Neurologic signs in relation to cognitive function in subcortical ischemic vascular dementia: a CREDOS (Clinical Research Center for Dementia of South Korea) study

Diagnostic performance of deep learning-based automatic white matter hyperintensity segmentation for classification of the Fazekas scale and differentiation of subcortical vascular dementia

Purpose

To validate the diagnostic performance of commercially available, deep learning-based automatic white matter hyperintensity (WMH) segmentation algorithm for classifying the grades of the Fazekas scale and differentiating subcortical vascular dementia.

Methods

This retrospective, observational, single-institution study investigated the diagnostic performance of a deep learning-based automatic WMH volume segmentation to classify the grades of the Fazekas scale and differentiate subcortical vascular dementia. The VUNO Med-DeepBrain was used for the WMH segmentation system. The system for segmentation of WMH was designed with convolutional neural networks, in which the input image was comprised of a pre-processed axial FLAIR image, and the output was a segmented WMH mask and its volume. Patients presented with memory complaint between March 2017 and June 2018 were included and were split into training (March 2017–March 2018, n = 596) and internal validation test set (April 2018–June 2018, n = 204).

Results

Optimal cut-off values to categorize WMH volume as normal vs. mild/moderate/severe, normal/mild vs. moderate/severe, and normal/mild/moderate vs. severe were 3.4 mL, 9.6 mL, and 17.1 mL, respectively, and the AUC were 0.921, 0.956 and 0.960, respectively. When differentiating normal/mild vs. moderate/severe using WMH volume in the test set, sensitivity, specificity, and accuracy were 96.4%, 89.9%, and 91.7%, respectively. For distinguishing subcortical vascular dementia from others using WMH volume, sensitivity, specificity, and accuracy were 83.3%, 84.3%, and 84.3%, respectively.

Conclusion

Deep learning-based automatic WMH segmentation may be an accurate and promising method for classifying the grades of the Fazekas scale and differentiating subcortical vascular dementia.

The Impact of Disease Registries on Advancing Knowledge and Understanding of Dementia Globally

To help address the increasing challenges related to the provision of dementia care, dementia registries have emerged around the world as important tools to gain insights and a better understanding of the disease process. Dementia registries provide a valuable source of standardized data collected from a large number of patients. This review explores the published research relating to different dementia registries around the world and discusses how these registries have improved our knowledge and understanding of the incidence, prevalence, risk factors, mortality, diagnosis, and management of dementia. A number of the best-known dementia registries with high research output including SveDem, NACC, ReDeGi, CREDOS and PRODEM were selected to study the publication output based on their data, investigate the key findings of these registry-based studies. Registries data contributed to understanding many aspects of the disease including disease prevalence in specific areas, patient characteristics and how they differ in populations, mortality risks, as well as the disease risk factors. Registries data impacted the quality of patients’ lives through determining the best treatment strategy for a patient based on previous patient outcomes. In conclusion, registries have significantly advanced scientific knowledge and understanding of dementia and impacted policy, clinical practice care delivery.

Hospitalizations and Mortality of Individuals with Dementia: Evidence from Czech National Registers

BACKGROUND

Facing an increasing prevalence of dementia, the Czech Republic is developing a new nationwide strategy for the management and prevention of dementia. Lack of evidence about characteristics of individuals with dementia in the country is a major obstacle.

OBJECTIVE

The study aimed to 1) characterize individuals with dementia, 2) compare their mortality with the general population, and 3) analyze differences in survival between different dementia disorders.

METHODS

The study capitalizes on two nationwide registers in the Czech Republic, from which information about individuals who were hospitalized with dementia or died from it between 1994 and 2014 was retrieved. Standardized intensity of hospitalizations was calculated for each year, mortality was studied using standardized mortality ratio, life-tables, Kaplan-Mayer curves, and Cox proportional hazard models.

