Time series changes in pseudo-R2 values regarding maximum glomerular diameter and the Oxford MEST-C score in patients with IgA nephropathy: A long-term follow-up study

Machine learning-based diagnosis and prognosis of IgAN: A systematic review and meta-analysis

Purpose

Plenty of studies have explored the diagnosis and prognosis of IgA nephropathy (IgAN) based on machine learning (ML), but the accuracy lacks the support of evidence-based medical evidence. We aim at this problem to guide the precision treatment of IgAN.

Methods

Embase, Pubmed, Cochrane Library, and Web of Science were searched systematically until February 24th, 2024, for publications on ML-based diagnosis and prognosis of IgAN. Subgroup analysis or meta-regression was conducted according to modeling method, follow-up time, endpoint definition, and variable type. Further, the rank sum test was applied to compare the discrimination ability of prognosis.

Results

A total of 47 studies involving 51,935 patients were eligible. Among the 38 diagnostic models, the pooled C-index was 0.902 (95 % CI: 0.878-0.926) in 27 diagnostic models. Of the 162 prognostic models, the C-index for model discrimination of 144 prognostic models was 0.838 (95 % CI: 0.827-0.850) in training. The overall discrimination ability of prognosis was as follows: COX regression > new ML models (e.g. ANN, DT, RF, SVM, XGBoost) > traditional ML models (logistic regression) > Naïve Bayesian network (P < 0.05). External validation of IIgAN-RPT in 19 models showed a pooled C-index of 0.801 (95 % CI: 0.784-0.817).

Conclusions

New ML models have shown application values that are as good as traditional ML models, both in diagnosis and prognosis. In addition, future models are desired to use a more sensitive prognostic endpoint (albuminuria), improve predictive ability in moderate progression risk, and ultimately translate into clinically applicable intelligent tools.

Glomerular hyperfiltration and hypertrophy: an evaluation of maximum values in pathological indicators to discriminate "diseased" from "normal"

The success of sodium-glucose cotransporter 2 inhibitors and bariatric surgery in patients with chronic kidney disease has highlighted the importance of glomerular hyperfiltration and hypertrophy in the progression of kidney disease. Sustained glomerular hyperfiltration and hypertrophy can lead to glomerular injury and progressive kidney damage. This article explores the relationship between obesity and chronic kidney disease, focusing on the roles of glomerular hyperfiltration and hypertrophy as hallmarks of obesity-related kidney disease. The pathological mechanisms underlying this association include adipose tissue inflammation, dyslipidemia, insulin resistance, chronic systemic inflammation, oxidative stress, and overactivation of the sympathetic nervous system, as well as the renin-angiotensin aldosterone system. This article explains how glomerular hyperfiltration results from increased renal blood flow and intraglomerular hypertension, inducing mechanical stress on the filtration barrier and post-filtration structures. Injured glomeruli increase in size before sclerosing and collapsing. Therefore, using extreme values, such as the maximal glomerular diameter, could improve the understanding of the data distribution and allow for better kidney failure predictions. This review provides important insights into the mechanisms underlying glomerular hyperfiltration and hypertrophy and highlights the need for further research using glomerular size, including maximum glomerular profile, calculated using needle biopsy specimens.

Visceral fat and attribute-based medicine in chronic kidney disease

Visceral adipose tissue plays a central role in obesity and metabolic syndrome and is an independent risk factor for both cardiovascular and metabolic disorders. Increased visceral adipose tissue promotes adipokine dysregulation and insulin resistance, leading to several health issues, including systemic inflammation, oxidative stress, and activation of the renin-angiotensin-aldosterone system. Moreover, an increase in adipose tissue directly and indirectly affects the kidneys by increasing renal sodium reabsorption, causing glomerular hyperfiltration and hypertrophy, which leads to increased proteinuria and kidney fibrosis/dysfunction. Although the interest in the adverse effects of obesity on renal diseases has grown exponentially in recent years, the relationship between obesity and renal prognosis remains controversial. This may be attributed to the long clinical course of obesity, numerous obesity-related metabolic complications, and patients' attributes. Multiple individual attributes influencing the pathophysiology of fat accumulation make it difficult to understand obesity. In such cases, it may be effective to elucidate the pathophysiology by conducting research tailored to individual attributes from the perspective of attribute-based medicine/personalized medicine. We consider the appropriate use of clinical indicators necessary, according to attributes such as chronic kidney disease stage, level of visceral adipose tissue accumulation, age, and sex. Selecting treatments and clinical indicators based on individual attributes will allow for advancements in the clinical management of patients with obesity and chronic kidney disease. In the clinical setting of obesity-related nephropathy, it is first necessary to accumulate attribute-based studies resulting from the accurate evaluation of visceral fat accumulation to establish evidence for promoting personalized medicine.

