Effects of a 6-week, whole-body vibration strength-training on depression symptoms, endocrinological and neurobiological parameters in adolescent inpatients experiencing a major depressive episode (the “Balancing Vibrations Study”): study protocol for a randomized placebo-controlled trial

Participants are recruited from the adolescent psychiatric wards of the Department of Child and Adolescent Psychiatry and Psychotherapy at the University Hospital of Cologne, Germany. On admission to the inpatient treatment, adolescents and their parents (or other caregivers) are contacted by the study coordinator and informed about the program. If interest in participation is expressed, patients will be screened for eligibility. Inclusion and exclusion criteria of the study are listed in Table 2. Patients are included in the study only after they and their parents (or other caregivers) give written informed consent. The participants are included chronologically.

Eligible patients are randomized to one of two treatments (see “Groups and treatments” section below). A stratified block randomization with permuted block length is conducted. The stratification factor is patients’ gender (male vs. female). Randomization of treatment groups is implemented based on sealed, opaque envelopes by the Institute of Medical Statistics and Computational Biology of the University of Cologne.

The present study is a double-blinded trial. Prior to participation, patients will be informed that two physical activity interventions (whole-body vibration strength-training/myofascial release training) will be compared regarding their effect on severity of depression symptoms in adolescent inpatients. Accordingly, patients do not know the specific hypotheses of the trial. They will not know if they are allocated to the experimental or to the control group and, therefore, can be considered blinded. Moreover, experimental testing procedures will be carried out by specially trained experimental staff that will also be blinded regarding the allocation of participants to the specific experimental group.

WBV-training and PCG-intervention are conducted at the Department of Child and Adolescent Psychiatry and Psychotherapy at the University Hospital of Cologne. The participants of both intervention groups exercise four times per week for 6 weeks during their inpatient stay. Each session lasts approximately 30 min. The intervention sessions are carried out individually or in small groups with up to four participants. Training sessions are always instructed by specially trained study staff. In the following, WBV-training and PCG intervention will be described in detail:

Participants perform static and dynamic exercises on the Galileo® Whole Body Vibration Plate Med M (Novotec Medical GmbH, Pforzheim, Germany). The training sessions consist of six standardized exercises: squats, lunges, upper-body rotation against resistance, butterfly reverse against resistance, standing in partial squatting position, rocking the feet between the ball and heel. Resistance is applied using a latex resistance band (TheraBand, Akron, OH, USA). Each exercise is performed for 2 min. at 24 Hz and standing with feet parallel on the vibration plate, feet hip width apart. Between each exercise, participants stay active conducting low-intense shoulder and hip mobilization exercises for 2 min. After completion of four training sessions, vibration frequency is increased to 26 Hz. After completion of 12 trainings, the time of each exercise is increased to 3 min with 3-min. active rest intervals between the exercises.

Patients allocated to the PCG-training receive a supervised myofascial release training program. The training sessions consist of seven standardized exercises using a foam roll (Blackroll, Bottighofen, Switzerland): self-massage of the soles of the feet (one leg stand), of the calves (lying in dorsal position), of the hamstrings (lying in dorsal position), of the thighs (manually while lying in dorsal position), of the neck (lying in dorsal position), of the lower and upper back (leaning backwards against a wall in standing position), as well as self-massage of the upper arm and shoulder (leaning sideways against a wall in standing position). Each exercise is repeated 10 times on each side of the body. Between each exercise, participants do low-intensity shoulder and hip mobilization exercises for 2 min. The PCG-intervention was designed to provide an amount of psychosocial stimulation comparable to the WBV-training. All myofascial release exercises are instructed and patients are corrected, if necessary. However, unlike the WBV-training, myofascial release training sessions applied to the PCG hardly induce any muscular effort and cardiovascular stimulation.

Contraindications for participation in a training session in the WBV-training and in the PCG-intervention are acute suicidality, injuries, infections, or fever. To increase adherence to the training, participants are picked up from the ward by study staff. Treatment participation and comments on each session will be recorded.

The WBV-training and PCG-intervention are conducted simultaneously with the patients’ TAU at the Department of Child and Adolescent Psychiatry and Psychotherapy at the University Hospital of Cologne. TAU is carried out by a multidisciplinary team, consisting of psychiatrists, psychotherapists, music therapists, art therapists, pedagogs and social workers. Patients receive psychotherapy in the form of individual sessions one to two times per week. Families are closely involved in the patients’ therapy process. Every 4 weeks, systemic family therapy sessions are conducted. In addition, patients participate in individual and group music-therapy and art-therapy sessions. Counseling by the social services is conducted once or twice during inpatient stay.

