CAJ | 학술논문

目的：应用受试者工作特征曲线（ROC）确定血清人附睾蛋白4（HE-4）对卵巢癌的最佳诊断界点，为卵巢癌的早期诊断提供依据。方法采用电化学发光免疫分析法（ECLIA）检测68例卵巢癌患者、42例卵巢良性肿瘤患者和30例健康妇女血清中 HE-4的水平，绘制 ROC 曲线，运用统计软件分析。结果血清 HE-4水平在卵巢癌组、卵巢良性肿瘤组和健康对照组中均呈非正态分布。卵巢良性肿瘤组和健康组 HE-4水平无显著差异，而卵巢癌组血清 HE-4水平与卵巢良性肿瘤组、健康对照组两者差异均有统计学意义（P ＜0.01）。血清 HE-4值为108 pmol／L 时，正确诊断指数（Youden’s in-dex）最大（0.713），敏感度和特异度分别为77.9％和93.1％，阳性预测值为91.4％，阴性预测值为81.7％，阳性似然比为11.6，阴性似然比为0.2，优势比（odds ratio）为47.3。结论 HE-4检测是卵巢癌诊断和排除诊断的较理想指标，选择108pmol／L 作为卵巢癌诊断界点比较合适。
목적：응용수시자공작특정곡선（ROC）학정혈청인부고단백4（HE-4）대란소암적최가진단계점，위란소암적조기진단제공의거。방법채용전화학발광면역분석법（ECLIA）검측68례란소암환자、42례란소량성종류환자화30례건강부녀혈청중 HE-4적수평，회제 ROC 곡선，운용통계연건분석。결과혈청 HE-4수평재란소암조、란소량성종류조화건강대조조중균정비정태분포。란소량성종류조화건강조 HE-4수평무현저차이，이란소암조혈청 HE-4수평여란소량성종류조、건강대조조량자차이균유통계학의의（P ＜0.01）。혈청 HE-4치위108 pmol／L 시，정학진단지수（Youden’s in-dex）최대（0.713），민감도화특이도분별위77.9％화93.1％，양성예측치위91.4％，음성예측치위81.7％，양성사연비위11.6，음성사연비위0.2，우세비（odds ratio）위47.3。결론 HE-4검측시란소암진단화배제진단적교이상지표，선택108pmol／L 작위란소암진단계점비교합괄。
Objective Using ROC curve to determine the best cut-off point of serum HE-4 in the diagnosis of ovarian cancer and provide important value to diagnosis early ovarian cancer.Methods The levels of serum HE-4 in 68 ovarian cancer pa-tients,42 ovarian benign tumor patients and 30 healthy female were detected by electrochemistry irradiance method.The ROC curve was drawn and the cut-off point of HE-4 was determined by statistical software.Results The levels of serum HE-4 were all non-normal distribution in the groups of ovarian cancer,ovarian benign tumor and healthy controls.Whats more,there was no significant difference between ovarian benign tumor group and normal control group.And compared with benign ovarian tumors and normal control group,the level of HE-4 in ovarian cancers was significantly increased (P <0.01).It would be best for diagnosis when the level of serum HE-4 was 108pmol/L in ovarian cancer.Youden’s index showed the maximum (0.713)and the sensitivity and specificity of diagnosing were 77.9% and 93.1% respectively.The positive predictive value was 91.4% and negative predictive value was 91.4%.At the same time,the positive likelihood ratio was 11.6 and the negative likelihood ratio was 0.2,odds ratio reached to 47.3.Conclusion The detection of HE-4 is an ideal mark for diagnosing and excluding ovarian cancer.Selecting 108 pmol/L as ovarian cancer diagnosis point is relative appro-priate.