Non-immersive virtual reality for cognitive rehabilitation of individuals with severe acquired brain injury (VR-sABI): study protocol for a multicentric randomized controlled trial

Abstract

Background

After a severe acquired brain injury (sABI), individuals might experience mild-to-severe cognitive impairments. In these patients, cognitive rehabilitation is provided as early as possible, to take advantage of and guide brain plasticity. According to a restorative approach, traditional cognitive training (TCT) usually involves repeated paper-and-pencil exercises of increasing difficulty and targeting specific cognitive domains. Recently, some evidence supported the use of virtual reality (VR) for ABI rehabilitation, particularly focusing on stroke cases. Nonetheless, there is limited evidence in sABI and in different etiologies. The present multicenter randomized controlled trial aims at exploring the effectiveness, in terms of clinical-functional, neurophysiological, and biomarker changes, of a cognitive rehabilitation focused on executive functions performed by a non-immersive VR device, compared to TCT, in a cohort of patients with sABI.

Methods

According to an a priori power analysis, 28 adult patients with sABI will be enrolled by 5 Italian neurorehabilitation units within the Fit for Medical Robotics (https://www.fit4medrob.it/) Consortium. Patients will be randomly assigned to receive either 30 min of VR or TCT sessions each day throughout the trial. At study entry (T0), patients will undergo clinical-functional evaluation, neurophysiological assessments, and serum blood sampling. Thereafter, VR or TCT will be provided daily, five times per week, for 5 weeks (25 total treatment sessions). Clinical-functional and neurophysiological assessments will be repeated at the end of the treatment (T1). A follow-up evaluation will be performed after 1 month from T1 (T2). Statistical analyses will be conducted blindly, according to the intention-to-treat principle.

Discussion

VR has recently been gaining popularity as a cognitive rehabilitation tool. Notwithstanding substantial evidence supporting the use of VR for ABI rehabilitation, there is limited evidence in sABI from traumatic or anoxic etiology. This multicenter pragmatic trial will provide new insights into the effectiveness of VR in sABI cognitive rehabilitation.

Trial registration

ClinicalTrials.gov NCT06474871. Registered on December 2, 2024.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13063-025-09128-7.

Administrative information

Note: the numbers in curly brackets in this protocol refer to SPIRIT checklist item numbers. The order of the items has been modified to group similar items (see http://www.equator-network.org/reporting-guidelines/spirit-2013-statement-defining-standard-protocol-items-for-clinical-trials/).

Title {1}	Non-immersive Virtual Reality for cognitive rehabilitation of individuals with severe Acquired Brain Injury (VR-sABI): study protocol for a multicentric randomized controlled trial
Trial registration {2a and 2b}.	ClinicalTrials.gov NCT06474871. Registered on 2nd Decemberh 2024
Protocol version {3}	30th August 2024, version 3
Funding {4}	This work was supported by the Italian Ministry of Research, under the complementary actions to the NRRP “Fit4MedRob - Fit for Medical Robotics” Grant (# PNC0000007).
Author details {5a}	1. IRCCS Fondazione Don Gnocchi ONLUS, Florence, Italy
Name and contact information for the trial sponsor {5b}	This trial is not funded by any private sponsor.
Role of sponsor {5c}	This trial is not funded by any private sponsor.

Introduction

Background and rationale {6a}

After an acquired brain injury (ABI), individuals might experience mild-to-severe cognitive impairments, in addition to motor disorders. The cognitive impairments can make motor rehabilitation programs difficult and negatively impact activities of daily living [1]. In these patients, cognitive rehabilitation should be provided as early as possible, once the acute phase of brain injury management is resolved, taking advantage of and guiding brain plasticity [2]. According to a restorative approach, traditional cognitive training (TCT) usually involves repeated paper-and-pencil exercises of increasing difficulty and targeting specific cognitive domains, assuming that impaired cognitive functions will respond to cognitive exercise in the same manner as muscles respond to physical exercise, and that repetitive training in a cognitive task may generalize to improved performance on other tasks within the same cognitive domain [3].

In the last decades, virtual reality (VR) has been used for rehabilitative purposes, especially in stroke patients [4], as VR encompasses enhanced ecological validity, the ability to control assessment and treatment standardization, and to manage task complexity. VR provides relatively naturalistic virtual environments for repetitive practice of functional tasks, aiding in generalizing targeted skills such as activities of daily living [5]. Furthermore, VR can boost patient motivation and active participation through visual and auditory feedback, vital for neurorehabilitation [6]. Additionally, VR tools offer the flexibility to adapt exercises to the patient’s abilities and monitor their performance [7]. VR encompasses computer-generated digital environments that can be experienced as if they were real [8]. VR systems are generally categorized as immersive, semi-immersive, or non-immersive. Immersive VR systems provide a changing field of view through head-mounted display (viewers) and use hardware such as head trackers, hand wearable sensors, and body motion sensors for tracking the user’s movements. All these devices allow for complete immersion of the user in a virtual environment that mimics a real scenario (e.g., cooking something or moving around a bedroom). Semi-immersive VR systems use projection-based systems, and non-immersive VR systems include basic computer screens and video games for providing real-based context, but there are no physical sensations to enhance the experience [9].

VR is recently utilized in rehabilitating functional impairments in patients with ABI, customizing cognitive-motor tasks to patients’ abilities, which enhances engagement and compliance [10].

