Background and aimsEndostatin, generated from collagen XVIII, and endorepellin, possess dual activity as modifiers of both angiogenesis and endothelial cell autophagy. Plasma endostatin levels are elevated in a large number of diseases, and may reflect endothelial cell dysfunction. Few data on endostatins are available for patients with chronic kidney disease (CKD). We tested whether serum endostatin values are predictive for all-cause mortality and cardiovascular events (CVEs) in a CKD population. Materials and methodA total of 519 CKD pre-dialysis patients were included. Baseline plasma endostatin levels were measured in all patients. All included patients were followed-up (time-to-event analysis) until occurrence of death, fatal or nonfatal CVEs. Fatal and nonfatal CVE including death, stroke, and myocardial infarction were recorded prospectively ResultsThe mean age of the patients was 52.2±12.3years. There were 241 (46.4%) males, 111 (21.4%) had diabetes, 229 (44.1%) were smokers and 103 (19.8%) had a previous CVE. After a median follow-up of 46months, 46 patients died and 172 had a new CVE. In the univariable Cox survival analysis, higher endostatin levels were associated with a higher risk for both outcomes. However, after adjusting for traditional (age, gender, smoking status, diabetes, systolic blood pressure, HDL and total cholesterol) and renal-specific (eGFR, proteinuria and hsCRP) risk factors, endostatin levels remained associated only with the CVE outcome (HR=1.88, 95% CI 1.37–2.41 for a 1 SD increase in log endostatin values). ConclusionEndostatin levels are independently associated with incident CVE in CKD patients, but show limited prediction abilities for all-cause mortality and CVE above traditional and renal-specific risk factors.