Construction of machine learning models for recognizing comorbid anxiety in epilepsy patients based on their clinical and quantitative EEG features.

BACKGROUND: This study aimed to construct prediction models for the recognizing of anxiety disorders (AD) in patients with epilepsy (PWEs) by combining clinical features with quantitative electroencephalogram (qEEG) features and using machine learning (ML). METHODS: Nineteen clinical features and 20-min resting-state EEG were collected from 71 PWEs comorbid with AD and another 60 PWEs without AD who met the inclusion-exclusion criteria of this study. The EEG were preprocessed and 684 Phase Locking Value (PLV) and 76 Lempel-Ziv Complexity (LZC) features on four bands were extracted. The Fisher score method was used to rank all the derived features. We constructed four models for recognizing AD in PWEs, whether PWEs based on different combinations of features using eXtreme gradient boosting (XGboost) and evaluated these models using the five-fold cross-validation method. RESULTS: The prediction model constructed by combining the clinical, PLV, and LZC features showed the best performance, with an accuracy of 96.18%, precision of 94.29%, sensitivity of 98.33%, F1-score of 96.06%, and Area Under the Curve (AUC) of 0.96. The Fisher score ranking results displayed that the top ten features were depression, educational attainment, α_P3LZC, α_T6-PzPLV, α_F7LZC, ß_Fp2-O1PLV, θ_T4-CzPLV, θ_F7-PzPLV, α_Fp2LZC, and θ_T4-PzPLV. CONCLUSIONS: The model, constructed by combining the clinical and qEEG features PLV and LZC, efficiently identified the presence of AD comorbidity in PWEs and might have the potential to complement the clinical diagnosis. Our findings suggest that LZC features in the α band and PLV features in Fp2-O1 may be potential biomarkers for diagnosing AD in PWEs.

Associations Between Plasma Orexin-A Level and Constipation in Cognitive Impairment.

BACKGROUND: Constipation is a common symptom in dementia, and the cause is controversial. Rare clinical studies focused on plasma orexin-A levels and constipation in dementia. OBJECTIVE: To evaluate the associations between orexin-A and constipation in patients with cognitive impairment. METHODS: A total of 21 patients with mild cognitive impairment (MCI), 142 with Alzheimer's disease (AD), and 57 with Lewy body dementia (LBD) were conducted. Besides informant-based history, neurological examinations or neuropsychological assessments, plasma levels of orexin-A, and constipation were assessed. The associations between orexin-A and constipation were evaluated by logistic regression models. RESULTS: There were 47/220 (21.36%) cognitive impairment patients having constipation, and the proportion of constipation in LBD (61.40%) was significantly higher than AD (5.63%) and MCI (19.05%). No significant age or sex differences in the prevalence of constipation were found in the MCI, AD, and LBD groups. We found the cognitive impairment patients with constipation had lower levels of plasma orexin-A [1.00 (0.86, 1.28) versus 1.29 (1.01, 1.50) ng/ml, p < 0.001] than those without. And the plasma levels of orexin-A were significantly associated with the occurrence of constipation after adjusting for all variables in all patients with cognitive impairment (OR = 0.151, 95% CI: 0.042-0.537, p = 0.003). And the same finding was more prominent in the LBD group (p = 0.048). CONCLUSIONS: The decrease of plasma level of orexin-A is closely associated with the occurrence of constipation. Orexin-A has an intestinal protective effect and is involved in the gastrointestinal symptoms of patients with cognitive impairment.

Serum lipid levels are associated with orthostatic hypotension in multiple system atrophy patients.

OBJECTIVES: Orthostatic hypotension (OH) is one of the most important autonomic features of multiple system atrophy (MSA). This study was established to confirm the correlation between lipid levels and OH in MSA. METHODS: A total of 580 patients with probable or possible MSA from neurological wards in six hospitals in Tianjin, Beijing, Hebei Province, and Henan Province, China, were included in this study. The tilt test or stand test was used to assess the severity of OH. Lipid contents, including total cholesterol, low-density-lipoprotein cholesterol (LDL-C), high-density-lipoprotein cholesterol (HDL-C), and triglyceride were evaluated. RESULTS: Serum levels of total cholesterol, LDL-C, and triglyceride in MSA-OH patients were significantly lower than those in MSA without OH. The risks of OH were significantly higher in the lowest quartiles of triglyceride and LDL-C than in the highest quartiles, after adjusting for confounders (OR = 2.17, 95% CI: 1.23-3.82, P = 0.008 and OR = 2.02, 95% CI: 1.16-3.47, P = 0.012). The risk of severe OH was significantly higher in the lowest quartile and the second quartile of triglyceride than in the highest quartile after adjusting for confounders (OR = 2.16, 95% CI: 1.20-3.87, P = 0.010 and OR = 2.25, 95% CI: 1.24-4.07, P = 0.007). Moreover, the risk of OH was significantly higher in the lowest quartile, and the third quartile of TC than in the highest quartile after adjusting for confounders (OR = 2.04, 95% CI: 1.18-3.52, P = 0.010 and OR = 2.06, 95% CI: 1.19-3.56, P = 0.010). CONCLUSION: Low levels of TC, LDL-C, and triglyceride increased the risk of OH in MSA. A low level of triglyceride predicted severe OH in MSA.

