Clinical risk factors to predict prognosis in wake-up stroke patients: A retrospective study

This study aimed to develop and validate a clinical risk model based on clinical factors to predict prognosis in patients with wake-up stroke (WUS) after multimodal magnetic resonance imaging combined with recombinant tissue plasminogen activator intravenous thrombolysis. The study enrolled 263 patients with WUS, who were divided into the training (n = 162) and validation cohorts (n = 101). In the training cohort, patients were stratified based on modified Rankin Scale (mRS) score at 90 days after thrombolysis, with mRS ≤ 2 indicating a good prognosis (n = 117), and mRS > 2 indicating a poor prognosis (n = 45). Multivariate regression analyses were employed to identify independent risk factors and develop clinical risk models. The performance and stability of the clinical risk model were evaluated using receiver operating characteristic analysis and Hosmer-Lemeshow test. The clinical risk nomogram was constructed based on this model, and evaluated using decision curve analyses. Patients with poor prognosis showed a higher proportion of hyperlipidemia and diabetes and showed a higher levels of National Institute of Health Stroke Scale (NIHSS) at admission, NIHSS at 24 hours, triglyceride, and total cholesterol. Diabetes (odds ratio [OR] = 3.823), hyperlipidemia (OR = 7.361), NIHSS at admission (OR = 5.399), NIHSS at 24 hours (OR = 2.869), triglyceride (OR = 13.790), and total cholesterol (OR = 9.719) were independent predictors of poor prognosis in patients with WUS. Hosmer-Lemeshow test showed that the clinical risk model had a good fit in the training (χ2 = 19.573, P = .726) and validation cohorts (χ2 = 19.573, P = .726). The clinical risk model had an area under the curve value of 0.929 (95% confidence interval, 0.886-0.978) in the training cohort and 0.948 (0.906-0.989) in the validation cohort. The decision curve analysis indicated clinical risk nomogram has application value. The clinical risk model can effectively predict WUS prognosis outcomes.