Introduction

Unilateral vocal fold paralysis (UVFP) is the most common neurogenic disorder affecting the larynx. The technique of medialisation of the paralysed vocal fold depends on several factors like the position of the paralysed vocal fold, severity of symptoms, the cause for paralysis, the overall medical condition of the patient, the chance for recovery or compensation, and the personal and professional vocal needs of the patient [1]. Despite the increasing popularity and availability of long-acting injection materials, laryngeal framework surgery (LFS) remains the gold standard for long-term treatment of unrecovering UVFP.

LFS for vocal fold medialisation was first conceptualised by Payr [2]. In 1974, Dr. Nobuhiko Isshiki described medialisation of the paralysed vocal fold using Silastic implant, placed between the thyroid cartilage and the inner perichondrium [3]. However, this did not gain general acceptance till the technique was further refined and popularised by Dr. James Koufman [1] and Dr. Netterville [4].

The purpose of this paper is to describe a series of patients who have undergone medialisation thyroplasty (with or without arytenoid adduction) at our centre with respect to demographics, aetiology for UVFP, size of the silastic implant used, amount of anterior and posterior medialisation required and complications.

Materials and methods

Of all the patients (n = 560) presented to the voice clinic for unilateral vocal fold paralysis at our centre from August 2008 to August 2014, 449 patients (80.18 %) responded to speech therapy, 44 patients (7.86 %) underwent injection laryngoplasty while the remaining 67 patients (11.96 %) underwent medialisation thyroplasty, with or without arytenoid adduction.

Data pertaining to age, gender, aetiology of UVFP and symptoms was noted. The side of the vocal fold paralysis, phonatory gap and arytenoid symmetry was recorded on videostroboscopic examination. Pre-operative and post-operative maximum phonation times (MPT) were noted.

All the medialisation thyroplasties were performed using the Netterville technique [4, 5] by a single surgeon (first author) using the Netterville thyroplasty instrument set, marketed by Medtronics. All cases were performed under local anaesthesia with an intra-operative flexible laryngoscopy monitoring.

A thyroplasty window of (13 × 6) mm was created in males and (11 × 5) mm in females or small larynges, 5–7 mm from the midline (5 mm in females and 7 mm in males). The window was made 3 mm from the inferior border of the thyroid cartilage, irrespective of the gender.

The inner perichondrium was incised along the four borders of the window, but not removed. Dexamethasone Injection (8 mg) was administered intravenously at this stage of the operation. Elevation was performed between the cartilage and inner perichondrium all around the window (anterior elevation was minimal). A medialisation probe was used to determine the optimum amount of anterior and posterior medialisation, and the appropriate plane of medialisation in this window. A partially pre-fabricated implant was used (Netterville implant, Medtronics), which was further carved depending on the medialisation specifications and fashioned into the final implant. A flexible laryngoscopic examination was performed after the placement of the implant, and the patient was asked to phonate. The quality of voice was subjectively assessed by both surgeon and patient, and the MPT was checked.

If the phonatory gap closed completely (assessed by flexible laryngoscopy) and the intra-operative voice was satisfactory with a reasonable increase in MPT (minimum increase of 5 s), then a Type I medialisation thyroplasty was performed.

Arytenoid adduction was performed only in case of a persistent posterior glottic gap. This posterior gap was either due to the arytenoid being malrotated or a level difference between the two vocal folds.

Arytenoid adduction was performed using the Netterville’s technique. A Maragos’ window was created in the posterolateral part of the thyroid cartilage. The pyriform sinus mucosa was dissected away gently in a posterior and superior direction, and the muscular process of the arytenoid cartilage was palpated for. The cartilage is always better felt than seen initially, and once it was clearly identified, a 4-0 Prolene suture was passed twice through the muscular process in a figure of 8 manner. The sutures were then pulled anteriorly through the thyroplasty window, and then through the thyroid cartilage midline near the lower thyroid border and tied using appropriate force, which was determined by asking the patient to phonate. Typically, a loose knot was made, following which the silastic implant was introduced into the thyroplasty window.

Prior to placing a drain, the pyriform mucosa integrity was confirmed by instilling few cc of saline over the field and asking the patient to blow as though “blowing out birthday candles”. Ballooning of the mucosa and the absence of bubbles confirmed pyriform mucosa integrity.

The size of the silastic implant used, the amount of anterior and posterior medialisation required and complications, if any, were noted. In case of revision surgeries, indication for revision surgery and the intervention needed was tabulated.

All the patients were given post-operative speech therapy to eliminate the compensatory mechanisms, which most patients of UVFP develop in a bid to decrease the breathiness and hoarseness.

The follow-up period varied, as most of the patients came from remote places. The median follow-up period for all patients was 2.3 years (range 1.1–5.4 years).

Observations and results

Demographics

A total of 67 patients underwent medialisation thyroplasty at our centre during the study period. Of the 67 patients, 51 were males while 16 were females.

Mean age of the study group was 50.40 years (SD 12.06 years).

