Combining prismatic adaptation and digital cognitive training: preliminary evidence on cognitive and biological effects in patients with mild cognitive impairment

doi:10.21203/rs.3.rs-5288152/v1

Download PDF

Research Article

Combining prismatic adaptation and digital cognitive training: preliminary evidence on cognitive and biological effects in patients with mild cognitive impairment

https://doi.org/10.21203/rs.3.rs-5288152/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

This study explored the efficacy of a new rehabilitation tool that combines prismatic adaptation (PA) and cognitive training through serious games (SGs) in patients with mild cognitive impairment (MCI) due to two aetiologies: prodromal to Alzheimer’s dementia or consequent to Parkinson’s disease. We investigated whether this tool could improve cognitive performances, with effects at least similar to programs used in clinical practice. Leveraging studies suggesting a neuromodulatory effect of PA, we explored if the PA+SGs combined treatment could influence plasticity-related mechanisms assessed through brain-derived neurotrophic factor (BDNF) serum levels, compared to cognitive training with only SGs and standard cognitive rehabilitation (SCR).

23 MCI patients were randomized into three intervention groups: PA+SGs, SG-only, and SCR, completing 10 treatment sessions. Before and after the treatment, patients underwent neuropsychological assessment and blood sampling acquisition.

At baseline, all groups showed similar demographic, clinical, and biological characteristics. Post-treatment, the PA+SGs group improved in memory, executive function, and visuospatial abilities, although these changes were not statistically different from the control groups. Increased BDNF serum levels were observed only in the PA+SG group and were positively correlated with improved memory and language performance.

Our findings suggest that combining PA with cognitive training may improve cognitive functioning in MCI patients, with results similar to SCR. Further, PA seems to enhance neuroplasticity mechanisms that may support the behavioral improvements of cognitive training. Future research should confirm these findings and delve deeper into the relationship between cognitive impairment and its rehabilitation, based also on the underlying neurobiological mechanisms.

Trial registration number(Clinicaltrials.gov): NCT05826626

cognitive impairment

prismatic adaptation

Alzheimer’s disease

Parkinson’s disease

brain-derived neurotrophic factor

brain plasticity

Mild Cognitive Impairment (MCI) is defined as a state of cognitive decline that is more prominent than expected for an individual’s age and educational level, and that does not significantly interfere with daily activities [1]. Historically, the term MCI has been applied to identify a prodromal stage of dementia in patients with Alzheimer’s disease (AD), but it has been later broadened to include also cognitive impairment due to other etiologies, such as Dementia with Lewy Body (LBD), Parkinson’s disease (PD), vascular dementia (VD) or frontotemporal dementia (FTD) [2]. Overall, MCI prevalence in older adults is estimated to range between 5 and 30%, with an incidence rate of 22.6 per 1,000 persons every year [3]. Although MCI may be a stable condition [4], [5], it represents one of the main risk factors for dementia [6]. Approximately 11 to 33% of cases of MCI convert to dementia within 2 years [7], with patients showing memory deficits (amnestic MCI) being twice as likely to progress to dementia than non-amnestic ones. Moreover, patients with MCI may show multi-domain cognitive deficits, affecting, for instance, language[8], calculation [9], [10], attention and executive function [11], [12] in addition to memory, which can subsequently impact also affective behaviors [13] and daily tasks, such as decision-making and financial independence [14]–[16]. Overall, given the ongoing growth of the elderly population[17], MCI is a matter of significant public health concern.

To date, there are no recognized pharmacological therapies for MCI, and behavioral interventions have been designed to improve cognitive performance or help individuals compensate for cognitive difficulties [18]. Among these, cognitive rehabilitation approaches have shown promising results (e.g., [19], [20]), especially when relying on the use of new technologies [21], [22]. In particular, in recent years, new programs for cognitive training were developed using a gamification approach. Serious Games (SGs) are defined as digital games created with the intention of achieving additional outcomes other than entertainment alone [23], [24]. SGs are used to overcome some limits of classical cognitive rehabilitation, as they can increase the patient’s motivation and involvement in the rehabilitation process [25]. Moreover, they allow therapists to easily personalize treatment to the patient’s needs, thanks, for instance, to a dynamic adaptation of difficulty parameters to the patient's performance [26]. In addition, several studies have been exploring the possibility of combining cognitive training with non-invasive brain stimulation (NIBS) techniques, such as transcranial magnetic stimulation (TMS) or transcranial direct current stimulation (tDCS). Although these techniques act differently on cortical excitability, they can both induce synaptic plasticity through Long-Term Potentiation (LTP) or Long-Term Depression (LTD) [27], [28], ultimately inducing changes that can support the reorganization of the brain networks involved in the impaired cognitive function [29], [30]. On this line, the combination of NIBS with cognitive training has shown promising results in different populations (see, for instance, [31], [32]). Recently, some studies suggested that Prismatic Adaptation (PA), a visuomotor procedure inducing a shift of the visual field using prismatic lenses, may induce higher cortical excitability in the frontoparietal network of the hemisphere ipsilateral to prism deviation [33], with tDCS-like effects [34]. Although PA cognitive effects in patients with MCI have been scarcely studied [35]–[37], a recent pilot study on stroke patients combining PA with cognitive training suggested that PA could prime cortical excitability in order to make it more responsive to cognitive training, thus supporting its positive effects [38].

