Prevalence of Upper Aerodigestive Symptoms in Patients Who Underwent Thyroidectomy With and Without the Use of Intraoperative Laryngeal Nerve Monitoring

Introduction

U pper aerodigestive symptoms (UADS) after thyroidectomy may have multiple causes and at least some of these may be related to laryngeal nerve injury. Hormonal alterations (1, 2), postorotracheal intubation trauma, local infections, postoperative bleeding, and temporary dysfunction of cervical musculature (3 –6) are among the reported risk factors for UADS. The use of intraoperative neuromonitoring (IONM) has recently been considered as a potential method of reducing UADS. It has been suggested that IONM facilitates the identification and preservation of both the recurrent and external branch of the superior laryngeal nerves (7, 8). However, some studies do not find that the rate of vocal fold palsy was reduced by neuromonitoring in patients undergoing thyroidectomy (9, 10). Thus, the impact of IONM on UADS after thyroidectomy is not clear.

According to Musholt et al. (11), patients who have a normal signal during monitoring of the laryngeal nerves and have a preserved vocal fold mobility confirmed by laryngoscopy may still report postoperative voice alterations, thought to be of complex origin and due to multiple factors.

The most common complaints of individuals with preserved vocal fold mobility post-thyroidectomy are vocal fatigue, hoarseness, difficulty speaking loudly and in high pitch, choking, the sensation of a foreign body in the pharynx, and neck soreness (5,6,12 –15). Hong and Kim (5), Stojadinovic et al. (16), and Soylu et al. (17) have reported that most of these subjective complaints are temporary and disappear within a few months after surgery. However, Aluffi et al. (6), Pereira et al. (12), and Rosato et al. (13) reported that patients may continue to experience UADS after 1–4 years post-thyroidectomy.

The long-term prevalence and impact of UADS is a topic that requires further evaluation. Rosato et al. (13) and Lombardi et al. (18) consider that a thorough investigation of the prevalence of and underlying reasons for these symptoms will promote a more accurate and effective therapeutic approach, thus enhancing patient satisfaction and quality of life.

The aim of the present study was to examine the prevalence of UADS one year after thyroidectomy and to determine the relationship of IONM to the development of UADS.

Materials and Methods

A cross-sectional study in a consecutive series of adult patients who underwent thyroid surgery in the Head and Neck Surgery and Otorhinolaryngology Department at the A.C. Camargo Hospital in Sao Paulo, Brazil, was performed. This study evaluated patients without alterations in laryngeal or thyroid hormonal function, whose evaluation by videolaryngoscopy indicated that vocal fold mobility was preserved both before and after thyroidectomy at least one year after this surgery. No patients in this study had a history of gastroesophageal reflux disease (GERD) or presurgical findings of laryngeal edema or hyperemia based on a videolaryngoscopic examination. Patients were assigned to two groups. The control group (CG) included patients who underwent a thyroidectomy without the use of IONM. The neuromonitored group (NMG) comprised patients who underwent a thyroidectomy with the use of IONM.

Data collected included the extent of surgery performed (total or partial), the performance of paratracheal neck dissection, size of their thyroid nodule, their histological diagnosis, and whether they received post-thyroidectomy radioiodine therapy. Patients in the CG and NMG answered a questionnaire regarding UADS designed by us based on a previous study (19). In the previous study (19), the Voice Handicap Index questionnaire was used; however, this instrument did not address some symptoms reported by thyroidectomized patients, and we therefore designed the UADS-related questionnaire to include these symptoms. The same examiner contacted patients by telephone (1–4 years after thyroidectomy) to complete the questionnaires. Symptoms that were temporary were considered negative responses.

The post-thyroidectomy UADS questionnaire was divided into two parts, one relating to voice symptoms and the other to swallowing symptoms. Voice symptoms include vocal fatigue, difficulty speaking loudly or in a high pitch, hoarseness, and a low- or high-pitched voice. Swallowing symptoms include burning during deglutition, pain during deglutition, pharyngeal pain during deglutition, choking, dry throat, discomfort during deglutition, throat-clearing during deglutition, scar stiffness during swallowing, a sensation of a foreign body in the pharynx, strangling during deglutition, and coughing during deglutition (Appendix I). This questionnaire asks patients about the presence of voice and deglutition symptoms. Positive responses to the questionnaire were quantified according to a 4-point Likert scale that ranked the symptoms as not a problem, a minor problem, a moderate problem, or a major problem. In addition, patients were asked whether there were other complaints not addressed by the questionnaire. UADS that qualified as not a problem or a minor problem were considered to cause a low degree of disturbance. UADS qualified as a moderate problem or a major problem were considered to cause a high degree of disturbance. It should be noted that at the time of this study the questionnaire instrument was a subjective assessment that had not been psychometrically validated.

