Ultrasound evaluation of tympanic bulla in calves

Abstract

An ultrasonographic approach for the tympanic bulla in calves is described. Ultrasound evaluation was performed through a lateral approach on 7 fetuses, 2 newborn calves, and 4 living calves of various ages. The probe was positioned caudal to the vertical mandibular ramus, ventral to the base of the ear. Three different positions of the probe were used. The tympanic bulla and adjacent structures were successfully identified and their ultrasonographic appearance described. Fluid within the tympanic bulla was visualized in 12 bullae in the cadavers and in 6 bullae in live calves. Ultrasound imaging has been shown to be useful but its diagnostic value remains to be studied in clinical cases of otitis media in calves.

Introduction

Otitis media occurs frequently in calves and causes economic losses (1). Otitis media often is concurrent with a respiratory infection, with the same bacteria isolated in both infections (2,3). The auditory tube is the primary route of infection of the middle ear even though the external acoustic meatus and bacteremia are other possibilities (3,4).

Clinical signs of otitis media in calves include dropped ear caused by facial paralysis and purulent auricular discharge when the tympanic membrane has ruptured (1–3). The condition can evolve into otitis interna and meningitis, in which case a head tilt and other neurological signs can be present. Macroscopically, accumulation of serous exudate and thickening of the mucosa are present in the early stage and advance to accumulation of pus, enlarged tympanic bulla (TB), thickening of the ventral wall, and necrosis of the trabeculations and of the temporal bone (1,4). Clinical diagnosis can be confirmed by diagnostic imaging such as radiography and computed tomography (CT), which have been previously evaluated for the assessment of the TB in calves (5). A radiographic diagnosis is based on increased opacity, thickening of the bulla wall and osteolysis of the trabeculations on a lateral-oblique view (2,5). Comparing both bullae improves diagnosis in cases of unilateral infection; however, the open-mouth view, as that performed in small animals to compare bullae simultaneously without superimposition, is not feasible in calves. Computed tomography has recently been suggested to be the gold standard for diagnosing otitis media in calves (5). Observed changes on CT are increased soft tissue opacity within the bulla, thickening of the mucosa and bulla wall, enlarged bulla, trabeculations and bulla wall osteolysis, and osteomyelitis of the temporal bone (5,6). However, the cost of CT is prohibitive in field situations. An early diagnosis by a field veterinarian supported by a practical diagnostic tool could prevent the condition from becoming chronic and the prognosis worse (2,7).

Ultrasonography of the TB has been described in the dog, cat, and rabbit (8–10). Ultrasonography is more sensitive and specific than radiography in detecting fluid in the TB in the absence of bone remodeling in cadaver specimens of dogs and cats (9,11). Greater agreement was observed between 2 ultrasonographers than 2 radiologists (9). However, in cases of chronic canine external otitis, ultrasonography was less sensitive and less specific than radiology (12). No descriptive or comparative studies are available for cattle.

The objective of this study was to establish an ultrasonographic approach for the TB and adjacent structures in calves and to describe their ultrasonographic appearance.

Materials and methods

Part 1: Identification of anatomic landmarks in cadaver specimens

The area around the TB of a neonatal bovine cadaver was dissected and the anatomical structures were identified and photographed. The head of a second neonatal bovine cadaver was boiled to aid in dissection leaving only the bony and cartilaginous structures around the TB for ultrasonographic observation. The head was submerged in water and examined using an Aloka 3500 ultrasound unit (Aloka, Tokyo, Japan) in conjunction with a linear 10 MHz probe in order to identify (ultrasonographically) the structures previously observed on dissection and establish references for the ultrasonography of cadavers.

Part 2: Ultrasonography on calf cadavers

Ultrasonography was performed on 14 TB of 7 fresh fetuses collected at a slaughterhouse. Fetal maturity was confirmed by eruption of the incisors. Fetal weight was recorded. The region caudal to the vertical mandibular ramus was clipped and coupling gel was applied. The fetuses were placed in lateral decubitus. An Aloka 3500 US unit was used at fixed frequency, depth, and focus. The region was examined using 3 different probes: linear (38 mm, Aloka UST-5546), finger-grip (convex 65°/20 mm, Aloka UST-995-7.5), and transrectal (linear 60 mm, Aloka UST-5821-7.5) at 10 MHz. Three different positions caudal to the mandible and ventral to the base of the ear were evaluated. Position 1 was parallel to the vertical mandibular ramus with the beam oriented perpendicular to the skin or slightly rostral. Position 2 was identical to the first except for a more pronounced rostral direction. Position 3 was perpendicular to the vertical mandibular ramus and just ventral to the base of the ear.

