SPECT/螺旋CT融合显像鉴别诊断颅骨良恶性病变的临床价值

摘要

目的:探讨SPECT/螺旋CT融合显像鉴别诊断颅骨良恶性病变的临床价值.方法:对19例恶性肿瘤患者全身骨扫描(WBS)显示颅骨放射性浓聚的21个病灶行SPECT/CT融合显像,由两名有经验的核医学科医生独立阅片.患者同期接受MRI和/或CT检查,各检查时间间隔不超过两周,并对所有病例进行随访,平均随访约9个月,随访结果主要根据随后的影像学检查(MRI、CT或随后的骨扫描)及临床信息.结果:根据同期影像及随访检查共10个病灶诊断为恶性病变,11个病灶诊断为良性病变.SPECT/CT融合显像检出全部10个恶性病灶及8个良性病变,诊断符合率为85.71％.1个病灶在SPECT/CT融合显像中诊断为恶性,而同期MRI及随访检查显示良性病变.另有两个病灶在SPECT/CT中难以确定其性质,同期影像及随访检查亦显示为良性.SPECT/CT融合显像在肯定性诊断为良、恶性病灶中诊断颅骨恶性病变的敏感性、特异性、阳性预测值、阴性预测值分别为100.00％、88.89％、90.91％、100.00％.结论:SPECT/螺旋CT融合显像对鉴别诊断颅骨良恶性病变有较高的敏感性和特异性.%Objective: To evaluate the value of single photon emission computed tomography (SPECT)/spiral computed to-mography (CT) fusion imaging in the differential diagnosis of benign and malignant skull lesions. Methods: Nineteen cancer patients with 21 skull lesions on whole-body bone scan (WBS) were recruited and underwent SPECT/CT fused imaging of the head. The images were evaluated by two independent reviewers. The patients underwent MRI and/or CT examinations within two weeks after SPECT/CT. An average follow-up interval after SPECT/CT was 9 months to correlate with additional radiologic imaging studies and clinical informations. Results: Ten lesions were classified as metastatic lesions and 11 lesions were classified as benign based on the other radiological examinations performed at the same period and follow-up. The SPECT/CT correctly identified 10 out of 10 metastatic lesions and 8 out of 11 benign lesions, 85.71% of lesions were correctly classified. One lesion was classified incorrectly as metastatic lesion with SPECT/CT while it was proven benign by other imaging modalities and follow-up. Two lesions remained indeterminate even after the SPECT/CT examination and were confirmed benign by other imaging modalities and follow-up. The sensitivity, specificity, positive and negative predictive values of SPECT/CT images in the diagnosis of malignant skull lesions were 100.00%, 88.89%, 90.91% and 100.00%, respectively. Conclusions: SPECT/ spiral CT can help identify benign versus metastatic skull lesions with high sensitivity and specificity.