The classical approach to parotid gland surgery was described more than one hundred years ago; it entails the identification and preservation of the facial nerve, and this technique should not change, regardless of whether the surgeon is using nerve monitoring or not. [1, 2]
Table 2 depicts a series of similarities and disparities between the recurrent laryngeal and facial nerve monitoring process. The recurrent laryngeal nerve (RLN) is a branch of the Vagus nerve, though the nerve might also divide itself into at least two or more sensitive ramifications. The facial nerve is a terminal nerve; as the nerve exits through the stylomastoid foramen, it gives origin to motor and sensory branches (posterior belly of digastric, stylohyoid, superior auricular, etc.) continued by the main trunk (MT) dividing shortly into a Temporozigomatic or superior branch (SB) and a Cervicofacial branch or inferior branch (IB), which – at the same time – divides into several tiny motor ramifications as they travel in an external to a medial fashion to innervate the facial musculature. [22–29]
The neuro- monitoring technology must be seen as an aid to try to ensure that no nerve damage will be caused and to allow for an early identification of a potential nerve injury at any time during the surgical process. Regarding the mechanism of injury to the facial nerve, similar to what has been described for the Vagus or RLN, lesions can be classified as general or diffuse; for example, accidentally traumatizing the main trunk by heat, section, pressing, etc., causing a complete face paralysis and loss of signal, or just punctually referring to a specific anatomical site where a particular nerve branch was affected. Even though there is no doubt that the technology certainly helps to expedite and look out for the nerve in high-volume lesions – tumors – it surely speeds up the process of finding the facial nerve main trunk. [8–12]
Loss of Signal
It has been widely described that when there is trauma to the recurrent laryngeal nerve during a thyroidectomy done under IONM, changes – such as an increase in latency (time) and reduced amplitude (energy) that may worsen if the trauma continues – are noted on the monitor, constituting what has been named loss of signal. Even though the nerve might seem intact, in reality – and thanks to the information provided by the monitoring process – it is not functional. A consensus was agreed upon by many authors, that if – at some point during a total or hemithyroidectomy – a recurrent laryngeal nerve signal drops more than fifty percent of its original maximum stimulation data in its amplitude along with a prolongation of the latency time, the term “Loss of signal (LOS)” is applied. At this point there is still a chance that the nerve might recover, depending on the severity of the trauma. The surgeon should decide if – in the presence of such event – a proven loss of signal that does not improve in a prudent period of time (usually 20 to 25 minutes) and in the face of a scheduled total thyroidectomy, said procedure should be stopped and a completion thyroidectomy done at a later date, waiting for the nerve to recover within the coming weeks, while avoiding at that moment the potential risk of a tracheostomy if the contralateral lobe is indeed approached and the nerve is injured. Without a doubt, this is certainly one of the benefits that intraoperative nerve monitoring offers to both the surgeon and the patient nowadays. A quite surprising finding of the study was to realize that after exploring the research engines, three articles briefly refer to the process of facial nerve injury and amplitude and latency changes respectively; however, there is no clear instruction in any of the articles on what to do if a surgeon performing a classical superficial parotidectomy under nerve monitoring is faced with an intraoperative facial nerve loss of signal. [6, 20, 23]
Even though that there are significant differences in each process of stimulation of the recurrent laryngeal and the facial nerve, a precise directive to advise what action to take if the parotid gland surgeon is faced with a loss of signal has not yet been described. In our case and based on the analysis of the experience and the data described here, what we do nowadays – in the event that a particular branch has lost its signal, meaning that a considerable difference is encountered in the obtained numbers/nerve stimuli feedback comparing the initial response once the superior or inferior branches are dissected due to trauma – the surgeon must continue and finish the surgery dissecting the remaining branches. Not only does it seem like the logical thing to do, but also – even if the main trunk is injured at the beginning of the procedure and a dramatic drop in the facial nerve amplitude is observed along with a disturbing prolongation of the latency, which in our patients was not the case (Table 1) – the surgeon will complete the surgery removing the entire gland, avoiding, if possible, any further or severe manipulation of the nerve or its branches. Also, if severe monitoring changes are observed in the nerve pattern for a particular branch, the surgeon might “leave this ramification to rest for a while” and dissect the other branch, giving that ramification the potential chance to recover as has been described for the recurrent laryngeal nerve. Besides, different nerve branches might have a system of interconnecting anastomosis that may “overpass” that trauma area and continue to transmit the stimuli, preserving the harmony to the face. This retrospective study did not have the tools, nor the complete data, to answer the important number to suggest at what point of the Amplitude reduction or the prolongation of the Latency, a surgeon should consider a true loss of signal during a facial nerve dissection. Undoubtedly, this will require a larger prospective study (Fig. 1). [30–32]
The Frontalis and Mentalis branches were chosen as the ones with more reliable information, representing the superior and inferior principal branches respectively, especially regarding the Amplitude as detailed in Table 1. In the complete information available, the data collected mostly showed an increase in the numbers obtained for the superior branches; for the marginal mandibular nerve post operative paresis, a slight decrease in the amplitude was noticed in five patients that had no permanent problem with their lower lip movement.