Executive dysfunction is a common cognitive impairment following traumatic brain injury, affecting around 57% of moderate-to-severe patients [11]. Executive functions (i.e., response inhibition, cognitive flexibility, and working memory) enable an individual to independently perform complex and goal-directed activities, including problem solving and inhibition of irrelevant processing [12]. Impairment of executive functions is associated with poor functional achievement and lower ability to perform daily living tasks [1].

A recent study by De Luca et al. [7] investigated the effects of non-immersive VR-based training vs a TCT (e.g., using paper and pencil) to enhance executive functions in patients with moderate to severe ABI (sABI) with traumatic etiology. The study indicated improvement in global cognitive and executive functions for all patients, with those undergoing VR treatment showing better outcomes [7]. However, the overall existing evidence on VR-based rehabilitation treatment shows several pitfalls and biases, including heterogeneous populations and outcomes, small sample sizes, and a lack of randomized controlled trials, which may impact the evidence level and result generalizability [13, 14].

The present multicenter randomized controlled trial aims at overcoming the limits of the previous literature and exploring the effectiveness of a cognitive rehabilitation by means of VR device in a cohort of patients with sABI.

Objectives {7}

The primary objective of the VR-sABI study is investigating the effectiveness of a VR-based training, compared to a TCT in promoting improvement in executive functions (i.e., response inhibition, cognitive flexibility, and working memory).

As secondary objectives, the study aims to compare the effectiveness of a VR-based training, compared to a TCT, in (i) maintaining the possible improvement in executive functions at 1 month after treatment and (ii) reducing functional disability. Furthermore, this study aims at investigating the usability and compliance of patients and of rehabilitation therapists in using VR-based devices for cognitive rehabilitation.

Exploratory objectives are to compare the effectiveness of a VR-based training, compared to a TCT, in (i) promoting the recovery of global cognitive functioning; (ii) reducing behavioral disorders; (iii) improving neurophysiological indices of brain functionality; and (iv) improving brain plasticity.

Trial design {8}

The VR-sABI study is a randomized 1:1, controlled, multicenter, exploratory, single-blind trial, with two parallel groups.

Methods: participants, interventions, and outcomes

Study setting {9}

This trial will be delivered through five Italian post-acute neurorehabilitation units within the Fit for Medical Robotics (https://www.fit4medrob.it/) Consortium, with expertise in the care of patients with sABI. The trial centers include (i) Polo Specialistico Riabilitativo, Fondazione Don Carlo Gnocchi ONLUS (Sant’Angelo dei Lombardi, Italy; coordinating center); (ii) IRCCS SM Nascente Fondazione Don Carlo Gnocchi ONLUS (Milan, Italy); (iii) Istituti Clinici Scientifici Maugeri IRCCS (Bari, Italy); (iv) Istituti Clinici Scientifici Maugeri IRCCS (Telese Terme, Italy); and (v) IRCCS Ospedale Policlinico San Martino (Genoa, Italy). Two further Italian centers will participate in the data analysis only: (i) SM della Provvidenza Fondazione Don Carlo Gnocchi ONLUS (Rome, Italy); (ii) IRCCS Fondazione Don Carlo Gnocchi ONLUS (Florence, Italy).

Eligibility criteria {10}

Patients with sABI consecutively admitted to the participating centers will be screened and enrolled if they meet the following inclusion criteria:

• Age between 18 and 75 years

• Traumatic, vascular, anoxic, or mixed sABI (i.e., Glasgow Coma Scale score ≤ 8 for at least 24 h after the brain injury)

• Level of Cognitive Functioning (LCF) ≥ 4 at study entry

• Stable clinical diagnosis (assessed by at least 4 evaluations by means of the LCF repeated in a time window of 1 week)

• Time post-injury between 28 days and 6 months at the time of the enrollment

• Ability to complete the pre-test A of the Trail Making Test (TMT) in its Italian version [15]

• Deficit in executive functions represented by a pathological performance on the B-A score of the TMT, according to normative data corrected for age and education [16]

• Written informed consent by the patient and/or by patient’s legal representative/primary caregiver

Exclusion criteria will be:

• Severe medical conditions hampering the participation in the rehabilitation sessions (e.g., severe paresis, praxis or language deficit, severe cardiac-vascular instability, photosensitive or reflex epilepsy, severe visual or auditory impairment)

• Medical conditions influencing clinical diagnosis and electroencephalographic (EEG) activity (e.g., severe hepatic or renal insufficiency, subcontinuous or abundant epileptiform activity detected on a standard EEG acquisition)

• Previous neurodegenerative or acquired neurological disorders affecting cognitive performances (e.g., dementia)

• Pregnancy

Who will take informed consent? {26a}

Delegated clinical research staff at sites who are trained on the study protocol and procedures will be able to receive informed consent from potential participants. Purposes, procedures, and time points of the study will be clearly explained after a semi-formalized interview, whose structure is shared across centers, to either the eligible patients (if the LCF is between 7 and 8) or the patient’s legal representative/primary caregiver (if the assessment of cognitive functioning suggests that the patient may not understand all the information, e.g., for patients with LCFs between 4 and 6) for their written consent to participation in the study. The original Italian version of the informed consent forms for both the patients and for their legal representatives/primary caregivers is reported in Additional file 1.

