Apt. Harmoinen et al., Evaluation of plasma cystatin C as a marker for glomerular filtration ratein patients with type 2 diabetes, CLIN NEPHR, 52(6), 1999, pp. 363-370
Aim: To evaluate plasma cystatin C as a marker of the glomerular filtration
rate in patients with type 2 diabetes and their age and sex-matched contro
ls. Materials and methods: Forty-seven patients with one decade of type 2 d
iabetes and 51 non-diabetic control subjects were studied. Plasma cystatin
C was measured by particle-enhanced turbidimetric immunoassay in a new appl
ication for the Hitachi 704 analyzer. For comparison, plasma creatinine and
creatinine clearance were measured. The plasma clearance of Cr-51-EDTA by
the single injection method was utilized as reference. Results: In patients
with type 2 diabetes the correlation coefficient between plasma cystatin C
and the plasma clearance of Cr-51-EDTA was 0.774 (Spearman's coefficient)
and that between plasma creatinine and the plasma clearance of Cr-51-EDTA w
as 0.556 (p = 0.001 for the difference). The correlation between creatinine
clearance and the plasma clearance of Cr-51-EDTA was 0.411. In receiver op
erating characteristic (ROC) curve analysis the diagnostic accuracy of plas
ma cystatin C was significantly better than that of plasma creatinine (p =
0.047) or creatinine clearance (p = 0.001). The best diagnostic efficiency
(98%) for cystatin C was obtained when the cut-off limit was set at 1.32 mg
/l. In the control group the correlation coefficients were: between cystati
n C and the plasma clearance of Cr-51-EDTA 0.627, between creatinine and th
e plasma clearance of Cr-51-EDTA 0.466 and between creatinine clearance and
the plasma clearance of Cr-51-EDTA 0.416. The area under the ROC plot curv
e of cystatin C was also greatest in the control group, but the diagnostic
accuracy of cystatin C was marginally better than that of either plasma cre
atinine (p = 0.05) or creatinine clearance (p = 0.08). Among the control su
bjects Various non-renal causes may have interfered with cystatin C concent
rations reducing the correlations. Conclusions: Cystatin C measurement is a
more sensitive and specific test for GFR in patients with type 2 diabetes
than plasma creatinine or its clearance, when GFR is normal or only slightl
y reduced. If an elevated cystatin C concentration is found, non-renal fact
ors have to be excluded. The turbidimetric application described here can e
asily be applied for most clinical chemistry analyzers and is therefore use
ful in daily clinical practice.