Which equations give best estimates of renal function in older patients?

Accurate estimation of glomerular filtration rates (GFR) in older patients is important to identify renal impairment and guide drug dosing. An analysis of four equations used in estimating GFR in patients aged 65 years and older showed that all four had similar diagnostic performances. These results, published in JAMA Internal Medicine, suggest that any of them can be used in older patients to estimate GFR.

Assessing renal function by measuring inulin or iohexol clearance is the most accurate method to estimate GFR, but this is rarely done in clinical practice because of inconvenience and cost. GFR-estimating equations that are plasma creatinine–based are used as an alternative, but there have been concerns about their accuracy.

“There is no perfect approach to estimating renal function in any patient, but it is particularly difficult to estimate renal function in [older adult] patients due to the wide heterogeneity seen within this population,” said Laura H. Waite, PharmD, BCPS, CLS, BC-ADM, adult internal medicine clinical pharmacy specialist at the Hospital of the University of Pennsylvania.

Waite added that when using any equation to estimate renal function, pharmacists must interpret the result using the context of each individual patient, “since the result is only an estimate based on population parameters and may not be accurate based on patient-specific factors.”

A retrospective cross-sectional analysis of 2,247 patients aged 65 to 90 years (mean age 71.5 y) who underwent a GFR measurement via a urinary inulin (reference method) was compared with results from four plasma creatinine-based GFR-estimating equations—the Chronic Kidney Disease–Epidemiology Collaboration (CKD-EPI), Lund-Malmö Revised (LMR), full age spectrum (FAS), and Berlin Initiative Study (BIS).

The analysis showed that despite small statistical differences, the performance of the CKD-EPI equation was not clinically different from that of the LMR, FAS, or BIS. Close alignment was observed for the four equations in terms of bias (difference between estimated and measured GFR) and precision (interquartile range of the median difference).

For the full article, please visit www.pharmacytoday.org for the July 2019 issue of Pharmacy Today.