Additional consent provisions for collection and use of participant data and biological specimens {26b}

Given the patient’s legal representative/primary caregiver consent, data will be pseudonymized and stored in paper format in an archive accessible only to the principal investigator of each center and then entered in pseudonymized form into a computerized database using the REDCap (Research Electronic Data Capture) application located on the Fondazione Don Carlo Gnocchi ONLUS server.

The data collection will be in pseudonymized form. Consequently, once the database is completed, this data will be definitively anonymized by proceeding to the destruction of the re-identification keys and the eventual removal of potentially identifying combinations as indicated by recital 26 of the GDPR.

In accordance with the principle of minimum data retention, data will be retained for a period of 10 years after the conclusion of the study. Any biological sample will be kept at the laboratory of the center SM della Provvidenza Fondazione Don Carlo Gnocchi ONLUS (Rome, Italy) for the purpose of the analyses provided for in the protocol and destroyed after such analyses. In the event that a Biobank is set up at the Fondazione Don Carlo Gnocchi ONLUS, these samples will not be destroyed but transferred to it.

Interventions

Explanation for the choice of comparators {6b}

Enrolled patients will be assigned randomly to two parallel arms (VR or TCT) to receive 30 min of either VR or TCT in the same period of the day throughout the experiment. Twenty-five VR or TCT sessions will be applied, 5 per week for 5 weeks.

The comparator in this study is the TCT of executive functions that are routinely applied in clinical practice of cognitive rehabilitation. The TCT uses mostly paper-and-pencil tools and is based on a face-to-face approach between the therapist and the patient, in line with the most recent guidelines on cognitive rehabilitation of executive functions [17, 18]. The training of the executive abilities is carried out by working on cognitive tasks with the same aim and similar structure as the ones implemented in VR, but using non-digital materials and, when possible, supported by metacognitive training and group-based interventions [17, 18]. Therapists will adapt the TCT session based on the patient’s characteristics, implementing the type, number, and duration of tasks that best fit the patient’s abilities, and according to a criterion of incremental difficulty and complexity of the cognitive task. References for the TCT tasks will be Powell [19] and Iannizzi et al. [20].

Intervention description {11a}

The intervention in this study is non-immersive VR-based cognitive training focused on executive functions. The device used is the COMPACT VRRS ENGINE (Khymeia®, Italy). The device consists of a central processing unit, complete with a capacitive touch screen LCD monitor. This instrumentation, together with a complete VR visualization and interaction system presented to the patient and therapist through the monitor, enables the implementation of an extremely advanced learning and rehabilitation framework based on operant conditioning. The exercises are grouped by functional modules. The functional module used in the project is the cognitive module, with more than 100 rehabilitation activities for the main cognitive functions, customizable and adaptable to the patient’s needs and abilities. Indeed, therapists can select the number of tasks to be completed in a session; can select the tasks that best fit the needs and the residual abilities of the patient; can modify the task complexity by changing the number of stimuli presented, the speed of the stimuli, or the time available for the patient to respond; and can modify the number of trials to be completed within each task.

During a VR session, the patient sits in front of the device screen and actively interacts with it (either by touch screen function, by an eye-tracker, or by inertial sensors), under the supervision of the therapist. The device contains several different exercises, and for each of them, the therapist can modify the virtual task scenario, increasing/decreasing its difficulty, creating a stimulating learning context in which the exercises are neither too easy nor too difficult.

For each patient, the therapist will select the 6 tasks, out of 18 tasks training executive functions, best suited for the patient characteristics. For example, in the task “Rearranging images in series,” patients are presented with several (the exact number depends of task difficulty) pictures illustrating different actions that represent daily routines (e.g., waking up, have a breakfast, take a shower, go to work); pictures are presented in random order, and the patient’s task is to rearrange pictures by putting them in the correct order. The list and brief description of available tasks is reported in the Appendix.

Patients will be involved in each task for 4 consecutive minutes. A pause of 1 min is planned across each task. In order to tailor the tasks to patients’ residual abilities, the difficulty of the tasks will automatically increase after the correct completion of each trial within the task. See Fig. 1 for the sequence of events in each VR session.

Fig. 1.

Sequence of events in each VR session

Criteria for discontinuing or modifying allocated interventions {11b}

Protocol violations include medication change, infection, surgery, functional deterioration, 2 consecutive missed sessions or more than 5 missed sessions, and seizure. Subjects violating the protocol will be excluded from the study, and the last observation will be considered for statistical analysis, according to the intention-to-treat principle.

Strategies to improve adherence to interventions {11c}

Adherence of patients: in order to improve patients’ adherence to the trial, clear and detailed information about study procedure, interventions, and aims will be provided at patients’ enrollment (as reported in the informed consent).

Adherence of participating centers: results of periodic clinical and instrumental evaluations included in the protocol (neurophysiology and laboratory tests) will be provided to the principal investigators of participating centers during and at the end of the study. The coordinator of the multicenter study will monitor the progression of the study through periodic call meetings with all participating clinical centers and with the laboratory for the serum markers. The laboratory will monitor the data entry in the REDCap database for verifying the consistency and completeness of the data.

Relevant concomitant care permitted or prohibited during the trial {11d}

VR-sABI is a pragmatic trial using cognitive training that is usually included in the routine care of patients with sABI. In this trial, usual care (e.g., motor rehabilitation) can continue without restriction. Indeed, regardless of assignment to the VR or TCT group, all patients will carry out the all-inclusive rehabilitation program as designed by the participating center for the entire study period (from T0 to T2). This program consists of active limb mobilization, training in different cognitive domains, occupational therapy, speech, and swallowing therapy based on the functional condition of the patient.

