Rationale

Guillain-Barré Syndrome (GBS) is a prevalent cause of acute flaccid paralysis, distinguished by symmetrical weakness of the limbs, and hyporeflexia or areflexia ^{[1] [2] [3] [4]}. Reports of GBS date back to the second half of the 19th century, and it has been formally investigated since 1916 ^[3]. Infection or other immune-related stimulation that causes an atypical autoimmune response that targets peripheral nerves and their spinal roots is often what precedes it ^{[2] [5]}; more than 90 % of patients reach the peak of the disease severity between 2 and 4 weeks since the sickness debuts ^[6]. There are multiple clinically distinguishable subtypes of GBS: Acute inflammatory demyelinating polyradiculoneuropathy (AIDP), the most common; Miller- Fisher Syndrome (MFS); Acute motor axonal neuropathy (AMAN); Acute motor sensory axonal neuropathy (AMSAN); Bickerstaff’s Encephalitis (EB); and Pharyngealcervical- brachial weakness ^{[1] [6] [7] [8]}.

The symptoms of GBS vary depending on the subtype. Taking AIDP, for example, the aberrant immune response consists mainly in the degradation of the patient’s myelin ^[9], this reports in paresthesia in feet and fingertips followed by symmetrical or slightly asymmetrical weakness in lower limbs that can ascend to the upper limbs in a matter of hours or days, and even affect the respiratory system muscles in severe cases ^{[1] [6] [10] [11]}.

GBS treatment generally combines multidisciplinary supportive medical care and immunotherapy. Recognized as the most effective treatments: Intravenous Immunoglobulin (IVIg) and plasma exchange are the most common and have proven to be very effective, especially if they are started within the first 2 weeks of the disease’s debut ^{[1] [12]}. Although the majority of cases report a complete or partial recovery of the patients, with very few resulting in death, many of the affected individuals obtain a residual deficit that impacts their living conditions. For instance, 20 % of GBS patients still require assistance walking six months after their sickness first manifested. Reduced muscular strength, sensory indications, weariness, and discomfort are the most typical remaining deficiencies ^{[1] [13] [14]}.

Currently, the GBS rehabilitation process is based primarily in a multidisciplinary set of interventions, conducted mainly by neurologists and rehabilitation physicians. The main goal is to give the patient back a partial or complete autonomy and the ability to carry out normal life activities. According to the requirements of the particular patient, this may also include nursing care, nutritionist guidance, psychotherapy, speech therapy, and social rehabilitation in addition to physical or occupational therapy and exercise regimens. One of the main problems is that since multidisciplinary rehabilitation varies markedly between regions, it is challenging to design research trials to properly assess the effectiveness of the rehabilitation ^{[10] [11]}.

Another common rehabilitation approach for GBS is exercise ^[15]. For example, in a Dutch study, 20 GBS patients who complaint mainly of fatigue had an intervention consisting of three 45-minute sessions weekly for 12 weeks, the target was to increase heart rate from 65 % to 90 % of maximal heart rate, and during that period the workload was gradually increased. As a consequence, there was less fatigue, more isokinetic muscular strength, and a higher peak oxygen uptake ^{[10] [15] [16]}.

Neuromuscular electrical stimulation has been a remarkable ally in GBS rehabilitation as well, especially for patients with a delicate state of health or an acute paralysis ^[10]. Without the patient's involvement, muscle contractions can be induced via neuromuscular electrical stimulation (NMES). In the acute stage of GBS, until patients have adequately recuperated to start a multidisciplinary rehabilitation effort, this may be an alternate treatment approach that can minimize inactivation and denervation waste ^{[10] [17]}.

Besides these approaches for GBS rehabilitation, there are some cases where biomedical engineering is applied in neurorehabilitation for this disease, however, these cases not only are scarce compared to the use of traditional rehabilitation but they are also little studied, hence the need for a systematic review for the analysis of the current state of the art.