Each probe and each position was evaluated with respect to the ease of identifying the anatomic landmarks set out in part 1 (bulla wall, stylohyoid bone and cartilage, paracondylar process of the temporal bone, occipito hyoidian muscle). We noted the structure first identified and whether or not the trabeculations were visible. In position 3 only the ease of identifying the cartilage, the TB, and the trabeculations was recorded. Length and thickness of the bullae, and the distance between different bony landmarks were noted. The distance between the bulla and the paracondylar process was calculated by substracting the distance between the skin and the bulla and the distance between the skin and the paracondylar process so as to eliminate the pressure applied to the soft tissue as a variable. The distance between the bulla and the stylohoid was not calculated given its mobility; its position varies with the pressure applied through the probe.

Part 3: Ultrasonography on live calves

Ultrasonographic examinations were performed on live calves using the Aloka 3500 US unit and the linear probe at 10 MHz. The calves were restrained either standing or in sternal recumbency. Ultrasonography of the TB was first performed with a linear probe in position 1 within 12 h following the birth of 2 clinically healthy calves at the Centre Hospitalier Universitaire Vétérinaire. The ability to identify the bulla was noted. Ultrasonography of the TB was also performed on 4 live calves of various ages. The ultrasonography was performed with the linear probe in the 3 positions described previously. The calf’s age, the examiner’s ability to identify the TB, various anatomical landmarks and the trabeculations were noted, as well as which structure permitted positive identification of the TB.

Results

Part 1: Identification of anatomic landmarks in cadaver specimens

The digastric muscle is the first anatomic structure identified when incising the skin along the vertical mandibular ramus. By retracting the digastric muscle, the occipitohyoidian muscle is exposed, as well as its origin on the paracondylar process of the temporal bone and its insertion on the stylohyoid. The stylohyoid bone courses lateral to the TB first caudally then rostrally, then ends in a cartilaginous attachment to the stylohyoidian process of the temporal bone. The wall of the TB forms a depression under the stylohyoid cartilage, lined rostrally with the vaginal crest. Dorsocaudal to the stylohyoid process is the paracondylar process of the temporal bone (Figure 1). The exposed and ultrasonographically visible portion of the TB is lateral, since the ventral aspect is covered by the stylohyoid bone.

Figure 1.

Anatomical drawing (lateral view) of the TB with adjacent bony structures and positions 1, 2, and 3 of the probe. TB — tympanic bulla; PP — paracondylar process; SB — stylohyoid bone; VC — vaginal crest.

When the dissected head is submerged in water and the probe placed in position 1, the TB wall appears as a convex interface with an acoustic shadow, and the paracondylar process and stylohyoid bone are visible dorsally and ventrally, respectively (Figure 2). In position 2, the stylohyoid cartilage is visible longitudinally, superposed on the TB over all of its length, connecting the stylohyoid bone to the stylohyoid process of the temporal bone under the vaginal crest (Figure 3). In position 3, the bulla wall is convex in its caudal half and concave in the rostral half with the stylohyoid cartilage in transverse section which appears round and hypoechoic. The mandibular ramus is visible rostrally and the paracondylar process caudally to the TB (Figure 4).

Figure 2.

Ultrasound image of the TB on cadaver in position 1, with corresponding diagram. D — dorsal; V — ventral; S — skin; SC — subcutaneous tissue; PP — paracondylar process; TB — tympanic bulla; SB — stylohyoid bone.

Figure 3.

Ultrasound image of the TB on cadaver in position 2, with corresponding diagram. D — dorsal; V — ventral; S — skin; SC — subcutaneous tissue; VC — vaginal crest; TB — tympanic bulla (trabeculations are visible); C — stylohyoid cartilage; SB — stylohyoid bone.

Figure 4.

Ultrasound image of the TB on cadaver in position 3, with corresponding diagram. R — rostral; Ca — caudal; S — skin; SC — subcutaneous tissue; M — mandibular ramus; TB — tympanic bulla; C — stylohyoid cartilage; PP — paracondylar process.