Nevertheless, the mechanisms underlying the effects of cognitive rehabilitation, alone or combined with other techniques are still not fully understood. In other words, it is not clear to what extent the rehabilitation may act only on the symptomatic component (i.e., cognitive impairment) or also on its underlying processes, such as neurodegenerative or neuroplastic ones. To shed light on this issue, some studies suggested the use of the brain-derived neurotrophic factor (BDNF) as a potential marker for the effectiveness of interventions [39]. BDNF is a neurotrophin secreted by neurons in the brain, crucial for maintaining both cognitive and motor functions [40]. It supports neurons and promotes synaptic plasticity through long-term potentiation [41]. Higher BDNF levels at baseline have been associated with decreased odds of developing dementia [42]. Moreover, increased BDNF serum levels have been linked to the effectiveness of non-pharmacological interventions in ameliorating cognitive performances in different populations [43], [44]. However, studies focusing on patients with MCI have yielded inconsistent results [45]–[47].

The aims of the present exploratory study were manifold. Firstly, we aimed to investigate the association between BDNF serum levels and the degree of cognitive impairment in a cohort of patients with MCI due to different underlying aetiologies, specifically AD or PD. Moreover, we aimed to explore the effectiveness of different rehabilitative approaches on the diverse cognitive impairments associated with the disease, and the promotion of brain plasticity. To this aim, we compared changes in cognitive performances and BDNF serum levels between patients who underwent traditional cognitive rehabilitation, SG-based cognitive training, or SG-based cognitive training combined with PA. In particular, we hypothesized that PA could be a useful tool to enhance the beneficial effects of cognitive training and promote plastic changes in patients with MCI.

The present study is part of a clinical trial registered on Clinicaltrials.gov (NCT05826626).

Participants

Study participants were recruited among patients admitted to IRCCS San Camillo Hospital or referred to a clinical screening for suspected cognitive impairment from September 2021 to August 2023. Inclusion criteria were: i) aged 40–85 years old; ii) diagnosis of MCI [1], [48]; iii) preserved use of at least one hand; iv) absence of psychiatric illnesses and/or comorbidity with other neurological pathologies; v) ability to provide informed consent. Eligible participants were randomized into three groups: combined cognitive training using PA and SGs (PA + SG, experimental group), cognitive training using only SGs (SG-only, control group), and conventional cognitive rehabilitation (SCR, control group). Before and after the treatment, patients underwent a cognitive and clinical assessment, as well as a blood sample acquisition. All evaluations were performed by experienced and trained neuropsychologists. Blood sample acquisitions were performed by trained nurses. Biological analyses were performed in blind. All participants took part in the study voluntarily and gave informed consent according to the Declaration of Helsinki. The study was approved by the Ethics Committee of Venice and IRCCS San Camillo Hospital (Venice, Italy), reference number 2021.12.

Cognitive and clinical assessment

Each patient completed a comprehensive neuropsychological evaluation before and immediately after the treatment. The assessment included measures of attention, executive functions, memory, visuospatial abilities, and language. In particular, the following tests were administered: (i) Mini-Mental State Examination (MMSE) for general cognitive functioning; (ii) Trail Making Test (TMT) for attention; (iii) Stroop test, Clock Drawing Test (CDT), Digit and Spatial Span backward, and Phonological fluencies for executive functions; (iv) Digit and Spatial Span forward, Rey Auditory Verbal Learning Test (RAVLT), Prose memory, and recall of the Rey-Osterrieth Complex Figure (ROCF-recall) for learning and memory domain; (v) copy of the Rey-Osterrieth Complex Figure (ROCF-copy) for the visuospatial and visuoconstructive domain; (vi) Semantic fluencies for the language domain.

Lastly, psychological concerns were assessed using the State-Trait Anxiety Inventory (STAI-Y) and Beck Depression Inventory-II (BDI-II).

Intervention

All groups completed ten 30-minute sessions in 2 weeks (5 sessions/week).

The first group (PA + SG) underwent the experimental treatment using Mindlenses Professional, a new tool that allows coupling the digital implementation of PA with the administration of SGs [38], [49], [50]. Specifically, the PA procedure was performed in the first five minutes of each session, followed by approximately 20 minutes of digital cognitive exercises (SGs). The pointing task was performed on a tablet under three conditions: pre-exposure, exposure, and post-exposure. The pre-exposure and the post-exposure conditions counted 30 trials each (10 for each target position), and the exposure condition counted 90 trials (30 for each target position). In the exposure condition, patients performed the pointing task while wearing prims inducing a 10° shift of the visual field. The direction of the shift (right vs. left deviation) was randomized across patients, as no clear effects of different deviations are reported in the literature in this population. Patients were instructed to keep their index finger at the sternum level and to place the fingertip on the target at a fast but comfortable speed. Pointing movements were executed with the right hand. The sight of the hand and the arm was allowed for the whole duration of the procedure (concurrent exposure procedure). The tablet resolution was 1200 x 2000 pixels, with a pixel density of 224 ppi. Spatial displacement of the pointing movements was recorded automatically by the tablet in terms of pixels. Immediately after PA, seven SGs were administered every day in the same order. The games were designed to tackle various components of attention, executive function, and language. In detail, the games addressed visual search, sustained and alternate attention, planning, inhibition, verbal and abstract reasoning, working memory, short-term memory, and calculation [38].

In the second group (SG-only), the same digital exercises of the MindLenses protocol were administered without first performing the PA procedure.

Lastly, patients in the SCR groups completed ten sessions of conventional cognitive training using the Erica® software (Giunti Psychometrics, Italy, 2013).

Blood samples collection and serum isolation

Blood samples were drawn from patients before and after completing the treatment and collected in BD Vacutainer® SST™ II Advance vials (#366882A, BD Vacutainer, Becton Dickinson and Company, New Jersey, USA). Samples were left to coagulate for 30’ at room temperature (RT) and successively centrifuged for 15’ at 1500 g. The serum was then divided into aliquots and stored at -80°C before use.

