To the Editor: In about 2% to 5% of patients with subclinical hypothyroidism, the condition progresses to overt hypothyroidism each year.1 It is currently recommended that thyroid function tests should be repeated at 6- to 12-month intervals to monitor improvement or worsening in level of thyroid-stimulating hormone (TSH).1 Testing for thyroid peroxidase antibody (TPOAb) is also common practice in subclinical hypothyroidism, as it has been shown that individuals with raised TSH and TPOAb levels have a 40-fold increased risk of developing overt hypothyroidism.2,3

We audited the management of patients who had a result indicating subclinical hypothyroidism from our hospital laboratory, focusing on follow-up thyroid function and TPOAb tests. In December 2003, we retrospectively inspected clinical case notes of patients who had been reported in November 2002 with a TSH level of 4.1–9.9 mIU/L (normal reference interval, 0.4–4.0 mIU/L) and a free thyroxine (FT4) level within the reference range of 10–23 pmol/L (subclinical hypothyroidism). We also contacted the patients’ general practitioners (GPs) for further information when necessary.

There were 72 patients with results suggesting subclinical hypothyroidism. Of these, we excluded 29 from other hospitals and three whose GPs were not able to be contacted. Of the remaining 43 patients, 18 had no previous history of thyroid disease (8 men and 10 women; age range, 24–90 years). Six patients were seen in the emergency department, four in the outpatient clinics, and eight in the wards. All the laboratory reports of subclinical hypothyroid results were accompanied by a comment advising repeat thyroid studies at a later date and thyroid antibody tests. However, only three of the 18 patients (17%) had TPOAb tested. Only seven of the 18 patients (39%) were followed up with repeat thyroid function tests, at intervals ranging from 3 days to 7 months. One patient, with a TSH level of 9.8 mIU/L, was started on thyroid replacement therapy. The GPs of 10 of the 11 patients who did not have follow-up testing were not informed of the initial TSH results.

The low rate of follow-up of hospital patients with a first-time diagnosis of subclinical hypothyroidism is of concern. While the increase in TSH level in some of these patients may have been related to sick euthyroidism, this can only be confirmed by normalisation of TSH level on repeat testing. TPOAb testing can be deferred until confirmation of persistently raised TSH level. Better strategies, such as a computerised system for selective copying of results to GPs whenever relevant, and inclusion of treatment advice dependent on TPOAb status in the reports,4 may be needed to improve follow-up.