摘要
目的探讨前列腺特异性抗原(PSA)、胱抑素C(Cys C)和中性粒细胞淋巴细胞比值(NLR)单个指标对前列腺癌的诊断效能,并通过Logistic回归模型结合受试者工作特征曲线(ROC)评估多指标联合对前列腺癌的诊断价值。方法收集2017年5月至2020年5月徐州医科大学附属医院151例经病理金标准首次确诊为前列腺癌患者的临床资料,并选取同期良性前列腺增生患者150例和健康体检者150例作为对照组。ROC曲线分析确定3项指标的最佳诊断界值及灵敏度和特异度,并通过Logistic回归模型建立预测模型,获取新的预测因子,进而结合ROC曲线获取联合诊断界值及曲线下面积(AUC)。结果实验组和对照组年龄、体重指数(BMI)差异无统计学意义(P>0.05),实验组PSA和Cys C表达水平及NLR均高于对照组(P<0.01)。通过Pearson相关分析证实三者有相关(PSA比Cys C:r=0.345,PSA比NLR:r=0.411,Cys C比NLR:r=0.303,P值均<0.01)。ROC曲线结果显示,三者的最佳诊断界值分别为PSA 4.84μg/L(AUC为0.836,灵敏度为80%,特异度为77%),Cys C 0.99 mg/L(AUC为0.708,灵敏度为57%,特异度为87%)和NLR 2.33(AUC为0.792,灵敏度为58%,特异度为87%)。利用三者建立Logistic模型为Logit(P)=-5.259+0.068×(PSA)+1.837×(Cys C)+0.863×(NLR),联合诊断的界值为0.24(AUC为0.913,灵敏度为83%,特异度为82%),与单项指标比较,AUC差异有统计学意义(P值均<0.05)。结论PSA、Cys C及NLR与前列腺癌的发生有关,并且可以协助前列腺癌的诊断。3项指标联合检测的诊断效能优于单项指标单独检测,有效降低了前列腺癌的误诊率和漏诊率。
Objective To explore the diagnostic efficacy of single indexe of prostate specific antigen(PSA),cystatin C(Cys C)and neutrophil lymphocyte ratio(NLR)on prostate cancer,and evaluate the diagnostic value of multi-indicator combination for prostate cancer by logistic regression model combined with receiver operating characteristic curve(ROC).Methods The clinical data of 151 patients diagnosed as prostate cancer for the first time by the gold standard of pathology in the Affiliated Hospital of Xuzhou Medical University from May 2017 to May 2020 were collected,and 150 patients with benign prostatic hyperplasia and 150 healthy people were selected as control group.ROC curve analysis was done to determine the best diagnostic Cut-off value,sensitivity and specificity of the three indicators,and a prediction model was established through logistic regression model to obtain new predictive factors,and then the prediction model was combined with the ROC curve to obtain the joint diagnostic Cut-off value and the area under the curve(AUC).Results There was no statistically significant difference in body mass index(BMI)between the experimental group and the control group(P>0.05).The expression levels of PSA and Cys C and NLR in the experimental group were significantly higher than those in the control group(P<0.01).Pearson correlation analysis confirmed that the three indicators were related(PSA vs.Cys C:r=0.345;PSA vs.NLR:r=0.411;Cys C vs.NLR:r=0.303;P<0.01).ROC curve results showed that the best diagnostic Cut-offs of the three indicators as follows:for PSA,4.84μg/L(AUC:0.836,sensitivity:80%,specificity:77%);for Cys C,0.99 mg/L(AUC:0.708,sensitivity:57%,specificity:87%)and for NLR 2.33(AUC:0.792,sensitivity:58%,specificity:87%).The logistic model established by the three indicators was logit(P)=-5.259+0.068×(PSA)+1.837×(Cys C)+0.863×(NLR).The cut-off value of the combined diagnosis was 0.24(AUC:0.913,sensitivity:83%,Specificity:82%).There was significant difference in the AUC between the single index and the combined index(all P<0.05).Conclusion PSA,Cys C and NLR are related to the occurrence of prostate cancer and can assist in the diagnosis of prostate cancer.The diagnostic efficacy of the combined detection of three indicators is better than that of a single indicator,which effectively reduces the misdiagnosis rate and missed diagnosis rate of prostate cancer.
作者
赵硕
李望
Zhao Shuo;Li Wang(Xuzhou Medical University,Xuzhou 221000,China;Department of Urology,the Affiliated Hospital of Xuzhou Medical University,Xuzhou 221000,China)
出处
《中华实验外科杂志》
CAS
CSCD
北大核心
2020年第8期1437-1441,共5页
Chinese Journal of Experimental Surgery
