Automated Electronic Phenotyping of Cardioembolic Stroke.

BACKGROUND AND PURPOSE: Oral anticoagulation is generally indicated for cardioembolic strokes, but not for other stroke causes. Consequently, subtype classification of ischemic stroke is important for risk stratification and secondary prevention. Because manual classification of ischemic stroke is time-intensive, we assessed the accuracy of automated algorithms for performing cardioembolic stroke subtyping using an electronic health record (EHR) database. METHODS: We adapted TOAST (Trial of ORG 10172 in Acute Stroke Treatment) features associated with cardioembolic stroke for derivation in the EHR. Using administrative codes and echocardiographic reports within Mass General Brigham Biobank (N=13 079), we iteratively developed EHR-based algorithms to define the TOAST cardioembolic stroke features, revising regular expression algorithms until achieving positive predictive value ≥80%. We compared several machine learning-based statistical algorithms for discriminating cardioembolic stroke using the feature algorithms applied to EHR data from 1598 patients with acute ischemic strokes from the Massachusetts General Hospital Ischemic Stroke Registry (2002-2010) with previously adjudicated TOAST and Causative Classification of Stroke subtypes. RESULTS: Regular expression-based feature extraction algorithms achieved a mean positive predictive value of 95% (range, 88%-100%) across 11 echocardiographic features. Among 1598 patients from the Massachusetts General Hospital Ischemic Stroke Registry, 1068 had any cardioembolic stroke feature within predefined time windows in proximity to the stroke event. Cardioembolic stroke tended to occur at an older age, with more TOAST-based comorbidities, and with atrial fibrillation (82.3%). The best model was a random forest with 92.2% accuracy and area under the receiver operating characteristic curve of 91.1% (95% CI, 87.5%-93.9%). Atrial fibrillation, age, dilated cardiomyopathy, congestive heart failure, patent foramen ovale, mitral annulus calcification, and recent myocardial infarction were the most discriminatory features. CONCLUSIONS: Machine learning-based identification of cardioembolic stroke using EHR data is feasible. Future work is needed to improve the accuracy of automated cardioembolic stroke identification and assess generalizability of electronic phenotyping algorithms across clinical settings.

Design and rationale of a pragmatic trial integrating routine screening for atrial fibrillation at primary care visits: The VITAL-AF trial.

Given the preventable morbidity and mortality associated with atrial fibrillation (AF), increased awareness of undiagnosed AF, and advances in mobile electrocardiogram (ECG) technology, there is a critical need to assess the effectiveness of using such technology to routinely screen for AF in clinical practice. VITAL-AF is a pragmatic trial that will test whether screening for AF using a single-lead handheld ECG in individuals 65 years or older during primary care visits will lead to an increased rate of AF detection. The study is a cluster-randomized trial, with 8 primary care practices randomized to AF screening and 8 primary care practices randomized to usual care. We anticipate studying approximately 16,000 patients in each arm. During the 1-year enrollment period, practice medical assistants will screen eligible patients who agree to participate during office visits using a single-lead ECG device. Automated screening results are documented in the electronic health record, and patients can discuss screening results with their provider during the scheduled visit. All single-lead ECGs are overread by a cardiologist. Screen-detected AF is managed at the discretion of the patient's physician. The primary study end point is incident AF during the screening period. Key secondary outcomes include new oral anticoagulation prescriptions, incident ischemic stroke, and major hemorrhage during a 24-month period following the study start. Outcomes are ascertained based on electronic health record documentation and are manually adjudicated. The results of this pragmatic trial may help identify a model for widespread adoption of AF screening as part of routine clinical practice.

Seed-competent high-molecular-weight tau species accumulates in the cerebrospinal fluid of Alzheimer's disease mouse model and human patients.

