Predictors and a prediction model for positive fine needle aspiration biopsy in C-TIRADS 4 thyroid nodules

Abstract

Objectives: To screen out the predictors and establish a prediction model of positive fine needle aspiration biopsy (FNAB) in the Chinese Guidelines for Malignant Risk Stratification of Thyroid Nodule Ultrasound (C-TIRADS) 4 thyroid nodules, and this nomogram can help clinicians evaluate the risk of positive FNAB and determine if FNAB is necessary.

Methods: We retrospectively analyzed data from 547 patients who had C-TIRADS 4 thyroid nodules and underwent fine-needle aspiration biopsy (FNAB) at the Second Affiliated Hospital of Chongqing Medical University between November 30, 2021 and September 5, 2022. Patients who met our inclusion criteria were divided into two groups based on positive or negative FNAB results. We compared their ultrasound (US) features, BRAF V600E status, thyroid function, and other general characteristics using univariate and multivariate logistic regression analyses to identify independent predictors. These predictors were then used to construct a nomogram. The calibration plot, area under the curve (AUC), and decision curve analysis were employed to evaluate the calibration, discrimination, and clinical utility of the prediction model.

Results: Out of 547 patients, 39.3% (215/547) had a positive result on fine-needle aspiration biopsy (FNAB), while 60.7% (332/547) had a negative result. Univariate logistic regression analysis revealed no significant differences in TPOAb, TgAb, TSH, Tg, nodule location, sex, or solid status between the two groups (P>0.05). However, age, nodule size, internal or surrounding blood flow signal, microcalcifications, aspect ratio, morphology, and low echo showed significant differences (P<0.05). Multivariate logistic regression analysis was conducted to explore the correlation between potential independent predictors. The results showed that only age (OR=0.444, 95% Cl=0.296~0.666, P<0.001), low echo (OR=3.549, 95% Cl=2.319~5.432, P<0.001), microcalcifications (OR=2.531, 95% Cl=1.661~3.856, P<0.001), aspect ratio (OR=3.032, 95% Cl=1.819~5.052, P<0.001), and morphology (OR=2.437, 95% Cl=1.586~3.745, P<0.001) were independent predictors for a positive FNAB. These variables were used to construct a prediction nomogram. An ROC curve analysis was performed to assess the accuracy of the nomogram, and AUC=0.793, which indicated good discrimination and decision curve analysis demonstrated clinical significance within a threshold range of 14% to 91%.

Conclusion: In conclusion, 5 independent predictors of positive FNAB, including age (≤45 years old), low echo (yes), microcalcifications (yes), aspect ratio (>1) and morphology (irregular), were identified. A nomogram was established based on the above 5 predictors, and the nomogram can be used as a complementary basis to help clinicians make decisions on FNAB of C-TI-RADS 4 thyroid nodules.

Keywords: 95% Cl=0.296~0.666; C-TIRADS; fine needle aspiration biopsy; low echo (OR=3.549); prediction model 444; predictor; thyroid nodules.

Figure 1

(A) Schematic diagram of ultrasound (US)-guided fine needle aspiration biopsy (thyroid nodule). (B) US image of fine needle aspiration (thyroid nodule).

Figure 2

(A, C) A C-TIRADS 4A thyroid nodules patient (31-year-old male) with positive fine needle aspiration biopsy (FNAB). (A) The thyroid nodule had microcalcifications on US images. (C) Cytological examination indicated suspicious malignant tumor. (B, D) A C-TIRADS 4A thyroid nodules patient (40-year-old female) with negative FNAB. (B) The thyroid nodule had irregular morphology on US images. (D) Cytological examination indicated follicular epithelial cells.

Figure 3

(A, C) A C-TIRADS 4B thyroid nodules patient (25-year-old female) with positive fine needle aspiration biopsy (FNAB). (A) The thyroid nodule had a low echo and irregular morphology on US images. (C) Cytological examination indicated suspicious malignant tumor. (B, D) A C-TIRADS 4B thyroid nodules patient (36-year-old male) with negative FNAB. (B) The thyroid nodule had a low echo and irregular morphology on US images. (D) Cytological examination indicated inflammatory.

Figure 4

(A, C) A C-TIRADS 4C thyroid nodules patient (32-year-old female) with positive fine needle aspiration biopsy (FNAB). (A) The thyroid nodule had a low echo, aspect ratio >1 and irregular morphology on US images. (C) Cytological examination indicated suspicious malignant tumor. (B, D) A C-TIRADS 4C thyroid nodules patient (28-year-old female) with negative FNAB. (B) The thyroid nodule had a low echo, aspect ratio >1 and irregular morphology on US images. (D) Cytological examination indicated inflammatory cell, follicular epithelial cells and gelatine.

Figure 5

A prediction nomogram. The nomogram is used for the prediction of positive FNAB in C-TIRADS 4 thyroid nodules patients. The prediction nomogram was developed in the cohort, with age, aspect ratio, low echo, microcalcifications and morphology.

Figure 6

Calibration curves of the prediction nomogram. The solid line is close to the diagonal dotted line, which represents a good prediction ability. The x-axis represents the predicted positive FNAB. The y-axis represents the actual positive FNAB. The diagonal dotted line represents the perfect prediction by the ideal model. The solid line represents the performance of the nomogram, and a closer fit to the diagonal dotted line means a better prediction.

Figure 7

(A) The receiver operating characteristics (ROC) curve and area under the curve (AUC) in the training cohort. (B) ROC curve and AUC in the validating 1 cohort. (C) ROC curve and AUC in the validating 2 cohort. Validating 1 and validating 2 were performed to evaluate the accuracy of this model.

Figure 8

Decision curve analysis (DCA) for the risk nomogram. DCA shows that the model is clinically useful when intervention is decided in the threshold range of 14% to 91%.