Our results prompted us to recognize, analyze and revise our strategy for every parotidectomy done under IONM in our department, proposing an initial guideline to follow, as shown in Table 3. It is worth stimulating the main trunk at the beginning of the procedure, observing, and saving the result of the four targeted muscle zones with their amplitude and latency responses in the monitor and comparing them to the following ones. If a surgeon wants to be more precise on the final integrity and functionality of the nerve, the feedback from the Superior and Inferior Branches or any particular branch might also be obtained if they were individually stimulated and saved early in the case. (Figs. 2 and 3)
Indeed, even though more prospective studies are needed, this is a lengthy process of building a guideline for facial nerve monitoring during parotid gland surgery that is not only accepted by most surgeons, but also offers practical and reliable information, thus ensuring the same protocol be used for each patient, just as it has been described and followed by thyroid surgeons with recurrent laryngeal nerve neuromonitoring. [10–14, 32]
Table 3 Proposed guideline to stimulate the facial nerve during a parotidectomy*
1. Preoperative: Inform the patient of the diagnosis and indication for surgery under intraoperative nerve monitoring. Do not present the technology as a warranty that the nerve will not be injured.
2. Anesthesia: Intubation with a short acting muscle relaxant; no other relaxation agents should be used throughout the procedure.
3. Four face electrodes (Frontalis, Orbicularis Oculi, Orbicularis Oris and Mentalis) are sub-dermally inserted in the face-targeted muscles, cleaning the area with alcohol swabs. Grounding and receiving electrodes are placed in the shoulder area. All six electrode ends are connected, matching the color to the provided receiving console that integrates all systems and transmits the electromyographic (EMG) response to the monitor.
4. Landmarks: After a classic approach is done, follow the known anatomical landmarks to find the facial main trunk. Once the tragal pointer is in view, use the nerve stimulator probe set at 1 to 1.5 milliamps (mA) to do “mapping” or a blind search for the main trunk; at this point, a general response of the nerve should be obtained, indicating the proximity of the nerve.
5. Main trunk: Once the trunk is identified, reduce the energy to 0.5 to 0.8 mA; dissect and follow the nerve up to the Pes Anserinus. Save a couple of responses of the main trunk stimulation process for a subsequent end-of-procedure comparison. (Fig. 2)
6. Dissect and expose the Superior (SB – Temporozigomatic)) and Inferior (IB – Cervicofacial) branches. Stimulate and save branch responses at 0.5–0.8 mA. (Fig. 2)
7. From this point on, the nerve branches should be stimulated at 0.5 mA or with even less energy. Each branch response should be saved.
a. At any given time during the dissection, the surgeon can stimulate a branch if faced with a suspected trauma or dramatic change in the response. If, indeed, a reduced signal – or even a loss of signal is observed – he should try to identify and correct the cause; leave said branch to rest a while, dissecting the next ramification and coming back at a later time to see if it has recovered. Even though the nerve did not recover, the procedure should finish dissecting the affected branch as gently as possible.
8. Once the gland is removed, each dissected branch and main trunk is stimulated, saved, and compared with its initial response. The system will provide the surgeon with a PDF form to save. (Figs. 2 and 3)
*Using the Nim 3 Medtronic Nerve Stimulation System (Jacksonville, FL USA)