Unexpected postoperative incidental recurrent laryngeal nerve palsy post total thyroidectomy after intraoperative nerve monitoring (overstimulation and fatigability): a case report

Abstract

Background

Postoperative recurrent or permanent recurrent laryngeal nerve injury is one of the most serious complications in the field of thyroid surgery, in either benign or malignant thyroid disease, significantly affecting patients’ quality of life. The importance of recurrent laryngeal nerve identification intraoperatively reduces the risk of injury. We report herein a young patient who underwent nerve monitoring in total thyroidectomy that led to recurrent laryngeal nerve injury.

Case presentation

We report a 30-year-old Arab male patient who presented to our clinic with a longstanding thyroid swelling, which was reported as papillary thyroid carcinoma, and was scheduled for total thyroidectomy and lymph node dissection. Nerve monitoring was used to identify the recurrent laryngeal nerve, leading to recurrent laryngeal nerve nerve fatigability/paresis that was seen during the postoperative course.

Conclusion

Visual identification of the recurrent laryngeal nerve is the gold standard for nerve protection. We recommend the use of nerve monitoring as an adjunct in challenging cases but not in routine settings, as it does not decrease the incidence of injuries compared with visualization alone in our experience.

Background

The incidence of permanent complications after thyroidectomy is generally low [1]. One of the classical complications that is mainly reported after performing thyroidectomy is dysphonia, which arises owing to the close anatomic proximity of the thyroid gland with the recurrent laryngeal nerves (RLNs). This dysphonia may be either temporary, occurring in 5–11% of cases, or permanent, in 1–3.5% of cases [1]. Postoperative recurrent or permanent RLN injury is one of the most serious complications in the field of thyroid surgery, in either benign or malignant thyroid disease, significantly affecting patients’ quality of life [1]. The importance of RLN identification intraoperatively reduces the risk of injury. Various methods of nerve identification have been described, including direct visualization of the vocal cords during dissection; intermittent monitoring techniques, such as palpation of the cricothyroid after stimulation of the nerve with a disposable stimulator; and continuous monitoring methods, such as intramuscular electromyographic (EMG) electrodes [2, 3]. RLN injury is usually discovered postoperatively with transient or permanent damage, with most of these occurring without a noticeable injury to the nerve intraoperatively, owing to intense stretching of the gland, direct suction, electrothermal injury, or anatomical variations of RLN [4]. Some challenging cases, such as a huge multinodular goiter and redo surgery, tend to increase the risk of injury [4]. The incidence of permanent injury ranges from 0.5% to 5%, whereas transient injury can occur with an incidence ranging from 1% to 30% [4]. Patients who experience RLN damage will develop dysphonia postoperatively, with or without swallowing problems, and shortness of breath, as the RLN innervates all intrinsic muscles of the larynx except the cricothyroid muscle. So, injury will cause a paresis or palsy to the vocal cords, and some cases might even require urgent airway intervention, such as intubation or tracheostomy if it is affected bilaterally. Intraoperative nerve monitoring (IONM) is considered a useful tool that aids in assessing the functional integrity of the RLN and aids the identification of the RLN. IONM was utilized in a study conducted on patients undergoing thyroid surgeries for malignant conditions to predict the factors that can influence temporary RLN palsy [6]. RLN palsy was found to be associated with low baseline amplitude and/or a requirement of higher current to maintain normal baseline amplitude during IONM or thyroid malignancy-related surgery [6]. These findings could be used as indicators to predict temporary RLN palsy [6]. However, even with utilizing IONM intraoperatively, the risk of RLN injury is not decreased when compared with direct visualization alone [5].

In this case, we share our experience where nerve monitoring was used and we believe might have induced nerve fatigability/paresis with higher stimulation current and frequency, seen in a young patient managed for thyroid malignancy in a tertiary care center in Riyadh, Saudi Arabia.

