^(18)F-FDG PET/CT显像对甲状腺病变的应用价值研究

The application value by ^(18)F-FDG PET/CT imaging in thyroid lesions

目的初步分析和探讨^(18)F-FDG PET/CT显像对甲状腺病变的评估价值。方法回顾性分析2014年10月～2017年10月检查发现甲状腺病变并有明确病理结果的18例患者资料,以同一患者对侧正常甲状腺或周围颈部血池为健康对照组,分别对甲状腺恶性病变组、良性病变组和健康对照组的SUVmax进行统计并进行差异性分析,对甲状腺良、恶性病变SUVmax行ROC曲线分析,同时观察病灶的CT形态及密度改变,综合分析并计算PET、PET/CT诊断甲状腺病变的效能指标。结果 1)18例患者共21个病灶,恶性病变14个,良性病变7个,甲状腺恶性病变组、良性病变组和健康对照组的SUVmax分别为10.12±9.42、5.47±2.97和1.43±0.38,甲状腺恶性病变组与良性病变组SUVmax差异无统计学意义(F=0.635,P=0.435),甲状腺恶性、良性病变组与健康对照组SUVmax差异均有统计学意义(F=10.748和18.270,均有P<0.01);2)分析甲状腺良、恶性病变SUVmax的ROC,该指标的AUC为0.515(<0.70),诊断准确性不高;3)^(18)F-FDG PET扫描诊断甲状腺良、恶性病变的灵敏度、特异性、阳性预测值、阴性预测值和准确性分别71.43%(10/14)、42.86%(3/7)、71.43%(10/14)、42.86%(3/7)、61.90%(13/21),而^(18)F-FDG PET/CT扫描的诊断效能分别为92.86%(13/14)、71.43%(5/7)、86.67%(13/15)、83.33%(5/6)、85.71%(18/21)。^(18)F-FDG PET扫描诊断甲状腺良、恶性病变的功效与标准组一致性很差(Kappa=0.143,μ=0.644,P=0.513),而^(18)F-FDG PET/CT扫描与标准组基本一致(Kappa=0.762,μ=4.792,P<0.01)。结论 ^(18)F-FDG PET/CT显像中SUVmax不能作为甲状腺病变良恶性鉴别诊断指标,同时结合PET代谢特点和CT形态学改变,可以初步诊断甲状腺病变性质,并且能为临床提供更多信息如转移或者合并其他良恶性疾病。

Objective To initially investigate the value of 18F-FDG PET/CT imaging in the evaluation of thyroid lesions. Methods The data of 18 patients with thyroid lesions and pathological results were analyzed retrospectively, who were found with such lesions in inspection from October 2014 to Dctober 2017. Patients with normal thyroid or peripheral blood pool on same side were taken as the healthy control group, and the analysis was made on the SUVmax of the malignant lesion group, the benign lesion group and the healthy control group, and on the ROC curve of SUVmax of benign and malignant thyroid lesions. The CT morphology and density of the lesions were observed. The comprehensive analysis and computation were also made on the value of PET, PET/CT in the diagnosis of thyroid lesions. Results 1) There were 21 lesions(including 14 malignant lesions and 7 benign lesions) in 21 patients. SUVmax of thyroid malignant group, benign lesion group and healthy control group were 10.12±9.42, 5.47±2.97, and 1.43±0.38, respectively. There was no significant difference in SUVmax between thyroid malignant group and benign lesion group(F=0.635, and P=0.435), and there was a significant difference in SUVmax between thyroid malignant group, benign lesion group and healthy control group(F=10.748 and 18.270, P<0.01);2) Analysis of SUVmax ROC of thyroid benign and malignant lesions was AUC 0.515(<0.70), so the diagnostic accuracy was low;3) The sensitivity, specificity, positive predictive value, negative predictive value and accuracy of 18F-FDG PET for the differentiation of thyroid benign and malignant lesions were 71.43%(10/14), 42.86%(3/7), 71.43%(10/14), 42.86%(3/7), and 61.90%(13/21), respectively. The diagnostic value of 18F-FDG PET/CT was 92.86%(13/14), 71.43%(5/7), 86.67%(13/15), 83.33%( 5/6),85. 71%( 18/21),respectively.18F-FDG PET was very poorly consistent with the standard group in diagnostic value of benign and malignant lesions( Kappa = 0. 143,μ = 0. 644,P = 0. 513),while18F-FDG PET/CT was basically consistent with the standard group( Kappa = 0. 762,μ = 4. 792,P < 0. 01). Conclusion SUVmax in 18F-FDG PET/CT imaging cannot be used as a diagnostic parameter of benign and malignant thyroid disease. The combination of PET metabolic characteristics and CT morphological changes can initially identify the nature of thyroid lesions,and provide more clinical information such as the metastasis or concurrent with other benign and malignant diseases.