BACKGROUND:Electrolyte abnormalities following chemotherapy are common and clinically significant complications in cancer patients and are often associated with treatment delays and adverse outcomes. This study aims to develop and validate machine learning models to predict eight electrolyte abnormalities in cancer patients receiving chemotherapy. METHODS:We retrospectively analyze medical records of cancer patients from two tertiary hospitals in Korea (n = 11,227). Four machine learning algorithms are used to predict eight electrolyte abnormalities occurring within 4 weeks of chemotherapy initiation. Model performance is evaluated using a comprehensive validation framework, including internal, external, and temporal validation, and model interpretability is assessed using Shapley additive explanations. RESULTS:Here we show that electrolyte abnormalities occur in 4451 patients (74.0%) in the internal cohort and 4414 patients (84.7%) in the external cohort, with in-hospital mortality rates of 35.9% and 32.8%, respectively. The best-performing models achieve an average area under the receiver operating characteristic curve of 0.798, with an average performance decline of 0.11 during external validation. Model interpretation identifies serum albumin, heart rate, and estimated glomerular filtration rate as the most important predictors across models. Risk stratification demonstrates that patients in the highest-risk quintile have 4.15-fold greater odds of developing electrolyte abnormalities, and a 2.02-fold higher risk of mortality compared with the moderate-risk. CONCLUSIONS:These machine learning models provide an effective approach for predicting electrolyte abnormalities in cancer patients receiving chemotherapy and may support risk stratification and monitoring prioritization. Future studies are needed to validate these models in more diverse populations, incorporate additional biomarkers, and explore their integration into clinical decision-support systems.

Abstract Acute pesticide poisoning frequently leads to acute kidney injury (AKI), which is strongly associated with increased mortality. However, predictive research in this area remains limited, and criteria for AKI detection in patients with pesticide poisoning are not well-defined. This study aimed to evaluate the Kidney Disease: Improving Global Outcomes (KDIGO) criteria and develop a model for early AKI prediction in patients with pesticide poisoning. This retrospective study analyzed 877 patients presenting with acute pesticide poisoning between 2015 and 2020. AKI was defined using KDIGO criteria, considering serum creatinine, urine output, and renal replacement therapy initiation. Six machine learning models with four feature selection methods were compared using fivefold cross-validation, stratified by pesticide category. The final model, Prediction of acute Kidney Injury in Pesticide intoxication (PKIP), was established. KDIGO-defined AKI was significantly associated with mortality, with AKI patients showing a 16.6% mortality compared to 4.7% in non-AKI patients. The PKIP model, incorporating 14 features selected via the Least Absolute Shrinkage and Selection Operator, demonstrated fair discrimination [AUROC 0.720 (95% CI: 0.692–0.747), AUPRC 0.513 (95% CI: 0.464–0.563)]. Furthermore, the model showed prognostic utility for mortality prediction [AUROC 0.839 (95% CI: 0.767–0.910), AUPRC 0.421 (95% CI: 0.246–0.595)]. At the predefined cutoff value of 0.420, the model achieved a sensitivity of 39.0% and a specificity of 89.7%. Risk stratification based on PKIP probabilities showed significant differences in outcomes between groups. The high-risk group demonstrated significantly higher risks of AKI occurrence, progression to higher AKI stages, and mortality compared to the low-risk group. PKIP exhibited superior risk stratification for both AKI and mortality prediction compared to the APACHE II score. This study validates the use of KDIGO criteria for AKI detection in pesticide poisoning and introduces the PKIP model as a tool demonstrating moderate discrimination for early AKI prediction and risk stratification. The web-based PKIP tool can serve as a practical instrument for clinical decision-making for patients with pesticide poisoning. Future research should focus on external validation of the PKIP model and assessment of its impact on patient outcomes in diverse clinical settings. Trial registration: Retrospectively registered.

