Abstract
AimsAtrial fibrillation (AF) is a risk for patients receiving thyroid hormone replacement therapy. No published work has focused on pharmacogenetics relevant to thyroid dysfunction and AF risk. We aimed to assess the effect of L-thyroxine on AF risk stratified by a variation in a candidate gene.Methods and ResultsA retrospective follow-up study was done among European Caucasian patients from the Genetics of Diabetes Audit and Research in Tayside Scotland cohort (Scotland, United Kingdom). Linked data on biochemistry, prescribing, hospital admissions, demographics, and genetic biobank were used to ascertain patients on L-thyroxine and diagnosis of AF. A GWAS-identified insulin receptor-INSR locus (rs4804416) was the candidate gene. Cox survival models and sensitivity analyses by taking competing risk of death into account were used. Replication was performed in additional sample (The Genetics of Scottish Health Research register, GoSHARE), and meta-analyses across the results of the study and replication cohorts were done. We analyzed 962 exposed to L-thyroxine and 5,840 unexposed patients who were rs4804416 genotyped. The rarer G/G genotype was present in 18% of the study population. The total follow-up was up to 20 years, and there was a significant increased AF risk for patients homozygous carriers of the G allele exposed to L-thyroxine (RHR = 2.35, P = 1.6e–02). The adjusted increased risk was highest within the first 3 years of exposure (RHR = 9.10, P = 8.5e–04). Sensitivity analysis yielded similar results. Effects were replicated in GoSHARE (n = 3,190).ConclusionHomozygous G/G genotype at the INSR locus (rs4804416) is associated with an increased risk of AF in patients on L-thyroxine, independent of serum of free thyroxine and thyroid-stimulating hormone serum concentrations.
Highlights
Hypothyroidism is the most common thyroid disorder affecting about 3–5% of the general population, and patients are nearly always treated with L-thyroxine (Flynn et al, 2004; Wiersinga et al, 2012)
Most published work has focused on identifying genetic variants associated with atrial fibrillation (AF) (Fatkin et al, 2017), less on pharmacogenetics relevant to AF management (Huang and Darbar, 2016), few on genetic determinants of thyroid function/dysfunction associated with AF (Ellervik et al, 2019; Larsson et al, 2019; Salem et al, 2019), but none on pharmacogenetics relevant to thyroid dysfunction on AF risk
Those exposed to L-thyroxine had a higher average serum thyroid stimulating hormone (TSH) (2.7 vs 1.8 mIU/L, P < 1e-03), average TSH levels were within the biochemical reference range
Summary
Hypothyroidism is the most common thyroid disorder affecting about 3–5% of the general population, and patients are nearly always treated with L-thyroxine (Flynn et al, 2004; Wiersinga et al, 2012). Patients usually respond well to treatment, and dosage is monitored in response to a combination of serum thyroid stimulating hormone (TSH) concentration and patients’ symptoms (Wiersinga et al, 2012). Increased risk of cardiovascular disease, atrial fibrillation (AF) and bone fractures have been described in patients receiving long-term replacement thyroxine therapy (Flynn et al, 2010; Chaker et al, 2015; Floriani et al, 2017). Risk of AF for patients taking L-thyroxine is partly related to their dose and serum TSH concentration (Flynn et al, 2010). Most published work has focused on identifying genetic variants (common and rare) associated with AF (Fatkin et al, 2017), less on pharmacogenetics relevant to AF management (Huang and Darbar, 2016), few on genetic determinants of thyroid function/dysfunction associated with AF (Ellervik et al, 2019; Larsson et al, 2019; Salem et al, 2019), but none on pharmacogenetics relevant to thyroid dysfunction on AF risk
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