RESULTS

Standardized intensity of hospitalizations for dementia increased more than 3 times from 1994 to 2014. Standardized mortality ratio was 3.03 (95% confidence interval 2.97-3.08). One-year survival rate was 45% and five-year survival rate 16%. Vascular dementia was the most common type of dementia disorders and was associated with higher hazard of death than Alzheimer's disease, even after adjusting for sociodemographic and clinical covariates (hazard ratio 1.04; 95% confidence interval 1.02-1.05).

CONCLUSION

The study provides estimates on demographic characteristics and mortality of the Czech hospitalized dementia population, which have not been so far available and which are unique also in the context of the entire region of Central and Eastern Europe.

Prediction of cognitive impairment via deep learning trained with multi-center neuropsychological test data

Background

Neuropsychological tests (NPTs) are important tools for informing diagnoses of cognitive impairment (CI). However, interpreting NPTs requires specialists and is thus time-consuming. To streamline the application of NPTs in clinical settings, we developed and evaluated the accuracy of a machine learning algorithm using multi-center NPT data.

Methods

Multi-center data were obtained from 14,926 formal neuropsychological assessments (Seoul Neuropsychological Screening Battery), which were classified into normal cognition (NC), mild cognitive impairment (MCI) and Alzheimer’s disease dementia (ADD). We trained a machine learning model with artificial neural network algorithm using TensorFlow (https://www.tensorflow.org) to distinguish cognitive state with the 46-variable data and measured prediction accuracies from 10 randomly selected datasets. The features of the NPT were listed in order of their contribution to the outcome using Recursive Feature Elimination.

Results

The ten times mean accuracies of identifying CI (MCI and ADD) achieved by 96.66 ± 0.52% of the balanced dataset and 97.23 ± 0.32% of the clinic-based dataset, and the accuracies for predicting cognitive states (NC, MCI or ADD) were 95.49 ± 0.53 and 96.34 ± 1.03%. The sensitivity to the detection CI and MCI in the balanced dataset were 96.0 and 96.0%, and the specificity were 96.8 and 97.4%, respectively. The ‘time orientation’ and ‘3-word recall’ score of MMSE were highly ranked features in predicting CI and cognitive state. The twelve features reduced from 46 variable of NPTs with age and education had contributed to more than 90% accuracy in predicting cognitive impairment.

Conclusions

The machine learning algorithm for NPTs has suggested potential use as a reference in differentiating cognitive impairment in the clinical setting.

A mouse model of subcortical vascular dementia reflecting degeneration of cerebral white matter and microcirculation

Subcortical vascular dementia(SVaD) is associated with white matter damage, lacunar infarction, and degeneration of cerebral microcirculation. Currently available mouse models can mimic only partial aspects of human SVaD features. Here, we combined bilateral common carotid artery stenosis (BCAS) with a hyperlipidaemia model in order to develop a mouse model of SVaD; 10- to 12-week-old apolipoprotein E (ApoE)-deficient or wild-type C57BL/6J mice were subjected to sham operation or chronic cerebral hypoperfusion with BCAS using micro-coils. Behavioural performance (locomotion, spatial working memory, and recognition memory), histopathological findings (white matter damage, microinfarctions, astrogliosis), and cerebral microcirculation (microvascular density and blood-brain barrier (BBB) integrity) were investigated. ApoE-deficient mice subjected to BCAS showed impaired locomotion, spatial working memory, and recognition memory. They also showed white matter damage, multiple microinfarctions, astrogliosis, reduction in microvascular density, and BBB breakdown. The combination of chronic cerebral hypoperfusion and ApoE deficiency induced cognitive decline and cerebrovascular pathology, including white matter damage, multiple microinfarctions, and degeneration of cerebral microcirculation. Together, these features are all compatible with those of patients with SVaD. Thus, the proposed animal model is plausible for investigating SVaD pathophysiology and for application in preclinical drug studies.

