The terms subclinical hypothyroidism (SubHypo) and subclinical hyperthyroidism (SubHyper) entered the literature in the early 1970s. Remarkably, nearly 50 years later, the health impact of both conditions remains uncertain.
SubHypo is the commonest disorder of thyroid function in iodine-sufficient populations. Diagnosis can be problematic, because of the effects of non-thyroidal illness on thyroid function and because serum TSH levels increase with age. A recent large, randomised controlled trial reported no symptomatic benefit of levothyroxine treatment in older, largely asymptomatic people with mild SubHypo, as did a subsequent meta-analysis, but this is of little relevance to managing younger patients and those with symptomatic ill health. There are no RCTs of levothyroxine on cardiovascular outcomes in SubHypo, but cohort studies (including several from the Asia-Oceania region) and RCTs with surrogate endpoints suggest that SubHypo is an independent cardiovascular risk factor. Cardiovascular risks of SubHypo appear highest in people who are younger (<65-70 y old), with higher TSH levels (>7 mU/L) and who have other cardiovascular risk factors.
The impact of subclinical hyperthyroidism (SubHyper) on health is becoming better defined. As for SubHypo, RCT data is largely lacking, but a consistent body of evidence suggests that endogenous SubHyper is a risk factor for atrial fibrillation, cardiovascular mortality and fracture. SubHyper is also associated with dementia, but the evidence for a causal relationship is less convincing than for CVD and fracture. Recent studies using Mendelian randomization have provided further evidence for a causal relationship between SubHyper and atrial fibrillation, but not for other cardiovascular diseases. The health risks of exogenous SubHyper (in patients on levothyroxine treatment) are poorly defined, but have probably been somewhat exaggerated.