Abstract: Objective  To establish a scoring system for predicting central lymph node metastasis （CLNM） in papillary thyroid carcinoma （PTC），then evaluated and verified its work efficiency. Methods Totals of 237 patients with PTC who underwent surgery in Gansu Province Cancer Hospital from June 2015 to June 2018 were selected as the modeling group，and 100 patients met the same criteria from July 2018 to June 2019 were selected as the verification group. Logistic regression model was used to screen independent risk factors and establish scoring system for predicting CLNM，and ROC curve was used to evaluate the predictive value of the scoring system. Results Male，tumor diameter＞1 cm，multifocality，lower location，and capsular invasion were independent predictive factors for CLNM. The area under ROC curve of the scoring system in modeling group was 0.857，and the best cutoff value for the diagnosis of CLNM was 6 points with sensitivity of 81.9%，specificity of 85.3%，positive predictive value of 78.6% and negative predictive value of 87.8%. The area under ROC curve of the verification group was 0.823. Conclusions The scoring system base on independent risk factors has good predictive value of CLNM in PTC. Prophylactic central neck dissection should be actively performed when the PTC patients with score ≥6 points have a high risk of CLNM.

Key words: Papillary thyroid carcinoma, Central lymph node metastasis, Prophylactic central neck dissection, Risk factors, Scoring system