The characteristic of nonmotor symptoms with different phenotypes and onsets in multiple system atrophy patients.

OBJECTIVES: The characteristic of nonmotor symptoms in patients with multiple system atrophy (MSA) has varied among previous studies. The objective was to investigatethe nonmotor characteristics in MSA patients with different phenotypes, sex and different onset patterns. METHODS: We performed a retrospective review of 1492 MSA patients. All cases were evaluatedby neurologists and assessed with motormanifestations, nonmotor symptoms, auxiliary examinationand brain MRI scans. RESULTS: Multiple system atrophy-cerebellar ataxia (MSA-C) was the predominant phenotype in 998 patients. Average age of onset (56.8 ± 9.2 years) was earlier, the disease duration (2.4 ± 2.2 year) was shorter and brain MRI abnormalities (49.2 %) were more frequently in MSA-C (P < 0.001). Multiple system atrophy-parkinsonism (MSA-P) patients were more likely to have nonmotor symptoms. After adjusted significant parameters, urinary dysfunction (OR 1.441, 95 %CI = 1.067-1.946, P = 0.017), constipation (OR 1.482, 95 %CI = 1.113-1.973, P = 0.007), cognitive impairment (OR 1.509, 95 %CI = 1.074-2.121, P = 0.018) and drooling (OR 2.095, 95 %CI = 1.248-3.518, P = 0.005) were associated with the MSA-P phenotype. Males were more likely to have orthostatic hypotension, urinary dysfunction, sexual dysfunction, drooling and females in constipation and probable RBD. In different onset patterns, constipation (59.2 %) and probable RBD (28.4 %) were more frequently in autonomiconset pattern. CONCLUSIONS: MSA-C is the predominant phenotype in Chinese patients, while many nonmotor symptoms are more common in MSA-P phenotype. Patients with different sex and onset patterns have different nonmotor characteristics. The different clinical features identified could help the physician counseling of MSA patients more easily and more accurately.

Clinical outcomes and cognitive impairments between progressive supranuclear palsy and multiple system atrophy.

BACKGROUND: Both progressive supranuclear palsy (PSP) and multiple system atrophy (MSA) belong to atypical parkinsonian syndromes. It is important to differentiate these diseases accurately. We compared clinical outcomes and cognitive impairments between PSP and MSA. METHODS: Eighty-five MSA parkinsonism type (MSA-P) patients and 76 PSP patients participated in this research. The Montreal Cognitive Assessment (MoCA) and the mini-mental state examination (MMSE) evaluated cognitive function. RESULTS: MSA-P patients had a significantly higher incidence of dyskinesia, fall, urinary symptoms, and constipation, whereas patients with PSP had a higher incidence of tremor and salivation. MSA-P patients had higher MMSE and MoCA scores than PSP patients. The MMSE score showed a diagnostic cut-off score of 24.5 in PSP versus MSA-P. The MoCA score showed a diagnostic cut-off score of 20.5 in PSP versus MSA-P. CONCLUSION: In conclusion, patients with PSP had differences in the clinical outcomes and cognitive impairments compared with MSA-P patients. PSP patients had more severe cognitive deficits. The score of MMSE and MoCA could be used in distinguishing MSA-P from PSP.

The vascular risk factors and vascular neuropathology in subjects with autopsy-confirmed dementia with Lewy bodies.

