence, the cerebellum seems to act as a modulator of cognitive processes performed by the cerebral cortex, which detects patterns and pattern changes as well as errors in both motor and cognitive functions and feedbacks the cerebrum (Andreasen and Pierson 2008).

Further support for this association lies in the theory of learning. Erez et al. (Erez et al. 2004) present a three-stage of learning theory, which includes an acquisition stage and two distinct stages of conditioned response (CR): the emotional CR, driven by the autonomic and hormonal systems, is fast and non-specific and the motor CR, provides an effective CR and is slow but specific. Despite the apparent differences, these two stages seem to be highly interrelated as can be explained by a frightening stimulus: fear CRs are acquired, emotional fear CRs take over and promote motor CRs. Contrariwise fear CRs may take over, if a subclinical balance disorder exists, it could be postulated that intervention of balance may reduce PD.

Balance maintenance itself occurs automatically and is involved in the execution of basic motions, which enable daily life activities (Keshner 1994; Kurtzer et al. 2005) as well as high performance skills such as ballet (Koutedakis and Jamurtas 2004) and athletics (Hrysomallis 2011). Normally, once a balance task becomes automatic, voluntary or cognitive action are enabled without a trade-off (Adi-Japha et al 2008). The success of performing any motor task lies on the ability to conduct task-related voluntary movements on top of constant involuntary postural adjustments (Keshner 1994; Kurtzer et al. 2005). Thus, each task-related maneuver involves the contraction of muscles in the limbs and trunk for the maintenance of body stability. Balance maintenance also requires constant integration between sensory input coming from vestibular, visual, somatosensory (mainly proprioceptive) channels and the motor system (Shefer et al. 2010). Converging from different sensory sources, this system addresses various balance maintenance aspects (Massion and Woollacott 1996) and so the loss of sensory input from either one can result in the alteration of posture (Le and Kapoula 2008).

Impaired vestibular function causes imbalance, nausea, vertigo and abnormal eye movements (Gilman and Newman 1996) and hinders the acquisition and emission of corrective movements. Such occasions are associated with limbic-related anxiety, perhaps through the parabrachial circuit (Balaban and Thayer 2001). The parabrachial nucleus (PBN) is reciprocally connected with the spinal cord, hypothalamus, amygdala and limbic cortex, and thus receives vestibular nuclear and visceral afferent information; therefore links vestibulo-visceral integrations especially in the adjustment to gravito-inertial challenges. In a recent study evaluating vestibular mutation in mice, it was found that an elevated plus-maze platform and an open-field test contributed greatly to anxiety levels (Shefer et al. 2010). While under rotation, extra-cellular recordings gave support to vestibulo-recipient areas of the PBN integrate signals from the vestibular nuclei and convey information about whole-body's orientation to pathways that produce homeostatic and affective responses (Balaban et al. 2002); giving further support to panic and impaired balance co-morbidity.

Individuals who suffer from vestibular or balance disorders many times demonstrate symptoms of anxiety, and conversely, patients with anxiety disorders often show abnormalities of clinical balance or vestibular function tests (Bronstein et al. 2004; Furman et al. 2005). Perna et al. (Perna et al. 2001) showed that PD patients demonstrate larger velocity and length in body sway in comparison to healthy controls. An additional study measured postural sway in anxiety patients as a response to a moving visual environment and found greater sway among the anxiety group comparing to the healthy controls (Redfern et al. 2007). A Parkinson's disease study has shown that in most cases they did not suffer from a specific vestibular disease, but do have subclinical abnormalities of the balance system. These patients control their posture relying on non-vestibular cues, proprioceptive and mostly visual input (Staab 2006; Caldirola et al. 2011).

Despite some considerable evidence in the literature, it is still difficult to isolate and verify balance impairment in PD. Thus the investigators propose the implementation of virtual reality exposure therapy. The term virtual reality (VR) refers to the use of an interactive computerized environment (software and hardware) simulating real world settings, which enable the immersion of the individual in trial (Weiss et al. 2005). With its high sensitivity, it is possible to create an environment where the individual is gradually led into a motivating and challenging environment, which requires his/ her alertness and physical attendance rather than imaginational capabilities. Within the VR it is possible to design a setting that consists of motivational and challenging elements, which indicate for the individual regarding his/ her own capability and competence (in contrast to his/ her impotence and passiveness) and from the amended experience (triumph is achieved). In addition, the therapist has the unique opportunity to be part of that environment in real-time and straightforward.