Objective

This systematic review targets to respond the question “How has biomedical engineering been used for the rehabilitation of patients with disability caused by Guillain-Barré?” and to provide a broad perspective on the knowledge that is presently available on the subject. It seeks to give a solid platform for decision-makers, therapists and future investigations for the development of better neurorehabilitation methods based on biomedical engineering in GBS.

Review design and search strategy

This systematic review was developed in concordance with the Preferred Reporting Items for Systematic Reviews and Meta- Analyses (PRISMA) 2020 statement guidelines ^[18], over three months: from April to June of 2023. Five databases were used in total: ScienceDirect, PubMed, IEEE Xplore, Springer and Dimensions. The guiding question for the review design and execution was: “How biomedical engineering has been used for the rehabilitation of patients with disability caused by Guillain-Barré?”.

The investigation was performed using a search equation combining Boolean operators and focused on three key terms, and was stated as follows: ("Guillain Barré" OR "Acute inflammatory demyelinating polyradiculoneuropathy" OR "Miller-Fisher Syndrome" OR "Acute motor sensory axonal neuropathy" OR “Acute motor axonal neuropathy”) AND (rehabilitation OR recovery) AND ("biomedical engineering" OR "electronic device" OR "haptic device" OR “neuromodulation” OR “virtual environment” OR “telemedicine” OR “artificial intelligence” OR “internet of things” OR “wearable device” OR “robot” OR “biomedical transducer”). The key terms related to biomedical engineering were selected by using the IEEE Taxonomy.

The overall procedure for recognizing relevant articles was: (I) literature search in designated databases applying eligibility criteria; (II) export of results in reference files and import to specialized software for literature reviews (Rayyan); (III) removal of duplicates; (IV) selection of articles by title and abstract; (V) selection of articles by full-text analysis; (VI) detection of any additional relevant papers using the snowball technique.

This academic work is exempt from Institutional Review Board clearance because it is a systematic review.

Eligibility Criteria

Selection

After the database search, each article was subject to a manual filtration process by analyzing the title and abstract, followed by a full-text selection. Every article was scrutinized independently by the four authors, and when conflict arose, consensus was achieved through discussion.

The selection of papers, in both processes, was conducted considering the eligibility criteria and the focus of the research question.

Data extraction

Data were obtained from eligible papers using a standardized format based on the PRISMA guidelines where a variety of key summarizing points of information were collected; this was carried out by the three reviewers. The variables determined useful to extract from each article were: Author(s), publication date, country (or region where the research was conducted), objective, patients, brief description of the biomedical engineering device, methodology, results and conclusions. The result of this extraction is exposed in Table 1.

Quality assessment of the included articles

To assess the quality of the articles it was used the “Checklist for assessing the quality of quantitative studies” of the “Standard Quality Assessment Criteria for Evaluating Primary Research Papers from a Variety of Fields”, developed by Kmet, Lee and Cook in collaboration with the Alberta Heritage Foundation for Medical Research ^[19]. This checklist consists of 14 items stated as follows: I1: Question/objective sufficiently described. I2: Study design evident and appropriate. I3: Method of subject/comparison group selection or source of information/input variables described and appropriate. I4: Subject (and comparison group, if applicable) characteristics sufficiently described. I5: If interventional and random allocation was possible, was it described? I6: If interventional and blinding of investigators was possible, was it reported? I7: If interventional and blinding of subjects was possible, was it reported? I8: Outcome and (if applicable) exposure measure(s) well defined and robust to measurement/ misclassification bias. Means of assessment reported. I9: Sample size appropriate. I10: Analytic methods described/justified and appropriate. I11: Some estimate of variance is reported for the main results. I12: Controlled for confounding. I13: Results reported in sufficient detail. I14: Conclusions supported by the results. This checklist consists of 14 items. All the papers with 80% or more of positive results were included.

