Predicting Functional Outcome Based on Linked Data After Acute Ischemic Stroke: S-SMART Score

Development of a new non invasive prognostic stroke scale (NIPSS) including triage elements for sleep apnea and peripheral artery disease

BACKGROUND

Although sleep apnea and peripheral artery disease are prognostic factors for stroke, their added benefit in the acute stage to further prognosticate strokes has not been evaluated.

OBJECTIVES

We tested the accuracy in the acute stroke stage of a novel score called the Non-Invasive Prognostic Stroke Scale (NIPSS).

PATIENTS AND METHODS

Prospective cohort with imaging-confirmed ischemic stroke. Clinical data, sleep apnea risk score (STOPBANG) and blood pressure measures were collected at baseline. Primary outcome was the 90-day modified Rankin Scale (mRS), with poor outcome defined as mRS 3-6. Area under the ROC curve (AUC) was calculated for NIPSS and compared to six other stroke prognostic scores in our cohort: SPAN-100 index, S-SMART, SOAR, ASTRAL, THRIVE, and Dutch Stroke scores.

RESULTS

We enrolled 386 participants. After 90 days, there were 56% with poor outcome, more frequently older, female predominant and with higher admission National Institute of Health Stroke Scale (NIHSS). Four variables remained significantly associated with primary endpoint in the multivariable model: age (OR 1.87), NIHSS (OR 7.08), STOPBANG category (OR 1.61), and ankle-braquial index (OR 2.11). NIPSS AUC was 0.86 (0.82-0.89); 0.83 (0.79-0.87) with bootstrapping. When compared to the other scores, NIPSS, ASTRAL, S-SMART and DUTCH scores had good abilities in predicting poor outcome, with AUC of 0.86, 0.86, 0.83 and 0.82, respectively. THRIVE, SOAR and SPAN-100 scores were fairly predictive.

DISCUSSION AND CONCLUSIONS

Non-invasive and easily acquired emergency room data can predict clinical outcome after stroke. NIPSS performed equal to or better than other prognostic stroke scales.

Feasibility of a Clinical-Radiomics Model to Predict the Outcomes of Acute Ischemic Stroke

Objective

To develop a model incorporating radiomic features and clinical factors to accurately predict acute ischemic stroke (AIS) outcomes.

Materials and Methods

Data from 522 AIS patients (382 male [73.2%]; mean age ± standard deviation, 58.9 ± 11.5 years) were randomly divided into the training (n = 311) and validation cohorts (n = 211). According to the modified Rankin Scale (mRS) at 6 months after hospital discharge, prognosis was dichotomized into good (mRS ≤ 2) and poor (mRS > 2); 1310 radiomics features were extracted from diffusion-weighted imaging and apparent diffusion coefficient maps. The minimum redundancy maximum relevance algorithm and the least absolute shrinkage and selection operator logistic regression method were implemented to select the features and establish a radiomics model. Univariable and multivariable logistic regression analyses were performed to identify the clinical factors and construct a clinical model. Ultimately, a multivariable logistic regression analysis incorporating independent clinical factors and radiomics score was implemented to establish the final combined prediction model using a backward step-down selection procedure, and a clinical-radiomics nomogram was developed. The models were evaluated using calibration, receiver operating characteristic (ROC), and decision curve analyses.

Results

Age, sex, stroke history, diabetes, baseline mRS, baseline National Institutes of Health Stroke Scale score, and radiomics score were independent predictors of AIS outcomes. The area under the ROC curve of the clinical-radiomics model was 0.868 (95% confidence interval, 0.825–0.910) in the training cohort and 0.890 (0.844–0.936) in the validation cohort, which was significantly larger than that of the clinical or radiomics models. The clinical radiomics nomogram was well calibrated (p > 0.05). The decision curve analysis indicated its clinical usefulness.

Conclusion

The clinical-radiomics model outperformed individual clinical or radiomics models and achieved satisfactory performance in predicting AIS outcomes.

