Machine learning algorithms for the prediction of adverse prognosis in patients undergoing peritoneal dialysis.

BACKGROUND: An appropriate prediction model for adverse prognosis before peritoneal dialysis (PD) is lacking. Thus, we retrospectively analysed patients who underwent PD to construct a predictive model for adverse prognoses using machine learning (ML). METHODS: A retrospective analysis was conducted on 873 patients who underwent PD from August 2007 to December 2020. A total of 824 patients who met the inclusion criteria were included in the analysis. Five commonly used ML algorithms were used for the initial model training. By using the area under the curve (AUC) and accuracy (ACC), we ranked the indicators with the highest impact and displayed them using the values of Shapley additive explanation (SHAP) version 0.41.0. The top 20 indicators were selected to build a compact model that is conducive to clinical application. All model-building steps were implemented in Python 3.8.3. RESULTS: At the end of follow-up, 353 patients withdrew from PD (converted to haemodialysis or died), and 471 patients continued receiving PD. In the complete model, the categorical boosting classifier (CatBoost) model exhibited the strongest performance (AUC = 0.80, 95% confidence interval [CI] = 0.76-0.83; ACC: 0.78, 95% CI = 0.72-0.83) and was selected for subsequent analysis. We reconstructed a compression model by extracting 20 key features ranked by the SHAP values, and the CatBoost model still showed the strongest performance (AUC = 0.79, ACC = 0.74). CONCLUSIONS: The CatBoost model, which was built using the intelligent analysis technology of ML, demonstrated the best predictive performance. Therefore, our developed prediction model has potential value in patient screening before PD and hierarchical management after PD.

A novel case of acute glomerulonephritis with concurrent acute non-rheumatic myocarditis following group a streptococcal infection.

Streptococcal infection is a common cause of acute glomerulonephritis. Cardiac damage associated with streptococcal infection commonly occurs in acute rheumatic fever. However, cases of acute non-rheumatic streptococcal myocarditis have been reported in recent years. We report a novel case of concurrent acute glomerulonephritis and non-rheumatic myocarditis following streptococcal infection. A good prognosis was achieved with antibiotic and immunosuppressive therapy, indicating that Streptococcus causes cardiorenal syndrome type 5 via an immune-mediated response. A better understanding of post-streptococcal cardiorenal syndrome is warranted to enable the early diagnosis and treatment of affected patients.

Case report of extranodal natural killer/T-cell lymphoma that induced secondary hemophagocytic syndrome-related histiocytic glomerulopathy.

Hemophagocytic syndrome (HPS) is a proliferative disease of the mononuclear macrophage system involving multiple organs and systems. We report a 50-year-old Asian woman who presented with unexplained fever and proteinuria. Laboratory tests showed cytopenia, considerably elevated serum ferritin and IL-2 receptor concentrations, and evidence of hemophagocytosis in the bone marrow. A renal biopsy showed macrophage infiltration into the glomerulus, resulting in podocyte and endothelial cell damage. We finally diagnosed the patient with extranodal natural killer/T-cell lymphoma, nasal type that induced HPS-related histiocytic glomerulopathy. Proteinuria and inflammation responded to treatment with high-dose pulsed methylprednisolone combined with VP-16 and cyclosporine. To the best of our knowledge, this is the first documented case of HPS-related histiocytic glomerulopathy triggered by a malignant tumor.

Aristolochic acid inhibits Slit2-induced migration and tube formation via inactivation of Robo1/Robo2-NCK1/NCK2 signaling pathway in human umbilical vein endothelial cells.

Robo1/Robo2-NCK1/NCK2 signaling pathway controls endothelial cell sprouting and migration induced by Slit2 or VEGF, but whether it is involved in peritubular capillary (PTC) rarefaction of Aristolochic acid nephropathy (AAN) is unclear. In the present study, we evaluated whether AA exerts antiangiogenic effects by targeting this signaling pathways in HUVECs. HUVECs or lentivirus-mediated NCK1-overexpressing HUVECs were stimulated with AA (1, 2 or 3 µg/ml) in the absence or presence of 6 nM Slit2. Our results showed that AAІ (1-3 µg/ml) dose-dependently inhibited the migration and tube formation of HUVECs. This inhibition was in parallel with down-regulated mRNA and protein expression of Slit2/Robo1/Robo2-NCK1/NCK2 signaling pathway. Importantly, overexpression of NCK1 rescued AAІ-impaired angiogenesis, as evidenced by the increase of cell migration and tube formation of HUVECs in response to Slit2. The down-regulation of NCK2 and decreased activation of Rac1 was also restored by overexpression of NCK1. Taken together, our findings show that AA inhibits Slit2-induced migration and tube formation via inactivation of Robo1/Robo2-NCK1/NCK2 signaling pathway in HUVECs, and NCK1 might be a potential agent for vascular remodeling in AAN and diseases associated with impaired angiogenesis.

Slit2/Robo1 signaling is involved in angiogenesis of glomerular endothelial cells exposed to a diabetic-like environment.

Abnormal angiogenesis plays a pathological role in diabetic nephropathy (DN), contributing to glomerular hypertrophy and microalbuminuria. Slit2/Robo1 signaling participates in angiogenesis in some pathological contexts, but whether it is involved in glomerular abnormal angiogenesis of early DN is unclear. The present study evaluated the effects of Slit2/Robo1 signaling pathway on angiogenesis of human renal glomerular endothelial cells (HRGECs) exposed to a diabetic-like environment or recombinant Slit2-N. To remove the effect of Slit2 derived from mesangial cells, human renal mesangial cells (HRMCs) grown in high glucose (HG) medium (33 mM) were transfected with Slit2 siRNA and then the HG-HRMCs-CM with Slit2 depletion was collected after 48 h. HRGECs were cultured in the HG-HRMCs-CM or recombinant Slit2-N for 0, 6, 12, 24, or 48 h. The mRNA and protein expressions of Slit2/Robo1, PI3K/Akt and HIF-1α/VEGF signaling pathways were detected by quantitative real-time PCR, western blotting, and ELISA, respectively. The CCK-8 cell proliferation assay, flow cytometry and the scratch wound-healing assay were used to assess cell proliferation, cycles, and migration, respectively. Matrigel was used to perform a tubule formation assay. Our results showed that the HG-HRMCs-CM with Slit2 depletion enhanced the activation of Slit2/Robo1, PI3K/Akt, and HIF-1α/VEGF signaling in HRGECs in time-dependent manner (0-24 h post-treatment). In addition, the HG-HRMCs-CM with Slit2 depletion significantly promoted HRGECs proliferation, migration, and tube formation. Pretreatment of HRGECs with Robo1 siRNA suppressed the activation of PI3K/Akt and HIF-1α/VEGF signaling and inhibited angiogenesis, whereas PI3K inhibitor suppressed HIF-1α/VEGF signaling, without influencing Robo1 expression. In the HRGECs treated with Slit2-N, Slit2-N time-dependently enhanced the activation of Robo1/PI3K/Akt/VEGF pathway but not HIF-1α activity, and promoted HRGECs proliferation, migration, and tube formation. The effects induced by Slit2 were also abolished by Robo1 siRNA and PI3K inhibitor. Taken together, our findings indicate that in a diabetic-like environment, in addition to mesangial cells, autocrine activation of Slit2/Robo1 signaling of HRGECs may contribute to angiogenesis of HRGECs through PI3K/Akt/VEGF pathway; therefore, Slit2/Robo1 signaling may be a potent therapeutic target for the treatment of abnormal angiogenesis in early DN and may have broad implications for the treatment of other diseases dependent on pathologic angiogenesis.