Objective To explore the method of constructing an early mortality risk prediction model for patients with sepsis-induced cardiorenal syndrome by machine learning algorithm, so as to provide a basis for early clinical identification of high-risk patients and accurate treatment. Methods Patients with sepsis-induced cardiorenal syndrome from January 1, 2015 to May 31, 2019 in Tongji Hospital, Tongji University were enrolled. Basic characteristics, laboratory indexes, hospitality treatment and other relevant baseline data were collected. Thirty-day mortality was defined as the primary end-point event after the enrolled patients were diagnosed. Python software was applied to establish different machine learning models, and the area under the receiver -operating characteristic curve (AUC) was used to evaluate the predictive value of models. Disease-related risk factors were selected according to the most optimal model. Importantly, visualized decision tree and semi-naive Bayesian (sNB) models were established to further explore the interrelationship between these risk factors. Results A total of 340 patients were included, of whom 114 patients (33.5%) died within 30 days after diagnosis. The AUC of support vector machine (SVM), random forest (RF), gradient boosting decision tree (GBDT), extreme gradient boosting (XGBoost), and light gradient boosting machine (LGBM) prediction models were 0.652, 0.868, 0.870, 0.754, and 0.852, respectively. The AUC of GBDT model had the most efficiency to predict end-point events, and the prediction AUC value was better. According to the feature ranking of GBDT model, the relevant influencing factors were selected, including total sequential organ failure assessment (SOFA) score, neural SOFA score, vasoactive drug application, cardiac troponin I (cTNI), age, myoglobin, circulation system SOFA score, chronic kidney disease, heart rate and baseline serum creatinine. Visualized decision tree model had 4 layers, 15 nodes and 8 terminal nodes as evidenced by total SOFA score, myoglobin, baseline serum creatinine and age. The total SOFA score, change rate of myoglobin, serum creatinine and age were included into the visualized decision model. The AUC value of the model for predicting end-point event was 0.690. sNB model revealed complex correlation between the risk factors, in which neural SOFA score was related to total SOFA score, vasoactive drug application was related to total SOFA score, and cTNI was related to baseline serum creatinine. Conclusions A risk prediction model for patients with sepsis-induced cardiorenal syndrome is established and the model showes that high SOFA score remains the primary risk factor for patients with sepsis-induced cardiorenal syndrome based machine learning. Visualized decision tree and sNB models help clinicians to further identify the dependence and logic relationship among these risk factors clearly and provide a novel method to predict mortality risk for patients with sepsis-induced cardiorenal syndrome.

Objective To explore the incidence, influencing factors and prognostic value of cardiac valve calcification (CVC) in chronic kidney disease (CKD) non-dialysis patients. Methods The non-dialysis patients with CKD stage 1-5 who were hospitalized and underwent echocardiography in the Department of Nephrology, Xuanwu Hospital, Capital Medical University from January 1, 2018 to December 31, 2019 were retrospectively admitted. The patients were divided into CVC group and non-CVC group, and the clinical data were compared between the two groups. The deadline for follow-up was November 1, 2021, and the follow-up end point event was all-cause mortality. Logistic regression model was used to analyze the risk factors of CVC in patients with CKD, and Cox proportional hazards regression model was used to analyze the risk factors of all-cause mortality in patients with CKD. Results A total of 563 patients with CKD were enrolled in the study, with age of (59.49±13.97) years old, and 352 males (62.52%). There were 325 patients (57.73%) with CKD stage 1-3 and 238 patients (42.27%) with CKD stage 4-5. The incidence of CVC in CKD stage 1-5 patients was 32.32%(182/563). Aortic valve calcification occurred in 30.73%(173/563), mitral valve calcification occurred in 9.77% (55/563), double valve (mitral and aortic valve) calcification occurred in 8.35% (47/563), and tricuspid valve calcification occurred in 0.18%(1/563). Age (t=12.223, P<0.001) and the proportions of CKD stage 4-5 ( χ²=10.854, P=0.001), hypertension ( χ²=7.811, P=0.005), diabetes ( χ²=8.424, P=0.004), hyperlipidemia ( χ²=9.331, P=0.002), and taking statins ( χ²=4.868, P=0.027) in CVC group were significantly higher than those in non-CVC group. Total cholesterol (t=2.243, P=0.025), low density lipoprotein cholesterol (t=2.025, P=0.043), platelet count (t=2.230, P=0.026) and estimated glomerular filtration rate (t=8.630, P<0.001) in CVC group were lower than those in the non-CVC group. Logistic regression analysis results showed that age≥60 years old (≥60 years old/<60 years old, OR=7.412, 95%CI 4.514-12.170, P<0.001), CKD stage 4-5 (stage 4-5/stage 1-3, OR=2.791，95%CI 1.730-4.505，P<0.001) and hyperlipidemia (OR=5.241, 95%CI 3.283-8.367, P<0.001) were the independent influencing factors of CVC in patients with CKD. Five hundred and sixty-three patients were followed up for an average of 26 months, including 68 cases (12.08%) of death, 436 cases (77.44%) of survival and 59 cases (10.48%) of loss to follow-up. Multivariate Cox regression analysis results showed that age≥60 years old (≥60 years old/<60 years old, HR=2.157, 95%CI 1.127-4.127, P=0.020), serum albumin<30 g/L (<30 g/L/≥30 g/L, HR=1.923, 95%CI 1.037-3.568, P=0.038) and double valve calcification (double valve calcification/no valve calcification, HR=2.516, 95%CI 1.279-4.950, P=0.008) were the independent influencing factors of all-cause death in patients with CKD. Conclusions CVC accounts for 32.32% in non-dialysis patients with CKD stage 1-5. Older age, worse renal function and hyperlipidemia are the independent risk factors of CVC in CKD patients. Older age, hypoproteinemia and double valve calcification are the independent risk factors of all-cause death in patients with CKD.

Objective To investigate the incidence, risk factors, and outcomes of acute kidney injury (AKI) in cancer patients receiving immune checkpoint inhibitors (ICIs). Methods A retrospective analysis was performed on the inpatients who received ICIs therapy in Beijing Shijitan Hospital, Capital Medical University from October 2015 to December 2020. According to the Kidney Disease: Improving Global Outcomes (KDIGO) definition of AKI, patients were divided into AKI group and non-AKI group, and the patients in the AKI group were further divided into ICIs related AKI (ICIs-AKI) and AKI due to other etiologies. The clinical characteristics of the patients were compared. Multivariate logistic regression was used to analyze the influencing factors of AKI, and sensitivity analysis was used to evaluate the influencing factors of ICIs-AKI. Results A total of 279 cancer patients over 18 years old were included in this study, in which 175(62.7%) were males. There were 41 patients (14.70%) in AKI group, including 25 patients (8.96%) in ICIs-AKI group and 16 patients (5.73%) in AKI due to other etiologies group. Patients in the AKI group were characterized by higher proportions of hypertension, diuretics use and baseline eGFR<60 ml·min^-1·(1.73 m²)^-1, extrarenal immune-related adverse events (irAEs) and a lower plasma albumin level (all P<0.05). The patients in the ICIs-AKI group had higher proportions of new aseptic leukocyturia, blood eosinophil count>500/ml, combined extrarenal irAEs, glucocorticoid use and discontinued ICIs treatment (all P<0.05). Multivariate logistic regression results showed that hypertension (OR=3.424, 95%CI 1.559-7.522, P=0.002), use of diuretics (OR=4.620, 95%CI 2.111-10.112, P<0.001), baseline eGFR<60 ml·min^-1·(1.73 m²)^-1 (OR=3.668, 95%CI 1.336-10.070, P=0.012) and extrarenal irAEs (OR=9.909, 95%CI 4.198-23.391, P<0.001) were associated with AKI in cancer patients receiving ICIs therapy. Sensitivity analysis indicated that the risk factors of ICIs-AKI included use of diuretics and baseline eGFR<60 ml·min^-1·(1.73 m²)^-1, similar to the results of the above analysis, extrarenal irAEs (OR=17.572, 95%CI 6.302-48.995, P<0.001) were also associated with ICIs-AKI independently. Conclusions AKI is not uncommon in patients treated with ICIs. Concomitant hypertension, baseline eGFR<60 ml·min^-1·(1.73 m²)^-1 and use of diuretics are independent risk factors for AKI in such patients. Patients should be alert to the risk of ICIs-AKI when appearing extrarenal irAEs. Distinguishing ICIs-AKI from AKI caused by other causes will present a frequent challenge to clinical practitioners.

