New screening model quickly spots acute kidney injuryby Jennifer Wetzel Jun. 8, 2017, 9:15 AM
Researchers at Vanderbilt University Medical Center (VUMC) have developed a screening tool intended to more quickly identify patients with acute kidney injury (AKI).
This model, described in the journal Pediatric Research, is currently being tested at Monroe Carell Jr. Children’s Hospital at Vanderbilt to determine if patients at risk for AKI can be more quickly identified and treated.
AKI is a common issue among intensive care patients, affecting up to 25 percent of patients admitted to the Pediatric Intensive Care Unit, largely due to their acuity level and the needed medications for treatment, fever or pain management.
When AKI is detected, treatment plans can be adjusted to reduce the severity of the problem, but testing for AKI requires a blood draw and is invasive for children.
Using three years of electronic health record information, Vanderbilt researchers and biostatisticians collaborated to develop a predictive model that can identify children at higher risk for AKI even if they have not yet had a blood test to check for elevated creatinine levels, which would signify impaired kidney function or kidney disease.
When children are admitted to Children’s Hospital, half are organized into routine clinical care, while half are randomized into the intervention. Using the statistical model developed from the electronic health records data, if the child is at higher than 50 percent risk for AKI, the provider gets an alert that the patient is at risk.
“This surveillance during the inpatient stay should identify who needs the serum creatinine test earlier than without the surveillance, thus allowing for more targeted laboratory testing, earlier identification and modification of care,” said corresponding author Sara Van Driest, M.D., Ph.D., assistant professor of Pediatrics and Medicine.
“We’re hopeful this process will allow us to really pinpoint which patients need additional screening. That’s precision medicine, using every bit of information you can about a patient to decide which ones require a specific intervention to improve their care.”
This trial will continue through November, when the data will be analyzed to determine if this screening process has impacted the amount of testing done for serum creatinine and, ultimately, if there is a reduction in AKI among these patients.
If the trial demonstrates a benefit to patients, VUMC has committed to building this tool into all electronic health records, making this part of routine care at Vanderbilt.
Study authors include Vanderbilt’s Li Wang, Tracy McGregor, Deborah Jones, Brian Bridges, Geoffrey Fleming, Jana Shirey-Rice, Michael McLemore, Lixin Chen, Asli Weitkamp, Daniel Byrne and Sara Van Driest.
This study is funded by the National Center for Advancing Translational Sciences and the National Institute of Child Health and Development, part of the National Institutes of Health [Grant numbers UL1 TR000445, KL2 TR000446 and K23 HD000001].