Data are collected during the process of enrollment (− t₀), at baseline (t₀), after completion of the 6-week intervention (t₁), 8 weeks after completion of the 6-week intervention (follow-up I, t₂), as well as 20 weeks after completion of the 6-week intervention (follow-up II, t₃). At –t₀, patients’ demographic (age, sex, education, socioeconomic status) and anthropometric data (weight, height, Body Mass Index (BMI)), as well as past medical history and current medication, are collected. To determine patients’ eligibility for study participation before allocation to treatment groups, the primary outcome measure of the planned study (the Children’s Depression Rating Scale-Revised (CDRS-R)) is also assessed at − t₀. A raw-score of 40 or more points in the CDRS-R must be present to take part in the study [43] (compare Table 2). Baseline measurement takes place 1–3 days after enrollment. At t₀, all secondary outcome measures of the study (see “Secondary endpoints” section below), as well as potential confounding factors (see “Potential confounding factors” section below), are protocolled. At t₁, the primary outcome measure and all secondary outcome measures are assessed. The t₁ measurement takes place 1–3 days after completion of the 6-week exercise-treatment phase. At t₂, the primary outcome measure and all secondary outcome measures of the study, except cortisol awakening response, are assessed. Finally, at t₃, only the CDRS-R and the Beck’s Depression Inventory -second edition (BDI-II) are applied. Time accuracy of both follow-up intervals lies within a range of ± 3 days. The Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) study schedule of enrollment, interventions and assessments is provided in Fig. 2. The SPIRIT Checklist [44] is provided as Additional file 2 to this publication.

The primary endpoint of the present study is patients’ change of severity of depression symptoms from baseline until the end of the 6-week intervention. Patients’ severity of depression symptoms is measured by the CDRS-R. In the context of clinical trials, the CDRS-R is one of the most frequently used clinician ratings for the assessment of depressive symptomatology in adolescents [45]. The CDRS-R is a semi-structured interview consisting of 17 items that are rated between 1 (“no difficulties”) and 5 or 7 points (“clinically significant difficulties”). Usually, a raw-score of 40 points is used as a cut-off for depressive symptomatology and higher raw-scores are interpreted as more severe depression symptoms [43]. For CDRS-R application in adolescents, excellent internal consistency and good convergent validity have been reported [45].

The following secondary endpoints are assessed in the present study:

HPA axis activity is assessed by measuring patients’ cortisol awakening response. Saliva samples are collected after forced awakening at 7.00 and at 7.30 a.m. on two consecutive weekdays at t₀ and at t₁. Patients are asked to remain in bed without going back to sleep and to refrain from drinking and eating during the 30-min sampling period. In case of non-adherence (i.e., awakening before forced waking, rising before the second sample is collected, eating, or drinking), the procedure is repeated on the following day. Saliva is collected using Salivette collection devices (Sarstedt, Nümbrecht, Germany). Saliva samples are frozen and stored at − 20 °C until analysis. After thawing, salivettes are centrifuged at 3000 rpm for 5 min. Salivary cortisol concentrations are measured using commercially available chemiluminescence immunoassay with high sensitivity (IBL International, Hamburg, Germany). The intra- and interassay coefficients for cortisol are required to be below 8%. To measure HPA axis activity, mean salivary cortisol concentration and the increase between baseline and the 30-min samples are determined.

Drawing of 8.5 ml venous blood from sedentary patients is performed at t₀, t₁ and t₂. Samples for BDNF, cytokine (insulin-like growth factor-1, tumor necrosis factor-alpha, interleukin-6) and CRP analysis will be taken from a peripheral vein of the arm and collected in serum gel tubes (S-Monovette® Serum, Sarstedt, Nümbrecht, Germany). After complete coagulation at 4 °C, samples will be centrifuged for 10 min. at 1000 g and 4 °C. The serum will be prepared and aliquoted within 24 h after blood drawing. Duplicate samples will be stored at − 80 °C until analyzed when samples of all included participants are collected. Serum levels of BDNF and cytokines are assessed at t₀, t₁ and t₂. BDNF and cytokines will be measured in duplicates in a multiplex analyzer (Bio-Plex 200®, Bio-Rad Laboratories) according to the manufacturer’s instructions. Sera will be thawed, centrifuged for 10 min. at 10,000 g and 4 °C and diluted in sample diluent (1:4). By using the median of the fluorescence intensity and the standard curve, the absolute concentration of BDNF and the cytokines (pg/ml) will be calculated (Bio-Plex Manager 6.1, Bio-Rad Laboratories). For determination of CRP, the blood will be drawn as described above in lithium-heparin tubes (S-Monovette®, lithium-heparin, Sarstedt, Nümbrecht, Germany). It will be centrifuged for 10 min. at 4000 g and 21 °C. Plasma will be aliquoted within 3 h after blood drawing and used fresh. The CRP level will be determined via latex agglutination assay according to the manufacturer’s instructions (C-Reactive Protein Gen.3, cobas®, Roche Diagnostics, Basel, Switzerland). Briefly, plasma will be diluted 1:100 and added on a slide, which is pre-coated with antibodies to monoclonal anti-human CRP and latex reagent. After 2 min incubation, clear agglutination will be observed on the slide and it will be examined turbidimetrically using the analytic system cobas® C702 (Roche Diagnostics). CRP values below 3 mg/l are considered clinically irrelevant and will be adjusted to 0 mg/l.