BACKGROUND: The frequency of vascular risk factors (VRFs) and the relationship between vascular pathology and cognitive function in neurodegenerative disease remains incompletely understood. OBJECTIVE: The purpose of this study was to describe the frequency of VRFs and vascular pathology and explore the relationship between vascular pathology and cognitive function in dementia with Lewy bodies (DLB). METHODS: This study included 363 autopsy-confirmed DLB and 753 Alzheimer's disease (AD) patients from the National Alzheimer's Coordinating Center (NACC) database. We used chi-squared test and analysis of variance to compare the VRFs and related factors in DLB and AD. Multinomial logistic regression and Spearman's correlation test were used to examine the relationship between vascular pathology and cognitive function. RESULTS: No significant differences of VRFs were identified between DLB and AD. Alzheimer's disease patients had higher rates of microinfarcts (23.5% vs. 16.3%, p = 0.005) and moderate to severe amyloid angiopathy (45.9% vs. 36.1%, p = 0.002). In DLB patients, only cerebral amyloid angiopathy (CAA) pathology was negatively correlated with memory domain (r = -0.263, p < 0.001) and language (r = -0.112,p = 0.034). The rates of APOE ε4 allele carriers (60.0% vs. 44.9%, p = 0.004) and CAA pathology (45.9% vs.23.4%, p < 0.001) were much higher in the group with an intermediate likelihood of DLB than in the group with a high likelihood. There was a negative correlation between CAA pathology and memory (logical memory) in the group with an intermediate likelihood of DLB. CONCLUSION: No difference of VRFs was identified between autopsy-confirmed DLB and AD. Cerebral amyloid angiopathy was shown to be an important pathology in DLB, which specifically correlated with memory and language. The groups with high and intermediate likelihood of DLB differed in terms of CAA pathology, and CAA pathology may play an important role in the development of DLB.

Prevalence and risk factors of stroke in a rural area of northern China: a 10-year comparative study.

BACKGROUND AND AIMS: Stroke is currently the leading cause of death in China; however, the past decade has produced no new epidemiological studies of stroke. Therefore, the current study aimed to compare the prevalence and risk factors of stroke between 2010 and 2019. METHODS: A comparative study was used to analyze the prevalence of risk factors for stroke in a population aged 65 years or older between 2010 and 2019. Demographic characteristics, risk factors, medical history, and other clinical characteristics were collected for all participants via door-to-door interviews and inpatient hospital records. RESULTS: The standardized prevalence of stroke was 7.9% in 2010 and 14.2% in 2019 (p < 0.001). The prevalence of stroke was significantly higher in men than in women (p < 0.05) for all age groups. The risk factors of stroke were being male, hypertension, and diabetes mellitus in both 2010 and 2019. When comparing the risk factors between 2010 and 2019, these risk factors were statistically significantly more strongly associated with stroke in 2019 than in 2010. CONCLUSION: The current study suggests that the prevalence of stroke increased nearly by twofold in a population aged 65 years or older within the past 10 years. Hypertension, diabetes mellitus, and being male were the primary risk factors. In addition, these factors were more significantly associated with stroke in 2019 compared to 2010.

Orexin-A in Patients With Lewy Body Disease: A Systematic Review and Meta-Analysis.

Background: Abnormal orexin-A levels in cerebrospinal fluid (CSF) have been identified in Alzheimer's disease (AD) and other neurodegenerative diseases. However, few studies have focused on Lewy body disease (LBD) and often with debatable outcomes. Thus, we performed this systematic review and meta-analysis to investigate orexin-A levels in LBD by incorporating data from different studies. Methods: We gathered studies comparing orexin-A levels in patients with LBD and controls (including healthy controls and other dementia subtypes). In the initial search, 117 relevant articles were identified. After a selection process, seven studies, conducted in Japan, USA, Spain, Switzerland, France, Italy, and Netherlands, were chosen. Results: In total, 179 patients with LBD and 253 controls were included. Patients with LBD had significantly lower mean orexin-A CSF levels when compared with patients with AD [standard mean difference (SMD): -0.35, 95% confidence interval (CI): -0.70 to -0.00, Z = 1.96, P = 0.05], whereas mean orexin-A levels were significantly higher when compared with patients with frontotemporal lobar degeneration (FTLD) (SMD: 0.61, 95% CI: 0.23-0.99, Z = 3.12, P = 0.002). Orexin-A CSF levels in LBD patients were approximately equal to levels in healthy elderly individuals, whereas they were significantly decreased in LBD patients with excessive daytime sleepiness (EDS) (SMD: -0.15, 95% CI: -0.59 to 0.29, Z = 0.67, P = 0.50). Conclusions: We showed that orexin-A levels in patients with LBD were not very different from those in normal elderly individuals, whereas they were lower than those in AD patients and higher than those in FTLD patients. The influence of hypersomnia on orexin-A levels should be carefully interpreted. Systematic Review Registration: https://www.crd.york.ac.uk/prospero/, identifier CRD42021265900.

The Impact of the COVID-19 Pandemic and Lockdown on Mild Cognitive Impairment, Alzheimer's Disease and Dementia With Lewy Bodies in China: A 1-Year Follow-Up Study.

