Thyroxine treatment is a serious business

Timo Strandberg is a Professor of Geriatrics who works at the Universities of Helsinki and Oulu.

Thyroxine_Structural_FormulaeAmid important clinical issues such as the dangers of dental amalgam fillings, the evils of chronic candida yeast syndrome, the big benefits of low-carb diets and the like, thyroxine has been on the headlines during recent years. We’ve heard especially about the lack of thyroxine, and even some distinguished colleagues seem to have thought that if you’re a bit depressed, tired, cognitively impaired, gaining weight etc. thyroxine is the drug for you. Accordingly, treatments with this hormone have clearly increased, for example in the UK 3-fold between 1998 and 2010, and treatment for marginally elevated thyroid-stimulating hormone (TSH) levels have become more common. And lo and behold: patients often get better – at least temporarily (ever heard of the placebo effect?).

But seriously, thyroxine is not an inert “placebo”, but a powerful substance, which especially in older persons may be really dangerous if you have too much of it. On the other hand, too little thyroxine is obviously not healthy either. The harms of true hypothyroidism are well-known and recognised by all physicians (even though to the best of my knowledge there are no double-blind, randomised, placebo-controlled trials of the effects of thyroxine on prognosis). In practice, the grey area for the clinician lies in the subclinical TSH range, between 3.6 and 10 mU/L.

Is the eager treatment of subclinical hypothyroidism truly justified, especially in older patients? Is the subclinical state associated with important harms cross-sectionally or long-term? That was the background for our goal to look at the relationships between the levels of TSH, and some clinically meaningful variables in a group of older people.

Our cohort is the DEBATE study, a population-retrieved group of individuals aged 75 years and over (n=400) recruited in 1999. TSH was measured and cognitive function assessed with the CERAD neuropsychology test battery (routinely used to diagnose memory diseases in Finnish geriatrics clinics) in 2002 and mortality follow-up continued through 2012. Because all individuals in DEBATE had a history of cardiovascular disease (mostly coronary heart or cerebrovascular disease), we anticipated that they would be especially sensitive even to milder thyroid dysfunction.

We could see no meaningful relationship between TSH range mostly below 10 mU/L and long-term mortality, nor with cognitive function at baseline. The results suggest that at least in the subclinical range, thyroxine replacement therapy would not be needed in people aged 75 years and older.

Our cohort was not very large, so it’s appropriate to look at other evidence, too. They seem to support our findings as far as conditions which are often fatal in older people are considered. Subclinical hypothyroidism has not been associated with risk of fractures, stroke or heart failure. On the other hand, in a very large and very recent study, mortality seemed to be increasing after a threshold TSH value of 6.4 mU/L. The population was 65+ , and we have no proof that their prognosis would have got better if they got thyroxine treatment.

Subclinical thyroid dysfunction (TSH between 3.6 and 10  mU/L) is not an infrequent finding and may be found in > 10% of people aged 80 and older. Looking at the totality of evidence and remembering the potential harms of thyroxine replacement in this age group (atrial fibrillation, angina pectoris, heart failure, osteoporosis…), I would be very careful to start thyroxine below TSH levels 10 mU/L among older people.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s