A New Nomogram Model for Individualized Prediction of Cognitive Impairment in Patients with Acute Ischemic Stroke.

Background: Cognitive impairment is a common symptom after ischemic stroke. Such symptom can cause effect on rehabilitation of patients and their quality of life and. As stroke rapidly growth on nowadays, a reliable scoring tool to detect the risk of cognitive impairment after stroke is now being put on the first place.

Methods: We enrolled patients with acute ischemic stroke (AIS) as samples and hospitalized all at the First Affiliated Hospital of Soochow University between October 2018 and June 2020. All patients were assessed by the Montreal Cognitive Assessment (MoCA) scales and MoCA score < 26 was defined as standard to have having cognitive impairment. All patients were randomly (7:3) divided into two cohorts: the primary ones and the validated ones. Based on multivariate logistic model, the independent predictors of cognitive impairment in the acute phase were identified. The predictive nomogram was generated and evaluated by Harrell's concordance index (C-index) and calibration plot both in two cohorts, respectively.

Results: A total of 191 patients were enrolled, of whom 135 comprised the primary cohort and 56 comprised the validated cohort. Gender, age, baseline NIHSS score, hyperhomocysteinemia (HHcy) and multiple lesions were independently associated with acute cognitive impairment after stroke and included to construct the nomogram. The nomogram derived from the primary cohort had an Area Under Curve (AUC) of 0.773 and the validated ones had an AUC of 0.859. Calibration plot revealed adequate fit of the nomogram in predictive value.

Conclusion: The new nomogram based on gender, age, baseline NIHSS score, HHcy and multiple lesions gave rise to an accurate and comprehensive prediction for cognitive impairment in AIS patients. After further validation, it could potentially be a reliable forecasting tool.