Machine learning for morbid glomerular hypertrophy

A practical research method integrating data-driven machine learning with conventional model-driven statistics is sought after in medicine. Although glomerular hypertrophy (or a large renal corpuscle) on renal biopsy has pathophysiological implications, it is often misdiagnosed as adaptive/compensatory hypertrophy. Using a generative machine learning method, we aimed to explore the factors associated with a maximal glomerular diameter of ≥ 242.3 μm. Using the frequency-of-usage variable ranking in generative models, we defined the machine learning scores with symbolic regression via genetic programming (SR via GP). We compared important variables selected by SR with those selected by a point-biserial correlation coefficient using multivariable logistic and linear regressions to validate discriminatory ability, goodness-of-fit, and collinearity. Body mass index, complement component C3, serum total protein, arteriolosclerosis, C-reactive protein, and the Oxford E1 score were ranked among the top 10 variables with high machine learning scores using SR via GP, while the estimated glomerular filtration rate was ranked 46 among the 60 variables. In multivariable analyses, the R² value was higher (0.61 vs. 0.45), and the corrected Akaike Information Criterion value was lower (402.7 vs. 417.2) with variables selected with SR than those selected with point-biserial r. There were two variables with variance inflation factors higher than 5 in those using point-biserial r and none in SR. Data-driven machine learning models may be useful in identifying significant and insignificant correlated factors. Our method may be generalized to other medical research due to the procedural simplicity of using top-ranked variables selected by machine learning.

Prediction models used in the progression of chronic kidney disease: A scoping review

Objective

To provide a review of prediction models that have been used to measure clinical or pathological progression of chronic kidney disease (CKD).

Design

Scoping review.

Data sources

Medline, EMBASE, CINAHL and Scopus from the year 2011 to 17th February 2022.

Study selection

All English written studies that are published in peer-reviewed journals in any country, that developed at least a statistical or computational model that predicted the risk of CKD progression.

Data extraction

Eligible studies for full text review were assessed on the methods that were used to predict the progression of CKD. The type of information extracted included: the author(s), title of article, year of publication, study dates, study location, number of participants, study design, predicted outcomes, type of prediction model, prediction variables used, validation assessment, limitations and implications.

Results

From 516 studies, 33 were included for full-text review. A qualitative analysis of the articles was compared following the extracted information. The study populations across the studies were heterogenous and data acquired by the studies were sourced from different levels and locations of healthcare systems. 31 studies implemented supervised models, and 2 studies included unsupervised models. Regardless of the model used, the predicted outcome included measurement of risk of progression towards end-stage kidney disease (ESKD) of related definitions, over given time intervals. However, there is a lack of reporting consistency on details of the development of their prediction models.

Conclusions

Researchers are working towards producing an effective model to provide key insights into the progression of CKD. This review found that cox regression modelling was predominantly used among the small number of studies in the review. This made it difficult to perform a comparison between ML algorithms, more so when different validation methods were used in different cohort types. There needs to be increased investment in a more consistent and reproducible approach for future studies looking to develop risk prediction models for CKD progression.

