Early Injection Laryngoplasty With a Long-Lasting Material in Patients With Potentially Recoverable Unilateral Vocal Fold Paralysis.

OBJECTIVES: Injection laryngoplasty (IL) is one of the major options for treatment of unilateral vocal fold paralysis (UVFP). Early IL in patients with abrupt-onset UVFP can reduce hoarseness and aspiration-related discomfort and complications. Temporary or short-lasting materials are recommended for the early IL since permanent or long-lasting materials may negatively affect the voice quality when vocal fold motion is spontaneously recovered.
METHODS: Patients who received IL with a long-lasting material (ArteSense) within 1 month following development of postoperative UVFP were enrolled for this study. They were categorized into a spontaneously recovered group (n=30) and unrecovered group (n=276) from UVFP. The subjective and objective voice parameters were collected before and 6 months after IL. Patients' demographics and collected voice parameters were compared between the two groups.
RESULTS: Age and sex distributions were not different between the two groups. Aspiration symptom, maximum phonation time (MPT), jitter, shimmer, noise to harmonic ratio (NHR), overall grade (G) and breathiness (B) of GRBAS (grade of dysphonia, roughness, breathiness, asthenia, strain) scale score, and voice handicap index (VHI) were significantly improved after IL in both groups. The amounts of improvement in MPT, jitter, shimmer, NHR and VHI were significantly greater in the recovered group. There was no IL-related complication including granuloma formation, hypersensitivity to injection material, or worsening of hoarseness.
CONCLUSION: Early IL with a long-lasting material can be safely performed to improve voice quality and to reduce aspiration episode for the patients with postoperative UVFP, regardless of recovery from the paralysis. The amounts of improvement in the subjective and objective voice parameters were significantly greater in patients who showed spontaneous recovery from UVFP when compared with those in patients who did not.

INTRODUCTION

Unexpected unilateral vocal fold paralysis (UVFP) is one of the embarrassing complications of the thoracic and neck surgery. Clinical courses of postoperative UVFP are diverse including wide range of spontaneous recovery, synkinesis and permanent paralysis. Most of the spontaneous recovery from UVFP is obtained within a year, but it is difficult to predict whether UVFP will recover and when it will recover [1].

Abrupt-onset UVFP may cause hoarseness and aspiration, which can result in poor quality of life, longer hospital stay, and delay in the planned treatments [2-4]. Patients with abrupt-onset UVFP have higher chances of invasive procedures including bronchoscopy and tracheostomy [4]. Injection laryngoplasty (IL) became a major treatment option for the UVFP because of its noninvasiveness and development of biocompatible injection materials [3]. Recent studies have supported for the early IL in case of postoperative UVFP, since it can be easily performed without adding further morbidities and it will help to prevent or reduce UVFP-related discomfort and complications [5,6].

When performing an early IL in patients with potentially recoverable UVFP, temporary or short-lasting injection materials are preferred since long-acting materials may adversely affect the voice quality when UVFP is spontaneously recovered. Short-lasting materials including hyaluronic acid, collagen, or cellulose are usually resolved within 3 to 6 months [3]. Regarding the early IL for postoperative UVFP, most of the previous studies used short-lasting materials, therefore, many patients required additional procedures [7,8]. There is a study which used a long-lasting material for the early IL in potentially recoverable UVFP patients; it reported that the patients achieved adequate voice quality without further needs of open phonosurgery, but it did not provide objective voice quality data [9].

In this study, we evaluated the voice outcomes of early IL with a long-lasting material in patients with potentially recoverable postoperative UVFP and compared the subjective and objective voice parameters between the recovered and unrecovered patients from UVFP.

MATERIALS AND METHODS

Patients

This retrospective study was approved by Samsung Medical Center Institutional Review Board (IRB No. 2018-08-062) and informed consent was exempted. The medical records of patients who received IL between May 2004 and May 2016 at the authors’ institution were reviewed. Enrolled patients received an IL within a month after onset of UVFP following surgeries of the thyroid, esophagus, thymus or lung cancers. The patients who had no possibility of spontaneous vocal fold motion recovery due to the transection of the vagus nerve, recurrent laryngeal nerve or direct tumor invasion to these nerves were excluded.

