Imaging of lamination patterns of the adult human olfactory bulb and tract: in vitro comparison of standard- and high-resolution 3T MRI, and MR microscopy at 9.4 T.

PURPOSE: Neurological and smelling disorders (e.g. Alzheimer's disease, sinonasal disease) negatively affect the microstructural integrity of the olfactory bulb's (OB) cortical layers. Recovery processes depend on active restoration of this microstructural integrity enabled by neuroneogenesis in the OB. The aim of this study was to evaluate lamination patterns of the OB and adjacent tract (OT) using high resolution MRI at 3 Tesla (T) as well as MR microscopy at 9.4 T in comparison with histological sections. MATERIAL AND METHODS: Twenty-four human OBs were imaged in vitro using standard (2mm slice thickness) and high resolution (0.2mm slice thickness) T1w and T2w MR imaging at 3T. Based on signal intensity differences, the number of OB layers and the OB lamination patterns were assessed by two observers in consensus. Results were compared using Wilcoxon test. Signal intensity profiles were compared to reference Nissl stained histological sections and imaging results of MR microscopy. OT lamination patterns were assessed and different configurations of cross sectional areas were compared to macroscopic results and OB/OT lamination patterns. RESULTS: Standard resolution at 3T identified three layers in 8.3%, two layers in 83.3%, and one layer in 8.3%. High resolution at 3T (4 layers in 91.7%, 3 layers in 8.3%) significantly performed better (P<0.001). Signal intensity profile analysis at 3T and 9.4 T (yielding up to 6 different signal intensities) correlated with histological sections and enabled quantitative evaluation of OB lamination patterns. 3T MRI of the OT revealed two separate signal intensities in T2w in 73%, a hyperintense core and a hypointense sheath, and the number of OT signal intensities positively correlated (ρ=0.541, P=0.006) with the increasing complexity of the OTs' cross sectional area configurations. Additionally, cross sectional area configurations correlated with macroscopic results (ρ=0.558, P=0.002) and OB lamination patterns (ρ=0.446, P=0.022). CONCLUSIONS: 3T MRI and MR-microscopy indicate the possibility to identify the lamination pattern of the human OB/OT and to reflect the histological status. If further development will be able to provide technical equipment that complies with the condition of human in vivo high resolution imaging achieving a good enough signal noise ratio, the method of signal intensity profile analysis could prospectively enable scientists to assess the OB's microstructural status in neurological and smelling disorders.