High-density lipoprotein cholesterol (HDL-C) is thought to be atheroprotective yet some patients with elevated HDL-C levels develop cardiovascular disease, possibly due to the presence of dysfunctional HDL. We aimed to assess the metabolic fate of circulating HDL particles in patients with high HDL-C with and without coronary artery disease (CAD) using in vivo dual labeling of its cholesterol and protein moieties. We measured HDL apolipoprotein (apo) A-I, apoA-II, free cholesterol (FC), and cholesteryl ester (CE) kinetics using stable isotope-labeled tracers (D3-leucine and 13C2-acetate) as well as ex vivo cholesterol efflux to HDL in subjects with (n = 6) and without (n = 6) CAD that had HDL-C levels >90th percentile. Healthy controls with HDL-C within the normal range (n = 6) who underwent the same procedures were used as the reference. Subjects with high HDL-C with and without CAD had similar plasma lipid levels and similar apoA-I, apoA-II, HDL FC, and CE pool sizes with no significant differences in fractional clearance rates (FCRs) or production rates (PRs) of these components between groups. Subjects with high HDL-C with and without CAD also had similar basal and cAMP-stimulated ex vivo cholesterol efflux to HDL. When all subjects were considered (n = 18), unstimulated non-ABCA1-mediated efflux (but not ABCA1-specific efflux) was correlated positively with apoA-I production (r = 0.552, p = 0.017) and HDL FC and CE pool sizes, and negatively with the fractional clearance rate of FC (r = -0.759, p = 4.1 × 10-4) and CE (r = -0.652, p = 4.57 × 10-3). Our data are consistent with the concept that ex vivo non-ABCA1 efflux capacity may correlate with slower in vivo turnover of HDL cholesterol moieties. The use of a dual labeling protocol provided for the first time the opportunity to assess the association of ex vivo cholesterol efflux capacity with in vivo HDL cholesterol metabolic parameters.