Increased Hindbrain Motion in Chiari Malformation I Patients Measured Through 3D Amplified MRI (3D aMRI)

Abstract

AbstractChiari Malformation type 1 (CM-I) is a neurological disorder characterized by morphological defects such as excessive cerebral tonsils herniation and vast associated symptomatology. Given that these structural defects cannot explain the underlying symptomatology, and might result in misdiagnosis, in this work, we studied the brain’s intrinsic motion to better understand the mechanisms of CM-I. We acquired 3D cine MRI of 14 healthy and 14 CM-I subjects and used 3D amplified MRI (3D aMRI) to visualize and measure the brain’s intrinsic motion during the cardiac cycle. We observed that the regional brain motion in CM-I was significantly higher than the healthy subjects, with anterior-posterior (AP) and superior-inferior (SI) displacements in cerebral tonsils and medulla having the highest differences between the healthy and CM-I (∽45% ∽73% increased motion in the CM-I group). The motion of the cerebellum, and brainstem in AP directions (∽42% and ∽31% increased motion in the CM-I group, respectively), followed by the motion of the cerebral tonsils and medulla in medial-lateral (ML) directions were other significant differences found between the two groups (∽16% increased motion in the CM-I group). Additionally, for the CM-I subjects, we measured morphological parameters including the tonsil herniation, ratio of neural tissue in the foramen magnum, and 4thventricle volume. We then used the morphometrics and brain’s intrinsic motion to analyze the symptomatology of the CM-I patients and their surgical outcomes. Interestingly, we found the ratio of neural tissue in the foramen to be directly correlated with the SI motion of the tonsils (r= 0.58). We also found the tonsil herniation to be directly correlated with the AP motion of the tonsils (r= 0.61), and AP and ML motions of the medulla (r= 0.66, andr= 0.57). Additionally, we found the ML motion of the tonsils to be the only indicator of the surgical outcome (AUC= 0.95), in which subjects with higher motion had an improved outcome. Although we did not observe a significant correlation between the brains motion and morphometrics on the CM-I symptoms due to our small sample size, illustrative cases increase our hope for the development of a future tool based on the brain biomechanics.