The paired electrodes may be placed through the cricothyroid space directly into the anterior aspect of the thyroarytenoid muscle. In our experience, the depth of needle insertion was not related to monitoring sensitivity but may be important to stability. Alternative recording sites include the anterior cricopharyngeus muscle (not shown).
A single paired electrode may be used to monitor both superior laryngeal nerves by placement across the cricothyroid space. For all electrode placement, individual experimentation is recommended to arrive at optimum performance to suit personal preference.
Petro ML, Schweinfurth JM, Petro AB. Transcricothyroid, Intraoperative Monitoring of the Vagus Nerve. Arch Otolaryngol Head Neck Surg. 2006;132(6):624-628. doi:10.1001/archotol.132.6.624
Copyright 2006 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.2006
To develop a reliable, user-friendly, intraoperative, electromyographic monitoring technique to decrease the incidence of injury to the recurrent and superior laryngeal nerves.
Prospective, nonrandomized, clinical trial of a nerve monitoring technique.
Private tertiary care community hospital.
A population-based sample of 31 patients scheduled to undergo thyroid surgery was enrolled consecutively. Included in the study were patients older than 18 years who were scheduled for surgery and who were able to provide informed consent. Exclusion criteria were pregnancy, implanted metallic devices, and history of laryngeal surgery, injury, paresis, hoarseness, or paralysis. No patients were excluded, and all completed the study and returned for follow-up visits.
Twenty-nine patients required total thyroidectomy, of which 10 involved malignancy, and the other 2 patients required lobectomy. Each patient completed the Voice Handicap Index and underwent a preoperative fiberoptic laryngeal examination. Continuous monitoring was performed using a widely available, commercial nerve integrity monitor and a paired electrode placed into the cricothyroid space under direct vision. Postoperatively, participants completed a follow-up Voice Handicap Index survey and underwent a laryngeal examination.
Main Outcome Measures
The incidence of vocal paresis, or paralysis, and the preoperative and postoperative voice handicap score were recorded. The usefulness of the device based on the surgeon's subjective and immediate postoperative impressions was rated on a visual analog scale.
Sixty-two recurrent laryngeal nerves were identified with continuous electromyographic monitoring. Vocal cord paresis or paralysis was not observed. Postoperative Voice Handicap Index scores were unchanged from preoperative assessment. The technique was given a rating of 1 (most useful) on a 5-point scale in 70% of cases.
The technique described is sensitive, easy to use, accurate, and associated with a high degree of surgeon satisfaction. This technique is not associated with additional risk to the patient and offers the potential to reduce injury. Monitoring provides assurance that the nerve is intact and functioning prior to extubation.
Iatrogenic injury to the recurrent laryngeal nerve (RLN) during open neck surgery is the leading cause of vocal cord paralysis throughout the world. The reported incidence of this complication ranges from 2.3% to 5.2%.1 Reports in which routine preoperative and postoperative laryngoscopic procedures were performed measured a 0.4% to 3.9% (mean, 2.2%) incidence of temporary RLN paralysis and a 0% to 3.6% (mean, 1.6%) incidence of permanent RLN paralysis per nerve at risk after thyroidectomy.2 Mechanisms of RLN injury include transection, traction, ischemia, ligation, crush, and electrothermal injury.3,4 The risk of injury is increased in cases of malignancy, secondary operation, reexploration for hemorrhage, anatomic variability, anatomic distortion from goiter or neoplasm, and primary failure to identify the recurrent nerve.4- 6 Nerve injury may be avoided with accurate anatomic localization during surgical dissection. Several nerve localization techniques have been described in the literature. Most of these methods rely on continuous monitoring of the mechanical or electrical activity of the laryngeal musculature via electromyography (EMG).
Previous studies7,8 have characterized the EMG responses of the RLN and the superior laryngeal nerve (SLN). The EMG responses may be monitored by an audible amplified signal. When EMG is used to measure nerve stimulation, a negative response denotes the absence of neuromuscular activity. A nerve or peripheral muscle may demonstrate spontaneous motor unit potentials that are easily discriminated from mechanically elicited responses. Significant responses may be a large burst of activity, typically the result of mechanical stimulation of the nerve root, or a pattern of sustained firing that can result from mechanical or thermal stimulation from electrocautery.