Background: While the lockdown strategies taken by many countries effectively limited the spread of COVID-19, those were thought to have a negative impact on older people. This study aimed to investigate the impact of lockdown on cognitive function and neuropsychiatric symptoms over a 1-year follow-up period in patients with mild cognitive impairment (MCI), Alzheimer's disease (AD) and dementia with Lewy bodies (DLB). Methods: We enrolled consecutive patients with MCI, probable AD or DLB who were receiving outpatient memory care before the COVID-19 pandemic and followed-up with them after 1 year by face-to-face during the COVID-19 pandemic to assess changes in physical activity, social contact, cognitive function and neuropsychiatric symptoms (NPS). Results: Total 105 probable AD, 50 MCI and 22 probable DLB patients were included and completed the 1-year follow-up between October 31 and November 30, 2020. Among the respondents, 42% of MCI, 54.3% of AD and 72.7% of DLB patients had a decline in MMSE scores and 54.4% of DLB patients had worsening Neuropsychiatric inventory (NPI) scores. Patients with DLB showed a more rapid decline of MMSE than those with AD. Diminished physical activity and social contact might have hastened the deterioration of cognition and the worsening of NPS. Conclusion: Social isolation and physical inactivity even after strict lockdown for at least 6 months were correlated with accelerated decline of cognitive function and NPS in patients with AD and DLB.

Predicting seizure freedom with AED treatment in newly diagnosed patients with MRI-negative epilepsy: A large cohort and multicenter study.

OBJECTIVE: We developed and validated a prediction score for predicting the probability of 6-month and 12-month seizure freedom of antiepileptic drug (AED) treatment in newly diagnosed patients with magnetic resonance imaging (MRI)-negative epilepsy. METHODS: The development cohort included 543 consecutive patients from the Epilepsy Center of Henan Provincial People's Hospital, while the validation cohorts included 493 consecutive patients in two independent cohorts. Univariate analysis and a forward and backward elimination of multivariate Cox regression analysis were used to select predictive factors. The performance of the score was evaluated with C-index, calibration plots, and decision curve analysis. The risk stratification was also performed. RESULTS: The score included five routinely available predictors including Circadian rhythms, Electroencephalography before AED treatment, Neuropsychiatric disorders, Perinatal brain injury, and History of central nervous system infection (CENPH score). When applied to the external validation cohort, the score showed good discrimination with C-index (development group: 0.83; validation group: 0.78), and calibration plots indicated well calibration, as well as the decision curve analysis showed good predictive accuracy and clinical values in four cohorts. The points of the score were categorized to the following three probability levels for predicting seizure freedom: high probability (0-83.11 points), medium probability (83.11-122.71 points), and low probability (>122.71 points). And online calculator was established to make this score easily applicable in clinical practice. CONCLUSIONS: We established a simple, practical, and evidence-based prediction score for predicting seizure freedom with AEDs to aid in the clinical consultation and treatment decision for the newly diagnosed patients with MRI-negative epilepsy.

Personalized prediction model for seizure-free epilepsy with levetiracetam therapy: a retrospective data analysis using support vector machine.

AIMS: To predict the probability of a seizure-free (SF) state in patients with epilepsy (PWEs) after treatment with levetiracetam and to identify the clinical and electroencephalographic (EEG) factors that affect outcomes. METHODS: Retrospective analysis of PWEs treated with levetiracetam for 3 years identified 22 patients who were SF and 24 who were not. Before starting levetiracetam, 11 clinical factors and four EEG features (sample entropy of α, ß, θ, δ) were identified. Overall, 80% of each the two groups were chosen to establish a support vector machine (SVM) model with 5-fold cross-validation, hold-out validation and jack-knife validation. The other 20% were used to predict the efficacy of levetiracetam. The mean impact value (MIV) algorithm was used to rank the relativity between factors and outcomes. RESULTS: Compared with SF patients, not SF patients displayed a specific decrease in EEG sample entropy in α band from the F4 channel, ß band from Fp2 and F8 channels, θ band from C3 channel (P < 0.05). The SVM model based on the clinical and EEG features yielded 72.2% accuracy of 5-fold cross-validation, 75.0% accuracy of jack-knife validation, 67.7% accuracy of hold-out validation in the training set and had a high prediction accuracy of 90% in test set (sensitivity was 100%, area under the receiver operating characteristic curve was 0.96). The feature of ß band from Fp2 weighs heavily in the prediction model according to the mean impact value algorithm. CONCLUSIONS: The efficacy of levetiracetam on newly diagnosed PWEs could be predicted using an SVM model, which could guide antiepileptic drug selection.