动态增强MRI在孤立性肺结节鉴别诊断中的应用

Dynamic contrast-enhanced magnetic resonance imaging in differential diagnosis of solitary pulmonary nodules

目的探讨动态增强MRI的血流动力学参数对孤立性肺结节性质的鉴别诊断价值。材料与方法搜集2013年8月至2014年10月本院胸外科62例孤立性肺结节患者(恶性结节42例,炎性结节7例,良性结节13例),分别行MRI肺部常规平扫和动态增强扫描,动态增强扫描采用三维快速容积扫描技术(LAVA)以30 s每期共扫描15期。计算时间-信号曲线中第1 min、2 min、4 min信号增强比(SI1%、SI2%、SI4%)和最大信号增强比(MER),达到峰值信号的时间Tmax,曲线最大上升斜率(SS),以及廓清率(washout)。结果恶性结节SI1%、SI2%、SI4%、MER、SS显著高于良性结节而低于炎性结节(P<0.05)。以MER为2.23、0.63分别作为鉴别炎性结节和恶性结节、恶性结节和良性结节的临界值(>2.23为炎性病变,≤0.63为良性病变),恶性病变与另外两组得以鉴别的敏感度、阴性预测值、准确度分别可达100%、100%、92%。结合病变形态学特征,可以进一步提高诊断准确度。结论应用动态增强MRI血流动力学参数和形态学特征,鉴别肺部恶性、良性、炎性结节有很高的准确性。

Objective： This study is to assess the value of hemodynamic parameters of DCE-MR imaging in differentiating solitary pulmonary nodules（SPNs）. Materials and Methods： A total of 62 patients （42 malignant nodules, 13 benign nodules and 7 active infections） with SPNs underwent DCE-MR1 exams before bolus injection and after 15 phases since bolus injection of Gd-BOPTA, at the rate of 30 seconds. From the temporal-signal curves, the following analysis were taken： signal enhancement ratios at 1, 2, 4 minutes after bolus injection （SI1%, SI2%, SI4%） and the maximum enhancement ratio（ MER ）; time at the maximum enhancement ratio （Tmax）; the steepest slope of time-signal curves （SS） and washout ratio. Results： It was found that SI1%, SI2%, SI4%, MER, SS for the malignant SPN group were significantly higher than those for the benign SPN group and significantly lower than those for the active infection group （P〈0.05）. With 2.23 and 0.63 as the threshold maximum enhancement ratio for distinguishing malignant nodules from benign nodules and active infection, the sensitivity, nagative predictive value, accuracy is 100%, 100%, 92%, respectively. Diagnostic accuracy is further improved combined with morphologic features of nodules. Conclusions： The hemodynamic parameters and morphologic characteristics of DCE-MRl may be a useful tool in differentiating malignant SPN from active infection SPN, as well as benign SPN.