American Journal of Neuroradiology, Vol 18, Issue 6 1115-1125, Copyright © 1997 by American Society of Neuroradiology
ARTICLES |
Contrast-enhanced MR of the facial nerve in patients with posttraumatic peripheral facial nerve palsy
S Sartoretti-Schefer, M Scherler, W Wichmann and A Valavanis
Institute of Neuroradiology, University Hospital of Zurich, Switzerland.
PURPOSE: To estimate the value of noncontrast and contrast-enhanced T1- weighted MR imaging in detecting the underlying mechanisms of injury and regeneration in immediate- or delayed-onset posttraumatic peripheral facial nerve palsy. METHODS: Twenty-four patients with posttraumatic peripheral facial nerve palsy were examined on a 1.5-T MR imaging unit with precontrast and postcontrast T1-weighted spin-echo and gradient-echo sequences. RESULTS: Abnormal enhancement of the distal intrameatal nerve segment was visible in 92% of the patients up to 2 years after their initial trauma. A hematoma within the geniculate ganglion was seen in 33% of the patients with a longitudinal fracture. The greater superficial petrosal nerve (in 32% of patients) and the geniculate ganglion (in 48% of patients) were thick and intensely enhancing. Hematoma within the cochlea/vestibule or enhancement of the cochlea/vestibule and the vestibulocochlear (eighth) nerve was observed in transverse fractures. CONCLUSION: MR images can show long-lasting abnormal nerve enhancement, especially in the distal intrameatal nerve segment, related to the long-lasting breakdown of the blood/peripheral nerve barrier associated with nerve degeneration and regeneration after traumatic stretching of the greater superficial petrosal nerve. Additionally, intraoperatively observed perineural and intraneural scar formation leads to thickening and intense enhancement of the affected nerve segments on MR images. A hematoma in the region of the geniculate ganglion can be seen in some but not all patients. Associated damage of the inner ear structures in patients with transverse fractures is also visible on MR images.