Hearing thresholds for pure tones from 500 to 16 000 Hz are shown for the MeroGel-treated (an esterified hyaluronic acid) (group 1) (A) and the absorbable gelatin sponge (AGS)–treated (group 2) (B) guinea pigs. A comparison of gains in auditory brainstem recording threshold sensitivities for the 500- to 4000-Hz range of frequency stimuli at postoperative week 6 is presented for group 1 and 2 animals (C). SPL indicates sound pressure level; single asterisk, P<.05; double asterisks, P<.01; and triple asterisks, P<.001. See "Materials and Methods section for explanation of groups.
A cross-section of the middle ear cavity of a MeroGel-treated (an esterified hyaluronic acid) animal (group 1) at postoperative week 6. A, View of the membranelike structure (MR) and normal-appearing mucosal membranes (MEM) (toluidine blue, original magnification ×62.5). RWM indicates round window membrane niche; ST, scala tympani. B, View of the inflammatory cells (ICs) with the MR (toluidine blue, original magnification ×500).
A cross-section of the middle ear cavity of an absorbable gelatin sponge (AGS)–treated animal at postoperative week 6. A, View of residua of AGS filling of the middle ear cavity at the lesion site (toluidine blue, original magnification ×500). ST indicates scala tympani; RWM, round window membrane; SV, scala vestibuli; and SM, scala media. B, View of newly formed bone (NB) within the AGS residua (toluidine blue, original magnification ×62.5). C indicates bony wall of the cochlea.
A cross-section of the middle ear cavity (MEC) of an untreated (group 3) animal at postoperative week 6. The MEC is free of debris. The mucosal membranes (MEM) appear normal (toluidine blue, original magnification ×125). BM indicates basilar membrane; ST, scala tympani.
Li G, Feghali JG, Dinces E, McElveen J, Van De Water TR. Evaluation of Esterified Hyaluronic Acid as Middle Ear–Packing Material. Arch Otolaryngol Head Neck Surg. 2001;127(5):534-539. doi:10.1001/archotol.127.5.534
Copyright 2001 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.2001
To evaluate the efficacy of esterified hyaluronic acid (MeroGel) as a middle ear (ME)–packing material.
Randomized controlled trial.
Twenty-four guinea pigs.
Group 1, MeroGel-treated animals (n = 10), bilateral wounding of ME mucosa with 5 of the animals receiving the MeroGel packing in the left ME and 5 of the animals receiving MeroGel in the right ME; group 2, absorbable gelatin sponge–treated animals (n = 10), with the same experimental protocol as in group 1 except that the absorbable gelatin sponge was the packing material; group 3, untreated animals (n = 4), unilateral wounding of the left ME mucosa in 2 animals and in 2 animals in the right ME, with no packing material. Auditory brainstem recordings were performed for all groups before the ME operation and 5 days and 6 weeks after the operation.
Auditory brainstem response recordings at postoperative day 5 showed that all ears with ME packing had hearing losses in the frequency range of 500 to 4000 Hz. The recovery of hearing acuity at postoperative week 6 was significantly better in group 1 (MeroGel-treated) guinea pigs compared with group 2 (the absorbable gelatin sponge–treated) animals. In group 2 animals, 20% of the packing material remained in the ME cavities and new bone formation was observed, while in group 1 animals, there was less packing material in the ME and no formation of new bone.
MeroGel is a nonototoxic packing material with a high level of biocompatibility for ME mucosa; it is an effective supportive material following ME surgery and is easily expelled from the ME cavity.
HYALURONIC ACID (HA) is a naturally occurring extracellular polysaccharide with a molecular weight in the range of 2 to 4 million daltons. Hyaluronic acid is a component of the extracellular matrix of many tissues within the body,1 making it highly biocompatible. Hyaluronic acid is frequently used as a supportive aid in ophthalmologic surgery.2 MeroGel (Medtronic Xomed Inc, Jacksonville, Fla) is made from an esterified form of HA (HYAFF; Fidia Advanced Biopolymer srl, Abano Terme, Italy).