Total 306 patients were enrolled, and they were classified into two groups; spontaneously recovered from UVFP (recovered group, n=30) and unrecovered group (n=276). Vocal fold mobility and subjective and objective voice parameters were evaluated at 6 months after IL procedures. Most of the patients were evaluated again at 1 year after IL to assess delayed complications including granuloma formation and to determine the need for further management of UVFP.

IL procedure

All of IL procedures were performed through cricothyroid membrane approach under local anesthesia. For local anesthesia, patients inhaled 4% lidocaine using nebulizer for 5 minutes, and 2% lidocaine injection was performed at the subcutaneous layer around the cricothyroid membrane. Flexible fiberoptic nasopharyngoscope (VNL-1530T; Pentax, Tokyo, Japan) was introduced through nasal cavity to visualize the vocal fold. The cricothyroid membrane was identified by palpation and 23-guage spinal needle was inserted into paralyzed vocal fold until the needle tip was bluntly seen at the lateral and anterior side of the vocal process of arytenoid cartilage. We used ArteSense (European Medical Contract Manufacturing B.V., Nijmegen, The Netherlands) which consists of 20% volume of polymethylmethacrylate (PMMA) microspheres suspended in 80% volume of denatured bovine collagen. In average, 0.7 mL of ArteSense was injected to get slight overcorrection of volume of the paralyzed vocal fold. Cautions were given to inject slowly, to keep the plane at deeper layer of vocal ligament and subglottic portion of the paralyzed vocal fold. Patients were observed for 2 hours after IL to check acute complications including allergic reaction, bleeding and dyspnea.

Voice evaluations

The objective and subjective voice evaluations were performed before and 6 months after IL. Subjective aspiration symptom was also checked. Laryngeal examination using stroboscope (Pentax, Lincoln Park, NJ, USA) was performed to evaluate the vocal fold mobility, patterns of glottic closure and vibratory phase symmetry. Maximum phonation time (MPT), jitter percentage, shimmer percentage, and noise to harmonic ratio (NHR) were evaluated for aerodynamic and acoustic analyses. These parameters were calculated using the software of Computerized Speech Lab (Kay Elemetrics, Pine Brook, NJ, USA). The GRBAS (grade of dysphonia, roughness, breathiness, asthenia, strain) scales [10] were used for perceptual evaluation by two experienced speech language pathologists. The patients replied Korean version of the voice handicap index (VHI)-30 and VHI-10 which is a patients-based self-rating questionnaires including the subscales of the functional, physical, and emotional aspects of the voice impairment [11,12].

Statistical analysis

Baseline characteristics and voice related clinical factors were compared between the recovered and unrecovered group. Voice parameters obtained before and after IL were compared by paired sample t-test in each group. The amounts of improvement in voice parameters (Δvoice parameter) after IL were compared between the two groups by the Mann-Whitney U-test. A P-value <0.05 was considered as statistically significant. Statistical analyses were performed using IBM SPSS ver. 20.0 (IBM Corp., Armonk, NY, USA).

RESULTS

Baseline characteristics of the enrolled patients

Baseline characteristics of the patients were shown at Table 1. The age at the time of IL and sex distribution were not different between the two groups. The etiologies of vocal fold paralysis were neck surgery (n=12, 40.0%) and thoracic surgery (n=18, 60.0%) in the recovered group, and neck surgery (n=139, 50.4%) and thoracic surgery (n=137, 49.6%) in the unrecovered group, which were not statistically different between the two groups (P=0.281).

Table 1.

Baseline characteristics of the spontaneously recovered group and unrecovered group from unilateral vocal fold paralysis

Clinical factor	Recovered (n=30)	Unrecovered (n=276)	P-value
Baseline characteristics
Age (yr)	53.4±11.7	55.3±14.2	0.286
Sex (male:female)	15:15 (50.0:50.0)	153:123 (55.4:44.6)	0.570
Cause of procedure			0.281
Neck surgery	12 (40.0)	139 (50.4)
Thoracic & esophagus surgery	18 (60.0)	137 (49.6)
Vocal fold recovery state
Full recovery	23 (76.7)
Partial recovery	7 (23.3)

Values are presented as mean±standard deviation or number (%).