Background: Catheter-related infections, such as exit-site infection and tunnel infection, are major complications in peritoneal dialysis (PD) patients, affecting their prognosis. This study investigates the association between skin conditions and catheter-related infections.Methods: Data from two distinct sources were analyzed: (1) 626 PD patients in the Korean arm of the Peritoneal Dialysis Outcomes and Practice Patterns Study (PDOPPS) and (2) skin microbiome data from 76 dialysis patients at Soonchunhyang University Cheonan Hospital. The relationship between catheter-related infection and self-reported xerosis and pruritus severity was assessed by Cox regression. Risk factors for xerosis and pruritus were evaluated by logistic regression. Furthermore, we discovered the relationship between the severity of pruritus and the relative abundance of Staphylococcus aureus on the skin.Results: The risk of catheter-related infections in PD patients increased with xerosis (hazard ratio [HR], 2.71; 95% confidence interval [CI], 1.19–6.18) and pruritus (HR, 2.57; 95% CI, 1.27–5.22), particularly increasing the risk of S. aureus-associated catheter-related infections (xerosis: HR, 5.66; 95% CI, 1.97–16.30; pruritus: HR, 5.93; 95% CI, 2.18–16.15). The relative abundance of S. aureus was notably higher in patients with severe pruritus. Moreover, patients were more likely to exhibit severe xerosis if they owned pets, had higher serum creatinine levels, and elevated calcium-phosphorus product levels.Conclusion: Xerosis and pruritus significantly increase the risk of catheter-related infections, especially those caused by S. aureus. Instead of relying solely on prophylactic antibiotics for infection prevention, this study highlights the need for new preventive strategies in PD patients, focusing specifically on effective skin management.

With rising pet ownership, concerns regarding pet-related infections during peritoneal dialysis (PD) have increased. This retrospective study analyzed the characteristics of PD-related infections according to pet ownership. A total of 162 PD patients treated at Soonchunhyang University Cheonan Hospital between 2016 and 2023 were reviewed. Patients were grouped by pet ownership and pet type (dog or cat) based on data obtained from PD nurse home visits. Peritonitis, exit-site infection (ESI), and tunnel infection (TI) were defined according to the International Society of Peritoneal Dialysis guidelines, and data on causative organisms and clinical outcomes were collected. Zoonotic microorganism-associated peritonitis episodes were identified in patients with pets. Staphylococcus-associated ESIs (55.6 vs. 16.2%, p = 0.006) were more frequently observed in patients with pets than in those without pets. However, the overall incidence of peritonitis and ESI did not differ significantly between patients with and without pets, and pet ownership was not associated with mortality or PD catheter removal. These findings suggest that while pet ownership may influence the microbial characteristics of PD-related infections, it does not appear to increase overall infection incidence or adverse clinical outcomes.

Background:The independent impact of type 2 diabetes mellitus (T2DM) on kidney outcomes beyond albuminuria remains unclear. We evaluated whether T2DM affects kidney outcomes in individuals with normal kidney function without albuminuria. Methods:Data from the National Health Insurance Service-National Sample Cohort of Korea (2009-2015) were analyzed. Individuals with normal kidney function were stratified by T2DM status. The primary outcome was a composite kidney outcome consisting of initiation of kidney replacement therapy and a sustained decline in estimated glomerular filtration rate (eGFR) of ≥40% from baseline. Results:Among 77,267 individuals with normal kidney function without albuminuria, patients with T2DM (n = 11,957) showed significantly steeper annual decline in eGFR than non-T2DM individuals (-0.113 mL/min per 1.73 m2 per year; 95% confidence interval [CI], -0.222 to -0.003). T2DM was associated with a 57% higher risk of composite kidney outcome (adjusted hazard ratio, 1.57; 95% CI, 1.28-1.92), independent of traditional risk factors. This association was strongest in individuals with glomerular hyperfiltration and longer T2DM duration (≥6 years). Conclusion:Normal kidney function T2DM was associated with accelerated kidney function decline and a 1.5-fold increased risk of adverse kidney outcomes compared with normal kidney function non-T2DM, particularly in individuals with glomerular hyperfiltration and longer duration.

Background:Operational variability in defining acute kidney injury (AKI) undermines the diagnostic reliability and impairs the generalizability of machine learning (ML) models. We evaluated whether the refined criteria based on the Kidney Disease: Improving Global Outcomes (KDIGO) guidelines enhance prognostic validity and predictive performance. Methods:Two conservative refinements to the KDIGO serum creatinine (SCr)-based criteria were proposed: 1) exclusion of transient SCr decreases from baseline estimation and 2) application of a minimum absolute SCr increase threshold. We generated 441 AKI labeling strategies by combining these refinements with baseline estimation methods. For each strategy, we calculated adjusted hazard ratios (aHRs) for adverse outcomes and assessed CatBoost model performance using the area under the receiver operating characteristic curve (AUROC) and area under the precision-recall curve (AUPRC). To assess real-world interpretability, ambiguous cases were reviewed. Results:Among 5,115 patients labeled by the standard KDIGO, 4,137 (80.9%) met the refined criteria, and 978 (19.1%) were considered ambiguous. The refined cohort had higher adverse event rates than the standard cohort (44.0% vs. 37.8%; aHR, 1.23). ML models trained with refined labels showed improved discrimination (AUROC, 0.935 vs. 0.926) and precision (AUPRC, 0.729 vs. 0.707) and reduced false-positive alerts by 27% with minimal loss in adverse outcome detection. Ambiguous cases often reflected transient measurement variability or frailty. Conclusion:This study presents a strategic AKI labeling approach based on KDIGO criteria and suggests a dual-alert system to distinguish refined AKI and ambiguous AKI cases. These approaches may enhance both predictive model performance and clinically meaningful detection.

