A nomogram for predicting early thrombolytic efficacy in stroke patients

Background The purpose of this study was to develop and verify a novel nomogram for predicting stroke patients’ early thrombolytic efficacy. Methods We collect basic facts and clinical data of stroke patients with intravenous thrombolysis. A nomogram was established for predicting early thrombolytic efficacy in these people. The LASSO regression method and multivariate logistic regression were used to filter variables and choose predictors. Predictors were applied to develop a model. The model’s discriminatory capacity was assessed by computing the area under the curve. In addition, the model’s calibration analysis and decision curve analysis were performed. Results Using multivariate logistic regression and LASSO regression techniques, five variables were chosen. These variables were age, NIHSS score before thrombolysis, prothrombin time, neutrophil count, and monocyte count. The AUC of the prediction model was 0.761 (95% CI, 0.717–0.805) in the training set and 0.744 (95% CI, 0.653–0.835) in the test set. The decision curve showed that the threshold probabilities for the effectiveness of early thrombolysis in cerebral infarction are 25–67% and 25–73% in the training set and test set, respectively. Conclusion A novel nomogram with age, NIHSS score before thrombolysis, prothrombin time, neutrophil count, and monocyte count as variables has the potential to predict early thrombolytic efficacy in stroke patients. Physicians could utilize this handy nomogram to make better decisions for stroke patients with intravenous thrombolysis.