Biofeedback Intervention in Adolescents With Autism Spectrum Disorder

Evaluating the Potential of Respiratory-sinus-arrhythmia Biofeedback for Reducing Physiological Stress in Adolescents With Autism: a Randomized Controlled Study

This study consists of two phases, a cross-sectional and a interventional. The cross-sectional phase will provide more insight into the differences in the autonomic functioning between adolescents with autism spectrum disorder and their typically developing peers. During the longitudinal phase, the efficacy of an RSA biofeedback intervention will be examined in adolescents with autism spectrum disorder.

In recent years, numerous studies have investigated the potential contributing role of the autonomic nervous system (ANS) in psychiatric disorders, including autism spectrum disorder (ASD). One of the most commonly adopted physiological parameters for assessing ANS functioning in psycho-physiological research is based on the assessment of heart rate variability (HRV). The heart's pacemaker is controlled by both the sympathetic and the parasympathetic nervous system, which results in the natural variation of the interval between heart beats. One important component of HRV is commonly known as respiratory sinus arrhythmia (RSA). This component is solely mediated by the parasympathetic subsystem through direct projections of the vagus nerve from the brain stem to the heart and is therefore also known as an index of cardiac vagal modulation. The term RSA is adopted since it defines the phenomenon of the relation between heart rate variability and the respiratory cycle in certain circumstances. The clinical importance of RSA has been reported in multiple studies as higher RSA is associated with higher cognitive abilities, effective social behavior, fewer internalizing symptoms (e.g. anxiety and depression) and appropriate emotion regulation. In addition, lower RSA has been linked to conduct problems, trait hostility, anxiety disorders and depression. With respect to children and adolescents with ASD, several studies have investigated the ANS functioning by assessing RSA during various activities (e.g. cognitive tasks, non-social auditory tasks, social cognition tasks and social interaction tasks) as well as during resting situations. While some studies revealed no significant differences in RSA levels between ASD and typically developing (TD) peers, the majority of studies seemed to converge on demonstrating lower levels of RSA in patients with ASD. Furthermore, in these studies, lower RSA levels in ASD were associated with more social problems, internalizing problems, higher levels of anxiety, problems with emotional control, the presence of more severe visual and/or auditory sensory problems and the presence of repetitive and restrictive behaviors. In line with the link between lowered RSA and more psychological and social problems, it has been posited that it would be of great relevance to develop and validate an intervention that specifically targets the up-regulation of RSA-values, or the cardiac vagal modulation, in individuals with ASD . To do so, biofeedback-based interventions have been put forward as these involve the recording of physiological signals (e.g. RSA or skin conductance) which are transformed into visual (or auditory) presentations and subsequently provided to the individual as direct feedback. This visualisation of ongoing bodily sensations would provide an added value for individuals with ASD, as previous studies have reported a lower ability to consciously perceive internal bodily sensations (interoceptive awareness) in this population. Considering that adolescence is a vulnerable period with rapid changes in distinct developmental domains, this population will be targeted during this research project. As previously mentioned, the clinical importance of RSA in both typically developing populations and populations with ASD has already been demonstrated. However, due to methodological differences, these studies are frequently incomparable and/or often lead to conflicting evidence. Therefore, a cross-sectional phase will be conducted first in this research project in which adolescents with and without ASD will be recruited. In this phase, they will perform a stress-inducing test in which physiological, endogenic and behavioral data will be collected. The main focus of this first part is on RSA in which lower RSA-values are hypothesized in adolescents with ASD as compared to their typically developing peers, based on previously mentioned study results. The second part of this research project consists of the evaluation of an RSA-based biofeedback approach for the up-regulation of RSA in adolescents with ASD. In other words, we will explore the potential of this intervention for enhancing cardiac vagal modulation and the influence on behavioral problems related to lower RSA in individuals with ASD as determined in previous research and based on findings during the first part of this study. Importantly, the validation of an RSA biofeedback intervention for adolescents with ASD may open novel avenues for home-based interventions in this population where therapist contact can be limited and integration into the daily routine can be supported using biofeedback apps. In addition, the current worldwide circumstances due to the corona virus disease has provided clinicians with new and useful information regarding their patients with ASD as these patients perceived the increase of home-based interventions as something positive. Therefore, the biofeedback intervention in this research project may have an additional advantage due to its home-based approach. Finally, the promising results of RSA biofeedback interventions in clinical populations, including ASD, are reported in various domains, going from decrements in anxiety and depression symptoms to improvements of social functioning. Therefore, this intervention may be valuable for adolescents with ASD given their vulnerable position in development.