Prediction Model of Conversion to Dementia Risk in Subjects with Amnestic Mild Cognitive Impairment: A Longitudinal, Multi-Center Clinic-Based Study

BACKGROUND

Patients with amnestic mild cognitive impairment (aMCI) have an increased risk of dementia. However, conversion rate varies. Therefore, predicting the dementia conversion in these patients is important.

OBJECTIVE

We aimed to develop a nomogram to predict dementia conversion in aMCI subjects using neuropsychological profiles.

METHODS

A total of 338 aMCI patients from two hospital-based cohorts were used in analysis. All patients were classified into 1) verbal, visual, or both, 2) early or late, and 3) single or multiple-domain aMCI according to the modality, severity of memory dysfunction, and multiplicity of involved cognitive domains, respectively. Patients were followed up, and conversion to dementia within 3 years was defined as the primary outcome. Our patients were divided into a training data set and a validation data set. The associations of potential covariates with outcome were tested, and nomogram was constructed by logistic regression model. We also developed another model with APOE data, which included 242 patients.

RESULTS

In logistic regression models, both modalities compared with visual only (OR 4.44, 95% CI 1.83-10.75, p = 0.001), late compared to early (OR 2.59, 95% CI 1.17-5.72, p = 0.019), and multiple compared to single domain (OR 3.51, 95% CI 1.62-7.60, p = 0.002) aMCI were significantly associated with dementia conversion within 3 years. A nomogram incorporating these clinical variables was constructed on the training data set and validated on the validation data set. Both nomograms with and without APOE data showed good prediction performance (c-statistics ≥ 0.75).

CONCLUSIONS

This study showed that several neuropsychological profiles of aMCI are significantly associated with imminent dementia conversion, and a nomogram incorporating these clinical subtypes is simple and useful to help to predict disease progression.

Update on Vascular Cognitive Impairment Associated with Subcortical Small-Vessel Disease

Subcortical small-vessel disease (SSVD) is a disorder well characterized from the clinical, imaging, and neuropathological viewpoints. SSVD is considered the most prevalent ischemic brain disorder, increasing in frequency with age. Vascular risk factors include hypertension, diabetes, hyperlipidemia, elevated homocysteine, and obstructive sleep apnea. Ischemic white matter lesions are the hallmark of SSVD; other pathological lesions include arteriolosclerosis, dilatation of perivascular spaces, venous collagenosis, cerebral amyloid angiopathy, microbleeds, microinfarcts, lacunes, and large infarcts. The pathogenesis of SSVD is incompletely understood but includes endothelial changes and blood-brain barrier alterations involving metalloproteinases, vascular endothelial growth factors, angiotensin II, mindin/spondin, and the mammalian target of rapamycin pathway. Metabolic and genetic conditions may also play a role but hitherto there are few conclusive studies. Clinical diagnosis of SSVD includes early executive dysfunction manifested by impaired capacity to use complex information, to formulate strategies, and to exercise self-control. In comparison with Alzheimer's disease (AD), patients with SSVD show less pronounced episodic memory deficits. Brain imaging has advanced substantially the diagnostic tools for SSVD. With the exception of cortical microinfarcts, all other lesions are well visualized with MRI. Diagnostic biomarkers that separate AD from SSVD include reduction of cerebrospinal fluid amyloid-β (Aβ)42 and of the ratio Aβ42/Aβ40 often with increased total tau levels. However, better markers of small-vessel function of intracerebral blood vessels are needed. The treatment of SSVD remains unsatisfactory other than control of vascular risk factors. There is an urgent need of finding targets to slow down and potentially halt the progression of this prevalent, but often unrecognized, disorder.

Age-Specific Cutoff Scores on a T1-Weighted Axial Medial Temporal-Lobe Atrophy Visual Rating Scale in Alzheimer's Disease Using Clinical Research Center for Dementia of South Korea Data

BACKGROUND AND PURPOSE

Visual assessment of medial temporal-lobe atrophy (MTA) has been quick, reliable, and easy to apply in routine clinical practice. However, one of the limitations in visual assessments of MTA is the lack of widely accepted age-adjusted norms and cutoff scores for MTA for a diagnosis of Alzheimer's disease (AD). This study aimed to determine the optimal cutoff score on a T1-weighted axial MTA Visual Rating Scale (VRS) for differentiating patients with AD from cognitively normal elderly people.

