Background

The goal of the Artificial Intelligence in Renal Scarring (AIRS) study is to develop machine learning tools for noninvasive quantification of kidney fibrosis from imaging scans.

Methods

We conducted a retrospective analysis of patients who had one or more abdominal computed tomography (CT) scans within 6 months of a kidney biopsy. The final cohort encompassed 152 CT scans from 92 patients, which included images of 300 native kidneys and 76 transplant kidneys. Two different convolutional neural networks (slice-level and voxel-level classifiers) were tested to differentiate severe versus mild/moderate kidney fibrosis (≥50% versus <50%). Interstitial fibrosis and tubular atrophy scores from kidney biopsy reports were used as ground-truth.

Results

The two machine learning models demonstrated similar positive predictive value (0.886 versus 0.935) and accuracy (0.831 versus 0.879).

Conclusions

In summary, machine learning algorithms are a promising noninvasive diagnostic tool to quantify kidney fibrosis from CT scans. The clinical utility of these prediction tools, in terms of avoiding renal biopsy and associated bleeding risks in patients with severe fibrosis, remains to be validated in prospective clinical trials.

Background

The KBindER (K⁺ Binders in Emergency Room and hospitalized patients) clinical trial is the first head-to-head evaluation of oral potassium binders (cation-exchange resins) for acute hyperkalemia therapy.

Methods

Emergency room and hospitalized patients with a blood potassium level ≥ 5.5 mEq/L are randomized to one of four study groups: potassium binder drug (sodium polystyrene sulfonate, patiromer, or sodium zirconium cyclosilicate) or nonspecific laxative (polyethylene glycol). Exclusion criteria include recent bowel surgery, ileus, diabetic ketoacidosis, or anticipated dialysis treatment within 4 h of treatment drug. Primary endpoints include change in potassium level at 2 and 4 h after treatment drug. Length of hospital stay, next-morning potassium level, gastrointestinal side effects and palatability will also be analyzed. We are aiming for a final cohort of 80 patients with complete data endpoints (20 per group) for comparative statistics including multivariate adjustment for kidney function, diabetes mellitus, congestive heart failure, metabolic acidosis, renin-angiotensin-aldosterone system inhibitor prescription, and treatment with other agents to lower potassium (insulin, albuterol, loop diuretics).

Discussion

The findings from our study will inform decision-making guidelines on the role of oral potassium binders in the treatment of acute hyperkalemia.

Trial registration

ClinicalTrials.gov Identifier: NCT04585542 . Registered 14 October 2020.