The purpose of the present study was to assess the effectiveness of the LMC® system in a semi-immersive VR protocol, as an adjunct to conventional rehabilitation treatment, in improving the functionality of the UL through grip strength, dexterity, and motor function compared to a conventional treatment group. In addition, the degree of satisfaction with the device was assessed, as well as adherence to treatment.
To our knowledge, the present investigation is the first RCT with follow-up to assess the effectiveness of the LMC® device in improving the functionality of the UL, through grip strength, dexterity, and motor function, in patients with stroke in the chronic phase. Satisfaction with the device and the rate of adherence to treatment were also considered.
Regarding the VR protocols with the LMC® device observed in the literature, each author handled different times, numbers of sessions, and frequencies [10]. The time spent in VR ranged from 15 to 45 minutes [11, 18–23], in interventions where it was supplemented with conventional therapy, the total treatment session time ranged from 45 to 60 minutes [11, 18–21] or even 90 [23] or 110 [22] minutes. The number of sessions also varied, although most studies conducted between 9 and 20 sessions, spread over 4 to 12 weeks [10, 11]. In summary, the amount of therapy with the LMC® device does not seem to be a determining factor in the improvement of the functionality of the UL, as significant improvements have been achieved with different times, numbers of sessions, and frequencies [10, 11]. Likewise, our work coincides with that proposed in the review by Jin et al. [4], where the 4-week period is established as the most widely adopted in VR studies in stroke patients.
With regard to grip strength, recovery is the primary goal. Muscle weakness is the most common impairment of the UL after stroke [24], leading, in addition to the impact on ADLs, to potential long-term problems such as decreased bone density and thus risk of fracture [25]. All authors who assessed it after intervention with the LMC® found significant improvements. Vanbellingen et al. [19] improved their participants’ grip strength by 11.3% in a single VR intervention. Similarly, but in combination with conventional therapy, improvements in grip strength were also achieved [11, 22]. Iosa et al. [22] attribute them to the fact that the LMC® system allows the capture and reproduction of movements with all degrees of freedom at the wrist and fingers. In our case, we obtained improvements in both groups, and the GE achieved significance between pre- and post-treatment assessments, and between pre-treatment and follow-up. Although the VR protocol did not include any specific muscle-strengthening exercises, the GE achieved significance. Improved stabilisation of the shoulder girdle in stroke patients is a necessary precursor to improved grip strength [26]. Following this premise, our results on improvements in gross movements (ARAT D) could justify improvements in grip strength.
Our results on gross motor dexterity assessed by the BBT show that both groups achieved the minimum detectable change necessary (1.99 blocks) to translate into clinical improvements in the affected hand [27]. The BBT was also the test of choice for other authors [11, 21] to measure gross manual dexterity, in both cases showing significance. In contrast, two other studies opted for the Nine Hole Peg Test as a method for assessing fine dexterity. Iosa et al. [22] did not obtain significance, although they did obtain improvements, while Vanbellingen et al. [19] did achieve significant differences. It should be noted that, in these two studies, some of the games developed involved selective finger movements, unlike those developed for our study.
In relation to motor function assessed by ARAT, there is strong scientific evidence of its improvement after VR protocols in stroke patients. It is one of the aspects most frequently addressed in research on VR, and there are different tools for its assessment [28, 29]. In the present investigation, despite the improvements in all subgroups and the significance obtained, no minimal clinically important differences were achieved in either treatment group. A score of 5.7 has been established as the value that determines these minimal differences [18], and in our case, all improvements were below this figure. Two papers [11, 18] used this test in their research on LMC® and stroke, in both cases showing significant differences. On the other hand, in one of them [18], in contrast to our results, there were minimal clinically important differences in GE, perhaps because their protocol was more intensive (three sessions per week for 6 weeks, using VR as the only therapy), and/or their intervention was immersive, as it combined the LMC® with Oculus Rift® (Oculus VR, USA) goggles. However, the authors did not break down their results into subgroups by differentiating between different motor functions. In order to gather more information about the participants’ DM motor function, we decided to include the DASH questionnaire. Although we relied on the patient’s subjective opinion, it allowed us to measure and quantify the degree of difficulty in performing certain ADLs. Despite being a validated questionnaire in stroke patients, its use is not very widespread in relation to VR, as we have seen both in different systematic reviews [3, 4, 28–31] as well as in articles where LMC® was used in relation to stroke[11, 18–23].
The patient’s opinion is fundamental in any neurorehabilitation process, and in our research, we used an ad-hoc Technology Satisfaction Questionnaire. “Technical problems during the intervention” was the item with the lowest score (2.56). In particular, the lack of a signal from the device was a recurrent complaint from patients. This is in line with Iosa et al. [22], who attribute it to the LMC® sensors’ lack of ability to follow the movements of the fingers when the hands are overlapped, or when there is a high level of spasticity. This same problem has also been reported in a previous study [11]. The next item with the lowest score (2.78) was “the use of electronic devices”, from which we wanted to learn the frequency with which participants used either mobile phones, computers, or video game consoles. Although the mean age of the EG was 60.33 years, we associated that the older the age, the lower the frequency of use of this type of device and, consequently, the greater the difficulty of use. However, this question was addressed by another article [22] in which the LMC® device was applied in older patients in the subacute phase, without the sample showing any problems with its use, although the participants’ satisfaction was not assessed.
On the other hand, the items with the highest scores were “the incorporation of new technologies into the rehabilitation process” and “the therapist’s help during the intervention”, especially during the first treatment sessions, both with a score of 3.83 points out of 4 [30, 31]. The help of a therapist during the play period also appeared in other similar studies [11, 19, 22, 23], concluding, as in our work, the advantage of having a clinician nearby at the beginning of the intervention.
Pain was another of the questionnaire items with the best score (3.61). Only a few patients reported occasional mild shoulder discomfort at the end of the treatment session, compatible with tendon overloads, which disappeared after the use of the therapy, and which at no time made it necessary to interrupt it. Nor was there any other type of discomfort, since, as indicated by Ogün et al. [18], the new generation of VR devices avoid symptoms such as dizziness, nausea, and headache, which is their biggest advantage compared to older devices. However, as no specific pain scale was used, the data provided by the questionnaire cannot be considered conclusive regarding pain.
Stroke, like the vast majority of chronic neurological diseases, given the care resources available, does not allow these patients to be treated continuously over time, and they are discharged from rehabilitation services after around 6 months [32, 33]. Current evidence has shown that improvements in functionality are still possible in the chronic patient [33]. This should force a change in the paradigm of care for these patients, but until this happens, and in the face of limited social and health care resources, the implementation of this type of treatment tool must be considered.
Limitations
The present study is not without limitations. The main limitation was the lack of stratification of patients in relation to their level of involvement. This would have allowed a more accurate assessment of the results, given the clinical heterogeneity inherent to UL involvement in stroke, as it is a way of homogenising the sample. For this same reason, together with the need for a larger sample size, our results cannot be extrapolated to the entire chronic stroke population.
Although common guidelines and directives were given, the therapy was applied by different therapists, and therefore the patients’ different backgrounds are a limitation.
To determine the full effectiveness of the LMC® as a therapeutic tool for chronic stroke patients, other comparisons, such as conventional treatment versus VR protocol as a single intervention or comparison with other types of VR devices, were not considered.