129 Automatic Maximal Resolution Heart Rate Adaptive Stress Perfusion Imaging: Cardiovascular Magnetic Resonance Study at 3.0T.

INTRODUCTION: Myocardial perfusion cardiovascular magnetic resonance (CMR) with vasodilator stress has high diagnostic accuracy for the detection of coronary artery disease (CAD). Current CMR perfusion pulse sequences use fixed acquisition parameters designed to acquire at least three slices heart beat and optimised for the heart rates that occur during pharmacological stress.In patients with lower heart rates, there can therefore be a significant amount of unused imaging time (Figure 1). In patients with higher heart rates, acquisition with fixed parameters may not be possible at every heart beat. A more flexible acquisition scheme could optimise acquisition parameters specifically for each patient and heart rate with potential improvements in image quality or temporal resolution. The aim of this study was to assess the feasibility of a perfusion pulse sequence which adapts to the heart rate, maximising imaging time and acquired in-plane spatial resolution. heartjnl;100/Suppl_3/A75-a/F1F1F1 Abstract 129 Figure 1 A: Fixed resolution pulse sequence and B: Adaptive resolution pulse sequence with acquisition duration maximised for heart rate. Blue: Pre-pulse; PD - Preparation pulse Delay time; k0: true centre of K space

METHODS: A new perfusion method, which automatically adapts the acquisition duration to maximise spatial resolution whist maintaining 3 slice imaging at every heart beat was developed (Figure 1). Ten healthy volunteers (mean age 21.5 ± 1.3 years) and two patients (mean 70 years) underwent adenosine stress and rest perfusion CMR on two separate occasions using a 3.0 T whole body scanner and dedicated 32 channel cardiac coil. On one occasion, a conventional "fixed" resolution perfusion sequence was used (3 short axis slices, SENSE acceleration, acquired in-plane resolution of 2.42 × 2.4 mm). On a second occasion, the adaptive method was used. heartjnl;100/Suppl_3/A75-a/F2F2F2 Abstract 129 Figure 2 Example perfusion first pass perfusion mid-slice images A: standard optimised fixed resolution protocol with acquired resolution of 2.42 x 2.42 mm and B: adaptive resolution protocol maximising imaging time with acquired resolution of 1.74 x 1.74 mm revealing less dark rim artefact Images were evaluated blinded to the sequence and image quality graded (1 = high, 2 = adequate, 3 = poor, 4 = unusable) and DRA was measured with electronic callipers at standardised windows settings.

RESULTS: Adaptive perfusion CMR was feasible in all subjects. Mean stress heart rate (HR) was 89 ± 11 in the fixed resolution group and 90 ± 18 in the adaptive resolution group. The standard perfusion sequence acquired in-plane resolution was 2.42 mm(2) and the mean HR adaptive sequence resolution was 1.91 × 1.91 mm ± 0.41 (range 1.53-2.89)(p = 0.001). In two cases the stress HR was too high for alternate R-R interval imaging with the fixed resolution sequence resulting in alternate heart beat imaging. This did not occur with the adaptive sequence which adjusted the resolution was adapted (to 2.84 and 2.89 mm(2) respectively). The mean DRA width was 3.0 ± 0.6 mm (95% CI: 2.57-3.51) with the standard perfusion sequence and 2.1 ± 0.6 mm (95% CI: 1.65-2.57) with the adaptive sequence (p < 0.001) (Figure 2). There was no statistical difference in median image quality score.

DISCUSSION: Optimising the use of available imaging time during CMR myocardial perfusion imaging with heart rate adaptive shot acquisition duration is feasible and improves the acquired resolution and reduces dark rim artefact whilst maintaining image quality. The effect on diagnostic performance of perfusion CMR should be investigated.