Clinicians can also prescribe other medication as necessary but will be asked to exercise caution in case they plan to prescribe drugs that may interfere with cognitive functions. Information on other drug or psychological treatments will be collected at baseline and at all follow-up time points. A list of proscribed medications can be found in the Appendix. In case of need to administer/change the proscribed drugs during the study, patients will be excluded from the study (drop-out) and the last assessment will be included in the analysis according to the intention to treat principle.

Provisions for post-trial care {30}

After the 5 weeks of treatment, all participants will be returned to standard care (i.e., TCT) for at least 1 month (i.e., until T2). After the study completion, the continuation of VR or TCT treatments is a responsibility of the participant’s clinician. This information will be discussed with the participant at enrollment and in the final study appointment.

Outcomes {12}

Primary, secondary, and exploratory outcomes are summarized in Table 1.

Table 1.

Summary of primary, secondary, and exploratory outcomes and measures

Outcome	Tool/method
Primary outcome:
1. Improvement in executive functions	• TMT B-A score, corrected for age and education, measured at T1
Secondary outcomes:
1. Maintaining an improvement in executive functions at 1 month after treatment	• TMT B-A score, corrected for age and education, measured at T2
2. Functional disability	• DRS and mBI total score, measured at T1 and T2
3. Usability and adherence of patients and of rehabilitation therapists	• SUS total score, measured at T1
Exploratory outcomes:
1. Global cognitive functioning	• LCF, GOAT, and OCS total score measured at T1 and T2
2. Behavioral disorders	• NPI total score measured at T1 and T2
3. Neurophysiological indices of brain functionality	• Qualitative (background activity, reactivity) and quantitative (power spectrum, microstates, connectivity) measures from a 30-min standard EEG recorded at T1 and T2
4. Blood biomarkers of brain plasticity	• BDNF, NF-L, GFAP from serum sample measured at T2
5. Adverse events	• AER

Abbreviations: AER adverse events report, BDNF brain-derived neurotrophic factor, DRS Disability Rating Scale, EEG electroencephalogram, GFAP glial fibrillary acid protein, GOAT Galveston Orientation and Amnesia Test, LCF Levels of Cognitive Functioning, mBI modified Barthel Index. NF-L neurofilament light chain, NPI Neuropsychiatric Inventory, OCS Oxford Cognitive Screening, SUS System Usability Scale, TMT Trail Making Test

The primary outcome is the improvement of executive functions, as measured by the B-A score of the TMT, according to normative data corrected for age and education [16], performed within 1 week after the completion of the 5-week trial (T1).

The secondary outcomes are the improvement of executive functions as measured by the B-A score of the TMT, according to normative data corrected for age and education [16], performed 1 month after the completion of the trial (T2); the improvement of functional disability as measured by the total score of the Disability Rating Scale (DRS) [21]; the improvement of performance in basic activities of daily living as measured by the total score of the modified Barthel Index (mBI) [22]; and the usability of VR or TCT by means of the total score of the System Usability Scale (SUS) [23].

Exploratory outcomes are the presence of neurobehavioral disturbances as measured by the total score of the Neuropsychiatric Inventory (NPI) [24]; the improvement of the level of cognitive functioning as measured by the Levels of Cognitive Functioning (LCF) [25, 26]; the improvement in the symptoms of post-traumatic amnesia, as measured by the total score of the Galveston Orientation and Amnesia Test (GOAT) [27]; the improvement of cognitive profile (including attention, praxis, memory, language, and visuo-spatial abilities) as measured by the total score of the Oxford Cognitive Screening (OCS) [28, 29]; the improvement of neurophysiological indices of brain functionality, measured by qualitative (background activity, reactivity) and quantitative (i.e., power spectrum, microstate and connectivity analysis) analysis of EEG activity; brain plasticity biomarkers (levels of serum biomarkers: brain-derived neurotrophic factor, BDNF; neurofilament light chain, NF-L; glial fibrillary acid protein, GFAP); and presence of adverse events (e.g., agitation, irritability).

Participant timeline {13}

The schedule of the study is reported in Table 2.

Table 2.

Schedule of enrollment, interventions, and assessments from the VR-sABI study. Each week includes 5 intervention sessions of 30-min duration provided on a daily basis

Time point/measure	Pre-randomization	Randomization	Post-randomization
	Enrollment	T0 (within 7 days prior to the start of intervention)	Week 1	Week 2	Week 3	Week 4	Week 5	T1 (within 7 days after the end of the intervention)	T2 (1 month after T1)
Informed consent	X
Eligibility screening	X
Allocation		X
Clinical and anamnestic data collection		X
Interventions:
VR			X	X	X	X	X
TCT			X	X	X	X	X
Outcomes:
TMT		X						X	X
DRS		X						X	X
mBI		X						X	X
SUS								X
NPI		X						X	X
LCF		X						X	X
GOAT		X						X	X
OCS		X						X	X
30-min standard EEG		X						X	X
Blood sampling		X							X
AER			X	X	X	X	X	X	X

