Reproducibility of quantitative dynamic MRI of normal human tissues

The aim of the study was to establish the normal ran-e and to evaluate the reproducibility of dynamic contrast enhanced MRI (DCE-MRI) parameter estimates in normal human pelvic tissues. Nineteen patients with prostate cancer. undergoing androgen deprivation treatment. had paired DCE-MRI examinations of the pelvis using spoiled gradient-echo sequences. Quantitative enhancement parameters were calculated for each examination: transfer constant (K-trans). leakage space (v(c)) and maximum contrast medium accumulation (MCMA) of pelvic muscles bone marrow and fat. Descriptive and reproducibility statistics were calculated: within-patient Standard deviation (wSD), repeatability and within-patient coefficient of variation (wCV). The femoral head and ischiorectal fat showed large numbers of non-enhancing pixels (81 and 88%. respectively). The ischial bone marrow had the highest values of kinetic parameter estimates (K-trans 0.554 min(-1), v(c) 18.5% and MCMA 0.164 mmol/kg). Muscle parameters values were lower (K-trans 0.126-0.137 min(-1), v(c) 10.6-11.5% and MCMA 0.077-0.086 mmol/kg). The mean difference between paired examination,, was not significantly different from zero for any parameter. v, and MCMA had the lowest wCV (between 19 and 29%). For individuals. a log(10) K-trans chance of approximately 0.90 in muscles and 0.52 in the ischium would be statistically significant. The corresponding absolute changes for v(c) are 6.7% in muscle and 13.6% in the ischium. For a group of 19 patients, small changes are statistically significant (muscle log(10) K-trans 0.208 and v(c) 1.5% and ischium log(10) K-trans 0.123 and v(c) 3.1%). Fat and the femoral head are unreliable tissues from which to obtain kinetic parameter estimates due to poor enhancement. v(c) and MCMA have smaller coefficient of variation than K-trans in muscles and ischium. Reproducibility studies of normal and pathological tissues should be incorporated into clinical research protocols that measure treatment effects by DCE-MRI techniques. Copyright (C) 2002 John Wiley Sons, Ltd.