Searchable abstracts of presentations at key conferences in endocrinology
Endocrine Abstracts (2007) 13 P242

SFEBES2007 Poster Presentations Neuroendocrinology and behaviour (including pituitary) (27 abstracts)

Subclinical hypothyroidism is associated with increased cardiovascular risk: A meta-analysis

Salman Razvi , Jolanta Weaver & Simon Pearce


University of Newcastle upon Tyne, Newcastle upon Tyne, United Kingdom; Queen Elizabeth Hospital, Gateshead, United Kingdom; Royal Victoria Infirmary, Newcastle upon Tyne, United Kingdom.


There are discrepancies between different studies with regards to the association of subclinical hypothyroidism (SCH), dyslipidaemia and cardiovascular (CV) disease. Therefore, we have conducted a systematic search and meta-analysis of population-based studies of CV disease in which thyroid status was examined and randomised controlled trials (RCTs) of treatment.

The PubMed and Embase databases were searched for relevant studies published between 1976 and 2006. A total of 1446 articles were identified, 206 abstracts were reviewed, 79 full-length articles were compared to preset criteria, and 24 population-based studies and 8 RCTs included in the analysis. Missing or indirectly reported data was obtained from 6 study authors. All analysis was performed according to Cochrane collaboration guidelines, using RevMan 4.2 software.

CV disease was more prevalent in people with SCH (n=2,079) as compared to euthyroid controls (n=23,289), OR (95% CI) of 1.21 (1.02–1.48), P=0.04. Incident CV event rates were not different in SCH patients (n=1,310) versus euthyroid individuals (n=12,464), OR of 1.28 (0.87–1.88), P=0.21. However, the incidence of CV events was significantly higher in SCH patients considering 5 studies in which the minimum age of subjects was at least 65 years (SCH <65), OR of 1.78 (1.15–2.77), P=0.01. Mortality (cardiovascular or all-cause) was no different between SCH and euthyroid subjects. But in the SCH <65 studies, CV mortality was elevated with an OR of 1.53 (1.03–2.29), P=0.04. LDL cholesterol levels were higher in SCH individuals (n=1,219) versus euthyroid individuals (n=85,373) (mean difference (95% CI) 0.2 mmol/L (0.08–0.33)), P=0.002. L-thyroxine treatment reduced LDL cholesterol by 0.24 mmol/L (0.01–0.46) and BMI by 0.38 kg/m2 (0.15–0.60) but not HDL cholesterol or triglyceride levels.

Our systematic analyses of these large datasets suggest that there may be real age-related differences in the risk of CV disease attributable to SCH. This has potentially important implications for therapy.

Article tools

My recent searches

No recent searches.