Abstract
Purpose
To compare the performance of absorbable gelatin sponge (AGS) with polyurethane foam (PUF) as middle ear packing material after mucosal trauma.
Materials and methods
Using a randomized, controlled and blinded study design fifteen guinea pigs underwent middle ear surgery with mucosal trauma performed on both ears. One ear was packed with either PUF or AGS while the contralateral ear remained untreated and used as non-packed paired controls. Auditory brainstem response (ABR) thresholds were measured pre-operatively and repeated at 1, 2, and 6 weeks postoperatively. Histological analysis of middle ear mucosa was done in each group to evaluate the inflammatory reaction and wound healing. Another eighteen animals underwent middle ear wounding and packing in one ear while the contralateral ear was left undisturbed as control. Twelve guinea pigs were euthanized at 2 weeks postoperatively, and six were euthanized at 3 days post-operatively. Mucosal samples were collected for analysis of TGF-β1 levels by enzyme-linked immunosorbent assay.
Results
ABR recordings demonstrate that threshold level changes from baseline were minor in PUF packed and control ears. Threshold levels were higher in the AGS packed ears compared with both control and PUF packed ears for low frequency stimuli. Histological analysis showed persistence of packing material at 6 weeks postoperatively, inflammation, granulation tissue formation, foreign body reaction and neo-osteogenesis in both AGS and PUF groups. TGF-β1 protein levels did not differ between groups.
Conclusion
PUF and AGS packing cause inflammation and neo-osteogenesis in the middle ear following wounding of the mucosa and packing.
1
Introduction
Packing of the middle ear cavity during otologic surgery is primarily done to serve as a scaffold for the tympanic membrane and ossicular grafts, achieve hemostasis, aerate the middle ear cavity, and to enhance wound healing in cases of mucosal disease or removal. A variety of middle ear packing agents exist to meet these aims, including both non-absorbable and absorbable materials . Since introduced as a complement to otologic surgery in the 1950s, absorbable gelatin sponge (AGS) (Pharmacia & Upjohn, Kalamazoo, MI) has become the most commonly used middle ear packing agent in clinical practice . Nevertheless, it has been shown that AGS is prone to initiating an extensive inflammatory process leading to fibrosis, adhesions, and new bone formation within the middle ear . This has served as the impetus for studying alternative packing materials and particularly those exhibiting better wound healing properties such as hyaluronic acid (HA) .
Otopore ® , i.e. Co-Polyether-Ester Urethane Foam (PUF) (Striker, Kalamazoo, MI), is a synthetic biodegradable foam currently marketed as a middle ear packing agent. Unlike the degradation of AGS, which relies on phagocytosis by macrophages, PUFs degradation occurs mainly via drainage through the Eustachian tube. Furthermore, cytocompatability studies have demonstrated a lack of cytotoxicity from degradation products of this compound . A histology study by Dogru, et al. compared short-term and long-term appearances of middle ears packed with either AGS or PUF in an atraumatic model of rat middle ear packing . Here it was shown that PUF is prone to mild inflammation and fibrosis in the long-term in contrast to the severe inflammatory process and fibrosis associated with AGS .
The transforming growth factor (TGF)-β supergene family of low molecular weight cytokines is involved in the recruitment, activation, and proliferation of tissue and cells during inflammatory processes. TGF-β cytokines are known to play a prominent role in wound healing and are important contributors to hypertrophic scarring when present in excess . Among the various isoforms, TGF-β1 has been implicated in the pathogenesis of otitis media with effusion and cholesteatoma formation, two diseases of the middle ear mucosa .
We hypothesize that PUF can serve as a better middle ear packing agent than AGS in the healing of the middle ear mucosa. The primary objective of this study is to investigate the characteristics of PUF packing material as an alternative to AGS in a guinea pig model of middle ear trauma-wound healing. Furthermore, we investigate the molecular environment of middle ear mucosal-wound healing via analysis of TGF-β1 signaling in an attempt to understand how this environment can be impacted by the type of middle ear packing agent used.
2
Materials and methods
2.1
Animals
Fifteen 250–350 g pigmented guinea pigs were obtained from a commercial supplier (Cady Ridge Farm, Natick, MA). All animals were treated according to the experimental protocol approved by the University of Miami Animal Care and Use Committee, and in full compliance with the Public Health Service Policy on Humane Care and Use of Laboratory Animals and the Animal Welfare Act (7 U.S.C. et seq.). The experimental ear of all animals received an intentional, defined pattern of wounding of their middle ear mucosa and a defined, reproducible level of packing material of their middle ear cavity, while the contralateral ear served as a nontreated surgical wound-healing control. Control ears underwent the same procedure as experimental ears but were left unpacked. Animals were euthanized at 6 weeks post-operatively and their temporal bones harvested and processed for histology.
2.2
Surgery
Guinea pigs were anesthetized with a combined mixture of xylazine hydrochloride (15 mg/kg) and ketamine hydrochloride (35 mg/kg). The corneas of anesthetized animals were protected using an ophthalmic ointment applied at the beginning of each case. The surgical site was shaved and cleansed using 30% Betadine in 70% ethanol. Local anesthesia was then provided using 1% lidocaine subcutaneously. A mid-sagital scalp incision was made and a dura touching stainless steel screw electrode fixed into place just anterior to lambda location in the calvarium. Baseline auditory brainstem response (ABR) recordings were then taken using this permanent electrode along with subcutaneous needle electrodes. Following this step, a retroauricular incision was made to reach the bulla, which was opened using a diamond burr. After visualizing the cochlea, the mucosa was scraped at the junction of basal turn of the cochlea and adjacent bulla wall using a periosteal elevator. The middle ear mucosal lesion was standardized and kept by the surgeon (SA) similar for all lesions. The middle ear cavity was then packed gently using 1-mm 2 pieces of either PUF or AGS being careful to preserve the integrity of the ossicular chain and tympanic membrane. Incisions were sutured closed and antibiotic ointment was applied to the wound site ( Fig. 1 ).
