Purpose: To investigate the effect of transcytolemmal water exchange on the dynamic contrast-enhanced (DCE) T(1)-weighted MRI of human squamous cell carcinomas of the head and neck (HNSCC).
Materials and methods: Nine patients with HNSCC nodal metastasis underwent pretreatment DCE-MRI with a temporal resolution of 2.5 seconds and a spatial resolution of 1 mm x 1 mm x 5 mm at 1.5T. We used two pharmacokinetic models for data analysis: generalized kinetic model (GKM) without considering transcytolemmal water exchange and the shutter-speed model (SSM), based on a two-site exchange model for transcytolemmal water exchange. The results were compared in three subgroups of voxels in the tumor depending on the level of contrast enhancement.
Results: SSM was found to be a better fit for more than 75% of pixels of all subjects (P < 0.01) in terms of residual size and Bayesian information criterion (BIC). For all three subgroups based on the contrast enhancement, the median K trans values of SSM were 42% to 55% higher than those of GKM and the median upsilon e values of SSM were 116% to 176% larger than those of GKM. The median K trans and upsilon e of two models were found significantly different (P < 0.01). The median tau i measured by SSM were from 211 to 364 msec.
Conclusion: The effect of transcytolemmal water exchange is an important factor that needs to be incorporated for adequate modeling of contrast enhancement dynamics measured by MRI of HNSCC.
(c) 2007 Wiley-Liss, Inc.