Impact of visceral fat area in patients with chronic kidney disease

Longitudinal studies evaluating the association between visceral fat area (VFA) and kidney function decline in patients with chronic kidney disease (CKD) are limited, and little is known about VFA interactions contributing to the kidney prognosis (e.g. interactions between VFA ≥ 100 cm² and age, sex, and CKD category). In this study, we stratified patients with CKD according to VFA category, as well as age, sex, CKD category, hyperglycemia, and diabetes mellitus, and determined the ability of obesity-related indicators (body mass index, waist circumference, subcutaneous fat area, visceral-to-subcutaneous fat ratio) to predict the renal prognosis. Kidney outcomes (≥ 50% estimated glomerular filtration rate decline or end-stage kidney disease) were examined in 200 patients with CKD (median follow-up, 12.3 years). On multivariable Cox analysis, an increase in VFA (10-cm² increase) was significantly associated with kidney outcomes in the entire cohort, and VFA was significantly associated with kidney disease progression even in the VFA < 100 cm² sub-cohort. Interestingly, the hazard ratio (HR) was higher for VFA (10-cm² increase) than for the VFA ≥ 100 cm² sub-cohort (HR 1.33 vs. 1.07). Overall, VFA was found to be the most versatile obesity-related indicator associated with kidney disease progression. VFA was associated with the primary outcome in the sub-cohorts of CKD stages 1-2, hyperglycemia, and diabetes mellitus. A high VFA was a significant kidney prognostic factor in the entire CKD cohort, with greater significance in patients with VFA < 100 cm² than in patients with VFA ≥ 100 cm². Our results may provide new insights into strategies for treating CKD.

Sex Differences in Time-Series Changes in Pseudo-R2 Values Regarding Hyperuricemia in Relation to the Kidney Prognosis

Studies on sex differences in time-series changes in pseudo-R² values regarding hyperuricemia (HU) in relation to the kidney prognosis among patients with chronic kidney disease (CKD) are scant. The kidney prognosis was evaluated in 200 patients with CKD (median follow-up, 12.3 years). Survival analyses and logistic regression analyses were conducted, generating time-series pseudo-R² values. We used four definitions of HU according to serum uric acid (SUA) levels (HU6, SUA ≥ 6.0 mg/dL; HU7, SUA ≥ 7.0 mg/dL; HU8, SUA ≥ 8.0 mg/dL) and antihyperuricemic agent use to calculate the mean and percentage of the change in pseudo-R² values from the 6th year until the end of the study (6Y–End Mean and 6Y–End Change, respectively). The multivariable Cox regression analysis showed that HU7 was significantly associated with kidney outcomes. When stratified by sex, the 6Y–End Mean was clearly higher in women than in men for all HU definitions, with the highest value (0.1755) obtained for HU7 in women. The pseudo-R² values for HU6 in women showed an increasing pattern, with a 6Y–End Change of 11.4%/year. Thus, it may be clinically meaningful to consider sex differences in the time-series pseudo-R² values regarding HU and kidney outcomes.

Visceral to subcutaneous fat ratio as an indicator of a ≥30% eGFR decline in chronic kidney disease

Whether the visceral-to-subcutaneous fat ratio (V/S ratio) is associated with renal prognosis in patients with chronic kidney disease (CKD) remains unclear. Furthermore, little is known about the effect of sex and the absolute amount of visceral fat accumulation such as visceral fat area (VFA) ≥100 cm² on the V/S ratio in relation to renal prognosis. In this study, 200 patients with CKD were evaluated for renal prognosis. Survival analyses and logistic regression analyses were conducted, generating time-series pseudo-R² values. The mean and percent change of the pseudo-R² values from the 6^th year to the 10^th year (6Y–10Y Mean and 6Y–10Y Change, respectively) were calculated for determining the cut-off points for the medium-term renal prognosis. Multivariate Cox regression analysis revealed that the V/S ratio was significantly associated with renal outcomes and that the VFA category (VFA ≥ 100 cm²) had significant interactions with the V/S ratio regarding renal prognosis. The hazard ratio (HR) of the V/S ratio was higher in the sub-cohort of VFA < 100 cm² than in the sub-cohort of VFA ≥ 100 cm² (HR: 6.42 vs. 1.00). Regarding sex differences, a strong association was noted between the V/S ratio and renal prognosis in women but not in men (HR: 2.40 vs. 1.10). On the other hand, 6Y–10Y Mean of the pseudo-R² values indicated differences in the cut-off points of the V/S ratio between men and women (V/S ratio: 0.75 vs. 0.5). Our findings indicate that it may be clinically meaningful to consider the differences in sex and the amount of VFA ≥100 cm² for the V/S ratio in relation to renal outcomes in patients with CKD. The 6Y–10Y Mean of the pseudo-R² values contributed to determining the cut-off points of the V/S ratio according to the sex difference.