Assessment of visual misperceptions in patients with Parkinson's disease using single and bistable percepts as testing tools
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.253587
Source of Support: None, Conflict of Interest: None
Keywords: Bistable percepts, cognitive tools, frontal assessment battery, frontal executive function, nonmotor symptoms, Parkinson's disease, trail making test, visual hallucinations, visual misperception
Visual perceptual abnormalities are reported in over one-half of the patients suffering from Parkinson's disease (PD), particularly in the later stages of the disease. Visual symptoms in PD typically progress along a clinical spectrum ranging from vivid dreams, visual misperception (VM) and hallucination (VH), to frank psychosis. VMs represent misinterpretation of visual stimuli caused by failure to successfully integrate stimuli physically presented, and VHs are perceptions in the absence of any stimulus. Patients with PD often do not report their VM and VH unless a direct enquiry is made. Selected cognitive testing batteries and novel behavioral approaches have been used by various authors to investigate VMs across a range of clinical conditions. Paradigms of single and bistable visual percepts have been used to detect visual dysfunction in patients with PD., Recently, the paradigm of bistable and single visual percepts was used to support the hypothesis that the mechanism of visual misperceptions in PD involves dysfunction of dorsal attentional networks of the brain. There is a need to further assess the utility of single and bistable percepts as tools for detecting visual misperception to better understand the underlying dysfunctions of the neuronal networks. This understanding will help in management of these visual perceptual disturbances, and in the future, likely stimulate ways to improve the quality of life of patients with PD. We conducted this study to determine the utility of single and bistable visual percepts presented to PD patients as tools to assess visual misperception.
This was a case-control study conducted at the outpatient Movement Disorder Clinic of our tertiary care neurology centre.
The study was duly approved by the Institutional Review Board and the Institute Ethics Committee. Thirty patients of Parkinson's disease aged ≥40 years and ≤65 years diagnosed as per the UK Brain Bank Criteria and educated at least up to 12th standard were selected. Thirty age- and sex-matched healthy volunteers educated at least up to class 12th standard were recruited as controls. Healthy volunteers included attendants accompanying the patients or staff members of our tertiary care centre. Patients with a score of ≤24 on Folstein Mini-Mental State examination (MMSE), visual acuity less than 6/6 even after correction by glasses, and those diagnosed with psychosis according to the International Classification of Diseases-10 diagnostic criteria for schizophrenia and schizoaffective disorders were excluded from the study. Patients with a history of migraine with aura, epilepsy, structural brain lesions, metabolic encephalopathy, and those on medicines likely to cause hallucinations (except antiparkinsonian drugs) were also excluded from the study. Written informed voluntary consent was taken from all the participants before inclusion. After collecting the demographic data, all the participants underwent detailed history taking and physical examination.
The severity and stage of PD was assessed by the examiner using Unified Parkinson Disease Rating Scale, Part-III (UPDRS- III), in “on-state.” UPDRS was supplemented by Modified Hoehn and Yahr (H and Y) staging of the disease.
Cognitive function assessment
Folstein MMSE, Frontal Assessment Battery (FAB) score, and Trail Making Tests (TMT-A and B) were used for assessing cognition. The MMSE is recommended by the Movement Disorder Society Task Force for level I testing to assess PD associated with a decreased global cognitive efficiency as well as for detecting impairment in more than one cognitive domain. The screening instrument has been previously used to characterize PD dementia in clinical trials., FAB takes approximately 10 minutes to be administered and consists of six subtests that explore neurocognitive processes related to the frontal lobes including conceptualization (similarities task), mental flexibility (phonological lexical fluency task), motor programming (Luria's motor series), sensitivity to interference (conflicting instructions task), inhibitory control (go-no go task), and environmental autonomy (evaluation of prehension behavior). Each subtest is scored between 0 and 3, and a composite score ranging between 0 and 18 indicates the presence and severity of executive dysfunction.
The TMT is a measure of attention, speed, and mental flexibility. It also tests spatial organization, visual pursuits, recall, and recognition. TMT-A requires the individual to draw lines to connect 25 encircled numbers distributed on a page. This tests visual scanning, numeric sequencing, and visuomotor speed. TMT-B is similar except that the person must alternate between numbers and letters; it is believed to be more difficult and takes longer to complete. TMT-B primarily reflects working memory and secondarily task-switching ability. Both parts A and B of TMT are timed and the score represents the amount of time required to complete the task. We also calculated the difference between TMT B and TMT A scores (TMT-B − A). TMT-B − A score provides a relatively pure indicator of executive control abilities by minimizing visuoperceptual and working memory demands measured by both the parts of TMT.
Assessment of visual misperceptions
All the participants were explained and trained about the visual percepts to be used for testing [Figure 1]. The study was conducted indoors in the same laboratory room under similar lighting conditions. Participants were allowed to wear spectacles as required. During the training, they were shown images and explained about single or bistable visual percept. Images from the training period were not presented during the test. Participants were shown a set of 40 monochromatic, black and white images on a computer screen through a powerpoint presentation. Each slide with an image was signalled by the appearance of a blank slide. Image in every slide represented either “single” or “bistable” percept (i.e. hidden images). The participants' responses were recorded at the end of having shown each image. After recording the response, the patient was shown the next image. They were told that they could take their time to recognize all the aspects of the image and were supposed to:
Participant responses were characterized as below, following the categorization used by Shine et al.