There are two intervention phases, a supervised intervention and a feasibility study for a non-supervised intervention (during the follow-up phase after the supervised intervention). During the supervised intervention (5weeks), participant will be randomly allocated to the RSA biofeedback intervention (1) or a sham-control intervention (2). Then a 5-week follow-up phase is included during which the feasibility study of a non-supervised RSA biofeedback intervention will take place. The two groups from the supervised intervention will each be split into two (a and b) where groups 1a and 2a will receive a non-supervised biofeedback intervention and groups 1b and 2b will receive no intervention. Allocation to groups a and b is also ad random. Important to note: only during the supervised intervention, blinding of participants is possible whereas during the non-supervised intervention, this is not feasible.

5 weeks with weekly supervised sessions of 30 minutes and daily home practice sessions of 20 minutes. The latter sessions can be divided into four bouts of 5 minutes which can be spread across the day. The main goal during this intervention is to breath at resonance frequency which is slower than the usual breathing frequency. Participants will be guided towards this lower breathing frequency by using appropriate breathing techniques and the provision of a breathing pacer during the supervised sessions. During home practice, the participants will be provided with an application which will visualise a breathing pacer at their resonance frequency. At the third session, participants will no longer receive a breathing pacer but are instructed to breath in phase with their heart rate which is visualised on the computer screen instead of the breathing pacer.

5 weeks with weekly supervised sessions of 30 minutes and daily home practice sessions of 20 minutes. The latter sessions can be divided into four bouts of 5 minutes which can be spread across the day. The sham-control treatment will follow the same steps as outlined in the supervised intervention arm. However, the participants in this control group will not receive any information regarding their own heart rate and they will not practice at their resonance frequency. Instead, a default mode will be shown during the sessions and in their application for home practice.

1 short guided session (30 minutes) and daily practice (20 minutes; 4 bouts of 5 minutes) for 5 weeks. The biofeedback protocol will be provided by an application, similar to the supervised protocol but without guidance throughout the sessions, except for the first session. During this first session, the information will be provided about the application and the heart rate sensor (Polar band) which will detect changes in heart rate during the training. In addition, the resonance frequency of the participant will be determined so they can practice breathing at this frequency for the next 5 weeks.

Change in Root Mean Square of Successive Differences between normal heart beats (RMSSD) as a measure of cardiac vagal modulation.

This measure will provide information concerning the cardiac vagal modulation of heart rate based on time domain analysis. This analysis will be performed based on heart rate recordings at multiple assessment timepoints (only for adolescents with autism spectrum disorder). In addition, during one assessment, change will be measured between the different tasks (baseline, stroop word-color interference task, rest, social stress recall task and rest) in both adolescent groups (with and without autism spectrum disorder).

At all assessment points: before supervised intervention (baseline) + after 5 weeks supervised intervention + after 5 weeks follow-up

Change in Logarithm of High-Frequency Heart Rate Variability (LnHF-HRV) as a measure of cardiac vagal modulation.

This measure is an additional primary outcome measure concerning cardiac vagal modulation of heart rate based on frequency domain analysis. This analysis will be performed based on heart rate recordings at multiple assessment points (only for adolescents with autism spectrum disorder). In addition, during one assessment, change will be measured between the different tasks (baseline, stroop word-color interference task, rest, social stress recall task and rest) in both adolescent groups (with and without autism spectrum disorder).

At all assessment points: before supervised intervention (baseline) + after 5 weeks supervised intervention + after 5 weeks follow-up

Physiological assessment of heart rate by using a ECG-monitoring with 3 electrodes placed on the chest. The unit measure will be expressed in beats per minute. The sensors and biofeedback device from 'nexus-10' will be used in combination with 'biotrace+' software. This measurement will be repeated at multiple assessment points (only for adolescents with autism spectrum disorder). In addition, during one assessment, change will be measured between the different tasks (baseline, stroop word-color interference task, rest, social stress recall task and rest) in both adolescent groups (with and without autism spectrum disorder).

Adolescents with autism spectrum disorder: At all assessment points: before supervised intervention (baseline) + after 5 weeks supervised intervention + after 5 weeks follow-up; TD adolescents: baseline only

Physiological assessment of the breathing pattern by using an elastic band with stretch-sensitive sensors around the waist. The unit measure will be expressed in breaths per minute. The sensors and biofeedback device from 'nexus-10' will be used in combination with 'biotrace+' software. This measurement will be repeated at multiple assessment points (only for adolescents with autism spectrum disorder). In addition, during one assessment, change will be measured between the different tasks (baseline, stroop word-color interference task, rest, social stress recall task and rest) in both adolescent groups (with and without autism spectrum disorder).