Voice evaluations before and after IL

The values of subjective and objective voice parameters were summarized in Table 2. In the recovered group, nine patients (30%) reported aspiration symptoms before IL, which improved in all patients after IL. In contrast, in the unrecovered group, aspiration symptom was observed in 54 patients (19.6%) before IL and seven patients (2.5%) after IL. The MPT significantly improved after IL in both groups (P<0.001). The MPT improved from 4.9±2.9 to 12.2±5.8 seconds in the recovered group and from 5.0±3.9 to 7.7±5.5 seconds in the unrecovered group. The jitter, shimmer and NHR also significantly improved in both groups (P<0.001). The jitter and shimmer more markedly improved in the recovered group than in the unrecovered group. Regarding GRBAS scores, overall grade of dysphonia (G), breathiness (B) and asthenia (A) improved significantly in both groups (P<0.001). The VHI-30 and VHI-10 improved significantly after IL in both groups (P<0.001).

Table 2.

Voice outcomes before and after early injection laryngoplasty in patients with spontaneously recovered and unrecovered unilateral vocal fold paralysis

Variable	Recovered (n=30)			Unrecovered (n=276)
	Before	After	P-value	Before	After	P-value
Aspiration	9 (30)	0	<0.001	54 (19.6)	7 (2.5)	<0.001
MPT (sec)	4.9±2.9	12.2±5.8	<0.001	5.0±3.9	7.7±5.5	<0.001
Jitter (%)	5.5±4.3	1.7±1.6	<0.001	5.8±5.2	3.4±3.9	<0.001
Shimmer (%)	11.3±8.4	4.0±3.2	<0.001	10.9±7.7	7.5±6.5	<0.001
NHR	0.2±0.2	0.1±0.0	<0.001	0.3±0.2	0.2±0.2	<0.001
GRBAS score
Grade of dysphonia	2.7±0.5	0.7±0.7	<0.001	2.5±0.7	1.3±0.9	<0.001
Roughness	0.3±0.8	0.4±0.6	0.424	0.5±0.9	0.4±0.7	0.272
Breathiness	2.7±0.5	0.5±0.7	<0.001	2.4±0.7	1.2±0.9	<0.001
Asthenia	0.1±0.5	0	<0.001	0.3±0.7	0.1±0.4	<0.001
Strain	0	0.0±0.2	0.500	0.1±0.3	0.1±0.3	0.424
VHI-30	82.5±28.8	27.7±29.0	<0.001	79.5±26.1	55.8±31.0	<0.001
VHI-10	29.0±8.3	9.4±10.3	<0.001	28.6±8.7	20.3±11.6	<0.001

Values are presented as number (%) or mean±standard deviation.

MPT, maximum phonation time; NHR, noise to harmonic ratio; GRBAS, grade of dysphonia, roughness, breathiness, asthenia, strain; VHI, voice handicap index.

Changes in voice parameters of each patient of the recovered group are depicted in Fig. 1. Most of the patients showed improvement of jitter, shimmer, NHR, MPT, VHI-30 and G. None of the recovered group patients experienced worsening of VHI-30 and G by the early IL with a long-lasting material. Stroboscopic evaluations confirmed that mucosal waves were not hampered in all of the patients. Follow-up laryngoscopic evaluation at 1 year after IL revealed no case of delayed complications including granuloma formation.

Fig. 1.

Voice parameters before and after injection laryngoplasty in each patient (n=30) of spontaneous recovery from unilateral vocal fold paralysis. MPT, maximum phonation time; VHI, voice handicap index; NHR, noise to harmonic ratio.

Amounts of voice improvement between the recovered and unrecovered group

The amounts of voice improvement were summarized in Table 3. ΔMPT (Δ, difference between before and after injection laryngoplasty) was significantly longer (P<0.001) in the recovered group (7.3±6.9 seconds) than in the unrecovered group (2.7±5.6 seconds). ΔJitter was significantly greater (P=0.026) in the recovered group (–3.8%±3.8%) than in the unrecovered group (–2.3%±5.2%). In contrast, ΔShimmer and ΔNHR were not different between the two groups (P=0.172 and P=0.315, respectively). Regarding the GRBAS scales, G and B improved more in the recovered group (P<0.001). ΔVHI-30 and ΔVHI-10 showed significantly greater improvement (P<0.001) in the recovered group than in the unrecovered group.

