常规超声联合超声造影对胰腺浆液性囊腺瘤与黏液性囊性肿瘤的鉴别诊断

Value of conventional ultrasound combined with contrast-enhanced ultrasound in differential diagnosis of serous cystic adenoma and mucinous cystic neoplasm of the pancreas

目的探讨常规超声及超声造影(CEUS)对胰腺浆液性囊腺瘤(serous cystic adenoma,SCA)和黏液性囊性肿瘤(mucinous cystic neoplasm,MCN)的鉴别诊断价值。方法回顾性分析浙江省人民医院2016年6月至2020年12月经手术病理证实的86例胰腺SCA和MCN的一般临床资料及常规超声和CEUS影像资料,对其进行单因素及多因素Logistic回归分析,并建立Logistic回归模型,绘制回归模型鉴别SCA与MCN的ROC曲线,评价常规超声及CEUS特征对SCA与MCN的鉴别诊断价值。结果86例胰腺囊性肿瘤病例中SCA共58例,MCN共28例。单因素分析结果显示:SCA与MCN的性别、肿瘤生长部位、生长形态、临床症状、CEUS动脉期增强强度及强化模式比较,差异均无统计学意义(P均>0.05);2组的发病年龄、瘤体最大径、病灶内部囊腔是否≥6个、有无钙化及钙化在病灶内的分布情况、单个囊腔的最大径是否≥20 mm、囊壁厚度及均匀性、囊壁结节的有无及CEUS静脉期增强强度比较,差异均有统计学意义(P均<0.05)。Logistic多因素回归分析结果显示:囊腔个数≥6个(OR=0.163,95%CI:0.047~0.561)、囊壁增厚或厚度不均(OR=9.388,95%CI:2.663~33.094)及单囊腔直径≥20 mm(OR=3.254,95%CI:1.037~10.213)是诊断MCN的独立危险因素,建立Logistic回归模型为:Logit(P)=-2.792-1.815X3(囊腔个数)+2.239X5(囊壁特征)+1.180X6(单囊腔直径),ROC曲线显示该回归模型鉴别SCA与MCN的曲线下面积为0.823(95%CI:0.734~0.913)。结论胰腺SCA和MCN具有各自的超声影像学特征,应用常规超声及CEUS可以较好地对二者进行鉴别,为临床进一步治疗提供依据。

Objective To investigate the routine ultrasound and contrast enhanced ultrasound(CEUS)manifestations of serous cystic adenoma(SCA)and mucinous cystic neoplasm(MCN)of the pancreas,in order to improve their diagnostic accuracy.Methods The clinical data and conventional ultrasound and CEUS imaging data of 86 cases of SCA and MCN confirmed by surgery and pathology at Zhejiang Provincial People's Hospital from June 2016 to December 2020 were retrospectively analyzed.Univariate and multivariate Logistic regression analyses were used to evaluate the diagnostic value of clinical data as well as conventional ultrasound and CEUS characteristics in SCA and MCN.Results There were 58 cases of SCA and 28 cases of MCN.Univariate analysis showed that there were no significant differences between SCA and MCN in gender,tumor site,tumor morphology,clinical symptoms,and enhancement intensity and mode of CEUS in the arterial phase(P>0.05).There were statistically significant differences between the two groups in age of onset,the largest diameter of the tumor,the presence of more than six cysts in the lesion or not,the presence of calcification or not and the distribution of calcification in the lesion,the largest diameter of a single cyst more than or equal to 20 mm or not,the uniformity of the thickness of the cyst wall,the presence of cyst wall nodules or not,and the intensity of enhancement in the middle venous phase of CEUS(P<0.05).Multivariate analysis showed that the number of cysts≥6,non-uniform cyst wall,and the diameter of single cyst≥20 mm were independent risk factors for the diagnosis of MCN,with odds ratio values of 0.163(95%confidence interval[CI]:0.047-0.561),9.388(95%CI:2.663-33.094),and 3.254(95%CI:1.037-10.213),respectively.The Logistic regression model was as follows:Logit(P)=-2.792-1.815X3(number of cystic cavities)+2.239X5(characteristics of cystic walls)+1.180X6(diameter of single cystic cavities).ROC curve analysis showed that the area under the curve of this regression model for identifying SCA and MCN was 0.823(95%CI:0.734-0.913).Conclusion Pancreatic SCA and MCN have appreciated imaging characteristics,and ultrasound examination allows for a better differential diagnosis of the two,thus providing a basis for clinical treatment.