The head and neck surgical team at our institution has a standardized surgical approach for thyroid surgery. A meticulous dissection and direct visualization of the parathyroid and recurrent laryngeal nerve is performed in all patients before inferior vascular pedicle ligation; the vessels of the superior thyroid pole are also individually dissected and ligated to avoid trauma to the external branch of the superior laryngeal nerve.

In the NMG, a Medtronic Xomed Nerve Integrity Monitor-2^® (NIM-2, Jacksonville, FL) electromyographic endotracheal tube was used for IONM. This endotracheal tube contains electrodes along its wall that are placed against the true vocal folds. The grounding wires were placed in the subcutaneous tissue of the presternum area, and the nerve was stimulated with the Medtronic Xomed Prass monopolar nerve stimulator, usually adjusted to 1.0 mA. When the nerve was directly touched by the nerve stimulator, a characteristic audible beep confirmed its identification, and the electromyographic signal was recorded by the monitoring device. The use of IONM helped in the identification of the recurrent laryngeal nerve, sometimes even before it was directly viewed. The use of this equipment decreased the need for extensive dissections and prolonged manipulations over and around the nerve, which would theoretically lead to reduced rates of nerve injury. In addition, its use allowed the identification of the external branch of the superior laryngeal nerve and any anatomical variations in the recurrent laryngeal nerve, such as nerve bifurcation and nonrecurrent laryngeal nerves, and made it possible to verify that the nerve was intact at the end of the procedure. The criteria for IONM use were at the convenience and discretion of the staff surgeon.

All patients undergoing thyroidectomy at the institution are routinely discharged the first day after surgery unless a major complication occurs and a prolonged hospital stay is needed, for example, re-operation, severe hypocalcemia, or respiratory insufficiency.

All patients included in this study signed a consent form, and the Institution Ethics Committee approved this study.

For statistical analyses, the chi-square test was applied because the variables compared between groups had no normal distribution, and p<0.05 was considered to be significant. All analyses were made using SPSS software (SPSS Corporation, Chicago, IL).

Results

All data were collected between May 2006 and July 2007 from 308 patients who underwent thyroidectomy. Of these patients, 208 were classified as the CG and 100 were classified as the NMG. Both groups were similar with regard to demographic, clinical, and surgical aspects (Table 1). The nodule(s) size in both groups was also quite similar; most of the nodules in both groups (>80%) were <3 cm (p=0.280). At the time of interview, all patients had normal thyroid hormone levels at their most recent examination.

There was no significant difference between groups with regard to surgical complication rates. In the CG and the NMG, 10% and 8% of patients experienced temporary hypoparathyroidism, respectively. In both groups, <2% of the patients had seroma/hematoma that did not require re-operation, 1% of patients had site infection, and no patients needed re-operation in either group. All patients with vocal fold immobility were excluded from this series (as described in the Materials and Methods section). No study patient required a prolonged hospital stay.

As shown in Table 2, there was no significant difference in the prevalence of overall UADS in CG and NMG patients. However, a separate analysis of voice and swallowing symptoms indicated swallowing complaints were more prevalent in the CG (p=0.036).

A higher proportion of patients reported only one UADS in the NMG, and none of the patients in this group had more than five complaints. In contrast, 23.6% of patients in the CG had more than five UADS, which was statistically significant (Table 3).

The most prevalent symptoms in both groups were a sensation of a foreign body in the pharynx, vocal fatigue, hoarseness, dry throat, choking, and throat-clearing during deglutition, a low-pitched voice, and difficulty speaking loudly (Table 4). Other reported symptoms included difficulty singing (4.8% in CG; 6% in NMG), absence of voice (1.4% in CG; 2% in NMG), and difficulty swallowing solid food (1% in CG; 2% in NMG). Patients in the NMG had less high degree UADS symptoms.

