Ix, Joachim H.

ABSTRACT Artificial intelligence (AI), encompassing machine learning, deep learning and generative AI, is poised to redefine nephrology by enabling earlier detection, more precise risk stratification and workflow-integrated clinical decision support across the spectrum of kidney disease. This state-of-the-art review synthesizes emerging applications of AI in acute kidney injury (AKI), chronic kidney disease (CKD), dialysis and kidney transplantation, with attention to clinical integration, real-world validation, workflow implementation and translational challenges. In AKI, predictive models trained on high-frequency electronic health record data and intensive care unit telemetry have demonstrated strong performance in forecasting critical events, yet translation into routine clinical workflows remains limited. In CKD, machine learning tools support progression risk stratification and phenotype clustering, with the potential to inform individualized surveillance and therapy. AI-enabled dialysis management systems optimize ultrafiltration, anemia control and vascular access surveillance, while generative AI and large language models offer novel capabilities for clinical documentation, triage and patient education. In transplantation, AI applications span organ allocation, dynamic graft monitoring and digital pathology-assisted rejection classification, with validated tools such as the iBox system gaining regulatory recognition. Implementation challenges include data heterogeneity, bias, interpretability, regulatory uncertainty and workflow integration. Looking ahead, multimodal integration of imaging, pathology and multi-omics data may support biologically informed precision nephrology. Reinforcement learning, digital twins and ambient intelligence are emerging as adaptive decision-support paradigms rather than autonomous care systems. Regulatory frameworks are evolving to accommodate adaptive algorithms, underscoring the need for clinician engagement in model development, validation and deployment. As AI matures from pilot to practice, nephrologists who embrace and help shape these tools will lead the transition toward a more personalized, efficient and equitable future for kidney care.

Several loci have been discovered and replicated for urine albumin-creatinine ratio (ACR), a measure of kidney function, in European populations; however, other ethnic populations such as African-Americans have been less studied for this trait. We examined the association of genetic variants on the Metabochip with natural log transformed ACR (ln(ACR)) in 4,629 African-Americans from the Population Architecture using Genomics and Epidemiology (PAGE) Study which includes participants from the Atherosclerosis Risk in Communities (ARIC), Cardiovascular Health Study (CHS), Coronary Artery Risk Development in Young Adults (CARDIA), and Epidemiologic Architecture for Genes Linked to the Environment (EAGLE-BioVU) studies. We then replicated our findings in 1,533 African-American participants from the Multi-Ethnic Study of Atherosclerosis (MESA). Linear regression models were used to examine associations, assuming additive effects of the SNPs. Within each study, models were adjusted for age, field center (where applicable), and principal components of ancestry, and stratified by sex. Fixed effects inverse variance weighted meta-analysis was used to combine results from contributing studies. Array-wide statistical significance was concluded for SNP associations of p-value less than 2.8 x 10-7. Mean ln(ACR) ranged from 1.4 to 2.9 mg/g, average age ranged from 44 to 77 years, and percent female participants ranged from 59-65% in the contributing studies. Four SNPs were significantly associated array-wide with ACR in the discovery sample: rs17399841 in GRHL2 (β=-0.67, MAF=0.01, p=2.7x10-9), rs153914 near SPATA9/RHOBTB3 (β=-0.59, MAF=0.01, p=1.0x10-7), rs17439845 in LEF1 (β=-0.49, MAF=0.02, p=1.3x10-7), and rs11836243 near ZNF641/ANP32D (β=0.23, MAF=0.12, p=1.8x10-7). Of these four loci, only the association of rs17439845 with ACR replicated in MESA (β=-0.36, p=0.038), with a combined discovery-replication beta of -0.46 and p value of 1.8x10-8. Combined discovery-replication for rs17399841, rs153914, and rs11836243 no longer achieved array-wide significance (p-values 9.84x10-7, 1.98x10-5, and 1.32x10-4, respectively). LEF1 is a mediator in the Wnt signaling pathway, and has been previously implicated in systemic lupus erythematosus and osteoporosis, as well as previously associated with waist circumference, echocardiographic traits, bone mineral density, and blood count traits. Continuing studies in diverse populations are warranted to identify a set of loci important in contributing to the variation in ACR.

