Understanding estimated glomerular filtration rate: Implications for identifying chronic kidney disease

Research output: Contribution to journalReview articlepeer-review

51 Scopus citations

Abstract

PURPOSE OF REVIEW: Glomerular filtration rate (GFR) can be estimated using serum markers such as serum creatinine (SCr) or cystatin C. This review presents new insights into estimated GFR based on theory, validation studies, SCr assay standardization, cystatin C, and longitudinal comparison with measured GFR. RECENT FINDINGS: The estimation of GFR by SCr differs in health and in chronic kidney disease (CKD) due to differences in GFR range and in creatinine production between these two populations. Among populations with normal baseline GFR, there is a more rapid decline in measured GFR than in SCr-based estimated GFR. While elevated SCr is specific for CKD, other disease processes may lead to elevated cystatin C. Validation is improved by refitting equation coefficients to compare populations, recognizing the asymmetry between estimated GFR and measured GFR, and using residual plots instead of Bland-Altman plots to assess bias. SUMMARY: As a screening test, SCr should be interpreted as a marker of CKD probability in the context of the patient's clinical presentation. Measured GFR or creatinine clearance may be helpful in high-risk patients with normal SCr levels. GFR estimating equations should be reserved for patients with identified CKD. Standardized SCr and cystatin C assays are needed.

Original languageEnglish (US)
Pages (from-to)242-249
Number of pages8
JournalCurrent opinion in nephrology and hypertension
Volume16
Issue number3
DOIs
StatePublished - May 2007

Keywords

  • Cystatin C
  • Estimating equations
  • Glomerular filtration rate
  • Serum creatinine

ASJC Scopus subject areas

  • Internal Medicine
  • Nephrology

Fingerprint

Dive into the research topics of 'Understanding estimated glomerular filtration rate: Implications for identifying chronic kidney disease'. Together they form a unique fingerprint.

Cite this