Nijeholt G, van Es A, van der Lugt A, Majoie C, Burke J, Roozenbeek B, Lingsma H, Dippel D. Outcome Prediction Models for Endovascular Treatment of Ischemic Stroke: Systematic Review and External Validation

Supplemental Digital Content is available in the text.

Prediction models for outcome of patients with acute ischemic stroke who will undergo endovascular treatment have been developed to improve patient management. The aim of the current study is to provide an overview of preintervention models for functional outcome after endovascular treatment and to validate these models with data from daily clinical practice.

Prediction of Hemorrhagic Transformation after Ischemic Stroke Using Machine Learning

Hemorrhagic transformation (HT) is one of the leading causes of a poor prognostic marker after acute ischemic stroke (AIS). We compared the performances of the several machine learning (ML) algorithms to predict HT after AIS using only structured data. A total of 2028 patients with AIS, who were admitted within seven days of symptoms onset, were included in this analysis. HT was defined based on the criteria of the European Co-operative Acute Stroke Study-II trial. The whole dataset was randomly divided into a training and a test dataset with a 7:3 ratio. Binary logistic regression, support vector machine, extreme gradient boosting, and artificial neural network (ANN) algorithms were used to assess the performance of predicting the HT occurrence after AIS. Five-fold cross validation and a grid search technique were used to optimize the hyperparameters of each ML model, which had its performance measured by the area under the receiver operating characteristic (AUROC) curve. Among the included AIS patients, the mean age and number of male subjects were 69.6 years and 1183 (58.3%), respectively. HT was observed in 318 subjects (15.7%). There were no significant differences in corresponding variables between the training and test dataset. Among all the ML algorithms, the ANN algorithm showed the best performance in terms of predicting the occurrence of HT in our dataset (0.844). Feature scaling including standardization and normalization, and the resampling strategy showed no additional improvement of the ANN's performance. The ANN-based prediction of HT after AIS showed better performance than the conventional ML algorithms. Deep learning may be used to predict important outcomes for structured data-based prediction.

Modification of Acute Stroke Pathway in Korea After the Coronavirus Disease 2019 Outbreak

Background: Since the global pandemic of coronavirus disease 2019 (COVID-19), the process of emergency medical services has been modified to ensure the safety of healthcare professionals as well as patients, possibly leading to a negative impact on the timely delivery of acute stroke care. This study aimed to assess the impact of the COVID-19 pandemic on the acute stroke care processes and outcomes in tertiary COVID-19-dedicated centers in South Korea.

Methods: We included 1,213 patients with acute stroke admitted to three centers in three cities (Seoul, Seongnam, and Daegu) through the stroke critical pathway between September 2019 and May 2020 (before and during the COVID-19 pandemic). In all three centers, we collected baseline characteristics and parameters regarding the stroke critical pathway, including the number of admitted patients diagnosed with acute stroke through the stroke critical pathway, door to brain imaging time, door to intravenous recombinant tissue plasminogen activator time, door to groin puncture time, and door to admission time. We performed an interrupted time series analysis to determine the impact of the COVID-19 outbreak on outcomes and critical pathway parameters.

Results: Three centers modified the protocol of the stroke critical pathway during the COVID-19 pandemic. There was an immediate decrease in the number of patients admitted with acute ischemic stroke after the outbreak of COVID-19 in Korea, especially in the center of Daegu, an epicenter of the COVID-19 outbreak. However, the number of patients with stroke soon increased to equal that before the Covid-19 outbreak. In several critical pathway parameters, door to imaging time showed a temporary increase, and door to admission was transiently decreased after the COVID-19 outbreak. However, there was no significant effect on the timely trend. Moreover, there was no significant difference in the baseline characteristics and clinical outcomes between the periods before and during the COVID-19 pandemic.

Conclusion: This study demonstrated that the COVID-19 outbreak immediately affected the management process. However, it did not have a significant overall impact on the trends of stroke treatment processes and outcomes. The stroke management process should be modified according to changing situations for optimal acute management.