Abbreviations: AER adverse events report, DRS Disability Rating Scale, EEG electroencephalogram, GOAT Galveston Orientation and Amnesia Test, LCF Levels of Cognitive Functioning, mBI modified Barthel Index, NPI Neuropsychiatric Inventory, OCS Oxford Cognitive Screening, SUS System Usability Scale, TCT traditional cognitive training, TMT Trail Making Test, VR virtual reality

At study entry (T0), each center will collect demographic and anamnestic data. The enrolled patients will undergo clinical-functional evaluation, neurophysiological assessments, and serum blood sampling at study entry (T0). Thereafter, interventions (duration: 30 min) will be provided daily, five times per week (excluding weekends), for 5 weeks (25 total treatment sessions). Further clinical-functional and neurophysiological assessments will be performed at the end of the fifth week of treatment, or at least within 7 days from the last treatment (T1). A follow-up evaluation will be performed after 1 month from T1. In total, each patient will be followed for up to 9 weeks (T0 to T2). An enrollment window of 12 months is expected, for a total study duration of 24 months.

Sample size {14}

The sample size calculation was based on data for the primary outcome from a previous study [7], which showed an improvement after 8 weeks in the B-A score of the TMT in a group of 10 patients with traumatic sABI undergoing VR-based training and in a group of 10 patients with traumatic sABI undergoing TCT (p < 0.02; effect size, Glass’s delta = 1.55). Considering a two-sided α of 0.05 and a power of 80%, at least 11 subjects per group plus 20% would be required to account for possible drop-outs, thus increasing to 14 subjects per group, for a total of 28 (about 6 patients per center).

Recruitment {15}

In order to reach the desired sample size, the participation in the trial will be proposed to the legal representative/primary caregiver of all patients with sABI consecutively admitted at the five Italian post-acute neurorehabilitation units with expertise in the care of patients with sABI: (i) Polo Specialistico Riabilitativo, Fondazione Don Carlo Gnocchi ONLUS (Sant’Angelo dei Lombardi, Italy); (ii) IRCCS SM Nascente Fondazione Don Carlo Gnocchi ONLUS (Milan, Italy); (iii) Istituti Clinici Scientifici Maugeri IRCCS (Bari, Italy); (iv) Istituti Clinici Scientifici Maugeri IRCCS (Telese Terme, Italy); (v) IRCCS Ospedale Policlinico San Martino (Genoa, Italy).

Assignment of interventions: allocation

Sequence generation {16a}

The random allocation sequence of eligible patients will be generated using REDCap. Patients will be randomized to one of two treatment groups in a 1:1 ratio to either VR (intervention arm) or TCT (control arm; see Fig. 2). The alphanumeric codes will be constructed according to the following order: numeric ID of the center responsible for the case; number of the recruited patient.

Fig. 2.

Study flowchart

Concealment mechanism {16b}

The web-based REDCap randomization system ensures allocation concealment because the randomized code is sequentially generated for each participating center and released only once the patient is enrolled.

Implementation {16c}

Each PI (or authorized delegate) of participating centers, blinded to the cognitive assessment, will enroll patients and will be able to access the patients’ allocation generated by REDCap and assign them to the interventions.

Assignment of interventions: blinding

Who will be blinded {17a}

Double-blinding cannot be applicable as neither participants nor cognitive rehabilitation therapists can be blinded due to the nature of the interventions. However, the outcome measures (clinical-functional, neurophysiological, biomarkers) assessment will be performed by examiners who will be blinded to the patient’s treatment group. The statistician in charge of the analysis will also remain blind.

Procedure for unblinding if needed {17b}

As stated above, participants and examiners who will administer the treatments will be always unblinded to the participant’s group, due to the nature of the interventions. Unblinding of the research team will be provided at the end of the data collection and analysis, for interpretative purposes.

Data collection and management

Plans for assessment and collection of outcomes {18a}

Prior to the study starting, the coordinating center presented, during several conference calls, the rationale, aims, and general methodology of the study to potential candidate neurorehabilitation centers within the Fit for Medical Robotics (https://www.fit4medrob.it/) Consortium. Five clinical centers with high expertise in the management of patients with sABI were involved in the study. In each participating center, a team of multidisciplinary investigators with expertise in using the selected clinical and neurophysiological tools was involved. The final version of standard operating procedures was shared among the skilled examiners from each center during a kick-off meeting, in order to ensure the consistency of (i) administration and scoring of the clinical scales; (ii) EEG acquisition and visual analysis of background activity and reactivity; (iii) acquisition and storage of the serum blood samples and transfer to the laboratory; and (iv) data entry procedures into the REDCap database.

After screening for inclusion/exclusion criteria, patients will undergo repeated (at least 4 times in a week) assessment by means of the LCF in order to confirm clinical diagnosis. Once clinical diagnosis is confirmed, the patient is enrolled in the trial and the following demographic and anamnestic variables are collected: admission date, enrollment date, birth date, age, sex, education, language, ethnicity, brain injury date, time post-injury, etiology, brain injury location (from available neuroimaging), clinical assessment date, presence of ophthalmoplegia, presence of visual deficit, presence of assistive devices for breathing, presence of tracheostomy tube. Within 1 week from enrollment, the enrolled patients will undergo clinical-functional evaluation, neurophysiological assessments, and serum blood sampling at study entry (T0).

