Roxadustat Versus Erythropoietin: The Comparison of Efficacy in Reversing Ventricular Remodeling in Dialysis Patients with Anaemia.

Background: Renal anaemia and left ventricular hypertrophy are the main complications of chronic kidney disease and are shared among dialysis patients. This retrospective study aimed to compare the efficacies of the hypoxia-inducible factor prolyl hydroxylase inhibitor roxadustat and recombinant human erythropoietin in reversing ventricular remodeling in dialysis patients with renal anaemia. Methods: A total of 204 participants underwent baseline examinations, including echocardiograms and laboratory tests, before being administered either treatment for at least 24 weeks from January 2018 to October 2021, after which follow-up examinations were conducted at 6 months. Propensity score matching based on key variables included age, gender, cardiovascular diseases, cardiovascular medications, dialysis course and the vascular access at baseline was performed to include populations with similar characteristics between groups. Results: In total, 136 patients were included with roxadustat or recombinant human erythropoietin. The left ventricular mass index after treatment with roxadustat and recombinant human erythropoietin both significantly decreased after 6 months, but there was no significant difference in the change in left ventricular mass index between the two groups. In addition, the left ventricular end-diastolic diameters and left ventricular wall thickness, systolic blood pressure, and diastolic blood pressure significantly decreased in the roxadustat group. Roxadustat and recombinant human erythropoietin also increased haemoglobin significantly, but there was no significant difference in the change in haemoglobin between the two groups. The results of multiple linear regression showed that the change in haemoglobin was independent factor affecting the improvement of left ventricular mass index. Conclusions: The increase of haemoglobin was associated with improving left ventricular hypertrophy in dialysis patients. However, the beneficial effects between roxadustat and recombinant human erythropoietin on left ventricular mass index did not show clear superiority or inferiority in six months.

The ratio of neutrophils to lymphocytes effectively predicts clinical outcomes in idiopathic membranous nephropathy.

BACKGROUND: Systemic inflammatory indicators are important in the prognoses of various diseases. Such indicators, including the neutrophil-to-lymphocyte ratio (NLR), can be meaningful in predicting the clinical outcome in patients diagnosed with idiopathic membranous nephropathy (IMN). MATERIALS AND METHODS: 112 IMN patients diagnosed by renal biopsy were recruited retrospectively. The endpoint was defined as a combination of partial and complete remission. Statistical analysis determined the independent factors associated with clinical remission and the predictive utility of NLR. RESULTS: Within the 12-month follow-up period, 72 patients achieved clinical remission after treatment. Univariate analysis identified significant differences in serum albumin, estimated glomerular filtration rate (eGFR), proteinuria, neutrophil count, and NLR between the remission group and the non-remission group (all p < 0.05). Cox proportional hazards indicated that elevated eGFR (HR 1.022, 95% CI (1.009 - 1.035), p = 0.001), lower NLR (HR 0.345, 95% CI (0.237 - 0.501), p = 0.0001), and decreased proteinuria (HR 0.826, 95% CI (0.693 - 0.984), p = 0.032) were protective elements for remission. With an optimal cut-off value of 2.61, the pre-treatment NLR had an excellent ability to identify the remission (area under the curve (AUC), 0.785). Participants were separated into low- and high-NLR groups by using 2.61. Kaplan-Meier survival curves revealed significantly higher remission rates in the lower group (p < 0.0001). CONCLUSION: The NLR is an effective indicator for predicting clinical remission in patients with IMN.

Application of cloud server-based machine learning for assisting pathological structure recognition in IgA nephropathy.

