Need for standardization of qualitative and
quantitative assays for rheumafactor detection
and quantitation
Ivan Brandslund and Erik D. Lund, Dept. of Clinical Chemistry,Vejle Hospital, Denmark, and J0rgen
Rasmussen, Dept. of Clinical Chemistry, Kolding Hospital, Denmark.
The problem:
In the county of Vejle, Denmark, with 330.000
inhabitants it was decided to centralise the
rheumatology service in 1992 to Give Hospital, a
satellite hospital to the county's main hospital,
Vejle. Until then, the four larger hospitals in the
county bad their own rheumatology departments,
and the local hospitallabs performed most of the
analyses on their patients, except the rarer or spe–
cial tests. Among these were the quantification of
rheumafactor. This test was done in Kolding and
Vej le, using two different methodologies. This had
not been (or been considered to be!) a problem
previously.
Both Kolding and Vejle had moved through a
process from using Rose-Waaler like assays with
titration andreporting results in titer or as "negative"
and "positive", to a quantitative determination.
Kolding chose the Behring latex-hased agglutina–
tion test with nephelometric reading in a Cobas
Fara analyzer. Vej le chose an ELISAmethod using
human IgG coated wells and detection with
peroxidase labeled rabbit F(ab) anti-human IgM.
Before 1992 the patients were generally not
transferred between hospitals, so the difference in
reference interval (K:<lOklU/L, and V:<8klU!L)
was not a problem. Further, it is generally accepted
that the value should be higher, at least above
30klU/I, to justity that the result should be taken as
a positive diagnostic indication for RA. And RF
detection is primarily used for the diagnosis of RA,
not for monitoring disease activity.
Klinisk Umi
i
Norden l, 1994
However, when the central department of
rheumatology began their activities, it was noted,
that patients exhibited considerable differences in
RF values, depending on which laboratory had
performed the test. This was a problem for the
doctor treating the patient, for in some patients the
value could change up to a factor 2! Even though
experienced rheumatologists do not request the
analysis more than 3-4 times per year or use the
result as an indicator for the disease activity or
success of treatment over a short time period, it is
still annoying not to obtain the same results, inde–
pendent ofmethod andperforming laboratory.And
over a longer time horizon, 1-2 years, the test is
useful in patient management. With the new
legislation in Denmark, making all test results
immediately available to patients, the doctor on top
of the professional problems will have to explain
the reason for the apparent changes of the test result
for the patient.
Results from proficiency testing programs.
It
became clear from the Labquality (Finland)
survey in 1991 that the two methods differed
significantly, and also that the ELISA method
differed considerably from the classical
Rose-Waaler method (Table 1).
19