The clinical-functional evaluation includes:

• The TMT [15, 16], a neuropsychological test assessing, in particular, visual exploration, selective attention, and set shifting. It is constituted of two parts, A and B. In part A, the patient is requested to connect with a line in progressive order 25 numbers printed in random locations on the paper, whereas in part B the patient is requested to connect alternatively 13 numbers and 12 letters in progressive order, similarly printed in random locations on the paper. The task ends when the patient connects in the correct way all numbers and letters. In the case of any error, the examiner stops the patient and helps her/him to find the right solution. The score is represented by the time the patient needs to complete the test. The score of the part B of the TMT is a good predictor of the return to work after severe traumatic brain injury [30, 31].

• The LCF [25, 26], an observer-rated scale (from coma to purposeful appropriate behavior) divided into eight levels that describe the patterns or stages of recovery typically seen after a brain injury.

• The GOAT [27], a structured interview measuring orientation to person, place, and time, and memory for events preceding and following the injury. It comprises 13 items. Incorrect response to each item is scored with error points. The total score is obtained by deducting the sum of the error points from 100.

• The OCS [28, 29], a first-line, domain-specific cognitive screening tool which can be administered at the bedside. For each domain evaluated, cut-off scores indicate a pathological performance compared to individuals matched for age and education.

• The NPI [24], an observer-rated scale for evaluation of psychiatric disorders, including ten behavioral and two neurovegetative areas. Each domain is scored on a 5-point Likert scale assessing frequency of symptoms and on a 3-point Likert scale assessing severity of symptoms. The score of each domain is computed by multiplying the frequency and the severity scores. A total score (maximum score = 144, minimum score = 0) can be computed by summing up the items scores, with the highest scores representing the higher frequency and severity of psychiatric symptoms.

• The DRS [21], an observer-rated, 30-point continuous scale that provides quantitative information to document the progress of patients with severe brain injury from coma to community reintegration. It evaluates 8 areas of functioning, organized in 4 categories: (1) consciousness (eye opening, verbal response, motor response); (2) cognitive ability (feeding, toileting, grooming); (3) dependence on others; (4) employability. Each area of functioning is rated on a scale of 0 to either 3, 4, or 5 (maximum score = 30—death, minimum score = 0—person without disability) with the highest scores representing the higher level of disability.

• The mBI [22], an observer-rated continuous scale for evaluation of ability to perform autonomously personal activities of daily living. It measures physical disability across 10 categories that are scored from 0 to either 5, 10, or 15 (maximum score = 100—independence, minimum score = 0—complete dependence) with the highest scores representing the higher level of independence.

The neurophysiological assessment includes:

• Two consecutive eyes-closed EEG recordings: (i) a first resting-state EEG with forced eye closing (duration: 15 min); (ii) a second reactivity EEG using the following randomized stimuli: eye opening and (forced) eye closing, proximal noxious stimulation (deep pressure applied to the trapezius muscle on each side), distal noxious stimulation (pressing fingernail beds on each hand), acoustic stimulation (hand clapping), acoustic stimulation (patient’s name), intermittent photic stimulation (IPS) with flashes at 1—2—8—10—15—18—20—25—40—50—60 Hz in 5-s trains presented through closed eyelids. Each stimulus is administered twice, with an inter-stimulus interval of at least 1 min (duration: 15–17 min). EEG will be recorded by means of 19 electrodes placed on the patients’ scalp according to the international 10–20 system (Fp1, Fp2, F7, F8, F3, F4, C3, C4, T3, T4, P3, P4, T5, T6, O1, O2, Fz, Cz, Pz) plus electrooculogram and electrocardiogram. Data will be sampled at a minimum of 512-Hz rate and with hardware LP-HP filters open; a notch filter will be used to eliminate frequencies around 50 Hz for online visualization. EEG recordings will take place with the patient sitting on her/his wheelchair, in morning time after customary nursing procedures and (at least) 10 h after administration of drugs acting on the central nervous system, such as myorelaxants and sedative drugs (e.g., benzodiazepines) to optimize vigilance. Qualitative (background activity, reactivity) [32, 33] and quantitative measures (i.e., power spectrum, microstate and connectivity analysis) [34] of EEG activity will be extracted.

The serum blood sampling includes:

• Two test tubes of blood per patient will be collected in order to analyze the following serum biomarkers: BDNF, NF-L, GFAP. Blood biomarkers have been correlated with cognitive performance and assist in prognosis in patients with sABI [35, 36]. BDNF is a neurotrophin involved in neurogenesis and synaptic plasticity [37]. NF-L, a marker of primary and secondary neurodegeneration in sABI, has been correlated with long-term axonal degeneration and poor outcome [38]. GFAP correlates with recovery of brain function [39].

Thereafter, interventions will be provided daily, five times per week, for 5 weeks. At the end of each week of treatment, any adverse event will be reported.

At the end of the fifth week of treatment or within 7 days from the last treatment (T1), the same clinical-functional and neurophysiological evaluations performed at T0 will be collected and, in addition, the SUS is administered to patients and to rehabilitation professionals. The SUS [23] is a simple, ten-item 5-point Likert scale giving a global view of subjective assessments of usability, including effectiveness, efficiency, and satisfaction of a device. Higher scores correspond to higher usability.

After 1 month from T1, a follow-up evaluation will be performed (T2), in which the clinical-functional and neurophysiological evaluations performed at T0 and a further serum blood sample are collected.

