摘要
目的:探讨表观弥散系数(ADC)值对儿童颅后窝常见肿瘤鉴别诊断的价值。方法:回顾性分析60例经手术病理证实的儿童颅后窝常见肿瘤,其中髓母细胞瘤25例,毛细胞星形细胞瘤23例,室管膜瘤12例。所有患儿术前行MRI平扫、增强和DWI检查。测量三组肿瘤的平均ADC值和最小ADC(ADCmin)值,绘制受试者工作特征(ROC)曲线,获取肿瘤鉴别诊断的最佳阈值及灵敏度、特异度和诊断准确率。结果:髓母细胞瘤、毛细胞星形细胞瘤和室管膜瘤的平均ADC值分别为(0.69±0.34)×10^(-3)mm^2/s、(1.68±0.23)×10^(-3)mm^2/s和(1.15±0.27)×10^(-3)mm^2/s;ADCmin值分别为(0.48±0.29)×10^(-3)mm^2/s、(1.59±0.32)×10^(-3)mm^2/s和(0.83±0.39)×10^(-3)mm^2/s,三组肿瘤的平均ADC值和ADCmin值差异具有统计学意义(P<0.05)。ROC曲线分析鉴别髓母细胞瘤与室管膜瘤的平均ADC值最佳阈值为0.87×10^(-3)mm^2/s,相应敏感度92.7%、特异度96.3%、诊断准确率94.5%;鉴别室管膜瘤与毛细胞星形细胞瘤的平均ADC值最佳阈值为1.36×10^(-3)mm^2/s,相应敏感度91.7%、特异度92.6%、诊断准确率92.7%。ADCmin值鉴别髓母细胞瘤与室管膜瘤的最佳阈值为0.67×10^(-3)mm^2/s,相应敏感度97%、特异度100%、诊断准确率97.8%。ADCmin值鉴别室管膜瘤与毛细胞星形细胞瘤的最佳阈值为1.06×10^(-3)mm^2/s,相应敏感度95.7%、特异度97.9%、诊断准确率94.7%结论:平均ADC值和ADCmin值在儿童颅后窝常见肿瘤的鉴别诊断中具有较高的临床应用价值,且ADCmin值对肿瘤的鉴别诊断能力优于平均ADC值。
Purpose:To investigate the value of apparent diffusion coefficient (ADC) in dltterenual diagnosis of common pediatric posterior fossa tumors. Methods: Sixty children with posterior fossa tumors confirmed by surgery and pathology results were retrospectively analyzed, including 25 medulloblastomas, 23 pilocytic astrocytomas, and 12 ependymomas. All the patients underwent preoperative MRI plain scan, enhanced scan and DWI. The mean and minimum ADC (ADCmin) values of tumors were measured and the receiver operating characteristic (ROC) curves were delineated. The optimal mean ADC and ADCmin values for differential diagnosis of three types tumors, and sensitivity, specificity and diagnostic accuracy rate of them were determined. Results: The mean ADC and ADCmin values for medulloblastoma, pilocytic astrocytoma and ependymoma were (0.69±0.34)×10^-3mm^2/s, (1.68±0.23)×10^-3mm^2/s, and (1.15±0.27)×10^-3mm^2/s; (0.48±0.29)×10^-3mm^2/s, (1.59±0.32)×10^-3mm^2/s and (0.83±0.39)- 10^-3mm^2/s, respectively. There were statistically significant differences of mean ADC and ADCmin values among these three kinds of tumors (P〈0.05). The optimal mean ADC value which could be used to distinguish medulloblastoma from ependymoma was 0.87×10^-3mm^2/s (sensitivity: 92.7%, specificity: 96.3% and diagnostic accuracy rate: 94.5%); The optimal mean ADC value to distinguish ependymoma from pilocytic astrocytoma was 1.36×10^-3mm^2/s (sensitivity: 91.7%, specificity: 92.6% and diagnostic accuracy rate: 92.7%); The optimal ADCmin value to distinguish medulloblastoma from ependymoma was 0.67 ×10^-3mm^2/s (sensitivity: 97%, specificity: 100% and diagnostic accuracy rate: 97.8%); The optimal ADCmin value to distinguish ependymoma from pilocytic astrocytoma was 1.06×10^-3mm^2/s (sensitivity: 95.7%, specificity: 97.9% and diagnostic accuracy rate: 94.7%). Conclusion: The mean and minimum ADC values are useful in differential diagnosis of common pediatric posterior fossa tumors, and ADCmin value is better than mean ADC value for the differential diagnosis.
出处
《中国医学计算机成像杂志》
CSCD
北大核心
2017年第3期201-206,共6页
Chinese Computed Medical Imaging
基金
河南省医学科技攻关项目(No.201602030)
河南省科技攻关项目(162102310104)~~