Adolescents with autism spectrum disorder: At all assessment points: before supervised intervention (baseline) + after 5 weeks supervised intervention + after 5 weeks follow-up; TD adolescents: baseline only

Physiological assessment of skin conductance by placing one sensor on the palmar side of the index finger and 1 on the palmar side of the ring finger. The unit measure will be expressed in Micro Siemen. The sensors and biofeedback device from 'nexus-10' will be used in combination with 'biotrace+' software. This measurement will be repeated at multiple assessment points (only for adolescents with autism spectrum disorder). In addition, during one assessment, change will be measured between the different tasks (baseline, stroop word-color interference task, rest, social stress recall task and rest) in both adolescent groups (with and without autism spectrum disorder).

Adolescents with autism spectrum disorder: At all assessment points: before supervised intervention (baseline) + after 5 weeks supervised intervention + after 5 weeks follow-up; TD adolescents: baseline only

Physiological assessment of the temperature of the fingertip by attaching a temperature sensor on the palmar side of the middle finger. The unit measure will be expressed in Degrees Celsius.The sensors and biofeedback device from 'nexus-10' will be used in combination with 'biotrace+' software. This measurement will be repeated at multiple assessment points (only for adolescents with autism spectrum disorder). In addition, during one assessment, change will be measured between the different tasks (baseline, stroop word-color interference task, rest, social stress recall task and rest) in both adolescent groups (with and without autism spectrum disorder).

Adolescents with autism spectrum disorder: At all assessment points: before supervised intervention (baseline) + after 5 weeks supervised intervention + after 5 weeks follow-up; TD adolescents: baseline only

Endogenic measurement of cortisol level in saliva samples (3 times during assessment procedure for both adolescent groups, with and without autism spectrum disorder). These 3 measurements will be repeated at multiple assessment points (only for adolescents with autism spectrum disorder).

Adolescents with autism spectrum disorder: At all assessment points: before supervised intervention (baseline) + after 5 weeks supervised intervention + after 5 weeks follow-up; TD adolescents: baseline only

Identifies the extent of autistic traits shown by a person of normal intelligence based on parent-report. The scale's total score can vary between 0 and 50 in which a higher score means a higher presence of symptoms related to the autism spectrum. A cut-off of 32 is used for the in- or exclusion of typically developing adolescents. This measure will only be used once in both adolescent groups (with and without autism spectrum disorder).

At baseline assessment (typically developing adolescents + adolescents with autism spectrum disorder)

A quantitative assessment of the severity of autism traits during the past six months in children based on parent-report. This questionnaire identifies social impairment associated with ASD and quantifies its severity by using 5 subscales (Social Awareness, Social Cognition, Social Communication, Social Motivation, and Restricted Interests and Repetitive Behavior). The total score can vary between 65 and 260. These scores are related to t-values from the questionnaires manual. Higher scores are related to a higher level of impairment. Change in score for this outcome measure is only measured in adolescents with autism spectrum disorder.

At baseline assessment (typically developing adolescents + adolescents with autism spectrum disorder) + after follow-up phase at the end of the trial (adolescents with autism spectrum disorder)

Scores the number and intensity of repetitive and restrictive behaviors of the past month in children based on parent-report. The scale's maximum score varies between 0 and 129 with higher scores relating to a higher presence of repetitive and restrictive behaviors. These behaviors are additionally categorised in 6 categories: Stereotyped Behavior; Self-injurious Behavior; Compulsive Behavior; Ritualistic Behavior; Sameness Behavior. Parent are also asked to rate the impact of these behaviors on daily functioning on a scale from 0 (meaning no interference with daily life) and 100 (extremely problematic related to functioning in daily life). Change in score for this outcome measure is only measured in adolescents with autism spectrum disorder.

At baseline assessment (typically developing adolescents + adolescents with autism spectrum disorder) + after follow-up phase at the end of the trial (adolescents with autism spectrum disorder)

A brief behavioral screen in which parents need to rate the presence of certain behaviors in their child during the past six months or during the past academic year. There are 5 categories of behavioral problems: emotional problems, conduct problems, hyperactivity, peer problems and prosocial behavior. The scale's total score varies between 0 and 40 with higher values indicating more behavioral problems and consists of the summation of all scales except the prosocial scale score. Change in score for this outcome measure is only measured in adolescents with autism spectrum disorder.

