Laryngeal Examination in Thyroid and Parathyroid Surgery

Introduction

Thyroid and parathyroid surgery is undertaken by surgeons from a variety of surgical disciplines and backgrounds. The implications of this type of surgery on the voice are a frequent patient concern and a major determinant of surgical quality outcomes. Although several neural and non-neural factors can influence voice outcomes, injury to the recurrent laryngeal nerve (RLN) has the greatest effect.

Incidence vocal cord paralysis (VCP) is a key performance indicator of surgical quality that requires gold-standard presurgical and postsurgical laryngeal examination to assess paralysis rate. After the recommendation by Randolph in 2010, many national guidelines and consensus documents included a recommendation for all patients to undergo an examination of the vocal cords both before and after surgery.

In a 2003 study, Saunders et al. found that 50% of thyroidectomies in the United States are performed by surgeons operating on fewer than five cases annually. There is overwhelming evidence that shows a correlation between volume and surgical outcomes in thyroidectomy. VCP and other complications have shown an inverse relationship with increased volume. An annual thyroid volume of higher than 50 has been shown to be protective. The literature reports a further decrease in complications when volumes exceed 100. This figure has been supported by the 5th BAETS (British Association of Endocrine and Thyroid Surgeons) National Audit, which is the largest endocrine surgical database in the world, containing in excess of 115,000 operations.

A preoperative knowledge of vocal cord function is vital for surgical planning and is imperative when obtaining informed consent from patients. Operations performed on patients with only one functioning cord involve a very different conversation with the patient. Patients are increasingly forward in asking surgeons about what their operative volume and outcomes will be. In the absence of a preoperative laryngeal examination given national and international guidelines, a patient-turned-litigant’s complaint may be likely to be supported in a court of law in the event of injury.

A change in voice is the most common complication after thyroid surgery. Although VCP rates in the extant literature have quoted low paralysis rates, there has been an increasing focus on patient outcomes in recent years. Often, these reports derive from high-volume surgeons and centers and may not have routine laryngeal examination postoperative evaluations. Paralysis rates after surgery have been shown to be significantly underreported. The variable, inconsistent adoption of presurgical and postsurgical laryngeal examination are primarily responsible.

RLN injury is the most common cause of litigation in endocrine surgery. A study by Abadin et al. revealed that RLN injury contributed to 46% of thyroidectomy malpractice litigation between 1989 and 2009. In a 2003 study, Lydiatt found that 67% of unilateral VCP lawsuits had a favorable outcome for the plaintiff with a deficiency in informed consent responsible in 78% of cases.

Having a better understanding of the implications of VCP from a patient’s perspective is valuable; a useful reading on the subject can be found in Munch and DeKryger’s article from 2001.

Anatomy and the voice

Voice is unique to the individual; it serves as an indicator of emotion, background, education, and many other personal qualities. The voice, speech, and language have been developed only in the human. The generation of voice is complicated and relies on the integration and coordination of designated areas within the cerebral cortex that direct the laryngeal muscles and resonance of air through the upper airway. Airflow through the cords leads to subtle movements of the inner epithelial cover of the vocal cords referred to as the mucosal wave . This process is attributed to the interplay of several factors ( Table 15.1 , Figure 15.1 ).

The larynx has a shield-shaped thyroid cartilage composed of a left and right lamina, which sits posteriorly on the inferior ring-like cricoid cartilage through the cartilaginous cricothyroid joints. The true vocal cords consist of a lining of squamous unciliated, nonkeratinized epithelium draping the fine laryngeal musculature. The cords project into the laryngeal lumen that is located halfway down the thyroid cartilage. The posteriorly placed arytenoid cartilages rest on the posterior cricoid lamina and contribute to the posterior cartilaginous third of the true cord. The cords insert anteriorly into the inner surface of the midline thyroid cartilage at Broyles’ ligament. The movements of the arytenoids are rotational, but they also glide subtly along the face of the cricoid. Abduction is caused by a rotation of the arytenoid laterally, whereas medial rotation leads to adduction of the cords. The laryngeal musculature can be divided into intrinsic or extrinsic muscles. They are of 4th and 6th branchial arch derivation and are innervated by the external branch of the superior laryngeal nerve (EBSLN) (4th arch) and the RLN (6th arch). Sensory supply of the supraglottis derives from the superior laryngeal nerve (SLN), whereas the infraglottic sensory supply is via the RLN. The cricothyroid muscle is the only external laryngeal muscle situated on the outer surface of the larynx; as a consequence, it is within the thyroidectomy field and is easily and frequently injured. The cricothyroid muscle tilts the thyroid cartilage forward, which lengthens and tenses the cords facilitating a higher vocal register and voice projection. The cricothyroid muscle is extremely important for singing and vocal projection; significant injury may be career-ending for the professional singer (see Chapter 35 , Surgical Anatomy of the Superior Laryngeal Nerve).

