Background: Early identification of patients at high-risk of postoperative acute kidney injury (AKI) can facilitate the development of preventive approaches. This study aimed to develop prediction models for postoperative AKI in noncardiac surgery using machine learning algorithms. The authors also evaluated the predictive performance of models that included only preoperative variables or only important predictors. Materials and methods: Adult patients undergoing noncardiac surgery were retrospectively included in the study (76 457 patients in the discovery cohort and 11 910 patients in the validation cohort). AKI was determined using the KDIGO criteria. The prediction model was developed using 87 variables (56 preoperative variables and 31 intraoperative variables). A variety of machine learning algorithms were employed to develop the model, including logistic regression, random forest, extreme gradient boosting, and gradient boosting decision trees. The performance of different models was compared using the area under the receiver operating characteristic curve (AUROC). Shapley Additive Explanations (SHAP) analysis was employed for model interpretation. Results: The patients in the discovery cohort had a median age of 52 years (IQR: 42–61 years), and 1179 patients (1.5%) developed AKI after surgery. The gradient boosting decision trees algorithm showed the best predictive performance using all available variables, or only preoperative variables. The AUROCs were 0.849 (95% CI: 0.835–0.863) and 0.828 (95% CI: 0.813–0.843), respectively. The SHAP analysis showed that age, surgical duration, preoperative serum creatinine, and gamma-glutamyltransferase, as well as American Society of Anesthesiologists physical status III were the most important five features. When gradually reducing the features, the AUROCs decreased from 0.852 (including the top 40 features) to 0.839 (including the top 10 features). In the validation cohort, the authors observed a similar pattern regarding the models’ predictive performance. Conclusions: The machine learning models the authors developed had satisfactory predictive performance for identifying high-risk postoperative AKI patients. Furthermore, the authors found that model performance was only slightly affected when only preoperative variables or only the most important predictive features were included.

Objectives Hyperuricaemia is a risk factor for gout attacks, kidney damage and cardiovascular events. Evidence on the trends in hyperuricaemia burden in Wuhan city, China, was limited. The present study aimed to estimate the prevalence of and a decade trend in hyperuricaemia in Wuhan city. Design Cross-sectional study. Setting Health Management Center of Tongji Hospital. Participants A total of 732 527 adult participants from the general population who took a physical examination in the Health Management Center between 2010 and 2019. Main outcome measures Prevalence of and trends in hyperuricaemia. Results The overall prevalence of hyperuricaemia was 25.8% (36.6% in men and 10.8% in women) in 2019. The hyperuricaemia prevalence and serum uric acid (SUA) levels were significantly higher in young men, old women and participants with obesity, hypertension, diabetes or dyslipidaemia (p<0.05). SUA levels among men and women gradually increased from 358.0 (313.0–407.0) umol/L and 250.0 (217.0–288.0) umol/L in 2010 to 388.0 (338.0–445.2) umol/L and 270.0 (233.0–314.0) umol/L in 2019, respectively, p<0.05. From 2010 through 2019, hyperuricaemia prevalence significantly increased in each age category and it increased most sharply among participants aged 20–39 years. The multivariate-adjusted prevalence among men was 26.1% (25.4% to 26.7%) in 2010, 30.9% (30.4% to 31.4%) in 2015 and 34.4% (34.1% to 34.8%) in 2019, while among women it was 5.8% (5.4% to 6.2%) in 2010, 7.2% (6.9% to 7.5%) in 2015 and 10.1% (9.9% to 10.3%) in 2019. Conclusions Hyperuricaemia was highly prevalent among adults in Wuhan city. More attention should be paid to the increasing burden of hyperuricaemia, especially for those at higher risks.

This study aimed to elucidate the role of USP25 in a mouse model of anti-glomerular basement membrane glomerulonephritis (anti-GBM GN). USP25-deficient anti-GBM GN mice were generated, and their nephritis progression was monitored. Naïve CD4+ T cells were isolated from spleen lymphocytes and stimulated to differentiate into Th1, Th2, and Th17 cells. This approach was used to investigate the impact of USP25 on CD4+ T lymphocyte differentiation in vitro. Furthermore, changes in USP25 expression were monitored during Th17 differentiation, both in vivo and in vitro. USP25−/− mice with anti-GBM GN exhibited accelerated renal function deterioration, increased infiltration of Th1 and Th17 cells, and elevated RORγt transcription. In vitro experiments demonstrated that USP25−/− CD4+ T lymphocytes had a higher proportion for Th17 cell differentiation and exhibited higher RORγt levels upon stimulation. Wild-type mice with anti-GBM GN showed higher USP25 levels compared to healthy mice, and a positive correlation was observed between USP25 levels and Th17 cell counts. Similar trends were observed in vitro. USP25 plays a crucial role in mitigating renal histopathological and functional damage during anti-GBM GN in mice. This protective effect is primarily attributed to USP25’s ability to inhibit the differentiation of naïve CD4+ T cells into Th17 cells. The underlying mechanism may involve the downregulation of RORγt. Additionally, during increased inflammatory responses or Th17 cell differentiation, USP25 expression is activated, forming a negative feedback regulatory loop that attenuates immune activation.