Thyroid Monitoring and Amiodarone-Induced Thyroid Disease in Australian General Practice: A Retrospective Cohort Study

TL;DR

This study found that many Australian patients on amiodarone for atrial fibrillation did not have regular thyroid tests, leading to a higher risk of thyroid dysfunction.

Contribution

The study provides real-world evidence on thyroid monitoring adherence and risk factors for amiodarone-induced thyroid disease in general practice.

Findings

Only 82.7% of amiodarone-treated patients had thyroid tests during 18 months of follow-up.

Amiodarone-treated patients had significantly higher rates of hypothyroidism and hyperthyroidism compared to untreated patients.

Baseline TSH levels and comorbidities like heart failure and chronic kidney disease were risk factors for hypothyroidism in amiodarone users.

Abstract

Background: Australian guidelines recommend conducting thyroid function tests (TFTs) before commencing amiodarone and every six months subsequently. This study sought to investigate thyroid monitoring in Australian general practice patients with atrial fibrillation (AF) who commenced amiodarone. Methods: We performed a retrospective observational analysis using a nationwide primary care dataset to examine whether TFTs were conducted according to guidelines following amiodarone initiation in euthyroid patients aged 18 years or older with AF. Secondary outcomes included the prevalence of amiodarone-induced thyroid dysfunction (AITD) and the identification of factors associated with its development. Results: In total, 12,932 patients with AF were included. Of these, 1306 (10.1%) had commenced long-term amiodarone. Two hundred twenty-six (17.3%) of the patients commenced on amiodarone did not have any recorded TFT results during an 18-month follow-up period. During follow-up, 18.1% and 4.4% of patients developed hypothyroidism in the amiodarone-treated and amiodarone-untreated groups, respectively (p < 0.0001). The corresponding values for hyperthyroidism were 7.3% and 2.5% in the amiodarone-treated and amiodarone-untreated groups, respectively (p < 0.0001). In the subset of patients commenced on amiodarone, after controlling for the number of TFTs within the follow-up, the risk factors independently associated with the development of hypothyroidism were baseline thyroid stimulating hormone (TSH) level (adjusted odds ratio/AOR: 3.80 (95% confidence interval: 3.00–4.82)) and the comorbidities heart failure (AOR: 1.64 (1.09–2.46)) and chronic kidney disease (AOR: 2.29 (1.26–4.18)). Baseline TSH (AOR: 0.43 (0.28–0.63)) was significantly associated with the development of hyperthyroidism in patients taking amiodarone. Conclusions: AITD was relatively common, occurring in one-quarter of patients within 18 months of initiation of amiodarone. Increased awareness is required amongst both clinicians and patients of the need for regular thyroid monitoring during therapy with amiodarone.

Figures