The mean age in females was 48.38 years (SD 11.93 years), and 51.04 years (SD 12.14 years) in males. The maximum numbers of patients were in the age group of 51–60 years (33.3 %).

Symptomatology

Hoarseness of voice was the most common presenting symptom, seen in all 67 cases, followed by an inadequate expectoration of mucous (57 %), air hunger (40 %), aspiration (27 %) and vocal fatigue (19 %).

Aetiology

The commonest cause for UVFP was post-operative surgical complication followed by idiopathic cause, as shown in Table 1.

Table 1 Aetiology for unilateral vocal fold paralysis in our study
Full size table

Laterality

In our study group, 73 % cases had left sided vocal fold paralysis, while 27 % cases had right sided paralysis.

Amount of anterior and posterior medialisation

We measured the amount of medialisation required anteriorly and posteriorly. The average anterior medialisation needed was 2.25 mm (SD 1.05 mm). The average posterior medialisation needed was 6.75 mm (SD 1.79 mm).

Further, we grouped the posterior medialisation needed into those requiring less than 3 mm of medialisation between 4 and 5 mm and more than 5 mm. About 77.6 % (52 cases) needed posterior medialisation of >5 mm while 16.4 % (11 cases) needed between 4 and 5 mm, and 6 % (4 cases) needed posterior medialisation of <3 mm.

We compared the amount of medialisation required in males and females and found no significant difference. In males, the thyroid cartilage is bigger in size, but the angle between the thyroid alae is acute (90°), conversely in a female, the cartilage is smaller in size, but the angle between the thyroid alae is obtuse (120°). Hence, the amount of medialisation required is similar in males and females.

The amount of medialisation required was directly related to the position of the vocal fold (e.g. Paramedian, cadaveric). We did not find a co-relation between the position of the vocal fold and the aetiology for the UVFP.

Thus, the amount of medialisation required was not influenced by the gender of the patient or the cause for UVFP.

Outcomes

Besides the subjective voice improvement, we used an intra-operative flexible laryngoscopic evaluation as well as MPT to evaluate the operative results.

In our study, the mean pre-operative MPT was 4.8 s, while the mean post-operative MPT on-table was 12.5 s with an average gain in MPT of 7.7 s. The MPT and the position of the implant remained stable in all patients during the study period.

None of the patients required any additional procedures for improving vocal outcomes.

Arytenoid adduction

Arytenoid adduction was needed in 13 patients (19.4 %). This was an on-table decision for all patients as already described in “Materials and methods”. The pre-operative MPT was found to be less than 5 s in all the 13 patients who required arytenoid adduction. These 13 patients, who required arytenoid adduction, also demonstrated arytenoid asymmetry on phonation on videostroboscopy. Interestingly, there were 27 patients with a pre-operative MPT of less than 5 s who did not require arytenoid adduction for adequate phonatory gap closure. These patients demonstrated arytenoid symmetry on phonation and they improved with only a Type 1 thyroplasty.

Complications

With regards to complications, we had a pyriform fossa mucosal tear in one patient (1.4 %), for which primary closure was done and a feeding tube was inserted. On the fifth day, FEES (functional endoscopic evaluation of swallowing) study was done and oral feeds started as there was no leakage of ingested fluids.

Revision thyroplasty

None of our patients, who underwent medialisation thyroplasty, needed a revision surgery during the study period. However, we had 7 (10.4 %) patients who had already undergone medialisation elsewhere and underwent revision surgery with us. Out of seven, one patient had a vertically placed implant. He needed a revision thyroplasty with the placement of a new implant along with arytenoid adduction.

In two cases, the implant that was placed was too big, causing a strained voice. It was replaced by an implant of a smaller size. In three patients, the implant was very small with no significant voice outcome, for which a bigger implant had to be kept. In another patient, a Gore–Tex implant had migrated and had to be removed. After 6 months, a revision surgery was performed by us with a silastic implant along with the arytenoid adduction.

Discussion

Silastic medialisation of vocal folds is largely indicated in glottal incompetence which results secondary to unilateral vocal fold paralysis, unilateral vocal fold paresis and selected traumatic defects. Bilateral silastic medialisation may be considered for symptomatic senile bowing or presbylaryngis. Contraindications include impaired abduction of contralateral vocal fold, previous laryngeal irradiation, traumatic laryngeal defects, and previous Teflon injection being among others [5].

The mean age in our study group was 52 years with a range of 16 to 81 years. In other studies the mean age was around 62 years with a wide age range [6, 7].

In our study group, 73 % cases had left sided paralysis, while 27 % cases had right sided paralysis. This finding was similar to other standard studies, which showed 82.2 and 85 % incidence of paralysis of left side [6, 7].

The most common cause for unilateral vocal fold paralysis was iatrogenic. Among the surgical procedures, thyroid surgery was the most common cause for trauma to the recurrent laryngeal nerve causing unilateral vocal fold paralysis in our study which is comparable to other studies done [8–11].

The various other aetiologies are compared with other studies in Table 2.

Table 2 Comparison of aetiology for UVFP with other studies
Full size table

We found hoarseness of voice as the most common complaint, which was also observed in earlier studies done [7, 12]. We have compared our observations with other studies in Table 3.