BACKGROUND: Machine learning (ML) models can help assisting diagnosis by rapidly localising and classifying regions of interest (ROIs) within whole slide images (WSIs). Effective ML models for clinical decision support require a substantial dataset of 'real' data, and in reality, it should be robust, user-friendly and universally applicable. METHODS: WSIs of primary IgAN were collected and annotated. The H-AI-L algorithm which could facilitate direct WSI viewing and potential ROI detection for clinicians was built on the cloud server of matpool, a shared internet-based service platform. Model performance was evaluated using F1-score, precision, recall and Matthew's correlation coefficient (MCC). RESULTS: The F1-score of glomerular localisation in WSIs was 0.85 and 0.89 for the initial and pretrained models, respectively, with corresponding recall values of 0.79 and 0.83, and precision scores of 0.92 and 0.97. Dichotomous differentiation between global sclerotic (GS) and other glomeruli revealed F1-scores of 0.70 and 0.91, and MCC values of 0.55 and 0.87, for the initial and pretrained models, respectively. The overall F1-score of multiclassification was 0.81 for the pretrained models. The total glomerular recall rate was 0.96, with F1-scores of 0.68, 0.56 and 0.26 for GS, segmental glomerulosclerosis and crescent (C), respectively. Interstitial fibrosis/tubular atrophy lesion similarity between the true label and model predictions was 0.75. CONCLUSIONS: Our results underscore the efficacy of the ML integration algorithm in segmenting ROIs in IgAN WSIs, and the internet-based model deployment is in favour of widespread adoption and utilisation across multiple centres and increased volumes of WSIs.

Using a machine learning model to predict the development of acute kidney injury in patients with heart failure.

Background: Heart failure (HF) is a life-threatening complication of cardiovascular disease. HF patients are more likely to progress to acute kidney injury (AKI) with a poor prognosis. However, it is difficult for doctors to distinguish which patients will develop AKI accurately. This study aimed to construct a machine learning (ML) model to predict AKI occurrence in HF patients. Materials and methods: The data of HF patients from the Medical Information Mart for Intensive Care-IV (MIMIC-IV) database was retrospectively analyzed. A ML model was established to predict AKI development using decision tree, random forest (RF), support vector machine (SVM), K-nearest neighbor (KNN), and logistic regression (LR) algorithms. Thirty-nine demographic, clinical, and treatment features were used for model establishment. Accuracy, sensitivity, specificity, and the area under the receiver operating characteristic curve (AUROC) were used to evaluate the performance of the ML algorithms. Results: A total of 2,678 HF patients were engaged in this study, of whom 919 developed AKI. Among 5 ML algorithms, the RF algorithm exhibited the highest performance with the AUROC of 0.96. In addition, the Gini index showed that the sequential organ function assessment (SOFA) score, partial pressure of oxygen (PaO2), and estimated glomerular filtration rate (eGFR) were highly relevant to AKI development. Finally, to facilitate clinical application, a simple model was constructed using the 10 features screened by the Gini index. The RF algorithm also exhibited the highest performance with the AUROC of 0.95. Conclusion: Using the ML model could accurately predict the development of AKI in HF patients.

Using Machine Learning to Evaluate the Role of Microinflammation in Cardiovascular Events in Patients With Chronic Kidney Disease.

Background: Lipid metabolism disorder, as one major complication in patients with chronic kidney disease (CKD), is tied to an increased risk for cardiovascular disease (CVD). Traditional lipid-lowering statins have been found to have limited benefit for the final CVD outcome of CKD patients. Therefore, the purpose of this study was to investigate the effect of microinflammation on CVD in statin-treated CKD patients. Methods: We retrospectively analysed statin-treated CKD patients from January 2013 to September 2020. Machine learning algorithms were employed to develop models of low-density lipoprotein (LDL) levels and CVD indices. A fivefold cross-validation method was employed against the problem of overfitting. The accuracy and area under the receiver operating characteristic (ROC) curve (AUC) were acquired for evaluation. The Gini impurity index of the predictors for the random forest (RF) model was ranked to perform an analysis of importance. Results: The RF algorithm performed best for both the LDL and CVD models, with accuracies of 82.27% and 74.15%, respectively, and is therefore the most suitable method for clinical data processing. The Gini impurity ranking of the LDL model revealed that hypersensitive C-reactive protein (hs-CRP) was highly relevant, whereas statin use and sex had the least important effects on the outcomes of both the LDL and CVD models. hs-CRP was the strongest predictor of CVD events. Conclusion: Microinflammation is closely associated with potential CVD events in CKD patients, suggesting that therapeutic strategies against microinflammation should be implemented to prevent CVD events in CKD patients treated by statin.