Search results

The database search using the Boolean equation yielded a total of 24,691 papers, with most articles found in Dimensions (23,818). However, after applying filtration based on the inclusion and exclusion criteria, only 2,237 remained. Of this total, 179 were duplicates, and the rest underwent analysis based on title and abstract, resulting in 154 papers. The next step involved a thorough examination of the full text of these articles to determine their usefulness for the review. Additionally, the snowball technique was employed by reviewing the references. Following this process, 19 articles were selected for inclusion in the systematic review. These selected articles were used to summarize the information using the pre-defined focus variables (refer to Table 1). The PRISMA flow diagram in Figure 1 illustrates the selection steps and results, providing reasons for excluding 135 out of the 154 articles.

Figure 1 PRISMA flow diagram for the steps followed to obtain the selection of eligible papers ^[18]. 

Quality and Reliability of Results

The included studies' quality was analyzed using the Checklist for assessing the quality of quantitative studies and the conclusions are reflected in Table 2. Figure 2 exhibits how many of the 19 articles are JCR and the number of articles that correspond to Quartile 1, 2, 3 or 4.

Figure 2 Pie chart of the JCR quartiles the journals of the presented articles belong to. Global distribution of the studies 

Summarized results obtained from the articles

The analyzed 19 studies have been described in Table 1 reporting the chosen variables: Author(s), publication date, country (or region where the research was conducted), objective, patients, a brief description of the biomedical engineering device, methodology, results, and conclusions.

The presented studies took place in 11 countries (Figure 3), although, in many cases, researchers of the same study came from different Universities, regions and even from different countries. The place where more studies were conducted was Zürich, Switzerland, with a total of 4.

Figure 3 Map of the countries where the studies took place. 

Discussion

The need to organize information on GBS and its treatments is necessary for progress and future advances. This systematic review started with the of how biomedical engineering has been utilized towards the rehabilitation of patients with Guillain-Barré and the path to answer it start with a search equation divided into three parts: 1.- The disease and its most common subtypes (GBS), 2.- the key objective in relation to this disease (to rehabilitate) and 3.- the instruments to achieve it via biomedical technologies. A minor detail worth mentioning is that one of the used databases, Science Direct, has a limit of Boolean operators, but by repeating the search while rotating the terms the same result was achieved as if it did not have this restriction. It is plausible to criticize this search equation for being slightly general, however, the researchers have determined that there is not enough information about this topic, thus warranting a reasonably general search equation that translates into more results and meticulous screening.

Regarding search results, it is interesting how few eligible studies were obtained, despite the 22-year span that was examined. This could indicate, contextualized within this systematic review, a lack of development of this field and, therefore, a significant area of future study.

Concerning the reliability of the obtained articles, 14 of the journals are included in the Journal Citation Reports ^[39], half of them in quartiles 1 and 2, which is a recognized standard for research quality. This indicates that the articles probably contain reliable information, but it does not mean they are not subject to possible bias. The use of the Checklist developed by Kmet, Lee and Cook also revealed the excellent quality of the articles.

Only 5 articles provided complete information on the sex, age, height and weight of the patients ^{[20] [21] [22] [23] [24]}. Considering articles that provided data on the sex of the GBS patients, there were in total 13 male ^{[20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30]} and 2 female patients ^{[31] [33]} examined throughout these investigations. Analyzing the articles that provided data about patient’s ages, the age ranged from 7 to 86 years old, with the mean being 39.5 years.

