Ventral bulla osteotomy (VBO) is a commonly performed procedure for the treatment of middle ear disease in feline patients. This procedure is most commonly performed for the treatment of nasopharyngeal polyps, either as a primary intervention or following failure of conservative management.1,2 Ventral bulla osteotomy may also be performed in the treatment of chronic otitis media without the presence of a nasopharyngeal polyp.3,4 Nasopharyngeal polyps are benign masses that arise from the epithelial lining of the tympanic bulla or auditory tube and may extend into the external acoustic meatus through the tympanic membrane or into the nasopharynx via the Eustachian tube.5–7 Depending on the degree and direction of extension, patients may present with clinical signs, such as sneezing, nasal discharge, stertor, vestibular signs, or chronic otorrhea.6–8 Several treatment options, including surgery, have been proposed for the management of nasopharyngeal polyps following diagnosis, with varying levels of success. Traction-avulsion is a commonly reported technique when conservative management is elected; however, blind traction-avulsion is limited to treating the portion of the polyp located in the nasopharynx or external ear canal. One report9 investigating blind traction-avulsion resulted in a 41% recurrence rate in patients treated with traction alone. The same study investigated the addition of prednisolone and found that recurrence was significantly reduced compared to traction alone; however, the majority of these cases also underwent surgical intervention. Other studies10,11 have reported recurrence as high as 50% to 71% with blind traction-avulsion and no significant difference with concurrent administration of glucocorticoids.10 As a result of the high recurrence rate, other methods of removal have been investigated, such as traction-avulsion following a lateral approach to the ear canal. This method resulted in a 20% recurrence rate, although recurrence rates were reported as low as 14.3% in the hands of an experienced surgeon.12 Minimally invasive techniques have also been proposed in order to reduce complications related to surgical approach, such as Horner syndrome, facial nerve paralysis, and vestibular signs. One study13 evaluated the use of a per-endoscopic transtympanic technique, which resulted in a recurrence rate of 13.5%. Given that these polyps arise from epithelium within the tympanic bulla, the major limitation with all of these techniques is that thorough curettage of the bulla is not possible, likely contributing to the moderate recurrence rate. In addition to this, the feline bulla possesses a bony septum, creating 2 separate compartments that must be investigated and debrided appropriately, which can only be accomplished via bulla osteotomy.
While VBO allows for the breakdown of the bony septum in feline patients and access to both the rostrolateral compartment and hypotympanum for curettage, recurrence rates following VBO for nasopharyngeal polyps has been reported to be up to 33%.4,9,11,14 Recurrence rates following VBO for otitis media in cats also range from 18% to 25%.3,4 Similar to concerns for blind and endoscopic-assisted traction-avulsion techniques, residual epithelium and polyploid tissue may contribute to recurrence, particularly given the knowledge that nasopharyngeal polyps generally arise from epithelium specifically within the rostrolateral compartment, where visualization is more limited compared to the spacious hypotympanum.2
To minimize the chances of recurrence and ensure thorough removal of epithelium within the tympanic bulla, endoscopic assistance via VBO may be considered. The use of endoscopy for visualization of the canine tympanic bulla following total ear canal ablation–lateral bulla osteotomy (TECA-LBO) has been previously reported with a high incidence of epithelial remnants.15 However, endoscopic assistance has not yet been reported to aid in thorough removal of all epithelium in a feline tympanic bulla following VBO. The objective of this study was to demonstrate the feasibility of a 1.9-mm, 30° rigid endoscope for use in evaluating the presence of epithelial remnants following VBO in a feline cadaveric model and to identify the section with the highest proportion of remnants. We hypothesized that the use of the 1.9-mm, 30° rigid endoscope would be effective in the detection of epithelial remnants. We also hypothesized that the rostral and lateral sections would have the highest prevalence of epithelial remnants.
Methods
Study population
Five frozen feline cadavers were used. The cadavers were donated from a local shelter and had been humanely euthanized prior to donation for reasons unrelated to the study. No gross abnormalities were detected prior to use. Each feline cadaver underwent bilateral VBO for a total of 10 procedures.
