A pilot study of leakage and compartmentalization of the contrast agent Ablavar

Methods Simulation study: In vitro studies have shown Ablavar to be 80-90% bound to albumin, with up to 10 fold relaxivity difference between bound and free fractions. We performed simulations to assess the effect of extravasation of the free fraction on signal. Vascular fraction measurements were simulated assuming slow two-compartment exchange for different contrast agent injection concentrations, binding fractions, bound and free relaxivity, and true vascular fractions. Volunteer study: five healthy volunteers (4 males, average age 33) underwent T1 measurement pre and 2 minutes post administration of five injections of 0.006 mmol/kg (a fifth of a single dose) Ablavar. Steady-state T1 was mapped using a cardiac gated Modified Look Locker Inversion Recovery (MOLLI) pulse sequence (slice thickness 8 mm, FOV 300 x 400 mm, matrix 256 x 172, effective TI 100 ms). Image Processing: Maps of vascular fraction were calculated from signal difference maps, according to a slow water exchange model. Fv was measured in the myocardium, dome of the liver, and skeletal muscle visible on the short axis MOLLI images, and was corrected for Ablavar extravasation based on the leakage study. The true fv and exchange rate of water protons was determined by chi square minimization between data and simulations of the effect of water exchange on fv according to the two compartment water exchange model presented by Donahue et al. (1996).