Details of the Thyroid Story
(written approximately 2007; updated 3/2009 and 12/2011)

TSH: 1.0 is normal;  3.5 is not?

What is "normal" for TSH? This is controversial. But all you need to know (as far as I'm concerned) is that at least some endocrinologists have spoken out, saying that the upper limit of the normal range ought to be 2.5 rather than upwards of 4.Wartofsky

This means that we can therefore consider using thyroid hormone as a possible mood treatment at much lower TSH levels without being thought  to be "radical", or practicing "cowboy medicine" (ready, fire, aim). Instead of waiting for values above 4.25, as in my lab; or values over 10, as some endocrinologists have said for years, we can justify a trial of adding thyroid hormone when TSH is 2.5 or higher.  That is not the usual practice. 

This different interpretation of "the normal range" is based on the fact that TSH is not centered around a middle value (not a "bell-shaped curve" ) as in this diagram: 


but instead is "skewed to the left" as in this diagram, so that the most common value is between 1.0 and 1.5: 


The blue cross marks the point at which the curve begins to flatten into a long tail. Some endocrinologists think this is the beginning of "abnormal", rather than 4.5 (or higher in some labs and interpretations. In years past, it was 10.0 ). 

(The mean values in the schematic diagrams above are based on the following real data for mean values in young adults (Hollowell et al, CDC, 2002)):




T4 acts directly on the brain, not just via T3

Generally it's taught that T4 thyroid hormone works by getting converted to T3. This makes people wonder about the logic of using very high doses of T4.  Why don't you just give T3? 

Well, as the diagram below (next section) shows, thyroid hormone actions and regulation are extremely complex. But a new study (Caria, 2009) makes it clear that thyroid hormone, in the T4 form, acts directly on the brain.  Indeed, this research shows that a region of the brain which seems to play a central role in mood, the hippocampus, is directly regulated by T4. And T3 had the opposite effects, acting more like epinephrine in this region. At least in rats. 

These data are consistent with the idea that T4 does not have its effects by changing the entire thyroid system, including the pituitary release of TSH. Maybe the effect is more direct, and maybe some people need a lot of that effect for some reason. In that case, high-dose thyroid might produce an effect which cannot be gauged by measuring TSH. That's important, you see, because endocrinologists and primary care doctors get very anxious about the idea of psychiatrists giving thyroid hormone at doses that completely suppress TSH. They think that's unwise in principle, not just because of the possible risks of atrial fibrillation and decreased bone density (those risks are discussed in detail on the high-dose thyroid page).  

Very interesting....  Thanks to MJ for the article. 

A working diagram of thyroid control in the brain

Check out this diagram for a sense of how complicated the thyroid controls are, and how many different factors are affecting thyroid stimulating hormone (TSH), our basic measure of thyroid function.  At least one thing seems clear:  TSH is involved in a lot more than just telling the thyroid gland to make more thyroid hormones.  Maybe this is why we see dramatic results sometimes with thyroid hormones -- somehow the whole TSH system seems to be changed when the hormones come from the outside instead of the inside.  Whatever explanation we come up with, it will have to explain why some patients seem to get better even when they start with TSH values that are clearly "normal" by standard lab tests.  For examples of that, see my patients' TSH values and their results: you'll see several people who had a clear response when their TSH prior to treatment was around 1! (e.g. see patient #4 and 16!)  

Figure 4-9. Schematic representation of the main factors interacting in the regulation of TSH synthesis and secretion (DA: dopamine; SS: somatostatin; a-AD: a adrenergic pathways).