2.3
Electrophysiological testing
ABR testing was done using the Intelligent Hearing Systems Hardware and Software (IHS, Miami, FL) to measure hearing thresholds to pure tone stimuli at baseline (pre-operatively), 1 week, 2 weeks, and 6 weeks post-operatively. Testing was conducted bilaterally to record responses at 0.5, 1, 4, and 16 kHz stimuli. At the time of hearing testing, animals were anesthetized with a combined mixture of xylazine hydrochloride (15 mg/kg) and ketamine hydrochloride (35 mg/kg). Electrical activity was recorded using the chronic screw electrode in the vertex of the skull, referenced to a needle electrode inserted into the neck deep musculature, and grounded to a second needle electrode inserted subcutaneously near the tympanic bulla (Grass Instrument Division, Astro-Medical Inc., West Warick, RI). Stimuli were introduced into each ear via an external ear canal microphone beginning with an intensity of 90 dB SPL and decreasing in increments of 10-dB until threshold responses were identified. Threshold was defined as the lowest intensity stimulus leading to an investigator identified reproducible ABR waveform. Those responses near or at threshold were confirmed using increments of 5 dB SPL.
2.4
Microscopic analysis
The fifteen audiologically tested animals were euthanized at 6 weeks post-operatively using an intraperitoneal injection of 1.5 ml Euthasol. Their temporal bones were rapidly dissected and fixed in 4% paraformaldehyde at 4 °C. The bones were subsequently transferred into formic acid and decalcified at room temperature for 4 days. Each bone was then dehydrated in a graded series of ethanol and embedded in paraffin for serial sectioning. Four μm coronal sections were obtained at a minimum of 2 levels from the site of interest and stained with hematoxylin and eosin. A pathologist (SV) blinded to the type of packing material used analyzed the histology specimens and assessed the degree of inflammation of the middle ear mucosa and the persistence of microscopic fragments of packing material.
2.5
Molecular analysis
A separate set of eighteen guinea pigs underwent middle ear wounding plus packing with the same technique as described earlier, with their contralateral ear undergoing no intervention in order to establish a group of naïve ears. Six animals were euthanized at 3 days post-operatively and twelve animals were euthanized at 2 weeks post-operative. All temporal bones were harvested. Using a periosteal elevator and a round knife, a technique was developed to extract uniform size middle ear mucosal samples from the site of wounding. Briefly, the bony shell of the bulla was fractured and peeled off the intact mucosa. The round knife was then used to collect the mucosal sample from the site of injury. These samples were placed in a sterile solution with proteinase and phosphotase inhibitors. The samples were then sonicated and centrifuged to yield supernatant for ELISA testing. Human TGF-β1 ELISA kits were used (RayBiotech, Norcross, GA) and a Bradford was performed prior to each experiment to control for amount of protein loaded into ELISA wells.
2.6
Statistical analysis
Post-operative responses recorded by ABR were compared to the pre-operative responses using a paired Student’s t -test within each group. A two-way ANOVA test was used to compare ABR recordings and TGF-β1 (adjusted with Bonferroni method) results between groups at each time interval and to compare threshold shifts between groups. A Fisher’s Exact Test adjusted with Bonferroni Method was used to compare histological findings between groups. Statistical significance was set at p < 0.05.
2
Materials and methods
2.1
Animals
Fifteen 250–350 g pigmented guinea pigs were obtained from a commercial supplier (Cady Ridge Farm, Natick, MA). All animals were treated according to the experimental protocol approved by the University of Miami Animal Care and Use Committee, and in full compliance with the Public Health Service Policy on Humane Care and Use of Laboratory Animals and the Animal Welfare Act (7 U.S.C. et seq.). The experimental ear of all animals received an intentional, defined pattern of wounding of their middle ear mucosa and a defined, reproducible level of packing material of their middle ear cavity, while the contralateral ear served as a nontreated surgical wound-healing control. Control ears underwent the same procedure as experimental ears but were left unpacked. Animals were euthanized at 6 weeks post-operatively and their temporal bones harvested and processed for histology.
2.2
Surgery
Guinea pigs were anesthetized with a combined mixture of xylazine hydrochloride (15 mg/kg) and ketamine hydrochloride (35 mg/kg). The corneas of anesthetized animals were protected using an ophthalmic ointment applied at the beginning of each case. The surgical site was shaved and cleansed using 30% Betadine in 70% ethanol. Local anesthesia was then provided using 1% lidocaine subcutaneously. A mid-sagital scalp incision was made and a dura touching stainless steel screw electrode fixed into place just anterior to lambda location in the calvarium. Baseline auditory brainstem response (ABR) recordings were then taken using this permanent electrode along with subcutaneous needle electrodes. Following this step, a retroauricular incision was made to reach the bulla, which was opened using a diamond burr. After visualizing the cochlea, the mucosa was scraped at the junction of basal turn of the cochlea and adjacent bulla wall using a periosteal elevator. The middle ear mucosal lesion was standardized and kept by the surgeon (SA) similar for all lesions. The middle ear cavity was then packed gently using 1-mm 2 pieces of either PUF or AGS being careful to preserve the integrity of the ossicular chain and tympanic membrane. Incisions were sutured closed and antibiotic ointment was applied to the wound site ( Fig. 1 ).
