Session: SUN 414-431-HPT Axis Biology
Poster Board SUN-422
METHOD: All patients, with very few exceptions, with atrial fibrillation in the county of Northern Halsingland, Sweden, are registered. The reported prevalence is high, 2.5%. Hyperthyroidism was defined as a TSH-level below 0.4 (0.4-4.0). Mortality data was obtained from the Swedish National Death Registry.
Appropriate non-parametric statistic methods were used to compare groups. Cox proportional hazard regression models were applied to evaluate the independent association between hyperthyroidism and mortality. Associations are given as Hazard Ratio and 95% confidential intervals (HR (95% CI)). A p<0.05 was considered as statistical significantly.
RESULTS: 790 patients were diagnosed with atrial fibrillation between 1 January 2009 and 30 June 2011. TSH was present in 615 cases. 53 patients (8.6%) had hyperthyroidism.
The overall 18-months mortality was 22.6% in patients with hyperthyroidism as compared to 11.0% in those without, p=0.013. Mortality rates were similar in patients <75 years (n=298), 95.0 vs. 94.2%, n.s. In patients ≥75 years (n=317) hyperthyroidism was associated with increased mortality, 33.3 vs. 16.2%, p=0.015.
In a Cox-proportional hazard model including patients ≥75 years old, hyperthyroidism (HR=2.87 (1.40-5.86), p=0.004), a history of smoking (HR=1.96 (1.10-3.49), p=0.023) and older age (5-year cohorts) (HR=1.42 (1.08-1.88), p=0.013) were associated with increased mortality. Treatment with warfarin was almost associated with improved survival (HR=0.57 (0.32-1.02), p=0.058).
CONCLUSION: Hyperthyroidism seems to be associated with increased 18-months mortality in patients with newly diagnosed atrial fibrillation and an age of 75 years of older, but not in younger patients.
Nothing to Disclose: PA, ML
*Please take note of The Endocrine Society's News Embargo Policy at http://www.endo-society.org/endo2013/media.cfm
See more of: Abstracts - Orals, Featured Poster Presentations, and Posters