OBJECTIVE: Cerebrospinal fluid (CSF) tau is an excellent surrogate marker for assessing neuropathological changes that occur in Alzheimer's disease (AD) patients. However, whether the elevated tau in AD CSF is just a marker of neurodegeneration or, in fact, a part of the disease process is uncertain. Moreover, it is unknown how CSF tau relates to the recently described soluble high-molecular-weight (HMW) species that is found in the postmortem AD brain and can be taken up by neurons and seed aggregates. METHODS: We have examined seeding and uptake properties of brain extracellular tau from various sources, including interstitial fluid (ISF) and CSF from an AD transgenic mouse model and postmortem ventricular and antemortem lumbar CSF from AD patients. RESULTS: We found that brain ISF and CSF tau from the AD mouse model can be taken up by cells and induce intracellular aggregates. Ventricular CSF from AD patients contained a rare HMW tau species that exerted a higher seeding activity. Notably, the HMW tau species was also detected in lumbar CSF from AD patients, and its levels were significantly elevated compared to control subjects. HMW tau derived from CSF of AD patients was seed competent in vitro. INTERPRETATION: These findings suggest that CSF from an AD brain contains potentially bioactive HMW tau species, giving new insights into the role of CSF tau and biomarker development for AD. Ann Neurol 2016;80:355-367.

Association between α-synuclein blood transcripts and early, neuroimaging-supported Parkinson's disease.

There are no cures for neurodegenerative diseases and this is partially due to the difficulty of monitoring pathogenic molecules in patients during life. The Parkinson's disease gene α-synuclein (SNCA) is selectively expressed in blood cells and neurons. Here we show that SNCA transcripts in circulating blood cells are paradoxically reduced in early stage, untreated and dopamine transporter neuroimaging-supported Parkinson's disease in three independent regional, national, and international populations representing 500 cases and 363 controls and on three analogue and digital platforms with P < 0.0001 in meta-analysis. Individuals with SNCA transcripts in the lowest quartile of counts had an odds ratio for Parkinson's disease of 2.45 compared to individuals in the highest quartile. Disease-relevant transcript isoforms were low even near disease onset. Importantly, low SNCA transcript abundance predicted cognitive decline in patients with Parkinson's disease during up to 5 years of longitudinal follow-up. This study reveals a consistent association of reduced SNCA transcripts in accessible peripheral blood and early-stage Parkinson's disease in 863 participants and suggests a clinical role as potential predictor of cognitive decline. Moreover, the three independent biobank cohorts provide a generally useful platform for rapidly validating any biological marker of this common disease.

Re-CHARGE-AF: Recalibration of the CHARGE-AF Model for Atrial Fibrillation Risk Prediction in Patients With Acute Stroke.

Background Performance of existing atrial fibrillation (AF) risk prediction models in poststroke populations is unclear. We evaluated predictive utility of an AF risk model in patients with acute stroke and assessed performance of a fully refitted model. Methods and Results Within an academic hospital, we included patients aged 46 to 94 years discharged for acute ischemic stroke between 2003 and 2018. We estimated 5-year predicted probabilities of AF using the Cohorts for Heart and Aging Research in Genomic Epidemiology for Atrial Fibrillation (CHARGE-AF) model, by recalibrating CHARGE-AF to the baseline risk of the sample, and by fully refitting a Cox proportional hazards model to the stroke sample (Re-CHARGE-AF) model. We compared discrimination and calibration between models and used 200 bootstrap samples for optimism-adjusted measures. Among 551 patients with acute stroke, there were 70 incident AF events over 5 years (cumulative incidence, 15.2%; 95% CI, 10.6%-19.5%). Median predicted 5-year risk from CHARGE-AF was 4.8% (quartile 1-quartile 3, 2.0-12.6) and from Re-CHARGE-AF was 16.1% (quartile 1-quartile 3, 8.0-26.2). For CHARGE-AF, discrimination was moderate (C statistic, 0.64; 95% CI, 0.57-0.70) and calibration was poor, underestimating AF risk (Greenwood-Nam D'Agostino chi-square, P<0.001). Calibration with recalibrated baseline risk was also poor (Greenwood-Nam D'Agostino chi-square, P<0.001). Re-CHARGE-AF improved discrimination (P=0.001) compared with CHARGE-AF (C statistic, 0.74 [95% CI, 0.68-0.79]; optimism-adjusted, 0.70 [95% CI, 0.65-0.75]) and was well calibrated (Greenwood-Nam D'Agostino chi-square, P=0.97). Conclusions Covariates from an established AF risk model enable accurate estimation of AF risk in a poststroke population after recalibration. A fully refitted model was required to account for varying baseline AF hazard and strength of associations between covariates and incident AF.