VR exposes the examinee as well as the therapist to salient components of the panic evoking situation and by visual, auditory and motor elements the stressing environment is reconstructed and reveals its panic-provoking roots (Roy et al. 2010). VR exposure enables the identification of intimidating moments as well as confident ones, providing the examinee with a competing instrument to deal with the stressing moments, while allowing the environment to change and adjust exposure levels and challenge the examinee (Wiederhold and Wiederhold 2010). VR exposure is done in laboratory conditions and under full supervision, thus it addresses individuals who are physically incapable of outdoors training as well. All in all, the utilization of the VR method allows the employment of a stepwise exposure approach, which is made by computerized and human fine-tuned modifications of the scene (Wiederhold and Wiederhold 2010).

Furthermore, VR exposure therapy, in practice for more than twenty years, supports the learning and training of neuro-motor tasks such as in flight, ski and surgery simulators. During the last few years, VR exposure method has made its way to clinical utilization in rehabilitation centers and hospitals, where balance and motor control are the main focus (Roy et al. 2010). The exposure to VR may contribute from several aspects, which are out of the reach of conventional therapy methods. VR exposure therapy provides an innovative and exceptional opportunity to reconstruct the scene and setting, which lead to panic behavior, and expose the examinee to critical factors consisting of the panic root. Altogether, VR systems enable the learning of simple and complex skills in a controlled environment, due to the endowment of monitored, graded, and adapted exposure of the various components constructing the therapy in a systematic manner (Rizzo et al. 1997a; Rizzo et al. 1997b; Schultheis and Rizzo 2001; Merians et al. 2002; Deutsch et al. 2004; Weiss et al. 2005).

Using VR, balance performance is evaluated precisely and efficiently by measuring body sway during quiet standing and during a few sets of perturbations (Nashner 1976; Allum and Carpenter 2005). The measured factor is the displacement of the center of pressure (CoP), i.e. a point at which a counter-reaction force vector of the vertical ground is applied. This force vector represents a weighted average of the pressure vectors of each foot on the force plates (Winter et al. 1990; Winter 1995). The CoP is a dynamic factor and displaces a few centimeters within a plane constantly (Winter 1995; Winter et al. 1996). Body sway is a kinematic term and is often estimated from the CoP displacement (in the anterior-posterior and medio-lateral axes) which is derived from force plate data, and is assumed to be synonymous with the CoP measure (Winter et al. 1998).

Balance maintenance and postural adjustments are modified by repeating experiences and improve with practice (Pollock et al. 2000), less muscles are recruited and responses refine and become subtle (Horak and Nashner 1986; Horak et al. 1989; Chong et al. 1999). Besides physical conditions, balance maintenance depends also on cognitive aspects such as attention, memory and dual tasking. Plotnik et al. (Plotnik et al. 2011) showed a relationship between dual tasking abilities and fall risk in patients suffering from Parkinson's disease, suggesting that Parkinsonian fallers are easily affected by dual tasking, with a great likelihood of loss of balance during the execution of daily attention-drawing activities. Thus, motor skills ranging form very simple to most complex ones can be performed and monitored for the characterization of learning processes; and training in virtual environments provides the opportunity to study learning skills and build learning curves (Holden 2005; Carelli et al. 2009). During VR, complete movement sequences are recorded as a pattern allowing full investigation motor skills, which is important in accordance with the notion that motor skill learning is task specific (Karni et al. 1998; Korman et al. 2007).