There was no relationship between having postoperative radioactive iodine treatment and the prevalence of UADS in patients in the CG (p=1.000) or patients in the NMG (p=0.741).

A relationship between total thyroidectomy and a higher prevalence of UADS (p=0.049) was found only in the CG (Table 5). When voice and swallowing complaints were analyzed separately, patients who underwent total thyroidectomy reported more swallowing complaints than those who underwent partial thyroidectomy in the CG (p=0.022). No significant relationship was observed between the prevalence of UADS and the extent of surgery in the NMG.

Discussion

Voice symptoms and swallowing complaints occur frequently after thyroidectomy, even in the absence of apparent laryngeal nerve injury, and these problems may have several causes (12 –14,17 –22). Pereira et al. (12) reported a 45% prevalence of UADS after an average of 4 years post-thyroidectomy in patients with no clinical evidence of neural injury. The present study confirms these results and shows that the prevalence of UADS is 42.8% in patients who underwent thyroidectomy without IONM and 39% in patients who underwent thyroidectomy with neuromonitoring. UADS may be described as a moderate or major problem to these individuals even after an average period of 2.8 years postsurgery.

These findings are relevant and emphasize the need for long-term evaluation of patients undergoing thyroidectomy, notwithstanding the absence of laryngeal nerve injury. Few individuals spontaneously report the presence of UADS in low or high degrees of disturbance. Patient reticence may be due to the fact that they consider UADS as an inevitable and irreversible outcome of surgery. In clinical practice, voice professionals, who likely have a more accurate perception of voice alterations and perceive their impact on postsurgery quality of life, have been found to spontaneously report complaints more often.

An analysis of complaints of permanent voice alterations in both groups determined a prevalence of 25%–27%. The most frequent symptoms reported with a high degree of disturbance and their respective prevalences in the CG and the NMG were vocal fatigue (14.4%; 9%), hoarseness (14.4%; 9%), difficulty speaking in a high-pitched (7.2%; 6%), and a low-pitched voice (8.6%; 5%). Other studies also confirm that there is a higher prevalence of these symptoms (5,6,12,14,18,20,21). Although symptoms were not significantly more prevalent in the CG, patients who underwent IONM of the laryngeal nerves during thyroidectomy had a reduced prevalence and a lower number of UADS of the higher degree of disturbances, thus suggesting that IONM promotes the reduction of long-term negative surgical effects. Similar results were found by Lombardi et al. (18) in patients who underwent video-assisted thyroidectomy as compared with conventional surgery.

In the CG, 4.8% of the patients spontaneously reported difficulties in singing as compared to 6% of patients in the NMG. However, the prevalence of this complaint may be underestimated because individuals were not questioned about this symptom directly. Alterations in singing ability have been mentioned by other authors, and a prevalence of 6%–7% was described in patients with preserved vocal fold mobility post-thyroidectomy (5,11,14,22). Musholt et al. (11) noted that a significant difference existed between pre- and postoperative parameters with regard to singing after thyroidectomy. These complaints indicate that thyroidectomy often has a negative effect on singing, especially in professional singers.

There are not many reports of post-thyroidectomy swallowing symptoms in the absence of neural injury. Pereira et al. (12) report that the sensation of a foreign body in the pharynx is a frequent post-thyroidectomy symptom; this complaint was also identified in the CG and the NMG with respective frequencies of 21% and 6%. Other symptoms included choking, dry throat, and throat clearing. Scar stiffness during deglutition was reported only in the CG. This symptom was probably related to extralaryngeal alterations, such as post-thyroidectomy scar retraction.

Lombardi et al. (14) reported that swallowing alterations are more frequent and tend to linger for a longer period than voice alterations after surgery. These findings are consistent with results obtained for the CG in the present study in which it was found a slightly, but significantly more prevalence of swallowing symptoms in the CG than in the NMG. However, the prevalence of voice symptoms in the NMG was quite similar to that in the CG.

The possible reasons for these symptoms to linger in patients even a year after thyroidectomy include alterations of the extrinsic perithyroidal neural plexus that innervate the pharyngeal and laryngeal structures (12), whose impact are unknown. Debruyne et al. (4) and Soylu et al. (17) reported that alterations in the laryngeal vascular supply occurred after removal of the thyroid gland. Other studies (4 –6,13,18,20) have observed malfunction of the extralaryngeal muscles, laryngo-tracheal fixation, or scar retraction. The extrinsic laryngeal muscles enable vertical movement of the hyolaryngeal skeleton (23), an essential movement in swallowing and phonation.

