Predictive Model and Mortality Risk Score during Admission for Ischaemic Stroke with Conservative Treatment

Abstract

Background: Stroke is the second cause of mortality worldwide and the first in women. The aim of this study is to develop a predictive model to estimate the risk of mortality in the admission of patients who have not received reperfusion treatment. Methods: A retrospective cohort study was conducted of a clinical–administrative database, reflecting all cases of non-reperfused ischaemic stroke admitted to Spanish hospitals during the period 2008–2012. A predictive model based on logistic regression was developed on a training cohort and later validated by the “hold-out” method. Complementary machine learning techniques were also explored. Results: The resulting model had the following nine variables, all readily obtainable during initial care. Age (OR 1.069), female sex (OR 1.202), readmission (OR 2.008), hypertension (OR 0.726), diabetes (OR 1.105), atrial fibrillation (OR 1.537), dyslipidaemia (0.638), heart failure (OR 1.518) and neurological symptoms suggestive of posterior fossa involvement (OR 2.639). The predictability was moderate (AUC 0.742, 95% CI: 0.737–0.747), with good visual calibration; Pearson’s chi-square test revealed non-significant calibration. An easily consulted risk score was prepared. Conclusions: It is possible to create a predictive model of mortality for patients with ischaemic stroke from which important advances can be made towards optimising the quality and efficiency of care. The model results are available within a few minutes of admission and would provide a valuable complementary resource for the neurologist.

External Validation and Recalibration of a Mortality Prediction Model for Patients with Ischaemic Stroke

“…In this analytical observational study, we perform an external validation and subsequent recalibration of the coefficients of the original variables from a prior model [11] used to predict the mortality at first admission of patients with non-reperfused ischaemic stroke.…”

“…The model was obtained using binary logistic regression, and the variables included were age, sex, 30-day readmission status, chronic ischaemic heart disease, diabetes mellitus, hypertension, dyslipidaemia, heart failure, and symptoms suggestive of posterior circulation stroke. More detailed information on the BM can be found in the original publication [11].…”

“…Several models have been proposed for evaluating the risk of mortality and sequelae [5][6][7][8]. Some have been applied to clinical-administrative registers and databases [9][10][11][12][13], but their utility is limited because these records do not contain the specific variables found in the main scale used worldwide for assessing the risk and severity of stroke, i.e., the National Institutes of Health Stroke Scale (NIHSS) [14].…”

“…Lea-Pereira et al (2022) [11] developed a mortality estimation model for use at first hospital admission based on the Minimum Basic Data Set (MBDS) (compiled by the Spanish Health Ministry) of patients admitted in Spain for non-reperfused stroke during the period 2008-2012. This model had moderate-high discriminant capacity (AUC: 0.742, 95% CI [0.74-0.75]), good visual calibration according to the representation of the risk deciles, and included most of the sociodemographic and clinical variables that are usually recorded during hospital admission in Spain.…”

Novel Survival Features Generated by Clinical Text Information and Radiomics Features May Improve the Prediction of Ischemic Stroke Outcome

“…Several studies have shown that stroke outcomes correlate with CTI and stroke parameters computed from images. Lea-Pereira et al [ 10 ] predicted mortality risk scores during admission for ischemic stroke with CTI such as age, sex, readmission, and neurological symptoms, and an AUC of 0.742 was obtained with a logistic regression model. Ref.…”