Safety Profile in Hyaluronic Acid‐Based Vocal Fold Injection Augmentation

ABSTRACT

Objective

The study's objective is to assess the safety profile and complication rates of hyaluronic acid‐based vocal fold injection augmentation (VFIA), specifically evaluating inflammatory reactions, injection volume, and COVID‐19 vaccination status.

Methods

A retrospective cohort study was conducted of 299 VFIA procedures performed using Restylane‐L at a single academic institution. Medical charts were reviewed to collect patient demographics, injection details (approach, location, and volume), videolaryngostroboscopy exams, and post‐injection outcomes, including complication rates and nature of inflammatory reactions.

Results

The overall complication rate was 4.0% (12/299), with inflammatory complications comprising 2.34% (7/299) of cases. Inflammatory symptoms developed a mean of 4.3 days after injection (median: 3 days; range: 12 h–10 days) and resolved in all cases with corticosteroid therapy. A statistically significant difference in injection volume was noted among groups (p = 0.037), although post hoc comparisons did not demonstrate significance. COVID‐19 vaccination status was not associated with complication rates (p = 0.84).

Conclusions

VFIA using HA‐based material demonstrates a low rate of complications, and all inflammatory reactions resolved with medical management. Injection volume may play a role in complication risk, but further multi‐site studies are needed. No association between COVID‐19 vaccination and adverse outcomes was identified.

Level of Evidence

3

Hyaluronic acid vocal fold injection augmentation (VFIA) is widely used to treat glottic insufficiency, but inflammatory complications remain incompletely characterized. In this retrospective cohort of 299 injections, the overall complication rate was low (4.0%), with inflammatory reactions occurring in 2.34% of cases and all resolving with corticosteroid therapy. Injection volume differed across complication groups on omnibus testing, suggesting a possible association with complication risk, while no relationship was identified between complications and prior COVID‐19 vaccination.

1. Introduction

Vocal fold injection augmentation (VFIA) with hyaluronic acid (HA) is a growing treatment for glottic insufficiency, including unilateral vocal fold paralysis (UVFP) and paresis, presbylarynges, and vocal fold atrophy [1]. HA is a biocompatible polysaccharide naturally present in the extracellular matrix. It is widely used for both cosmetic and otolaryngologic applications due to its viscoelastic properties, ease of injection, and ability to improve phonatory outcomes. However, despite its generally favorable safety profile, post‐injection complications, particularly inflammatory reactions, have been reported in multiple studies, raising concerns about hypersensitivity, vascular compromise, and material‐related reactions.

HA is a naturally occurring glycosaminoglycan and a key structural component of the vocal fold extracellular matrix, where it contributes to tissue viscosity, shock absorption, and preservation of the mucosal wave [2]. HA is particularly abundant in the intermediate layer of the lamina propria and plays an important role in vocal fold viscoelasticity and wound healing [2]. Experimental, in vivo, and histologic studies have demonstrated that HA‐based biomaterials are generally well tolerated within the vocal folds, preserving viscoelastic properties and eliciting only mild, self‐limited inflammatory responses without evidence of foreign body reaction or tissue necrosis [3, 4]. In vitro studies further support the biocompatibility of HA‐based materials, showing minimal inflammatory signaling and favorable extracellular matrix remodeling in vocal fold fibroblasts [5]. Despite this favorable safety profile, rare inflammatory reactions have been reported with specific HA formulations, suggesting that such events may be product‐ or formulation‐dependent rather than inherent to HA itself [6].

Previous literature has documented inflammatory complications following HA injection in both aesthetic and laryngologic uses, with reported rates of 0.6%–3.8% in vocal fold augmentation [7, 8, 9]. These reactions range from mild erythema and edema to severe glottic inflammation, dyspnea, and prolonged dysphonia, in some cases requiring hospitalization or airway intervention. The degree of HA cross‐linking, which varies by product and affects both structural integrity and degradation rate, has also been proposed as a potential contributor to immunogenicity and adverse reactions [10]. The precise etiology of these reactions remains unclear but is hypothesized to be related to immune‐mediated hypersensitivity or vascular congestion [11, 12, 13]. Additionally, acute and delayed inflammatory reactions to HA have been documented in dermatologic literature, with some studies suggesting an association with viral infections, immune dysregulation, and stress‐related immune activation. Recently, some of these factors that have become associated with COVID‐19 vaccination as well [14, 15].

Despite the growing body of research on HA injections, there is limited data on the incidence of complications, risk factors, and management strategies for inflammatory reactions in vocal fold augmentation. This study aimed to characterize the rate, presentation, and resolution of inflammatory reactions following Restylane‐L VFIA and to evaluate whether injection volume or COVID‐19 vaccination status were associated with increased complication risk.

