Objectives: Stress hyperglycemia, the acute and transient hyperglycemia has been studied in many critical illnesses like stroke and myocardial infarction. THRIVE score proved to be an impressive predictor of in-hospital mortality in previous studies but data on Indian population are lacking. This observational study was conducted to assess the different short term mortality predictors in acute stroke. Materials __ampersandsignamp; Methods: A total of 150 acute stroke patients presenting within 24 hours of onset, who had CT scan evidence and/or met minimum WHO criteria for diagnosis of stroke were included. Patients with TIA, recurrent stroke, secondary hyperglycemia, traumatic hematomas were excluded. Blood glucose and THRIVE score values were obtained on admission. Stroke severity assessed by NIHSS score, while mRS score was used for disability assessment. Results: We divided our study cohort into 4 groups. A cut off blood glucose of 140mg/dl was set for defining hyperglycemia. Out of 71 patients in group C(non diabetics with stress hyperglycemia), 54 expired (76.1%). The burden of stress hyperglycemia was more in ischemic stroke(65.1%) as compared to hemorrhagic stroke (57.1%). In a multivariate model, where age, GCS, NIHSS score were kept as predictors of mortality, stress hyperglycemia had been found to be an independent predictor of in-hospital post stroke mortality. THRIVE score of 6 or above predicted mortality in majority of non survivors. Conclusion: Stress hyperglycemia was found to be a significant poor prognostic determinant. While THRIVE score proves to be an impressive prognostic tool, its validation as independent mortality predictor in acute stroke needs further research.