In humans at least 75% of insulin secretion arises from punctuated insulin secretory bursts.

Detection of insulin secretory bursts in peripheral blood is hampered by hepatic insulin extraction, dilution in the systemic insulin pool, and time-delayed damping of secretory burst amplitude. Previous studies in dogs in vivo and other experiments in vitro have shown that approximately 70% of all insulin is released within distinct insulin secretory bursts. To establish a method for detection and quantification of pulsatile insulin release in humans on the basis of peripheral insulin concentration measurements, we used a high-sensitivity, -specificity, and -precision insulin enzyme-linked immunosorbent assay (ELISA) and optimized an established deconvolution methodology to quantify the frequency, mass, and amplitude of insulin secretory bursts as well as to estimate the relative contribution of pulsatile insulin release to overall insulin secretion. By use of minutely sampled serum insulin concentrations measured by a highly sensitive insulin ELISA and insulin kinetics of 2.8 min (first half-life), 5.0 min (second half-life), and a fractional slow component of 0.28, the deconvolved insulin secretion rates in 20 healthy subjects during glucose infusion (4.5 mg.kg-1.min-1) could be resolved into a series (4.7 +/- 0.1 min/pulse) of approximately symmetric insulin secretory bursts with a mean mass of 87 +/- 12 pmol.l-1 pulse-1 and a mean amplitude (maximal release rate) of 35 +/- 4.7 pmol.l-1.min-1. The relative contribution of pulsatile to overall insulin secretion was 75 +/- 1.6% (range 59-85%). We conclude that in vivo insulin secretion in humans during nominal glucose stimulation consists of a series of punctuated insulin secretory bursts accounting for > or = 75% of total insulin secretion.