2. Methods

This study is a single‐institution, retrospective cohort analysis of patients who underwent HA vocal fold injections over a 7‐year period. Data were collected from a single academic healthcare center with University of Virginia Institutional Review Board (IRB) approval (HSR240059) before participation. CPT code 31574 was used to identify 383 patients who received flexible laryngoscopy with injection(s) for augmentation. The study included 299 patients who received in‐office or operating room‐based HA injections for glottic insufficiency, including unilateral and bilateral vocal fold paresis, paralysis, and presbylarynges. Minimally crosslinked (< 1%) HA 20 mg/mL and 0.3% lidocaine (Restylane‐L, Galderma Laboratories, Dallas, TX) formulation was used for all patients receiving VFIA in this study. All procedures were performed by one of three fellowship‐trained laryngologists within the same academic practice. Eighty‐four patients with incomplete medical records, missing follow‐up data, injection into the velopharynx for velopharyngeal insufficiency or those who received non‐HA injectables were excluded.

Patient demographics, medical comorbidities (Charlson Comorbidity Index), history of allergies and atopy, and prior COVID‐19 vaccination status were recorded. Vaccination status was obtained from chart review of immunization records. For patients without explicit documentation, status was recorded as “unknown.” Allergy and atopy were extracted from chart review of documented past medical history. Procedural details included injection approach, laterality, volume of HA injected, and the location of the procedure (Operating Room, Inpatient Bedside, or In‐Office). The number of injection sites within the vocal fold was not systematically recorded. The primary outcome was overall complication rate, with secondary outcomes assessing complication characteristics, injection volume, and potential associations with COVID‐19 Vaccination. All complications were verified via laryngoscopic assessment, with inflammatory cases confirmed based on erythema, edema, and steroid responsiveness.

Data analysis was conducted using R Studio 2024.12.1. Comparisons between patients with and without complications were made using chi‐squared or Fisher's exact tests for categorical variables and Wilcoxon rank sum tests for continuous variables. Statistical significance was set at p < 0.05. Differences in injection volume across complication groups were assessed using a Kruskal–Wallis rank‐sum test, followed by pairwise Wilcoxon rank‐sum tests with Bonferroni correction for multiple comparisons.

3. Results

A total of 299 vocal fold injection augmentations using HA were performed. The average age was 63.9 years (±16.2), and the cohort was evenly distributed by sex (51.5% Female) (Table 1). The majority of patients identified as White/Caucasian (82.3%), followed by Black/African American (7.2%), Hispanic (4.9%), Asian/Pacific Islander (2.6%), American Indian or Alaskan Native (1.0%), and Other/unspecified (2.0%). The mean Charlson Comorbidity Index (CCI) was 4.11 (±2.70). Nearly half of the patients (47.2%) had a documented history of allergy, and 17.8% reported a history of atopic conditions, such as asthma, eczema, or allergic rhinitis.

TABLE 1.

Population demographics of patients undergoing vocal fold injection augmentation (N = 299).

Characteristics	n	% or mean (±SD)
Age (years)	299	63.7 ± 16.1
Sex
Female	154	51.5%
Male	145	48.5%
Race/ethnicity
White/Caucasian	251	82.3%
Black/African American	22	7.2%
Hispanic	15	4.9%
Asian/Pacific Islander	8	2.7%
American Indian/Alaska Native	3	1.0%
Other/unspecified	6	2.0%

Note: Baseline demographics of patients undergoing hyaluronic acid vocal fold injection augmentation (VFIA). Values are reported as mean ± standard deviation (SD) for continuous variables and number (percentage) for categorical variables. Race/ethnicity categories are not mutually exclusive; totals exceed 299 because some patients identified with more than one category.

With respect to COVID‐19 vaccination status, 113 patients (37.8%) were vaccinated prior to their injection, 156 (52.2%) were unvaccinated, and vaccination status was unknown for 30 patients (10.0%). A total of 62 procedures (20.7%) were performed in patients who had previously undergone HA vocal fold injection, while the remaining 237 (79.3%) were first‐time recipients. In terms of laterality, 45.2% of injections were performed on the left vocal fold, 27.1% on the right, and 27.7% were bilateral.

Most procedures were performed in‐office in the outpatient setting (82.8%), with the rest being conducted during inpatient admissions, either in the operating room (7.1%) or at the bedside (10.1%). Various injection approaches were utilized, including thyrohyoid (41.5%), transoral (20.1%), cricothyroid (19.4%), and direct laryngoscopy (11.0%), with the remainder categorized as unspecified (8.0%).

