BACKGROUND: Insulin adsorbs to plastics used for intravenous (IV) tubing. As a result, clinical IV insulin infusion procotols advise an initial priming volume of up to 50 mL, which may be wasteful-especially since most institutions use 100-mL IV solution bags. In this brief report, we sought to determine the optimal priming volume required for clinical IV insulin infusions. METHODS: One hundred units of regular human insulin was dissolved into 100 mL of 0.9% NaCl. Employing a standard polypropylene infusion set, a priming infusion was started. At 10- mL intervals, from 0 to 50 mL, effluent was collected directly into glass vials. After dilution (1:10,000) using a proprietary buffer, insulin concentrations were then measured using a double antibody radioimmunoassay. Twenty individually prepared insulin bags were tested in this manner. RESULTS: Insulin levels without prime were 15.8% [95% confidence interval (CI), 9.1-22.6%] lower than insulin levels following 50 mL of prime (designated as "maximal values"). After a priming volume of 10 mL, insulin adsorption losses fell to a marginally significant 6.6% (95% CI, 0.1-13.1%). Following 20 mL of prime, insulin concentrations were indistinguishable from maximal values (3.4% loss, 95% CI, -0.2% to 7.1%). CONCLUSIONS: For standard IV insulin infusions, a priming volume of 20 mL is sufficient to minimize the effect of insulin adsorption losses to IV lines. Priming volumes exceeding 20 mL are wasteful, increase costs, and generate unnecessary work for nurses and pharmacists.
BACKGROUND:Insulin adsorbs to plastics used for intravenous (IV) tubing. As a result, clinical IV insulin infusion procotols advise an initial priming volume of up to 50 mL, which may be wasteful-especially since most institutions use 100-mL IV solution bags. In this brief report, we sought to determine the optimal priming volume required for clinical IV insulin infusions. METHODS: One hundred units of regular humaninsulin was dissolved into 100 mL of 0.9% NaCl. Employing a standard polypropylene infusion set, a priming infusion was started. At 10- mL intervals, from 0 to 50 mL, effluent was collected directly into glass vials. After dilution (1:10,000) using a proprietary buffer, insulin concentrations were then measured using a double antibody radioimmunoassay. Twenty individually prepared insulin bags were tested in this manner. RESULTS:Insulin levels without prime were 15.8% [95% confidence interval (CI), 9.1-22.6%] lower than insulin levels following 50 mL of prime (designated as "maximal values"). After a priming volume of 10 mL, insulin adsorption losses fell to a marginally significant 6.6% (95% CI, 0.1-13.1%). Following 20 mL of prime, insulin concentrations were indistinguishable from maximal values (3.4% loss, 95% CI, -0.2% to 7.1%). CONCLUSIONS: For standard IV insulin infusions, a priming volume of 20 mL is sufficient to minimize the effect of insulin adsorption losses to IV lines. Priming volumes exceeding 20 mL are wasteful, increase costs, and generate unnecessary work for nurses and pharmacists.