Establishing a prediction model for lateral neck lymph node metastasis in patients with papillary thyroid carcinoma.

This study aimed to establish a model for predicting lateral neck lymph node metastasis in patients with papillary thyroid carcinoma. A total of 106 patients with papillary thyroid carcinoma undergoing initial treatment of neck lymph node dissection (including central and lateral neck lymph nodes) at the thyroid surgery department were enrolled from January 2009 to April 2017. Logistic regression analysis was used to screen the factors influencing lateral neck lymph node metastasis and develop a prediction model. The receiver operating characteristic curve was used to evaluate the predictive power and boundary value of the model for lateral neck lymph node metastasis. Prediction model: Logistic(P) = -5.699 + 0.681 × (TSH)  + 0.342 × (Metastatic rate of central lymph nodes)  + 1.463 × (Combined with Hashimoto's disease)  + 1.525 × (Number of tumors) . When logistic (P) was ≥ 0.821, it was predicted that lateral neck lymph node metastasis occurred in patients with papillary thyroid carcinoma. When logistic (P) was <0.821, it was predicted that no metastasis was found in the lateral neck lymph node. The prediction accuracy was 78.3%. The model helped in evaluating lateral neck lymph node metastasis in patients with papillary thyroid carcinoma. Also, it had significance in determining reasonable surgical range, reducing unnecessary lateral neck lymph node dissection, and further improving the quality of life of patients.