Plans to promote participant retention and complete follow-up {18b}

Eligible patients will be selected among all patients with sABI consecutively admitted at the five Italian post-acute neurorehabilitation units. According to the organization of the Italian National Health System, post-acute rehabilitation for patients with sABI involves an inpatient period of about 3 to 6 months. In this trial, efforts will be made to enroll patients as early as possible from the time of admission, as soon as the stability of clinical conditions allows, so that the average period of hospitalization can cover the entire duration of the study. Moreover, as specified below, all analyses in this trial will be conducted according to the intention-to-treat principle; then all enrolled patients will be included in the statistical analysis, regardless of protocol violations or discontinuation.

Data management {19}

The investigators in charge of data collection will also take care of the data entry. Patients’ data will be collected and entered in a dedicated paper-based case report form at each participating center. All paper-based copies of the data collected by participating centers will be safely stored at each center. Thereafter, patient data will be entered in a pseudonymized form into a centralized, password-protected, electronic database on REDCap, following as much as possible a forced-choice format. REDCap complies with the Health Insurance Portability and Accountability Act in compliance with security measures and will therefore only contain data that has already been pseudonymized. Indeed, the personal data of each patient will be de-identified and replaced by an alphanumeric ID code in an “association key” document accessible exclusively to the enrolling center in a password-protected Excel file and then entered into REDCap. Similarly, the patients’ EEG raw data will be transmitted in a pseudonymized European Data Format to the coordinating center via REDCap.

Each center will be granted to review only data of their own patients and will be blinded to data entered by other centers.

The completeness and quality of the data entry in the REDCap database will be checked regularly by the team of the coordinating center, who will have the possibility to review data from all participating centers. If any missing data or inconsistencies are detected by the coordinating center, the participating center in question will be encouraged to check it thoroughly. Patients’ data collected and rendered pseudonymized by the participating centers will be analyzed in aggregate by the coordinating center.

Confidentiality {27}

This trial collects personal data also belonging to the special categories referred to in Art. 9 of the European Data Protection Regulation 2016/679 (GDPR) and adheres to the principles and rules of the GDPR and TU 196/2003 in order to take into continuous and due consideration the risks from technological development.

Plans for collection, laboratory evaluation, and storage of biological specimens for genetic or molecular analysis in this trial/future use {33}

On admission (T0) and 1 month after the end of the treatment (T2), during the routine blood sampling, in the morning on an empty stomach, 2 blood test tubes for a total of 15 mL will be taken (biochemistry test tubes, yellow cap); the sample will then be processed, which consists of:

• i.Centrifugation of the sample at 3000 rpm for 15 min
• ii.Collection and division of the serum into aliquots of 500 µl
• iii.Labeling with reference ID code and storage of the sample aliquots at a controlled temperature (−20 °C) until dispatch
• iv.Shipment of the serum aliquots at controlled temperature (dry ice)

Sample aliquots will then be transferred to the biological research laboratory located at the SM della Provvidenza center of the Fondazione Don Carlo Gnocchi ONLUS (Rome, Italy) where serum levels of BDNF, NF-L, and GFAP will be analyzed (Ella Automated Immunoassay System, Biotechne).

The samples will be kept for the purpose of the analyses stipulated in the protocol and destroyed after these analyses. In the event that a Biobank is established at the Fondazione Don Carlo Gnocchi ONLUS, samples will not be destroyed but transferred to it.

Statistical methods

Statistical methods for primary and secondary outcomes {20a}

Frequencies (percentages) or mean ± standard deviation will be used for descriptive statistics regarding socio-demographic and clinical variables, as well as for the level of satisfaction with the experimental intervention and the occurrence of side effects. The statistical plan will include a 2 × 3 mixed ANOVA performed for each outcome with group (VR, TCT) as the between-subjects factor and time (T0, T1, T2) as the within-subjects factor. Post hoc comparisons corrected by the Bonferroni method will be performed. Missing data will be handled by pairwise deletion. The significance level will be set at 0.05.

Interim analyses {21b}

No interim analyses are planned.

Methods for additional analyses (e.g., subgroup analyses) {20b}

No subgroup analyses are planned.

Methods in analysis to handle protocol non-adherence and any statistical methods to handle missing data {20c}

All analyses will be conducted blind, according to the intention-to-treat principle. All randomized and followed subjects will be included in the statistical analysis, regardless of protocol violations or treatment compliance.

Plans to give access to the full protocol, participant-level data, and statistical code {31c}

Any information about the protocol is reported here and in the Appendix. Anonymized data can be requested from the corresponding author by bona fide researchers following the submission of a proposal of intended use and after their host institution has signed a data access agreement.

Oversight and monitoring

Composition of the coordinating center and trial steering committee {5d}

The trial steering committee of the project comprises Dr. Anna Estraneo (neurologist, principal investigator) and Dr. Alfonso Magliacano (neuropsychologist, co-PI, responsible for treatment protocol and co-responsible for clinical data collection). Both researchers contributed to the conception of the project and developed the design and methods of the protocol.

