Abstract
OBJECTIVE: Whether routine central lymph node dissection (CLND) is necessary for T1-T2 papillary thyroid carcinoma (PTC) patients without certain lateral lymph node metastases (LLNM) remains controversial. This study aims to construct a nomogram that predicts central lymph node metastasis (CLNM) for T1-T2 PTC patients without LLNM. METHODS: We retrospectively reviewed adult T1-T2 PTC patients with no LLNM retrieved from the Surveillance, Epidemiology, and End Results (SEER) database from 2010 to 2015. We also collected data from patients treated at the First Hospital of China Medical University between February and April 2021 for external validation. Logistic regression model was used to construct a risk prediction model nomogram. The receiver-operating characteristic (ROC) curve, calibration plot, and decision curve analyses (DCA) were used for assessing the nomogram. RESULTS: 5,094 patients from the SEER database and 300 patients from our department were finally included in this study. Variables such as age, gender, race, tumor size, multifocality, and minimal extrathyroidal extension (mETE) were found to be associated with CLNM and were subsequently incorporated into our nomogram. The C-index of our constructed model was 0.704, while the internal and external validation C-indexes were 0.693 and 0.745, respectively. The nomogram was then evaluated using calibration and decision curve analyses. CONCLUSION: A visualized nomogram was successfully developed to predict CLNM in T1-T2 PTC patients without LLNM and assist clinicians in making personalized clinical decisions.