Previous studies had showed that MTCs patients were more common in females, with single nodule in the majority of cases [9, 10]. In this study, there were no significant differences in age and tumor size between MTMCs and PTMCs. There was a slightly higher number of women than men in MTMCs patients in this study, however, there was also no statistical significance. In accord with previous reports [9, 10], most patients with MTMCs had a single lesion, and only 5 cases had two MTMCs nodules in the present study. PTCs often arise as multiple tumor foci, which are considered as the result of multiple synchronous primary tumors arising from independent clones. The incidence of multifocal PTMCs has been reported as 23.5%-36.1% [11-13]. In our study, multiple PTMCs was detected in 29.81% (31/104) cases, which was higher than that in MTMCs group. Our results showed that there was significantly different in lymph node metastases between MTMCs group and PTMCs group. In PTMCs group, 21 (20.19%) patients had lymph node metastases, the incidence of which was similar to those in previous reports [14-16]. MTMCs were more likely to have lymph node metastases, which indicated the strong invasive characteristics of MTCs. This also suggests that although the incidence of MTCs is very low, early diagnosis is of great significance.
Zhou et al  found that MTMCs had no significant difference from PTMCs in internal composition, calcifications, echogenicity, margin, and shape (P > 0.05). In our study, there were no significant differences in location, margin, boundary, echogenicity, peripheral halo ring, the presence of calcifications between the two groups (P > 0.05). But there were statistic differences in the characteristics as follows: internal composition, shape, calcification types and vascularity, which were not exactly consistent with the results in Zhou’s. The possible reasons may be the different sample size, imaging instrument and evaluation standard of the ultrasonic characteristics.
Cystic change was not common in PTMCs and MTMCs [ 7, 10]. Our results also showed that the majority of cases in both MTMCs (86.96%) and PTMCs (98.53%) exhibited solid composition, however, >50% solid composition was significantly more frequent in MTMCs than in PTMCs. In 46 MTMCs, >50% solid composition was observed in 6 (13.04%) lesions. Although the mechanism of cystic change is unclear, solid-cystic nodules might result from degeneration of solid nodules, with the accumulation of fluid probably due to intranodular necrosis. MTCs tended to display less differentiated and grow faster than PTCs, and cystic change in MTCs was most likely induced by necrosis as a result of cell proliferation that exceeds the available blood supply. Lee et al  also reported that cystic change was significantly more common in MTCs compared with in PTCs, and the tumor size had no significant relationship with cystic changes.
PTC usually had a large intensive fibrosis, and its compressibility was reduced, resulting in its standing-like morphology . A taller-than-wide shape was regarded as an ultrasongraphic feature of PTMCs [18, 19]. In our study, it was also a common sign in PTMCs group, and the proportion was as high as 59.56%. However, there was only 9 (19.57%) cases with a taller-than-wide shape in MTMCs group, which indicated that it did not apply to MTMCs. Previous studies have shown that, regardless of size, the shape of MTCs was mostly round or ovoid rather than taller than wide [20, 21]. Similar results were also observed in our series. The MTMCs nodules were more likely to show ovoid to round comparing with PTMCs, and the difference was statistically significant (P=0.000).
Calcifications, especially microcalcifications, were regarded as a specific sign associated with thyroid malignancy. Kim et al  found that calcifications were not significantly different between PTCs and MTCs. The results of our study showed that calcifications were not common in both MTMCs group and PTMCs group, and the incidence of calcification was similar (P = 0.674). However, the type of calcification was different between the two groups. We found that microcalcifications were frequently seen in PTMCs, and macrocalcifications appeared more commonly in MTMCs. This might be related to the different mechanisms of calcification formation in MTCs and PTCs. Calcifications in MTCs are mainly due to the deposition of local calcium salts surrounded by amyloid substances, leading to the formation of coarse calcifications. However, calcifications in PTCs are mainly caused by psammoma bodies with the smaller diameter of 10~100 μm, which are commonly manifested in round or concentric under the light microscope.
To the best of our knowledge, there were very limited literatures reported the blood supply of MTCs. Trimboli et al  reported that the percentage of intranodular vascularization in MTCs (25%) was higher than that in PTCs (15%), but with no significant difference. However, in the present study, MTMCs were more likely to show hypervascularity than PTMCs (P=0.000). PTMCs were predominantly characterized by reduced blood supply, which might be due to the incompletely developed neoangiogenetic vascular bed in comparison with the uncontrolled cell proliferation typical of carcinoma . In the present series, 80.15% of PTMCs were deficient in vascularity. According to the study reported by Lai et al , 72.4% of MTCs showed hypervascularity, which might be induced by the rapid division and growth of MTCs cells, resulting in a faster rate of proliferation than PTCs. In this study, abundant blood supply was also commonly detected in MTMCs (58.70%), but the proportion was not as high as reported in the literature. This may be due to the relatively small number of blood vessels caused by the small size of the tumor.
There were some limitations in our study. First, this was a retrospective study with statistical bias. Second, because of the low incidence of MTCs, it took a long time to collect MTMCs cases. During this time, there has been a significant development in US technology, which is likely to have an impact on the evaluation of sonographic features. In addition, the sample size was small in MTMCs group, although there were more cases than those in the previous studies. A further study with a larger sample size should be performed to identify the sonographic features of MTMCs.