Table 3.

Variance of voice parameters between before and after injection laryngoplasty in patients with spontaneously recovered and unrecovered unilateral vocal fold paralysis

Variable	Recovered (n=30)	Unrecovered (n=276)	P-value
∆MPT (sec)	7.3±6.9	2.7±5.6	<0.001
∆Jitter (%)	–3.8±3.8	–2.3±5.2	0.026
∆Shimmer (%)	–6.1±8.8	–3.4±8.5	0.172
∆NHR	–0.1±0.2	–0.1±0.4	0.315
∆GRBAS score
Grade of dysphonia	–2.0±0.9	–1.1±1.0	<0.001
Roughness	0.1±1.0	–0.1±0.9	0.175
Breathiness	–2.2±0.8	–1.2±1.0	<0.001
Asthenia	–0.1±0.5	–0.2±0.7	0.509
Strain	0.0±0.2	–0.0±0.4	0.341
∆VHI30	–54.9±31.5	–23.7±30.9	<0.001
∆VHI10	–19.6±9.9	–7.7±11.3	<0.001

Values are presented as mean±standard deviation.

Δ, difference between before and after injection laryngoplasty; MPT, maximum phonation time; NHR, noise to harmonic ratio; GRBAS, grade of dysphonia, roughness, breathiness, asthenia, strain; VHI, voice handicap index.

DISCUSSION

Postoperative UVFP is not uncommon complication of the extensive neck or thoracic surgery. The prognosis of UVFP is difficult to predict. Laryngeal electromyography is a good predictor of poor prognosis for UVFP, but accuracy of prediction for complete recovery is only 12.8% [13,14]. Because it is difficult to clearly foretell whether and when UVFP will recover, management of postoperative UVFP should be focused to reduce the patients’ discomfort and related complications. IL, medialization thyroplasty, and arytenoid adduction are available surgical options for the management of UVFP [15,16]. Early IL has become a major treatment option for the acute-onset postoperative UVFP because of its effectiveness, noninvasiveness and low rate of complications [5,6]. The best timing for early IL is controversial. Previous studies defined an early injection as from 1 day to 6 months after occurrence of UVFP [5-7]. In our institution, we usually performed IL within 1 month in the hope to provide the better quality of life and to avoid a delay in the planned treatment for their main diseases.

Injection materials can be categorized into temporary (short-lasting) and permanent (long-lasting) materials according to the duration of effects in the body. Permanent (long-lasting) injection materials include autologous fat, calcium hydroxylapatite (Radiesse), PMMA (ArteSense) and polydimethylsiloxane (or particulate silicone). Temporary injection materials include bovine gelatin (Gelfoam, Surgifoam), collagen-based products (Cymetra, Zyplast, Cosmoplast/Cosmoderm), hyaluronic acid (Restylane, Hylaform), and carboxymethylcellulose (Radiesse Voice Gel). According to the literatures, “permanent/long-lasting” materials typically refer to lasting more than 1 year [17,18]. In 2006, ArteFill (the same product of ArteSense) became the first permanent injectable wrinkle filler to receive U.S. Food and Drug Administration approval [19]. ArteSense is a mixture of denatured bovine collagen (short-lasting) and PMMA (long-lasting component). The PMMA microspheres become encapsulated with the collagen fibers secreted by the patients’ own fibroblasts and stabilized for more than several years [20]. A 5-year follow-up study in U.S. has shown the same wrinkle improvement as seen at 6 months. Long-term follow-up in European patients has shown wrinkle correction lasting up to 15 years [19].