Surgical technique
Five board-certified surgeons each performed 2 VBOs (left and right performed on the same cat). The VBO was performed via a standard ventral approach.2 A 3-to-5-cm incision was made in the ventral cervical region, over the location of the tympanic bulla. Dissection was performed through the platysma and sphincter colli and between the digastricus and mylohyoid muscles until the tympanic bulla was exposed. A bulla osteotomy was performed using a Steinmann pin of variable diameter and was subsequently expanded with Lempert Rongeurs. The bony septum was penetrated using a Steinmann pin (IMEX), and this osteotomy was also enlarged using Lempert Rongeurs (Sontec). Curettage was completed until the performing surgeon was satisfied. Following penetration of the bony septum and completion of curettage, a 1.9-mm, 30° rigid endoscope was introduced into the bulla, and images were obtained of the rostral, caudal, dorsal, lateral, and medial sections, utilizing rotation of the light post to facilitate observation of each section (Figure 1). The rostrolateral compartment was included in both the rostral and lateral section, whereas the caudal, dorsal, and medial sections were limited to the hypotympanum. An observer (FPS), blinded to the surgeon performing the procedure and section being evaluated, measured the proportion of each section occupied by epithelial remnants using ImageJ (version 1.53; NIH). This was performed by measuring the number of pixels comprising the area of each epithelial remnant and then compared to the total surface area of each section measured in pixels, reported as a percentage.
Otoscopic images taken with a 1.9-mm rigid 30° endoscope during a ventral bulla osteotomy. Arrows indicate residual epithelium following curettage. A—Rostral. B—Caudal (Ca). C—Dorsal (D). D—Lateral (L). E—Medial. Cr = Cranial. P = Promontory. RW = Round window. S = Septum.
Citation: American Journal of Veterinary Research 86, 3; 10.2460/ajvr.24.11.0331
Statistical analysis
Descriptive statistics were used to report the distribution of epithelium within each section (mean, SD, minimum, and maximum) and to compare the amounts of epithelium remaining in each section. The resulting percentage of epithelium was then square root transformed prior to further analysis. The transformed data was evaluated using a Tukey multiple comparisons procedure to determine pairwise comparisons. A P value less than .05 was considered statistically significant.
Results
Remnant epithelium was noted in at least 1 section in each cadaver following VBO. The distribution of epithelial remnants is reported in Table 1. Epithelial remnants were noted most frequently associated within the dorsal section (10 of 10 [100%]) and were less likely to be noted in the lateral and medial section (5 of 10 [50%]; 5 of 10 [50%]; Table 2). However, when epithelial remnants were present, the lateral section was found to have the highest percentage of surface area of bulla occupied (6.9 ± 11.42%), whereas the lowest was found in the caudal section (1.2 ± 1.42%). The only significant difference with regard to epithelial remnant distribution when comparing between sections was noted between caudal and dorsal sections (Table 3).
Percentage of tympanic bulla occupied by epithelial remnants in respective sections following ventral bulla osteotomy.
| Location | Mean (± SD) |
|---|---|
| Caudal | 1.23 (1.42) |
| Dorsal | 5.01 (2.04) |
| Lateral | 6.92 (11.42) |
| Medial | 3.77 (1.80) |
| Rostral | 3.47 (0.76) |
Number of times epithelial remnants were identified in each section of the tympanic bulla following ventral bulla osteotomy.
| Location | No. of times epithelial remnants noted (n = 10) |
|---|---|
| Caudal | 7 |
| Dorsal | 10 |
| Lateral | 5 |
| Medial | 5 |
| Rostral | 9 |
Pairwise comparisons of the percentage of epithelial remnants within each section following ventral bulla osteotomy in 10 cadavers.
| Location | Difference between means (%) | P value |
|---|---|---|
| Caudal vs dorsal | −13.62 | .0034 |
| Caudal vs lateral | −8.53 | .7654 |
| Caudal vs medial | −4.56 | .9302 |
| Caudal vs rostral | −8.60 | .2251 |
| Dorsal vs lateral | 5.09 | .9444 |
| Dorsal vs medial | 9.05 | .4547 |
| Dorsal vs rostral | 5.01 | .5300 |
| Lateral vs medial | 3.96 | .9891 |
| Lateral vs rostral | −0.07 | 1.0000 |
| Medial vs rostral | −4.04 | .9543 |
Discussion
The study confirms the feasibility of the use of a 1.9-mm, 30° endoscope to evaluate the feline tympanic bulla for epithelial remnants following VBO in a cadaveric model. All surfaces were successfully imaged in all bullae. The main hypothesis of epithelial remnant detection using the endoscope was accepted. Epithelial remnants were able to be visualized in every cadaver in at least 1 section with the rigid endoscope despite surgeon satisfaction with the degree of curettage. Additionally, the lateral section was identified as possessing the highest proportion of epithelial remnants, whereas the caudal possessed the least. As a result, we reject our hypothesis that the rostral and lateral sections of the tympanic bulla will have the highest proportion of epithelial remnants.
