Kashani, Kianoush

Measured glomerular filtration rate (mGFR) remains the reference standard for precise assessment of kidney function when estimated approaches are insufficient. This second part of a two-part series reviews exogenous filtration markers used to obtain mGFR, including inulin, iothalamate, iohexol, and radionuclide-based tracers, with emphasis on their physiologic handling, analytic methods, accuracy, and clinical applicability. The review further examines emerging fluorescent and near-infrared tracers coupled with transcutaneous monitoring technologies that enable rapid or continuous assessment of GFR. Together, these approaches illustrate the evolution of mGFR measurement toward methods that are more practical, scalable, and responsive to dynamic changes in kidney function, while maintaining analytical accuracy.

Abstract Background Cardiac dysfunction is a common complication in septic population. Though traditionally considered an acute and reversible phenomenon, the true prevalence of reversible cardiac dysfunction and its long-term prognostic implications remain unclear. Purpose Using serial echocardiographic assessments, this study aimed to investigate the prevalence and reversibility of cardiac dysfunction in septic patients, and to determine whether the recovery of cardiac function is associated with improved 5-year mortality. Methods We retrospectively screened 16,041 septic patients who underwent transthoracic echocardiography during their ICU stay. After excluding patients with a history of heart failure, pre-existing cardiac dysfunction, or lacking follow-up echocardiographic assessments post-ICU discharge, 3,112 patients were enrolled for analysis. To assess the reversibility of cardiac dysfunction, paired-sample t-tests were used to evaluate changes in echocardiographic parameters between in-ICU and post-ICU measurements. Furthermore, 5-year mortality were compared among controls and patients with or without recovery of echocardiographic abnormalities. Results Totally 1,604 patients (51.5%) were found to have newly recognized cardiac dysfunction during ICU stay. The mean interval between ICU admission and follow-up echocardiography was 425 days. Most in-ICU echocardiographic abnormalities demonstrated partial recovery over time, with the exception of left atrial volume index (LAVi) and right ventricular fractional area change (RV FAC) (Figure 1). Recovery of left ventricular ejection fraction (LVEF), average E/e', and tricuspid regurgitation (TR) velocity was significantly associated with reduced 5-year mortality (Figure 2). Older age [adjusted hazard ratio (aHR) 0.99, p=0.034], male gender (aHR 0.54, p=0.003), and elevated troponins (aHR 0.87, p=0.006) were negative predictors for LVEF recovery; while older age (aHR 0.099, p=0.043), higher APACHE III scores (aHR 0.99, p=0.018), atrial fibrillation (aHR 0.59, p=0.046), and chronic pulmonary disease (aHR 0.68, p=0.042) were negative predictors for TR velocity recovery. Conclusions Both the presence and recovery of newly recognized echocardiographic abnormalities during ICU stay were significantly associated with long-term mortality. Certain abnormalities, such as increased LAVi, persisted despite resolution of sepsis.Figure 1 Figure 2

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.

Background: Adequate medication dosing in the intensive care unit (ICU) remains challenging due to rapid physiological variability, narrow therapeutic windows, and evolving standards of care. Artificial intelligence (AI) offers promise for personalized dosing, but its accuracy and clinical utility remain uncertain. Methods: We conducted a systematic scoping review (2011-2025) of PubMed, Embase, and Cochrane, following the Joanna Briggs Institute (JBI) and preferred reporting Items for systematic reviews and meta-analyses extension for scoping reviews (PRISMA-ScR) guidelines (open science framework [OSF]-registered). Eligible studies applied AI methods ( e.g. , machine learning [ML], deep learning [DL], reinforcement learning [RL], and natural language processing [NLP]) to ICU medication dosing or monitoring. Results: Data extraction included study design, drug class, model inputs and outputs, algorithms, validation approach, and reported outcomes. Forty-seven studies met the inclusion criteria, which were mostly retrospective and focused on adult ICU populations over the past five years. Antibiotics and anticoagulants were the most studied classes, with vancomycin being the most common. Models primarily used labs, vitals, and demographics, with limited incorporation of time series, pharmacokinetic data, or clinical notes. Outputs spanned dose prediction, serum concentration, risk stratification, adverse events, and decision support. Algorithms evolved from classical ML and ensembles to RL and long short-term memory (LSTM)-based models, although a few advanced beyond retrospective validation. Discussion: AI systems showed strong predictive performance and some clinical benefits ( e.g. , heparin target attainment, sedative reduction, and adverse event detection), but the field remains fragmented and clinically immature. Newer architectures and systematic explainability are lacking, and generalizability is uncertain. AI-enabled dosing holds promise for precision pharmacotherapy in critical illness but requires prospective multicenter validation, inclusion of population pharmacokinetic (PK) parameters, diverse cohorts, explainable design, and clinical integration.

