甲状腺结节的CT灌注评价

The evaluation of perfusion CT imaging in thyroid nodule

目的探讨甲状腺结节的CT灌注成像价值。方法搜集经手术后病理证实的甲状腺结节40例，其中结节性甲状腺肿22例，甲状腺腺瘤6例，甲状腺癌12例，术前行CT灌注扫描，计算病灶ROI的血流量（BF）、血容量（BV）、平均通过时间（MTT）和表面渗透性（PS）。经两样本比较的秩和检验，比较各灌注参数值的差异性，然后将有统计学意义的参数值进行判别分析，建立判别函数，对甲状腺良恶性结节进行判别。结果28例甲状腺良性结节MTT中位数为4．33s（范围1．42～10．93s），12例恶性结节MTF中位数为2．18S（范围1．95～2．87S），差异有统计学意义（P=0．00）。甲状腺癌BF中位数为560．23（范围330．66—1000．00）ml·100g^-1·min^-1良性结节BF中位数为374．79（范围117．47～1000．00）ml·100g^-1·min^-1，差异有统计学意义（P=0．01）。经判别分析，良、恶性结节之间MTF的差异对结节性质判别有统计学意义（P=0．00），以此建立判别函数，经Bayes判别分析，总体诊断准确度为87．5％（35／40）。结论甲状腺良、恶性结节的CT灌注扫描中，MTT和BF对其鉴别诊断具有价值，但以MTT的诊断准确度更高。

Objective To investigate the value of perfusion CT imaging in thyroid nodule. Methods Of the 40 cases of post-surgical thyroid nodule identified by pathology, nodular goiter was verified in 22 cases, thyroid adenoma in 6 cases, and thyroid carcinoma in 12 cases. All cases underwent CT perfusion scan at preoperative. The perfusion parameters including the blood flow （BY）, blood volume （BV）, mean transit time （MTT）, and permeability of surface （PS） of region of interest （ROI） were calculated. The Mann-whitney lest was used to comparing the differences of the different perfusion parameters. The ones with statistical significance would be introduced into the discriminatory analysis to distinguish the benign and malignant thyroid nodule. Results The median MTT in 28 cases with benign thyroid nodule were 4. 33 s （Min 1.42 s,Max 10. 93 s）, and that in 12 cases with malignant nodules were 2. 18 s（ Min 1.95 s, Max 2. 87 s）. The difference had statistical significance （ P = 0. 00 ） . The median BF in cases with thyroid carcinoma was 560. 23 （Min 330. 66, Max 1000. 00） ml ·100g^-1·min^-1, and that in cases with the benign nodular were 374. 79 （ Min 117. 47, Max 1000. 00 ） ml ·100g^-1·min^-1 There was a significant statistical difference （ P = 0. 01 ）. Through the discriminatory analysis, the difference of MTT between benign and malignant nodules were statistically significant （ P = 0. 00 ）. After establing the Bayes discriminatory function, the overall diagnostic accuracy was 87.5% （35/40）. Conclusions MTr and BF are useful parameters in CT perfusion imaging to distinguish the benign and malignant thyroid nodules and the MTT is more accuracy.