摘要
目的探讨利用CT良性征象校正FDG假阳性的PET/CT综合分析方法在胰腺癌诊断中的价值。方法胰腺原发良恶性疾患40例,其中胰腺癌27例(手术6例,临床随访21例),急、慢性胰腺炎13例(腹腔镜2例,临床随访11例)。分别用单纯SUV测量法和PET与CT相结合的综合分析法分析,比较两种方法诊断胰腺恶性肿瘤的敏感性、特异性及准确性。SUV测量法以病灶SUVmax值大于2.5为恶性病变诊断指征,反之为良性。PET/CT综合分析法以PET为主导,CT具有否决权,利用PET/CT融合图像的精确对位,当浓聚热点相应CT表现为明显的良性病变征象时,无论FDG浓聚程度如何,均诊断为良性病变。结果综合分析法对胰腺癌诊断的敏感性、特异性、准确性都是100%;单纯SUV测量法敏感性、特异性、准确性分别为96.3%、76.9%、90.0%。结论CT的良性征象,尤其是炎性渗出征象,对FDG假阳性有校正作用,通过这种校正能够提高PET/CT在胰腺癌诊断的特异性和准确性。
Objective We studied the value of synthetic analysis interpretation,which used CT benign feature to correct false positive findings on positron emission tomography/computed tomography(PET/CT)with fluorine-18 labeled fluorodeoxyglucose(FDG)in patients with pancreatic lesion.Methods We examined forty patients whom had primary pancreatic lesions by FDG PET/CT.Among them,27 cases of patients had pancreatic neoplasm proved by surgery(n=6)and clinical follow-up(n=21).Thirteen cases were pancreatitis proved by abdominoscopy(n=2)and clinical follow-up(n=11).PET/CT images were interpreted by simple half quantitative method(the standard uptake value,SUV)and synthetic analysis method,respectively.The sensitivity,specificity and accuracy of the two methods were calculated and compared.The lesion was diagnosed malignant on the simple SUV methods,if the maximum SUV(SUVmax)of lesion greater than 2.5.The synthetic analysis methods combined PET and CT,PET was the luder,but CT can pass PET.If the lesion showed hot spot on PET image,but correspondence CT image showed a benign feature,it will be diagnosed as benign lesion,even if the lesion's SUVmax was greater than 2.5.Results The sensitivity,specificity and accuracy of synthetic analysis methods were 100%.The sensitivity,specificity and accuracy of the simple SUV methods were 96.3%,76.9%,90.0%,respectively.Conclusion Pancreas CT benign feature,especially inflammatory exudation CT findings,can be used to correct some FDG false positive,and can improve diagnosis specificity and accuracy of PET/CT in pancreatic cancer.
出处
《中国医学影像技术》
CSCD
北大核心
2007年第11期1709-1712,共4页
Chinese Journal of Medical Imaging Technology