腮腺多形性腺瘤MRI特征评分对定性诊断的有效性评估

Diagnostic Efficacy of MRI Characteristic for Parotid Pleomorphic Adenoma

目的评估腮腺多形性腺瘤（PA）的特征性MRI表现对术前定性诊断的有效性。方法回顾性分析经手术病理证实并于术前行MRI扫描的腮腺肿瘤患者87例,将全部病例分为PA组（37例）和非PA组（50例）。分别对两组肿瘤的形态、边界、T_2WI信号、包膜、强化特点、ADC值赋予分值,分析其差异。建立改良多形性腺瘤MRI特征评分系统,并判断该评分系统诊断PA的准确率。结果两组肿瘤的形态、边界、T_2WI信号、包膜、强化特点、ADC值有明显的差异。采用改良多形性腺瘤MRI特征评分诊断PA的敏感度64.86%,特异度94.00%,阳性预测值88.89%,阴性预测值78.33%,准确率81.64%。结论完全符合多形性腺瘤MRI特征评分的腮腺肿瘤,术前定性诊断有较高的特异度和准确率。但仍有一部分PA影像表现不典型,与其他肿瘤存在一定的重叠,需要进一步检查定性。

Objective To evaluate the diagnostic efficacy of an MRI algorithm characteristic of pleomorphic adenoma（ PA）. Methods MR of 87 cases with parotid gland tumors confirmed by histopathology were analyzed retrospectively.These 87 cases were divided into 2 groups including PA（ n = 37） and other parotid gland tumor（ n = 50）. Six of the MR imaging features characteristic for PA（ lobulated contours,sharp margins,bright T2-signal,T2-dark rim,heterogeneous nodular enhancement,ADC value） of these 2 groups were compared and analyzed. MRI scores of each patient were acquired using a revised MRI algorithm system,which was based on initial scoring system of five＂high probability＂criteria,and added the scores of the ADC value. We then identified MRI images to test the diagnostic accuracy of the proposed algorithm. Results There were statistical differences in these MR imagingcharacteristic features between the two groups. Twenty-four of 27 cases that met all six ＂high probability＂MRI criteria were consistent with PA on final histopathology. MRI had a specificity of 94. 00% and sensitivity of 64. 86% for PA. The positive predictive value was 88. 89% and the negative predictive value was 78. 33%. The overall diagnostic accuracy was 81. 64%. Conclusion The revised MRI algorithm characteristic of PA is about 94% specific for pleomorphic adenoma. A subset of patients with MR imaging that is highly suggestive of PA may reliably avoid further workup. A significant proportion of PAs,however,have indeterminate imaging features that overlap considerably with other benign and malignant lesions.