Brian-Derived Neurotrophic Factor (BDNF) assay

Serum isolated from patients’ blood was thawed in ice prior to analysis. BDNF concentration in the serum was tested with the ChemiKine BDNF Sandwich ELISA Kit (#CYT306, CHEMICON International, Inc., California, USA) according to the manufacturer’s instructions. Briefly, samples were diluted (1:100) in sample diluent, added to the plate along with the standard samples, and run in duplicate. After overnight incubation at 4°C and washing with washing solution, incubation with the primary antibody was performed for 2h at room temperature (RT). Wells were then incubated with the horseradish peroxidase-conjugated secondary antibody for 1h at RT. After washing, tetramethylbenzidine solution was added for 15’ at RT; the reaction was stopped with the stop solution to allow visualization. Absorbance was measured at 450 nm with the Infinite F50 Plus plate reader (TECAN Trading AG, Switzerland).

Statistical analysis

As preliminary steps, scores at cognitive tests were corrected for age, education, and gender using normative data for the Italian population. The Shapiro-Wilk test was then used to assess the data distribution. As the majority of variables exhibited a deviation from normality, non-parametric tests were used for further analyses. We examined differences in demographic variables between groups using the Kruskal-Wallis test for continuous variables or the Chi-square for nominal ones. One-way non-parametric ANOVAs using the Kruskal-Wallis test were then conducted to investigate baseline differences between groups at the cognitive and clinical evaluations.

As an exploratory step, paired-sample t-tests using the Wilcoxon signed-rank test were computed separately for each group to explore within-group changes in cognitive performances after the treatment. Afterward, to investigate the presence of differences between groups in the pre vs. post-treatment changes, non-parametric ANOVAs using the Kruskal-Wallis test were applied to delta scores, computed for each measure by subtracting pre-treatment scores from post-treatment ones. Post-hoc tests using Mann-Whitney U were then applied to explore differences in changes in pre vs post-treatment cognitive performances between the experimental and the two control groups. Bonferroni’s correction for multiple comparisons was then applied to control for inflated type I error rates. Effect sizes were computed following the method described by Rosenthal [51].

Concerning BDNF levels, preliminary correlations were performed to investigate whether baseline BDNF levels and BDNF changes at post-treatment may be associated with demographic characteristics, such as age and education. Similarly, non-parametric one-way ANOVAs using the Mann-Whitney U test were conducted to explore the association between BDNF levels and categorical variables, such as gender, and clinical characteristics, such as MCI pathology (AD or PD), and MCI type (single vs multiple domain). Correlation analyses were then performed using Spearman's rho to investigate the baseline association between cognitive scores and BDNF serum levels and to explore if changes at the cognitive level could be associated with changes in BDNF depending on the type of rehabilitation received.

All statistical analyses were conducted with SPSS 23 [52].

Sociodemographic and baseline characteristics of the sample

Thirty patients were enrolled in the study. However, seven patients did not complete the rehabilitation for reasons unrelated to the study (tested positive for COVID-19 n = 3; early discharge n = 3; or personal reasons n = 1). Therefore, a total of 23 patients were included in the analysis. No statistical differences emerged when comparing drop-outs and included patients in the baseline characteristics (see Table S1 in the supplementary materials). Table 1 reports the sociodemographic characteristics of the final sample and separately for each treatment group.

Table 1

Sociodemographic characteristics of the patients enrolled in the study. Mean and standard deviation, or number of patients and percentages, are reported. Results of the Kruskal-Wallis (H) or the Chi-square test (X²) are reported for continuous and categorical variables, respectively.
	Whole group (n = 23)	PA + SG (n = 7)	SG-only (n = 7)	SCR (n = 9)	H / X² (p-value)
Age, years (SD)	73.91 (8.00)	72.43 (11.12)	72.14 (8.47)	76.44 (4.19)	0.795 (.672)
Education, years (SD)	10.43 (3.09)	12.71 (0.76)	9.43 (2.99)	9.44 (3.54)	5.253 (.072)
Gender, n. females (%)	9 (39.1)	2 (28.6)	3 (42.9)	4 (44.4)	0.475 (.789)
MCI pathology, n. AD (%)	11 (47.8)	3 (42.9)	3 (42.9)	5 (55.6)	1.168 (.558)
MCI type, n. single-domain (%)	8 (34.8)	2 (28.6)	3 (42.9)	3 (33.3)	0.329 (.849)

All groups had similar age (p = .672), educational level (p = .072), and gender distribution (p = .906). Concerning patients’ cognitive profiles (Table 2), at baseline, no significant differences emerged between groups in all the administered neuropsychological tests (all ps > .050). Notably, most of the sample showed difficulties in more than one cognitive domain (multi-domain MCI). Moreover, no significant differences were observed in baseline BDNF serum levels between the three intervention groups (H = 3.297, p = .192).