The use of endoscopy is known for enhanced visualization due to magnification and illumination, resulting in widespread use in veterinary surgery. A previous study16 investigated the use of a 2.7-mm, 30° rigid endoscope and optimal endoscope and light post positioning for identification of structures within the canine middle ear following both a ventral and a lateral approach. The authors of this study determined that this was a straightforward technique that allowed for identification of structures with the middle ear that are often not visible during an open approach. Similarly in our study, it is likely that the use of endoscopy facilitated visualization of sections of the tympanic bulla that are challenging to view, enhancing identification of epithelial remnants that were previously unable to be seen by the operating surgeon. At least a portion of this enhanced visualization is likely due to the use of a 30° endoscope, allowing the operator to rotate the light post to look around corners, which is particularly useful in a constrained space, such as a feline bulla.
Given the advantages conferred by the use of endoscopy, more targeted debridement may be possible in clinical cases, which may result in a lower rate of iatrogenic complications, such as Horner syndrome and vestibular signs. Horner syndrome is observed in 57% to 95% of feline patients following VBO and is suspected to be secondary to curettage over the promontory, injuring the tympanic plexus, or creating an osteotomy into the rostrolateral compartment that is too far medial.4,11,14,17 Vestibular signs are observed in 4% to 42% postoperatively.11,14 This is suspected to be due to overly aggressive curettage in the location of the round or oval windows, resulting in damage to the vestibulocochlear apparatus. By using the advantages gained with endoscopic assistance (magnification, illumination, and the ability to see around corners in a confined space), it is theoretically possible to identify a more appropriate location to enter the rostrolateral compartment in feline patients and also more effectively ensure the removal of diseased or polyploid tissue while limiting iatrogenic injury to the tympanic plexus and vestibulocochlear apparatus. Previous studies16,18 have demonstrated improved identification of structures within the canine and feline bulla, such as the round and oval windows, the chorda tympani, and the ossicles. However, the tympanic plexus was unable to be identified in feline cadavers and is only consistently found within canine cadavers. This is possibly due to the multiple freeze-thaw cycles that the feline cadavers underwent, leading to destruction and increased fragility of delicate soft tissue structures, such as the tympanic plexus.18 Further studies are needed in fresh cadavers or clinical cases to determine the frequency with which more delicate structures, such as the tympanic plexus, can be identified with endoscopy and the impact on postoperative complications.
While the lateral section did possess the highest proportion of epithelial remnants (6.9%), the rostral was not among the highest. Additionally, the high degree of variability in the percentage of bulla possessing retained epithelium limits the ability for the authors to state that any 1 section is much more likely to contain epithelial remnants than any other. In pair-wise comparisons, the only difference found to be statistically significant was when comparing the caudal and dorsal locations. This information suggests that surgeons must perform a deliberate and thorough evaluation of the entire feline tympanic bulla in order to minimize the likelihood that residual diseased tissue is residing in the bulla postoperatively and not rely on 1 section of the bulla being more likely to possess epithelial remnants or debris. The information presented here also suggests that the use of endoscopy may be beneficial to minimize the risk of disease recurrence in clinical cases.