No significant difference was observed in the overall prevalence of UADS between the CG and the NMG. Nevertheless, swallowing complaints were significantly less prevalent in the NMG; fewer UADS were reported by the same individual, and a decrease in the severity of UADS was also reported. A significant difference in UADS of those who had total as compared to partial thyroidectomy was noted only in the CG, particularly with regard to swallowing complaints.

Although the surgical technique was identical in both groups, the lesser prevalence observed in swallowing symptoms in the NMG might be related to implementation of a less extensive dissection and thus less manipulation and denervation of surrounding thyroid tissues resulting from better identification of the recurrent and superior laryngeal nerves. This likely reduces scarring and fibrosis formation in the operating field, decreasing a potential negative impact in the laryngeal vertical movement.

Our results suggest that, aside from documented recurrent laryngeal nerve lesions, the use of IONM could reduce, but not eliminate long-term voice and swallowing surgical complications. Notably, in our previous prospective series we did not find a difference in vocal fold mobility between the CG and the NMG. We observed a vocal fold immobility rate (partial or total) of 3.1% and 3.4% for the nerves at risk, respectively, at 3 months after surgery (24).

Song and Shemen (25) indicate that IONM does not interfere with the surgical technique. IONM can assist in laryngeal nerve identification (7, 8) and can preclude significant manipulation of the area by the surgeon, which can result in minor injury to the muscle and laryngeal vascular supply. Long-term results suggest that the prevalence of UADS is reduced in conjunction with a reduction in the degree of disturbance caused by these symptoms and a minor impact on the incidence of major surgical procedures. However, the role of IONM in reducing the incidence of laryngeal nerve injury, and the reasons for permanent UADS in the absence of neural injury, require further studies.

Some reports have shown that vocal alterations are associated with hypothyroidism, leading to alterations in the vocal fold mucosa, such as myxedema or an increased accumulation of hyaluronic acid (1, 2). However, Page et al. (22) did not report any contribution of hypothyroidism to the incidence of voice symptoms post-thyroidectomy. According to Stemple et al. (1), vocal alterations due to hyperthyroidism are mainly present as a slight instability, including vocal trembling, air shortage, reduced loudness, and a dry cough. In the present study, all patients had normal thyroid hormone levels, and no correlations between UADS and thyroid hormone levels could be detected. However, it is important to note that only those medical records related to hormonal alterations were examined, which could be incomplete or inadequate because some patients did not undergo regular follow-up evaluations.

The relationship between the prevalence of long-term UADS and radioiodine therapy was investigated. No relationship between isotope Iodine-131 radiation and voice or swallowing complaints was detected. Isolan-Cury et al. (26) obtained the same result in a study of patients with Graves' disease who did not undergo thyroidectomy, but who underwent treatment with iodine therapy. Few other studies on the short- and long-term effects of iodine therapy on the larynx are available.

One of the limitations of the present study is the lack of perceptual-auditory, acoustic, and laryngological analyses one year after thyroid surgery. However, previous studies (14,16,27) show that inconsistencies exists between voice complaints and the results of these evaluations. According to Lombardi et al. (14), this discrepancy is noted in about one-third of the evaluated patients. Pereira et al. (12) stated that many UADS cannot be related to laryngological alterations. Another possible limitation of this study is the lack of a routine esophageal pH-test because most UADS could be similar to GERD. However, patients with laryngeal and deglutition symptoms before thyroidectomy as well as patients with hyperemia or edema of the larynx identified in the preoperative videolaryngoscopic examination were excluded from the present study. These exclusion factors could eliminate most patients with possible GERD.

In summary, the results of this study reveal that UADS are common long after thyroidectomy in patients in whom the vocal fold is preserved. IONM during thyroidectomy of the laryngeal nerve did not eliminate symptoms, but the use of this technique reduced the prevalence, degree of disturbance, and the impact of extensive surgeries on UADS. The use of IONM during thyroid surgery seems to have a positive impact in terms of reducing surgical complications other than apparent laryngeal nerve injury.