The coordinating center is the Research Unit of Fondazione Don Gnocchi at Sant’Angelo dei Lombardi and involves Dr. Maria Rosaria Fiorentino (geriatrician, co-responsible for clinical data collection), Dr. Giovanni Scarano and Dr. Antonia Monda (neurophysiopathology technicians, co-responsibles for EEG protocol). Four additional researchers from 3 institutes of Fondazione Don Gnocchi ONLUS were involved based on their specific expertise: Dr. Angela Comanducci (neurophysiologist, co-responsible for EEG quantitative analysis), Dr. Piergiuseppe Liuzzi (biomedical engineer, co-responsible for EEG quantitative analysis), Dr. Mariacristina Siotto (biologist, responsible for serum blood biomarkers collection and analysis), Dr. Andrea Mannini (biomedical engineer, responsible for data entry and management). Members of the steering committee and of the coordinating center met several times to define the trial protocol and, after the study started, monthly meetings were organized to supervise the flow of the events.

Composition of the data monitoring committee, its role and reporting structure {21a}

An independent data monitoring committee is not needed as this trial is not funded by any private organization.

Adverse event reporting and harms {22}

No contraindications for both intervention and comparator are reported. However, possible adverse events will be evaluated and reported during the study, and their causal relationship with the interventions will be assessed. Any adverse event could be reported in the REDCap database, where investigators could select pre-specified possible adverse events (i.e., dizziness, nausea, drowsiness, pain, epilepsy, psychomotor agitation, misperception, worsening of the clinical condition, discharge, exitus, patient refusal, consent withdrawal) or could specify any other adverse event.

If an adverse event is found, the researcher will evaluate the intensity and frequency of each episode and decide whether additional treatments are needed. These adverse events will be managed according to the 90/385/EEC and 93/42/EEC regulations. If the physician considers it appropriate, the patient will be withdrawn from the study.

Frequency and plans for auditing trial conduct {23}

Monthly call meetings between the steering committee, the coordinating center, and all participating centers will be planned for ensuring the appropriateness of the patients’ enrollment, of the informed consent collection, and of data privacy management and quality of data acquisition.

Plans for communicating important protocol amendments to relevant parties (e.g., trial participants, ethical committees) {25}

The trial has received approval by the National Ethics Committee of the Italian National Institute of Health. Any amendments which affect the safety (physical or mental integrity) of the participants, the scientific value of the study, or the conduct or management of the study will constitute a substantial amendment and a request to the National Ethics Committee for approval will be submitted. All trial sites and investigators will be notified of amendments and their date of implementation. All participants will provide consent using the procedures in the latest version of the protocol at the time of their enrollment. Important modifications made to the initial protocol will be registered in https://clinicaltrials.gov/.

Dissemination plans {31a}

Study results will be communicated to clinical stakeholders and disseminated in communications at scientific national and international conferences and congresses and through publications in peer-reviewed scientific journals. Patients and patients’ family representative associations will be involved in the dissemination of results. The final results will also be disseminated to the health professionals involved in the care of this type of patients through the internal communication of Italian scientific societies involved in neurorehabilitation.

Discussion

Virtual reality has recently been gaining popularity as a cognitive rehabilitation tool. Notwithstanding substantial evidence supporting the use of VR for ABI rehabilitation, there is limited evidence in sABI from traumatic or anoxic etiology. This multicenter pragmatic trial aims at investigating the effectiveness of a VR-based training, compared to a TCT, for cognitive rehabilitation of patients with sABI, and in particular for promoting improvement in executive functions. Given the paucity of solid previous evidence and the novelty of the device used, providing pre-specified hypothesis on a possible difference and on the direction of this difference between VR and TCT on the outcomes would be strongly speculative.

This trial has some limitations. Firstly, it is impossible to perform a double-blind procedure due to the nature of the intervention. However, all outcome measurements will be performed by examiners blind to the patient clinical diagnosis and intervention group (i.e., single blind). Secondly, the absence of a long-term follow-up evaluation will not allow us to investigate whether treatment effects are maintained over time. For this reason, further studies should follow the patients’ progress after discharge from post-acute rehabilitation.

Notwithstanding these limitations, this trial will provide new insights into the effectiveness of VR in sABI cognitive rehabilitation.

Trial status

The present version of the protocol has been approved on 14th October 2024. The recruitment started on 28th October 2024. The approximate date for study end is 31 st July 2026.

Abbreviations

ABI

Acquired Brain Injury

BDNF

Brain-Derived Neurotrophic Factor

DRS

Disability Rating Scale

EEG

Electroencephalogram

GFAP

Glial Fibrillary Acid Protein

GOAT

Galveston Orientation and Amnesia Test

LCF

Levels of Cognitive Functioning

mBI

modified Barthel Index

NF-L

Neurofilament Light Chain

NPI

Neuropsychiatric Inventory

OCS

Oxford Cognitive Screen

REDCap

Research Electronic Data Capture

sABI

severe Acquired Brain Injury

SUS

System Usability Scale

TCT

Traditional Cognitive Training

TMT

Trail Making Test

VR

Virtual Reality

Authors’ contributions

AE, the principal investigator, conceived the study with AlM. AE, AlM, MRF, GS, AnM, MC, CF, and MS contributed to the study design. AE, AlM, and MRF provided methodological and statistical expertise in the design of the clinical trial. AlM, MRF, GS, AnM and AE developed the study procedures manual. All the authors read and approved the final manuscript.

Funding

This work was supported by the Italian Ministry of Research, under the complementary actions to the NRRP “Fit4MedRob—Fit for Medical Robotics” Grant (# PNC0000007).

Data availability

Anonymized data can be requested from the corresponding author by bona fide researchers following the submission of a proposal of intended use and after their host institution has signed a data access agreement.

Declarations

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.