Table 2

Scores at the cognitive and clinical evaluation across the three groups. Means and standard deviations of corrected scores are reported for each test/clinical scale and BDNF serum levels (ng/mL).
Domain/test	PA + SG		SG		SCR
	T0	T1	T0	T1	T0	T1
General cognitive functioning
Mini-Mental State Examination	26.46 (2.01)	27.20 (0.91)	26.50 (2.10)	26.84 (2.56)	26.64 (2.21)	27.82 (2.34)
Attention
Trail Making Test - A	44.57 (21.95)	35.29 (38.61)	37.17 (30.24)	34.83 (26.81)	42.78 (34.91)	37.33 (31.69)
Trail Making Test - B	113.0 (73.45)	134.0 (91.75)	71.8 (114.03)	95.6 (78.02)	92.25 (76.96)	109.5 (98.59)
Executive function
Phonological Fluency	33.90 (15.86)	38.87 (25.30)	38.13 (8.93)	37.73 (12.45)	32.46 (11.53)	35.56 (9.40)
Stroop test - n. errors	0.57 (1.77)	0.321 (0.59)	3.58 (7.01)	3.21 (5.51)	0.78 (1.35)	2.056 (3.48)
Stroop test - time	27.07 (28.84)	15.00 (6.64)	12.08 (10.76)	18.68 (16.20)	23.42 (16.16)	20.89 (15.13)
Clock drawing test	8.14 (2.27)	8.21 (1.50)	7.20 (4.09)	8.33 (3.60)	8.19 (2.31)	9.06 (0.77)
Digit span backward	4.23 (1.11)	3.79 (0.86)	4.00 (0.82)	4.14 (0.90)	4.00 (0.97)	3.67 (1.12)
Spatial span backward	4.07 (1.27)	4.64 (0.85)	3.86 (1.46)	4.07 (1.48)	3.78 (1.00)	3.83 (0.66)
Memory
Digit span forward	5.57 (0.73)	5.43 (1.54)	5.93 (1.64)	5.29 (0.81)	4.78 (1.28)	5.17 (1.20)
Spatial span forward	4.29 (0.93)	4.71 (0.81)	4.29 (1.41)	4.86 (1.57)	3.94 (0.88)	4.72 (0.51)
Rey Auditory Verbal Learning Test - immediate recall	38.73 (11.15)	43.97 (14.27)	42.58 (5.79)	48.20 (14.61)	44.04 (15.15)	51.26 (14.73)
Rey Auditory Verbal Learning Test - delayed recall	8.27 (3.95)	9.23 (4.03)	7.72 (3.97)	9.89 (4.65)	9.41 (4.13)	10.63 (4.25)
Rey-Osterrieth Complex Figure - recall	14.18 (8.28)	17.00 (9.40)	16.04 (7.24)	17.50 (8.98)	16.06 (4.85)	18.46 (7.49)
Prose memory	12.14 (5.15)	18.92 (7.95)	18.66 (8.44)	27.57 (9.29)	19.38 (7.59)	24.63 (9.78)
Language
Semantic fluency	33.86 (17.85)	35.57 (17.86)	38.33 (8.76)	39.50 (14.75)	36.67 (11.15)	40.11 (10.88)
Visuospatial Abilities
Rey-Osterrieth Complex Figure - copy	33.07 (5.29)	34.32 (4.55)	34.71 (1.50)	32.12 (5.69)	31.67 (4.47)	33.21 (3.86)
Clinical scales
State-Trait Anxiety Inventory - Y2	34.86 (8.23)	38.00 (6.26)	36.33 (10.46)	34.00 (5.57)	40.67 (11.14)	40.78 (12.03)
Beck Depression Inventory-II	9.29 (3.90)	6.67 (3.88)	7.83 (4.26)	7.00 (6.38)	10.13 (4.52)	4.38 (4.24)
Instrumental activities of daily living	6.71 (1.70)	7.17 (1.17)	5.00 (2.16)	5.80 (1.48)	5.75 (2.06)	5.83 (1.60)
Biological variables
BDNF	28.27 (10.51)	33.01 (9.30)	45.54 (18.02)	43.15 (14.67)	36.82 (21.98)	34.57 (19.73)

Exploratory analyses were performed to investigate whether baseline BDNF serum levels could be associated with the demographic or clinical characteristics of the sample. No significant correlations were observed with age (rho=-0.224, p = .484) or educational levels (rho=-0.331, p = .143). Similarly, baseline BDNF levels were not significantly associated with gender (U = 41.00, p = .456) or other clinical characteristics of the sample (MCI pathology: U = 48.00, p = .670; MCI type: U = 47.00, p = .913). Conversely, a significant positive correlation emerged at baseline between BDNF levels and scores at the phonological fluency test (rho = .509, p = .019).

Post-treatment cognitive changes

Concerning within-group changes between the two timepoints, participants in the experimental group (PA + SG) obtained significantly increased scores in the Prose memory test (z=-2.197, p = .028) with medium-large effect sizes (r=-0.59). Moreover, trends of improvements with medium effect sizes were observed in the delayed recall of the RAVLT test (z=-1.841, p = .066, r=-0.49), the copy of ROCF (z=-1.826, p = .068, r=-0.49), spatial span back (z=-1.947, p = .052, r=-0.52), and BDI-II scores (z=-1.897, p = .058, r=-0.55). The SG-only group showed significant improvements in Prose memory (z=-2.207, p = .027), with a large effect size (r=-0.64), and a trend in RAVLT delayed recall (z=-1.841, p = .066, r=-0.53). Conversely, the group who underwent standard cognitive rehabilitation obtained medium-large increased scores in semantic fluency (z=-2.033, p = .042, r=-0.48) and prose memory (z=-2.201, p = .028, r=-0.59), alongside a significant reduction of scores in the BDI-II scale (z=-2.383, p = .017, r=-0.60). Moreover, the SCR group showed trends of improvements in RAVLT immediate (z=-1.955, p = .051, r=-0.46) and delayed recall (z=-1.955, p = .053, r=-0.46).

When analyzing the presence of between-group differences in changes pre vs. post-treatment at cognitive and clinical levels, no significant differences were observed.

