Researchers develop all-cause mortality risk prediction model for adults with advanced CKD
Click Here to Manage Email Alerts
Key takeaways:
- Using common variables, the model can identify patients at a higher risk for mortality, kidney failure and cardiovascular events.
- External validation in a separate dataset confirmed the tool’s robustness.
PHILADELPHIA — Researchers presented data at ASN Kidney Week from a prediction model that can forecast all-cause mortality in adults with advanced chronic kidney disease.
“We were able to develop and externally validate a risk prediction model that predicted all-cause mortality in patients with advanced CKD,” Ranveer Singh Brar, of Seven Oaks General Hospital in Canada, told Healio. “Hopefully, our model will be able to aid in shared clinical decision making when making choices on aggressive therapies (ie, dialysis, transplantation) when patients with advanced CKD reach end-stage renal disease.”
Using common variables, the model can identify patients at a higher risk for mortality, kidney failure and cardiovascular events. Image: Adobe Stock.
By considering commonly measured variables, the tool aims to identify patients at a higher risk for mortality, kidney failure and cardiovascular events, according to Brar.
He joined researchers at the Chronic Disease Innovation Centre to study adult patients from Manitoba, Canada, with an eGFR below 30 mL/min/1.73m2 from 2012 to September 2020 showing no history of dialysis or transplant. Investigators identified patients who had at least two eGFRs below 30 at least 90 days apart. The development cohort included 397 patients.
Researchers built the prediction model by leveraging demographic, clinical and lab data from the patient group, with a primary outcome of time to all-cause mortality. If dialysis was initiated in follow-up, all-cause mortality within 1 year of initiation was ascertained. Assessments were made at 2 and 5 years using the area under the receiver characteristic operating curve. The final prediction model encompassed variables such as age, sex, eGFR, hemoglobin, serum albumin and congestive heart failure.
According to the findings, the model achieved a 2- and 5-year area under the curve of 74.3 and 80.2, respectively. Brar told Healio that external validation in a separate dataset from Ontario, Canada, confirmed the tool’s robustness with 2-year and 5-year area under the receiver characteristic operating curve scores of 71.4 and 73.
“This equation may aid as a support tool for nephrologists in dialysis decision making, especially in patients who are at high risk [for] mortality,” Brar and colleagues wrote in the study abstract. “Decision-making may be improved by weighing the risk [for] mortality against aggressive dialysis therapies, thus allocating the appropriate resources for care.”
Collapse
Brar RS, et al. FR-PO939 Presented at: American Society of Nephrology Kidney Week; Nov. 2-5, 2023; Philadelphia.
Click Here to Manage Email Alerts