Abstract
OBJECTIVE: To determine the predictive ability of multifocality for central lymph node metastasis in initially treated 18-55 years old female patients with unilateral papillary thyroid microcarcinoma. STUDY DESIGN: Retrospective review. SETTING: Tertiary medical center. METHODS: We retrospectively collected clinical data from initially treated papillary thyroid microcarcinoma (PTMC) patients at Cancer Hospital Chinese Academy of Medical and sciences between January 1st, 2018, and December 31st, 2018. Data from 975 initially treated 18-55 years old female patients with unilateral PTMC was collected. We also collected data from 340 initially treated 18-55 years old male patients with unilateral PTMC patients to compare the results between genders. Clinicopathological factors associated with central lymph node metastasis (CLNM) were investigated by univariate and multivariate analysis. RESULTS: (1) In the female group, there were 196 (20.1%) cases that had tumor multifocality, including 126 (12.9%) with 2 foci and 70 (7.2%) with >2 foci. The risk of CLNM in patients with 2 foci was not significantly higher than patients with 1 focus (37.3% vs 38.6%, P=0.775). However, diagnosed with >2 foci were independently and positively correlated with CLNM (OR=2.708, 95%CI=1.592-4.607, P<0.001), as was tumor diameter >0.55cm (OR=2.047, 95%CI=1.535-2.730, P<0.001). (2) In the male group, the risk of CLNM with 2 foci was significantly higher than 1 focus (P=0.008). Compared to female patients, the risk of CLNM was significantly higher in patients with 1 focus (P<0.001) or 2 foci (P<0.001). CONCLUSION: In summary, the risk of CLNM in patients with 2 foci was not significantly higher than patients with 1 focus, while multifocality with over 2 foci was an independent risk factor of CLNM. Therefore, multifocality in this subgroup should not be simply defined as "more than 1 focus". Future models that include multifocality as a predictive factor for cervical lymph node metastasis could consider stratifying the cohort into smaller subgroups for more accurate conclusions.