In patients with potentially recoverable UVFP, IL with a temporary or short-lasting material is usually preferred because there are concerns that a permanent or long-lasting material may disturb the natural course of vocal fold recovery and/or injection material remained in vocal fold may reduce vocal fold mucosal wave, which can result in unfavorable voice quality. However, even collagen, a representative of short-lasting materials not infrequently have long term effects on voice quality [21]. According to previous reports, satisfactory voice outcomes were maintained after IL with a long-lasting material in patients who later showed spontaneous recovery from UVFP [9,22]. In addition, IL with a long-lasting material is a useful procedure not only for UVFP but also for the other types of glottal insufficiency including vocal atrophy, in which bilateral vocal folds are fully mobile. It means that IL with a long-lasting material will be beneficial in selected cases of glottal insufficiency regardless of vocal fold mobility. A meta-analysis reported that complete recovery occurred in 13%–83% and partial recovery occurred in 19%–83% of UVFP [1]. Partially recovered vocal folds often result in glottal insufficiency because of the atrophy of intrinsic vocal fold muscles as well as incomplete adduction during phonation. Therefore, IL will be helpful for the more complete glottal closure of these partially recovered vocal folds.

In this study, none of the patients experienced acute adverse reactions including dyspnea, bleeding, hematoma, or allergic reaction to the injection materials. Theoretically bovine collagen allergy testing is required to minimize the risk of hypersensitivity reactions. The substance used as a carrier for the microbeads of ArteSense is the collagen derivative, atelocollagen. Enzymatic digestion of the antigenic end of natural bovine collagen minimizes the risk of allergic reactions (product inserts). Reports of allergic reactions to ArteFill are extremely rare (0.2%) [19]. PMMA microsphere of ArteSense has a smooth, round surface and is suspended in collagen; it allows tissue ingrowth into the interstitial space around the microspheres. This decreases the displacement of the filler. PMMA microsphere is big enough (30 to 50 μm in diameter) to avoid phagocytosis by macrophages and giant cell formation resulting in inflammation [23]. During the last 15 years, the authors have used ArteSense (formerly labeled as Artecoll) for more than 1,500 cases of IL. There has been no long-term or delayed complication including migration, granuloma or foreign body reaction. Mucosal waves of the vocal fold did not decrease regardless of spontaneous recovery from UVFP.

A previous study of potentially recoverable UVFP investigated voice related quality of life but not the other objective voice parameters [9]. According to a meta-analysis study, MPT, GRBAS, and mean airflow rate were recognized as significant indicators of voice quality in patients with UVFP [24]. In this study, MPT, jitter, shimmer, NHR, GRBAS, and VHI significantly improved after early IL with a long-lasting material in both groups. The patients of recovered group showed significantly greater improvement in MPT, jitter, and VHI than unrecovered patients. Notably, none of the recovered patients reported worsening of VHI-30 or G in GRBAS. Based on these results, we support that early IL with a long-lasting material is effective in improving voice quality even in patients who may show spontaneous recovery from postoperative UVFP.

There are several limitations of this study. Comparing voice outcomes between the recovered and unrecovered group does not necessarily tell us the true effects of early IL with a long-lasting material, especially for the patients of recovered group. In the recovered group, their voice outcomes will be affected by both IL procedure and natural course of recovery from UVFP. To specify the effects of IL in the recovered group, data should be compared in patients between with and without IL procedure. Moreover, we used only a long-lasting injection material and performed IL within a month of UVFP onset. Further studies comparing short-lasting with long-lasting materials and early with delayed timing of IL will provide more reliable conclusions regarding the use of a long-lasting material for the early IL in potentially recoverable UVFP patients. In addition, we analyzed only short-term follow-up data (6 months). If evaluated at more than 1 year of UVFP onset, more patients in the unrecovered group may be re-classified into the recovered group and voice outcomes may be different according to the degree of recovery from UVFP. Moreover, long-term follow-up will provide us more information regarding advantages as well as disadvantages of using long-lasting material for the early IL in patients with acute-onset UVFP.

When patients have significant discomfort in their voice or swallowing because of abrupt-onset UVFP after the neck or thoracic surgery, early IL with a long-lasting material can be safely performed to improve their voice quality and to reduce aspiration-related discomfort, even in patients who will show a spontaneous recovery from UVFP.

▪ • Early injection laryngoplasty (IL) with a long-lasting material can be safely performed in patients with postoperative unilateral vocal fold paralysis (UVFP).

• Subjective and objective voice parameters significantly improved after early IL, regardless of recovery status from the paralysis.

• The amounts of improvement in the subjective and objective voice parameters were significantly greater in patients who showed spontaneous recovery from UVFP.