A retrospective study of 29 cases of otitis media/interna in dairy calves

Abstract

Epidemiological data, clinical findings, laboratory data, medical imaging, and outcomes were reviewed in 29 dairy calves with otitis media/interna. Age at admission ranged from 1 to 24 wk. The majority of calves were referred during winter. Clinical signs included drooping ear, ptosis, head tilt, abnormal nystagmus, strabismus, dysphagia, regurgitation, stiff neck, opisthotonos, facial hyperesthesia, and purulent aural discharge. Intranasal endoscopic examination of 5 animals revealed nasopharyngeal collapse in 4. Cerebrospinal fluid (CSF) was abnormal in all of 7 cases. Mycoplasma bovis was cultured from all but 1 case with external ear or tympanic bullae samples (n = 12), and Mycoplasma arginini was cultured from the remaining ear sample. Radiographs of the tympanic bullae were performed in 24 calves, tomodensitometry (CT) in 3 calves and ultrasound in 4 calves. According to medical imaging techniques or necropsy, 69% of the cases were classified as chronic. Mean duration of treatment was 23.3 d. The rate of clinical recovery was 75%.

Introduction

Otitis media is a common disease in dairy calves. Although its prevalence has not been established, it is probably underestimated (1). Complications that may arise from the condition include concomitant diseases such as arthritis and pneumonia, poor growth (2,3), meningitis and abscess from extension to the surrounding tissue (1), and permanent neurological signs (3). In valuable animals, the permanent head tilt remains a problem even if the animal seems to adapt and perform well. Vital prognosis is variable (0% to 100%) (1). Poor response to treatment has been related to chronicity, intracranial invasion, multisystemic disease, and etiological agent (3,4). Recently, Mycoplasma spp. was frequently isolated in otitis media (5).

Clinical findings include auricular purulent discharge and involvement of cranial nerves: VII (facial) and VIII (vestibulo-cochlear). Involvement of cranial nerve IX (glossopharyngeal) and X (vagus), supported by endoscopic examination of upper airways, has also been reported (6).

The objective of this retrospective study was to review the clinical findings, laboratory data, medical imaging results, and outcomes in dairy calves diagnosed with otitis media/interna referred to the Centre Hospitalier Universitaire Vétérinaire (CHUV) of the Université de Montréal.

Materials and methods

Medical records of calves admitted to the CHUV between February 2003 and April 2010 with a final diagnosis of otitis media or otitis interna were reviewed. Case selection was based on physical examination including: ear droop, head tilt, and/or auricular purulent discharge, or on necropsy results. Clinical diagnosis was confirmed by tympanic bullae digital radiography, tomodensitometry (CT), ultrasonography, or necropsy.

For each animal, initial data retrieved from the medical records included age, sex, breed, month, and year of admission. Case history, clinical signs, and concurrent diseases diagnosed were also recorded. Calves were considered to be chronically infected if osteolytic lesions of the tympanic bullae were identified by medical imaging techniques or at necropsy.

Laboratory results available included complete blood (cell) count (CBC), serum globulin concentration, cerebrospinal fluid (CSF) cytology, and bacteriological culture (aural swab, transtracheal wash, CSF, and synovial fluid). Bacteriologic culture results from tympanic bullae, joints, or lungs sampled during necropsy were also recorded. Reports of microscopic cytological evaluation of CSF were read and standardized according to current definitions used in the clinical pathology laboratory of the Faculté de médecine vétérinaire de l’Université de Montréal. Pleocytosis was defined as increased cellular count > 3 cells/μL. For calves that had > 1 CSF analysis, only the first sample was included. Cerebrospinal fluid was collected from the lumbosacral space and placed into collection tubes with or without EDTA. Analyses were performed within 1 h of collection. Total nucleated cell count (TNCC) and red blood (cell) count (RBC) were determined using a Neubauer hema-cytometer (Hausser Scientific, Horsham, Pennsylvania, USA). Cytocentrifugation of 200 to 400 μL of CSF was prepared using a cytocentrifuge (Cytospin 4; Shandon, Pittsburg, Pennsylvania, USA) set at 155 × g for 5 min. Slides were air-dried and stained with a 2-part Romanowski stain (Wescor, Logan, Utah, USA). Differential cell counts were made with 200 cells or with all cells when less than 200 cells were available. Microprotein concentration was determined with a spectrophotometric technique using a pyrogallol red-molybdate complex method on an automated chemistry analyzer (Synchron CX5 and Synchron DX; Beckman Coulter, Brea, California, USA).