METHODS

The 3,430 recruited subjects comprising 1,427 with no cognitive impairment (NC) and 2003 AD patients were divided into age ranges of 50-59, 60-69, 70-79, and 80-89 years. Of these, 446 participants (218 in the NC group and 228 in the AD group) were chosen by random sampling for inclusion in this study. Each decade age group included 57 individuals, with the exception of 47 subjects being included in the 80- to 89-year NC group. The scores on the T1-weighted axial MTA VRS were graded by two neurologists. The cutoff values were evaluated from the area under the receiver operating characteristic curve.

RESULTS

The optimal axial MTA VRS cutoff score from discriminating AD from NC increased with age: it was ≥as ≥1, ≥2, and ≥3 in subjects aged 50-59, 60-69, 70-79, and 80-89 years, respectively (all p<0.001).

CONCLUSIONS

These results show that the optimal cutoff score on the axial MTA VRS for diagnosing of AD differed according to the decade age group. This information could be of practical usefulness in the clinical setting.

Group- and Home-Based Cognitive Intervention for Patients with Mild Cognitive Impairment: A Randomized Controlled Trial

BACKGROUND

We examined the efficacy of group-based cognitive intervention (GCI) and home-based cognitive intervention (HCI) in amnestic mild cognitive impairment (aMCI) and intervention effects on serum brain-derived neurotrophic factor (BDNF).

METHODS

In this randomized and rater-blinded trial, 293 patients with aMCI from 18 nationwide hospitals were randomized: 96 to the GCI group, 98 to the HCI group and 99 to the control group. For 12 weeks, subjects receiving GCI participated twice per week in group sessions led by trained instructors, and those receiving HCI completed homework materials 5 days per week. They were assessed at baseline, postintervention (PI) and at the 6-month follow-up after the intervention. The primary endpoint was the change from baseline to PI in the modified Alzheimer's Disease Assessment Scale-cognitive subscale (ADAS-Cog).

RESULTS

In comparison to the controls (a 0.8-point decrease), the subjects receiving GCI (a 2.3-point decrease, p = 0.01) or HCI (a 2.5-point decrease, p = 0.02) showed significant improvements in the modified ADAS-Cog at PI, respectively. By the 6-month follow-up, those receiving GCI or HCI had better scores in the modified ADAS-Cog than the controls. The changes in BDNF levels significantly correlated with the changes in the modified ADAS-Cog in the GCI (r = -0.29, p = 0.02 at PI) and HCI (r = -0.27, p = 0.03 at 6-month follow-up) groups, respectively.

CONCLUSIONS

The GCI and HCI resulted in cognitive improvements in aMCI. An enhanced brain plasticity may be a component of the mechanism underpinning the cognitive improvements associated with the cognitive interventions.

Macular Ganglion Cell -Inner Plexiform Layer Thickness Is Associated with Clinical Progression in Mild Cognitive Impairment and Alzheimers Disease

PURPOSE

We investigated the association of the macular ganglion cell-inner plexiform layer (GCIPL) and peripapillary retinal nerve fiber layer (RNFL) thicknesses with disease progression in mild cognitive impairment (MCI) and Alzheimer's disease (AD).

METHODS

We recruited 42 patients with AD, 26 with MCI, and 66 normal elderly controls. The thicknesses of the RNFL and GCIPL were measured via spectral-domain optic coherent tomography in all participants at baseline. The patients with MCI or AD underwent clinical and neuropsychological tests at baseline and once every year thereafter for 2 years.

