Construction of a nomogram prediction model for early neurological deterioration after intravenous thrombolysis in acute ischemic stroke
LI Baiyan
WANG Lei
MI Xiaolu
WANG Yuanyuan
ZHU Baoyue
XIAO Qingmian
WANG Weizhan
Abstract：Objective To construct a nomogram prediction model for early neurological deterioration(END)after intravenous thrombolysis in acute ischemic stroke(AIS).Methods This retrospective study included AIS patients who underwent intravenous thrombolysis with recombinant tissue plasminogen activator(rt-PA)from January 2021,to December 2023.Patients were randomly divided into training and validation sets at a 3:7 ratio.LASSO regression and multivariate logistic regression were applied to screen variables in the training set to construct a nomogram model to predict END after intravenous thrombolysis in AIS.The model's discrimination,calibration,and clinical utility were compared with the baseline National Institutes of Health Stroke Scale(NIHSS)score in both the training and validation sets.Results A total of 338 patients were included in the study,among whom 55 experienced END.Baseline NIHSS score,presence of large hemispheric infarction,atrial fibrillation,and platelet to lymphocyte ratio(PLR)at 24 h after thrombolysis were ultimately screened out to be independent risk factors for END after intravenous thrombolysis in patients with AIS(P<0.05,P<0.01).The area under receiver operating characteristic(ROC)curve(AUC)of the training set nomogram model and baseline NIHSS score for predicting END after intravenous thrombolysis in AIS patients was 0.818(95%CI:0.746,0.890)and 0.755(95%CI:0.681,0.830)respectively;the AUC of the validation set and the baseline NIHSS score for predicting END after intravenous thrombolysis in AIS patients was 0.912(95%CI:0.832,0.993)and 0.783(95%CI:0.630,0.935)respectively.Compared with the NIHSS score,the nomogram showed improved predictive ability in the training set[net reclassification index(NRI)=0.1853,P=0.002;integrated discrimination improvement(IDI)=0.131,P<0.001]and the validation set(NRI=0.3077,P=0.017;IDI=0.365,P<0.001).Hosmer-Lemeshow tests indicated good calibration(P=0.673 in the training set;P=0.538 in the validation set).Decision curve analysis demonstrated better clinical net benefit for the novel nomogram compared with the baseline NIHSS score in both sets.Conclusion Baseline NIHSS score,presence of large hemispheric infarction,atrial fibrillation,and PLR at 24 h after thrombolysis are independent risk factors for END in AIS patients.The nomogram model based on these factors outperforms the NIHSS score in discrimination,calibration,and clinical utility.
Keywords：acute ischemic strokeearly deterioration of neurological functionintravenous thrombolysisnomogramNational Institutes of Health stroke scaleatrial fibrillationplatelets to lymphocytes ratio
Publication Date：2025-12-13
Online Publishing Date：2025-12-25(First online date of this platform, not the publication date of the document)
Pages：10( 95-104 )