Influence of scan duration on dynamic contrast -enhanced magnetic resonance imaging pharmacokinetic parameters for...

Gadolinium-based contrast agent needs time to leak into the extravascular-extracellular space, leak back into the vascular space, and reach an equilibrium state. For this reason, acquisition times of <10 min may cause inaccurate estimation of pharmacokinetic parameters. Since no studies have been conducted on the influence of long scan times on DCE-MRI parameters in brain tumors, the aim of this study is to investigate the variation of DCE-MRI-derived kinetic parameters as a function of acquisition time, from 5 to 10 min in brain tumors.

Fifty-two patients with histologically confirmed brain tumors were enrolled in this retrospective study, and examination at 3 T, DCE-MRI, with scan duration of 10 min, was used for retrospective generation of 6 sets of quantitative DCE-MRI maps (K^trans, V_e and K_ep) from 5 to 10 min. Features were extracted from the DCE-MRI maps in contrast enhancement (CE) volumes. Kruskal-Wallis with post-hoc correction and coefficient of variation (CoV) were used as statistical test to compare DCE-MRI maps obtained from 6 data sets.

p < 0.05.

No differences in K^trans features in CE volumes between different scan durations. V_e, K_ep features in CE volumes were influenced by different data length. The highest number of significantly different V_e and K_ep features in CE volumes were between 5 min and 10 min (p < 0.013), 5 min and 9 min (p < 0.044), 6 min and 10 min (p < 0.040). CoV of K_ep was reduced from 5 min to 10 min, going from highly variable (CoV = 0.70) to mildly variable (CoV = 0.42).

K_ep and V_e were time-dependent in brain tumors, so a longer scan time is needed to obtain reliable parameter values. K^trans was found to be time-independent, as it remains the same in all 6 acquisition times and is the only reliable parameter with short acquisition times.