Despite the association between high-sensitivity C-reactive protein (CRP) and recurrent events in non-ST elevation acute coronary syndromes (ACS), routine determination of this marker has not been recommended. In order to verify whether the current scientific evidence justifies the inclusion of CRP for risk stratification at hospital admission of patients with ACS, we carried out a systematic review and meta-analysis of the studies indexed in MEDLINE, SciELO or LILACS, with the following inclusion criteria: prospective cohort design and assessment of the prognostic value of CPR, as measured using a high-sensitivity method at the moment of hospital admission of patients with ACS. Nineteen studies met the inclusion criteria. In relation to the long-term follow-up, there was a consistent association between CRP and cardiovascular events, with an overall odds ratio (OR) of 4.6 (95% CI = 2.3 - 7.6) and overall multivariate OR of 2.5 (95% CI = 1.8-3.4). As for the short-term, nine studies were positive and six were negative, with an overall OR of 1.65 (95% CI = 1.2-2.3). The overall multivariate OR was not obtained for the short-term follow-up, because this measurement was described only in three heterogeneous studies. Only two short-term studies analyzed the incremental predictive value of CRP in relation to multivariate models, with contradicting results. In conclusion, the small number of assessments of the incremental value of CRP, in conjunction with controversial results regarding the independent predictive value of CRP for short-term events does not support the recommendation of the routine use of CRP for risk stratification at admission of patients with ACS.