The supportive material commonly used since the late 1950s in otological surgery is the absorbable gelatin sponge (AGS) that is manufactured from a selected grade of animal skin gelatin. There are, however, some reports that indicate possible deleterious effects secondary to the use of AGS in the middle ears (MEs) of animals3 and in patients following a stapedectomy.4,5
There is an interest in using HA as a supportive material in ME surgery. It is highly biocompatible and seems to have no ototoxic effect on the inner ear.6- 8 Hyaluronic acid has also been used as packing in clinical otological surgery and in animal experiments.9,10 These studies have suggested that HA was equivalent to or superior to AGS.
The aims of this study were 3-fold to: evaluate MeroGel as a supportive substance during wound healing in the ME, determine if MeroGel has any toxic effects on either ME mucosa or the hearing threshold, and compare the effects of MeroGel packing with those of AGS packing on wound healing under identical laboratory conditions.
All animals in this study were treated in accord with the guidelines of the Albert Einstein College of Medicine, Institutional Animal Care and Use Committee, Bronx, NY. Twenty-four, 4-week-old guinea pigs (Dunkin Hartley, Charles River, Wilmington, Mass), weighing from 300 to 350 g, were divided into 3 groups. Group 1 (MeroGel recipients) consisted of 10 guinea pigs that underwent bilateral wounding of ME mucosa. Five animals received MeroGel packing in their left ME and the remaining 5 animals received MeroGel packing in their right ear. Group 2 (AGS packing recipients; Pharmacia & Upjohn Inc, Stockholm, Sweden) consisted of 10 guinea pigs, with the same experimental plan as in group 1 except AGS was used to pack the ME cavities. The volume and positioning of packing material used in the MEs was standardized for the animals in groups 1 and 2. Group 3 (untreated) consisted of 4 guinea pigs with unilateral wounding of the ME mucosa. In 2 guinea pigs the left ear was wounded and in the remaining 2 animals the right ear was wounded. No packing material was placed in the MEs of group 3 animals.
A series of 3 ABR tests for pure-tone stimuli were performed just prior to the ME wounding operations, on postoperative day 5 and at postoperative week 6 week. For each ABR recording session, guinea pigs were anesthetized with a mixture of intramuscular xylazine hydrochloride (15 mg/kg) and ketamine hydrochloride (35 mg/kg). The ABRs were measured using the Tucker-Davis Technology System II (Tucker-Davis Technology, Gainsville, Fla) connected to a personal computer (model E3110; Gateway, San Diego, Calif). Pure-tone stimuli were digitally synthesized using SigGen software (Tucker-Davis Technology) and presented through an earphone (Etymotic ER-2; Etymotic Research, Elk Grove Village, Ill). Tone pips were presented at octave intervals from 500 to 16 000 Hz. Animals were presented with a stimulus intensity series, from a 90- to 10-dB sound pressure level (SPL) in 5-dB steps. Each ABR data point for an individual animal consisted of an average of 500 stimulus presentations recorded over a 10-millisecond period. Electrical activity was recorded via a platinum needle electrode (Grass Telefactor, West Warwick, RI) inserted into the scalp at the vertex, referenced to another needle electrode in a neck muscle. A third needle ground electrode was placed at the tympanic bulla. The responses at or near threshold were confirmed on a second testing. Threshold was defined as the lowest intensity capable of producing a reproducible ABR waveform with good morphology of wave III or V response. These ABR data were collected bilaterally from all guinea pigs.
Prior to surgery guinea pigs were anesthetized with a combined mixture of intramuscular xylazine hydrochloride (15 mg/kg) and ketamine hydrochloride (35 mg/kg). Following a retroauriclular incision, tympanic bullae were opened with a cutting bur. Middle ear lesions were bilateral in groups 1 and 2 and unilateral in group 3. A 3 × 3-mm area of promontory ME mucosa close to the round window niche was scraped away with a right angle elevator. Equivalent volumes of isotonic sodium chloride-moistened packing material (ie, MeroGel and AGS) were determined by side-by-side comparison as viewed using a dissecting microscope and then each volume was subdivided for packing the MEs of animals from groups 1 and 2. Packing material was placed in the posterior inferior quadrant of each ME cavity. The packed area started anterior to the wound margin on the promontory and anterosuperior to both the posterior crus of the stapes and long process of the incus and continued until the opening of the bulla. Thus, the packing material partially covered the promontory area near the round window membrane niche, partially surrounded the posterior crus of the stapes and long process of the incus, and completely filled the whole round window membrane niche. Care was taken to ensure the continuity of the ossicular chain and the integrity of the tympanic membrane in all animals. To control for variations in surgical technique all procedures were performed by only one of us (G.L.).
