CT定量特征在鉴别胰腺实性假乳头状瘤与乏血供神经内分泌肿瘤中的价值

CT Quantitative Characteristics in Differentiation of Pancreatic Solid Pseudopapillary Neoplasm and Hypovascular Pancreatic Neuroendocrine Neoplasm

目的探讨多层螺旋CT(MSCT)定量特征对胰腺实性假乳头状瘤(pSPN)与乏血供胰腺神经内分泌肿瘤(hypo-pNEN)的鉴别诊断价值。方法回顾性分析郑州大学第一附属医院2017年1月至2022年10月经手术病理证实的96例pSPN和52例hypo-pNEN患者的临床信息、病理结果及影像学检查资料,采用卡方检验比较两组患者定性参数(性别、部位、形态等),采用两独立样本t检验比较两组患者定量参数(年龄、长径、动脉比等)。应用二元Logistic回归模型筛选出pSPN和hypo-pNEN的独立预测因素,并采用受试者工作特征曲线(ROC)分析MSCT特征鉴别pSPN和hypo-pNEN的诊断效能。结果pSPN患者与hypo-pNEN患者相比,多为女性(80.2%vs.59.6%),年龄较小[(27.7±12.7)岁vs.(43.8±13.9)岁],发病中位时间短[(80.1±183.4)天vs.(407.6±869.7)天]、无临床症状(55.2%vs.9.6%),肿瘤多呈类圆形或椭圆形(84.4%vs.61.5%)、边界清楚(83.3%vs.63.5%),有钙化、有“浮云征”、密度不均匀、无转移或侵犯(51.0%vs.13.5%、54.2%vs.17.3%、80.2%vs.44.2%、94.8%vs.78.8%),肿瘤长径较大[(6.8±3.5)cm vs.(4.9±2.9)cm],动脉比、静脉比、动脉强化率、静脉强化率较大[(50.6±15.3)%vs.(65.2±21.6)%,(68.3±24.7)%vs.(81.9±39.2)%,(66.4±51.4)%vs.(83.2±38.2)%,(69.4±37.9)%vs.(84.7±45.8)%],差异均有统计学意义(P均<0.05)。二元Logistic回归模型分析结果显示,诊断pSPN和hypo-pNEN的独立预测因素包括动脉强化率、静脉比、“浮云征”、钙化。当联合动脉强化率、静脉比、“浮云征”、钙化等指标进行ROC曲线分析时,其曲线下面积(AUC)为0.872,诊断敏感度、特异度分别为100.0%、80.8%。各独立预测因素和联合诊断模型的ROC分析结果显示,动脉强化率、静脉比、“浮云征”、钙化和联合诊断模型的AUC分别为0.633、0.623、0.684、0.688、0.872,联合诊断模型具有较高的诊断效能(P<0.001)。结论动脉强化率、静脉比、“浮云征”、钙化等MSCT特征可用于鉴别pSPN和hypo-pNEN。

Objective To investigate the differential diagnostic value of MSCT quantitative characteristics in the differential diagnosis of pancreatic solid pseudopapillary neoplasm(pSPN)and hypovascular pancreatic neuroendocrineneoplasm(hypo-pNEN).Methods The clinical information,pathological results and imaging data of 96 patients with pSPN and 52 hypo-pNEN patients confirmed by surgical pathology from January 2017 to October 2022 in the First Affiliated Hospital of Zhengzhou University were retrospectively analyzed,and the qualitative parameters(sex,location,morphology,etc.)of the two groups were compared by chi-square test,and the quantitative parameters(age,length diameter,arterial ratio,etc.)of the two groups were compared by two independent sample t-tests.The binary logistic regression model was used to screen out the independent predictors of pSPN and hypo-pNEN,and the MSCT feature was analyzed by using the receiver operating characteristic curve(ROC)to identify the diagnostic performance of pSPN and hypo-pNEN.Results Compared with hypo-pNEN patients,pSPN patients were mostly female(80.2%vs 59.6%),younger[(27.7±12.7)years old vs.(43.8±13.9)years old],short incidence time[(80.1±183.4)days vs.(407.6±869.7)days],no clinical symptoms(55.2%vs.9.6%),and tumors were more circular or oval(84.4%vs.61.5%),clear boundaries(83.3%vs.63.5%),more with calcification,more with floating cloud sign,density inhomogeneous,more without transfer or invasion(51.0%vs.13.5%,54.2%vs.17.3%,80.2%vs.44.2%,94.8%vs.78.8%),tumor length diameter was larger[(6.8±3.5)cm vs.(4.9±2.9)cm],arterial ratio,venous ratio,arterial enhancement rate and venous enhancement rate were larger[(50.6±15.3)%vs.(65.2±21.6)%,(68.3±24.7)%vs.(81.9±39.2)%,(66.4±51.4)%vs.(83.2±38.2)%,(69.4±37.9)%vs.(84.7±45.8)%],all of which were statistically significant(all P<0.05).The results of binary Logistic regression model analysis showed that the independent predictors for the diagnosis of pSPN and hypo-pNEN included arterial enhancement rate,venous ratio,floating cloud sign,calcification.When the ROC curve was analyzed by combining arterial enhancement rate,venous ratio,floating cloud sign,calcification and other indicators,the area under the curve of the ROC chart was 0.872,and the diagnostic sensitivity and specificity were 100.0%and 80.8%,respectively.The ROC analysis results of each independent predictor and the combined diagnostic model showed that the area under the curve of the arterial strengthening rate,venous ratio,floating cloud sign,calcification and combined diagnostic model were 0.633,0.623,0.684,0.688,0.872,respectively,and the diagnostic efficacy of the combined diagnostic model was higher(P<0.001).Conclusion MSCT features such as arterial enhancement rate,venous ratio,floating cloud sign,and calcification can be used to distinguish pSPN and hypo-pNEN.