Part 2: Ultrasonography on calf cadavers

The fetal weight of calf cadavers ranged from 27.5 to 50.5 kg and averaged 37.5 kg. The image obtained was similar to that of the submerged head in each of the 3 positions. However, liquid in the bulla allowed the ultrasound beam to go through the tympanic wall, revealing the trabeculations as thin hyperechoic lines perpendicular to the tympanic wall in anechoic fluid (Figure 5).

Figure 5.

Ultrasound image of trabeculations visible in a TB in position 1 with a fingergrip probe.

Landmarks surrounding the TB were in position 1 the stylohyoid bone and the paracondylar process, in position 2 the stylohyoid cartilage and the vaginal crest, and in position 3 the stylohyoid cartilage. All TB and landmarks respective to the 3 positions were identified ultrasonographically with the 3 probes. The occipitohyoidian muscle and the TB trabeculations were not visualized in all the positions with each probe. The number of evaluations where those structures were visualized is shown in Figure 6.

Figure 6.

Number of TB in which the trabeculations were visible (A) and for which the occipito hyoidian muscle was visible over the TB (B), with the linear, fingergrip and transrectal probes in all positions (trabeculations) and in positions 1 and 2 (occipito hyoidian muscle).

Overall, the linear probe offers superior handling ability compared to the fingergrip probe. Both the linear and fingergrip probes offer superior image quality and handling ability over the transrectal probe.

The trabeculations helped in identifying the TB in 12/14 cases (Figure 6). In position 1 or 2, the first structure identified with the linear and fingergrip probes was the TB. The occipitohyoidian muscle was the first structure identified when using the transrectal probe. However, in position 3, the stylohyoid cartilage was always the first structure identified, regardless of the probe used.

The visible lengths of the bulla with the linear and fingergrip probes were 1.98 ± 0.26 cm and 1.89 ± 0.28 cm, respectively, in position 1, and 1.54 ± 0.15 cm and 1.54 ± 0.14 cm, respectively in position 2. The thickness of the bulla wall was 0.077 ± 0.010 cm with the linear probe and 0.078 ± 0.016 cm with the fingergrip probe in position 1, and 0.092 ± 0.014 cm and 0.08 ± 0.013 cm, respectively, in position 2. The distance or depth differences between the bulla and the dorsal landmark (paracondylar process or vaginal crest) were 0.15 ± 0.15 cm and −0.07 ± 0.36 cm with the linear and fingergrip probes, respectively, in position 1, and 0.45 ± 0.16 cm and 0.29 ± 0.11 cm in position 2. The values for the length, thickness, and the distance were not calculated for the transrectal probe in positions 1 and 2 given the low number of measurements obtained.

Part 3: Ultrasonography on live calves

In newborn calves, identification of the bulla was possible and liquid and trabeculations were visible. The ages of the 4 oldest calves were 2, 3, 4, and 10 wk. For the 8 bullae, all were identified as well as the same adjacent structures as identified in the cadavers. In every case the first structure identified was the stylohyoid cartilage in position 3. Liquid and trabeculations were visible in both TB in the 10-week-old calf. The examination was well-tolerated by the calves.

Discussion

Ultrasonography permits the visualization of the TB and adjacent structures in cadavers and live calves. The ultrasonographic examination of the TB has been described for the dog with a ventral and lateral approach and for the cat and rabbit with a ventral approach (8–10). In the present study, only the lateral approach provided an adequate acoustic window in near-term fetuses and calves.

In position 3, particularly for the live calves of various ages, the structure first visible and permitting the identification of the bulla was the stylohyoid cartilage, visible in the transverse view as a hypoechoic round structure. A practitioner would most likely use the stylohyoid cartilage in position 3 as a reference point to find the bulla when performing ultrasonography in a clinical case of otitis media. However, the first recognized structures in the neonatal bovine cadavers and newborn calves in positions 1 and 2 were the trabeculations. As in the cat and the dog (8,9), the air-filled bulla appears as a convex interface between the bone surface and the acoustic shadow formed by the air in the TB. When the bulla is liquid filled, ultrasound beams are propagated through the bony surface and the TB content is anechoic. In this situation, identification of the bulla is easier as the trabeculations appear in anechoic liquid. In the cat, visualization of the bony septum separating the bulla in 2 compartments helps distinguish between air and liquid in the bulla. The image of liquid in the ventral compartment, from a ventral approach, shows an anechoic content and a hyperechoic convex septum with the structures deep to the septum obscured (9).

