American Association for Clinical Chemistry
Better health through laboratory medicine
NACB - Scientific Shorts
NACB - Scientific Shorts (formerly NACB Blog)
By Steven Soldin, PhD, DABCC, FACB

Candidate reference methods for FT4 first separate the T4 binding proteins (TBG, prealbumin, albumin, thyroid antibodies) by either equilibrium dialysis or ultrafiltration at 37 C. This is followed by measurement of FT4 using either immunoassays or mass spectrometry. Using these approaches the resulting FT4 values correlate very well with log TSH, an important requirement for any FT4 assay.

Unfortunately neither the diagnostic companies nor the FDA have used this important relationship to test the validity of the direct analogue FT4 methods on a variety of chemistry platforms. In a study over several years we showed that the direct analogue FT4 method disagreed with the TSH value approximately 50% of the time in patients with either hypo- or hyper-thyroidism. In contrast the FT4 employing ultrafiltration at 37C followed by tandem mass spectrometry was found to consistently correlate well with log TSH.

There are several conclusions that can be drawn from all these studies. You cannot depend on the direct analogue FT4 result in patients with thyroid disease. Instead of recommending that all FT4's be performed by candidate reference methods employing either equilibrium dialysis or ultrafiltration, we suggest that a candidate reference approach employing either immunoassay or mass spectrometry be used whenever the TSH is above the 90th percentile or below the 10th percentile. This approach should optimize the correct diagnosis of these patients.













Please log in to leave a comment!

Posted by
On 12/9/2011

Are you including first and second generation FT4 assays under the title of "direct analogue" methods? While first generation assays had significant problems, second generation assays were much improved (but not perfect). This comment was approved by the NACBLOG editorial board. Please remember to add your name and affiliation!

Posted by
On 11/21/2011

This blog highlights a point known to most clinical chemists for years and yet no manufacturer has risen to the challenge. Dr. Soldin's reasoned approach of using the reference methods (Lc-MS/MS) on the most extreme cases should be adopted with haste by all state of the art labs. Jim Ritchie Emory University Atlanta, GA