The aim of this study is to compare two distinct modes of NIM Vital application in thyroid surgery: NerveTrend vs. NerveAssure mode with respect to prevalence of early postoperative RLN injury.

The hypothesis explored in this study is that NerveTrend mode may be not inferior than NerveAssure mode in intraoperative identification of impending neural injury and in prognostication of postoperative glottis function in monitored bilateral thyroid surgery. Hence, NerveTrend mode may be considered a bridge between i-IONM and NerveAssure modes, and particularly in health care environments with limited financial resources it can be considered a substantial step forward representing a modern alternative to the NerveAssure technique.

A prospective, randomized study with 2 arms: NeveTrend vs. NerveAssure mode (n=132 patients and 264 nerves at risk, each).

The primary outcome measure is prevalence of recurrent laryngeal nerve (RLN) injury (%) on postoperative day 1 assessed by direct laryngoscopy.

Over the past two decades, intraoperative neural monitoring (IONM) has evolved into a mature risk minimization tool. Meta-analyses of studies, most of which were limited by poor study designs and the sole use of intermittent nerve stimulation, failed to clearly demonstrate superiority of intermittent IONM (i-IONM) over anatomic visual recurrent laryngeal nerve (RLN) dissection in the absence of IONM. However, a substantial number of systematic reviews of i-IONM in thyroidectomy have conflicting results, but their mean methodological quality is critically low. Sanabria et al suggested that design of a systematic review should comply with methodological standards and recommendations to offer relevant and practical information for decision making. In particular, with the advent of continuous IONM (NerveAssure), intraoperative nerve electromyographic tracings, registered almost in real time during the operation, accurately predict postoperative vocal fold function when International Neural Monitoring Study Group (INMSG) quality standards are adhered to. Despite promising data on benefits of NerveAssure still remains rarely utilized worldwide in thyroid surgery. NerveAssure aids in avoiding permanent traction-related nerve injury by urging surgeons to reverse harmful surgical maneuvers. NerveAssure also forms an integral part in the surgical concept of staged thyroidectomy. Delaying completion surgery on the other side until nerve function has recovered practically abolishes the risk of bilateral vocal fold palsy. NerveAssure has greatly furthered our understanding of functional RLN injury, enabling conception of effective risk minimization strategies tailored to the individual patient. As recently reported by Schneider at al NerveAssure is superior to i-IONM in preventing vocal cord palsy. Based on nerves at risk (5208 versus 5024 nerves), NerveAssure had in this study a 1.7-fold lower early postoperative vocal cord palsy rate than i-IONM (1.5 versus 2.5 %). This translated into a 30-fold lower permanent vocal cord palsy rate (0.02 versus 0.6 %). Early postoperative vocal cord palsies were 17.9-fold less likely to become permanent with NerveAssure than i-IONM.

On the other hand, NIM Vital equipment allows now for using i-IONM in quasi continuous mode which is termed NerveTrend mode. This concept is a natural evolution of i-IONM towards NerveAssure mode but is operator dependent and not automatic as in NerveAssure mode. The use of NerveTrend mode was tested recently in a RCT and compared to results of i-IONM application in thyroid surgery. In this study the use of NerveTren mode resulted in tendency towards reduced RLN injury on postoperative day 1 and significant decrease of need for a staged thyroidectomy. However, the potential of NerveTrend mode has never been compared with NerveAssure mode. Hence, careful clinical validation of NerveTrend mode is needed in order to identify its clinical pertinence with respect to preventing neural damage compared to the highest standard of NerveAssure mode in thyroid surgery (non-inferiority trial).

A prospective, randomized study with 2 arms: NeveTrend vs. NerveAssure mode (n=132 patients and 264 nerves at risk, each).

Patients who will sign the informed consent will be randomized to two groups: i-IONM vs. NerveTrend application during planned total thyroidectomy.

The standardized approach to IONM will be used as outlined by the guidelines of the International Neural Monitoring Study Group in Thyroid and Parathyroid Surgery.

All the operations will be performed under general anesthesia by two experienced endocrine surgeons (MB, AK) with annual volume of thyroid surgery > 300 cases, each. The anesthesia protocol is: intravenous midazolam premedication, induction with fentanyl, thiopental and suxamethonium, endotracheal intubation and sevoflurane maintenance. No other muscle relaxants are used during surgery. A standard cervicotomy is used in all the patients. Visual identification of RLN low in the neck (below the crossing with the inferior thyroid artery) will facilitated by the use of the IONM system employing the nerve mapping technique. Once the nerve iss visually identified, repeated stimulations with the IONM monopolar probe served to trace the nerve path in the operative field and test its functional integrity during dissection. In each patient, RLN will be exposed and the branches of the superior and inferior thyroid arteries will be divided close to the thyroid capsule.