Background:Tau, a microtubule-associated protein, has a well-established role in neurodegenerative disorders, but its function in the kidney and contribution to chronic kidney disease (CKD) remain unclear. Although emerging data suggest the role of tau in CKD pathogenesis, the mechanisms of tau-mediated renal injury are not well defined. In this context, we investigated tau expression in CKD and its potential link to disease progression. Methods:Tau protein levels were measured in kidneys from CKD patients and three murine CKD models: unilateral ureteral obstruction (UUO), folic acid-induced CKD (FA-CKD), and adenine diet-induced CKD (AD-CKD). Renal functional and histological parameters were assessed. Disease progression was further evaluated in P301S tau transgenic mice. To assess potential therapeutic effects, a tau-specific antibody was administered to the AD-CKD mouse model. Results:Expression of tau-including total, phosphorylated, and acetylated forms-was markedly elevated in the kidneys of CKD patients and across all three CKD mouse models (UUO, FA-CKD, and AD-CKD). This elevation correlated with worsening renal function, tubular injury, and kidney fibrosis. Tau was upregulated as CKD progressed, and was predominantly observed in proximal tubular epithelial cells, but was also detected in glomerular podocytes and mesangial cells. In tau transgenic mice, tau accumulation worsened renal dysfunction and pathological changes, while administering a tau-specific antibody in AD-CKD mice reduced renal tau levels and improved both tubular dilatation and fibrosis. Conclusion:Tau is upregulated during CKD progression and contributes to tubular injury and fibrosis. Targeting tau could offer a new therapeutic approach for reducing renal injury in CKD.

Pesticide poisoning remains a significant public health issue, characterized by high morbidity and mortality, particularly among patients presenting to the emergency department. This study aimed to develop a 14-day in-hospital mortality prediction model for patients with acute pesticide poisoning using early clinical and laboratory data. This retrospective cohort study included 1056 patients who visited Soonchunhyang University Cheonan Hospital between January 2015 and December 2020. The cohort was randomly divided into train (n = 739) and test (n = 317) sets using stratification by pesticide type and outcome. Candidate predictors were selected based on univariate Cox regression, LASSO regularization, random forest feature importance, and clinical relevance derived from established prognostic scoring systems. Logistic regression models were constructed using six distinct feature sets. The best-performing model combined LASSO-selected and clinically curated features (AUC 0.926 [0.890–0.957]), while the final model—selected for interpretability—used only LASSO-selected features (AUC 0.923 [0.884–0.955]; balanced accuracy 0.835; sensitivity 0.843; specificity 0.857; F1.5 score 0.714 at threshold 0.450). SHapley Additive exPlanations (SHAP) analysis identified paraquat ingestion, Glasgow Coma Scale, bicarbonate level, base excess, and alcohol history as major mortality predictors. The proposed model outperformed the APACHE II score (AUC 0.835 [0.781–0.888]) and may serve as a valuable tool for early risk stratification and clinical decision making in pesticide-poisoned patients.