Association between changes in BDNF serum levels, demographic characteristics, and cognitive performances

Exploratory analyses were performed to investigate whether changes in BDNF serum levels may depend on the demographic or clinical characteristics of the sample. At the post-treatment acquisition, changes in BDNF levels after the treatment were not associated with age (rho=-.175, p = .587), gender (U = 31.00, p = .140), or clinical characteristics (MCI pathology: U = 47.00, p = .619; MCI type: U = 29.00, p = .149). Instead, a positive correlation was observed between increased BDNF levels and educational level (rho = .478, p = .028).

Concerning within-group changes between the two timepoints, participants in the experimental group showed increased BDNF serum levels (z=-2.201, p = .028) with a large effect size (r=-0.64). When analyzing the presence of between-group differences in changes pre vs. post-treatment, a trend toward significance emerged for changes in BDNF levels (X² = 5.570, p = .062). In the post-hoc analysis, no significant differences were observed after Bonferroni’s correction for multiple comparisons. However, a trend toward significance emerged when comparing changes in BDNF levels in the experimental group versus the SCR group (U = 9.00, p = .068, r=-0.55).

Concerning the association between changes at the biological and cognitive levels separately for each treatment group, positive correlations were observed between increased BDNF levels and scores at the semantic fluency (rho = 0.943, p = .005), prose memory (rho = .829, p = .042), and spatial span forward tests (rho = .820, p = .046) in the experimental group. Similarly, in the SG-only group, a positive association was found between scores at the spatial span forward test (rho = .943, p = .005) and BDNF levels, whereas no significant correlation emerged in the SCR group.

The present study aimed to explore the effects of a new tool for cognitive training (MindLenses Professional), which combines prismatic adaptation (PA) and serious games (SGs), in a cohort of patients with MCI. In particular, we wanted to investigate whether this new tool could improve cognitive performances, at least with effects similar to programs routinely used in clinical practice. Moreover, as previous studies suggested that PA may induce neuromodulatory effects on cortical excitability [33], [34], we also explored if changes at the behavioral level may also reflect changes in plasticity-related mechanisms assessed through BDNF serum levels.

Our findings show that MindLenses may be an effective tool for cognitive rehabilitation in patients with MCI on different cognitive domains, including verbal memory, spatial working memory, and visuospatial abilities. Additionally, a slight decrease in depressive symptoms was observed as well. Of note, the experimental group did not show statistical differences in cognitive changes compared to the group undergoing standard cognitive rehabilitation. This highlights that Mindlenses has similar effects to already validated programs that are used in clinical practice.

Moreover, patients in the experimental group did show a significant increase in BDNF serum levels post-treatment, which was not observed in the other two control groups. Notably, on a qualitative level, 100% of patients in the PA + SG group had increased BDNF levels after the treatment, whereas this percentage was lower in the other two groups (50% in SGs and 66.6% in SCR). This result indicates that the neuromodulation exerted by Mindlenses has a significant effect on plasticity. Our findings may also be consistent with a previous study reporting increased BDNF levels in patients with PD after stimulation of fronto-central regions using anodal tDCS [53].

The exploration of the association between BDNF levels and cognitive functioning shows interesting interactions. A positive correlation, for example, emerged at baseline with scores in the phonological fluency test, a proxy for mental flexibility. Of note, verbal fluency tests have been reported to have high sensitivity in detecting MCI [54].

Furthermore, in the experimental group, increased BDNF levels were associated with improved performances in language and memory functioning. Previous studies have linked BDNF expression with hippocampal functioning (see [55] for a review), suggesting that non-pharmacological interventions, especially aerobic ones, may increase BDNF peripheral levels and memory functioning while reducing depressive symptoms and hippocampal atrophy.

Lastly, a significant association emerged between BDNF changes and educational levels. In particular, education may be indicative of the overall cognitive reserve of the individual, although other factors, such as occupation or type of leisure activities, are involved [56]. Cognitive reserve refers to the set of mechanisms that the brain can enact “to cope with damage by using pre-existing cognitive processes or enlisting compensatory strategies” [56], and it has been linked to the flexibility of synaptic reorganization after brain damage [57]. Similarly, higher cognitive reserve has been identified as a protective factor against the development of dementia in patients with MCI [58], [59]. Hence, our findings suggest that patients with higher plastic resources could benefit more from the neuromodulation induced by prisms.

Nevertheless, this finding must be taken cautiously, especially when considering its possible clinical implications. For instance, one may speculate that higher BDNF levels post-treatment may translate into higher retention of the rehabilitation effects or slowing the disease progression, but no causal implication may be drawn, and future studies are needed to explore these hypotheses further.

A primary limitation of the present exploratory study was the limited sample size, which prevented us from conducting more refined analyses. For instance, due to a lack of power, we were not able to stratify patients based on the underlying MCI pathology (AD vs PD) or MCI type (amnestic/non-amnestic single/multi-domain). Future studies should replicate these results to validate the effect and trends observed on larger samples and analyze in depth how clinical factors may influence cognitive recovery after cognitive training or how they may mediate the effects of neuroplasticity mechanisms. Additional research is also needed to investigate possible PA effects on brain functional connectivity in MCI patients, which may underlie cognitive recovery. Lastly, concerning biological analyses, since the concentration of BNDF observed in the serum may depend on genetic polymorphisms [60], future studies should also clarify the role of genetic factors or the possible involvement of other biomarkers in the effects of cognitive rehabilitation for patients with MCI.

Captions

Table 1. Sociodemographic characteristics of the patients enrolled in the study. Mean and standard deviation, or number of patients and percentages, are reported. Results of the Kruskal-Wallis (H) or the Chi-square test (X2) are reported for continuous and categorical variables, respectively.

