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.

Endothelial and Leukocyte-Derived Microvesicles and Cardiovascular Risk After Stroke: PROSCIS-B

OBJECTIVE

To determine the role of circulating microvesicles (MV) on long-term cardiovascular outcomes after stroke, we measured them in patients with first-ever stroke with a 3-year follow-up.

METHODS

In the Prospective Cohort With Incident Stroke Berlin (PROSCIS-B), patients with first-ever ischemic stroke were followed up for 3 years. The primary combined endpoint consisted of recurrent stroke, myocardial infarction, and all-cause mortality. Citrate-blood levels of endothelial MV (EMV), leukocyte-derived MV (LMV), monocytic MV (MMV), and platelet-derived MV (PMV) were measured with flow cytometry. Kaplan-Meier curves and adjusted Cox proportional hazards models were used to estimate the effect of MV levels on the combined endpoint.

RESULTS

Five hundred seventy-one patients were recruited (median age 69 years, 39% female, median NIH Stroke Scale score 2, interquartile range 1-4), and 95 endpoints occurred. Patients with levels of EMV (adjusted hazard ratio [HR] 2.5, 95% confidence interval [CI] 1.2-4.9) or LMV (HR 3.1, 95% CI 1.4-6.8) in the highest quartile were more likely to experience an event than participants with lower levels with the lowest quartile used as the reference category. The association was less pronounced for PMV (HR 1.7, 95% CI 0.9-3.2) and absent for MMV (HR 1.1, 95% CI 0.6-1.8).

CONCLUSION

High levels of EMV and LMV after stroke were associated with worse cardiovascular outcome within 3 years. These results reinforce that endothelial dysfunction and vascular inflammation affect the long-term prognosis after stroke. EMV and LMV might play a role in risk prediction for stroke patients.

CLINICALTRIALSGOV IDENTIFIER

NCT01363856.

CLASSIFICATION OF EVIDENCE

This study provides Class II evidence of the effect of MV levels on subsequent stroke, myocardial infarction, or all-cause mortality in survivors of mild stroke.