The RLN is a mixed nerve that carries motor, sensory, and autonomic fibers. It innervates all intrinsic muscles of the larynx except the cricothyroid muscle, which is innervated by the EBSLN. The RLN also innervates the inferior constrictor muscle of the pharynx and cricopharyngeus muscle carrying sensory fibers from the larynx, upper esophagus, and trachea. These RLN sensory fibers may be injured during thyroidectomy; this is why swallowing problems can be a postoperative complication with or without a VCP (see Chapter 36 , Surgical Anatomy and Monitoring of the Recurrent Laryngeal Nerve).

Trauma to the epithelial lining of the vocal cords is invariable after insertion of an endotracheal tube and may influence the subtle mucosal wave movements; this may result in a change in voice quality generically described as hoarseness. Voice change from a VCP usually has a more distinctive character more accurately termed breathiness. This is a weak voice and cough caused by glottic incompetence during phonation because of the lateral resting position of the cord, which is typical for most paralyzed vocal cords (see Chapter 42 , Pathophysiology of Recurrent Laryngeal Nerve Injury, and Chapter 43 , Management of Recurrent Laryngeal Nerve Paralysis).

Iatrogenic RLN injury is most commonly mechanical (compression and stretching) or thermal (diathermy and energy devices). Minimal injury of the myelin sheath of the RLN causes temporary blockage (i.e., neuropraxia) of nerve conduction, which usually recovers fully after 6 to 8 weeks (see Chapter 36 , Surgical Anatomy and Monitoring of the Recurrent Laryngeal Nerve). More severe trauma to the RLN, particularly from thermal injury, may damage the myelin sheath neural fibers (axonotmesis). Recovery is still possible, but with a greater delay, and is often suboptimal manifested by reduced mobility and dysfunctional reinnervation known as synkinesis; the manifestation of these symptoms have a negative effect on the voice. Greater RLN injury (neurotmesis) leads to interruption of endoneurial, perineurial, or epineurial sheaths. This type of injury is associated with incomplete or absent nerve regrowth and connectivity.

Unilateral vocal fold paralysis can be well tolerated, particularly if the cord remains in a paramedian position. Factors that control the final resting position of the cord after RLN injury are not understood. This and intubation edema can often allow for a near-normal voice in the initial days after surgery. Patients may not, therefore, complain of a change in voice until weeks have passed after their surgery. In more severe injuries, symptoms usually appear as a consequence of laryngeal muscle and cord atrophy. Patients complain of a weak, breathy, and hoarse voice. Laryngeal examination often reveals an immobile cord in a lateralized position. Injuries to the EBSLN can be extremely difficult or impossible to diagnose by laryngeal examination, even by expert laryngologists using stroboscopy. In pure SLN injuries, the cord can be apparently thinner and shorter, and the posterior glottis may “point” to the injured side. Injury to the cricothyroid muscle as a gold-standard test can only be confirmed by electromyography (EMG) (see Chapter 35 , Surgical Anatomy of the Superior Laryngeal Nerve).

Lateral displacement of the paralyzed vocal cord leads to a glottic gap that lets air escape during phonation (breathy voice). Cord level can be also be different compared with the normally moving side, which further contributes to poor voice quality. Because of impaired glottic competence, the patient can complain of swallowing difficulties, which may predispose them to coughing spasms secondary to aspiration. Recent big data studies revealed associated morbidity and long-term mortality associated with VCP. VCP is significantly associated with emergency admission within 12 months after surgery, the development of pneumonia, and a ten-fold increase in the risk of gastrostomy for feeding issues. Furthermore, there was a long-term increase in mortality associated with VCP from bronchopneumonia.

As discussed previously, the EBSLN innervates the cricothyroid muscle. Injury to the muscle causes a problem when projecting the voice because it is a tenser of the cords important in attaining higher registers; this injury is especially significant to the singing voice. The cricothyroid muscle has two components: the oblique and transverse bellies. The external nerve inserts between the two bellies and lies in the normal position 2 mm below the insertion of the superior head of the sternothyroid muscle along the oblique line of the thyroid cartilage lamina. An attempt should always be made, where possible, to identify the nerve. Division of the superior aspect of the sternothyroid muscle often assists by improving exposure to the superior thyroid pole.

The SLN has also been known as the nerve of Amelita Galli-Curci , named after the famous soprano whose career was brought to a ruinous end in 1935 because of thyroid surgery performed under local anesthesia. Newspapers at the time wrote that after thyroid surgery, “the surprising voice is gone forever. The sad spectre of a ghost replaces the velvety softness.”

Bilateral VCP usually, but not always, presents acutely at extubation. If it is not recognized immediately after surgery, respiratory arrest can occur (in several hours or in the early days after surgery), which makes this even more dangerous. In cases of severe injury, both cords are usually in a paramedian position causing variable airway obstruction. In cases of patients with a bilateral VCP, 50% of patients will require some form of airway surgery; half of these surgeries will entail tracheostomy. The patient may have biphasic stridor and be in frank respiratory distress. As mentioned previously, patients do not necessarily have airway symptoms in the immediate postoperative period. The recently intubated glottis may be stented open for some period of time, so the glottis may narrow progressively in the ensuing hours and days postoperatively. The patient may, therefore, present at outpatient follow-up complaining of shortness of breath or stridor on exertion.

