Objective To assess the predictive value of triglyceride-glucose index（ TyG） combined with C-reactive protein（ CRP） in the prognosis of patients with ischemic stroke. Methods This was a retrospective cohort study that included 1 125 patients with ischemic stroke in Cangzhou People's Hospital from 2018 to 2021. Clinical and follow-up data were collected from patients， and TyG and CRP were determined and calculated. Patients were divided into recurrent group（ 79 cases） and non-recurrent group（ 1 046 cases） according to whether the follow-up outcome was recurrent ischemic stroke or not， and the clinical data of the two groups were compared. The relationship between TyG and CRP and ischemic stroke recurrence was explored using Cox proportional risk regression， and the value of these two markers individually and in combination in predicting stroke recurrence was assessed using receiver operating characteristic（ ROC） curve， and the covariance between the markers was analyzed using variance inflation factor（ VIF）. Results The proportion of diabetes， proportion of hypertension， proportion of atrial fibrillation， proportion of heart failure， serum uric acid（SUA），serum creatinine（SCr），homocgsteme（HCY），fasting blood glucose（ FBG），hemoglobinalc（HbA1c），total cholesterol （TC），triglyceride（TG），low-density lipoprotein cholesterol（LDL-C）， CRP， and TyG were higher in recurrent group compared with non-recurrent group， and the differences were statistically significant（ all P＜0.05）. After adjusting for all confounding factors， multifactorial Cox proportional risk regression model analysis showed that elevated CRP［ HR=1.028， 95%CI（ 1.017，1.039）， P＜0.001］ and TyG［ HR=2.450， 95%CI（ 1.487，4.039）， P ＜ 0.001］ were independent risk factors for recurrent ischemic stroke， and the difference was statistically significant. The area under the ROC curve（ AUC） was 0.689［ 95%CI（ 0.662，0.716）］ for CRP， 0.756［ 95%CI （0.730，0.781）］ for TyG， and 0.838［ 95%CI（ 0.815，0.859）］ for CRP combined with TyG. The combined predictive value of the two was statistically higher than the predictive value of CRP and TyG alone（ all P＜0.05）. CRP and TyG were analyzed for covariance in the univariate analysis， and the results showed that there was no covariance between them（ VIF=1.133）. Conclusions TyG and CRP have a certain independent predictive value for the prognosis of ischemic stroke patients， and the combined assessment of the two can further improve the accuracy of prediction.