On Thu, 10 Jan 2002 16:06:22 +0000, [log in to unmask]
<[log in to unmask]> wrote:
>
>Not great, but if I've got chromogenic junk in my serum I'd rather have my
>creatinine measured by a hydrolase method than by Jaffe. Unfortunately,
>one can't plug an OCD creatinine (or BCP albumin) into the Andrew Levey et
>al GFR formula (AIM, 1999, 130:461-470) since they used an Astra 8 (kinetic
>alkaline picrate, BCG).
>
Unfortunately the between manufacturer agreement for serum creatinine is
depressingly poor and a sad indictment of the diagnostics industry in the
21st century.
Nigel Lawson in Nottingham has recently completed an exercise for the Trent
Audit Panel in which he sent out 48 serum pools of varying creatinine
concentration to 10 labs. He also sent out the NIST standard and arranged
ID-MS analysis of 10 samples. In short, the agreement was poor. Perhaps
surprisingly, given the "chromogenic junk" argument, the Ortho results were
consistently higher than Roche, Olympus or ID-MS, although Ortho gave the
closest result to the NIST standard. No method agreed with ID-MS. From a
clinical standpoint, how does 701 compared to 561 umol/l for the average of
two methods for the same pool sound? Or 253 vs 300? Not good, and rather
undermines the rather precise correction formulae previously discussed
doesn't it?
Our renal colleagues will become increasingly aware of these discrepancies,
as patients move between hospitals. We need to be aware of the problem and
force the manufacturers to improve the situation.
Paul Masters
Chesterfield
|