Osteoprotegerin improves risk detection by traditional cardiovascular risk factors and hsCRP
- Rasmus Mogelvang1,2,
- Sune Haahr-Pedersen3,
- Mette Bjerre4,
- Jan Frystyk4,
- Allan Iversen3,
- Søren Galatius3,
- Allan Flyvbjerg4,
- Jan Skov Jensen2,3,5
- 1Department of Cardiology, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark
- 2Copenhagen City Heart Study, Bispebjerg University Hospital, Copenhagen, Denmark
- 3Department of Cardiology, Gentofte University Hospital, Hellerup, Denmark
- 4Department of Endocrinology and Internal Medicine, The Medical Research Laboratories, Institute of Clinical Medicine, Aarhus Faculty of Health Sciences and Aarhus University Hospital, Aarhus, Denmark
- 5Faculty of Health Science, Clinical Institute of Surgery and Internal Medicine, University of Copenhagen, Copenhagen, Denmark
- Correspondence to Dr Rasmus Mogelvang, Department of Cardiology, University of Copenhagen, 2.14.1, Rigshospitalet, Blegdamsvej 9, Copenhagen DK-2100, Denmark;
- Received 17 April 2012
- Revised 13 September 2012
- Accepted 14 September 2012
- Published Online First 7 November 2012
Objective To evaluate the association of plasma osteoprotegerin (OPG) to hospitalisation for ischaemic heart disease (IHD), ischaemic stroke and all-cause mortality, and the effect of combining plasma OPG and high-sensitivity C-reactive protein (hsCRP).
Design OPG and hsCRP concentrations were measured at baseline in a large Danish prospective community-based population study.
Setting The 4th Copenhagen City Heart Study.
Participants 5863 men and women aged 20–95 were recruited from the general population.
Main outcome measures Combined end-point of IHD, ischaemic stroke or all-cause mortality.
Results During a median follow-up of 7.8 years (IQR 7.3–8.3), 1270 subjects (21.7%) reached the combined end-point. A twofold increase in plasma OPG was a significant predictor of the combined end-point (univariable HR 1.85, 95% CI 1.75 to 1.96; p<0.001). In a multivariable Cox-regression model containing age, male sex, hypertension, diabetes, hypercholesterolaemia, present or former smoking, glomerular filtration rate, prior IHD, prior ischaemic stroke, hsCRP and plasma OPG, high concentrations of hsCRP and plasma OPG were independent predictors of the combined end-point. The two biomarkers interacted statistically (p<0.001). Compared to low hsCRP and low OPG (n=1927), either high hsCRP or high OPG (univariable HR 2.38, 95% CI 2.02 to 2.80, p<0.001; n=2816), or both high hsCRP and high OPG (univariable HR 5.13, 95% CI 4.29 to 6.13, p<0.001; n=775) conferred increased risk of the combined end-point.
Conclusions OPG is an independent predictor of the combined end-point of hospitalisation of IHD, ischaemic stroke and all-cause mortality. The combination of plasma OPG and hsCRP provides more prognostic information than the individual effect of the two biomarkers.