Chronic kidney disease (CKD) is increasingly recognized as a major global health burden, with neurological manifestations such as cognitive impairment and depression significantly contributing to patient disability. Despite the rising prevalence of cognitive decline in CKD, its underlying mechanisms remain poorly understood. This study aims to elucidate the mechanisms associated with cognitive impairment in CKD by utilizing single nucleus RNA sequencing of CKD-brain, providing novel insights into potential pathways and therapeutic targets. A rat model of chronic kidney disease (CKD) was established through 5/6 nephrectomy, followed by a 10-week observation period. Behavioral analyses were performed to assess cognitive function prior to brain tissue collection. The frontal lobe of the brain was then harvested for single-nucleus RNA sequencing (snRNA-seq), conducted using the 10× Chromium Single Cell 3’ v3.1 protocol (10× Genomics, document no. CG000315), to investigate transcriptional changes linked to cognitive impairment. The study identified eight cell type clusters, with astrocytes, excitatory neurons, and inhibitory neurons showing significant transcriptional changes in CKD. Total cell counts were higher in controls, but cell type proportions were comparable between groups. Astrocytes exhibited the most differentially expressed genes (DEGs), including upregulated Kcnj3 and Nmnat2 and downregulated Zmiz2. These findings, supported by pseudobulk analysis, highlight astrocytes’ key role in cognitive impairment in CKD. The results align with their involvement in other brain diseases, such as Alzheimer's and inflammatory conditions. Sub-clustering and trajectory analysis identified six astrocyte sub-clusters, with clusters 2, 3, and 5 showing increased proportions in CKD rats, indicating their role in CKD-related cognitive impairment. Cluster 4, with no proportional change, was linked to reactive astrocytes and inflammatory pathways. In contrast, cluster 5, unique to CKD, was associated with synaptic signaling, neurotransmitter receptors, and ion channel dysfunction. Pathway analysis revealed metabolic derangement in cluster 2, similar to Alzheimer's disease, and ferroptosis in cluster 3, a novel cell death mechanism. Collectively, CKD astrocytes exhibited distinct mechanisms of cognitive impairment, including metabolic disruption, ferroptosis, and synaptic dysfunction. Astrocytes, especially CKD-specific sub-clusters, were identified as one of the most affected cells by snRNA-seq analysis. Our study demonstrated for the first time the heterogeneity of reactive astrocytes in response to the CKD environment. The presence of the CKD-specific astrocyte population suggested that the heterogeneity of reactive astrocytes could be a crucial target for future research. In addition, our findings would provide valuable insights into the pathogenesis of cognitive impairment in patients with CKD.

BACKGROUND:Chronic kidney disease (CKD)-associated pruritus is a severe distressing condition that frequently occurs in patients undergoing dialysis. In this study, the profile of the skin microbiome was analyzed to understand the underlying etiology and potential treatments. METHODS:Seventy-six end-stage kidney disease (ESKD) patients (hemodialysis, 40; peritoneal dialysis, 36) and 15 healthy controls were enrolled and swabbed at three sites: back, antecubital fossa, and shin. The pruritus severity of the enrolled subjects was validated by the Worst Itch Numeric Rating Scale (WI-NRS), 5-D itch scale, and Uremic Pruritus in Dialysis Patients (UP-Dial). The 16S genebased metagenomics method was applied to skin microbiome analysis. RESULTS:In the comparison of bacterial communities of ESKD patients and the control group, there was a significant difference on back. Specifically, the average composition ratio of the Cutibacterium in the back samples was significantly lower in ESKD patients than in healthy controls (p < 0.01). In further analysis of ESKD patients, Cutibacterium was significantly lower in the high pruritus group than in the low pruritus group (p < 0.05), even though other clinical parameters such as age, calcium-phosphorus product, and intact parathyroid hormone showed no significance difference between the groups. CONCLUSION:In ESKD patients, the skin microbiome of the back was significantly altered, and the severity of itching was related to the reduction of Cutibacterium. This research reveals the relationship between skin microbiota and CKD-associated pruritus in multiple skin sites for the first time. The results of this study suggest a potential data basis for the diagnosis and treatment of CKD-associated pruritus.

Background: In patients with type 2 diabetes mellitus (T2DM), diabetic kidney disease (DKD) is diagnosed based on clinical features. A kidney biopsy is used only in selected cases. This study aimed to reconsider the role of a biopsy in predicting renal outcomes. Methods: Clinical and laboratory parameters and renal biopsy results were obtained from 237 patients with T2DM who underwent renal biopsies at Soonchunhyang University Cheonan Hospital between January 2000 and March 2020 and were analyzed. Results: Of 237 diabetic patients, 29.1% had DKD only, 61.6% had non-DKD (NDKD), and 9.3% had DKD with coexisting NDKD (DKD/NDKD). Of the patients with DKD alone, 43.5% progressed to end-stage kidney disease (ESKD), while 15.8% of NDKD patients and 36.4% of DKD/NDKD patients progressed to ESKD (p < 0.001). In the DKD-alone group, pathologic features like ≥50% global sclerosis (p < 0.001), tubular atrophy (p < 0.001), interstitial fibrosis (p < 0.001), interstitial inflammation (p < 0.001), and the presence of hyalinosis (p = 0.03) were related to worse renal outcomes. The Cox regression model showed a higher risk of progression to ESKD in the DKD/NDKD group compared to the DKD-alone group (hazard ratio [HR], 2.73; p = 0.032), ≥50% global sclerosis (HR, 3.88; p < 0.001), and the degree of mesangial expansion (moderate: HR, 2.45; p = 0.045 and severe: HR, 6.22; p < 0.001). Conclusion: In patients with T2DM, a kidney biopsy can help in identifying patients with NDKD for appropriate treatment, and it has predictive value.

