Purpose: Most prognostic systems for differentiated carcinoma have been designed for papillary carcinoma. Objective: To analyze the value of the existing prognostic systems for evaluating follicular carcinoma and to determine whether any of them have a better predictive effect. Methods: A total of 66 follicular carcinomas were analyzed. The following prognostic systems were studied: EORTC, AGES, AMES, MACIS, TNM, and NTCTCS. Results: The AGES and AMES systems did not demonstrate a good prognostic correlation. In the EORTC system, the rate of disease-free patients was 89% in group 1, 75% in group 2, 69% in group 3, and 0% in group 4. The MACIS system showed 83, 60, 67, and 0% of disease-free patients respectively. The TNM system showed 81, 71, 50, and 0% of disease-free patients respectively. Finally, the NTCTCS system demonstrated 100, 84, 53, and 0% of disease-free patients respectively. Cox's regression analysis was used to calculate the proportion of variation in survival time explained (PVE). The prognostic classification system with the greatest survival prediction was EORTC at 67.64% of PVE, followed by TNM at 62.5% of PVE, and MACIS at 57.82% of PVE. Conclusions: MACIS and TNM are good prognostic systems for evaluating follicular thyroid carcinoma, although the one with the most prognostic value was the EORTC system.