Introduction

Imaging of the middle ear is often performed for diagnosis of inflammatory and noninflammatory middle ear diseases such as otitis media-interna, neoplasia, and polyps.^1,2 Diagnostic imaging may provide general information on the condition of the middle ear cavity, such as the presence of abnormal fluid or soft tissue in the bulla, and may allow evaluation of boney structures surrounding the bulla. Abnormalities such as bone lysis or bone proliferation or sclerosis are commonly identified in association with middle ear disease. Soft tissue changes in the tissues external to the bulla may also be identified with diagnostic imaging. Still, currently available diagnostic imaging modalities have a low specificity for identifying changes to specific soft tissue structures in the middle ear.³

Otitis media is the most common disease of the middle ear in dogs.¹ It is usually the result of progressive chronic otitis externa causing attenuation of the vertical and horizontal ear canals and associated chronic inflammation and accumulation of bacteria in the external acoustic meatus.^1,2,4,5 Surgical intervention is indicated in dogs with septic otitis media when medical management has failed or is unlikely to succeed. Surgery is also indicated in patients with end-stage changes leading to neurologic signs. Surgical management eliminates diseased tissue and decompresses the tympanum. Total ear canal ablation with lateral bulla osteotomy (TECA-LBO) and ventral bulla osteotomy (VBO) are surgeries commonly performed to treat middle ear disease. Both TECA-LBO and VBO are routinely performed through an open approach, which limits visualization and identification of structures inside the middle ear cavity.⁶

When performing either TECA-LBO or VBO in dogs, one of the main goals is to remove all epithelial tissue from within the tympanic bulla. The removal of all epithelium and debris within the osseous tympanic cavity is thought to reduce the risk of persistent deep infection, which may result in fistulous tracts or peri-bulla abscess formation months to years after surgery.⁷ Traditionally, the epithelial layer is removed by blindly scraping the inner portion of the bulla with a variety of curettes until the soft tissues lining the bulla have been deemed to be fully removed; however, this blind debridement may inadvertently lead to postoperative morbidity.^6,8

The middle ear is a complex anatomic region containing many irregularly shaped structures. Owing to the intricate anatomy of the middle ear, a thorough understanding of the orientation and location of each structure relative to all the surrounding structures may be difficult to obtain from an anatomy text. The major components of the middle ear include the tympanic cavity and its membranes, auditory ossicles (malleus, incus, and stapes), cochlear and auditory windows, and important muscles, nerves, and vessels within the bulla.⁸ During middle ear surgery, normal structures should be preserved to the greatest extent possible; however, iatrogenic damage during surgical debridement remains a substantial risk and may result in hemorrhage or neurologic deficits.⁶

Endoscopy provides a minimally invasive option for improving visualization of middle ear structures, detecting normal versus abnormal anatomy, and assessing the extent of disease. Enhanced ability to directly visualize structures in the middle ear during surgery would be expected to decrease the frequency of iatrogenic middle ear damage. The use of endoscopy to guide debridement of the middle ear epithelium during TECA-LBO has recently been described in a cadaver model.⁵ To our knowledge, the normal anatomy of the canine middle ear as it appears when viewed with a telescope has not been described. The objective of the study reported here was to describe the anatomic structures of the canine middle ear visible during endoscopic examination through a lateral (as when performing TECA-LBO) or ventral (as when performing VBO) approach as well as determine the optimal telescope and light post positions with which to view the major anatomic structures in the middle ear.

Materials and Methods

Five canine cadaver heads (10 bullae) without gross evidence of middle ear disease were used in the study. Specimens were obtained from a commercial supplier (Skulls Unlimited International Inc). Two were from pit bull–type dogs, and 1 each was from a Labrador Retriever, an American Bulldog, and a German Shepherd Dog mix. For each skull, a lateral approach was performed on one bulla and a ventral approach was performed on the contralateral bulla to obtain endoscopic images. The lateral approach was designed to simulate the surgical approach performed for a TECA-LBO, and the ventral approach was the described surgical approach used to perform a VBO.