Case presentation

Initial presentation

A 30-year-old Arab male presented to our endocrine surgery clinic complaining of neck swelling for more than 2 years. He noted that the swelling was mainly on the left side, but he never sought medical attention as he was never symptomatic before. He was only investigated after he noted that the swelling had become more progressive in size. He denied having any voice changes, difficulty or pain during swallowing or choking, difficulty sleeping, or any compressive symptoms. He also denied having any generalized constitutional symptoms. He had no history of previous radiation exposure, upper airway or neck surgery or medications, any thyroid or endocrine diseases or any hormonal imbalance, or any previous malignancy. His social history was significant for smoking for more than 8 years. His family history was significant as he had two siblings complaining of thyroid-related issues (his parents have no consanguinity). He had done thyroid laboratory testing and thyroid ultrasound (US) outside. His laboratory blood tests showed thyroid stimulating hormone (TSH) of 1.43 mIU/L (reference range 0.35–4.94 mIU/L), free triiodothyronine (T3) of 2.79 pg/dL (reference range 1.58–3.91 pg/dL), and free thyroxine (T4) of 1.10 ng/dL (reference range 0.7–1.48 ng/dL). The thyroid US report showed that the right thyroid was of normal size, with a homogeneous echo pattern with no discrete nodules evident. The left thyroid gland was of large size and heterogeneous echogenicity, containing multiple complex nodules and internal calcification, the largest measuring 39.33 mm × 21.93 mm. The gland showed internal hypervascularity under Doppler, and showed a large complex cyst that was rounded and oval with a smooth margin that contained a low level of echogenicity that might be hemorrhagic, measuring 43.40  mm × 25.7 mm × 34.79 mm. Neck computed tomography (CT) with intravenous contrast was done to rule out retrosternal extension, which showed an enlarged left thyroid lobe with the presence of a complex thyroid nodule measuring 9 cm × 5 cm, with central calcification. The cranial cystic component reached up to the level of carotid bifurcation, the trachea was mildly compressed and deviated to the right side, and there was retrosternal extension of the left lobe by 1.3 cm. Other findings were unremarkable. The patient was admitted electively for surgery, undergoing total thyroidectomy with central neck dissection.

Intraoperative observations

The patient underwent total thyroidectomy and central neck dissection. Intraoperatively, we proceeded as per the standard surgical approach. A specialized neuromonitoring tube was inserted under direct vision with the guidance of a Glidescope. Clear instructions were to avoid any long-acting muscle relaxant agents. The patient was successfully intubated, and the tube was precisely placed between the vocal cords. The electrophysiology of the nerve monitoring was to stimulate the nerve by a standard current through an electrical probe that was transmitted after the contraction of the vocal muscles, which was stimulated by the electrodes placed on the endotracheal tube, and these contractions were reflected in the form of an electromyographic (EMG) wave.

We were able to identify the superior vascular pedicle and separate it from the external branch of the superior laryngeal nerve, which was protected. The RLN was identified and protected throughout the whole procedure. On the left side (the site of the tumor), there were extensive adhesions and fibrosis, and the thyroid nodule at the site of malignancy was also very adhesive, fibrocystic, and adherent to the muscles. The RLN had been followed to the entrance, and nerve monitoring was used, but it was not helpful. The stimulation current used was between 0.8 and 1 µV, and at some point the threshold was downgraded from 100 to 75 to increase the sensitivity, and the current was increased above 1 µV, but never exceeded 3 µV. Initially, we had a satisfactory signal, but with repeated stimulation, the signal was inaccurate and misleading. Also, IONM was used on the right side, but it was much weaker with the same current and threshold. Intraoperatively, the patient was extubated and tolerated the procedure well. An awake Glidescope was done by the anesthesia team post-extubation in the operating room and found vocal cord abducting and adducting bilaterally. After that, the patient was transferred safely to the post-anesthesia care unit (PACU) in safe condition.

