We aimed to compare the association of high-sensitivity C-reactive protein (CRP) and National Institutes of Health Stroke Scale (NIHSS) score with mortality risk and to determine the optimal threshold of CRP for prediction of mortality in ischemic-stroke patients. A series of 162 patients with first-ever ischemic-stroke admitted within 24 h after onset of symptoms was enrolled. CRP and NIHSS score were estimated on admission and their predictive abilities for mortality at 7 days were determined by logistic-regression analyses. Receiver-Operating Characteristic (ROC) curves were depicted to identify the optimal cut-off of CRP, using the maximum Youden-index and the shortest-distance methods. Deceased patients had higher levels of CRP and NIHSS on admission (8.87 ± 7.11 vs. 2.20 ± 4.71 mg/l for CRP, and 17.31 ± 6.36 vs. 8.70 ± 4.85 U for NIHSS, respectively, P < 0.01). CRP and NIHSS were correlated with each other (r (2) = 0.39, P < 0.001) and were also independently associated with increased risk of mortality [odds ratios (95 % confidence interval) of 1.16 (1.05-1.28) and 1.20 (1.07-1.35) for CRP and NIHSS, respectively, P < 0.01]. The areas under the ROC curves of CRP and NIHSS for mortality were 0.82 and 0.84, respectively. The CRP value of 2.2 mg/l was identified as the optimal cut-off value for prediction of mortality within 7 days (sensitivity: 0.81, specificity: 0.80). Thus, CRP as an independent predictor of mortality following ischemic-stroke is comparable with NIHSS and the value of 2.2 mg/l yields the optimum sensitivity and specificity for mortality prediction.