Many researches have recognized the potential of using VR in clinical treatment of psychopathologies, especially in the anxiety field. Some evidence on the application of VR exposure therapy to the treatment of specific phobias and PD can be recently found in the literature (Botella et al. 2004; Pull and Damsa 2008; de Carvalho et al. 2010; Gerardi et al. 2010; Meyerbroker and Emmelkamp 2010; Perez-Ara et al. 2010; Meyerbroker et al. 2011). A Belgian research studying motor vehicle accident trauma examinees showed that the use of VR was the most efficient method in treating phobias and that VR exposure provided a safe controlled therapeutic environment unrelated to treatment circumstances (Wiederhold and Wiederhold 2010). Therefore, VR exposure is applied interactively, since there is more value to an exposure where the examinee is active and the environment is dynamic and challenging; as exposure of this nature enhances alertness and emotional excitement, which are valuable to the treatment's success. The investigators will examine prevalence and severity in PD patients taking into consideration that they have higher prevalence of balance impairment.

The implementation of the VR method involves research and development costs as well as maintenance ones, and sometimes technical impediments may take place due to the relying on advanced and multisensory method. VR method calls for the recruitment of a technician beyond the professional therapist, which will be thoroughly trained and will operate a large number of software test-runs (pilots) before introducing it to patients. The common belief that multisensory exposure, triggering several systems at the same time distract the examinee's mind and prevent him/ her to be fully immersed (Roy et al. 2010) is refuted. In Sheba Rehabilitation Center for Advanced Technologies there is a VR laboratory running for years, enabling the rapid and successful rehabilitation of many patients. Salient and relevant specification of the VR laboratory can be adjusted to the treatment of balance disorder and panic. The investigators intend to implement VR method in the treatment of balance disorder patients, and to investigate its role in ameliorating panic behavior. The investigators further hypothesize that VR treatment can aid patients with challenging situations and improve neuro-motor functioning and thus better management in their everyday experiences.

Objectives

Stage 1 (evaluation) - to investigate whether there is balance impairment within PD patients in comparison to matched healthy subjects.

Stage 2 (treatment) - to investigate whether balance training using VR exposure therapy is significantly more effective in reducing panic symptoms, than the exposure to still pictures from the same scene without balance challenge, or from CBT standard protocol for the treatment of PD.

Study population

Stage 1 Sixty PD patients, men and women at the age of 18-45, will be recruited from Sheba Medical Center psychiatric clinics and an ad. PD will be assessed using SCID-1 checklist. Patients and 20 healthy controls matched by age, sex and BMI will complete the MINI.

Stage 2 The same 60 PD patients will be randomized to three treatment groups: (1) 20 through VR-based balance training; (2) 20 through exposure to still pictures from the same VR scene but no balance challenge; (3) 20 through CBT for PD.

Ethics

The following study includes a neurological examination which focuses on balance assessment based on questionnaires and a psychiatric examination for PD diagnosis. In addition, this research exposes the participants to a medical device (VR), which has been certified by the FDA and has received the Israeli Ministry of Health approval. This advanced method is used daily at the Advanced Technologies Department of the Rehabilitation Hospital at Sheba Medical Center. Furthermore, the identification and contact details of patients and healthy controls will be encoded to follow anonymity.

Methods

Stage 1: The investigators will collect demographic data such as age, sex and education, basic information such as height and weight for BMI calculation for the matching with healthy controls. The investigators will also collect medical history and concomitant medicine, clinical state as well as general morbidity and sports-related questions.

Stage 2: The investigators will ask the participants of this research not to take any part in psychotherapeutic treatment apart from the ones in this study.

Panic Disorder Diagnosis

Patients who participate will be diagnosed using the SCID-1 checklist and MINI for PD diagnosis and CGI. In addition the participants will go through the following for the assessment of PD severity and functionality under panic attacks:

• Hamilton Rating Scale for Anxiety (HAM-A).
• Panic Disorder Severity Scale (PDSS)
• Panic and Agoraphobia Scale (PAS) In addition, the investigators will use visual analogue scale (VAS) for the assessment of the panic severity level from one week to the other (will be applied in each visit).

Balance Assessment

Participants who are recruited will pass a neurological examination which includes a thorough balance assessment. Within this assessment the investigators will implement two self-rating questionnaires: Dizziness Handicap Inventory and the The Activities-specific Balance Confidence (ABC) Scale, as well as the functional assessment tools implemented by the neurologist:

• Berg Balance Scale
• Dynamic Gait Index, DGI
• Functional Gait Assessment, FGA Additionally, objective balance assessment will be done using VR as given in the detailed description.

