In the UK NEQAS for steroid hormones, the Abbott Architect
assay is the most positively biased method amongst a number of
others with varying degrees of positive bias.
In a recent distribution (sample 274A) set to probe the upper
reference range border, the all laboratory trimmed mean (ALTM)
was 3.56 nmol/L and the Architect method mean was 4.26 nmol/L.
The GCMS reference method value for this sample was 2.72
nmol/L. (The few remaining in-house extraction users had a
median of 3.1 nmol/L - well done chaps!)
For a 'mid range' sample (275C) the ALTM was 1.99 nmol/L, the
Architect method mean was 2.99 nmol/L and the GCMS value was
1.225nmol/L.
These data appear to be consistent with John's experience and the
findings of the Australian scheme. On this basis, his laboratory's
upper limit of normal for the Architect system there would appear to
be 'about right' but I am not at all surprised that his
endocrinologists are unhappy with these numbers!
The 'numbers' laboratories issue for this analyte are highly
dependent on the method, several of which have high variability
amongst users so that, depending on the sample and its
testosterone concentration you can get any number you like
between half and double the 'true' value! Bias against the reference
method (ID-GCMS) is concentration dependent (higher at lower
concentrations).
I have not surveyed reference ranges recently, but past experience
suggests that many laboratories may not take method bias fully
into account when setting these.
Such is the poor state of the art of 'measurement' of this analyte in
the female matrix, that I am holding a meeting with the diagnostics
industry and other interested parties to discuss it (and oestradiol)
in Birmingham on the 16th December 2002. (Bit far for you to
come John!). Those interested in attending should visit
http://www.ukneqas.org.uk/saf01.htm
and complete the web form. Some 30 places have already been
filled out of a maximum of 70.
Hope this helps
Jonathan Middle
On 15 Oct 2002, at 12:46, Beilby, John wrote:
> Dear All
>
> We are using the Abbott Architect Immunoassay system to determine
> testosterone levels in males and females. On the AQAP external QC
> program we consistently get high results for low levels of
> testosterone. For example, with a median values of 1.6 nmol/L we get
> 3.0 nmol/L. The CV for the assay is excellent at 2.7% over a range of
> 3.0 to 46.0 nmol/L.
>
> We have established a new reference interval for females at <4.8
> nmol/L but the Endocrinologists do not like the numbers.
>
> What do other Architect users quote for their female testosterone
> reference interval and what reference intervals do you use for the
> free androgen index in females?
>
> Thanks
>
>
> John
>
> Dr John Beilby
> Senior Biochemist
> Clinical Biochemistry
> PathCentre
> Locked Bag 2009
> Nedlands
> Western Australia 6009
>
> ------ACB discussion List Information--------
> This is an open discussion list for the academic and clinical
> community working in clinical biochemistry.
> Please note, archived messages are public and can be viewed
> via the internet. Views expressed are those of the individual and they
> are responsible for all message content.
>
> ACB Web Site
> http://www.acb.org.uk
> List Archives
> http://www.jiscmail.ac.uk/lists/ACB-CLIN-CHEM-GEN.html
> List Instructions (How to leave etc.)
> http://www.jiscmail.ac.uk/
>
============================================
Jonathan Middle, UK NEQAS Birmingham
tel 0121 414 7300 fax 0121 414 1179
This message is intended only for the above
recipient(s). The opinions expressed are
mine alone and do not necessarily represent
those of UK NEQAS Birmingham, the University
Hospital Birmingham NHS Trust or the UK NEQAS
Organisation.
==================================================
------ACB discussion List Information--------
This is an open discussion list for the academic and clinical
community working in clinical biochemistry.
Please note, archived messages are public and can be viewed
via the internet. Views expressed are those of the individual and
they are responsible for all message content.
ACB Web Site
http://www.acb.org.uk
List Archives
http://www.jiscmail.ac.uk/lists/ACB-CLIN-CHEM-GEN.html
List Instructions (How to leave etc.)
http://www.jiscmail.ac.uk/
|