New Hyperkalemia Risk Model for CKD and Diabetes

Medscape

2 days ago

A new risk model may help identify which patients with chronic kidney disease (CKD) and diabetes are more likely to develop hyperkalemia, granting physicians more confidence in prescribing medications like finerenone. However, researchers caution that further validation is needed.
The model is based on pooled data from two phase 3 trials of finerenone: FIDELIO-DKD (Finerenone in Reducing Kidney Failure and Disease Progression in Diabetic Kidney Disease) and FIGARO-DKD (Finerenone in Reducing Cardiovascular Mortality and Morbidity in Diabetic Kidney Disease). The combined dataset, known as FIDELITY, includes 6355 patients in the placebo groups.
Researchers used this dataset to develop the risk score, identifying seven factors independently associated with the primary outcome of new onset of hyperkalemia (incident serum potassium level > 5.5 mmol/L):
Serum potassium > 4.5 mmol/L
Prior history of hyperkalemia
No use of sodium-glucose co-transporter 2 inhibitor
Urine albumin-to-creatinine ratio > 1000 mg/g
Hemoglobin < 12 g/dL
No use of thiazide-type diuretics
Estimated glomerular filtration rate < 45 mL/min/1.73 m2
The model was subsequently validated using data from the finerenone groups in FIDELITY. In their paper, published online in the European Heart Journal , João Pedro Ferreira, MD, PhD, of the Cardiovascular Research and Development Center at the University of Porto, Porto, Portugal, and coauthors noted that the model could be used to reduce the risk for hyperkalemia among high-risk patients receiving finerenone.
'This could include tailoring of individualized treatment and follow-up strategies (eg, frequency of visits and serum potassium assessments, and use of potassium binders), thus optimizing patient management and potentially improving outcomes,' they noted.
Welcome Results, Validation Required
CKD, which affects more than 800 million people worldwide, is a common complication of diabetes. An estimated 15%-40% of people with CKD in the setting of diabetes may have hyperkalemia, noted Bernard G. Jaar, MD, MPH, clinical director for Nephrology at Johns Hopkins School of Medicine, Baltimore, who was not involved in the study.
In an email exchange with Medscape Medical News , Jaar described the study as 'welcome and much needed.' He noted that hyperkalemia poses a challenge for clinicians, as many medications required for optimal care in CKD — including renin-angiotensin-aldosterone system (RAAS) blockade agents and finerenone — can raise serum potassium.
'This is a true barrier to care,' Jaar wrote. 'Clinicians are worried about the potential complications associated with hyperkalemia, such as arrhythmias and even death.'
Jaar noted that the FIDELITY analysis has several strengths, including its large sample size of patients with varying stages of CKD and degrees of proteinuria, and a risk score model that incorporated readily available clinical variables. He added that it reports a stepwise increase in hyperkalemia risk across risk score tertiles in both placebo and finerenone groups, even though all patients were already on a maximally tolerated RAAS blockade agent.
However, he questioned the study's generalizability, given that it excluded patients if their serum potassium was > 4.8 mmol/L.
'This risk model in predicting hyperkalemia needs to be validated in other populations where diet may be different,' Jaar wrote. 'Also, this needs to be validated in at-risk populations during real-life experience and not only in patients enrolled in clinical trials, who are typically highly motivated and selected.'
Future Applications
Rajiv Agarwal, MD, MS, professor emeritus of medicine at Indiana University and a lead researcher in the finerenone clinical studies, echoed the need for further validation of this tool. Although not involved with the development of the risk model, Agarwal was an author on key publications from the FIGARO-DKD, FIDELIO-DKD, and FIDELITY studies.
Agarwal suggested that this model's approach could be applied to other therapies, including aldosterone synthase inhibitors in a similar population with CKD. He also noted the importance of incorporating validated risk models for hyperkalemia into electronic medical records or apps, particularly in countries with limited access to care, like China and India. This would allow more nuanced approaches for following up patients taking drugs with established hyperkalemia risk.
'If the patient cannot come after a month, can I make a clinical judgment and say, 'Okay, if you came back after 3 months, your risk of hyperkalemia is low enough that I'm willing to take that risk,'' he said. 'Or if the risk of hyperkalemia is high, you can tell the patient who can't come back in 30 days, 'I don't think I want to prescribe this drug for you'.'
The work on the risk model was supported by Bayer, which also funded the FIDELIO-DKD and FIGARO-DKD studies and pooled analysis.
The study's authors included Bayer employees. Other financial relationships of the authors included research support, consulting and other fees from companies including:
Abbott Vascular, Amgen, Astellas, AstraZeneca, Bayer, BioVentrix, Boehringer Ingelheim, Brahms, Brainstorm Medical, Cardiac Dimensions, Cardior, Cereno Scientific, CSL Vifor, CVRx, Edwards, Eli Lilly, G3 Pharmaceuticals, Gilead, GSK, Impulse Dynamics, Janssen, KBP Biosciences, Mundipharma, Novartis, Novo Nordisk, Occlutech, PhaseBio, Proton Intel, Respicardia, Sanofi, Sarfez, scPharmaceuticals, Servier, SQ Innovation, Tricida, Vectorious, Vifor International, and V-Wave.
Agarwal had received support from Bayer. He also had received consulting fees and other support from Boehringer Ingelheim, Novartis, Akebia, Intercept Pharma, Alnylam, and Vertex.
Jaar reported no relevant financial disclosures.