Background: Currently, physicians remain uncertain whether to screen and/or treat subclinical thyroid disease. Many observational studies, including cross-sectional, casecontrol, and prospective cohort studies, have reported on the association between subclinical thyroid disease and an outcome of interest. All cause mortality, coronary heart disease, fracture risk, and pregnancy outcome may drive the decision to screen or treat subclinical thyroid disease.
As systematic reviews usually provide more comprehensive and robust conclusions, we conducted a critical synopsis of published meta-analyses in the field of subclinical thyroid disease.
Methods: We searched Medline, Web of Science, EMbase and the Cochrane Library to identify relevant meta-analyses investigating the association between subclinical thyroid disease and the previously mentioned endpoints.
Results: We identified a total of 10 meta-analyses.
The currently best evidence suggests that subclinical hypothyroidism (S Hypo) is a risk factor for coronary heart disease in young persons and a protection factor for all-cause mortality in older patients. S Hypo is associated with higher cholesterol values, and intervention with thyroid hormones has some beneficial effects on blood lipids and heart function. Data on health-related quality of life and symptoms, on the other hand, do not demonstrate significant differences between placebo and thyroid hormone therapy groups, even though cognitive function improves. Adverse effects are inadequately addressed.
Subclinical hyperthyroidism (S Hyper) may be a risk factor for all cause mortality, especially in men and with advancing age, but the evidence is limited to one meta-analysis.
In postmenopausal women S Hyper is a risk factor for osteoporosis (BMD measurements), but data on fracture risk are lacking.
We found no meta-analyses addressing pregnancy outcomes.
Conclusions: The evidence to screen and treat subclinical thyroid disease is mainly observational. The benefit of therapeutic interventions needs further validation by randomised controlled trials with long-term follow-up.
Prague, Czech Republic
24 - 28 Apr 2010
European Society of Endocrinology