The most common indications for injection included vocal fold immobility or paralysis (177 patients), vocal fold atrophy (86 patients), and vocal fold hypomobility or paresis (54 patients) (Table 2). Less frequently, patients were injected for presbylarynges (6), sulcus vocalis/vocal fold scar (3), or other reasons (26). It is worth noting that multiple indications were present for some patients, and the total number of indications exceeds the number of patients accordingly.

TABLE 2.

Indications for vocal fold injection augmentation (n = 299).

Indication	Number of patients
Vocal fold immobility/paralysis	177
Vocal fold hypomobility/paresis	54
Vocal fold atrophy	86
Presbylaryngis	6
Sulcus vocalis/vocal fold scar	3
Other	26

Note: Clinical indications for hyaluronic acid vocal fold injection augmentation. Percentages are based on the total cohort (n = 299). Some patients had more than one indication recorded: specifically, 50 patients (16.7%) carried two indications, and three patients (1.0%) had three indications. Therefore, totals exceed 299.

The mean total injection volume across all procedures was 0.74 ± 0.45 mL, with a median of 0.675 mL and a range of 0.1–2.2 mL. Among complication groups, mean total injection volume was 0.73 ± 0.44 mL in patients without complications, 1.15 ± 0.62 mL in those with inflammatory complications, and 1.22 ± 0.66 mL in other complications (Table 3). Median total injection volumes were 0.65 mL (range 0.1–2.2 mL), 1.20 mL (range 0.4–2.0 mL), and 1.10 mL (range 0.5–2.0 mL), respectively. A statistically significant difference was observed among groups on omnibus testing (p = 0.037); however, no pairwise comparisons reached significance after correction for multiple testing.

TABLE 3.

Patient and procedural characteristics by complication status.

	All patients (n = 299)	No complications (n = 287)	Inflammatory complications (n = 7)	Other complications (n = 5)	p
Mean CCI value	4.11 (±2.70)	4.02 (±2.66)	4.29 (±2.98)	7.00 (±3.54)	0.1173 a
Allergy	47.2% (n = 141)	47.4% (n = 136)	14.3% (n = 1)	80.0% (n = 4)	0.07692 b
Atopy	17.8% (n = 53)	17.8% (n = 51)	28.6% (n = 2)	0%	0.4899 b
COVID19 vaccinated
Yes	37.8% (n = 113)	37.6% (n = 108)	28.6% (n = 2)	60.0% (n = 3)	0.8423 b
No	52.2% (n = 156)	52.3% (n = 150)	57.1% (n = 4)	40.0% (n = 2)
Unknown	10.0% (n = 30)	10.1% (n = 29)	14.3% (n = 1)	0%
Hx of prior injection
Yes	20.7% (n = 62)	21.3% (n = 61)	14.3% (n = 1)	0%	0.7324 b
No	79.3% (n = 237)	78.7% (n = 226)	85.7% (n = 6)	100% (n = 5)
Laterality
Left	45.2% (n = 135)	46.0% (n = 132)	14.3% (n = 1)	40.0% (n = 2)	0.1018 b
Right	27.1% (n = 81)	26.8% (n = 77)	57.1% (n = 4)	0%
Bilateral	27.7% (n = 83)	27.2% (n = 78)	28.6% (n = 2)	60.0% (n = 3)
Total injection amount c	0.74 (± 0.45)	0.73 (± 0.44)	1.15 (± 0.62)	1.22 (± 0.66)	0.0373 a

Note: General characteristics of patients undergoing HA‐based vocal fold injection augmentation, grouped by complication status. CCI represents the Charlson Comorbidity Index. p‐values calculated using Kruskal–Wallis or Fisher's exact tests as appropriate. Three patients had missing injection volume data (n = 296 for that analysis). All prior injections included in this category were with the same product, hyaluronic acid (Restylane‐L).

^a Kruskal–Wallis rank sum test.

^b Fisher's exact test.

^c Three patients had no injection amount recorded/found in their EMR. The sample sizes are total = 296, no complication = 285, Inflammation = 6, Other = 5.

To identify which specific groups differed, we conducted pairwise Wilcoxon Rank‐Sum tests with Bonferroni correction for multiple comparisons. The adjusted p‐values were as follows:

• Inflammation versus no complication: p = 0.22

• Inflammation versus other: p = 1.00

• No Complication versus other: p = 0.19

None of the pairwise comparisons were statistically significant at the alpha 0.05 level after Bonferroni correction. Despite the conservative Bonferroni adjustment likely contributing to the loss of significance, even less conservative corrections (Holm and Benjamini–Hochberg) did not yield any significant results.