Physiological measures

The investigators will also collect the following measures: blood pressure, Galvanic Skin Conductance and Heart Rate

CAREN system

The VR system the investigators will be using is the CAREN™ (Computer Assisted Rehabilitation Environment) Integrated Reality System, using the CAREN III software. This system is designed by MOTEK BV, (Amsterdam, The Netherlands, www.motekmedical.com). This system works in real-time and enables the creation of a variety of controlled and repeatable simulated environments via dedicated software with 3D visual, sound, and proprioceptive stimuli. It consists of the following components:

Motion Platform - CAREN™ consists of an electro-hydraulic motion 2 meter diameter motion platform (Rexroth Hydraudyne, MOTEK, Micro motion, www.rexroth-hydraudyne.com ) that can be manipulated with 6 degrees of freedom (x-y-z translation and pitch-roll-yaw rotation). The movement of the platform is either driven by the subject's movements, or preprogrammed in synchrony with function curves that define a specific pathway in the virtual environment.

Force Plates - Two 51 × 47 centimeter force plates (AMTI - Advanced Mechanical Technology, INC, Massachusetts, USA, www.amtiweb.com) are embedded in the motion platform. The output of the force plates is three dimensional force and moment data from each platform as well as the CoP displacement.

Motion Capture - Up to 41 passive markers are placed on anatomical landmarks and sampled at 120 Hz to record 3D head, trunk and limb motion. The markers are detected by an optokinetic system which consists of 12 Vicon infra-red cameras having a resolution of 2 megapixels (www.vicon.com).

Projection: The Virtual scene is projected on a large screen (3m × 2.5m). Stereo projection may be used.

The 'Road scene' In a virtual reality setting, standing on a moving platform, subjects were required to maintain their balance, while advancing along a pre-defined road (the 'road scene'). The scene was projected onto a large screen, in front of the subject. The scene consisted of a virtual road, bounded on both sides by walls. The road itself had flat, straight sections, vertical bumpy sections (movement along the y-axis), right and left tilts (rotation of 'z axis' - "roll") and right and left translations (movement along the 'x axis' - "sway"). The platform's movement was correlated with the visual stimulus (i.e., when the subject arrived at a bump on the screen, the platform elevated. When the road tilted, the platform tilted accordingly in the same direction). The length of the road is 1230m. The road's velocity was 30 m/sec, which equals to 1.8 km/min or 108 km/h. The specific parameters of the platform excursions in the 'X' (pitch) and 'Z' axis will be determined following a pilot study.

The 'Road scene' with an additional task The voluntary, secondary task consisted of intercepting 18 moving targets (colored ball of 12.5 cm diameter) appearing above the road. Each target could be intercepted when appearing no further than 20 cm from the subject's hands (which are reflected on the screen as a 4.5 cm in diameter ball). The targets appear, one at a time, at pre-set points along the road alternately on the right and left sides of the subject's body. Each ball is displayed for 5 seconds from its appearance and then disappears (unless it has been intercepted). After intercepting a target, the subject was instructed to return to the 'home position'.

On either of the scenes, the challenge level may be heightened by increasing the platform's tilt.

Study Protocol

Stage 1 A 3-4 hours introductory meeting will take place and will include a couple of hours assessment using the SCID-1 checklist for the diagnosis of PD, as well as the implementation of the MINI for psychopathology elimination. A general neurological examination for both healthy controls and PD patients will include functional balance assessment and self-rating questionnaires, VR-based evaluation of objective measures of balance maintenance (described in detail). VR exposure will be done in three consecutive exposures: (a) VR road scene (movie only) - in case the movie itself evokes anxiety and panic symptoms; (b) balance training using VR (no movie); (c) movie +balance challenge.

The evaluation will be done in conditions, such as entering a dark closed room, assuring the minimum of anxiety is evoked. Therefore, the investigators will take VAS subjective report and physiological measurements from the participant for cardio-vascular fluent monitoring.

Stage 2 The same 60 PD patients will be randomized to three treatment groups: (1) 20 through VR-based balance training; (2) 20 through exposure to still pictures from the same VR scene but no balance challenge; (3) 20 through CBT for PD. At this stage too the investigators will use VAS subjective report of the panic levels during the last week as well as objective physiological measurements at each visit.