RESULTS

The Clinical Dementia Rating scale-Sum of Boxes (CDR-SB) score exhibited significant negative relationships with the average GCIPL thickness (β = -0.15, p < 0.05) and the GCIPL thickness in the superotemporal, superonasal, and inferonasal sectors. The composite memory score exhibited significant positive associations with the average GCIPL thickness and the GCIPL thickness in the superotemporal, inferonasal, and inferotemporal sectors. The temporal RNFL thickness, the average and minimum GCIPL thicknesses, and the GCIPL thickness in the inferonasal, inferior, and inferotemporal sectors at baseline were significantly reduced in MCI patients who were converted to AD compared to stable MCI patients. The change of CDR-SB from baseline to 2 years exhibited significant negative associations with the average (β = -0.150, p = 0.006) and minimum GCIPL thicknesses as well as GCIPL thickness in the superotemporal, superior, superonasal, and inferonasal sectors at baseline.

CONCLUSIONS

Our data suggest that macular GCIPL thickness represents a promising biomarker for monitoring the progression of MCI and AD.

Chronic pain in "probable" vascular dementia: preliminary findings

BACKGROUND

In a previous study, the levels of pain reported by patients with "possible" vascular dementia (VaD) were higher than those reported by older individuals without dementia.

OBJECTIVE

To examine experienced pain in patients with "probable" VaD, confirmed by brain imaging.

STUDY DESIGN

Observational, cross sectional.

SETTING

Nursing home.

METHODS

The participants were 20 nursing home residents (14 females, 6 males) who met the NINDS-AIREN criteria for "probable" VaD and 22 nursing home residents with a normal mental status (18 females, 4 males). The patients were in a mild to moderate stage of dementia. All of the participants were suffering from arthritis/arthrosis or osteoporosis. Global cognitive functioning was measured by the Mini-Mental State Examination. Pain was assessed by the Coloured Analogue Scale (CAS: original and modified version) and the Faces Pain Scale. The Geriatric Depression Scale and the Symptom Checklist-90 were used to assess mood.

RESULTS

The main finding was that, after controlling for mood, the pain levels indicated by patients with "probable" VaD (M = 102.32; standard deviation [SD] = 53.42) were significantly higher than those indicated by the control group (M = 59.17; SD = 38.75), only according to the CAS modified version (F[1,29]) = 5.62, P = 0.01, η2  = 0.16).

CONCLUSION

As VaD patients may experience greater pain than controls, it is essential for prescribers to be aware of the presence of this neuropathology if these patients are to receive adequate treatment.

Recent updates on subcortical ischemic vascular dementia

Vascular dementia (VaD) is a history-laden disease entity that dates back to the 19th century when arteriosclerotic brain atrophy due to hardening of the arteries was perceived as the major cause of senile dementia. Its existence had been overshadowed by the emergence of Alzheimer's disease (AD) in the past century and research on AD dominated the field of dementia. Interest in VaD has been revived in recent years as vascular lesions have been shown to make great contributions to the development of dementia, particularly in the elderly. VaD has now evolved into the concept of vascular cognitive impairment (VCI), which encompasses not only VaD but also AD with cerebrovascular disorder and VCI with no dementia. The concept of VCI is intended to maximize the therapeutic potential in dementia management because the vascular component may be amenable to therapeutic intervention particularly in the early stages of cognitive impairment. Subcortical ischemic vascular dementia (SIVD) is pathologically driven by severe stenosis and the occlusion of small vessels that culminate into white matter ischemia and multiple lacunar infarctions in the subcortical structures. The relatively slow progression of symptoms and clinical manifestations associated with cholinergic deficits often make the differentiation of SIVD from AD difficult. The recent development of in vivo amyloid imaging enabled further pathological breakdown of SIVD into pure SIVD and mixed dementia with subcortical ischemia based on the absence or existence of amyloid pathology in the brain. In this article, the authors reviewed the emerging concepts of VaD/VCI and the clinical manifestations, biomarkers, treatments, and preclinical models of SIVD based on the pathophysiologic mechanisms of the disease.