Injury to the facial nerve has profound repercussions. The facial nerve doesn't have a uniform pattern and its path through the parotid gland can be displaced into unexpected positions by tumors growing within the gland. As the nerve trunk breaks into its terminal fibres they become particularly delicate and it can be difficult to discern within the gland during surgery.

Facial nerve monitoring can only warn proximity to the nerve. There was until recently no method of visualising the actual path of the nerve. For the first time, 3+ Tesla MRI imaging can show us the exact path of the nerve in the parotid.

Hololens2 (Microsoft R) is an augmented reality headset which allows the holographic projection of images onto the wearers view of their surroundings creating a mixed image.

In this study we assessed the ability of 3T MRI and Hololens2 to visualise the path of facial nerve around parotid tumors and the impact on surgical planning.

METHODOLOGY:

15 patients with parotid neoplasms underwent 3T MRI imaging. Each patient was diagnosed with the neoplasm using clinical examination, ultrasound and FNA. The facial nerve was not compromised in any patient pre-operatively. The facial nerve MRI images were subsequently segmented using the 3D slicer software and uploaded into HoloLens2 (Microsoft R) using ApoQlar software. This created a 3D hologram view of the facial nerve and the relations to the surrounding structures (images and video will be presented).

The facial nerve - mainly the main trunk, pes anserinus and origin of main branches- pattern and its relationship to the underlying tumor as predicted on hologram were later compared to intraoperative anatomical findings and the quality of nerve course prediction was determined by A) Predicted contact with nerve b) gross deviation of nerve due to tumour. C) Prediction of nerve involvement and nerve sparing in malignant disease.

RESULTS:

A total of 15 patients with parotid gland neoplasms were included in this study. 12 females and 3 males between the ages of 18 and 66 with a median age of 42 with different parotid neoplasms; 10 benign neoplasm and 5 were malignant neoplasms. The pre-operative 3D hologram predicted nerve displacement or direct contact by the tumor in 7 cases. 100% of these cases had identical intra-operative anatomy (images and video will be presented). The rate of post-operative facial nerve injury was only 1 injury. Age, sex, tumor type, size, site did not make a change in the predicted images of the neoplasms.

CONCLUSION:

This is preliminary data on a limited case series which shows great prospect of accurate facial nerve visualisation using advanced MRI systems. Augmented reality allows surgeon friendly view of the 3D anatomy to guide surgery.

Although the accuracy of prediction depends on radiology interpretations of MRI, in our experience the accuracy easily meets the tolerance required for parotid surgery.

This allows surgeons to be forewarned of unusual configuration of the facial nerve and also opens up the opportunity to plan and execute operations in a way that minimizes nerve contact.