Variability in absolute apparent diffusion coefficient values across different platforms may be substantial: a multivendor, multi-institutional comparison study.

PURPOSE To determine whether and to what degree absolute apparent diffusion coefficient (ADC) values vary between different imagers, vendors, field strengths, and intraimager conditions. MATERIALS AND METHODS Informed consent and institutional review board approval were obtained. Diffusion-weighted (DW) images with nearly identical parameters were obtained at 1.5 and 3.0 T from 12 healthy volunteers at seven institutions by using 10 magnetic resonance (MR) imagers provided by four different vendors. ADC maps were generated from isotropic DW maps, and images with a b value of 0 sec/mm(2) were generated by using in-house software. The mean pixel values for the brain tissues were calculated for evaluating the differences among coil systems, imagers, vendors, and magnetic field strengths. RESULTS The absolute ADC values of gray and white matter from the same vendor varied substantially: 4%-9% at 1.5 and 3.0 T. With the exception of one vendor, the intervendor variability at 1.5 T was as high as 7%. Moreover, there was substantial intraimager variability, up to 8%, depending on the coil systems in certain imagers. CONCLUSION There is significant variability in ADC values depending on the coil systems, imagers, vendors, and field strengths used for MR imaging. The relative ADC values may be more suitable than absolute ADC values for evaluating diffusion abnormalities in patients enrolled in multicenter acute ischemic stroke trials.