Parents
The clinical data of 273 PCT patients undergoing thyroid surgery in the First Affiliated Hospital of Nanjing Medical University from September 2020 to August 2021 were retrospectively collected. All these patients underwent total thyroidectomy or lobectomy, with preventive or therapeutic central lymph node dissection, and some of them underwent lateral cervical lymph node dissection. All patients were reexamined by ultrasound and serum thyroglobulin (TG) at 3 and 6 months after operation. When there was no obvious abnormality in the two examinations, it was considered that there was no lateral cervical lymph node metastasis. Inclusion criteria: (1) two-dimensional ultrasound examination before thyroid surgery with complete examination data, (2) The diagnosis of PTC was confirmed by pathological examination, (3) He did not receive any form of thyroid nodule treatment before ultrasound examination. Exclusion criteria: (1) Patients with other head and neck tumors. (2) There was no follow-up for 3 and 6 months after operation. The recruitment pathway for patients in this study was displayed in Figure 1. This study followed the rules of the Helsinki Declaration and was approved by the Ethics Committee of the First Affiliated Hospital of Nanjing Medical University(2022-SR-512). Because of the retrospective nature of the study, the need for informed consent from the participants was waived by the Ethics Committee of the First Affiliated Hospital of Nanjing Medical University.
Ultrasonic machines and methods
The ultrasonic diagnosis is performed by a senior ultrasonic doctor with more than 5 years of work experience by using the same machine (L15-4 linear-array transducer, Supersonics Aixplorer, Supersonic Imagine, France) and the same settings about gain, depth, frequency. The patient took the supine position, fully exposed the neck area, and the examination sequence was right, isthmus, and left. The probe was perpendicular to the skin surface, and scanned in the transverse and longitudinal sections respectively. The clearest section of the tumor was located and the image was saved. The following parameters were recorded: lob (left or right), number of nodules (single or multiple), the location of the tumor (1.upper or middle or lower pole, 2. medial or middle or lateral,3.dorsal or middle or ventral), size (1.longitudinal diameters, 2.transverse diameters, 3.antero-posterior diameters ), the aspect ratio (< or ≥ 1), boundary (smooth or blurring or angulation), micro-calcification in the nodule (no or yes), blood flow (not obvious or obvious). In this study, the size of thyroid nodules is the largest diameter on each section. In addition, when the number of lesions is large, the ultrasound images of the nodules with the largest diameter or the highest TI-RAS score are used for analysis.
Clinical Characteristics
Clinical characteristics including gender, age, thyroid stimulating hormone (TSH), TG, thyroglobulin antibody (TGAB), and thyroid peroxidase antibody (TPOAB) were collected from the electronic medical records. The results of TSH, TG, TGAB, and TPOAB were detected within 1 week before surgery. According to the clinical experience and previous researches the thresholds set for TSH, TG, TGAB, and TPOAB were as follows: TSH 0.27-4.2 ng/ml, TG<77 ng/ml, TGAB<115 IU/ml, and TPOAB <35 IU/ml. According to preoperative puncture pathology and postoperative pathology, the BrafV600E gene was divided into wild type and mutant type. Some missing values are defined as “Missing”.
Statistical analysis
All statistical analysis was performed by using R software (version 4.2.2) and spss (version 24.0). The queue was randomly divided into training cohort and validation cohort using SPSS. The least absolute contraction and selection operator (Lasso) method was used to select factors. Based on the characteristics of non-zero coefficients in the lasso regression model, the multivariate logistic regression analysis was used to screen independent risk factors to build the prediction model. Use the Concordance index (C-index) and calibration curve to test and evaluate the distinguishability and calibration of the nomogram. In order to quantify the discrimination performance of the nomogram, Bootstrap resampling (1000 times) internal validation was used to calculate the C-index, and decision curve analysis (DCA) was drawn to determine the clinical usefulness of the nomogram. A two-sided p value < 0.05 was considered statistically significant.