In addition to the neurologist and psychiatrist at the 5th and last visits, there will be a CAREN specialist and a physiotherapist from the VR crew in each visit.

1. Protocol of imbalance treatment using VR exposure therapy (10 visits) 9 visits of VR-training with/ without parallel task: road scene repeating, which takes 24 minutes, including preparation and the installment of up to 41 sensors; 45 minutes per visit. The challenge increases if possible and will consist of sharp slopes and bumpy roads and/ or cognitive load (additional task). In addition, specific balance parameters will be collected, HR, GSC and a physiological examination and a verbal conversation at the end of each visit will take place and inform the participant on his/ her improvement.

   During the 5th visit there will be another psychiatric evaluation using the same questionnaires and assessment tools.

   Last visit: a summary meeting with no VR exposure. The investigators will implement all the assessment tools and questionnaires, as well as a neurological examination. In this visit the investigators will repeat sway and perturbation measurements and will make a summary.

2. Protocol of imbalance treatment using VR exposure therapy (10 visits) 9 visits of exposure to still pictures from the same VR scene: still pictures in a loop, which takes 24 minutes, including preparation and the installment of up to 41 sensors; 45 minutes per visit. In addition, specific balance parameters will be collected, HR, GSC and a physiological examination and a verbal conversation at the end of each visit will take place and inform the participant on his/ her improvement.

   During the 5th visit there will be another psychiatric evaluation using the same questionnaires and assessment tools.

   Last visit: a summary meeting with no VR exposure. The investigators will implement all the assessment tools and questionnaires, as well as a neurological examination. In this visit the investigators will repeat sway and perturbation measurements and will make a summary.

3. CBT standard protocol for the treatment of PD (10 visits) A CBT standard protocol for the treatment of PD will be applied in all 10 visits. Each visit will last 1 hour, except for the 5th and last visits, which include a psychiatric assessment using the same evaluation scales and questionnaires as in the evaluation, and therefore will last 1.5 hours. CBT will be implemented by a therapist, i.e., a psychologist or psychiatrist trained for the treatment of panic disorder. The therapist evokes emotional feelings such as vulnerability, disappointment, frustration, joy etc. This supportive therapy helps the individual to normalize his/ her behavior at the time of a panic-evoking situation and thereafter. The therapist tries to lead the individual to self-empowerment by providing him/ her self-defense mechanisms and inner-strengthening tools.

Handling the Data

The investigators will assure anonymity of all data collected and medical discretion.

Length of Study

Stage 1: The investigators intend to interview 60 participants in order to recruit 60 PD patients, and at the same time the investigators will interview 20 matched healthy controls.

Stage 2:

The same 60 PD patients from Stage 1 will participate in a 10-12 weeks experiment. In addition, there will be a follow-up within 6 months and 12 months from end of trial, when the participants will be asked to fill assessment questionnaires regarding their PD.

Statistical Analysis

The investigators will test for normal distribution of the dependent variables and then will examine the differences between the groups using univariate analyses: for continuous independent variables, such as scores, the investigators will apply t-test and for categorical variables, such as sex - chi square. All using SPSS software.

Any data to be collected during this study will be used for research purposes only and does not have diagnostic significance at the moment. However, this study may have clinical realizations and application in the future. Should the investigators have important findings, this research group has intentions to publish them.

Inclusion Criteria:

Stage 1

Patients

• Clinical diagnosis of panic disorder
• Stable on the same drug and dosage for at least one month

Healthy controls

• Determined as healthy upon completion of the MINI
• At the age of 18-45 years
• Matching patients by age, sex and BMI

Stage 2

Patients only

• The same patients, who have participated in Stage 1.

Exclusion Criteria:

Stage 1

Patients

• Co-morbidity with another psychiatric disorder (phobia is acceptable)
• Cognitive dysfunction or a neurological disorder
• Patients reacting with extreme anxiety to VR exposure at screening
• History of substance and/ or alcohol abuse

Healthy controls (in addition to the exclusion criteria above)

• Any psychiatric diagnosis

Stage 2

Patients only

The same patients from Stage 1, therefore they follow the same exclusion criteria mentioned in Stage 1.