Six weeks after creating a lesion in ME mucosa, animals were deeply anesthetized and transcardially perfused with 4% paraformaldehyde. Temporal bones were harvested and tympanic bullae carefully opened. At postoperative week 6, the amount of MeroGel and AGS packing material left in the ME was evaluated, using the following system of scoring: 0, no packing material remained; 1, a membranelike sheet was present; and 2, a mass less than 20% of the original packing material volume remained. Evaluation of the MEs was done using a surgical microscope and was performed by the same observer (G.L.). Dissected temporal bones were then fixed overnight in 4% paraformaldehyde at 4°C. The specimens were transferred into a formic acid decalcification solution (Immunocal; Decal Corp, Congers, NY) and were decalcified for 3 days at room temperature. Decalcified temporal bones were dehydrated in increasing concentrations of ethyl alcohol and subsequently embedded in paraffin. Seven-micrometer-thick serial sections of the area of interest were cut, stained with toluidine blue, and inspected using a light microscope (ZeissAxiophot; Carl Zeiss Microscope Systems, Oberkochen, Germany).
The χ2 test was used to determine the level of statistical significance between the amounts of MeroGel and AGS left in the ME cavities. A t test was used to evaluate the level of significance between ABR changes in the animals in all 3 groups. Statistical significance was set at P<.05.
For group 1 (unilateral MeroGel packing, bilateral wounding) a low-frequency hearing loss (ie, 30- to 40-dB SPL) was noted in the MeroGel-packed ears at postoperative day 5 when compared with the preoperative ABR thresholds. This was statistically significant for 500-, 1000-, 2000-, and 4000-Hz pure-tone stimuli (P<.001). Eighty percent of the guinea pigs with MeroGel-packed ears (group 1) had a full recovery from these depressed ABR thresholds by postoperative week 6 recovering to preoperative ABR threshold levels of sensitivity (Figure 1A). Twenty percent of the animals in group 1 had a partial recovery of the ABR threshold (ie, 15- to 20-dB–SPL gain). There were no significant threshold differences between preoperative and 6 weeks' postoperative ABRs for all frequencies (all values P>.05).
For group 2 (unilateral AGS packing, bilateral wounding) a statistically significant low-frequency hearing loss (ie, 30- to 40-dB SPL) was also noted in the AGS-packed ears at postoperative day 5 compared with preoperative ABR thresholds, for 500-, 1000-, 2000-, and 4000-Hz pure-tone stimuli (all values P<.001). At postoperative week 6 the AGS-packed ears showed some recovery of the hearing loss (ie, a 15- to 20-dB SPL gain), but none of these ears recovered an ABR hearing threshold that was equivalent to their preoperative levels of sensitivity (Figure 1B). There were statistically significant differences between the low-frequency preoperative ABR hearing thresholds and the postoperative 6-week thresholds for group 2, for 500-, 1000-, 2000-, and 4000-Hz pure-tone stimuli (P<.001). However, the gains in the ABR hearing threshold between the low-frequency, postoperative 5-day ABR hearing thresholds, and the postoperative 6-week ABR hearing thresholds for group 2 were sufficient to show statistical significance (all values P<.001).
All animals with packed MEs in group 1 (MeroGel) and group 2 (AGS) had similar frequency ranges and hearing losses at postoperative day 5 (Figure 1, respectively). There was a significant recovery of ABR hearing thresholds in both group 1 and group 2 animals at postoperative week 6; however, a comparison of the ABR hearing thresholds showed that the animals in group 1 had a greater improvement in their recovery of ABR hearing thresholds than those in group 2, ie, P<.05 for 500 and 4000 Hz; P<.001 for 1000 Hz; and P<.01 for 2000 Hz (Figure 1C).
For groups 1 through 3 all nonpacked ears (ie, either wounded or unoperated on) showed no significant changes in ABR hearing thresholds throughout the course of this study (all values P>.05). In addition, there were no observed high-frequency hearing deficits in any of the animals tested, ie, 8000- to 16 000-Hz range, and the ultrahigh frequencies (ie, >16 000 Hz) were not measured in this study.
