The emerging plasma biomarker Dickkopf-3 (DKK3) and its association with renal and cardiovascular disease in the general population
Arnold Piek, Leonie Smit, Navin Suthahar, Stephan J. L. Bakker, Rudolf A. de Boer, Herman H.W. Silljé
Abstract
Abstract Dickkopf-3 (DKK3) is an emerging biomarker for cardiovascular disease (CVD) and chronic kidney disease (CKD). Herein, baseline DKK3 plasma levels were measured in 8420 subjects from the Prevention of Renal and Vascular ENd-stage Disease (PREVEND) cohort, a large general population cohort, using enzyme-linked immunosorbent assays. Associations with clinical variables and outcomes were analysed. Median DKK3 level was 32.8 ng/ml (28.0–39.0). In multivariable linear regression analysis, the strongest correlates for plasma DKK3 were age, body mass index and estimated glomerular filtration rate (eGFR). At baseline, 564 (6.7%) subjects had CVD (defined as a myocardial infarction and/or cerebrovascular accident) and 1361 (16.2%) subjects had CKD (defined as eGFR < 60 ml/min/1.73m 2 and/or urinary albumin excretion (UAE) > 30 mg/24 h). Of subjects with known CVD and CKD follow-up status (respectively 7828 and 5548), 669 (8.5%) developed CVD and 951 (17.1%) developed CKD (median follow-up respectively 12.5 and 10.2 years). Crude logistic regression analysis revealed that DKK3 levels were associated with prevalent CVD (Odds ratio: 2.14 [ 1.76–2.61] per DKK3 doubling, P < 0.001) and CKD (Odds ratio: 1.84 [1.59–2.13] per DKK3 doubling, P < 0.001). In crude Cox proportional hazard regression analysis, higher DKK3 levels were associated with higher risk for new-onset CVD (Hazard ratio: 1.47 [1.13–1.91] per DKK3 doubling, P = 0.004) and CKD (Hazard ratio: 1.45, [1.25–1.69] per DKK3 doubling, P < 0.001). However, these associations remained no longer significant after correction for common clinical variables and risk factors, though independently predicted for new-onset CKD in a subgroup of subjects with the lowest UAE values. Together, DKK3 plasma levels are associated with cardiovascular risk factors, but are generally not independently associated with prevalent and new-onset CVD and CKD and only predicted for new-onset CKD in those subjects with the lowest UAE values.