The lateral and ventral surgical approaches were performed as previously described.⁹ For the lateral approach, the head was positioned in lateral recumbency, and a skin incision was made from dorsal to ventral over the vertical external ear canal and extended circumferentially through the auricular cartilage at the level of the conchal cavity (just proximal to the entrance into the external acoustic meatus). The subcutis and muscle were dissected from around the vertical and horizontal portions of the external ear canal, and annular cartilage was dissected to expose the osseous external auditory meatus. The annular cartilage and horizontal ear canal were amputated as close as possible to the osseous external auditory meatus. Rongeurs were used to widen the opening of the osseous external auditory meatus into the lateral aspect of the tympanic bulla.¹⁰ In 2 bullae, a variation of the previously described approach was initially used whereby the tympanic membrane was initially left intact and the tympanic bulla was entered ventral to the pars tensa, which is the main portion of the tympanic membrane, so that images of the medial aspect of the intact tympanum could be obtained from a lateral approach. Once images of the intact tympanum were obtained, the tympanum was removed, the bulla osteotomy was enlarged, and evaluation and image collection were performed similarly to that for the remaining specimens.

For the ventral approach, the head was positioned in dorsal recumbency, and a paramedian skin incision was made in a rostral-to-caudal direction centered over the tympanic bulla (midway between the angular process of the mandible and the wing of the atlas). Blunt dissection was used to separate the digastricus muscle from the hyoglossal and styloglossal muscles to expose the ventral aspect of the tympanic bulla. The periosteum was stripped from the ventral aspect of the tympanic bulla with a periosteal elevator. A ventral bulla osteotomy was performed initially with a pneumatic burr (Hall Surgairtome Two Pneumatic System; Conmed Corp), followed by enlargement of the osteotomy with a rongeur.⁹

A 2.7-mm, 30° telescope attached to a 4K camera system (Synergy UHD4 Imaging Platform; Arthrex Inc) was used to evaluate and capture images of the interior of the tympanic bulla for analysis. The interior of the middle ear was evaluated, and images were captured while the telescope was inserted in the tympanic bulla with the light post in sequentially different positions based on a clock face orientation. For the lateral approach, the camera buttons (and the 12 o’clock light post position) were oriented dorsally toward the most proximal aspect of the skull, whereas for the ventral approach, the camera buttons (and the 12 o’clock light post position) were oriented medially toward the contralateral bulla. Fluid irrigation was not used during endoscopy and image acquisition, as we found that clearer images were obtained when irrigation was not used. The images obtained during endoscopy were organized by surgical approach, bulla evaluated, and light post orientation.

Images were reviewed by all authors, which included board-certified surgeons with extensive soft tissue experience, and structures in the middle ear visible on the images were identified on the basis of the described anatomy of the middle ear.^{2,7,8,10–13} The identity of each visible structure in the middle ear was confirmed and recorded, along with the relative position of each structure in relation to other structures in the middle ear. Telescope and light post positions at which each anatomic structure was most easily identified were recorded as well as whether the structure was identifiable from a lateral approach, ventral approach, or both approaches. Anatomy was identified at the time the images were obtained, during which all authors were present. Images were reviewed by authors individually later to determine optimal light post position. The highest-quality images were chosen, and optimal light post position was determined, with no disagreement between authors. All authors in the study reached a consensus on the identification of all anatomic structures as well as the optimal scope position and surgical approach for identifying all structures.

Results

Endoscopic evaluation of the middle ear allowed 20 anatomic structures to be viewed and identified (Figures 1 and 2). Most structures in the middle ear were identifiable from both the ventral and lateral approaches (Tables 1 and 2), but there were a few exceptions. The primary neural structures, including the tympanic plexus, which runs over the promontory, were visible with both approaches. Structures that were located more ventrally in the middle ear, such as the ventral bulla and septum bulla, were only visible through the ventral approach. The tensor veli palantini muscle was also only visible through the ventral approach because the muscle attaches at the rostral margin of the tympanic bulla and then passes ventromedially over the wall of the pharynx.⁸ The location of the auditory tube within the tympanic bulla was identified from the ventral approach but not from the lateral approach owing to the skeletal boundary of the bulla preventing the correct angle of the scope with the lateral approach. The branches of the rostral tympanic artery were only identified from the ventral approach. Variability in blood vessel anatomy inside the middle ear was not identified. The external acoustic meatus, pars flaccida, and pars tensa were the only structures that were visible from the lateral approach alone.

View Full Size

Images obtained during endoscopic evaluation of the middle ear through a lateral approach in a Labrador Retriever (left column; A, B, and C) and a pit bull–type dog (left column; D and E), and corresponding photographs of a skull showing the telescope and light post orientation for each image (right column). Images were obtained with the light post in the 6 o’clock (A), 11 o’clock (B), 7 o’clock (C), 5 o’clock (D), and 8 o’clock (E) positions. Ca = Caudal. CT = Chorda tympani. CW = Cochlear window. D = Dorsal. EAM = External acoustic meatus. I = Incus. MM = Manubrium of malleus. MP = Muscular process of malleus. P = Promontory. R = Rostral. S = Stapes. SP = Spicules. TM = Tympanic membrane. TP = Tympanic plexus. V = Ventral. VW = Vestibular window. *Branches of the rostral tympanic artery.

