To compare the sensitivity, specificity and accuracy of the 2010 and 2017 revisions of the McDonald criteria in a Croatian cohort of patients with a clinically isolated syndrome (CIS).Prospectively collected data from 113 patients were retrospectively analyzed. Sensitivity, specificity and accuracy for both criteria were calculated regarding conversion to clinically definite multiple sclerosis (Poser CDMS) or multiple sclerosis (MS) (defined as fulfilment of clinical or MRI evidence for dissemination in space and the development of a second relapse and/or ≥1 new T2 lesions on the follow-up MRIs) during a two-year follow-up. Survival analysis was performed to estimate the cumulative risk of patients developing Poser CDMS. Binary logistic regression model was used to determine which variables are statistically significant predictors for the conversion to MS.The 2017 revision had higher sensitivity (85 vs. 30% and 85 vs. 41%) and lower specificity (33 vs. 63% and 63 vs. 85%) compared to the 2010 revisions, for conversion to Poser CDMS and MS, respectively. Patients who did not meet the 2017 McDonald criteria had a higher chance of conversion-free survival for Poser CDMS than those who met the 2017 McDonald criteria (p = 0.037). Results of the multivariate regression analysis revealed that patients who at baseline fulfilled 2017 revisions of the McDonald criteria have the increased likelihood of conversion to MS (Exp(B) 9.68, 95%CI 3.62-25.90, p < 0.00001).This study provides new information about the application of the 2017 revisions of the McDonald criteria in a Croatian cohort of patients with typical CIS.