The same images were shown to all the participants in the study. After recording the responses, the total VMs including VMS, VMB, and missed images were calculated. Mean of the total visual misperceptions (VMm) for cases and controls was determined. VMS, VMB, and reporting of missed images were also separately analyzed for both the groups. A visual misperception error score (VMES) was calculated using the data for visual misperceptions in controls. A cut-off score for VM was defined using the 1.5 standard deviation (SD) value above the average VM score of the control group. Using this cut-off score, the cases with PD were split into those with an error score below the cut off value, “normal visual perception,” and those with an error score above the cut off value, “impaired visual perception.”
Statistical analysis was done using IBM Statistical Package for Social Sciences (SPSS) (IBM, New York) 21.0. Quantitative variables were compared using independent t-test/Mann–Whitney test (for nonparametric data) for cases and controls, and for those classified as having “normal visual perception” and “impaired visual perception” based on the cut-off value of VMES. Qualitative variables were compared using chi-square test/Fisher's exact test. Pearson correlation coefficient was used to find the association between VMs with other parameters. Linear regression was used to determine the predictability of various parameters for VMs. Binary logistic regression was used to find out the association of risk of VMES-impaired patients with various factors.
The mean age (±SD) of cases and controls were 59.23 (±5.11) and 59.27 (±5.11) years, respectively. Both the groups comprised 24 male (80%) and 6 female (20%) patients.
Disease severity and treatment characteristics of patients
The mean (±SD) durations of disease and treatment were 35.97 (±23.06) and 25.5 (±20.72) months, respectively. Twenty-nine of our patients were in H and Y stages 1–3. There was only one patient who had H and Y stage 4 disease. The mean UPDRS-III score of the cases was 20.37(±6.52). While 27 (90%) patients were on various antiparkinsonian drugs, only 3 patients with PD (10%) were drug naïve. Twenty-five of the patients were on l-dopa alone or l-dopa with other antiparkinsonian drugs. One patient was on pramipexole alone and another patient was on ropinirole along with trihexyphenidyl. The mean (±SD) of Levodopa Equivalent Daily Dose (LEDD) for the patients was 294.25 (±158.32) mg.
Comparison of cognitive function scores between cases and controls
The cases and controls did not have significant difference in their MMSE score (P = 0.678). The FAB score was significantly higher for controls in comparison to cases. The mean scores of TMT-A, TMT-B, TMT-B − A were significantly lower in the control group compared to cases [Table 1].
Comparison of visual misperception scores of cases and controls
PD patients had higher total visual misperception score and misperceptions of bistable percepts (VMB) compared to healthy controls [Table 1]. The number of misperceptions for monostable percepts (VMS) and missed images were similar in cases and controls. Twenty-six of the 30 patients were able to correctly identify monostable percepts. Nine patients were unable to identify both the images in a bistable percept or reported a new image for the same (missed image). Only one control participant was unable to identify the monostable percept image. The participant was able to identify the image as a monostable percept but incorrectly identified the single image of the monostable percept (VMS). Two control participants (6.67%) missed an image as they were unable to identify both the images of a bistable percept or reported a new image which was not present.
Relation between cognition and visual misperception
For cases, there was a significant positive correlation of VMs with the TMT-A and TMT-B scores implying that patients taking longer time to complete TMT A and B tests had increased frequency of VMs. An inverse correlation was also observed between MMSE and VMs, and between FAB scores and visual misperceptions, suggesting that VMs are less in patients with preserved cognition [Table 2]. Among controls, only TMT-A correlated positively with VMs (correlation coefficient: 0.765; P < 0.0001).
Regression analysis further showed that TMT-A, TMT-B, FAB scores were strong predictors of VMs. Variance of 37.2% and 34.3% in visual misperceptions was observed to be explained by TMT-A and TMT-B, respectively. One unit decrease in FAB and MMSE score was noted to increase VM score by 1.226 and 1.304, respectively. In the control group, TMT-A (coefficient of determination r2: 0.586; P < 0.005), was a significant predictor for VMs.
Relation between disease severity, treatment, and visual misperceptions
For cases, there was a significant positive correlation of VMs with disease duration, treatment duration, UPDRS-III, and H and Y stage [Table 2]. Regression analysis revealed that disease duration, treatment duration, UPDRS-III score, and H and Y stage were significant predictors of VM. LEDD and visual misperceptions did not show statistically significant correlation (r2: 0.072; P = 0.70).
Visual hallucinations in cases and controls
All the participants were specifically asked questions regarding any unusual visual experiences or hallucinations in the previous 3 months. An enquiry was also made for formed and unformed VHs. None of the participants in our study reported these complaints.
