Endocrine Abstracts (2019) 65 P395 | DOI: 10.1530/endoabs.65.P395

Thyroid dysfunction is common in hospitalised patients with pre-existing levothyroxine treated hypothyroidism

Barbara Torlinska, Jamie J Coleman & Kristien Boelaert


University of Birmingham, Birmingham, UK


Background: Levothyroxine replacement for hypothyroidism is safe and effective but may be associated with reduced quality of life and serious adverse effects if administered inappropriately. We set out to investigate if thyroid hormone profiles in hospitalised patients on levothyroxine are similar to those in inpatients without pre-existing thyroid dysfunction.

Methods: We conducted a cross-sectional study in patients admitted for non-endocrine reasons to a large tertiary centre between 2007 and 2011. All initial thyroid function tests (TFT) were included in the analysis. Patients with admissions for endocrine reasons and with co-morbidities of hyperthyroidism or thyroid cancer were excluded.

Results: We collected data on 15 710 inpatients with no thyroid dysfunction (TD) and 2240 inpatients treated for hypothyroidism with levothyroxine. The proportions of TFTs, based on the TSH and fT4 results combined, varied significantly between patients on levothyroxine and those with no TD. Only 41.7%(95%CI: 39.7–43.8) levothyroxine-treated patients were biochemically euthyroid (vs. 81.8% with no TD, P<0.001), 32.1%(30.2–34.1) had subclinical hypothyroidism (vs. 8.2%, P<0.001), 2.0% (1.4–2.6) overt hypothyroidism (vs. 0.2%, P<0.001), 7.3%(6.3–8.5%) subclinical hyperthyroidism (vs. 0.2%, P<0.001) and 3.4%(2.7–4.2) overt hyperthyroidism (vs. 0.6%, P<0.001). Atypical TFT results, not fitting any of the above categories, were found in 13.4%(12.1–14.9) of patients treated with levothyroxine and in 6.7%(6.3–7.1, P<0.001) without TD. The mean TSH in those on levothyroxine replacement was 2.68 mIU/l (2.58–2.80) compared with 1.81 mIU/l (1.79–1.83) in those without TD. The difference was statistically significant in univariate (P<0.001) and multivariable analysis (P<0.001) following correction for sex, age, severity of comorbidities (Charlson Comorbidity Index) and type of admission (elective or acute).

Conclusions: The results of the study highlight the difficulties of achieving good control with levothyroxine in hypothyroid inpatients. It remains unclear if thyroid dysfunction in these patients contributes to the need of hospitalisation and this requires further investigation.

Article tools

My recent searches

No recent searches.