Development and validation of the nomogram based on ultrasound, thyroid stimulating hormone, and inflammatory marker in papillary thyroid carcinoma: a case-control study

Abstract

Background: The increase in the number of thyroid cancer cases in recent years has increased not only the medical burden but also the potential for overtreatment. Therefore, it is crucial to distinguish papillary thyroid cancer from benign thyroid nodules before surgery when treating thyroid nodules.

Methods: The patients were divided into two groups: 117 patients made up the validation cohort and 414 patients made up the primary cohort. As a result of the primary cohort, a preoperative prediction model was developed, which was then validated externally in the validation cohort. Preoperative thyrotropin (thyroid stimulating hormone, TSH), systemic immune-inflammation index (SII), lymphocyte-to-monocyte ratio (LMR), and ultrasonographic features were recorded in both groups.

Results: As predictors for the model, the preoperative blood levels of TSH, SII, LMR, echogenicity, margin, calcification, composition, taller-than-wide, and age were chosen. This was the regression equation: Y = -0.070 × (age) + 1.511 × (echogenicity) + 1.664 × (margin) + 1.003 × (calcification) + 0.939 × (composition) + 2.964 × (tall than wide) + 0.305 × (TSH) + 0.558 × (SII) - 1.271 × (LMR) + 0.327. Papillary thyroid carcinoma (PTC) was predicted positively with values of Y ≥0.808. The prediction model's accuracy, sensitivity, and specificity were 88.2%, 85.1%, and 94.9%, respectively. The area under the receiver operating characteristic (ROC) curve was 0.961. The model's external validation produced satisfactory results with accuracy, sensitivity, and specificity of 85.5%, 90.9%, and 75.5%, respectively.

Conclusions: Using the preoperative TSH, SII, LMR, and ultrasonographic characteristics, a straightforward and accurate preoperative prediction model for PTC has been developed and validated. The preoperative assessment of PTC in clinical application is enhanced by this approach.

Keywords: Papillary thyroid carcinoma (PTC); diagnosis model; lymphocyte-to-monocyte ratio (LMR); systemic immune-inflammation index (SII); thyroid stimulating hormone (TSH).

Figure 1

RCS. Association between continuity variables (age, TSH, SII, and LMR) and papillary thyroid carcinoma. OR, odds ratio; RCS, restricted cubic splines; TSH, thyrotropin; SII, systemic immune-inflammation index; LMR, lymphocyte-to-monocyte ratio.

Figure 2

Developed nomogram. The nomogram was developed in the primary cohort. LMR, lymphocyte-to-monocyte ratio; TSH, thyrotropin; SII, systemic immune-inflammation index.

Figure 3

Calibration curve (A) and ROC curve (B) of nomogram and ROC curve (C) of Thyroid Imaging Reporting and Data System classification in the primary cohort. The solid dot indicates sensitivity and specificity with the maximal value of Youden’s index. ROC, receiver operating characteristic; AUC, area under the curve.

Figure 4

Calibration curve (A) and ROC curve (B) of nomogram in the validation cohort. ROC, receiver operating characteristic; AUC, area under the curve.

Figure 5

Decision curve analysis for the nomogram. The y-axis measures the net benefit. The blue line represents the nomogram. The grey line represents the assumption that all patients are diagnosed with papillary thyroid carcinoma. The black line represents the assumption that no patients are diagnosed with PTC. TC, thyroid carcinoma; PTC, papillary thyroid carcinoma.