Article Text

Atrial fibrillation and the thyroid
  1. Registrar in Cardiology,
  2. Research Fellow in Cardiology,
  3. Consultant Cardiologist,
  4. University Department of Medicine and Department of Cardiology,
  5. City Hospital, Birmingham B18 7QH, UK;
  6. email: greg{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Sir,—We read with interest in the editorial by Forfar1 that both clinical and subclinical hyperthyroidism are associated with the subsequent development of atrial fibrillation. The association of hypothyroidism with atrial fibrillation is less recognised.2 ,3 For example, the Canadian Registry of Atrial Fibrillation Investigators4reported that 1.5% of 726 patients with atrial fibrillation had hypothyroidism over a period of 1.7 years. However, Tajiri et al reported that up to 8% of the 75 elderly patients with atrial fibrillation (mean age 75.6 years) studied were found to be hypothyroid, following a thyrotropin releasing hormone test.5

We recently encountered a 61 year old man who was referred by his general practitioner with a two month history of peripheral oedema and hypertension. He was found to be in atrial fibrillation, with blood pressure of 160/100 mm Hg. There was no evidence of heart failure, and he was clinically hypothyroid. Thyroid function tests revealed a thyroid stimulating hormone (TSH) concentration of 160 mIU/l (normal range 0.4–5.5) and free T4 of 3 pmol/l (normal range 9–19) confirming hypothyroidism. Transthoracic echocardiography showed normal left ventricular systolic function with borderline enlargement of the left atrium (4.1 cm) and the absence of left ventricular hypertrophy or pericardial effusion. After treatment with aspirin 300 mg, and thyroxine replacement for four months (dose titrated to 100 μg daily) he was found at outpatient review to be in sinus rhythm. A repeat 24 hour Holter monitor recorded a few short episodes of paroxysmal atrial fibrillation. At four months he appeared clinically euthyroid and repeat thyroid function tests showed a TSH of 20.8 mIU/l and free T4 of 15 pmol/l.

In the west Birmingham atrial fibrillation project, we reported the presence of hyperthyroidism in 16 of 213 white patients (7.5%) who were admitted to hospital with atrial fibrillation.6 No cases of thyroid disease were found among black or Asian patients. In our recent general practice survey, we found 17 cases (15.3%) of previous hyperthyroidism among 111 patients with atrial fibrillation, although only about half of the patients in the survey had thyroid function blood tests at any time.7

Although excessive chronotropic stimulation by thyroid hormones and activation of the sympathetic system may lead to the development of atrial fibrillation in thyrotoxicosis, our patient developed atrial fibrillation despite the low chronotropic “drive” with his hypothyroidism. Hypothyroidism may also be associated with cardiomyopathy or pericardial effusion, but there was no evidence for this on the echocardiogram. However, our patient was hypertensive, which may have led to a high left atrial pressure as a result of a reduced compliance in the left ventricle, although the latter is more common when left ventricular hypertrophy is present. Although our patient did not have left ventricular hypertrophy, the possibility remains that he may have had underlying coronary artery disease. This case should remind clinicians that hypothyroidism, as well as hyperthyroidism, can be associated with the development of atrial fibrillation, and careful vigilance is necessary.