Background: Insulin-like growth factor 1 (IGF1) plays a central role in metabolism and growth regulation. High IGF1 levels are associated with an increased risk of cancer, low IGF1 levels with increased risk for cardiovascular disease. We hypothesized that the relationship between circulating total IGF1 levels and mortality is U-shaped.
Methods: We performed a systematic review searching PubMed, EMBASE, Web of Science, Cochrane Library (1985 to September 2010) and undertook a random-effects meta-analysis of population-based studies assessing the association between IGF1 and mortality. Mortality in subjects with low or high IGF1 levels was compared with mid-centile reference categories. A dose-response meta-regression model was constructed to quantify the effect of IGF1 on mortality across the full IGF1 range.
Results: Twelve studies with 14 906 participants were included in the analysis. All-cause mortality was increased in subjects with low as well as high IGF1 categories, hazard ratio (HR) 1.27 (95% CI 1.081.49) and HR 1.18 (95% CI 1.041.34), respectively. Dose-response meta-regression confirmed a U-shaped relation of IGF1 and all-cause mortality (P=0.003). The predicted HR for the increase in mortality comparing the 10th IGF1 to the 50th percentile was 1.56 (95% CI 1.311.86); the predicted HR for the increase in mortality comparing the 90th to the 50th percentile was 1.29 (95% CI 1.061.58). A U-shaped relationship was present for both cancer mortality and cardiovascular mortality.
Conclusions: This meta-analysis showed that both low and high IGF1 concentrations are associated with increased mortality in the general population.
Rotterdam, The Netherlands
30 Apr - 04 May 2011
European Society of Endocrinology