Citation: Journal of the American Veterinary Medical Association 260, 13; 10.2460/javma.22.05.0192

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Images obtained during endoscopic evaluation of the middle ear through a ventral approach in a pit bull–type dog (left column; A and B), a German Shepherd Dog mix (left column; C), and a Labrador Retriever (left column; D), and corresponding photographs of a skull showing the telescope and light post orientation for each image (right column). Images were obtained with the light post in the 11 o’clock (A), 12 o’clock (B), 5 o’clock (C), and 1 o’clock (D) positions. TT = Tensor tympanic muscle. TVP = Tensor veli palatini muscle. See Figure 1 for remainder of key.

Citation: Journal of the American Veterinary Medical Association 260, 13; 10.2460/javma.22.05.0192

• Download Figure
• Download figure as PowerPoint slide

Discussion

Results of the present study may serve as a reference for optimizing telescope and light post positions to view middle ear structures as well as help identify structures within the middle ear. Scope positioning information may be particularly beneficial if visualization of specific structures is desirable during middle ear evaluation or surgery. Recommendations on scope and light post positioning are commonly provided for minimally invasive surgery, such as arthroscopy or laparoscopy, and are thought to be beneficial for improving efficacy, accuracy, and efficiency in intraoperative imaging.^5,14

A recent study¹¹ that was focused on identifying branches of the external carotid and maxillary arteries that supply the canine middle ear identified several anatomic variations of the blood vessels located outside the bulla that supply the middle ear.¹⁵ In contrast, we did not identify any variability in blood vessel anatomy inside the middle ear.

Although there were structures that were only visible from the lateral approach in the present study, visualization of these structures when using a ventral approach would typically not be necessary, in that diseases of the middle ear that are treated with a VBO commonly do not involve the external acoustic meatus or tympanum.

The advantages of minimally invasive surgery have been well documented in both human and veterinary medicine. With the advantage of magnification, arthroscopy has improved the ability to provide early diagnosis of veterinary orthopedic disease, especially cranial cruciate ligament disease. Furthermore, the identification of structures within the canine stifle joint seen with various light post positions has aided in the diagnosis and treatment of cranial cruciate ligament and meniscal tears. With a 30° telescope, the viewing range is enlarged by rotating the light post around the telescope’s long axis.^14,16,17 In a similar fashion, the identification of middle ear structures may be valuable for continuing advancements in treatment of canine middle ear disease.

The authors found endoscopic assessment of the canine middle ear to be technically straightforward and found that endoscopy allowed subjectively easy identification of middle ear structures that typically would not be identified with an open surgical approach.⁸ In the present study, no additional dissection or widening of the bulla osteotomy over what would be performed during a traditional surgical approach was necessary prior to telescope insertion or during evaluation of the middle ear with the telescope and image acquisition. Although middle ear evaluation in this study was performed without fluid irrigation because hemorrhage was not a factor during the procedure, in a clinical setting, fluid irrigation during middle ear evaluation would likely provide improved clarity of the field and would result in thorough lavage of the middle ear cavity. Although not specifically evaluated as part of the present study, the authors also found that instruments such as a probe or curette were subjectively easily introduced into the bulla simultaneously with the endoscope.

When the telescope was first introduced into the bulla, most of the anatomic structures described could be visualized regardless of light post position. This finding reinforces the ease of endoscopic evaluation of the middle ear and means that the learning curve for proper telescope and light post positioning will likely be short. Appropriate telescope and light post positioning helps to center the described anatomic structure in the field of view and may assist with structure identification in clinical patients.

The present study had several limitations. Only 5 cadaver heads were included; thus, the results may not be representative of the general canine population. A larger number of specimens from a wider variety of breeds, including more samples from brachycephalic breeds, would have been optimal to account for normal anatomic variance. Within the small sample size evaluated in this study, the authors found the ability to identify specific structures in the middle ear cavity was consistent among specimens.

Given the results of this study, we conclude that endoscopic evaluation of the canine middle ear is feasible from both ventral and lateral approaches. Many of the major anatomic structures that exist in the middle ear could be viewed and identified through both the lateral and ventral approaches. The findings of this study may be useful as an anatomic reference guide for future video-assisted surgical procedures of the middle ear.