A stimulus artifact is characterized by a short, high-frequency click, whereas an evoked EMG response results in a more intense, low-frequency, triphasic signal. Responses of the SLN have been further differentiated from those of the RLN by a complex wave (biphasic-triphasic) with an average latency of 0.5 ms. These responses differ dramatically from the interference of general anesthesia and respiration on background EMG activity both visually and sonically. Finally, unlike interference and spontaneous activity, evoked responses are readily and immediately reproducible and should produce the same response to a repetitive stimulus.
Initially, EMG monitoring was described using electrodes placed via direct laryngoscopy into the vocal cords and laryngeal musculature prior to surgery. This method requires separate preoperative and postoperative laryngoscopic procedures and is therefore cumbersome or impossible for the nonotolaryngologist. Risks include hematoma; local abscess formation; electrode displacement during surgery, often without the surgeon's knowledge; and the additional risks and discomfort associated with direct laryngoscopy.9 The use of surface electrodes has subsequently been described; however, a significant drawback is the inconsistent EMG response secondary to the larger area of muscle being sampled.7,9 In addition, the EMG responses using surface electrodes differ significantly from needle electrodes and have not been well characterized.9
Electromyography via contact electrodes integrated into the endotracheal tube has been used to monitor the RLN.10 Although this method produces very clear EMG responses, it has several disadvantages. A specially modified laryngoscope blade is required for accurate placement and proper positioning. The quality of recording depends on maintaining contact with the vocal cords; however, the large endotracheal tube necessary to maintain optimal contact also significantly increases intraoperative pressure on the glottis. If contact is lost secondary to endotracheal tube movement during surgery, invalid recordings may result.9,11,12 Furthermore, the use of the tube is contraindicated by factors that interfere with electrode contact, including anatomic distortion (common in thyroid goiter) or scarring. Tube insertion can injure and potentially dislocate the arytenoids. Other contraindications are those that make injury to the RLN more likely. Limitations of the specialized endotracheal tube include cost, necessity of a stylette for introduction, and high flexibility with considerable risk of intraoperative tube obstruction. In addition, studies13,14 of cervical spinal surgery–associated vocal paralysis demonstrated significant nerve compression within the endolarynx.
Yet another technique combines the use of a laryngeal mask airway and intraoperative, flexible laryngoscopy. This requires continuous observation of the vocal folds for movement and electrical stimulation of the RLN for localization and incurs the increased risks of aspiration and laryngospasm associated with use of the laryngeal mask airway.15 Compared with EMG techniques, the sensitivity is lower and there is no integrated mechanism to alert the surgeon.
These methods are invasive, technically challenging, and time consuming, and they require special skills and instrumentation with an associated increase in set-up time and cost.8 In addition, all require electrical stimulation of the nerve for intraoperative identification. Finally, none of these techniques allow for monitoring of the SNLs. We designed a study employing the nerve integrity monitor (NIM-2; Medtronic Xomed, Jacksonville, Fla), commonly used to monitor the facial nerve, and wire electrodes placed directly into the cricothyroid space within the operative field. To our knowledge, this technique has not been previously described in the literature. Wire electrodes have been demonstrated to record an order-of-magnitude higher-amplitude, compound motor action potential than do contact electrodes with a higher signal-to-noise ratio.10 Improved sensitivity is expected to consistently alert the surgeon to the presence of the nerve at a greater distance from the dissection, and it may be especially important when nerve dysfunction and decreased amplitude are expected. We hypothesize that monitoring the RLN in this fashion will prevent and/or decrease the risk of iatrogenic injury without additional risk to the patient and that it will be especially useful in procedures associated with increased risk to the RLN. The proposed technique provides a simple method of highly sensitive, continuous EMG monitoring using equipment readily available in most hospitals, and it requires no increased set-up or operative time and no additional surgical dissection or exposure.
Thirty-one consecutive patients classified as anesthesia risk class 1 or 2 according to the criteria of the American Society of Anesthetists and scheduled to undergo thyroid surgery performed by a very experienced surgeon (A.B.P., who has performed approximately 4000 thyroidectomies) at a tertiary care institution were enrolled prospectively. The protocol and consent were approved by the institutional review board at the University of Mississippi, Jackson, and written informed consent was obtained from each patient in advance. Included in the study were patients older than 18 years who were scheduled to undergo surgery and were able to provide informed consent. Exclusion criteria were pregnancy; implanted metallic devices that might have focused current density; and history of laryngeal surgery, injury, paresis, hoarseness, or paralysis. Preoperative and postoperative examination included fiberoptic laryngeal examination. Each patient completed the Voice Handicap Index preoperatively and postoperatively.