Reported prevalence of recurrent laryngeal nerve paralysis

The true incidence of VCP is unknown due to a failure of long-term consistency in adoption of the gold-standard of performing both a pre- and postoperative laryngeal examination in all patients undergoing thyroid and parathyroid surgery. What is known is that the true incidence is considerably underestimated and underreported.

A bedside assessment of the voice has poor sensitivity and specificity for vocal cord palsy. Validated thyroid voice questionnaires have been developed, but they fail to meet the accuracy of a laryngeal examination and cannot be recommended as an alternative.

A systematic review in 2009 by Jeannon et al. of 27 articles and 25,000 patients showed a wide variation in cord palsy based on the laryngeal examination method used. An average initial paralysis rate of 9.8% (range 2.3% to 26%) was found in their study.

The goal of the review was that anonymous reporting in national audit databases might provide a more accurate picture of palsy. In Sweden, the Scandinavian Quality Register (SQR) of 2008 reviewed 3660 thyroidectomies performed during a 12-month period. This analyzed data from 26 specialist endocrine surgical units from Sweden and Denmark and revealed an immediate 4.3% paralysis rate. Furthermore, this rate doubled when patients were submitted to routine laryngeal examination (as opposed to postoperative laryngoscopy performed only in patients with persistent and severe voice changes). The 2009 3rd BAETS National Audit report of 10,814 thyroidectomies reported a permanent palsy rate of 2.5% with 4.9% of patients reporting a change in voice. For first-time surgery, the reported incidence of a cord paralysis was 1.4% after lobectomy and 3.7% after total thyroidectomy. A significant increase in cord palsy was associated with revision surgery with rates of 5.4% for lobectomy and 6.9% for total thyroidectomy. This was from self-reported and unvalidated data entry from surgeons focusing on thyroid surgery; the surgeons in question did not uniformly conform to the gold-standard practice of laryngeal examination in all operations. As a result, these data are likely significant underestimations of true paralysis rates. Only a minority (21.5%) of patients in this 2009 audit underwent a postoperative laryngeal examination.

The 2012 4th BAETS National Audit reported a preoperative laryngeal examination rate of 60.9% in primary and 86.7% in revision surgery, respectively. Postoperative laryngeal examination occurred in less than 20% of patients with a higher than 20% rate of missing field data. As one might expect, postoperative paralysis rates were shown to increase with increasing laryngoscopy rates. When the postoperative laryngeal examination rate was less than 30%, 30% to 80%, and > 80%, respectively, which correlates with the rates revealed in the SQR 1.7%, 2.5%, and 4.2% paralysis rates were reported. Finally, the 2017 5th BAETS National Audit showed an increase in the preoperative laryngeal examination rate to 73% and 86% in primary and revision surgery, respectively; however, a postoperative laryngeal examination was performed in only 40% of patients, and there was more than 20% of missing field data. Therefore the true incidence of vocal cord palsy in the UK registry report of 2017 that analyzed 47,493 thyroidectomies remains unknown and could not be accurately reported.

The early diagnosis of cord paralysis is important because there is evidence that treatment within 3 months of surgery leads to significantly better patient outcome. There have been advances in the management of the paralyzed vocal cord with the proven safety and effectiveness of transcutaneous in-office injection.

Vocal cord injection is one of a number of techniques available to manage cord paralysis (see Chapter 42 , Pathophysiology of the Recurrent Laryngeal Nerve Injury, and 43, Management of Recurrent Laryngeal Nerve Paralysis).

An open thyroplasty method involves a skin incision and partial thyroid cartilage removal with placement of an implant or spacer material lateral to cord in the paraglottic space. This procedure may be combined with the arytenoid adduction, a technically difficult operation, where the arytenoid cartilage is repositioned through suture placement to reposition the cord; this is indicated where injection thyroplasty may be less suitable in patients with a large posterior glottic gap and, in particular, if the cord is at a different level than the opposite normal non paralyzed cord.

Glottic examination and the voice

Preoperative VCP can be caused by benign and malignant thyroid disease. It can also be present in the absence of any voice symptoms. This knowledge supports the recommendation for a preoperative laryngeal examination of all patients to help deliver and uniform gold-standard care for patients. The discrepancies between cord function on laryngeal examination and voice symptoms are due to many variables. Whether paralysis or paresis exist, the variability of contralateral cord compensation and other mechanisms are at play. Commonly in known cases of VCP, symptoms improve over time. This may be due to either a resumption of normal function or an evolution of cord position into a more favorable medial position. Only laryngeal examination will make a distinction between these two scenarios of “apparent or true” functional vocal cord recovery.