Prediction of cognition in Parkinson's disease with a clinical-genetic score: a longitudinal analysis of nine cohorts.

BACKGROUND: Cognitive decline is a debilitating manifestation of disease progression in Parkinson's disease. We aimed to develop a clinical-genetic score to predict global cognitive impairment in patients with the disease. METHODS: In this longitudinal analysis, we built a prediction algorithm for global cognitive impairment (defined as Mini Mental State Examination [MMSE] ≤25) using data from nine cohorts of patients with Parkinson's disease from North America and Europe assessed between 1986 and 2016. Candidate predictors of cognitive decline were selected through a backward eliminated Cox's proportional hazards analysis using the Akaike's information criterion. These were used to compute the multivariable predictor on the basis of data from six cohorts included in a discovery population. Independent replication was attained in patients from a further three independent longitudinal cohorts. The predictive score was rebuilt and retested in 10 000 training and test sets randomly generated from the entire study population. FINDINGS: 3200 patients with Parkinson's disease who were longitudinally assessed with 27 022 study visits between 1986 and 2016 in nine cohorts from North America and Europe were assessed for eligibility. 235 patients with MMSE ≤25 at baseline and 135 whose first study visit occurred more than 12 years from disease onset were excluded. The discovery population comprised 1350 patients (after further exclusion of 334 with missing covariates) from six longitudinal cohorts with 5165 longitudinal visits over 12·8 years (median 2·8, IQR 1·6-4·6). Age at onset, baseline MMSE, years of education, motor exam score, sex, depression, and ß-glucocerebrosidase (GBA) mutation status were included in the prediction model. The replication population comprised 1132 patients (further excluding 14 patients with missing covariates) from three longitudinal cohorts with 19 127 follow-up visits over 8·6 years (median 6·5, IQR 4·1-7·2). The cognitive risk score predicted cognitive impairment within 10 years of disease onset with an area under the curve (AUC) of more than 0·85 in both the discovery (95% CI 0·82-0·90) and replication (95% CI 0·78-0·91) populations. Patients scoring in the highest quartile for cognitive risk score had an increased hazard for global cognitive impairment compared with those in the lowest quartile (hazard ratio 18·4 [95% CI 9·4-36·1]). Dementia or disabling cognitive impairment was predicted with an AUC of 0·88 (95% CI 0·79-0·94) and a negative predictive value of 0·92 (95% 0·88-0·95) at the predefined cutoff of 0·196. Performance was stable in 10 000 randomly resampled subsets. INTERPRETATION: Our predictive algorithm provides a potential test for future cognitive health or impairment in patients with Parkinson's disease. This model could improve trials of cognitive interventions and inform on prognosis. FUNDING: National Institutes of Health, US Department of Defense.

Unrecognized vitamin D3 deficiency is common in Parkinson disease: Harvard Biomarker Study.

OBJECTIVE: To conclusively test for a specific association between the biological marker 25-hydroxy-vitamin D3, a transcriptionally active hormone produced in human skin and liver, and the prevalence and severity of Parkinson disease (PD). METHODS: We used liquid chromatography/tandem mass spectrometry to establish an association specifically between deficiency of 25-hydroxy-vitamin D3 and PD in a cross-sectional and longitudinal case-control study of 388 patients (mean Hoehn and Yahr stage of 2.1 ± 0.6) and 283 control subjects free of neurologic disease nested in the Harvard Biomarker Study. RESULTS: Plasma levels of 25-hydroxy-vitamin D3 were associated with PD in both univariate and multivariate analyses with p values = 0.0034 and 0.047, respectively. Total 25-hydroxy-vitamin D levels, the traditional composite measure of endogenous and exogenous vitamin D, were deficient in 17.6% of patients with PD compared with 9.3% of controls. Low 25-hydroxy-vitamin D3 as well as total 25-hydroxy-vitamin D levels were correlated with higher total Unified Parkinson's Disease Rating Scale scores at baseline and during follow-up. CONCLUSIONS: Our study reveals an association between 25-hydroxy-vitamin D3 and PD and suggests that thousands of patients with PD in North America alone may be vitamin D-deficient. This finding has immediate relevance for individual patients at risk of falls as well as public health, and warrants further investigation into the mechanism underlying this association.