At Belfast City Hospital all routine chemistry profiles are analysed on
Ortho VITROS 950 analysers. The creatinine method is enzymatic. Samples for
creatinine clearance are analysed on a Beckman CX7 (Jaffe).
At result validation the creatinine result from an electrolyte profile is
beside the result from the creatinine clearance request. Several large
discrepancies have been noted between the two results. In one particular
instance samples from a renal clinic were analysed on the Vitros analysers
for electrolyte profile on Friday pm but the blood and urine samples for
creatinine clearance were not collected from the clinic until Monday am.
Analysis on Monday am produced:-
Vitros 950 result Beckman CX7 result (umoL/L)
84 173
168 234
97 171
90 150
78 132
112 199
130 178
88 140
The delayed samples when re-analysed on Vitros produced similar results to
those obtained on Friday pm.
In the first sample serum creatinine of 173 umoL/L produced a creatinine
clearance of 61ml/min compared to the correct value of 125ml/min.
Previous papers and letters in the Annals (1992:29:307-309 , 1993:30:108-109
, and 1993:30:508-509) have noted this problem with creatinine measurements
using the Jaffe reaction and proposed the interfering component to be
Pyruvate which indeed is the case. At BCHT pyruvate levels up to 960umoL/L
were found in an experiment which produced apparent increases in creatinine
values up to 55umoL/L. When the pyruvate was removed from these samples and
re-analysed the original results were obtained.
As stated by Simpson this effect was eliminated by storing the unseparated
sample in a refrigerator rather than at room temperature.
Tom Hayes
Clinical Chemistry Department
Belfast City Hospital Trust
Tel. 028 9032 9241 Ext.2615
|