Periprocedural bleeding is one of the most frequent complications of percutaneours coronary interventions. We assessed the relation between blood transfusion and all-cause mortality or incident cardiovascular events (death, MI, stroke) among 6103 patients of the Evaluation of Oral Xemilofiban in Controlling Thrombotic Events (EXCITE)-trial. Subjects were followed for 7 months after enrollment for the occurrence of events. Multivariate Cox-regression analysis evaluated the independent association of blood transfusion with each outcome adjusted for age, gender, race, diabetes mellitus, hypertension, hypercholesterolemia, history of MI, PCI, CABG, heart failure, LVEF<30%, use of beta-blockers, statins, ACE-inhibitors, platelet inhibitors and allocation to treatment with xemolifiban. In addition, propensity score analyses were performed (ROC 0.80). Mean age was 59.2 years, 21.7% were female, and 18.9% had diabetes mellitus. Of the169 patients who received blood transfusion, 14 (8.3%) died and 42 (24.9%) experienced a CVD event. Of the 5934 patients without transfusion, 65 (1.1%) died (p-value: <0.001) and 555 (9,4%) experienced a CVD event (p-value: <0.001) In multivariate analysis, blood transfusion was associated with a 5.3 fold increased risk of mortality (HR 5.3; 95% CI 2.8 –10.2), and a 2.5 fold increased risk of incident CVD (HR 2.5; 95% CI 1.7–3.4.) Noteworthy, patients who were US citizens had a higher transfusion rate then non-US citizens (OR 1.45, 95%CI 1.02–2.06) The need of blood transfusion is a strong and independent predictor of all-cause mortality and incident CVD events among patients undergoing PCI.

Background and objectives eGFR and albuminuria primarily reflect glomerular function and injury, whereas tubule cell atrophy and interstitial fibrosis on kidney biopsy are important risk markers for CKD progression. Kidney tubule injury markers have primarily been studied in hospitalized AKI. Here, we examined the association between urinary kidney tubule injury markers at baseline with subsequent loss of kidney function in persons with nondiabetic CKD who participated in the Systolic Blood Pressure Intervention Trial (SPRINT). Design, setting, participants, & measurements Among 2428 SPRINT participants with CKD (eGFR<60 ml/min per 1.73 m 2 ) at baseline, we measured urine markers of tubule injury (IL-18, kidney injury molecule-1 [KIM-1], neutrophil gelatinase-associated lipocalin [NGAL]), inflammation (monocyte chemoattractant protein-1 [MCP-1]), and repair (human cartilage glycoprotein-40 [YKL-40]). Cox proportional hazards models evaluated associations of these markers with the kidney composite outcome of 50% eGFR decline or ESKD requiring dialysis or kidney transplantation, and linear mixed models evaluated annualized change in eGFR. Results Mean participant age was 73±9 (SD) years, 60% were men, 66% were white, and mean baseline eGFR was 46±11 ml/min per 1.73 m 2 . There were 87 kidney composite outcome events during a median follow-up of 3.8 years. Relative to the respective lowest quartiles, the highest quartiles of urinary KIM-1 (hazard ratio, 2.84; 95% confidence interval [95% CI], 1.31 to 6.17), MCP-1 (hazard ratio, 2.43; 95% CI, 1.13 to 5.23), and YKL-40 (hazard ratio, 1.95; 95% CI, 1.08 to 3.51) were associated with higher risk of the kidney composite outcome in fully adjusted models including baseline eGFR and urine albumin. In linear analysis, urinary IL-18 was the only marker associated with eGFR decline (−0.91 ml/min per 1.73 m 2 per year for highest versus lowest quartile; 95% CI, −1.44 to −0.38), a finding that was stronger in the standard arm of SPRINT. Conclusions Urine markers of tubule cell injury provide information about risk of subsequent loss of kidney function, beyond the eGFR and urine albumin.

The American Geriatrics Society convened a conference in Bethesda, Maryland, to explore models and studies of aging. This was the second of three conferences, supported by a U13 grant from the National Institute on Aging, to aid recipients of Grants for Early Medical/Surgical Specialists Transition to Aging Research (GEMSSTAR) in integrating geriatrics into their specialties. Recognizing that aging is the largest risk factor for multiple chronic diseases and age‐related loss of resilience, the conference organizers focused scientific sessions on how targeting age‐related mechanisms can delay, prevent, or reverse geriatric syndromes, age‐related chronic diseases, and loss of resilience. The rationale for studying models of aging as well as study designs, strategies, and challenges of studying human aging were reviewed. This article provides a summary of the full conference report, Models and Studies of Aging: Report from the U13 Conference Series, and summarizes key take‐home messages that were designed to support GEMSSTAR awardees in developing their research careers focused on aging research (see supplementary text for the full report). J Am Geriatr Soc 67:428–433, 2019.

