MRI征象联合炎症指标对肿块型早期浸润性乳腺癌淋巴血管侵犯预测价值

Value of MRI signs combined with inflammatory indexes in predicting lymphovascular invasion in lumpy early invasive breast cancer

目的:探讨术前磁共振成像(MRI)征象联合炎症指标对肿块型早期浸润性乳腺癌淋巴血管侵犯(LVI)的预测价值。方法:收集2018~2020年广西医科大学附属肿瘤医院收治的89例术前行磁共振增强检查的肿块型早期浸润性乳腺癌女性患者。根据术后病理检查结果将患者分为LVI(+)组与LVI(-)组,单因素分析评估两组患者术前MRI征象、炎症指标及病理学特征与LVI的关系,采用多因素logistic回归筛选危险因素并建立MRI征象模型和MRI征象联合炎症指标模型。采用受试者工作特征(ROC)曲线下面积(AUC)评价模型的预测效能。AUC的差异比较采用Z检验。结果:LVI(+)组和LVI(-)组间术前外周血中性粒细胞与淋巴细胞比值(NLR)、T分期、毛刺征、时间信号曲线(TIC)类型、瘤周水肿及表观扩散系数(ADC)值差异具有统计学意义(均P<0.05)。多因素logistic回归分析显示Ⅲ型TIC、毛刺征及瘤周水肿是LVI的独立危险因素(OR值分别为18.765、7.965、4.018,均P<0.05)。MRI征象模型(TIC类型+瘤周水肿+毛刺征)和MRI征象联合炎症指标模型(TIC类型+瘤周水肿+毛刺征+NLR)的AUC分别为0.866、0.879。Z检验显示两个模型之间AUC比较,差异无统计学意义(Z=-0.248,P=0.804)。结论:MRI征象联合炎症指标模型(TIC类型+瘤周水肿+毛刺征+NLR)有助于预测肿块型早期浸润性乳腺癌患者的LVI情况。

Objective:To explore the value of preoperative magnetic resonance imaging(MRI)signs combined with inflammatory indexes in predicting lymphovascular invasion(LVI)in lumpy early invasive breast cancer.Methods:A total of 89 female patients with lumpy early invasive breast cancer who underwent preoperative magnetic resonance enhancement examination at the Guangxi Medical University Cancer Hospital from 2018 to 2020 were collected.The patients were divided into LVI(+)and LVI(-)groups according to the postoperative pathological findings.The relationship between preoperative MRI signs,inflammatory indexes and pathological features with LVI in the two groups was evaluated by univariate analysis.Multivariate logistic regression was used to screen for risk factors,and the model of MRI signs and the model of MRI signs combined with inflammatory indexes were established.The diagnostic efficacy of the model was assessed using the area under the curve(AUC)of the receiver operating characteristic(ROC).The z-test was used to compare the differences in AUC.Results:The differences in preoperative peripheral blood neutrophil to lymphocyte ratio(NLR),T-stage,spicule sign,type of time intensity curve(TIC),peritumoural oedema and apparent diffusion coefficient(ADC)values were statistically significant between the LVI(+)and LVI(-)groups(all P<0.05).Multivariate logistic regression analysis showed that typeⅢTIC,spicule sign and peritumoural oedema were independent risk factors for LVI(OR=18.765,7.965 and 4.018,respectively;all P<0.05).The AUC of the MRI signs model(TIC type+peritumoural oedema+spicule sign)and the MRI signs combined with inflammatory indexes model(TIC type+peritumoural oedema+spicule sign+NLR)were 0.866 and 0.879,respectively.The Z-test showed that there was no statistically significant difference in the AUC between the two models(Z=-0.248,P=0.804).Conclusion:MRI signs combined with inflammatory indexes model(TIC type+peritumoural oedema+spicule sign+NLR)can help predict LVI in patients with lumpy early invasive breast cancer.