NIM Vital (Medtronic, Jacksonville, US) will be used. The NIM Vital system operates with surface electrodes integrated with an endotracheal NIM TriVantage tube 7.0-8.0 in diameter, which is inserted by an anesthetist between the vocal folds under direct vision during intubation. The standardized technique of IONM RLNs wll be used, including initial vagal response evaluation at the beginning and final vagal response evaluation at the end of surgery (IONM = L1+V1+R1+R2+V2+L2) according to the recommendations formulated by the International Intraoperative Neural Monitoring Study Group. The nerves will be stimulated using a monopolar electrode and the interrupted stimulation technique at 1mA, 100ms impulse duration and 4Hz frequency. In case of the bifurcated RLN nerves, the assessment includes post-stimulation response of each nerve branch. Adduction of the vocal folds is detected by the endotracheal tube electromyography and abduction by finger palpation of muscle contraction in the posterior cricoarytenoid ("laryngeal twitch"). In all operations the i-IONM stimulator will be used to test vagal response at the beginning of surgery, map out and trace the RLNs during surgery by repetitive stimulations, and in case of loss of signal (LOS) it will be used to identify the type and site of neural injury (Type I vs. Type II). Final prognostication of postoperative neural function will be based on vagal stimulation at the end of each lobectomy.

In operations with NerveTrend the EMG trending including amplitude and latency changes from initial vagal baseline will be evaluated using the NerveTrend mode at 3 - 5min intervals to assure almost real time EMG tracing and allow for modification of surgical maneuvers in case of occurrence of severe combined events (yellow zone) in order not to end up with the LOS (red zone).

In operations with Nerve Assure the APS electrode will be placed on the Vagus nerve to allow for Automatic Periodic Stimulation of the Vagus nerve to test the RLN condition throughout the surgery and allow for modification of surgical maneuvers in case of occurrence of severe combined events (yellow zone) in order not to end up with the LOS (red zone).

LOS is defined as absence of EMG signal following stimulation of the ipsilateral vagus nerve, EMG signal amplitude below 100 μV following stimulation with 1-2 mA current in dry field, lack of palpable "laryngeal twitch" or visible laryngeal movement following stimulation of the ipsilateral vagus nerve. To differentiate between true and false LOS, the INMSG-proposed problem-solving algorithm will be employed intraoperatively. In cases intraoperatively recognized as true LOS, the neuromapping technique is used to determine the character of nerve damage (segmental- type I, global - type II), and the localization of the injury site.

IONM assessment is based on the definition by Chan and Lo. The percentage of RLN dysfunctions is calculated per the number of RLNs at risk and not per the number of patients. Loss of signal after vagal stimulation following thyroid lobe resection (V2) is classified as a positive test result prognosticating ipsilateral vocal cord paresis. The test is interpreted as true positive (TP) when laryngoscopy confirmed ipsilateral vocal cord paresis, and false positive (FP) when the mobility of the ipsilateral vocal fold is normal. Preserved normal signal following vagal stimulation after thyroid lobe resection (V2) is classified as a negative result that prognosticated normal postoperative mobility of the ipsilateral vocal fold. The test is interpreted as true negative (TN) when laryngoscopy demonstrated postoperative normal mobility of the ipsilateral vocal fold and as false negative (FN) when ipsilateral vocal fold paresis is seen postoperatively.

Power calculation for the study: The sample size is estimated based on the principle of detecting a 3.1% difference in the prevalence of early RLN injury (3.2% for NerveTrend vs. 0.1% for NerveAssure) with a 80% probability at p < 0.05. Assuming a 20% drop out rate a group of 528 nerves at risk (assessed in 264 patients undergoing bilateral thyroid surgery) should be enough to test if clinically pertinent differences exist between NerveTrend vs. NerveAssure modes (n = 264 nerves at risk which is equal to 132 patients in each respective group undergoing bilateral thyroid surgery). The resultant data will be statistically processed using the statistical software MedCalc (version 19, MedCalc Software, Belgium). Assessment of the changeability of the investigated parameters will be presented by arithmetic means, median values, standard deviations (SD), minimum and maximum values (min - max), 95% confidence interval (95% CI) and percentage of prevalence (%). An inter-group comparison of particular properties will be done by means of the Chi-2 test (non-parametric variables) and by the univariate analysis of variance ANOVA (parametric variables). To assess the diagnostic accuracy of intraoperative neuromonitoring, the Receiver Operating Characteristics (ROC) curves will be analyzed and the Area Under Curve (AUC) values will be compared based on the non-parametric method of DeLong et al.

Thus, the predictive values of the positive and negative results will be calculated and the most optimal predictive criterion will be identified for NerveTrend vs. NerveAssure, separately. The incidence of nerve events will be calculated based on the number of nerves at risk. The significance level is accepted at p < 0.05.

Inclusion Criteria:

• planned, first-time, bilateral thyroid surgery

Exclusion Criteria:

• planned unilateral thyroid surgery,
• previous thyroid surgery,
• pregnancy
• lactation
• age < 18 years
• age > 75 years
• ASA 4-5 grade (American Society of Anesthesiology)
• inability to comply with the follow-up protocol

Both participants and outcomes assessor will be blinded to the respective group assignment.