In the study by Watt et al15 evaluating epithelial remnants following TECA-LBO, the highest percentage of epithelial remnants identified were found in the rostral section, whereas the least were identified in the medial section. This contrasts with the results found in this study, in which the lateral section possessed the highest proportion of epithelial remnants and the caudal the least. Differences for this distribution may include variation in the surgical procedure performed as well as species differences. A TECA-LBO is performed from a lateral approach, and so it is logical that the minority of epithelial remnants would be identified in the medial section as that would be the compartment with maximal visualization following a lateral approach. The feline bulla is separated into a hypotympanum and smaller rostrolateral compartment by a bony septum.2 This separation seen in cats requires the surgeon to perform an additional osteotomy to access the rostrolateral compartment. As stated previously, the rostrolateral compartment is markedly smaller than the hypotympanum, contributing to decreased visualization and reduced access with both a lateral and a ventral approach to the bulla. It is this species difference that may explain the high percentage of remnant epithelium noted in the lateral section in this study.15 Additionally, in some canine patients, the mandible may overlie the rostral aspect of the bulla, presenting a challenge for bulla osteotomy as well as subsequent curettage.19,20 While any breed-specific anatomy or brachycephaly is not reported in the aforementioned study, it is possible that this may have contributed to an increased distribution of epithelial remnants in the rostral section following TECA-LBO. It is also important to note that the extent and depth of soft tissue dissection required for bulla osteotomy following TECA in either species is much more extensive compared to a VBO in a feline patient. In a feline patient, the bulla can often be palpated prior to surgical approach and the entire ventral surface of the bulla exposed before osteotomy, which may allow for a larger osteotomy and improved visualization of structures within the bulla. This aspect may also explain why the distribution of epithelial remnants in the previous study tended to be higher in all sections than in this study. Further research is needed to draw a direct comparison between both VBOs and TECA-LBOs performed in canine and feline patients with respect to residual epithelium following curettage as well as the impact on video otoscopy for assistance in removal. Additional avenues for investigation may also include the impact of osteotomy size on residual epithelium when comparing between these 2 procedures in both dogs and cats.
There are several limitations to this study. Most notable is the use of a feline cadaveric model. These cadavers were noted to be grossly free of external and middle ear disease, which is not the likely population to undergo VBO in a clinical setting. Feline patients likely to undergo VBO in a clinical setting most often have a space-occupying lesion within the bulla, such as a polyp, which may affect the overall surgical approach as well as the size of the osteotomy. In addition to this, the size, shape, and thickness of the bulla in clinical cases may be altered by the underlying pathology, and this was unable to be replicated in this study. Additionally, total coverage of the tympanic bulla with epithelium was not confirmed with the endoscope prior to curettage. Surgeons were encouraged to perform the procedure in accordance with their own preference without a regimented stepwise procedure or guidelines for the size of the osteotomy. This may have affected the overall surface area available for epithelial remnants as well as the ability to visualize these remnants with and without endoscopic assistance. Additionally, surgeons performing the procedure were not blinded to the nature of the study, and this may have resulted in more aggressive curettage as a result, thus falsely lowering the proportion of epithelial remnants across all sections. Surgeons performing VBO in this study also did not utilize any additional magnification, such as surgical loupes, which may alter the ability to visualize remnant epithelium. Further prospective studies are needed to determine the likelihood of epithelial remnants in clinical cases as well as the impact on postoperative outcomes following endoscopic assistance for retrieval.
In conclusion, epithelial remnants are commonly located within all sections of the feline tympanic bulla following VBO. The use of a 1.9-mm, 30° endoscope may facilitate the visualization and retrieval of these epithelial remnants as well as provide the ability to reduce iatrogenic injury throughout dissection and curettage in order to improve postoperative outcomes and decrease recurrence rates.
Acknowledgments
The authors thank James Colee, PhD, for substantial contribution to data analysis and interpretation.
Disclosures
The authors have nothing to disclose. No AI-assisted technologies were used in the generation of this manuscript.
Funding
The authors have nothing to disclose.
ORCID
F. P. Solari https://orcid.org/0000-0002-4202-8701
References
- 1.↑
Fan T, de Lorimier LP. Inflammatory polyps and aural neoplasia. Vet Clin North Am Small Anim Pract. 2004;34(2):489–509. doi:10.1016/j.cvsm.2003.10.008
- 2.↑
White R. Middle and inner ear. In: Spencer J, Tobias K, eds. Veterinary Surgery: Small Animal Expert Consult. 2nd eds. Saunders; 2017:2328–2340.