A total of 12 complications (4.0%) were identified in the cohort, including seven inflammatory reactions (2.34%) (Figure 1). Of the five complications that were not attributed to an inflammatory response, there were two cases of dysphagia, one vocal fold hematoma, one vocal fold hemorrhage, and one instance of postprocedural dyspnea.

FIGURE 1.

A breakdown of complication types following HA‐based VFIA. Pie chart showing distribution of complications among 299 HA‐based VFIA procedures. Colors denote complication types; numbers indicate case counts per category.[Color figure can be viewed in the online issue, which is available at www.laryngoscope.com]

Inflammatory reactions were defined by the presence of erythema, edema, and/or decreased vocal fold pliability on laryngoscopic examination, with all patients demonstrating clinical improvement following corticosteroid treatment. All seven patients with inflammatory reactions presented with airway‐related symptoms that prompted repeat evaluation and laryngoscopy, most commonly dyspnea, shortness of breath, and a sensation of throat or airway swelling. One patient additionally reported dysphagia in conjunction with dyspnea. Of the seven patients with inflammatory reactions to the HA‐based filler, five experienced symptom resolution with per oral steroids, with or without antibiotics, taken on an outpatient basis. Oral corticosteroid regimens consisted of prednisone tapers ranging from 6 to 12 days, beginning at 40 mg daily with gradual dose reduction. Two patients were additionally treated with amoxicillin–clavulanate 875 mg/125 mg twice daily for 10 days. Two patients presented to the emergency department where they received intravenous steroids before being discharged home, neither requiring admission. One of these patients received a single 8‐mg dose of intravenous dexamethasone in the emergency department followed by a 6‐day oral prednisone taper (40 mg daily for 1 day, then tapered over 5 days). The other patient was evaluated at an outside hospital, where intravenous corticosteroids were administered without a steroid taper to follow; specific dosing details for this encounter were unavailable. In the patients with an inflammatory reaction, the average reported onset of complications was 4.33 days after the procedure, with a median of 3 and range from 12 h to 10 days. Symptom resolution occurred in all cases following corticosteroid therapy, with resolution documented within 1 week in three patients and within 3 months in the remaining four patients, although delayed follow‐up in some cases limited more precise determination of resolution timing.

The non‐inflammatory complications were attributed to hematoma formation, injection misplacement, or contribution from active anticoagulation (cited for the patient with Vocal Fold Hemorrhage). One patient who developed a vocal fold hematoma did require endotracheal intubation and was extubated 2 days later, eventually making a full recovery.

The average length of time to the post‐procedural follow‐up visit for all patients was 5.66 weeks, with a standard deviation of 5.95 weeks. The follow‐up duration ranged from a minimum of 1 day to a maximum of 41 weeks. There was no statistically significant difference in complication rates across procedure years (p = 0.72). Inflammatory complications were observed in one to three patients per year between 2019 and 2024, with no cases documented in 2017 or 2018. Other complications were infrequent and distributed sporadically across the study period.

4. Discussion

In this retrospective cohort study of 299 VFIA procedures with a HA based product, the overall complication rate was low, with only 12 adverse events (4.0%) identified. The majority of complications (58%) were inflammatory in nature, characterized by erythema, edema, and diminished vocal fold pliability on stroboscopic examination. These findings are consistent with previously reported inflammatory complication rates to HA, ranging from 0.6% to 3.8% in the literature [7, 8, 9]. Importantly, all inflammatory cases responded well to steroid therapy without long‐term sequelae, supporting the use of steroids as a first‐line intervention in these cases.

Although HA‐based fillers are generally well tolerated, these rare inflammatory reactions have been reported with specific HA formulations. Cosmetic filler literature suggests that these reactions are product‐ or formulation‐specific rather than inherent to HA itself, with proposed mechanisms including manufacturing‐related protein contamination or formulation‐dependent immunogenicity [6, 13]. Importantly, these inflammatory reactions are mechanistically distinct from vascular ischemic complications, which arise from extravascular compression or intravascular embolization and involve secondary inflammation rather than primary immune activation [16, 17]. The delayed onset, sterile presentation, and corticosteroid responsiveness observed in the present study are consistent with these inflammatory mechanisms.