Postoperatively

The patient developed a sudden stridor upon his arrival in the PACU, for which nasopharyngolaryngoscopy was done under anesthesia for interval assessment. It revealed that both vocal cords were visualized and noticed to be tense, but no abduction or adduction movement was detected bilaterally. The vocal cords were in abduction, and with the scope advanced in the trachea (beyond the glottic area), no cough/sensation or movement was seen. The otorhinolaryngology (ENT) team was consulted immediately. Upon their assessment, he was vitally stable, kept on 3 L oxygen (O₂) nasal cannula, kept on nothing by mouth, and arranged to go for an intensive care unit (ICU) monitoring bed. Another nasolaryngoscopy was done by ENT as well, showing bilateral immobility of the vocal cords in the paramedian position from their side. They recommended to keep nothing by mouth, start on dexamethasone 8 mg intravenously three times a day (TID), racepinephrine nebulization as needed (PRN), vitamin B complex, to place the patient on enteral feeding via a nasogastric tube (NGT), a swallowing assessment once the patient stabilized, and if the patient developed any signs of respiratory distress to proceed with intubation. During his ICU admission, two trials of NGT insertions failed, as the patient was gagging and did not tolerate it, for which he was kept on maintenance intravenous fluids. He was having aphonia with a low-pitched voice. In the ICU, on his postoperative day (POD) 1, he was on an O₂ nasal cannula, before he was weaned off oxygen, tolerated it, and was sating well on room air. On POD 3, the patient was able to swallow his own saliva without choking, but his voice was still hoarse. Overall, he remained completely asymptomatic during his ICU stay, never developing any choking episodes, aspiration, or any signs of respiratory compromise. After stabilization of his general condition, the ENT swallowing team was contacted, for which the patient underwent fiberoptic endoscopic evaluation of swallowing (FEES) on POD 7. FEES revealed aphonic voice with weak cough. An oral-motor examination was unremarkable. FEES study (with the scope passed through the right nostril) showed normal palate movement, bilateral vocal cord fold paralysis in the paramedian position with respiratory chink about 10–12 mm, phonatory gap big at about 10 mm, and no pooling of saliva in the valleculae, pyriform fossae, or laryngeal inlet. With thick liquid (2–3 ml), there was good oral control of the bolus, with mild penetration, but no aspiration, mild vallecular residual that was cleared with double swallowing and improved with chin tuck posture. With puree (5 ml) effortful swallowing, there was no penetration, no aspiration, but mild vallecular residue that was cleared with a double swallow. Other consistencies were not tested at that time for patient’s safety. The diagnosis concluded neurogenic pharyngeal dysphagia, normal palate movement, bilateral vocal fold paralysis in the paramedian position, pharyngeal muscle weakness, and fair laryngeal sensation. After the first session of FEES, the patient was allowed to start on a liquid diet (except water) alongside intravenous maintenance, while on steroids and vitamin B complex. He was transferred from the ICU to a general surgical ward in a stable condition. He was rescheduled for another FEES assessment, which was done 14 days after the surgery, and showed that his voice had improved with dysphonic voice, that he still did not choke with his saliva, with no drooling, no dyspnea, no respiratory distress, and no stridor. FEES study (passed through the right nostril) showed normal palate movement, bilateral vocal fold paralysis in the paramedian position with respiratory chink about 10–12 mm, and phonatory gap improved from 10 mm to 2–3 mm. There was improvement of adductor movement of the right vocal fold, with no pooling of saliva in valleuclae, pyriform fossae, or laryngeal inlet. With thick liquids (2–5 ml) there was good oral control with bolus, with no penetration, no aspiration, no residue, and that improved with chin tuck posture, but penetration and aspiration without chin tuck posture. With puree (5 ml) and effortful swallow, there was no penetration, no aspiration, but mild vallecular residue that was cleared with a double swallow. With soft solids, there was no penetration, no aspiration, and no residue. The diagnosis this time revealed a neurogenic pharyngeal dysphagia, left vocal cord paralysis, right vocal fold paresis, and pharyngeal muscle weakness with still fair laryngeal sensation. He was planned for a third FEES assessment after 6 weeks from the second FEES session. Overall, the patient was doing fine, remained completely asymptomatic, was able to tolerate orally, and was fit to be discharged home safely on POD 15.

Follow-up

Postoperative pathology revealed a diagnosis of papillary thyroid carcinoma (PTC), with tall cell variant. The patient was referred for both endocrinology and radiation oncology for further follow-up and management, as well as follow-up with the ENT swallowing team for further assessment.

Seen after 6 weeks in the ENT clinic, he was improving clinically and was able to tolerate a regular diet. Cleared from ENT, he was to proceed to radioactive iodine ablation, as planned by the endocrinology and radiation oncology teams, and kept on vitamin B complex tablet (neurobion) for 3 months. After an interval of 3 months, the patient was seen again in the ENT clinic; naso-fiberoptic laryngeal examination was done and showed normal vocal cord movement. Thus reassured, the patient was discharged from the clinic with no intervention or further follow-up needed.

