Abstract: Objective To investigate the correlation between ultrasonographic features, tumor size of medullary thyroid carcinoma, and the causes of misdiagnosis. Methods The clinical data and ultrasonographic features of 48 patients （60 lesions in total） with medullary thyroid carcinoma confirmed by pathology were retrospectively analyzed. Results According to the diameter, the 60 lesions were divided into group A （n=16, <1 cm）, group B （n=38, <4.0 cm） and group C （n=6, >4.0 cm）. 81.25% of the lesions in group A were misdiagnosed as papillary carcinoma; in group B, 65.79% of the lesions were misdiagnosed as papillary carcinoma and 15.79% as follicular carcinoma; in group C,33.3% of the lesions were misdiagnosed as papillary carcinoma and 33.3% as undifferentiated carcinoma. There were statistically significant differences in tumor morphology, acoustic volume, orientation, posterior echo, and blood flow among 3 groups of medullary carcinomas （all P values <0.05）, pairings between groups showed that there were two ultrasonic signs （posterior echo and blood flow） in group A and group B, and there was one ultrasonic sign （blood flow） in group A and group C, and the difference was statistically significant （P<0.012 5）. Cervical lymph node metastasis （presence, absence, single region, multiple regions） was not correlated with tumor diameter （P>0.05）. Conclusions Ultrasonographic features of medullary thyroid carcinoma were correlated with tumor size. Enhancing the clinical cognition of medullary carcinoma was beneficial in improving the accuracy of ultrasonic diagnosis.