Myocardial flow is closely associated with cardiac function¹. However, it is unclear how the perfusion changes at the tissue level in response to hemodynamic alteration, such as in patients undergo chemo therapies. In this study we sought to investigate the relationship of myocardial resting perfusion in patients in chemo therapy versus those of age matching normals using quantitative contrast 1^st pass perfusion by MRI.

Cardiac function and regional myocardial resting dual bolus perfusion images were analyzed in 5 age-matching normal subjects (age: 57.2±5.5) and 6 patients (age: 55.7±14.8) who underwent chemo on a 1.5 T MRI scanner. The study was approved by the hospital IRB and written consents were collected from all subjects.

Breath holding first-pass perfusion studies were performed at rest on a 1.5 T Siemens Sonata or Avanto scanner (Siemens Medical Solutions, Malvern, PA), with a CP body array flex coil. After scout and cine LV functional imaging, perfusion imaging was obtained with data acquisition starting two cardiac cycles prior to administration of contrast agent. Gadodiamide (Omniscan, Amersham Health Inc., Princeton, NJ) was injected first at a dose of 0.005 mmol/kg (diluted to maintain the same volume as the standard dose), then at a dose of 0.05 mmol/kg bodyweight, followed immediately by a 15-20 cc saline flush on both injections. Perfusion imaging was performed using a partial Fourier saturation recovery steady state free precession (SSFP) sequence to acquire 4 slices per heartbeat over 50 heartbeats for an imaging time of 160 ms per slice. One short axis slice at mid ventricle and three rotational long axis slices (as horizontal long axis (HLA), vertical long axis (VLA), and left ventricular outflow tract (LVOT) views of the left ventricle, all perpendicular to a left ventricular short axis) were prescribed. The typical parameters were as follows: TR/TE/TI/FA = 2.9ms/1.3ms/90ms/50°, data matrix 90×192, and voxel size 1.9×2.8×8 mm³. All image acquisition was ECG triggered, and the non slice-selective 90° saturation pulses for each slice.

Using MASS (Medis, Leiden, the Netherlands) software, the myocardial contours were manually drawn on an image showing good contrast and then automatically propagated to each time point. The myocardium were divided clockwise into 6 equal segments and the mean signal intensities of all pixels in each myocardial segment at every time point were recorded. A custom developed program read the segmental signal intensity saved from MASS and calculated absolute perfusion according to the quantitative method using Fermi model as deconvolution technique. Myocardial perfusion was determined as the average of all 24 regions for each subject and 2 tailed student t-test was used for statistical analysis.

The patients all have normal ejection fraction (LVEF was 54.3±5.4%). The myocardial perfusion was significantly lower in patients 0.51±0.16 ml/min/g vs 0.77±0.24 ml/min/g in the age matching normals (p<0.05), as shown in Fig 1.

Reduced myocardial resting perfusion was associated with patients underwent chemo therapy even when their LVEFs are normal. Further studies are needed to fully understand the etiology of this change. This technique may potentially provide an early indication of cardiotoxicity.