Estimating GFR prior to contrast medium examinations-what the radiologist needs to know!
Author
Summary, in English
Creatinine-based equations to estimate glomerular filtration rate (GFR) are increasingly used in radiological practice and in studies on contrast medium-induced acute kidney injury (CIAKI). Their use is recommended in guidelines and contrast medium textbooks to identify patients at risk of CIAKI or nephrogenic systemic fibrosis. There is also an increased interest in cystatin C-based equations. Adopting GFR equations requires local creatinine and cystatin C assay calibrations to equal those used in developing the equations to avoid overestimation or underestimation of renal function. Methods should preferably be traceable to international standards, and assay traceability should be defined in CIAKI studies. Absolute GFR (mL/min) should be used when dosing contrast media and relating the dose to CIAKI instead of commonly used relative GFR (mL/min/1.73 m(2)) estimates. Accuracy of creatinine and cystatin C equations (percentage of GFR estimates within 30 % of measured GFR) ranges between 75 % and 85 %. Equations combining creatinine and cystatin C may reach 90 %, an accuracy similar to clearance methods (used as a reference test when developing and validating equations) when compared to the gold standard, renal clearance of inulin. The local laboratory or nephrology experts should be consulted in matters of method calibration and choice of GFR equation. Key Points • Traceability of creatinine/cystatin C assays used in GFR equations must be defined. • Absolute, not relative, GFR should be used when dosing contrast media. • Consult the local laboratory or nephrologist to choose the proper GFR equation.
Department/s
Publishing year
2016
Language
English
Pages
425-435
Publication/Series
European Radiology
Volume
26
Issue
2
Links
Document type
Journal article
Publisher
Springer
Topic
- Radiology, Nuclear Medicine and Medical Imaging
Status
Published
Research group
- Radiology Diagnostics, Malmö
- Internal Medicine - Epidemiology
ISBN/ISSN/Other
- ISSN: 0938-7994