In calves, the ultrasound beam is able to penetrate the multiple bony interfaces of the wall and trabeculations if fluid is present, even to the medial wall in some cases. This difference may be explained by the low degree of ossification in the mature fetus and newborn calves compared to the cadavers of adult cats. The thickness is also a factor influencing the ability of the beam to penetrate the osseous surface. In the rabbit, the thickness of the lateral surface of the TB compared to the ventral surface prevents an adequate image from the lateral approach (10). It is possible that in older calves a greater degree of ossification or thicker bulla wall may hinder visualizing the TB content. Increased age also augments the distance that the ultrasound beam must travel through soft tissue before reaching the bulla, possibly decreasing the image quality and thus preventing diagnosis. However, liquid was observed in 1 calf at 10 wk of age. This could indicate that the ossification by 10 wk may be insufficient to prevent observation of fluid in the TB, which is beyond the most susceptible period for otitis media in calves. Nonetheless, in a calf affected with bilateral otitis media, thinning of the TB wall could occur and facilitate beam penetration.

All 3 positions of the probe are useful. Position 1 permits a better visualization of the bulla and trabeculations than position 2, and the bulla is visible over a greater distance, with all 3 probes. This distance varies with the orientation of the probe, position of the head and larynx, and the pressure applied to the hyoid apparatus. The difference in measurement between positions 1 and 2 is associated with the different surrounding structures; in position 1, the dorsal osseous landmark is the paracondylar process while in position 2 it is the vaginal crest. The observable part of the bulla in position 2 is the rostro-dorsal portion. The proportion of bulla with observable trabeculations was greatest with the different probes in position 1. However, position 2 allows visualization of the stylohyoid cartilage in a longitudinal view, therefore confirming adequate positioning of the probe. Position 3 can help to identify the bulla in the live calf even if the image is limited and the trabeculations not always visible. Rotating the probe from position 3 can allow for a more thorough examination of the bulla.

The ease of handling and image quality with the linear probe was superior when compared with the other probes evaluated in this study. Ultrasonography of the TB with a transrectal probe is not recommended as it yields a low quality image and a low ability to visualize the trabeculations. This may be explained by its increased length compared with the other probes, preventing good placement and skin contact.

The presence of liquid in the fetal and neonatal (< 12 h old) bullae was shown. This is the first recorded observation of liquid in the bulla. The overlap in time periods in which fluid is naturally present after birth and when calves most commonly have otitis media can confound the diagnosis, resulting in false positives. The extent of this overlap remains to be studied.

Accurate probe placement can be achieved in live calves standing or in sternal recumbency but is more difficult than in the neonatal bovine cadaver because the viewing angle behind the mandible varies with the type of physical restraint. The examination is nonetheless well-tolerated in non-sedated calves, giving it an advantage over CT.

The most reliable imaging diagnosis criteria of otitis media in calves include increased soft tissue opacity of the bulla content and lysis of the trabeculations (5). The dissection demonstrated that the most ventral portion of the bulla is not observable by ultrasound. Therefore, the lateral approach described in this study may not allow ultrasonographic observation of a small amount of fluid in the bulla, but mucosal edema should be visible. Unlike CT, ultrasonography can be used to distinguish between anechoic fluid in the acute phase of otitis media and heterogenous and hyperechoic purulent material in a more chronic case. In the absence of observable fluid in the bulla, a change in the bone surface must be proven to confirm a diagnosis of otitis media. A clinical study in dogs has shown a significant association between ultrasonography and CT results for bulla wall changes, but not regarding its content. The described ultrasonographic changes include an altered outline and reduced beam penetration (12). Besides subjectively evaluating bone surface irregularity, measuring its thickness can reveal an anomaly, particularly by comparing the 2 bullae in an animal that is unilaterally affected. A slight difference in probe orientation can result in measurable variation. Measures of width and height showed no significant difference on CT images between affected and unaffected bullae (5). Furthermore, it has been reported that healthy calves often have asymmetrically sized TB (1).

The duration of the disease and etiologic agent would influence the duration of treatment and prognosis (2,7). Ultrasonography could not only confirm a clinical diagnosis, but could also help the practitioner to give a better prognosis and choose an appropriate duration of treatment in light of cellular exudate and bone changes seen in chronic cases. Ultrasonography has the potential to be a useful tool but its diagnostic value remains to be studied in clinical cases of otitis media in calves.