Subclinical Hypothyroidism in the Elderly - A Small Victory in Addressing Overdiagnosis

New England Journal of Medicine June 2017

Florence is 82. She’s starting to get a bit frail, lives out in the countryside and getting to the surgery can be difficult. She’s on a couple of antihypertensives and a couple of inhalers for her COPD and has had some blood tests for mild cognitive impairment. She sleeps a bit more in the day time (so will I if I get to 82). Her TSH comes back at 7.5 with a normal T4. What do we do? First thought ‘rats, why did I do the TFT in the first place’. 

I’m sure we have all been reading a lot about overdiagnosis and overtreatment over the past cou- ple of years, and one of the aspects that has really hit home with me is the idea of the ‘treatment burdens’ we place on patients. Not just medications, but testing, appointments etc. We have for many years been pushed down the path of ever increasing testing, monitoring and treatment, mainly based on evidence that was often not really applicable to the real world patients we are seeing in General Practice. We have known for some time we are just doing too much ‘stuff’ to our patients for minimal gains but have felt duty bound to ‘follow the guidelines’. 

However, the recent NICE guideline on multimorbidity has been a total game changer in my view. It has given us the backing to make those judgement calls on whether testing or treating the patient in front of us is really going to improve their quality of life or not. One of the difficulties though is that the evidence base on when NOT to treat is still in it’s infancy. 

However, one small victory came in the shape of an RCT published in the NEJM in June 2017 looking at subclinical hypothyroidism in the elderly. Richard Lehman in his blog nicely suggested that ‘the thyroid gland could be described as the playground of overdiagnosis’. It was an RCT of >700 patients >65 years old with subclinical hypothyroidism (TSH 4.6 to <20 with normal T4). They were randomised to treatment aiming for normal TSH levels, or placebo (which also had sham blood tests and dose adjustments). At 1 year there was no difference in either hypothyroid symptom score or tiredness score. 

So back to Florence and what we do about her subclinical hypothyroidism. Previously I would have been tempted to treat it just in case it was causing the cognitive impairment, but let’s be honest, it’s much more likely to be age related and/or vascular as she was a heavy smoker. Given the backing of the multimorbidity guideline I’m now not going to increase her treatment burden by adding more pills, and the hassle of another half a dozen appointments for blood tests and follow- ups, for something that is highly unlikely to improve her quality of life. I’m going to suggest we just watch and wait and check it again as part of her annual review next year.

Rob Walker