Table 2. Scores at the cognitive and clinical evaluation across the three groups. Means and standard deviations of corrected scores are reported for each test/clinical scale and BDNF serum levels (ng/mL).

Competing Interests: The authors declare the following financial interests / personal relationships which may be considered as potential competing interests: Massimiliano Oliveri and Agnese Di Garbo report a relationship with Restorative Neurotechnologies srl that includes: Massimiliano Oliveri involved as a shareholder and board member; Agnese Di Garbo involved as a shareholder and employee

Funding: This work was supported by the Italian Ministry of Health (Ricerca Corrente)

Ethics approval: This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Ethics Committee of Venice and IRCCS San Camillo Hospital (Venice, Italy) – Date: 29/06/2021, reference number 2021.12

Data availability statement:Data supporting the findings of this study are available on request from the corresponding author. Data are not publicly available due to privacy or ethical restrictions

Authors' contribution (CRediT):

Conceptualization: Danesin Laura, Oliveri Massimiliano, Burgio Francesca; Methodology: Danesin Laura, D’Este Giorgia, Barresi Rita, Semenza Carlo, Bottini Gabriella, Oliveri Massimiliano, Burgio Francesca; Investigation: Danesin Laura, D’Este Giorgia, Piazzalunga Elena; Formal analysis: Danesin Laura, D’Este Giorgia; Writing - original draft preparation: Danesin Laura, D’Este Giorgia, Barresi Rita, Burgio Francesca; Writing - review and editing: Danesin Laura, D’Este Giorgia, Barresi Rita, Piazzalunga Elena, Di Garbo Agnese, Semenza Carlo, Bottini Gabriella, Oliveri Massimiliano, Burgio Francesca; Resources: Di Garbo Agnese, Oliveri Massimiliano, Burgio Francesca; Supervision: Bottini Gabriella, Burgio Francesca

Acknowledgments: We thank Ms. Pazienza Elena and Dr. Spagna Mattia for their contribution to the initial analysis of biological samples.

