Significance Statement: Because AKI in hospitalized children is associated with poor outcomes, a tool allowing early identification of children at risk of developing AKI may facilitate timely interventions. The authors describe various machine learning techniques used to build a parsimonious model predictive of pediatric AKI. From an initial pool of 720 potential variables, they evaluated multiple feature selection techniques to create a ten-feature logistic regression model that could predict, in time-updated fashion, the risk of AKI in the next 48 hours. A machine learning-based genetic algorithm (reflecting the process of natural selection) was the best variable selection method, using ten factors extracted from electronic health records to use for AKI prediction. Risk-stratifying hospitalized children might allow clinicians to implement targeted and timely interventions prior to AKI development.Background: Timely prediction of AKI in children can allow for targeted interventions, but the wealth of data in the electronic health record poses unique modeling challenges.Methods: We retrospectively reviewed the electronic medical records of all children younger than 18 years old who had at least two creatinine values measured during a hospital admission from January 2014 through January 2018. We divided the study population into derivation, and internal and external validation cohorts, and used five feature selection techniques to select 10 of 720 potentially predictive variables from the electronic health records. Model performance was assessed by the area under the receiver operating characteristic curve in the validation cohorts. The primary outcome was development of AKI (per the Kidney Disease Improving Global Outcomes creatinine definition) within a moving 48-hour window. Secondary outcomes included severe AKI (stage 2 or 3), inpatient mortality, and length of stay.Results: Among 8473 encounters studied, AKI occurred in 516 (10.2%), 207 (9%), and 27 (2.5%) encounters in the derivation, and internal and external validation cohorts, respectively. The highest-performing model used a machine learning-based genetic algorithm, with an overall receiver operating characteristic curve in the internal validation cohort of 0.76 [95% confidence interval (CI), 0.72 to 0.79] for AKI, 0.79 (95% CI, 0.74 to 0.83) for severe AKI, and 0.81 (95% CI, 0.77 to 0.86) for neonatal AKI. To translate this prediction model into a clinical risk-stratification tool, we identified high- and low-risk threshold points.Conclusions: Using various machine learning algorithms, we identified and validated a time-updated prediction model of ten readily available electronic health record variables to accurately predict imminent AKI in hospitalized children.

Purpose of review: The clinical diagnosis of acute kidney injury (AKI) relies largely on changes in serum creatinine; a delayed biomarker. Research in children has been focused on developing novel AKI biomarkers, which can improve the prediction, early detection and diagnosis of kidney injury, as well as our understanding of AKI pathophysiology. In this review, we describe recently published studies on urine or blood biomarkers of AKI. The mechanistic relevance of neutrophil gelatinase-associated lipocalin (NGAL), kidney injury molecule-1, interleukin (IL)-18, liver-type fatty acid binding protein, tissue inhibitor of metalloproteinase (TIMP)-2/insulin-like growth factor-binding protein (IGFBP)-7, uromodulin, as well as other inflammatory biomarkers are discussed in the context of AKI pathophysiology, as well as their performance predicting or diagnosing AKI. Recent findings: Biomarkers of tubular injury, cell cycle arrest and inflammation are presented in this review. NGAL continues to be the most frequently studied biomarker and continues to have good performance in a variety of clinical settings, most notably after cardiopulmonary bypass. We also found promising results with less studied biomarkers for the prediction of AKI in children, including TIMP2, IGFBP7, uromodulin, tumor necrosis factor-α and IL-8. Summary: Identifying new AKI biomarkers is a priority in pediatric nephrology research because of the morbidity associated with AKI, as well as the lack of therapies for AKI. Recent research suggests that novel AKI biomarkers have the potential to predict the development of AKI and diagnose AKI earlier than changes in serum creatinine. The diverse causes of AKI, the different settings where patients develop AKI and the changing biomarker reference ranges throughout childhood remain challenges in biomarker development.

Monogenic mutations may be a significant cause of steroid-resistant nephrotic syndrome. NUP93 is a gene previously reported to cause isolated steroid-resistant nephrotic syndrome. Here we describe a case of recessive, syndromic, steroid-resistant nephrotic syndrome caused by NUP93 mutation. NUP93 may convey a phenotype that has not only SRNS, but also other syndromic features.

Background: End-stage renal disease (ESRD) is a growing global health concern, and hemodialysis (HD) remains the most common life-sustaining therapy for patients with advanced kidney failure. Both humoral and cellular immunity are impaired post hemodialysis, leading to immune system dysfunction. Methods: We utilized flow cytometry to quantify cell populations based on surface markers, including CD3 (total T lymphocytes), CD4 (helper T-cells), CD8 (cytotoxic T-cells), CD19 (B lymphocytes), and CD16/CD56 (natural killer (NK) cells). EDTA-blood samples were collected intravenously immediately before and after dialysis. Results: A consistent decline in CD3⁺ T lymphocytes was observed post hemodialysis. This reduction occurred across both male and female cohorts: p = 0.0342 and p = 0.0002, respectively. CD8⁺ cytotoxic T-cells decreased significantly post HD, p = 0.0003. Conversely, CD4⁺ helper T-cells exhibited a paradoxical increase, p = 0.0321. The divergent trends in CD4⁺ and CD8⁺ cells led to a statistically significant increase in the CD4/CD8 ratio post dialysis, p = 0.0005. Notably, stratification by gender uncovered that the post-HD changes in CD4⁺ and CD8⁺ T-cells were exclusive to female patients. Females demonstrated a pronounced increase in CD4⁺ cells and a sharper decline in CD8⁺ cells compared to males. CD19⁺ B lymphocytes showed a statistically significant decline post hemodialysis (p < 0.0001). While both genders exhibited reduced B-cell percentages, female patients experienced a more pronounced reduction than males. NK cells were severely depleted post dialysis in both male and female cohorts. Conclusions: Overall, the immune alterations observed in HD patients, including T-cell reduction, B-cell lymphopenia, and changes in NK cell populations, contribute to the increased risk of infections, malignancy, and cardiovascular disease in this population.

Background: Due to the complexity of chronic kidney disease (CKD) pathophysiology, biomarkers representing different mechanistic pathways have been targeted for the study and development of novel biomarkers. The discovery of clinically useful CKD biomarkers would allow for the identification of those children at the highest risk of kidney function decline for timely interventions and enrollment in clinical trials. Summary: Glomerular filtration rate and proteinuria are traditional biomarkers to classify and prognosticate CKD progression in clinical practice but have several limitations. Over the recent decades, novel biomarkers have been identified from blood or urine with metabolomic screening studies, proteomic screening studies, and an improved knowledge of CKD pathophysiology. This review highlights promising biomarkers associated with the progression of CKD that could potentially serve as future prognostic markers in children with CKD. Key Messages: Further studies are needed in children with CKD to validate putative biomarkers, particularly candidate proteins and metabolites, for improving clinical management.