Horm Metab Res 2018; 50(07): 543-550
DOI: 10.1055/a-0630-5239
Endocrine Care
© Georg Thieme Verlag KG Stuttgart · New York

A Risk Score Model for Evaluation and Management of Patients with Thyroid Nodules

Yongwen Zhang
1  Department of Endocrinology, Nanjing Integrated Traditional Chinese and Western Medicine Hospital Affiliated with Nanjing University of Chinese Medicine, Nanjing, China
,
Fanrong Meng
2  Department of Radiology, Nanjing Integrated Traditional Chinese and Western Medicine Hospital Affiliated with Nanjing University of Chinese Medicine, Nanjing, China
,
Lianqing Hong
3  Department of Pathology, Nanjing Integrated Traditional Chinese and Western Medicine Hospital Affiliated with Nanjing University of Chinese Medicine, Nanjing, China
,
Lanfang Chu
4  Department of Integrated Traditional Chinese and Western Medicine, Nanjing General Hospital of Nanjing Military Command, Nanjing, China
› Author Affiliations
Further Information

Publication History

received 10 December 2017

accepted 08 May 2018

Publication Date:
12 June 2018 (online)

Abstract

The study is aimed to establish a simplified and practical tool for analyzing thyroid nodules. A novel risk score model was designed, risk factors including patient history, patient characteristics, physical examination, symptoms of compression, thyroid function, ultrasonography (US) of thyroid and cervical lymph nodes were evaluated and classified into high risk factors, intermediate risk factors, and low risk factors. A total of 243 thyroid nodules in 162 patients were assessed with risk score system and Thyroid Imaging-Reporting and Data System (TI-RADS). The diagnostic performance of risk score system and TI-RADS was compared. The accuracy in the diagnosis of thyroid nodules was 89.3% for risk score system, 74.9% for TI-RADS respectively. The specificity, accuracy and positive predictive value (PPV) of risk score system were significantly higher than the TI-RADS system (χ2=26.287, 17.151, 11.983; p <0.05), statistically significant differences were not observed in the sensitivity and negative predictive value (NPV) between the risk score system and TI-RADS (χ2=1.276, 0.290; p>0.05). The area under the curve (AUC) for risk score diagnosis system was 0.963, standard error 0.014, 95% confidence interval (CI)=0.934–0.991, the AUC for TI-RADS diagnosis system was 0.912 with standard error 0.021, 95% CI=0.871–0.953, the AUC for risk score system was significantly different from that of TI-RADS (Z=2.02; p <0.05). Risk score model is a reliable, simplified and cost-effective diagnostic tool used in diagnosis of thyroid cancer. The higher the score is, the higher the risk of malignancy will be.