A modified version of the interlocking finger test as a bedside screening test for visuospatial deficits and dementia in Parkinson's disease

Abstract Introduction Objective of this study was to examine if the Interlocking Finger Test (ILFT) is a suitable bedside screening test for visuospatial functions and/or dementia in Parkinson's disease (PD) patients aiming to facilitate the diagnosis of a dementia syndrome associated with posterior cortical and temporal lobe dysfunction according to the dual syndrome hypothesis (frontostriatal vs. posterior cortical cognitive impairment). Methods Forty‐seven PD patients were assessed with the ILFT and an extensive cognitive test battery. The ILFT was carried out in the original version as well as in three modified versions of the test including a fifth figure and/or a more complex rating system, leading to four different ILFT scores (named after the maximum achievable scoring result: ILFT 4, ILFT 5, ILFT 12, and ILFT 15). We conducted a correlation analysis to reveal associations between the ILFT scores and cognitive as well as motor impairments. Receiver operating curve (ROC) analyses were calculated to evaluate the ability of the ILFT scores to predict visuospatial impairments and dementia. Results ILFT scores correlated significantly with global cognition, visuospatial functions, memory, attention, and age (p < .0125) but not with executive functions, language, education, depression, and motor impairment. The ROC analyses revealed ILFT 15 as best predictor for visuospatial deficits and dementia with an area under the curve of .82 and .88, respectively. Conclusion The ILFT is suitable for detecting symptoms of the posterior cortical degeneration syndrome according to the dual syndrome hypothesis. We recommend the use of the modified test version ILFT 15.


INTRODUCTION
Cognitive impairment is a common nonmotor symptom in Parkinson's disease (PD). The prevalence of mild cognitive impairment in PD is 40% (Baiano et al., 2020) and the prevalence of dementia in PD is 24-31% (Aarsland et al., 2005). Besides executive dysfunctions and impairments in memory and attention, PD patients often show visuospatial deficits (Aarsland et al., 2010;Curtis et al., 2019;Fernandez-Baizan et al., 2020;Muslimovic et al., 2005) including impairments in visuospatial perception, orientation, or construction. Deficits in visuospatial abilities increase in the course of the disease (Muslimović et al., 2007) and initially more severe impairments are predictive for the progression of cognitive impairment in PD (Stepkina et al., 2010). Furthermore, visuospatial deficits not only discriminate patients with mild cognitive impairment from those with dementia (Biundo et al., 2014) but are also of prognostic importance regarding the later conversion to development of PD dementia. It was shown that early deficits in posterior cortically based cognitive (e.g., visuoconstructive) tasks that are associated with Lewy body deposition in these areas lead to subsequent dementia while cognitive deficits that are associated with a dopamine modulated frontal-striatal network dysfunction (e.g., executive functions) do not (Williams-Gray et al., 2007;Williams-Gray et al., 2009). Based on these and other study results on longitudinal cognitive impairment patterns, Kehagia et al. (2013) proposed the dual syndrome hypothesis which differentiates between two different, partly overlapping syndromes: (1) a dopamine modulated frontal-striatal network dysfunction in nondemented PD patients which is present at early disease stages and leads to executive and working memory impairments and (2) a dementia syndrome associated with more posterior cortical degeneration, temporal lobe dysfunction, and cholinergic loss characterized by prodromal visuospatial and semantic fluency deficits.
Distinguishing between these two cognitive syndromes in PD at an early disease stage is important to identify specific cognitive risk  Figure   Test) and showed that the ILFT can predict parietal lobe dysfunction with a good sensitivity and a moderate specificity in a heterogeneous patient group (Moo et al., 2003). In PD patients, the ILFT correlated significantly with visuospatial functions (Clock Drawing Test) as well as with other cognitive domains (e.g., executive functions, memory), and was able to discriminate patients with dementia from those without it with a good specificity and a moderate sensitivity (Souza et al., 2016). In this study, we examined whether the ILFT is a suitable bedside screening test for visuospatial functions and/or dementia aiming to facilitate the diagnosis of the posterior cortical cognitive syndrome in PD. Furthermore, we examined if the predictive ability of the ILFT can be improved by modifying the rating system and adding an additional figure as the original test version has a small scoring range of only 0 to 4 points.

Patients
Forty-seven patients with PD diagnosed according to the UK Parkinson's Disease Society Brain Bank criteria (Hughes et al., 1992) were included in the analyses. Exclusion criteria were any neurological disorder other than PD and deep brain stimulation. The study was approved by the local ethics committee. All procedures contributing to this work comply with the ethical standards of the relevant national and institutional committees on human experimentation and with the Helsinki Declaration of 1975, as revised in 2013. All participants gave their informed consent to participate in the study in written form.

