Objective: Previous studies reported associations between insulin-like growth factor I (IGF-I) serum concentration and cardiac morbidity and mortality, but the association between IGF-I serum concentration and cardiac repolarization has not been investigated in a population-based study so far. Therefore, we analyzed the impact of IGF-I concentrations on QTc, QT and RR intervals in two population based studies, The Study of Health in Pomerania (SHIP) and the Rotterdam Study. Design: 457 individuals from SHIP and 155 individuals from the Rotterdam Study older than 55 years and without cardiovascular diseases and a left ventricular hypertrophy were investigated. IGF-I was determined by automated two-site chemiluminescence immunoassays and electrocardiograms were recorded by an ACTA electrocardiograph at a sampling frequency of 500 Hz. The association of IGF-I with QTc, QT and RR intervals was investigated by multivariable linear regression analyses adjusted for age, gender, diabetes mellitus, myocardial infarction, hypertension, body mass index, serum potassium and calcium in both studies separately and in pooled analysis. Results: There were no significant associations between log-transformed IGF-I and QTc interval in the single populations, whereas a significant inverse association was detectable in the pooled population (beta, -15.6; 95%-confidence interval, -25.7, -5.5). The QTc interval was significantly higher in the first tertile of IGF-I compared to the third tertile (beta, 5.4; 95%-confidence interval, 9.5-1.3) in the pooled analysis. Conclusion: The inverse association between IGF-I serum concentrations and QTc interval in our study is suggestive of a higher risk for cardiac arrhythmias and thus might provide additional evidence for increased cardiovascular mortality in subjects with low IGF-I secretion. (C) 2011 Elsevier Ltd. All rights reserved.