Our analysis revealed a statistically significant difference in the total injection volume among the complication groups (p = 0.037). However, follow‐up pairwise comparisons did not identify significant differences between specific groups, limiting definitive conclusions. While these findings suggest that injection volume may contribute to complication risk, this association remains speculative. One potential explanation is that higher injection volumes may increase local tissue pressure within the confined paraglottic space, which could secondarily exacerbate inflammatory responses rather than serve as a primary immune trigger. Larger datasets are required to determine whether injection volume independently contributes to inflammatory risk following VFIA.

Non‐inflammatory complications in our cohort included dysphagia, hematoma, vocal fold hemorrhage, and dyspnea, with several likely attributable to technical considerations. For example, one dysphagia case was directly linked to injection misplacement, and the hematoma was attributed to traumatic injury during needle passage. These outcomes underscore the importance of meticulous technique and anatomical awareness during VFIA, especially in patients with distorted laryngeal anatomy or coagulopathy. Complications due to superficial injection into the superficial lamina propria (SLP) have been reported in the literature, resulting in altered vibratory function, suboptimal voice outcomes, or need for revision procedures [18].

Despite early case reports suggesting an association between COVID‐19 vaccination and delayed inflammatory reactions, our study found no evidence to support this link [13, 14, 15]. Among vaccinated patients, the rate of inflammatory complications did not differ significantly compared to unvaccinated individuals (p = 0.84), and the timing of vaccination relative to injection also showed no impact on complication risk. These findings offer reassurance regarding the immunologic safety of VFIA in the context of prior COVID‐19 vaccination and begin to challenge earlier concerns raised regarding vaccine‐induced reactions to HA‐based fillers.

Additionally, while previous investigations have explored the durability and efficacy of HA materials based on particle size, there has been limited discussion on how particle characteristics may relate to complication rates [19]. Our data is limited to only one formulary of Restylane‐L, and these findings suggest that volume and technique may influence complication development alongside structural and chemical filler properties. However, due to the lack of diversity of HA formulations assessed in this study, additional work is warranted to assess whether material‐specific variables play a role in inflammation or hypersensitivity. One study found 9/150 (6%) of patients receiving Restylane‐L injections for facial cosmetic procedures were found to have preprocedure antibodies to Restylane but there was no increased risk of site reaction compared to patients without antibodies [20]. Preprocedural antibody formation has been hypothesized to relate to antigenic similarity or residual proteins from streptococcal fermentation [20].

A significant limitation of this study is the relatively low incidence of complications, particularly inflammatory reactions (n = 7). Although adequate for describing clinical presentation and outcomes, this small event count substantially limits statistical power and prevents meaningful subgroup analysis. With so few inflammatory cases, pairwise comparisons are underpowered and at high risk of Type II error. These limitations also reduce generalizability, especially given the single‐institution, retrospective design and exclusive use of Restylane‐L. Larger, multicenter datasets are needed to validate potential risk factors for inflammatory complications following VFIA.

The low rate of complications in this large cohort supports the overall safety profile of VFIA using HA. However, caution is advised when using higher injection volumes or treating complex anatomy. Clinicians should maintain a high index of suspicion for adverse events when patients present with post‐procedure changes in voice, breathing, or laryngeal sensation, particularly within the first week following the procedure.

5. Conclusion

This large, single‐institution retrospective study evaluating the safety profile of HA based vocal fold injection augmentations demonstrated a low rate (2.34%) of inflammatory reactions to the HA‐based filler. All inflammatory complications resolved following treatment with steroids, and none resulted in any long term complications. Although a statistically significant difference in total injection volume was observed between complication and no complication groups (p = 0.037), the post hoc analysis did not confirm significance between specific pairs, suggesting that although injection volume may contribute to the risk, the findings may be underpowered. Larger, multi‐centered studies are needed to validate the association.

No significant associations were found between complications and COVID‐19 vaccination status which challenges previously suggested links in the literature, supporting the continued safety of HA‐based fillers in VFIA. While our results reinforce the overall safety profile of HA, they do acknowledge the importance of cautious injection technique and awareness of injection volume. Ongoing multi‐site investigations are underway to expand upon these findings with the goal of better characterizing predictors of adverse outcomes in VFIA.

Funding

The authors have nothing to report.

Conflicts of Interest

The senior author (J.J.D.) is a co‐inventor of intellectual property related to biomaterial development for vocal fold injection augmentation.

Thornton N., Stepp R., Nguyenkhoa V., Leder Macek A., and Daniero J. J., “Safety Profile in Hyaluronic Acid‐Based Vocal Fold Injection Augmentation,” The Laryngoscope 136, no. 8 (2026): 3541–3547, 10.1002/lary.70513.

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.