S. Gauthier et al., “Mild cognitive impairment,” Lancet, vol. 367, no. 9518, pp. 1262–1270, 2006.
R. C. Petersen, “Mild cognitive impairment as a diagnostic entity,” pp. 183–194, 2004.
M. Overton, M. Pihlsgård, and S. Elmståhl, “Prevalence and incidence of mild cognitive impairment across subtypes, age, and sex,” Dement. Geriatr. Cogn. Disord., vol. 47, no. 4–6, pp. 219–232, 2019.
P. J. Visser and H. Brodaty, “MCI is not a clinically useful concept,” Int. Psychogeriatrics, vol. 18, no. 3, pp. 402–409, 2006.
G. Ravaglia et al., “Mild cognitive impairment: epidemiology and dementia risk in an elderly Italian population,” J. Am. Geriatr. Soc., vol. 56, no. 1, pp. 51–58, 2008.
R. C. Petersen, “Mild cognitive impairment,” Contin. lifelong Learn. Neurol., vol. 22, no. 2, pp. 404–418, 2016.
K. Glynn et al., “Clinical utility of mild cognitive impairment subtypes and number of impaired cognitive domains at predicting progression to dementia: A 20‐year retrospective study,” Int. J. Geriatr. Psychiatry, vol. 36, no. 1, pp. 31–37, 2021.
S. Belleville, C. Fouquet, C. Hudon, H. T. V. Zomahoun, J. Croteau, and C. for the E. I. of A. Disease-Quebec, “Neuropsychological measures that predict progression from mild cognitive impairment to Alzheimer’s type dementia in older adults: a systematic review and meta-analysis,” Neuropsychol. Rev., vol. 27, pp. 328–353, 2017.
F. Burgio et al., “Neurocognitive correlates of numerical abilities in Parkinson’s disease,” Neurol. Sci., vol. 43, no. 9, pp. 5313–5322, 2022, doi: 10.1007/s10072-022-06228-z.
S. Benavides-Varela et al., “Anatomical substrates and neurocognitive predictors of daily numerical abilities in mild cognitive impairment,” Cortex, 2015, doi: 10.1016/j.cortex.2015.05.031.
I. Corbo and M. Casagrande, “Higher-level executive functions in healthy elderly and mild cognitive impairment: a systematic review,” J. Clin. Med., vol. 11, no. 5, p. 1204, 2022.
N. L. J. Saunders and M. J. Summers, “Longitudinal deficits to attention, executive, and working memory in subtypes of mild cognitive impairment.,” Neuropsychology, vol. 25, no. 2, p. 237, 2011.
F. Burgio et al., “Facial emotion recognition in individuals with mild cognitive impairment: An exploratory study,” Cogn. Affect. Behav. Neurosci., pp. 1–16, 2024.
F. Burgio et al., “Financial and numerical abilities: patterns of dissociation in neurological and psychiatric diseases,” Neurol. Sci., pp. 1–9, 2024.
F. Burgio et al., “Predicting financial deficits from a standard neuropsychological assessment: preliminary evidence in Mild Cognitive Impairment,” Neurol. Sci., Apr. 2021, doi: 10.1007/s10072-021-05304-0.
L. Danesin, A. Giustiniani, G. Arcara, and F. Burgio, “Financial Decision-Making in Neurological Patients,” Brain Sci., vol. 12, no. 5, 2022, doi: 10.3390/brainsci12050529.
W. H. Organization, “Global strategy and action plan on ageing and health (2016–2020),” World Heal. Organ., 2016.
M. Huckans, L. Hutson, E. Twamley, A. Jak, J. Kaye, and D. Storzbach, “Efficacy of cognitive rehabilitation therapies for mild cognitive impairment (MCI) in older adults: working toward a theoretical model and evidence-based interventions,” Neuropsychol. Rev., vol. 23, pp. 63–80, 2013.
J. Reijnders, C. van Heugten, and M. van Boxtel, “Cognitive interventions in healthy older adults and people with mild cognitive impairment: a systematic review,” Ageing Res. Rev., vol. 12, no. 1, pp. 263–275, 2013.
A. Giustiniani, L. Maistrello, L. Danesin, E. Rigon, and F. Burgio, “Effects of cognitive rehabilitation in Parkinson disease: a meta-analysis,” Neurol. Sci., vol. 43, no. 4, pp. 2323–2337, 2022.
N. T. M. Hill, L. Mowszowski, S. L. Naismith, V. L. Chadwick, M. Valenzuela, and A. Lampit, “Computerized cognitive training in older adults with mild cognitive impairment or dementia: a systematic review and meta-analysis,” Am. J. Psychiatry, vol. 174, no. 4, pp. 329–340, 2017.
H. Coyle, V. Traynor, and N. Solowij, “Computerized and virtual reality cognitive training for individuals at high risk of cognitive decline: systematic review of the literature,” Am. J. Geriatr. Psychiatry, vol. 23, no. 4, pp. 335–359, 2015.
D. R. Michael and S. L. Chen, Serious games: Games that educate, train, and inform. Muska & Lipman/Premier-Trade, 2005.
R. Dörner, S. Göbel, W. Effelsberg, and J. Wiemeyer, Serious Games. Springer, 2016.
M. van der Kuil, A. Evers, A. Visser-Meily, and I. van der Ham, “Serious games in cognitive rehabilitation of spatial navigation impairment,” Ann. Phys. Rehabil. Med., vol. 61, p. e91, 2018.
C. Luque-Moreno, A. Oliva-Pascual-Vaca, P. Kiper, C. Rodriguez-Blanco, M. Agostini, and A. Turolla, “Virtual reality to assess and treat lower extremity disorders in post-stroke patients,” Methods Inf Med, vol. 55, no. 1, pp. 89–92, 2016.
G. Cirillo et al., “Neurobiological after-effects of non-invasive brain stimulation,” Brain Stimul., vol. 10, no. 1, pp. 1–18, 2017.
C. Miniussi and G. Vallar, “Brain stimulation and behavioural cognitive rehabilitation: a new tool for neurorehabilitation?,” Neuropsychol. Rehabil., vol. 21, no. 5, pp. 553–559, 2011.
C. Miniussi and P. M. Rossini, “Transcranial magnetic stimulation in cognitive rehabilitation,” Neuropsychol. Rehabil., vol. 21, no. 5, pp. 579–601, 2011.
A. Giustiniani, L. Maistrello, V. Mologni, L. Danesin, and F. Burgio, “TMS and tDCS as potential tools for the treatment of cognitive deficits in Parkinson’s disease: a meta-analysis,” Neurol. Sci., pp. 1–14, 2024.
T. Yang et al., “The cognitive effect of non-invasive brain stimulation combined with cognitive training in Alzheimer’s disease and mild cognitive impairment: a systematic review and meta-analysis,” Alzheimers. Res. Ther., vol. 16, no. 1, pp. 1–19, 2024.
T.-C. Cheng, S.-F. Huang, S.-Y. Wu, F.-G. Lin, W.-S. Lin, and P.-Y. Tsai, “Integration of virtual reality into transcranial magnetic stimulation improves cognitive function in patients with Parkinson’s disease with cognitive impairment: a proof-of-concept study,” J. Parkinsons. Dis., vol. 12, no. 2, pp. 723–736, 2022.
B. Magnani, C. Caltagirone, and M. Oliveri, “Prismatic adaptation as a novel tool to directionally modulate motor cortex excitability: evidence from paired-pulse TMS,” Brain Stimul., vol. 7, no. 4, pp. 573–579, 2014.
