Influence of adenosine on ventricular function measurements as part of a comprehensive stress perfusion magnetic resonance imaging study

Abstract

Background Stress-rest perfusion magnetic resonance imaging (MRI) has become one of the major indications for cardiovascular MRI. Standard multibreath hold short-axis image acquisition is the gold standard for LV function assessment and part of a comprehensive stress-rest perfusion MRI study, but takes up to 10 min. The total scan time could significantly be shortened when cine imaging would be performed between stress and rest perfusion. Purpose To systematically study whether adenosine, albeit its short half-life time, influences LV function measurements in this setting. Material and Methods Thirty-nine patients with suspected and/or known coronary artery disease underwent 3-Tesla adenosine (140 µg/min/kg) stress-rest perfusion MRI. A stack of short-axis slices covering the entire ventricles was acquired before and ≥3 min after adenosine infusion using standard multibreath-hold 2D steady-state free precession imaging. LV volumes (EDV, ESV), EF, and mass were calculated by the slice summation method, and wall motion was assessed using a 4-point scoring scale. Reproducibility was assessed by Bland-Altman statistics. Results EDV, ESV, EF, and mass were not significantly different between the two imaging time points (before versus after adenosine administration) in all patients ( P ≥ 0.346) and patients with impaired LV function ( P ≥ 0.718). Wall motion scores did not differ either ( P ≥ 0.124). Inter-observer variability before ( P ≥ 0.468) and after ( P ≥ 0.451) adenosine infusion was low. Conclusion Short-axis cine imaging for LV function assessment can accurately be performed between stress and rest imaging. These data demonstrate that potential effects of adenosine on LV function are of short duration, and propose a new time-saving imaging protocol without compromising accuracy.