Gesundheitswesen 2010; 72 - P136
DOI: 10.1055/s-0030-1266643

Are changes of serum TSH levels associated with alteration of cardiac electrical activity? Results from a large population-based cohort study

M Dörr 1, T Ittermann 1, S Baumeister 1, T Reffelmann 1, J Kors 2, S Felix 1, H Völzke 1
  • 1Ernst-Moritz-Arndt Universität, Greifswald
  • 2Erasmus Medical Center, Rotterdam, The Netherlands

Background: Thyroid dysfunctions have been link to various arrhythmias. However, there are no studies available that investigated possible effects between alterations of thyroid state and changes of general cardiac vulnerability over time. We analyzed whether changes of serum TSH levels (ΔTSH) were associated with alterations of the cardiac electrical activity as assessed by different ECG parameters. Methods: The relationship between ΔTSH and changes of selected ECG parameters over the following five years was analyzed in subjects aged 20–79 years from the population based, longitudinal Study of Health in Pomerania. ECG recordings were digitally stored and processed by the Modular ECG Analysis System (MEANS). Subjects with complete ECG data and without diseases or drugs that may influence cardiac electrical activity (n=1873, 1022 women) were included. The association of ΔTSH with change in ECG parameters was investigated by linear regression models using fractional polynomials with adjustment for the following baseline characteristics: sex, age, height, weight, hypertension, smoking status, physical activity, chronic heart failure and time between the examinations. Results: Median values of P duration, PR interval, QRS duration, RR interval, and QT interval increased significantly from baseline to follow-up (p<0.001), and whereas the median of QTc interval decreased (p<0.001). ΔTSH was independently associated with the PR interval change (β 25.52, 95%-CI 6.45, 44.60; p=0.009). An increase of one interquartile range of TSH corresponded to an increase of 0.6ms in the PR interval. In contrast, ΔTSH was not associated with changes of any other ECG variables. Conclusion: In this population based sample changes of serum TSH levels were positively associated with changes of the PR interval. Given that PR interval prolongation is a predictor of increased risk for atrial fibrillation this association might point towards increased atrial vulnerability.