IGFBP-1 predicts all-cause mortality in elderly women independently of IGF-I

• The manuscript describes associations between high and low serum insulin-like IGFBP-1 levels and increased risk of all-cause mortality compared with moderate levels in a cohort of elderly Swedish women followed during 10 years.
• To our knowledge this is the first population based study indicating that both high and low levels of IGFBP-1 may be associated with an all- cause mortality which makes our study represent a meaningful advance in the field.
• No association was found between the levels of IGF-I and mortality in this cohort.

ObjectivePrevious studies on the insulin-like growth factor (IGF) system and mortality have shown ambiguous results. We investigated the association between IGF-I and insulin- like growth factor binding protein-1 (IGFBP-1) with all-cause mortality in an elderly female Swedish population.Design and methodsA prospective cohort study of elderly women (n = 338) aged between 68 and 79 years (mean age 72 years) with a mean follow-up time of 9.9 years. Baseline data in the PRIMOS (Primary Health Care and Osteoporosis) study were collected between 1999 and 2001. Data of risk factors for cardiovascular disease were collected. Death rates were registered from the Swedish Cause of Death register for the period 1999–2009. Cox regression models were used to calculate hazard ratios. IGF-I and IGFBP-1 levels were separately divided into 3 groups (high, medium and low), with cut offs at the 30th and the 70th percentiles.ResultsIn a fully adjusted Cox proportional hazard model, increased risk of mortality was shown for women with high serum levels of IGFBP-1, HR 3.03 (95% CI 1.64–5.63) and also with low serum levels of IGFBP-1, HR 1.98 (95% CI 1.03–3.81), compared to women with moderate levels. No significant association between IGF-I and mortality was observed.ConclusionsHigh and low serum insulin-like IGFBP-1 levels were associated with an increased risk of all-cause mortality in elderly women, compared to moderate levels.