For the self-rating procedure, the German version of the BDI-II [46] is applied. The BDI-II was developed as an indicator of the presence and severity of depression in psychiatric patients from 13 years of age onwards. Twenty-one symptoms are rated by the patient from 0 to 3, with higher scores indicating more severe occurrence of the particular symptom. Accordingly, the sum score ranges from 0 to 63, with higher scores indicating more severe depressive symptomatology. The BDI-II is sensitive for changes in severity of depressive symptomatology. Excellent indices of reliability, as well as good discriminant and convergent validity, have been reported in adolescent psychiatric inpatients [47].

CDRS-R is additionally applied 8 and 20 weeks after completion of the treatment to assess sustainability of potential effects of WBV-training compared to placebo.

The “Balancing Vibrations Study” also comprises magnetic resonance imaging, as well as combined transcranial magnetic stimulation and electroencephalographic recordings. For clarity, the more complex protocols and hypotheses, regarding these neuroimaging/neurophysiological parameters, will be illustrated separately and reported elsewhere.

Several additional factors, potentially impacting the primary outcome of the study (severity of depression symptoms), are assessed at t₀. To confirm depression diagnosis and to capture current psychiatric comorbidities, as well as patients’ history of mental disorders, the “Schedule for Affective Disorders and Schizophrenia – Present and Lifetime Version adapted for school-aged children” [48] and the “Diagnostic System for Mental Disorders in Childhood and Adolescence” [49] are applied. For assessment of emotional and/or behavioral problems, the “Child Behavior Checklist/4–18,” as well as the “Youth Self-Report/11–18” of the Achenbach System of Empirically Based Assessment, are used [50]. The patients’ socioeconomic status is assessed with a modified questionnaire developed by Trinkl and colleagues [51].

Adverse events and serious adverse events, which are related to the interventions, are assessed by the clinicians. During our pilot study, neither adverse events nor severe adverse events occurred [31]. Compliance to WBV-training and PCG-intervention is protocolled by scientific staff. It was pre-defined that patients need to participate in 20 sessions for complete compliance. This was shown feasible in our pilot study [31].

The Department of Child and Adolescent Psychiatry and Psychotherapy at the University Hospital of Cologne has insurance to cover for non-negligent harm associated with the protocol. This will include cover for additional health care, compensation, or damages. Incidences judged to arise from negligence (including those due to major protocol violations) will not be covered by study insurance policies.

All data will be entered electronically by two independent student assistants. Any deviations will be clarified and corrected. Original study forms will be kept on file. All study-related information is stored in a secure and accessible place and manner. All participant information will be stored and locked in file cabinets in areas with limited access. Participant files will be kept in storage for a period of 3 years after completion of the study. Blood samples will be disposed after analyses.

All laboratory samples, reports, data collection, process and administrative forms will be identified by a coded identification number, to maintain participant confidentiality. All records that contain names or other personal identifiers, such as locator forms and informed consent forms, will be stored separately from study records, which are also identified by code numbers. All local databases will be secured with password-protected access systems. Forms, lists, logbooks, appointment books and any other listings that link participant identification numbers to other identifying information will be stored in a separate, locked file in an area with limited access.

Sample size calculation was done for the effect of experimental treatment (WBV-training vs. PCG intervention) on the primary endpoint (severity of depression symptoms) measured by CDRS-R raw-scores in an analysis of covariance (ANCOVA) model using G*Power 3.1.9.2 software [52]. It was hypothesized that changes in the CDRS-R raw-scores from baseline to completion of the 6-week intervention differ significantly between the WBV-training group and the PCG-training group. As clinically relevant effect size, a medium effect of d = .5 was defined. Alpha was set at 5%. Test power (1 − β) was set at 80%. Participants’ CDRS-R scores at baseline will be included in the model as covariate. Required sample size in such an ANCOVA model can be calculated as (1 − ρ²) × n, with ρ representing the correlation between participants’ baseline and post-treatment outcome scores, and n representing the sample size that would have been required if a t test of post-treatment outcome scores was applied [53]. Based on the results of our pilot study, we estimated the correlation between patients’ severity of depression symptoms at baseline and at post treatment with ρ = .6. Under the presuppositions made, sample size calculation showed that 41 patients in each group (N = 82) would be required. We accounted for a 15% dropout rate, leading to a total sample size of N = 94 patients.

An intention-to-treat analysis will be performed. We will also assess the effect of the complete treatment in a per-protocol analysis. To determine the effects of between-subjects factor “treatment,” within-subjects factor “measurement time point” and their interaction on primary and secondary outcomes, separate 2 × 3 (hypothesis A, E, F)/2 × 2 (hypothesis C, D) mixed ANCOVAs will be conducted. For secondary outcome “HPA axis activity” (hypothesis B) separate 1-factor ANCOVA will be conducted. As covariate, patients’ baseline scores of each specific parameter will be used. A significant main effect of within-subjects factor measurement time point will be further investigated through post hoc pairwise comparisons. Simple effects analyses will be conducted to determine potential group differences at each time point. Due to the exploratory character of the secondary hypotheses, alpha error adjustment will not be applied.