Part 1 - Coronary angiography with gadofosveset trisodium: a prospective feasibility study evaluating injection techniques for steady-state imaging

Abstract

Background: The purpose of this study was to define an optimal injection protocol for 5-10 min duration navigator-based coronary MR angiography using an intravascular gadolinium-based contrast agent (GBCA), which is better suited for steady-state coronary MR angiography than conventional GBCAs.

Methods: Using projections from pharmacokinetic models of the intravascular concentration of gadofosveset, a dual-injection protocol was formulated and tested on 14 healthy human subjects. Modified Look-Locker inversion recovery (MOLLI) sequences were used for T1 mapping at 3 Tesla to evaluate the concentration of tracer in the aorta over the scanning interval.

Results: Pharmacokinetic models for a bolus plus slow infusion technique at a 5, 10, and 15 min steady state intravascular concentration was compared to single bolus curves. The 70 %/30 % bolus/slow infusion technique resulted in the highest intravascular concentration over a 5 min scan duration. Similarly, the 60 %/40 % bolus/slow infusion technique was projected to be ideal for image acquisition duration of 5-10 min. These models were confirmed with T1 maps on normal volunteers. Arterial-venous mixing of contrast was achieved within 90 s of the beginning of the bolus.

Conclusions: Gadofosveset injection is optimized for the lowest intravascular T1 time for 5-10 min duration MR angiography by bolus injection of 60-70 % of the total dose followed by slow infusion of the remainder of the total dose. This protocol achieves rapid and prolonged steady state intravascular concentrations of the GBCA that may be useful for prolonged image acquisition, such as required for navigator-based coronary MR angiography at 3 Tesla.

Trial registration: ClinicalTrials.gov identifier: NCT01130545 NCT01130545 , registered as of May 25, 2010.

Figures

Fig. 1

Example of axial MOLLI images acquired of the heart. T1 time (in milliseconds) were derived from T1 maps using the MOLLI pulse sequence in the descending aorta (DAo, white arrow) before injection and at subsequent time points after injection. MOLLI – modified Look-Locker inversion recovery, LV left ventricle, RV right ventricle, RA right atrium, DAo descending aorta

Fig. 2

Pharmacokinetic model for the intravascular concentration of gadofosveset. Using various dual injection techniques, a fixed percentage of the total volume is injected as the bolus phase, followed by remainder as a slow infusion over a targeted duration to maintain a steady state concentration. A 100 % bolus is also plotted for reference. The 2-compartment model assumes immediate injection and arterial-venous mixing (equilibrium)

Fig. 3

In vivo behavior of dual-injection technique with gadofosveset. Examples of intravascular T1 time with injection protocols in normal volunteers for the (a) 70 %/30 % bolus/infusion, (b) 60 %/40 %, (c) 100 % bolus, and a composite of all subjects imaged (d). Time zero indicates the beginning of infusion