The main area targeted by the studies was the lower limbs, eleven studies ^{[20] [21] [22] [23] [24] [26] [27] [28] [30] [33] [34]} worked with biomedical devices that helped patients’ rehabilitation in this area. Six studies focused on the upper limbs ^{[25] [29] [32] [35] [36] [37]}. Céspedes et al.^[38] worked on a Socially Assistive Robot using a NAO platform, and even though the subjects were focused on lower-limb rehabilitation, the purpose of the platform is to be an assistant therapist that makes observations and gives feedback to the patient. Wosnitzer et al.^[31] used neuromodulation to attend urinary retention aiming the study at the sacral nerves. The area that more studies targeted was the lower limbs, which is consistent with the data that the most common disability amongst GBS patients is weakness or partial or complete paralysis in legs ^{[5] [7]}. Most studies only included male GBS patients, solely two women were mentioned as subjects. This observation aligns with the fact that, unlike other autoimmune diseases, GBS is significantly more frequent in men than in women ^{[11] [40]}.

Of the 19 studies, 12 took place in countries that, according to the United Nations’ “World Economic Situation and Prospects 2022” ^[41] are developed economies: Switzerland, Japan, United States of America (USA), Australia, Finland and United Kingdom (UK), the first one being the one where more investigations were conducted: four. The remaining 7 took place in countries with developing economies, all of which hosted one study, except for South Korea, where 3 took place. A possible connection between the number of studies for each continent and the density of incidences of GBS cases could be theorized, nonetheless, the data does not reveal one: The continents with more studies were Asia and Europe, each with 7, however, the statistics show that in these regions the range of incidences per 100,000 habitants is 0.44-3.25 and 0.841.91, respectively. In South America, where only 2 researches were found, the range is 5.6-7.63 per 100,000 habitants ^[41]. Probably the real reason behind the differences in the number of studies are population and economic development: In Asia lives more than half of the planet’s population ^[42], Japan is one of the world’s leading economies and India, South Korea and Iran are thriving developing economies ^[41] that excel in technology and science; Europe’s population is almost twice of that of South America ^[42], and Finland, UK and Switzerland are three formidable developed economies with high per capita income ^[41].

The variety excels in the presented studies in terms of the biomedical engineering devices utilized. Ten of the devices were custom-made systems, tested in the same article ^{[20] [21] [22] [24] [25] [29] [32] [35] [36] [37]}. The other seven devices were acquired commercially ^{[23] [26] [27] [28] [30] [31] [33] [34] [38]}. Two of these devices were used in different studies: Morning Walk was the protagonist in two papers ^{[33] [34]}, and Lokomat was the main device in two studies as well ^{[26] [28]}. However, it was also employed to test the NAO platform as a therapy assistance robot in a third study ^[38].

There were a total of four devices based on haptics and virtual reality, with one focusing on lower limb rehabilitation ^[20] and the other three concentrating on upper limb rehabilitation ^{[29] [32] [36]}. All of them paid close attention to the issue of patient motivation during therapy, particularly the PITS ^[29], which centered on pediatric rehabilitation. The developers applied principles of serious game development, as the main objective and the entire experience are designed to aid in rehabilitation while still maintaining motivation through an enjoyable gaming experience. The positive results obtained in the studies demonstrate the usefulness of specific-purpose video games in rehabilitation, a concept supported by research such as the one conducted by Ong et al.^[43].

Three devices were orthoses: two of them were wearables designed for the ankle-foot area ^{[21] [22]}. Contrary to most orthoses, these devices had integrated electronic systems that improved the patients' ability to move their feet. The third one is the Lokomat ^{[26] [28]}, which functions as an integrated system providing body weight support and rehabilitating gait by driving both legs. The Biodex Unweighting Support System ^[27] only provides weight support but is adjustable, aiding in gait rehabilitation. The Morning Walk devicen^{[33] [34]} had great relevance in gait rehabilitation. Even though, in some cases, the saddle caused discomfort, the use of the end-effector footplates has proven to be beneficial for the patients.

It is worth mentioning that four of the studies addressed limb-related issues indirectly. Jamshidi et al.^[23] developed a computerized model using the SimMechanics toolbox to simulate human gait, aiming to contribute to the improvement of leg orthosis.