Table 3 Comparison of symptoms of UVFP reported in various studies
Full size table

There has been considerable debate over the location of the thyroplasty window. Many formulas have been described to accurately locate this window just lateral to the vocal fold.

In 1986, Dr. Koufman reported a series of 11 patients who had undergone medialisation laryngoplasty. The window size was determined using the Koufman formula. He reported the amount of medialisation required as 2–4 mm in his series [1].

Table 4 Complication rates reported in various studies
Full size table

We performed all the thyroplasties using the Netterville technique. Silastic medialisation described by Netterville et al. [4, 5] is an easy and adaptable technique. In this technique, the window is used as an entry into the paraglottic space, using an implant system that allows one infinite variability in placing the vocal fold in the physiologic phonating position, no matter where the position of the vocal fold is in relation to the window.

According to Netterville, the average depth of medialisation needed in most men and women is approximately 5 mm in the posterior aspect of the implant, with 1 mm or less medialisation needed in the anterior aspect of the window [13].

In our study, the average anterior medialisation needed was 2.25 mm (SD 1.05 mm), and the average posterior medialisation needed was 6.75 mm (SD 1.79 mm).

MPT is used to evaluate the competence of the phonatory control and respiratory support, and is influenced by many factors such as sex, age and physical characteristics of the respiratory system. Hence, the post-operative MPT may still fall outside the normal range. According to Bielamowicz, he finds MPT as the single most reliable indicator of vocal function, and we agree with the same [14]. Lundy et al. concluded that the intra-operative measurement of MPT was predictive of their post-operative counterparts, and that MPT is a good measure to predict post-operative outcome of medialisation thyroplasty [8].

Lu et al. compared the pre- and post-operative MPT and found significant improvement from a mean of 6.21 to 12.80 s which almost remained stable when tested at 1, 3 and 6 months post-operatively [11].

Leder et al. also noted similar results with a mean pre-operative MPT of 5.7 s and post-operative MPT of 9.8 s [15]. Chrobok et al. also found an improvement of MPT from 6.5 s pre-operatively to 12.5 s post-operatively [12].

Kraus et al. studied the MPT changes in a group undergoing both silastic medialisation and arytenoid adduction, and found that from 6.7 s, the MPT improved to 13.2 s [16].

Previous studies have shown that asymmetry of the arytenoid complex during adduction is common and can be seen in persons without any vocal symptoms [17, 18]. We, therefore, conclude that a pre-operative MPT of less than 5 s with arytenoid asymmetry should alert the clinician for the possible need for arytenoid adduction.

The overall complication rate for both AA and ML was found to be 8.5 % by Rosen [19]. Of which 13.8 % was reported as having post-operative airway compromise for ML or AA (9.5 % for ML alone). The extrusion rate of implant was 0.81 % (Table 4).

Chrobok et al. reported a complication rate of 5.6 % and sub-optimal voice outcome in 5.6 % cases. Lu et al. had no wound infections or airway compromise, but 6 % had implant extrusions. Young et al. [20] had an overall complication rate of 15 % with airway compromise in 2.2 % requiring intervention and sub-optimal voice outcomes in 4 %. Of the implants extruded, 0.5 % was extruded medially and 0.3 % was laterally extruded. Cotter et al. [21] reported 9.8 % extrusion rate.

Sub-optimal outcomes are the cause for revision medialisation thyroplasty, which may present as persistence of symptoms or worsening of symptoms. The cause of which may be a displaced implant, a very small or a very large implant, incorrectly placed implant or inability to detect the contralateral paralysis of vocal fold.

According to Rosen, the revision rate was 5.5 %, a revision rate 6 % was observed by Young et al., Koufman [22] reported 12.5 %, while Anderson et al. [23] had 24 % revision rate.

Lu had to perform a revision surgery in 13 % patients at a later date for either implant repositioning or placement of larger implant. Young had to revise the surgery to place a larger implant in 37 %, a smaller implant was required in 7.9 % and implant repositioned in 23.7 %. In 10.3 %, arytenoid adduction was added, and in 19.7 % vocal fold injection augmentation was added.

The association of complication rates and overall experience of the surgeon have already been studied by Rosen and Young et al.

Conclusion

  • We saw a significant improvement in MPT (mean gain of 7.7 s) after surgery.

Intra-operative measurement of MPT is a reliable predictor of the post-operative outcome.

  • Arytenoid adduction was needed in 20 % of cases. Pre-operative MPT of less than 5 s with arytenoid asymmetry should alert the clinician to the possible need for arytenoid adduction. It is important to anticipate the need for arytenoid adduction as it decreases the chances of a revision surgery.

  • Our study is the first to determine the amount of anterior and posterior medialisation needed in the Indian population. The average anterior medialisation needed was 2.25 mm (SD 1.05 mm) while the average posterior medialisation needed was 6.75 mm (SD 1.79 mm).

Mean anterior and posterior medialisation required was found to be the same, regardless of age, gender and side of surgery.