摘要
目的:通过病例回顾分析标准摄取值(SUV)和以此为诊断标准时18F—FDG PET灵敏度的影响因素。方法:回顾性分析具有CT不能诊断的肺部实性变而行PET检查的病例,用半定量的方法进行良恶性的鉴别诊断。参照最终的病理或随访结果,计算灵敏度、特异性,绘制工作曲线ROC。分别选用单因素和多因素分析寻找SUV和18F—FDG PET诊断肺癌的灵敏度的影响因素。结果:入选病例121例,最后证实恶性94例,良性27例。以SUV≥2.5为诊断标准,PET正确诊断恶性69例,良性23例,其灵敏度和特异性分别为73.4％和85.2％,细支气管肺泡癌在假阴性病例中占68％。析因结果显示,SUV值受病理类型和病灶大小的影响意义显著,年龄和性别对其影响的意义不显著。结论:病理类型和病灶大小是影响SUV值的两大独立因素,它们影响了以SUV值为诊断标准的灵敏度,细支气管肺癌是影响灵敏度的主要病理类型。
Objective: In order to distinguish the factors that have effects on the value of standard uptake value(SUV) and the diagnostic sensitivity of 18F-FDG PET based on it, we reviewed the patients with undetermined pulmonary lesions who underwent FDG-PET for further diagnosis. Methods: We diagnosed the pulmonary lesions with semiquantitative method. With this method, pulmonary lesions with SUV no less than 2.5 was diagnosed as malignant. Pathological results obtained after tho-ractomy or TTNA was referred to as the definite diagnostic-standard. For cases without pathological results, we referred to the follow-up result instead. The duration of the follow-up was at least one year. Then we calculated the sensitivity, specificity and drawed the ROC. For statistical analysis, we explored single and then multivariance analysis with SPSS. Results: In total, 121 cases were registered in our review. With SUV, we correctly diagnosed 69 of the 94 malignant and 23 of the 27 benign lesions. The sensitivity and specificity was 73.4% and 85.2% respectively. According to the analytical results, we identified pathological pattern and lesion size as the two independent factors that had effects on the value of SUV. Conclusion: Pathological pattern and lesion size had indepent effects on the value of SUV and thus the sensitivity based on it. In this review, bron-chioalveolar carcinoma comprised the most part of the negative cases.
出处
《中国临床医学影像杂志》
CAS
2003年第3期169-172,共4页
Journal of China Clinic Medical Imaging