Comparison between subsets of cases based on VMES cut-off score
Using the VMES cut-off score calculated from the control group data, patients having VMES ≤6.75 were categorized as those with “normal visual perception function.” Patients with VMES >6.75 were grouped as those with “impaired visual perception function” [Table 3].
Twenty-two patients out of thirty (73.34%) had a mean score above the cut off (VMES >6.75) and 8 patients were below this level (VMES ≤6.75). The two subgroups of patients did not have any statistically significant difference with respect to age and gender. The subgroup with “impaired visual perception” had longer disease and treatment duration, higher mean UPDRS-III scores, and advanced modified H and Y stages compared to the patient subgroup with normal visual perception. No significant difference for LEDD was observed between the two subgroups.
There was no significant difference in the MMSE scores of the two subsets of patients (P = 0.199).
However, the two subsets differed significantly with respect to TMT-A, TMT-B, TMT-B − A, and FAB scores. There was a significant difference for identification of bistable percepts (VMB) between the two subsets. The mean VMB scores were 7.73 (±1.16) and 5.25 (±1.04) for those with impaired visual perception (VMES >6.75) and normal visual perception (VMES ≤6.75), respectively (P value <0.0005). VMS and missed images were not different for the two subsets of patients. Logistic regression analysis showed that the risk of increased VMs increases significantly with increase in disease duration, treatment duration, and scores on UPDRS, H and Y scale, TMTAs, TMTBs, TMT-B − A, and FAB.
A comprehensive understanding of associated factors that predict future development of hallucinations and misperceptions is indispensable and a prerequisite for the adequate management of PD patients. Our study demonstrated significantly greater number of VMs detected by using single and bistable visual percepts in patients with PD even in the absence of any visual complaints. Misperception for bistable percepts was significantly higher in patients, suggesting that they have greater difficulty when confronted with ambiguous precepts. Our study sample mostly comprised of patients with mild-to-moderately severe disease. The observations suggest that VMs are present in patients with PD even in early stages when patients may not have or report VHs. We observed that patients with longer disease duration were more likely to have VMs. Patients with higher VMES (>6.75) had significantly more severe motor impairment on the UPDRS score compared to those with normal VMES (≤6.75). Disease duration was observed to be a strong predictor for VMs on regression analysis, which is similar to the observations made in the earlier reports., Hallucinations were once considered to be a drug-induced phenomenon, leading to designations such as “dopaminomimetic psychosis” or “levodopa psychosis,” although anticholinergics could also induce analogous complications., We did not observe significant correlation between LEDD and visual misperceptions. Goetz et al., refuted the concept of dopaminergic psychosis. Historical descriptions of PD from the pre-levodopa era also suggest that hallucinations may be part of PD itself, especially in the context of late dementia, depression, or nonspecific encephalopathy.
Clinical studies have shown that cognitive impairment in PD patients is associated with the occurrence of VH., Several cognitive domains (executive functioning, visuospatial abilities, attention) have been shown to be impaired in nondemented PD patients with VH when compared to PD patients without VH., Deficiency in rapid attentional set shifting was reported as a specific predilection for VH in a recently proposed attentional network model of VHs in PD. To explore the role of attentional set shifting, all our patients performed the TMT-A and B. Patients with PD took significantly more time to complete the tasks. Patients with impaired visual perception (VMES >6.75) took significantly more time in performing TMT-A and B, underscoring the importance of impaired visuoperceptual abilities, attentional set shifting, and executive function in causing VMs and VHs. However, interestingly the TMT-A score of the patients compared to controls was not significantly different. Cubillo et al., conducted a study to clarify the cognitive mechanisms that underlie the TMT scores. The results suggested that TMT-A mainly requires visuoperceptual abilities; TMT-B primarily reflects working memory and secondarily task-switching ability. TMT-B − A minimizes visuoperceptual and working memory demands and possibly provides a relatively pure indicator of executive control abilities. We observed a significant difference in TMT-B − A scores of cases and controls. The patients with VMES above the upper cut-off limit for normal, calculated from the control group, had a higher TMT-B − A score compared to those with VMES ≤6.75. This finding is similar to other reports in the literature. Association between VMs and TMT-B and TMT-B − A could not be found in the control group. One possible explanation could be that for normal healthy subjects, the various mechanisms underlying visual misperceptions could be differently represented with a more robust contribution from visuoperceptive abilities rather than executive dysfunction, working memory, and task switching abilities. Further studies might be required in this regard.
We evaluated the cases and controls for frontal executive dysfunction using FAB and observed a significantly lower mean FAB score among cases compared to controls. Recent studies have demonstrated that nondemented PD patients with VHs have lower scores on frontal executive tasks than PD patients without hallucinations, which is most likely due to defective cholinergic circuitry. Our study supports the hypothesis that VH may be related to a specific dysfunction of prefrontal corticosubcortical circuits.,,,
Our study was a case-control study. Studies with a larger sample size, longitudinal design, and follow-up are required for further confirmation of the utility of the paradigm of monostable and bistable visual percepts.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Table 1], [Table 2], [Table 3]