Objective To investigate the characteristics and risk factors of infection in newly diagnosed patients with antineutrophil cytoplasmic antibody (ANCA)-associated vasculitis (AAV). Methods The clinical data of AAV patients (followed up for at least 6 months) in Affiliated Hospital of Qingdao University from September 2012 to September 2020 were retrospectively collected. According to whether infection occurred during follow-up, the patients were divided into infection group and non-infection group. The clinical characteristics and infection status of the two groups were analyzed, and the Cox regression analysis model was used to explore the influencing factors of infection. Results A total of 236 AAV patients were enrolled in this study, including 128 females (54.2%) and 108 males (45.8%), with a median age of 66.00 (59.76, 71.99) years. There were 202 patients (85.6%) with positive myeloperoxidase (MPO)-ANCA and 34 patients (14.4%) with positive protease 3 (PR3) -ANCA. There were 77 cases in the infection group and 159 cases in the non-infection group. A total of 121 infections occurred in 77 patients, and 54 infections (44.6%) occurred within 6 months after initial diagnosis. In the infection group the proportion of patients with hypertension history, pulmonary underlying diseases and patients who received hormone pulse therapy or plasma exchange, the incidence of lung, kidney, heart and gastrointestinal involvement, the level of serum creatinine and five factors score (FFS) at initial diagnosis were significantly higher than those in the non-infection group (all P<0.05), while the estimated glomerular filtration rate (eGFR) was significantly lower (P<0.05). Lung (73.6%) was the main infection organ of AAV patients. The most common pathogenic microorganisms were bacteria (64.0%), mainly Pseudomonas aeruginosa and Staphylococcus aureus, followed by fungi (33.7%, mainly Candida albicans). Multivariate Cox regression analysis showed that lung involvement (HR=1.682, 95%CI 1.034-2.734, P=0.036) and gastrointestinal involvement (HR=2.976, 95%CI 1.219-7.267, P=0.017) were the independent influencing factors for infection in AAV patients. Conclusions AAV patients have a higher incidence of infection within 6 months after initial diagnosis. The most common organ of infection in AAV patients is the lung, and the common pathogens are Pseudomonas aeruginosa, Staphylococcus aureus and Candida albicans. Lung involvement and gastrointestinal involvement are the independent risk factors for infection in AAV patients.

Objective To investigate the effect of autophagy inhibitor 3-methyladenine (3-MA) on uric acid (UA)-induced apoptosis of renal tubular epithelial cells. Methods (1) Totally 24 SD rats were randomly divided into 4 groups: control group, treatment with 3-MA group, hyperuricemic nephropathy (HN) group, and HN+3-MA group, with 6 rats in each group. According to the body weight of the rats, adenine (100 mg/kg) and potassium oxonate (1 500 mg/kg) were mixed with distilled water to make a suspension, and the rats were given intragastrically once daily for consecutive 21 days to establish HN rat model. The control group and the 3-MA treatment group were fed an equivalent amount of distilled water. At the same time, the 3-MA treatment group and the HN+3-MA group were intraperitoneally injected with 3-MA (15 mg/kg), and the control group and HN group were intraperitoneally injected with an equal volume of saline once daily for 21 consecutive days. Terminal deoxynucleotidyl transferase-mediated dUTP-biotin nick end labeling assay (TUNEL) was used to observe renal cell apoptosis. Western blotting was used to detect the expression levels of cleaved caspase-3 and Bax, and immunofluorescence staining was used to detect the expression and localization of cleaved caspase-3 in renal tissue. (2) Human renal tubular epithelial cells (HK-2) were stimulated with UA (800 μmol/L), and cells were administrated with different concentrations of 3-MA or Beclin-1 small interfering RNA (siRNA). The apoptosis of renal tubular epithelial cells was detected by Western blotting and immunofluorescence staining. Results Compared with the normal rats, the apoptosis of renal tubular epithelial cells in the HN group was significantly increased (P<0.01), and the expression levels of cleaved caspase-3 and Bax were significantly up-regulated (both P<0.05). Compared with the HN group, the apoptosis of renal tubular epithelial cells in the HN+3-MA group was significantly decreased (P<0.01). In addition, high level of uric acid could significantly increase the levels of apoptosis associated proteins in HK-2 cells (all P<0.05), and using different concentrations of 3-MA or transfecting with Beclin-1 siRNA could significantly reduce the expression of cleaved caspase-3 and Bax (all P<0.05). Conclusion Autophagy plays an important role in uric acid-induced apoptosis of renal tubular epithelial cells. Inhibiting the excessive activation of autophagy may be a new strategy to prevent the progression of HN.