Some microscopic residua of packing material was identifiable in the MEs for both group 1 and 2 animals at postoperative week 6. Both the amount and quality of the remaining ME-packing material differed greatly between the animals in group 1 and group 2. A transparent, membranelike sheet was found in 80% of MeroGel (group 1)-packed MEs at or near the wounding site (Figure 2A). Only 2 of the group 1 MEs had a masslike residua of MeroGel that was less than 20% of the original volume. In the AGS-packed MEs, a retained AGS mass was present in the ME of every animal and the volume of this retained AGS ranged from more than 20% to 50% of the original volume of packing material. All of the residual AGS-packing material were localized to the wounding site where the original packing was placed. Statistical analysis showed a significant difference (all values P<.01) in the amount of packing material remaining in the ME at postoperative week 6 between the group 1 (MeroGel)- and group 2 (AGS-)treated animals (Table 1).
No abnormalities were found in the MEs of any of the untreated animals with lesions from groups 1 and 2 or in the control animals (group 3). The histologic appearance of MeroGel-treated (group 1) MEs was characterized by a membranelike sheet infiltrated with inflammatory cells that covered the center of the area with the lesion(Figure 2). There was a slight increase in the thickness of the mucous membrane at the wound site but there was no new bone formation in these ME cavities (Figure 2A).
The AGS-treated (group 2) MEs retained a mass of AGS matrix in the area of the lesion that surrounded the round window membrane niche (Figure 3A). These MEs all had inflammatory cells, marked new bone formation (ie, either in the AGS matrix or attached to the wall of ME), new vessel formation, and fibroblasts within the AGS matrix (Figure 3B). Normal healing of the lesion site was present in the untreated ears with lesions (groups 1 through 3). Neither adhesions nor thickening of the healing mucosa was observed in any of these control ears (Figure 4).
All animals (groups 1 through 3) survived the duration of the experiment. None of these animals exhibited ME or inner ear infections or balance disorders.
Supportive materials are commonly used in the ME as packing to enhance healing of tympanic membrane repairs and to maintain the integrity of the ME space during postsurgical wound healing. Hyaluronic acid is a naturally occurring extracellular polysaccharide with a molecular weight in the range of 2 to 4 million daltons.1 Although purified HA has been shown to have a high level of tissue biocompatibility and to be nontoxic to the inner ear,6- 8 its viscous physical properties1 has limited its use as a surgical packing material for the ME cavity. Because MeroGel is an esterified form of HA, its physical properties are more like that of a sponge, which make it an ideal packing material for ME application. Like AGS, compressed MeroGel can be easily trimmed to allow for the insertion into a surgical site within the ME cavity. When moistened with an isotonic sodium chloride solution, MeroGel showed a high level of expandability that is a necessary property for any supportive packing material to be used during ME surgery. In this study, minimal residual MeroGel was encountered in the ME cavity at postoperative week 6 in 8 of the 10 ME cavities examined. This resorption period provides adequate time for support to aid in healing of ME tissues, while limiting the amount of time the ME has to produce an inflammatory response.
Changes in auditory thresholds for pure-tone stimuli were evaluated by ABR recordings which is a reliable test and allows for the evaluation of the integrity of both the inner ear and MEs. Increases in auditory thresholds for only the low frequencies in both the MeroGel- and AGS-treated ears at postoperative day 5 indicate that both of these packing materials effect the mobility of the ossicular chain and/or the round window and oval window membranes. Recovery of hearing, therefore, most likely implies a recovery of the mobility of the ossicular chain and/or of the round and oval window membranes. The observed hearing recovery correlated with the disappearance of most of the original volume of MeroGel packing and a lack of major tissue adhesions that could compromise ossicular chain function. The histologic evaluation showed scant adhesions in 20% of the animals and in 80% of the treated ears (group 1) that only a very small amount of the original MeroGel material (ie, membranelike material) had remained in the ME cavity. In the AGS-packed ears, at postoperative week 6, there was a persistence of a mass of AGS with adhesions in the ME cavity and new bone deposition that resulted in a hearing loss that was most probably of a conductive nature.
