Combined Fine-Needle Aspiration and Selective Intraoperative Frozen Section to Optimize Prediction of Malignant Thyroid Nodules: A Retrospective Cohort Study of More Than 3000 Patients
Zhuochao Mao,Yongfeng Ding,Liping Wen,Yu Zhang,Guofa Wu,Qihan You,Jie Wu,Dingcun Luo,Lisong Teng,Weibin Wang
DOI: https://doi.org/10.3389/fendo.2023.1091200
IF: 6.055
2023-01-01
Frontiers in Endocrinology
Abstract:Background Preoperative fine-needle aspiration (FNA) is widely used to differentiate malignant from benign thyroid nodules, while intraoperative frozen sections (FS) are suggested as a systematic supplement for intraoperative decision-making, but limitations still remain for both procedures. Methods Medical records of 3807 patients with thyroid nodules who underwent both pathological diagnoses (FS and FNA) at our hospital were reviewed. The diagnostic accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of FNA and FS were also evaluated. We further designed an optimal integration scheme (FNA+selective FS) to predict thyroid nodule malignancy. Finally, the efficiency of the proposed integrated diagnostic model was validated using an independent external cohort. Results For distinguishing malignant nodules, FNA had an accuracy of 90.3%, sensitivity of 90.7%, specificity of 85.2%, PPV of 98.8% and NPV of 40.4%. In contrast, the FS represented higher discriminative power (Accuracy, 94.5%; Sensitivity, 94.1%; Specificity, 100%; PPV, 100%; and NPV, 55.6%). we proposed the selective usage of FS (removed nodules with Bethesda category VI from routine FS, ~1/3 of total). The integrated new diagnostic model of FNA plus selective FS (FNA+sFS) achieved accuracy of 96.9%, sensitivity of 97.3%, specificity of 92%, PPV of 99.4%, and NPV of 71.6% (NRI=0.135, 95% CI 0.103-0.167, P <0.001) and was successfully applied to an external cohort (N=554). Conclusion Compared with the FNA diagnostic system, FS has an increased ability to distinguish benign and malignant thyroid nodules. The newly proposed integrated diagnostic model of FNA + selective FS can optimize the accuracy of diagnosis.