Objective To investigate the effect of triglyceride-glucose index (TyG) on the short-term prognosis of patients with initial acute ischemic stroke. Methods A total of 391 patients with initial acute ischemic stroke who were hospitalized in Kunshan Second People’s Hospital and The Affiliated Suzhou Hospital of Nanjing Medical University from Jun. 2020 to Jun. 2023 were retrospectively included. According to the modified Rankin scale (mRS) scores at 90 d follow-up, they were assigned to good prognosis group (286 cases) or poor prognosis group (105 cases). Logistic regression and receiver operating characteristic (ROC) curve were used to evaluate the effect of TyG on the short-term prognosis of patients with initial acute ischemic stroke. Results Compared with the good prognosis group, the patients of poor prognosis group had older age, higher proportion of atrial fibrillation, higher levels of homocysteine, triglyceride, total cholesterol, low-density lipoprotein, and TyG, and higher National Institutes of Health stroke scale (NIHSS) score (all P＜0.05). Multivariate logistic regression analysis showed that age, homocysteine, TyG and NIHSS score were independent risk factors for poor prognosis in patients with initial acute ischemic stroke (all P＜0.05). ROC curve analysis showed that TyG combined with NIHSS score had good predictive value for poor prognosis of patients with initial acute ischemic stroke, and the area under curve value was 0.795. Conclusion The combination of TyG and NIHSS score is an independent influencing factor for poor short-term prognosis in patients with initial acute ischemic stroke.