Succinylcholine chloride (1 mg/kg) was used in the initial phase of the general endotracheal anesthesia. No additional neuromuscular blocking agents were used following intubation. Prior to placement of the electrodes, transcutaneous neuromuscular stimulation was used to ensure that the patients were not paralyzed. The surgical approach to the thyroid was performed with concurrent EMG monitoring. No additional dissection beyond the standard operative technique was performed or required.
The NIM-2 was used for recurrent laryngeal nerve monitoring. Under direct visualization, a 4-mm paired electrode was inserted directly into the thyroarytenoid muscle through the cricothyroid membrane (Figure 1). Placement was confirmed by correlation with respiratory activity. Reference ground electrodes were placed separately outside the surgical field on the patient's shoulders. An event threshold of 100 μV was used. Nonpathologic and pathologic EMG responses were recorded under continuous monitoring. A background EMG was observed in all patients during dissection, and the surgeon was asked throughout the procedure to update the location of the dissection relative to nerve location.
Electrode impedance and imbalance values were periodically checked during the surgical procedure to ensure continued optimal electrode placement. Monitoring was conducted using consistent recording parameters. Significant EMG events elicited during surgery and artifactual responses were recorded. Episodes of direct nerve disturbance were recorded by amplitude in millivolts and overall duration of motor unit activity in milliseconds. This information was verified by direct visualization of the surgical field and position of the RLN.
All significant EMG activity was assessed through burst and sustained electrical events. The EMG events were compared with voice outcomes, which were subsequently assessed by direct laryngeal examination and Voice Handicap Index score. The usefulness of the device in identification and preservation of the nerve was rated subjectively by the surgeon using a 5-point visual analog scale. This is an ordinal scale that was defined preoperatively: 1, very useful; 2, useful; 3, neutral; 4, not helpful; and 5, a hindrance.
Thirty-one patients underwent thyroid surgery with concurrent EMG monitoring of the RLN, 29 underwent total thyroidectomy and 2 underwent thyroid lobectomy. Ten patients had malignant thyroid disease (Table). Twenty-seven were women and 4 were men. The subjects ranged in age from 18 to 80 years (mean age, 49 years; median age, 47 years). No patients experienced a temporary, delayed, or permanent dysfunction of the vocal folds. Voice indices remained unchanged between preoperative and postoperative measurements.
The operating surgeon rated the contribution of the monitor to surgical progress as a 1 of 5 (very useful) in 74% of cases and in 8 of the 10 cases involving malignancy, for an overall rating of 1.3. In the remaining 26% the monitor was rated useful (n = 7) or neutral (n = 1). Neuromuscular activity below the 100-mV threshold was found to be incidental because dissection within the vicinity of the RLN consistently produced identifiable electrical activity over the 100-mV baseline. Direct stimulation of the nerve at the 0.5-mA level resulted in evoked responses higher than 500 mV in all cases. There were no equivocal responses to stimulation experienced.
Despite the numerous techniques and commercially available products designed for intraoperative monitoring, iatrogenic RLN injury continues to occur with regularity. A presentation by Batniji and Batniji16 at the 2004 annual meeting of the American Laryngological Association described bilateral RLN paralysis post–total thyroidectomy. Neck exploration revealed a transected left nerve and suture and a hemoclip on the right nerve. It is unlikely that such a complication could have occurred under continuous EMG monitoring, which indicates that this technology may not enjoy widespread use.
In this prospective series, no complications were observed either during surgery or from the use of the monitoring system. In addition, there were no cases of temporary or permanent paralysis of the RLN. Last, the minimal voltage used to stimulate the nerve did not induce temporary paresis. It is not our intent, however, to imply that monitoring should replace standard systematic dissection and careful anatomic localization of the nerves. The information provided by EMG monitoring may be used to supplement information gained by the operating surgeon during the course of dissection. Severe injury to the RLN without warning is certainly possible despite the most sensitive monitoring techniques.
