Preoperative perforator imaging is a key component in state of the art tissue transfer (“flap”) reconstructive plastic surgeries ¹. Up to now, MR Angiography (MRA) has not had an important role in this clinical application. Most of the existing efforts in literature focused on contrast enhanced MRA (CE-MRA) techniques, and were considered inferior to Computed Tomography Angiography (CTA) in both spatial resolution and perforator-to-fat contrast ^{2, 3}. Phase Contrast MRA (PC-MRA) can be acquired with very high contrast and resolution ⁴. However, to our knowledge, no effort has ever been made to image perforators with PC-MRA. In this study, we explore the feasibility of visualizing deep inferior epigastric perforators (DIEP) with PC-MRA, and report our experiences in a few initial demonstration cases.

Our perforator PC-MRA sequence was implemented on a 3T Achieva scanner (Philips Healthcare, Cleveland, OH) equipped with a 32-channel cardiac coil. The subjects were scanned in prone position so that the coil physically restricted respiratory motion of the abdominal wall. Only the anterior piece of the coil was used for RF signal reception. A saturation band was placed after the anterior edge of aorta to further reduce motion artifact. The PC-MRA sequence was acquired with 0.5 mm in-plane resolution, which is slightly better than the axial resolution of advanced multi-detector CT scanners. The other scan parameters were: 350x100x200 mm³ Field-Of-View, 1.5 mm slice, TR/TE = 8.7/5.2 ms, flip angle 11°, 1.5x SENSE acceleration, V_max = 15 cm/s, respiratory triggered. In order to get better details in stationary soft tissues, the PC-MRA images are fused with a T2-TSE data set acquired with the following sequence parameters: 420x300x198 mm³ Field-of-View, 0.7 mm in-plane resolution, 5 mm slice with 0.5 mm gap, TR/TE = 838/80 ms, flip angle 90°, 2x SENSE acceleration, respiratory triggered. In consideration of the reviewer’s reading habit, the grayscale of the T2-TSE data was reversed to mimic soft tissue contrast in CTA. Both sequences were respiratory triggered. The total acquisition time was 15-25 minutes.

PC-MRA data were acquired from three female patients (42/50/51 years old) prior to DIEP flap surgery, and compared with their standard-of-care DIEP CTA. The CTA data were acquired on a Definition AS+ CT (Siemens Healthcare, Erlangen, Germany) with a 100 kVp technique after intravenous injection of Omnipaque (GE Healthcare, Princeton, NJ), and reconstructed to 0.75 mm slice thickness. PC-MRA and CTA maximum intensity projection (MIP) were compared at various thicknesses on an Intellispace Portal workstation (Philips Healthcare, Cleveland, OH). Image quality and perforator visualization were evaluated by a senior Plastic Surgeon. Signal-to-noise ratio (SNR) of major perforators and contrast-to-noise ratios (CNRs) in the rectus abdominis muscle and the subcutaneous fat layer were measured. Statistical comparison was performed with the R language (The R Foundation, Vienna, Austria).