The appropriate target for BP in patients with CKD and hypertension remains uncertain. We report prespecified subgroup analyses of outcomes in participants with baseline CKD in the Systolic Blood Pressure Intervention Trial. We randomly assigned participants to a systolic BP target of <120 mm Hg (intensive group; n =1330) or <140 mm Hg (standard group; n =1316). After a median follow-up of 3.3 years, the primary composite cardiovascular outcome occurred in 112 intensive group and 131 standard group CKD participants (hazard ratio [HR], 0.81; 95% confidence interval [95% CI], 0.63 to 1.05). The intensive group also had a lower rate of all-cause death (HR, 0.72; 95% CI, 0.53 to 0.99). Treatment effects did not differ between participants with and without CKD ( P values for interactions ≥0.30). The prespecified main kidney outcome, defined as the composite of ≥50% decrease in eGFR from baseline or ESRD, occurred in 15 intensive group and 16 standard group participants (HR, 0.90; 95% CI, 0.44 to 1.83). After the initial 6 months, the intensive group had a slightly higher rate of change in eGFR (−0.47 versus −0.32 ml/min per 1.73 m 2 per year; P <0.03). The overall rate of serious adverse events did not differ between treatment groups, although some specific adverse events occurred more often in the intensive group. Thus, among patients with CKD and hypertension without diabetes, targeting an SBP<120 mm Hg compared with <140 mm Hg reduced rates of major cardiovascular events and all-cause death without evidence of effect modifications by CKD or deleterious effect on the main kidney outcome.

Objective— To evaluate the predictive value of abdominal aortic calcium (AAC) for incident cardiovascular disease (CVD) independent of coronary artery calcium (CAC). Approach and Results— We evaluated the association of AAC with CVD in 1974 men and women aged 45 to 84 years randomly selected from the Multi-Ethnic Study of Atherosclerosis participants who had complete AAC and CAC data from computed tomographic scans. AAC and CAC were each divided into following 3 percentile categories: 0 to 50th, 51st to 75th, and 76th to 100th. During a mean of 5.5 years of follow-up, there were 50 hard coronary heart disease events, 83 hard CVD events, 30 fatal CVD events, and 105 total deaths. In multivariable-adjusted Cox models including both AAC and CAC, comparing the fourth quartile with the ≤50th percentile, AAC and CAC were each significantly and independently predictive of hard coronary heart disease and hard CVD, with hazard ratios ranging from 2.4 to 4.4. For CVD mortality, the hazard ratio was highly significant for the fourth quartile of AAC, 5.9 ( P =0.01), whereas the association for the fourth quartile of CAC (hazard ratio, 2.1) was not significant. For total mortality, the fourth quartile hazard ratio for AAC was 2.7 ( P =0.001), and for CAC, it was 1.9, P =0.04. Area under the receiver operating characteristic curve analyses showed improvement for both AAC and CAC separately, although improvement was greater with CAC for hard coronary heart disease and hard CVD, and greater with AAC for CVD mortality and total mortality. Sensitivity analyses defining AAC and CAC as continuous variables mirrored these results. Conclusions— AAC and CAC predicted hard coronary heart disease and hard CVD events independent of one another. Only AAC was independently related to CVD mortality, and AAC showed a stronger association than CAC with total mortality.

Abstract We tested whether the association between bone mineral density (BMD) and coronary artery calcification (CAC) varies according to dyslipidemia in community-living individuals. Between 2002 and 2005, 305 women and 631 men (mean age of 64 years), who were not taking lipid-lowering medications or estrogen were assessed for spine BMD, CAC, and total (TC), HDL- and LDL-cholesterol and triglycerides. Participants were a random sample from the Multi-Ethnic Study of Atherosclerosis (MESA) without clinical cardiovascular disease. Spine BMD at the L3 vertebrate was performed by computer tomography (CT). CAC prevalence was measured by CT. The total cholesterol to HDL ratio (TC:HDL) ≥ 5.0 was used as the primary marker of hyperlipidemia. The association of BMD with CAC differed in women with TC:HDL &lt; 5.0 versus higher (p-interaction = 0.01). In age- and race-adjusted models, among women with TC:HDL &lt; 5.0, each SD (43.4 mg/cc) greater BMD was associated with a 25% lower prevalence of CAC (prevalence ratio [PR] 0.75, 95% confidence interval [CI] 0.63–0.89), whereas among women with higher TC:HDL, higher BMD was not significantly associated with CAC (PR 1.22, 95% CI 0.82–1.82). Results were similar using other definitions of hyperlipidemia. In contrast, no consistent association was observed between BMD and CAC in men, irrespective of the TC:HDL ratio (p interaction 0.54). The inverse association of BMD with CAC is stronger in women without dyslipidemia. These data argue against the hypothesis that dyslipidemia is the key factor responsible for the inverse association of BMD with atherosclerosis. © 2011 American Society for Bone and Mineral Research