Rationale: Peritonitis caused by Pantoea agglomerans is a rare occurrence in patients undergoing peritoneal dialysis. Cases potentially linked to pet dogs are even rarer, and there is limited literature available. Patient concerns: A patient undergoing peritoneal dialysis presented with symptoms of peritonitis, including abdominal pain and cloudy dialysis fluid. Diagnoses: Microbiological analysis identified P agglomerans as the causative organism. Interventions: The patient was treated with targeted antibiotic therapy and showed a positive response. Outcomes: During a subsequent medical interview, it was revealed that the patient had close contact with their pet dog, raising the possibility that the infection may have been associated with this exposure. Lessons: This case highlights the importance of considering zoonotic transmission as a potential source of infection in peritoneal dialysis patients, particularly when there is close contact with pets. Healthcare providers should educate patients about the potential risks posed by pets and implement preventive strategies to mitigate such risks.

BACKGROUND:Receptor-interacting protein kinase (RIPK)3 is an essential molecule for necroptosis and its role in kidney fibrosis has been investigated using various kidney injury models. However, the relevance and the underlying mechanisms of RIPK3 to podocyte injury in albuminuric diabetic kidney disease (DKD) remain unclear. Here, we investigated the role of RIPK3 in glomerular injury of DKD. METHODS:We analyzed RIPK3 expression levels in the kidneys of patients with biopsy-proven DKD and animal models of DKD. Additionally, to confirm the clinical significance of circulating RIPK3, RIPK3 was measured by ELISA in plasma obtained from a prospective observational cohort of patients with type 2 diabetes, and estimated glomerular filtration rate (eGFR) and urine albumin-to-creatinine ratio (UACR), which are indicators of renal function, were followed up during the observation period. To investigate the role of RIPK3 in glomerular damage in DKD, we induced a DKD model using a high-fat diet in Ripk3 knockout and wild-type mice. To assess whether mitochondrial dysfunction and albuminuria in DKD take a Ripk3-dependent pathway, we used single-cell RNA sequencing of kidney cortex and immortalized podocytes treated with high glucose or overexpressing RIPK3. RESULTS:RIPK3 expression was increased in podocytes of diabetic glomeruli with increased albuminuria and decreased podocyte numbers. Plasma RIPK3 levels were significantly elevated in albuminuric diabetic patients than in non-diabetic controls (p = 0.002) and non-albuminuric diabetic patients (p = 0.046). The participants in the highest tertile of plasma RIPK3 had a higher incidence of renal progression (hazard ratio [HR] 2.29 [1.05-4.98]) and incident chronic kidney disease (HR 4.08 [1.10-15.13]). Ripk3 knockout improved albuminuria, podocyte loss, and renal ultrastructure in DKD mice. Increased mitochondrial fragmentation, upregulated mitochondrial fission-related proteins such as phosphoglycerate mutase family member 5 (PGAM5) and dynamin-related protein 1 (Drp1), and mitochondrial ROS were decreased in podocytes of Ripk3 knockout DKD mice. In cultured podocytes, RIPK3 inhibition attenuated mitochondrial fission and mitochondrial dysfunction by decreasing p-mixed lineage kinase domain-like protein (MLKL), PGAM5, and p-Drp1 S616 and mitochondrial translocation of Drp1. CONCLUSIONS:The study demonstrates that RIPK3 reflects deterioration of renal function of DKD. In addition, RIPK3 induces diabetic podocytopathy by regulating mitochondrial fission via PGAM5-Drp1 signaling through MLKL. Inhibition of RIPK3 might be a promising therapeutic option for treating DKD.