M. Bracco, G. R. Mangano, P. Turriziani, D. Smirni, and M. Oliveri, “Combining tDCS with prismatic adaptation for non-invasive neuromodulation of the motor cortex,” Neuropsychologia, vol. 101, pp. 30–38, 2017.
J. S. Paulsen, N. Butters, D. P. Salmon, W. C. Heindel, and M. R. Swenson, “Prism adaptation in Alzheimer’s and Huntington’s disease.,” Neuropsychology, vol. 7, no. 1, p. 73, 1993.
Y. Stern, R. Mayeux, A. Hermann, and J. Rosen, “Prism adaptation in Parkinson’s disease.,” J. Neurol. Neurosurg. Psychiatry, vol. 51, no. 12, pp. 1584–1587, 1988.
A. Swainson et al., “Slower rates of prism adaptation but intact aftereffects in patients with early to mid-stage Parkinson’s disease,” Neuropsychologia, vol. 189, p. 108681, 2023.
M. Oliveri et al., “A novel digital approach for post-stroke cognitive deficits: A pilot study,” Restor. Neurol. Neurosci., no. Preprint, pp. 1–11, 2023.
P. Komulainen et al., “BDNF is a novel marker of cognitive function in ageing women: the DR’s EXTRA Study,” Neurobiol. Learn. Mem., vol. 90, no. 4, pp. 596–603, 2008.
M. Miranda, J. F. Morici, M. B. Zanoni, and P. Bekinschtein, “Brain-derived neurotrophic factor: a key molecule for memory in the healthy and the pathological brain,” Front. Cell. Neurosci., vol. 13, p. 472800, 2019.
G. Leal, P. M. Afonso, I. L. Salazar, and C. B. Duarte, “Regulation of hippocampal synaptic plasticity by BDNF,” Brain Res., vol. 1621, pp. 82–101, 2015.
G. Weinstein et al., “Serum brain-derived neurotrophic factor and the risk for dementia: the Framingham Heart Study,” JAMA Neurol., vol. 71, no. 1, pp. 55–61, 2014.
C. M. Nicastri et al., “BDNF mediates improvement in cognitive performance after computerized cognitive training in healthy older adults,” Alzheimer’s Dement. Transl. Res. Clin. Interv., vol. 8, no. 1, p. e12337, 2022.
F. Angelucci et al., “A pilot study on the effect of cognitive training on BDNF serum levels in individuals with Parkinson’s disease ,” Front. Hum. Neurosci. , vol. 9, no. 4, p. 130, Dec. 2015, doi: 10.1007/s11065-017-9363-3.
F. Angelucci et al., “Alzheimer’s disease (AD) and Mild Cognitive Impairment (MCI) patients are characterized by increased BDNF serum levels,” Curr. Alzheimer Res., vol. 7, no. 1, pp. 15–20, 2010.
T. Casoli, C. Giuli, M. Balietti, B. Giorgetti, M. Solazzi, and P. Fattoretti, “Effect of cognitive training on the expression of brain-derived neurotrophic factor in lymphocytes of mild cognitive impairment patients,” Rejuvenation Res., vol. 17, no. 2, pp. 235–238, 2014.
A. Damirchi, F. Hosseini, and P. Babaei, “Mental training enhances cognitive function and BDNF more than either physical or combined training in elderly women with MCI: a small-scale study,” Am. J. Alzheimer’s Dis. Other Dementias®, vol. 33, no. 1, pp. 20–29, 2018.
I. Litvan et al., “MDS task force on mild cognitive impairment in Parkinson’s disease: Critical review of PD-MCI,” Movement Disorders. 2011. doi: 10.1002/mds.23823.
L. Danesin, M. Oliveri, C. Semenza, G. Bottini, F. Burgio, and A. Giustiniani, “Prism adaptation in patients with unilateral lesion of the parietal or cerebellar cortex: A pilot study on two single cases using a concurrent exposure procedure,” Neuropsychologia, vol. 184, p. 108557, 2023.
G. Conte, L. Quadrana, L. Zotti, A. Di Garbo, and M. Oliveri, “Prismatic adaptation coupled with cognitive training as novel treatment for developmental dyslexia: a randomized controlled trial,” Sci. Rep., vol. 14, no. 1, p. 7148, 2024, doi: 10.1038/s41598-024-57499-9.
R. Rosenthal, “Effect sizes: Pearson’s correlation, its display via the BESD, and alternative indices.,” 1991.
I. SPSS, “IBM SPSS statistics for windows, version 23.0. Armonk: IBM Corp.” 2015.
H. Hadoush, S. A. Banihani, H. Khalil, Y. Al-Qaisi, A. Al-Sharman, and M. Al-Jarrah, “Dopamine, BDNF and motor function postbilateral anodal transcranial direct current stimulation in Parkinson’s disease,” Neurodegener. Dis. Manag., vol. 8, no. 3, pp. 171–179, 2018.
M. McDonnell et al., “Verbal fluency as a screening tool for mild cognitive impairment,” Int. Psychogeriatrics, vol. 32, no. 9, pp. 1055–1062, 2020.
K. I. Erickson, D. L. Miller, and K. A. Roecklein, “The aging hippocampus: interactions between exercise, depression, and BDNF,” Neurosci., vol. 18, no. 1, pp. 82–97, 2012.
M. Nucci, D. Mapelli, and S. Mondini, “Cognitive Reserve Index questionnaire (CRIq): a new instrument for measuring cognitive reserve,” Aging Clin. Exp. Res., vol. 24, no. 3, pp. 218–226, 2012.
Y. Stern, “Cognitive reserve,” Neuropsychologia, vol. 47, no. 10, pp. 2015–2028, 2009.
R. F. Allegri et al., “Role of cognitive reserve in progression from mild cognitive impairment to dementia,” Dement. Neuropsychol., vol. 4, no. 1, pp. 28–34, 2010.
M. Poletti, M. Emre, and U. Bonuccelli, “Mild cognitive impairment and cognitive reserve in Parkinson’s disease,” Parkinsonism Relat. Disord., vol. 17, no. 8, pp. 579–586, 2011.
M. F. Egan et al., “The BDNF val66met polymorphism affects activity-dependent secretion of BDNF and human memory and hippocampal function,” Cell, vol. 112, no. 2, pp. 257–269, 2003.

The authors declare potential competing interests as follows: The authors declare the following financial interests / personal relationships which may be considered as potential competing interests: Massimiliano Oliveri and Agnese Di Garbo report a relationship with Restorative Neurotechnologies srl that includes: Massimiliano Oliveri involved as a shareholder and board member; Agnese Di Garbo involved as a shareholder and employee

Supplementary.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

Combining prismatic adaptation and digital cognitive training: preliminary evidence on cognitive and biological effects in patients with mild cognitive impairment

Status:

Version 1

Abstract

Introduction

Methods

Participants

Cognitive and clinical assessment

Intervention

Blood samples collection and serum isolation

Brian-Derived Neurotrophic Factor (BDNF) assay

Statistical analysis

Results

Sociodemographic and baseline characteristics of the sample

Post-treatment cognitive changes

Association between changes in BDNF serum levels, demographic characteristics, and cognitive performances

Discussion

Captions

Statements and Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1