Background There is limited evidence on the epidemiology and prognostic significance of acid-base disorders in the cardiovascular intensive care unit (CICU). This study examines the association of acid-base status at admission with in-hospital mortality among CICU patients. Methods We conducted a retrospective analysis of adults admitted to the Mayo Clinic CICU from 2007-2018 with available blood gas data, utilizing values obtained closest to CICU admission. Arterial pH, serum bicarbonate, base excess, and partial pressure of carbon dioxide (PaCO 2 ) were examined as predictors of in-hospital mortality. Multivariable logistic regression was used to assess associations, with adjustment for demographics, comorbidities, illness severity, and interventions. Results Among 3229 patients included for analysis, acidemia (pH < 7.35) emerged as the strongest predictor of in-hospital mortality (adjusted odds ratio [aOR] 1.60, 95% confidence interval [CI] 1.29-1.98, P < .003). Metabolic acidosis (HCO3 < 20 mEq/L, aOR 1.55, 95% CI 1.24-1.95, P < .001) and respiratory acidosis (PaCO2 > 45 mm Hg, aOR 1.44, 95% CI 1.14-1.81, P = .002) were associated with higher in-hospital mortality, whereas metabolic and respiratory alkalosis were not. After adjustment, lower pH and more negative base excess were associated with higher in-hospital mortality (both P < .001), whereas HCO3 and PaCO2 were not ( P = .053 and P = .051, respectively). Patients with combined metabolic and respiratory acidosis had the highest in-hospital mortality (56.3%). Conclusions Short-term survival in CICU patients decreases progressively with worse acidemia, especially in the context of combined metabolic and respiratory acidosis. Incorporating metabolic acid-base disorders as key therapeutic targets in randomized cardiogenic shock trials may improve outcomes in this complex population by addressing hemometabolic shock.

Purpose of review The diuretic agents are commonly used in ICUs for various purposes. Despite the widespread utilization of diuretics, there remains substantial ambiguity regarding their indications, ICU-related changes in their pharmacodynamics and pharmacokinetics, dosing, form, and timing of administration, the differences in their utilities based on each organ failure, and their monitoring and safety issues. Recent findings In the recent past, there have been several clinical trials and large registries or systematic reviews with a focus on the use of diuretic agents in ICUs. Summary In this review article, we outline the essential changes in drug behavior during critical illnesses and organ failures, describe the indications of their use in ICUs (management of fluid overload, electrolyte imbalances, or diagnostic or prognostic tests), assess the impact of different organ failures on the utility and effectiveness of diuretics, review major recent clinical trials related to diuretic comparison, use of multiple classes of diuretics, and monitoring and safety of their use. We also provide some information regarding the safety and adverse effects of diuretic use in the ICU and outline the importance of individualizing their use during critical illnesses.

Background: Paroxysmal atrial fibrillation (AF) is a common arrhythmogenic complication in the intensive care unit (ICU) associated with significant costs and morbidity. AF detection models in the literature have utilized entropic analyses of telemetry to diagnose AF, but they have not been utilized to forecast its onset. Hypothesis: Deep learning techniques applied to high-frequency telemetry data can be utilized to forecast AF with clinically significant lead time. Methods: Telemetry data were collected from 72 ICU monitored beds in a tertiary care center from 4/2018-12/2023. Coefficient of sample entropy (COSEn), a metric of heart rate variability complexity, was calculated on successive 1-minute windows and utilized to identify areas of AF initiation as well as control windows without AF. ECG lead II (256Hz) and beat-to-beat (B2B) measurements were analyzed by the MOMENT time-series foundational model to generate representation embeddings. A binary classification multilayer perceptron (MLP) was trained on these embeddings to forecast AF onset 5m in advance. All results were reported on separate test sets. Group-based trajectory modelling (GBTM) was performed on risk score trajectories from the 20m before AF onset or absence. Results: Over 230,000 hours of B2B data from 2263 patients was analyzed using COSEn to identify 287 transitions to AF from 136 patients. The same number of AF-absent windows were randomly selected. COSEn alone as a forecaster produced an area under the receiver operating characteristic curve (AUC) of 0.65 (95% CI: 0.61-0.69). The MOMENT + MLP model (MMM) achieved an AUC of 0.81 (0.78-0.83) (Fig. 1). COSEn and MMM were subsequently applied to the two hours preceding AF onset or absence. For COSEn, the mean risk trajectory dipped in the baseline period whereas the MMM risk trajectory sustained a high level through the baseline, highlighting its capacity to forecast AF (Fig. 2). For AF patients, GBMT yielded two types of trajectories: one with elevated risk throughout, and another with a rising trend 5 minutes prior to AF onset (Fig. 3). Conclusions: Deep learning techniques can forecast the onset of AF with up to 5 minutes of lead-time and GBTM allowed for the identification of different pre-AF risk trajectory phenotypes.The ability to forecast AF onset has potential to clinically prevent poor outcomes in the ICU, though further refinements to augment the amount of lead time are warranted.