Machine learning-based prediction of clinical outcomes after first-ever ischemic stroke.

Background: Accurate prediction of clinical outcomes in individual patients following acute stroke is vital for healthcare providers to optimize treatment strategies and plan further patient care. Here, we use advanced machine learning (ML) techniques to systematically compare the prediction of functional recovery, cognitive function, depression, and mortality of first-ever ischemic stroke patients and to identify the leading prognostic factors. Methods: We predicted clinical outcomes for 307 patients (151 females, 156 males; 68 ± 14 years) from the PROSpective Cohort with Incident Stroke Berlin study using 43 baseline features. Outcomes included modified Rankin Scale (mRS), Barthel Index (BI), Mini-Mental State Examination (MMSE), Modified Telephone Interview for Cognitive Status (TICS-M), Center for Epidemiologic Studies Depression Scale (CES-D) and survival. The ML models included a Support Vector Machine with a linear kernel and a radial basis function kernel as well as a Gradient Boosting Classifier based on repeated 5-fold nested cross-validation. The leading prognostic features were identified using Shapley additive explanations. Results: The ML models achieved significant prediction performance for mRS at patient discharge and after 1 year, BI and MMSE at patient discharge, TICS-M after 1 and 3 years and CES-D after 1 year. Additionally, we showed that National Institutes of Health Stroke Scale (NIHSS) was the top predictor for most functional recovery outcomes as well as education for cognitive function and depression. Conclusion: Our machine learning analysis successfully demonstrated the ability to predict clinical outcomes after first-ever ischemic stroke and identified the leading prognostic factors that contribute to this prediction.

White matter hyperintensities are an independent predictor of cognitive decline 3 years following first-ever stroke-results from the PROSCIS-B study.

BACKGROUND: White matter hyperintensities (WMH) are the result of cerebral small vessel disease and may increase the risk of cognitive impairment (CI), recurrent stroke, and depression. We aimed to explore the association between selected cerebrovascular risk factors (CVRF) and WMH load as well as the effect of increased WMH burden on recurrent vascular events, CI, and depression in first-ever ischemic stroke patients. METHODS: 431 from the PROSpective Cohort with Incident Stroke (PROSCIS) were included; Age-Related White Matter Changes (ARWMC) score was used to assess WMH burden on FLAIR. The presence of CVRF (defined via blood pressure, body-mass-index, and serological markers of kidney dysfunction, diabetes mellitus, and hyperlipoproteinemia) was categorized into normal, borderline, and pathological profiles based on commonly used clinical definitions. The primary outcomes included recurrent vascular events (combined endpoint of recurrent stroke, myocardial infarction and/or death), CI 3 years post-stroke, and depression 1-year post-stroke. RESULTS: There was no clear association between CVRF profiles and WMH burden. High WMH lesion load (ARWMC score ≥ 10) was found to be associated with CI (adjusted OR 1.05 [95% CI 1.00-1.11]; p < 0.02) in a mixed-model analysis. Kaplan-Meier survival analysis showed a visible increase in the risk of recurrent vascular events following stroke; however, after adjustment, the risk was non-significant (HR 1.5 [95% CI 0.76-3]; p = 0.18). WMH burden was not associated with depression 1-year post stroke (adjusted OR 0.72 [95% CI 0.31-1.64]; p = 0.44). CONCLUSION: Higher WMH burden was associated with a significant decline in cognition 3 years post-stroke in this cohort of first-ever stroke patients.

Cardiovascular Diseases in the Patients With Psoriatic Arthritis.

There are limited data regarding the burden and trend of cardiovascular diseases (CVD) in psoriatic arthritis (PsA). We analyzed the National Inpatient Sample database from January 2005 to December 2018 to examine the hospitalization trends amongst adults with PsA primarily for heart failure (HF), acute myocardial infarction (AMI), and stroke. The primary outcomes of interest included in-hospital mortality, length of stay (LOS), and inflation-adjusted cost. The age-adjusted percentage of HF hospitalizations among PsA patients decreased from 2.5% (2005/06) to 1.4% (2011/12; P-trend 0.013) and subsequently increased to 2.0% (2017/18; P-trend 0.044). The age-adjusted percentage of AMI hospitalizations among PsA patients showed a non-statistically significant decreasing trend from 2.1% (2005/06) to 1.7% (2011/12; P-trend 0.248) and showed a non-statistically significant increase to 2.3% (2017/18; P-trend 0.056). The age-adjusted stroke hospitalizations increased from 1.1% (2005/06) to 1.3% (2017/18; P-trend 0.036). Apart from a decrease in adjusted inflation-adjusted cost among heart failure hospitalizations, there was no significant change in inpatient mortality, length of stay or hospital cost, during the study period. We found an increasing trend of cardiovascular hospitalizations in patients with PsA. These findings will raise awareness and inform further research and clinical practice for PSA patients with CVD.

The smoking paradox in ischemic stroke patients treated with intra-arterial thrombolysis in combination with mechanical thrombectomy-VISTA-Endovascular.

BACKGROUND: The smoking-paradox of a better outcome in ischemic stroke patients who smoke may be due to increased efficacy of thrombolysis. We investigated the effect of smoking on outcome following endovascular therapy (EVT) with mechanical thrombectomy alone versus in combination with intra-arterial (IA-) thrombolysis. METHODS: The primary endpoint was defined by three-month modified Rankin Scale (mRS). We performed a generalized linear model and reported relative risks (RR) for smoking (adjustment for age, sex, hypertension, atrial fibrillation, stroke severity, time to EVT) in patient data stemming from the Virtual International Stroke Trials Archive-Endovascular database. RESULTS: Among 1,497 patients, 740(49.4%) were randomized to EVT; among EVT patients, 524(35.0%) received mechanical thrombectomy alone and 216(14.4%) received it in combination with IA-thrombolysis. Smokers (N = 396) had lower mRS scores (mean 2.9 vs. 3.2; p = 0.02) and mortality rates (10% vs. 17.3%; p<0.001) in univariate analysis. In all patients and in patients treated with mechanical thrombectomy alone, smoking had no effect on outcome in regression analyses. In patients who received IA-thrombolysis (N = 216;14%), smoking had an adjusted RR of 1.65 for an mRS≤1 (95%CI 0.77-3.55). Treatment with IA-thrombolysis itself led to reduced RR for favorable outcome (adjusted RR 0.30); interaction analysis of IA-thrombolysis and smoking revealed that non-smokers with IA-thrombolysis had mRS≤2 in 47 cases (30%, adjusted RR 0.53 [0.41-0.69]) while smokers with IA-thrombolysis had mRS≤2 in 23 cases (38%, adjusted RR 0.61 [0.42-0.87]). CONCLUSIONS: Smokers had no clear clinical benefit from EVT that incorporates IA-thrombolysis.