BACKGROUND:Neurologic complications, such as cognitive and emotional dysfunction, have frequently been observed in chronic kidney disease (CKD) patients. Previous research shows that uremic toxins play a role in the pathogenesis of CKD-associated cognitive impairment. Since astrocytes contribute to the protection and survival of neurons, astrocyte function and brain metabolism may contribute to the pathogenesis of neurodegeneration. Indoxyl sulfate (IS) is the most popular uremic toxin. However, how IS-induced astrocyte injury brings about neurologic complications in CKD patients has not been elucidated. METHODS:The rate of extracellular acidification was measured in astrocytes when IS (0.5-3 mM, 4 or 7 days) treatment was applied. The hexokinase 1 (HK1), pyruvate kinase isozyme M2 (PKM2), pyruvate dehydrogenase (PDH), and phosphofructokinase (PFKP) protein levels were also measured. The activation of the apoptotic pathway was investigated using a confocal microscope, fluorescence- activated cell sorting, and cell three-dimensional imaging was used. RESULTS:In astrocytes, IS affected glycolysis in not only dose-dependently but also time-dependently. Additionally, HK1, PKM2, PDH, and PFKP levels were decreased in IS-treated group when compared to the control. The results were prominent in cases with higher doses and longer exposure duration. The apoptotic features after IS treatment were also observed. CONCLUSION:Our results showed that the inhibition of glycolysis by IS in astrocytes leads to cell death via apoptosis. Specifically, longterm and higher-dose exposures had more serious effects on astrocytes. Our results suggest that the glycolysis pathway and related targets could provide a novel approach to cognitive dysfunction in CKD patients.

Background: Chronic kidney disease is a significant health burden worldwide, with increasing incidence. Although several genome- wide association studies (GWAS) have investigated single nucleotide polymorphisms (SNP) associated with kidney trait, most studies were focused on European ancestry. Methods: We utilized clinical and genetic information collected from the Korean Genome and Epidemiology Study (KoGES). Results: More than five million SNPs from 58,406 participants were analyzed. After meta-GWAS, 1,360 loci associated with estimated glomerular filtration rate (eGFR) at a genome-wide significant level (p = 5 × 10–8) were identified. Among them, 399 loci were validated with at least one other biomarker (blood urea nitrogen [BUN] or eGFRcysC) and 149 loci were validated using both markers. Among them, 18 SNPs (nine known ones and nine novel ones) with 20 putative genes were found. The aggregated effect of genes estimated by MAGMA gene analysis showed that these significant genes were enriched in kidney-associated pathways, with the kidney and liver being the most enriched tissues. Conclusion: In this study, we conducted GWAS for more than 50,000 Korean individuals and identified several variants associated with kidney traits, including eGFR, BUN, and eGFRcysC. We also investigated functions of relevant genes using computational methods to define putative causal variants.

Acute kidney injury (AKI) is a significant health challenge associated with adverse patient outcomes and substantial economic burdens. Many authors have sought to prevent and predict AKI. Here, we comprehensively review recent advances in the use of artificial intelligence (AI) to predict AKI, and the associated challenges. Although AI may detect AKI early and predict prognosis, integration of AI-based systems into clinical practice remains challenging. It is difficult to identify AKI patients using retrospective data; information preprocessing and the limitations of existing models pose problems. It is essential to embrace standardized labeling criteria and to form international multi-institutional collaborations that foster high-quality data collection. Additionally, existing constraints on the deployment of evolving AI technologies in real-world healthcare settings and enhancement of the reliabilities of AI outputs are crucial. Such efforts will improve the clinical applicability, performance, and reliability of AKI Clinical Support Systems, ultimately enhancing patient prognoses.

