MRI对肾脏占位性病变的定性诊断误诊分析

目的　分析磁共振成像对肾脏占位性病变的定性诊断误诊原因。方法　 89例肾脏占位性病变的患者 ,75例经手术病理证实 ,13例经随访证实。均行MR检查 ,评价MR对这些疾病的定性诊断价值。结果　 89例患者中 ,肾癌 49例 ,肾盂癌 10例 ,肾转移癌 2例 ,肾非何杰金氏淋巴瘤 1例 ,非典型中胚层肾瘤 1例 ,肾腺瘤 2例 ,肾囊肿 8例 ,肾错构瘤 12例 ,黄色肉芽肿性肾盂肾炎 3例 ,肾结核瘤 1例。MR对肾脏恶性占位性病变的定性诊断准确率为 88.9% (5 6/ 63 ) ,其中肾癌误诊 6例 ,肾盂癌误诊 1例。MR对肾脏良性占位性病变的定性诊断准确率为 69.2 % (18/ 2 6) ,其中肾腺瘤误诊 1例 ,肾囊肿误诊 1例 ,肾错构瘤误诊 4例 ,黄色肉芽肿性肾盂肾炎误诊 2例。MR对肾脏占位性病变总的定性诊断准确率为 83 .1% (74/ 89)。MR共误诊 15例 ,误诊的原因有 :病变体积较小4例 ,病变内信号不典型 6例 ,二者均有 3例 ,病灶边界不典型 2例。结论　MR可用于肾脏占位性病变的定性诊断 ,病变体积较小、病变信号和边界特征不典型是MR误诊的主要原因。

Analysis of Misdiagnosis of MRI in the Differential Diagnosis of Renal Masses

Objective To analyse the reason for magnetic resonance imaging in the misdiagnosis of renal masses.Methods 89 patients with pathologically proved diseases (75 cases) and follow-up records (13 cases) were examined with MR. The utility of MR in the differential diagnosis was analyzed. Results The final diagnosis of the 89 patients was as following: 49 renal cell carcinomas, 10 carcinomas of renal pelvis, 2 metastasis, 1 non Hodgkin's lymphoma, 1 atypical fetal renal harmatoma, 2 adenomas, 8 cysts, 12 angiomyolipomas, 3 xanthogranulomatous pyelonephritis and 1 tuberculoma. For the malignant lesions, the qualification accuracy of MR was 88.9% (56/63). The misdiagnosed cases were 6 renal cell carcinomas and 1 renal pelvis carcinoma. For the benign lesions, the qualification accuracy of MR was 69.2% (18/26). The misdiagnosed cases were as following: 1 adenoma, 1 cyst, 4 angiomyolipomas and 1 xanthogranulomatous pyelonephritis and 1 tuberculoma. The total accurate rate of MR in qualification diagnosis of the renal masses was 83.1% (74/89). The reason for misdiagnosis was because of the small size of the lesions (4 cases), or because of the atypical signal intensities (6 cases), or because of the both above reasons (3 cases). The other reason for misdiagnosis of MR was because of the atypical marginal features of the masses (2 cases).Conclusion MR is useful in the qualification of renal masses. The masses of small size with atypical signal and border features are likely to be misdiagnosed.