Céspedes et al.^[38] designed their experiment around patients using the Lokomat for gait rehabilitation, with the main focus on developing an automated assistant for rehabilitation. Kauhanen et al.^[35] addressed upper limb rehabilitation, targeting tetraplegic patients. The EEG device 'TTK-BCI' provided screen feedback based on left-hand or right-hand movement attempts. Lastly, Lee et al.^[25] created a passive shoulder joint tracker that can be integrated into various rehabilitation mechanisms to aid in upper limb rehabilitation. An outlier in the investigations was conducted by Wosnitzer et al.^[31]; they did not focus on arm or leg rehabilitation but successfully applied sacral nerve neuromodulation to treat urinary retention.

Almost every GBS patient treated showed positive rehabilitation results; one of the few exceptions occurred in the study conducted by Choi et al.^[33], where one patient was unable to complete the process due to discomfort with the saddle. The biomedical engineering devices proved to be useful, although, in most cases, further analysis and validation are required, especially in the long term.

A wide variety of parameters, scales, and methods were used to measure the patients' evolution during research directly dedicated to rehabilitation. The most commonly used tests ^{[20] [26] [28] [30]} included the 10-meter walk test, which is essentially a speed evaluation measuring the time it takes for the patient to walk 10 meters ^{[43] [44] [45] [46]}. The 'timed up and go test' also calculates speed but involves additional activities such as standing up, turning around, and sitting down ^{[45] [46] [47]}. A related test designed to estimate endurance is the 6-minute walk test ^{[21] [26] [28]}, which records the distance the patient walks in 6 minutes ^{[48] [49] [50]}.

The Medical Research Council Scale for Muscle Strength was also applied ^{[27] [33] [34]}. This scale focuses on muscle strength, encompassing 6 muscle systems in total ^{[51] [52] [53]}. However, since these studies were focused on the lower limbs, only 3 systems were considered: hip flexors, knee extensors, and foot dorsiflexors. Typically, therapists used the Functional Ambulation Categories ^{[26] [33] [34]} in conjunction with this form of examination.

Besides velocity and endurance, equilibrium is a crucial parameter to observe. For this purpose, the Berg Balance Scale ^[21], Tinetti test, and Unipedal stance time were employed ^[20]. These methods were useful in assessing patients' improvement, and all objectively revealed a correlation between the use of biomedical devices and the subjects' progression.

Multiple studies reveal the importance of motivation and comfortability during the neurorehabilitation process ^{[54] [55] [56] [57] [58]}. In order to take in consideration this topic, both the physical and psychological acceptance of the devices were taken in consideration in the studies, two of them used scales like the Suitability Evaluation Questionnaire ^[20] or the User Experience Questionnaire ^[37], but most authors reported subjective observations made by the therapists. In the bulk of the studies, the authors observed that the improvement of the patient’s autonomy and motor abilities had a remarkable incidence in their motivation and psychological well-being.

The usability of biomedical devices was tested in two studies by means of the System Usability Scale ^{[24] [37]}, which is a highly reliable tool for measuring the usability of a system or device. While this scale can be somewhat subjective, it is also very versatile ^{[58] [59]}. However, in most research, authors opted for custom-made scales.

Living with the disabilities caused by GBS can be a significant challenge, as basic day-to-day activities become much harder to execute. Both patients and caregivers are psychologically affected, and their quality of life is reduced. This impact can be even more challenging for those who don’t have good economic conditions.

Biomedical engineering devices are being used as auxiliary technologies that serve as extensions of human capacities. They enhance traditional rehabilitation methods, and the benefits for the patients multiply. There is a vast range of possibilities with this equipment, from wearables capable of providing firmness to steps to haptic devices that train fine motor abilities. Additionally, there are very complex mechanical systems that can take a patient from not being able to walk 4 meters to almost complete independence.

We must highlight the importance of studying the long-term effects of these rehabilitation methods. There should be more follow-up studies to understand the consequences in the long term and how well patients have evolved, in order to pursue a holistic comprehension of the benefits of these devices.