- 3.↑
Swales N, Foster A, Barnard N. Retrospective study of the presentation, diagnosis and management of 16 cats with otitis media not due to nasopharyngeal polyp. J Feline Med Surg. 2018;20(12):1082–1086. doi:10.1177/1098612X17746282
- 4.↑
Trevor P, Martin R. Tympanic bulla osteotomy for treatment of middle-ear disease in cats: 19 cases (1984-1991). J Am Vet Med Assoc. 1993;202(1):123–128. doi:10.2460/javma.1993.202.01.123
- 5.↑
Lane J, Orr C, Lucke VM, Gruffydd-Jones T. Nasopharyngeal polyps arising in the middle ear of the cat. J Small Anim Pract. 1981;22(8):511–522. doi:10.1111/j.1748-5827.1981.tb00638.x
- 6.↑
Reed N, Gunn-Moore D. Nasopharyngeal disease in cats: 2. Specific conditions and their management. J Feline Med Surg. 2012;14(5):317–326. doi:10.1177/1098612X12444998
- 7.↑
Greci V, Mortellaro C. Management of otic and nasopharyngeal, and nasal polyps in dogs. Vet Clin North Am Small Anim Pract. 2016;46(4):643–661. doi:10.1016/j.cvsm.2016.01.004
- 8.↑
Allen H, Broussard J, Noone K. Nasopharyngeal diseases in cats: a retrospective study of 53 cases (1991-1998). J Am Anim Hosp Assoc. 1999;35(6):457–461.
- 9.↑
Anderson D, Robinson R, White R. Management of inflammatory polyps in 37 cats. Vet Rec. 2000;147(24):684–687.
- 10.↑
Hoppers S, May E, Frank L. Feline bilateral inflammatory aural polyps: a descriptive retrospective study. Vet Dermatol. 2020;31(5):385–e102. doi:10.1111/vde.12877
- 11.↑
Kapatkin A, Matthiesen D, Noone K, Church E, Scavelli T, Patnaik A. Results of surgery and long-term follow-up in 31 cats with nasopharyngeal polyps. J Am Anim Hosp Assoc. 1990;26:387–392.
- 12.↑
Janssens S, Haagsman A, Ter HG. Middle ear polyps: results of traction avulsion after a lateral approach to the ear canal in 62 cats (2004-2014). J Feline Med Surg. 2016;19(8):803–808. doi:10.1177/1098612X16660356
- 13.↑
Greci V, Vernia E, Mortellaro CM. Per-endoscopic trans-tympanic traction for the management of feline aural inflammatory polyps: a case review of 37 cats. J Feline Med Surg. 2014;16(8):645–650.
- 14.↑
Faulkner J, Budsberg S. Results of ventral bulla osteotomy for treatment of middle ear polyps in cats. J Am Anim Hosp Assoc. 1990;26(5):496–499.
- 15.↑
Watt MM, Regier PJ, Ferrigno CRA, Mcconkey MJ, Fox-Alvarez WA, Bertran J. Otoscopic evaluation of epithelial remnants in the tympanic cavity after total ear canal ablation and lateral bulla osteotomy. Vet Surg. 2020;49(7):1406–1411. doi:10.1111/vsu.13492
- 16.↑
Viani EC, Hudson CC, Coleman KA, Allen PI. Anatomic structures of the canine middle ear visible during endoscopic examination through a ventral or lateral approach. J Am Vet Med Assoc. 2022;260(13):1663–1667. doi:10.2460/javma.22.05.0192
- 17.↑
Jacobson LSJK, Kennedy SK, Lockwood GA, Mackenzie SD, Porter CD, Ringwood PB. A Pandora’s box in feline medicine: presenting signs and surgical outcomes in 58 previously hoarded cats with chronic otitis media-interna. J Feline Med Surg. 2023;25(9). doi:10.1177/1098612X231197089
- 18.↑
Enright D, Cole G, Hatfield J. Evaluation of endoscopic-assisted feline lateral bulla osteotomy: a cadaveric study. J Feline Med Surg. 2023;25(10). doi:10.1177/1098612X231200375
- 19.↑
Banks C, Beever L, Kaye B, Foo M, Ter Haar G, Rutherford L. Influence of extreme brachycephalic conformation on perioperative complications associated with total ear canal ablation and lateral bulla osteotomy in 242 dogs (2010–2020). Vet Surg. 2023;52(5):661–673.
- 20.↑
Mielke B, Lam R, Ter Haar G. Computed tomographic morphometry of tympanic bulla shape and position in brachycephalic and mesaticephalic dog breeds. Vet Radiol Ultrasound. 2017;58(5):552–558. doi:10.1111/vru.12529
