Inaccuracy of GFR predictions by plasma cystatin C in patients without kidney dysfunction and in advanced kidney disease

Omran Bakoush, A. Grubb, B. Rippe

Research output: Contribution to journalArticlepeer-review

9 Citations (Scopus)


Background: In clinical practice there is need for a simple and reliable test for determination of impaired renal function. With reductions in GFR, the plasma cystatin C concentration (C, mg/l) will increase earlier than serum creatinine, and it is generally agreed that plasma cystatin C is only little affected by body weight, age or sex. However, some reports indicate that cystatin C may be influenced not only by GFR, but also by malignancy, inflammation and high doses of corticosteroids. The aim of the present study was to investigate how plasma cystatin C predicts GFR in distinct subcategories of patients with various disorders as well as in organ transplant patients. Methods: Plasma cystatin C was measured in 536 patients (age range 0.3 - 96 years, 262 females, 274 males), consecutively referred to our hospital for determination of GFR by iohexol clearance. Correlations of log GFR vs. log cystatin C were used to compare plasma cystatin C and measured GFR for the following categories: individuals with no known kidney disease (No-KD), malignant patients with (mostly) normal GFR, solid organ-transplanted patients, and patients with native chronic kidney disease (CKD). Results: In patients with normal kidney function and cystatin C level ≤ 1 mg/l, the cystatin C was poorly correlated with GFR (R2 = 0.13). By contrast, in patients with chronic kidney disease (log) plasma cystatin C was highly correlated with (log) GFR (R2 = 0.87). This correlation was more or less unchanged whether the cause of the reduction in GFR was CKD at Stages 1-3 (90 > GFR > 30 ml/min-1 (1.73 m2)-1) or solid organ transplantation (GFR = 84.55 C1.7666 and GFR = 83.95C-1.5968, respectively). Conclusion: Therefore, for these categories, a common equation for all patients with increased cystatin C, irrespective of cause of renal impairment, could be used, namely that presented by Grubb et al. [2005] (GFR = 83.93 C-1.676). However, at marked reductions of renal function (GFR < 30 or cystatin C > 2), i.e. for CKD Stages 4 and 5, the Grubb prediction equation is less accurate. Based on our data, we suggest the equation GFR = 50.52 C-1.26 for this category of patients.

Original languageEnglish
Pages (from-to)331-338
Number of pages8
JournalClinical Nephrology
Issue number5
Publication statusPublished - May 2008
Externally publishedYes


  • GFR predictions
  • Kidney disease
  • Plasma cystatin C

ASJC Scopus subject areas

  • Nephrology


Dive into the research topics of 'Inaccuracy of GFR predictions by plasma cystatin C in patients without kidney dysfunction and in advanced kidney disease'. Together they form a unique fingerprint.

Cite this