Visualization of the morphology and pathology of the peripheral branches of the cranial nerves using three-dimensional high-resolution high-contrast magnetic resonance neurography

Highlights

• HRHC-MRN provide a 3D view of the extracranial nerves via good background suppression.

• HRHC-MRN is a reliable nerve-selective imaging method for assessing nerve morphology.

• HRHC-MRN has potential for assessment of extracranial nerve pathology.

Abstract

Purpose

To assess the feasibility and advantages of high-resolution high-contrast magnetic resonance neurography (HRHC-MRN) for visualizing the morphology and pathology of the peripheral branches of cranial nerves.

Materials

cMRN (3D SPACE STIR sequence) and HRHC-MRN (contrast enhanced 3D SPACE STIR sequence) were performed at 3 T MR unit on 16 volunteers and 12 patients with head and neck tumors. Quantitative measurements such as SNR, CNR and CR were calculated. Three readers evaluated the continuity of the 10 major peripheral branches of cranial nerves using a 5-score scale (scores 0–4). Interobserver variability was tested. Quantitative measurements and scores were compared between cMRN and HRHC-MRN. The imaging features of the nerve pathology were analyzed.

Results

The CRs of nerve to bone marrow, nerve to muscle, and nerve to gland were significantly higher with HRHC-MRN than with cMRN (P = 0.014, P = 0.02, P <0.001, respectively). The scores of all nerve trunks were significantly higher with HRHC-MRN than with cMRN (all, P < 0.001). For all nerves on HRHC-MRN, the interobserver consistency was excellent across the three readers (all κ > 0.8). The scores of the inferior alveolar nerve, hypoglossal nerve, lingual nerve, facial nerve, infraorbital nerve, masseteric nerve, glossopharyngeal/vagus nerve, supraorbital nerve, auriculotemporal nerve and buccal nerve were 3.95, 3.77, 3.63, 3.25, 3.15, 3.04, 3.04, 2.87, 2.79, 1.88, respectively.

Conclusion

HRHC-MRN provides improved visualization of the peripheral branches of cranial nerves and is a promising nerve-selective imaging method for evaluating cranial nerve morphology and pathology.

Introduction

Peripheral branches of the cranial nerves are vulnerable to various pathological processes such as tumors, inflammation, and trauma. These processes induce certain symptoms and affected a patient’s treatment plan and prognosis [1,2]. Nerve involvement impact the prediction of outcome for patient with head and neck tumors. In a previous study, 18F-Fludeoxyglucose positron emission tomography (18F-FDG PET) was reported 100 % detection rate (n = 28/28) of perineural spread (PNS), and increased cranial nerve SUV_max was significantly associated with poorer overall survival [3]. Routine MRI showed prognostic value of cranial nerve involvement in the T4a/b subclassification of nasopharyngeal carcinoma (NPC) [4]. However, it is still difficult to achieve highly selective neurography using traditional method because of the poor contrast between peripheral branches of the cranial nerves and soft tissues [5]. A nerve-selective imaging method is supposed to improve the detection and evaluation of cranial nerve involvement. Therefore, advanced magnetic resonance neurography (MRN) should be developed for acquiring anatomic details of cranial nerves.

Visualizing the peripheral branches of the cranial nerves is limited because of their widespread distribution, tortuous segmentation, complicated surrounding structures, and defects in high tissue contrast imaging when using traditional magnetic resonance (MR) sequences [1,2,6,7]. With the rapid development of magnetic resonance imaging (MRI), different high-resolution sequences have been developed to depict the structures of the cranial nerves [[8], [9], [10], [11], [12]]. However, in these studies, only short portions of the large branches of certain nerves [e.g., inferior alveolar nerve (IAN)] could be visualized simultaneously, whereas certain tiny tortuous peripheral branches of the cranial nerves could not be visualized clearly. The cranial nerves and surrounding structures are complex. Thus, an accurate assessment of the relationship between lesions and nerves require substantial training, even for radiologists [13]. Current high-resolution sequences have to be reconstructed for each nerve to acquire integrated image [13,14]. Therefore, a 3D high resolution high contrast MRN that depict all major branches of the cranial nerves simultaneously will be very helpful, even for clinicians without special training, to evaluate the condition of nerve involvement.

In this study, a novel MR neurography was applied to image the peripheral branches of the cranial nerves. It was based on the three-dimensional (3D) turbo spin echo with variable flip-angle (SPACE) short-tau inversion recovery (STIR) sequence with the T2-shortening effect of a paramagnetic contrast agent for robust background suppression. We aimed to assess the feasibility and advantages of the three-dimensional high-resolution high-contrast magnetic resonance neurography (HRHC-MRN) for visualizing the morphology and pathology of the peripheral branches of the cranial nerves.