In most of the cases, the monitor was felt to be very useful. In one operation during which this technique was used, the nerve was identified embedded within the thyroid gland. During reexplorations, the monitoring system was also judged to be very helpful. Its usefulness in the removal of large goiters was also consistently rated highly as well. Most of the cases in which the monitor was not rated a “1” were performed near the beginning of the study, and the lower high rating may have been due to technical issues related to the monitor. For example, the monitor's performance was found to be affected by placement adjacent to the electrocautery unit and anesthesia equipment; eventual repositioning corrected this issue. In other cases, the nerve was more easily found secondary to size or location, and the monitor had a relatively smaller impact. The cases in which the rating of the technique was neutral were straightforward operations that would likely not need monitoring in the hands of most experienced surgeons. Finally, there was a learning curve in interpretation of the signals.
In some cases, the monitor was used to help localize the nerve before its identification, especially where the nerve was found embedded within the thyroid. In most cases, the nerve was identified and the monitor used to confirm identification and to assist in mapping its course. The EMG recording may also be useful in reducing cumulative surgical stimulation by calling attention to and thereby minimizing unnecessary manipulation of the nerve. It was noted that retraction in the vicinity of the nerve that elicited a response from the monitor resulted in altered technique that was likely protective. The ability to alter technique in response to the monitor may be responsible for improved outcomes. Last, EMG can be employed at the completion of surgery to ensure nerve integrity.
Although set-up time was not one of the measured outcomes in this study, it was found to be minimal in each case. Electrode placement did not require additional dissection and did not restrict surgical exposure or impede the surgeon. Electrodes did not become dislodged during the surgery and were directly visualized at all times, which provided further assurance as to their location. The EMG response was clearly discernible over background neuromuscular activity, and there were minimal false-positive EMG events. In evoked EMG, response is defined as a compound motor action potential. Noise is defined as all other electrical activity not related to signal. The laryngeal electrode in this study is anatomically close to the dissection site and therefore stimulation achieves a large-amplitude electrical event that is clearly identifiable over events that might occur with movement of the electrode or retraction.
Initially, an electrode was placed across the cricothyroid space (Figure 2) to monitor the external branches of the SLNs. Because the senior author (A.B.P.) does not routinely identify the SLN, it was not consistently possible to verify that monitor activity during dissection around the superior pole was due to direct manipulation of the SLN because its location was not verified with direct visualization. The potential certainly exists for this technique to be useful for monitoring the SLN when indicated. In this instance, the postoperative integrity of the SLN could be verified with direct stimulation while monitoring the cricothyroid muscle. Although monitoring multiple nerves through 1 or 2 channels is feasible, it is impossible to distinguish between nerves by the tone or quality of the signal and therefore the area of dissection becomes the only reliable guide.
In conclusion, the technique described herein is sensitive, easy to use, accurate, and associated with a high degree of surgeon satisfaction. Despite the superficial appearance of similarity to existing techniques, to our knowledge, direct intramuscular monitoring of the larynx from within the operative field has not previously been described and deserves careful consideration. It requires minimal set-up time, can be used by any surgeon without specialized knowledge, and provides an order of magnitude more sensitivity than currently available techniques. At no additional risk, significant potential to reduce or eliminate iatrogenic nerve injury to the recurrent laryngeal nerve is obtained. Monitoring provides assurance that the nerve is intact and functioning prior to extubation. As with any safety equipment, a significant reduction in injury is unlikely unless the device is used consistently. The simplicity of the technique makes it adaptable and applicable to other surgical subspecialists who perform thyroid surgery as well. The combined features of the technique described should make it easy to adopt, although we anticipate that it will be most valuable in surgical procedures that have to be redone or for cases in which the nerve is anticipated to be involved in the disease process and that most cases will not require nerve monitoring.
Correspondence: John M. Schweinfurth, MD, Department of Otolaryngology, The University of Mississippi Medical Center, 2500 N State St, Jackson, MS 39216-4505 (firstname.lastname@example.org).
Submitted for Publication: July 1, 2005; final revision received October 26, 2005; accepted January 3, 2006.
Financial Disclosure: None.
Previous Presentation: This study was presented in part at the Southern Section of the Triological Society Meeting; January 15, 2005; Miami, Fla.