Abstract Kidney fibrosis causes irreversible structural damage in chronic kidney disease and is characterized by aberrant extracellular matrix (ECM) accumulation. Although glutamyl-prolyl-tRNA synthetase 1 (EPRS1) is a crucial enzyme involved in proline-rich protein synthesis, its role in kidney fibrosis remains unclear. The present study revealed that EPRS1 expression levels were increased in the fibrotic kidneys of patients and mice, especially in fibroblasts and proximal tubular epithelial cells, on the basis of single-cell analysis and immunostaining of fibrotic kidneys. Moreover, C57BL/6 EPRS1tm1b heterozygous knockout (Eprs1 +/−) and pharmacological EPRS1 inhibition with the first-in-class EPRS1 inhibitor DWN12088 protected against kidney fibrosis and dysfunction by preventing fibroblast activation and proximal tubular injury. Interestingly, in vitro assays demonstrated that EPRS1-mediated nontranslational pathways in addition to translational pathways under transforming growth factor β-treated conditions by phosphorylating SMAD family member 3 in fibroblasts and signal transducers and activators of transcription 3 in injured proximal tubules. EPRS1 knockdown and catalytic inhibition suppressed these pathways, preventing fibroblast activation, proliferation, and subsequent collagen production. Additionally, we revealed that EPRS1 caused mitochondrial damage in proximal tubules but that this damage was attenuated by EPRS1 inhibition. Our findings suggest that the EPRS1-mediated ECM accumulation induces kidney fibrosis via fibroblast activation and mitochondrial dysfunction. Therefore, targeting EPRS1 could be a potential therapeutic target for alleviating fibrotic injury in chronic kidney disease.

BACKGROUND:Acute kidney injury (AKI) is a significant challenge in healthcare. While there are considerable researches dedicated to AKI patients, a crucial factor in their renal function recovery, is often overlooked. Thus, our study aims to address this issue through the development of a machine learning model to predict restoration of kidney function in patients with AKI. METHODS:Our study encompassed data from 350,345 cases, derived from three hospitals. AKI was classified in accordance with the Kidney Disease: Improving Global Outcomes. Criteria for recovery were established as either a 33% decrease in serum creatinine levels at AKI onset, which was initially employed for the diagnosis of AKI. We employed various machine learning models, selecting 43 pertinent features for analysis. RESULTS:Our analysis contained 7,041 and 2,929 patients' data from internal cohort and external cohort respectively. The Categorical Boosting Model demonstrated significant predictive accuracy, as evidenced by an internal area under the receiver operating characteristic (AUROC) of 0.7860, and an external AUROC score of 0.7316, thereby confirming its robustness in predictive performance. SHapley Additive exPlanations (SHAP) values were employed to explain key factors impacting recovery of renal function in AKI patients. CONCLUSION:This study presented a machine learning approach for predicting renal function recovery in patients with AKI. The model performance was assessed across distinct hospital settings, which revealed its efficacy. Although the model exhibited favorable outcomes, the necessity for further enhancements and the incorporation of more diverse datasets is imperative for its application in real- world.

It has been reported that a scenario-based cognitive behavioral therapy mobile app including Todac Todac was effective in improving depression in the general public. However, no study has been conducted on whether Todac Todac is effective in dialysis patients. Therefore, this study was intended to determine whether the use of this app was effective in improving depression in dialysis patients. Sixty-five end-stage kidney disease patients receiving dialysis at Soonchunhyang University Cheonan Hospital were randomly assigned to the Todac Todac app program (experimental group) or an E-moods daily mood chart app program (control group) for 3 weeks. The degree of depression was measured before and after using the app.After the end of the 3-week program, a small but significant improvement was observed in the Trait anxiety (p < 0.05) and Beck depression index (p < 0.05) in E-moods group and DAS-K scores (p < 0.05) in Todac Todac group. However, no differences were seen in any parameters between the two groups. In addition, Todac Todac was not statistically more effective than the control intervention in the subgroup analysis. The Todac Todac, a scenario-based cognitive behavioral therapy mobile app, seemed to have a limited effect on improving depression in dialysis patients. Therefore, it is necessary to develop new tools to improve depression in dialysis patients.

Applying deep learning to medical research with limited data is challenging. This study focuses on addressing this difficulty through a case study, predicting acute respiratory failure (ARF) in patients with acute pesticide poisoning. Commonly, out-of-distribution (OOD) data are overlooked during model training in the medical field. Our approach integrates OOD data and transfer learning (TL) to enhance model performance with limited data. We fine-tuned a pre-trained multi-layer perceptron model using OOD data, outperforming baseline models. Shapley additive explanation (SHAP) values were employed for model interpretation, revealing the key factors associated with ARF. Our study is pioneering in applying OOD and TL techniques to electronic health records to achieve better model performance in scenarios with limited data. Our research highlights the potential benefits of using OOD data for initializing weights and demonstrates that TL can significantly improve model performance, even in medical data with limited samples. Our findings emphasize the significance of utilizing context-specific information in TL to achieve better results. Our work has practical implications for addressing challenges in rare diseases and other scenarios with limited data, thereby contributing to the development of machine-learning techniques within the medical field, especially regarding health inequities.