At baseline assessment (typically developing adolescents + adolescents with autism spectrum disorder) + after follow-up phase at the end of the trial (adolescents with autism spectrum disorder)

A brief behavioral screen to rate the presence of certain behaviors during the past six months, based on adolescent self-report. There are 5 categories of behavioral problems: emotional problems, conduct problems, hyperactivity, peer problems and prosocial behavior. The scale's total score varies between 0 and 40 with higher values indicating more behavioral problems and consists of the summation of all scales except the prosocial scale score. Change in score for this outcome measure is only measured in adolescents with autism spectrum disorder.

At baseline assessment (typically developing adolescents + adolescents with autism spectrum disorder) + after follow-up phase at the end of the trial (adolescents with autism spectrum disorder)

Measures the degree to which situations in the adolescent's life are appraised as stressful, that is, how unpredictable, uncontrollable, and overloaded respondents find their lives. This measure is based on adolescent self-report and items are adapted according to the specific problems adolescents can encounter in their lives such as school-related stress. The scale's total score varies between 0 and 40 in which higher scores relate to higher perceived stress levels. Change in score for this outcome measure is only measured in adolescents with autism spectrum disorder.

Adolescents with autism spectrum disorder: At all assessment points: before supervised intervention (baseline) + after 5 weeks supervised intervention + after 5 weeks follow-up; TD adolescents: baseline only

Measures the presence of symptoms of depression, anxiety and stress based on adolescent self-report. The scale's total score varies between 0 and 63 in which higher scores relate to a higher presence of symptoms. Separate scale scores are calculated for symptoms related to depression, anxiety and stress. Change in score for this outcome measure is only measured in adolescents with autism spectrum disorder.

Adolescents with autism spectrum disorder: At all assessment points: before supervised intervention (baseline) + after 5 weeks supervised intervention + after 5 weeks follow-up; TD adolescents: baseline only

Measures the level of physical activity of the last week based on adolescent self-report. This measures consists of two short questions related to (1) the amount of days the adolescent was physically active and (2) the amount of minutes (on average) this psychical activity was endured. Change in score for this outcome measure is only measured in adolescents with autism spectrum disorder.

Adolescents with autism spectrum disorder: At all assessment points: before supervised intervention (baseline) + after 5 weeks supervised intervention + after 5 weeks follow-up; TD adolescents: baseline only

Measures the self-reported impact of sensory hyper-responsivity of the adolescent on a Visual Analogue Scale going from 0 (meaning no perceived sensory hyper-responsivity) to 100 (meaning presence of severe sensory hyper-responsivity and negative impact on functioning). Change in score for this outcome measure is only measured in adolescents with autism spectrum disorder.

Adolescents with autism spectrum disorder: At all assessment points: before supervised intervention (baseline) + after 5 weeks supervised intervention + after 5 weeks follow-up; TD adolescents: baseline only

To rate the self-reported perceived stress-level of the adolescent on a Visual Analogue Scale per stressor during the stress-inducing test. The scale score can vary between 0 (meaning no stress) and 100 (meaning highest level of perceived stress possible). Change in score for this outcome measure is only measured in adolescents with autism spectrum disorder.

Adolescents with autism spectrum disorder: At all assessment points: before supervised intervention (baseline) + after 5 weeks supervised intervention + after 5 weeks follow-up; TD adolescents: baseline only

Inclusion Criteria: Written informed consent/assent Male/female 13-18 years old For adolescents with ASD only: Confirmed ASD-diagnosis (DSM-IV/V) Exclusion Criteria: Presence of intellectual disability (DSM-IV/V) Hearing- or vision impairment without appropriate correction Presence of congenital heart diseases, diagnosed cardiovascular abnormalities or somatic diseases Pregnancy Insufficient knowledge of Dutch language Participation in other Clinical Trial(s) For typically developing adolescents only: Presence of neurodevelopmental disorder or psychiatric disorder (DSM-IV/V) + Presence of a sibling with a neurodevelopmental disorder (DSM-IV/V)

Thoen A, Steyaert J, Alaerts K, Van Damme T. Evaluating the potential of respiratory-sinus-arrhythmia biofeedback for reducing physiological stress in adolescents with autism: study protocol for a randomized controlled trial. Trials. 2021 Oct 21;22(1):730. doi: 10.1186/s13063-021-05709-4.