Interlocking Finger Test
All patients executed the ILFT. In this test, the investigator demonstrates consecutively nonsymbolic bimanual gestures, and the partic-

Cognitive functioning and clinical data
We used the Mini Mental State Examination (MMSE; Folstein et al., 1975)  Based on these tests, the patients were classified into PD with and without visuospatial impairments, and with and without dementia, respectively. Visuospatial impairment was diagnosed if a patient scored ≥ 1.5 standard deviations below normative data in at least one test assigned to the visuospatial domain. Dementia was diagnosed according to the Movement Disorder Society Task Force criteria (Emre et al., 2007) including (1) cognitive test scores ≥ 1.5 standard deviations below normative data in at least two different cognitive domains, (2) cognitive decline reported by the patient or a relative, and (3) significant impairment in activities of daily living.
Disease severity was rated with the Unified Parkinson's disease rating scale motor score (UPDRS III; Fahn et al., 1987). The short form of the Geriatric Depression Scale (GDS; Sheikh & Yesavage, 1986) was used to assess depression. Levodopa equivalent daily dose (LEDD) was calculated according to Tomlinson et al. (2010).  (Youden, 1950) was computed, defined as the sum of sensitivity and specificity minus 1. To determine the interrater reliability of the ILFT scores, the ILFT rating was carried out by two independent raters (one psychologist and one physician) and Kendall's tau-b coefficient was calculated.

RESULTS
Twenty-eight of the PD patients were men and 19 women. the ROC curve analyses can be seen in Table 2 and Figure 1. Interrater reliability ranged from τ = .755 to .891 for the four ILFT scores.

DISCUSSION
We found significant correlations between ILFT scores and global cognition, visuospatial functions, memory, attention, and age but not between ILFT scores and executive functions, language, education, depression, and disease related variables such as disease duration and LEDD. Remarkably, the ILFT did not reflect motor impairment, given the lack of significant correlations between ILFT and motor scores. This specific property classified the ILFT as a cognitive rather than a motor task. The ROC analyses revealed ILFT 12 and ILFT 15 as best predictors for visuospatial deficits and dementia with good negative and moderate positive prediction values.
The correlation analysis showed that the ILFT highly correlated with global cognition, visuospatial functions, memory, and attention. These results are in line with a previous study in which significant correlations with tests of the same neuropsychological domains were found (Souza et al., 2016 (Souza et al., 2016) and AD patients (Sanin & Benke, 2017). The fact that motor impairments did not affect the ILFT result and its easy implementation with a test duration of a few minutes maximum and no need for test material or equipment (e.g., table, pencil, watch) argue for the ILFT as an appropriate screening instrument.
Furthermore, the test showed good interrater reliability.
There were significant improvements in the prediction of visuospatial deficits and dementia when using the modified versions of the ILFT.
For predicting visuospatial deficits, ILFT 15 turned out to be the best predictor according to AUC and Youden index. Therefore, the ILFT 15 is recommended as bedside screening test for the diagnosis of visuospatial deficits. At a result of 12 points or lower, an extensive neuropsychological testing should be carried out. Regarding the prediction of PD dementia, ILFT 15 is the best predictor with a slightly higher Youden index as the ILFT 12 which is why we here also recommend the modified version ILFT 15 that contains an additional figure and a more complex rating system than the original. Cut-off for prediction of PD dementia is 10.5, meaning that a score of 10 or lower entails further diagnostic procedure.
A limitation of the study is that visuospatial functions are a multidimensional construct which could not be completely represented in the cognitive test battery. However, even in a formal neuropsychological testing are often isolated tests used that are not covering the entire spectrum of visuospatial deficits. Furthermore, long-term studies examining if patients with deficits in the ILFT will develop a dementia syndrome in the course of the disease are necessary to verify our results.
In summary, the ILFT significantly correlated with visuospatial functions, memory, attention, global cognitive abilities, and age indicating that it is suitable for detecting symptoms of the posterior cortical degeneration syndrome according to the dual syndrome hypothesis.
We recommend the use of ILFT 15 with cut-off scores of 12.5 for predicting visuospatial deficits or 10.5 for predicting PD dementia, respectively.

ACKNOWLEDGMENTS
For this study, part of the data was generated within the LAND-SCAPE study. The LANDSCAPE study is part of the Competence Network Degenerative Dementias (KNDD), which was funded by the German Federal Ministry of Education and Research (project number 01GI1008C).

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request.