South China Journal of Preventive Medicine ›› 2023, Vol. 49 ›› Issue (10): 1213-1217.doi: 10.12183/j.scjpm.2023.1213

• Original Article •     Next Articles

Early poor prognosis and predictive model construction in patients with acute ischemic stroke

YAN Chun, GONG Yuan, HUANG Ping, CHEN Juan, HUANG Miao   

  1. Hubei University for Nationalities Affiliated Hospital, Enshi 445000, China
  • Received:2023-05-29 Online:2023-10-20 Published:2023-11-28

Abstract: Objective To explore the influencing factors of early poor prognosis in patients with acute ischemic stroke (AIS) and further construct a risk prediction model. Methods AIS patients admitted to a tertiary general hospital in Enshi from January 2019 to December 2021 were selected for this study. After 6 months of admission, the patients were followed up and evaluated for early poor prognosis based on the modified Rankin scale (mRS) score. Based on the follow-up results, they were divided into a good prognosis group and a poor prognosis group. The case characteristics of the two groups were compared and the influencing factors for early poor prognosis were analyzed using multivariate logistic regression, and the risk prediction model was constructed and ROC curve was drawn to verify and evaluate the model. Results The results included a total of 836 AIS patients, 515 males and 321 females, aged 42-89 years, with an average age of (65.37±11.96) years. Stroke location: 227 cases of internal carotid artery, 301 cases of middle cerebral artery, 263 cases of posterior cerebral artery and others. Among them, 279 AIS patients had poor prognosis during a short-term follow-up of 6 months after admission, with a poor prognosis rate of 33.37%. Multivariate logistic regression analysis showed that age (OR=1.234), atrial fibrillation (OR=2.992), admission NIHSS score (OR=1.906), NLR (OR=1.770), D-dimer (OR=1.614), and UA (OR=0.834) were independent influencing factors for poor prognosis in AIS patients. The area under the ROC curve (AUC) was 0.817 (95% CI: 0.739-0.934). When the cut off value was 0.469, the accuracy of the model was 85.19%, sensitivity was 81.43%, and specificity was 76.13%. Conclusion The risk prediction model constructed based on age, atrial fibrillation, admission NIHSS score, NLR, D-dimer, and UA has good predictive efficacy for early poor prognosis in AIS patients, and the model is stable and well fitted.

Key words: Acute ischemic stroke, Early prognosis, Influencing factor, Prediction model construction

CLC Number: 

  • R195