OBJECTIVE  To investigate the value of serum cytokine profiles combined with CT features in the early diagnosis of calculi-induced obstructive pyonephrosis, and to develop and validate a nomogram prediction model based on multimodal data.

METHODS  A retrospective analysis was conducted on the clinical data of 280 patients with urinary calculi and obstructive renal hydronephrosis admitted to the Affiliated Hospital of Hebei University from Jan. 2022 to Jun. 2025. Based on the nature of the renal pelvic drainage fluid during surgery, patients were divided into a pyonephrosis group （n=68） and a non-pyonephrosis group （n=212）, and randomly assigned to a training set （n=196） and a validation set （n=84） at a ratio of 7:3. We collected clinical characteristics, imaging data and preoperative serum cytokine levels interleukin （IL）-1β, IL-2, IL-4, IL-6, IL-10, IL-12p70, IL-17, tumor necrosis factor （TNF）-α, C-reactive protein （CRP）, interferon （IFN）-γ. LASSO regression was applied to screen variables, and multivariate logistic regression analysis was conducted to identify independent predictors and construct a nomogram model. The discrimination, calibration and clinical applicability of the model were evaluated by receiver operating characteristic （ROC） curves, calibration curves and decision curve analysis （DCA）.

RESULTS  In both the training and validation sets, the pyonephrosis group exhibited higher values in terms of the maximum stone diameter, CT value of renal pevlvic effusion, proportion of patients with comorbid diabetes mellitus and blood neutrophil count compared to the non-pyonephrosis group, while the CT value of stones was lower in the pyonephrosis group （P＜0.05）. Additionally, CRP, PCT, IL-6, IL-10 and TNF-α levels were all elevated in the pyonephrosis group （P＜0.05）. Multivariate logistic regression analysis revealed that comorbid diabetes mellitus （OR=4.251）, CT value of renal pevlvic effusion （OR=1.360）, CRP （OR=1.050）, IL-6 （OR=1.029） and TNF-α （OR=1.109） were risk factors for calculi-induced obstructive pyonephrosis （P＜0.05）. The constructed nomogram model demonstrated AUC values of 0.953 （95%CI: 0.922－0.983） in the training set and 0.895 （95%CI: 0.796－0.995） in the validation set, outperforming individual indicators. The calibration curve indicated good agreement between predicted probabilities and actual incidence rates （MAE=0.021）, while the DCA curve suggested significant clinical net benefit across a wide range of threshold values.

CONCLUSION  Comorbid diabetes mellitus, elevated CT values of hydronephrosis, and abnormal levels of CRP, IL-6 and TNF-α are predictive factors for calculi-induced obstructive pyonephrosis.