The ABR recordings also showed that MeroGel had no toxic effect on cochlear hair cells based on the observation that there were no high-frequencies hearing losses (ie, 8000-16 000 Hz) and that there was a total recovery of hearing thresholds in 80% of the MeroGel-treated ears at postoperative week 6. Other studies6,8 also had a similar result from the application of a viscous form of HA to the ME cavity of guinea pigs and rats. Hearing thresholds of the HA-packed MEs were recovered to normal levels in either 4 weeks or 3 months after HA application.
The comparison of MeroGel and AGS as packing materials for the ME is limited by the fact that the experiment was terminated and analyzed at 6 weeks after the placement of both materials in the ME cavity. A comparison of hearing results between the MeroGel and AGS groups of animals at postoperative day 5 showed that the results at this stage were equivalent in groups 1 and 2, respectively. Presumably, this hearing loss was of a conductive origin because in both groups, the ME cavities were filled with packing material. Difference in hearing results (ie, ABR thresholds) between groups 1 and 2 became significant at postoperative week 6 (Figure 1C). At this time, most of the animals that had their ears packed with MeroGel had regained their baseline hearing sensitivity, while the AGS-treated animals continued to exhibit a residual hearing impairment at low frequencies. This hearing loss is significantly less severe than the hearing loss that the AGS-treated animals had at postoperative day 5 (compare Figure 1A with 1B) demonstrating a partial recovery of hearing acuity in the AGS-treated animals. There was no observable high-frequency hearing loss in any of the animals and ultrahigh-frequency hearing, ie, above 16 000 Hz, was not measured in this study. Therefore, our conclusions about ototoxic reactions are limited only to the tested frequencies, ie, 500 to 16 000 Hz.
When the MeroGel- and AGS-treated MEs were compared with surgical microscopic and histologic evaluations, the following findings were noted: in the MeroGel-treated ears, a thin transparent membranelike material was still present in the ME at postoperative week 6 (Figure 2A). This membranelike material showed some inflammatory cell infiltration (Figure 2B). In the AGS-treated group, the ME still contained a mass of the material that covered the floor of the ME and round window membrane niche area (Figure 3A). The mass of retained AGS was surrounded by inflammatory changes and showed new bone formation (Figure 3B). Most of the histologic studies in the literature have shown that use of the viscous form of HA in the ME cavity has not caused either new bone or adhesions to form in the ME cavity.6,7 The study11 evaluating the "foam" esterified form of HA has reported new bone formation, mucosa thickness, and adhesions, although they were moderate when compared with the adverse effects caused by the use of AGS. The severity of pathologic change for both MeroGel and AGS might be dependent on the severity and extent of ME mucosa wounding as well as the position and amount of packing material applied within the ME cavity. Variations in extent of wounding of the ME mucosa and amount and placement of packing material may also have implications for the clinical application of the tested packing materials. Therefore, the finding of the present study and of a previous study11 both show a similar pattern in the ME cavities response to esterified HA as compared with AGS-packing material (Table 1 and Figure 1, Figure 2, and Figure 3).
New bone formation in the AGS-packed MEs may not necessarily correlate with a similar reaction in human ears. The favorable results encountered in the MeroGel-treated ears indicate that, in guinea pigs, MeroGel is a safe packing material for applications in the ME cavity. Clinical studies will be required to establish the clinical usefulness and safety of MeroGel as a surgical packing material in the MEs of patients as an adjunct to aid healing following ME surgery.
The results show that MeroGel is a nonototoxic substance that has a high level of tissue biocompatibility with ME mucosa. It is an effective support material for packing the ME during surgery and is easily expelled from the ME cavity causing only a mild inflammatory response.
Accepted for publication September 22, 2000.
This study was supported by a grant from the Carolina Ear Institute, Raleigh, NC (Dr Van De Water).
Presented as a poster (abstract 31) at the 23rd Association for Resarch in Otolaryngology Midwinter Research Meeting, St Petersburg Beach, Fla, February. 20-24, 2000.
Corresponding author and reprints: Thomas R. Van De Water, PhD, Department of Otolaryngology, Albert Einstein College of Medicine, 1410 Pelham Pkwy S, Kennedy Center, Room 302, Bronx, NY 10461 (e-mail: email@example.com).