Competition for thyroid hormone binding to plasma proteins
  In contrast to the binding proteins for corticosteroids, vitamin D and the sex hormones, which are highly specific for a single family of ligands, TBG shows wide cross reactivity with unrelated hydrophobic ligands, such as NEFA and numerous drugs (table 1). The substances that interfere with T4 and T3 binding are themselves highly bound to albumin. Because of differences in therapeutic concentration and albumin binding, the true hierarchy of in vivo drug competitor potency in undiluted serum differs markedly from the rank order of drug affinity for isolated TBG (Lim et al, 1988).


Dilution effects

The literature on competitive effects of drugs is often unclear because the serum dilution and albumin concentration in the assay systems is poorly defined. It is technically easier to estimate the free concentration of a highly bound ligand such as T4 in diluted serum, but diluted serum does not maintain the concentrations of free hormone, competitor(s) and unoccupied binding sites in the relationship that exists in vivo. The terms "pre-dilution" and "co-dilution" are useful in considering these artefacts.

Pre-dilution occurs when the concentration of albumin is decreased by dilution before a competitor is added (fig. 2). At lower concentrations of albumin, there is a disproportionate increase in the free competitor concentration. Hence, pre-dilution before addition of competitor magnifies apparent competition, particularly when the effects of NEFA are considered (Mendel et al, 1986).


Figure 2:



This schematic contrasts pre-dilution of a serum sample, which will falsely accentuate the inhibitory effect of a binding competitor that is added to the diluted sample, and codilution, which will obscure inhibition of T4 and T3 by a competitor that is present in the original serum sample. (Reproduced with permission from Stockigt et al, 1996).


The opposite artefact, an under-estimate of competitor potency, occurs with co-dilution (fig 2). If whole serum is serially diluted so that total concentrations of binding proteins, hormone and competitors diminish in parallel, the free concentrations do not change in parallel. For example, in studies of the T4-displacing effect of a drug that has a free fraction in serum of 1:50, the free drug concentration has decreased markedly at a serum dilution of only 1:10, while progressive dissociation from its much larger reservoir of bound ligand sustains the free T4 concentration up to at least 1:100 dilution. Free T4 will be underestimated after co-dilution because competitor potency has been lost.

The importance of this type of artefact has been shown for therapeutic concentrations of phenytoin and carbamazepine. These drugs increased the free fraction of T4 by 40-50% using ultrafiltration of undiluted serum (Surks & Defesi, 1996), but no increase in T4 free fraction was seen with a commercial single-step free T4 assay after 1:5 serum dilution. During continuing drug therapy, total T4 was lowered by 25-50%, but free T4 concentrations were within the normal range, when calculated from estimates made without dilution . Notably, the ability of three commercial free T4 assays to detect the T4-displacing effect of furosemide was most obvious (fig 3) in the method with least sample dilution (Hawkins, 1998).


Figure 3:



Effecft of furosemide on estimates of serum free T4 by three commercial methods that involvwe different sample dilution factors. The competitor effect that increases free T4 is progressively lost with increasing sample dilution. The least dilution sample most closely reflects the in vio effect. (Redrawn from Hawkins et al, 1999).


Kinetics of the competitor

The kinetics of the competitor itself will determine its influence on T4 and T3 binding in vivo. A competitor of long half-life will eventually result in a new steady state with a normal free hormone concentration and a lowered total hormone concentration, i.e. an increased free fraction. In contrast, a competitor of short half-life such as furosemide (Newnham et al, 1987) or salsalate (Wang et al 1998) will lead to variations in the free hormone concentration, depending on the time between dosage and sampling, thus making it difficult to accurately assess thyroid status.



It is not yet known whether intermittent drug-induced increases in serum free T4 and T3 concentration can augment thyroid hormone action in humans, for example in patients with severe cardiomyopathy receiving high dosage of furosemide, where such an effect could be damaging. It is notable that a T3 and T4-displacing synthetic flavonoid has a transient thyromimetic effect in rats (Leuprasitsakul et al, 1990).


Albumin occupancy

Increasing concentrations of any substance that shares albumin binding sites with a competitor can increase the free concentration of that competitor. A substance that can exert such a "cascade" effect on T4 binding in serum is 3-carboxy-4-methyl-5-propyl-2-furanpropanoic acid (CMPF), a naturally occurring furanoid acid that accumulates in renal failure. Serum concentrations of CMPF that occur in renal failure had only a minimal direct effect on T4 binding, but the T4-displacing effects of therapeutic concentrations of furosemide, diflunisal and aspirin were augmented by 20-60% (Lim et al, 1993). Thus, by displacement of a direct drug competitor from albumin, free hormone concentrations can be influenced by substances that have negligible direct interaction with the hormone binding site.


Spurious competition

The effect of heparin to increase the apparent free T4 concentration in vitro is an example of spurious competition. Increases in free T4 as a result of heparin treatment appear to result from in vivo release of lipase, followed by in vitro generation of NEFA during sample storage (Mendel et al, 1987). Low molecular weight heparin preparations have a similar effect (Stevenson et al, 1998). As a result of this effect (fig 4), serum NEFA concentrations at the time of assay can be much higher than they were in vivo. Heparin doses as low as 10 units can cause this artefact after prolonged incubation of samples at 37C, especially if serum triglyceride is increased (Jaume et al, 1996).


Figure 4:



Summary of the heparin-induced changes that can markedly increase the apparent concentration of serum free T4. Heparin acts in vivo (left) to liberate lipoprotein lipase from vascular endothelium. Lipase acts in vitro (right) to increase the concentration of free fatty acids to level >3 mmol/l, resulting in displacement of T4 and T3 from TBG. In vitro generation of free fatty acids is increased by sample storage at room temperature, or incubation at 37° and by high concentration of serum triglyceride. The T4-displacing effect of free fatty acids is accentuated at low albumin concentrations. (Reproduced with permission from Oxford Textbook of Endocrinology, 2000).


This artefact, which becomes especially marked during incubation at 37C for equilibrium dialysis, (generally regarded as the "gold standard" for free T4 measurement), appears to account for some reports of method-dependent increases in apparent free T4 concentrations in critically-ill subjects (Sapin et al, 2000). Where heparin has been given, assays of total T4 and T3 are likely to be more informative than assays of free T4 and T3, unless laborious precautions are taken to avoid in vitro generation of NEFA.