Discussion and conclusion

Although injury to the RLN during thyroid surgery has become less reported and is no longer a common complication, it is still considered to be one of the major causes of high morbidity, and can greatly impact patients’ quality of life [1]. RLN injury can be classified as either direct or indirect. Direct injury can result secondary to direct clamping, ligature device entrapment, or electrothermal injury [8]. These injuries can be preventable by adjusting surgical techniques and improving surgical skills. On the contrary, indirect injuries occur from excess traction of the nerve by stretching or distortion during thyroid dissection [8]. These injuries cannot be eliminated because the handling of thyroid tissue must be mobilized in an appropriate fashion before resection during surgery [8]. In the field of thyroid surgery, continuous nerve monitoring has gained wide popularity in benign, but mostly malignant, thyroid cases. IONM was established to be used in many tertiary centers worldwide to identify the RLN and external branch of the superior laryngeal nerve (EBSLN) and thus reduce the rate of injury, as well as for early prediction of vocal cord paralysis intraoperatively [6]. Overall, few studies have reported whether RLN monitoring reduces the risk of postoperative paralysis, and have shown conflicting results. One retrospective study that aimed to determine the role of intraoperative neuromonitoring and the parameters used to estimate/predict temporary RLN palsy in malignant thyroid cases showed that patients with low baseline amplitude and/or requiring higher current to maintain normal baseline amplitude were often associated with higher incidence of temporary RLN palsy in the postoperative period [6]. In correlation to TNM staging of the thyroid malignancy, patients with advanced pT stage and intraoperative nerve injury necessitated higher current to obtain a baseline EMG amplitude, which ultimately can affect the results of INOM [6]. One reported case showed that traction and other forms of mechanical stress impair signal transmission from RLN to vocal cord muscles, with characteristic electrophysiological alterations of EMG signals [8]. This RLN injury was caused by traction on the thyroid, which may lead to indirect injury by stretching Berry’s ligament at the entrance to the larynx during an operation [8]. IONM is a helpful method to reduce the rate of RLN palsy with less experienced surgeons, with no superiority of IONM over the visual identification of the nerve. IONM will aid in decreasing the rate of transient, but not permanent, RLN paresis compared with visualization alone, particularly in high-risk patients [6, 7]. Whether its use truly reduces the risk of RLN injury has yet to be proven [11]. Prior studies have reported on the benefit of continuous RLN monitoring. However, very few studies have compared the outcome of RLN monitoring versus no RLN monitoring [11, 12]. A systematic review of randomized control studies concluded that visual identification is the safest approach to prevent nerve injury [9]. The rate of nerve injury is generally higher in thyroid carcinoma, goiter, and re-operation [9]. During thyroidectomy, both intraoperative IONM and visual nerve identification alone are widely used. The choice of the approach depends on many factors, such as the surgeon’s experience and technical resources. However, there is no clear evidence that IONM should be preferred over visual nerve identification only in all patients undergoing thyroid surgery or whether it should be used in selected high-risk patients, for example, during revision thyroid surgery [9]. RLN paralysis was recorded with an incidence of 3.4% in all thyroid pathologies. The incidence is higher in malignant tumors (5.7%) and varies depending on the type of thyroid malignancy, ranging from 1.4% for differentiated cancer to 16.5% for anaplastic or undifferentiated cancer, where direct invasion of the nerve can occur [10, 11]. The main pitfalls associated with continuous nerve monitoring include the potential of having a false sense of security. Lack of stimulated EMG signals may indicate that the structure being stimulated is not the RLN or that the nerve is transected or neurapraxic (true negative). However, it could be due to inappropriate position of the electrodes in relation to the vocal cords, which will lead to false negatives. This can be secondary to advancing the tube too far inferiorly or superiorly or because of rotation. Therefore, if the surgeon is convinced that the RLN has been identified correctly but cannot obtain an EMG signal with electrical stimulation, it would be beneficial to check the tube position [11, 12].

In conclusion, visual identification of RLN is the gold standard for nerve protection, and to reserve the use of nerve monitoring as an adjunct for challenging cases, not in routine settings, as it does not decrease the incidence of injuries compared with visualization alone. Furthermore, a good understanding of the electrophysiological background of nerve monitoring is a must to avoid possible complications related to nerve fatigability and overstimulation. In our case, the patient required higher current to maintain normal baseline amplitude, which resulted in temporary RLN palsy in the postoperative period.