Preliminary Establishment And Clinical Application Of Prostate Biopsy Positive Prediction Model In Shihezi Area

Objective:Collection of shihezi university school of medicine in the first affiliated hospital of prostate biopsy in patients with data,joint literature reports,preliminary build shihezi prostate biopsy positive prediction model.Method:1.Retrospective study clinical data of patients who had undergone transrectal ultrasonography for prostate biopsy from 2012 to 2017 in our hospital,patients with prostate cancer and non-prostate cancer who received puncture results were compared,looking for risk factors for prostate cancer.2.Combine the above data,the prediction model of positive prostate cancer biopsy was preliminarily established.And retrospectively validate and evaluate its clinical application value.3.The patients who plans to undergo a prostate biopsy were perspectively studied with using the established prediction model to predict the risk of prostate biopsy positivity,analyzed the difference between predictive value and observed value.Result:1.A total of 179 patients in this study,a total of 98 cases were diagnosed with prostate ca ncer(54.75%),total mean age(74.92±5.54),total prostate-specific antigen(54.01±37.57)ng/ml.Univ-ariate analysis showed that age,t PSA,PSAD,Serum creatinine(Scr),digital rectal examination(DRE),Magnetic Resonance Imaging(MRI)were related to the occurrence of positive prostate biopsy(P<0.05).Multivariate analysis showed that age(OR=2.909,95%CI:1.445-5.854),t PSA(OR=9.549,95%CI:3.540-25.761),DRE(OR=6.166,95%CI:1.923-19.771),and MRI(OR=14.143,95%CI:5.127-39.013)were associ-a ted with positive prostate biopsy(P<0.05).2.Incorporate age,t PSA,DRE,MRI into that model,Preliminary establishment of positive prediction model of prostate biopsy in shihezi area.Retrospective analysis showed that the area under the ROC curve of the predictive model was 0.927,Its sensitivity and specificity were 84.7% and 88.9% respectively at the best cut-off point of 0.60.The prediction model was divided into high-risk group(>0.60)and low-risk group(<0.60),according to the cut-off point.3.According to the prediction model,80 patients were predicted in preoperative,the results showed that there were 40 patients in the high-risk group and 40 patients in the low-risk group.Puncture results showed 45 positive cases and 35 negative cases.There was no significant difference between the predicted group and the actual group(P > 0.05).The area under the ROC curve is 0.894,95%CI(0.817~0.971).Conclusion:1.Age,t PSA,DRE,MRI are risk factors for prostate cancer when prostate biopsy is positive.2.The predictive model can effectively evaluate the positive risk of prostate biopsy patients before surgery.Clinicians should pay enough attention to high-risk patients above 0.60.3.The prediction model can be used as a risk prediction tool for prostate biopsy positive in shihezi area.