Objective To investigate the clinical characteristics of patients with chronic abacterial pros-tafitis(CAP),the CAP related factors were analyzed,and a risk prediction model for CAP were constructed and validated.Methods The clinical dataes of 252 suspected CAP patients admitted to the hospital from June 2022 to December 2024 were collected,the patients were divided into modeling set(ni=177)and validation set(n=75)by 7∶3 ratio.Based on the modeling set dataes,the Lasso was used to screen CAP related predictive factors,a logistic multiple factor model was used to analyze the independent influence factors of CAP and a risk prediction model was constructed.The validation set patient dataes were used to plot ROC and DCA and validate the predic-tion model.Results There were 86 cases of CAP in the modeling set,accounting for 48.59％;32 cases of CAP in the validation set,accounting for 42.67％.The Logistic multiple regression analysis showed that BMI,waist to hip ratio,abnormal elevation of IL-8,COX-2,and PGE2 in prostate fluid were independent influence factors of CAP(P＜0.05),a Nomogram column chart based on this was established.The ROC analysis showed that the sensitivity of the model for detecting CAP in the modeling and validation sets were 0.814 and 0.802,respectively,and the specificity were 0.673 and 0.703,respectively.The DCA analysis showed that the net benefit thresholds for model-ing and validation sets by column charts are 0.1～0.9 and 0.2～1.0,respectively.Conclusions The occurrence of CAP is related to the patient's BMI,waist to hip ratio,the levels of IL-8,COX-2,and PGE2 in prostate fluid.The predictive model established based on this is highly accurate and it can help for CAP screening.

Objective To investigate the clinical characteristics of patients with chronic abacterial pros-tafitis(CAP),the CAP related factors were analyzed,and a risk prediction model for CAP were constructed and validated.Methods The clinical dataes of 252 suspected CAP patients admitted to the hospital from June 2022 to December 2024 were collected,the patients were divided into modeling set(ni=177)and validation set(n=75)by 7∶3 ratio.Based on the modeling set dataes,the Lasso was used to screen CAP related predictive factors,a logistic multiple factor model was used to analyze the independent influence factors of CAP and a risk prediction model was constructed.The validation set patient dataes were used to plot ROC and DCA and validate the predic-tion model.Results There were 86 cases of CAP in the modeling set,accounting for 48.59％;32 cases of CAP in the validation set,accounting for 42.67％.The Logistic multiple regression analysis showed that BMI,waist to hip ratio,abnormal elevation of IL-8,COX-2,and PGE2 in prostate fluid were independent influence factors of CAP(P＜0.05),a Nomogram column chart based on this was established.The ROC analysis showed that the sensitivity of the model for detecting CAP in the modeling and validation sets were 0.814 and 0.802,respectively,and the specificity were 0.673 and 0.703,respectively.The DCA analysis showed that the net benefit thresholds for model-ing and validation sets by column charts are 0.1～0.9 and 0.2～1.0,respectively.Conclusions The occurrence of CAP is related to the patient's BMI,waist to hip ratio,the levels of IL-8,COX-2,and PGE2 in prostate fluid.The predictive model established based on this is highly accurate and it can help for CAP screening.