Aerobic bacteriologic cultures were initiated on Columbia agar that contained 5% sheep blood and was incubated at 35°C with 5% CO₂ for 48 h. Anaerobic cultures were done on anaerobic Columbia blood agar with gentamicin, incubated anaerobically at 35°C for 5 d. Mycoplasma cultures were done on Hayflick agar and broth at 35°C in a candle jar, and incubated for 7 to 10 d. Mycoplasma cultures were identified by immunofluorescence (7–9). When multiple samples (transtracheal wash and both tympanic bullae) were Mycoplasma positive, all isolates were speciated except for 3 cases in which only 1 was studied.

Results from medical imaging were recorded. Radiographic examination was performed with left and right lateral oblique views. Computed tomography was done using a GE Hi-speed ZXli scanner (GE Healthcare, Minneapolis, Minnesota, USA). Ultrasound examination of both tympanic bullae was performed using a 10 MHz linear probe (10). Intranasal endoscopic examination was performed with an Olympus Evis Exera II 10-mm-diameter endoscope (Olympus Canada, Richmond Hill, Ontario).

Antimicrobial drugs used and treatment regimen were recorded. Duration of hospitalization and improvement of clinical signs during hospitalization were noted. Follow-up information was collected by telephone interview with the owners in June 2010 (minimum of 1 mo after discharge). Owners were asked about resolution of neurological signs and the time elapsed before resolution, growth, milk production (when milk producing age was reached), and overall satisfaction with the handling of the case. Clinical recovery was considered to be achieved if growth or production were normal according to the owner.

Chi-squared exact test was used to assess seasonal and yearly variation of prevalence and effect of age and length of disease on clinical recovery. T-test analysis was used to compare duration of treatment between this study and a previous one in which treatment at the CHUV was attempted on 13 cases (3). Statistical analysis was performed using computer software SAS version 9.2 (Cary, North Carolina, USA). The level of statistical significance was set at P < 0.05.

Results

Medical records from 29 dairy calves (23 Holstein, 3 Ayrshire, 3 Jersey) were included in the study. There were 28 females and 1 male. Age on admission ranged from 1 to 24 wk (mean = 7.4 wk; median = 6 wk). One to 6 cases were referred each year (yearly prevalence of 0.3% to 1.9% among all admitted calves less than 6 mo). The prevalence was significantly different from year to year (P = 0.03). Five calves were referred in spring, 2 in summer, 5 in fall, and 17 in winter. The seasonal prevalence showed statistical heterogeneity (P = 0.02) and post-hoc tests revealed a statistically higher prevalence in winter than in summer months. Differences between other seasons were not statistically significant.

Neurological signs included droopy ear, unilateral (n = 7) or bilateral (n = 13), palpebral paresis (n = 16), head tilt (n = 14), stiff neck (n = 3), vestibular strabismus (n = 2), regurgitation (n = 2), opisthotonos (n = 2), facial hyperesthesia (n = 2), dysphagia (n = 1) and abnormal nystagmus (n = 1). One calf was admitted for septic arthritis and another for lumbar neurological pathology; both had no clinical signs of otitis. When these calves were submitted to necropsy, bilateral otitis was diagnosed. Two other calves with no neurological signs were referred with fever and depression but, based on ultrasonography, they were suspected to have otitis media. Eventually, facial paralysis developed.

One animal had inspiratory and expiratory dyspnea. Severe bronchopneumonia was diagnosed using radiographic imaging, and nasopharyngeal collapse was diagnosed by upper airway endoscopy. This animal dramatically improved following tracheotomy and insertion of a tracheostomy tube. The respiratory distress disappeared and the animal seemed more comfortable. On arterial blood gas analysis, both PaO₂ and PCO₂ markedly improved. Intranasal endoscopic examination was performed on 5 calves that had dyspnea, stridor, regurgitation or dysphagia. All 5 calves had abnormalities, which included nasopharyngeal edema and collapse (n = 4) and laryngeal hemiplegia (n = 2). Aural examination was performed in 11 calves and purulent discharge was observed in 8. Concurrent diseases included pneumonia (n = 21) and septic arthritis (n = 3).

Analysis of the CSF was performed in 10 calves. Two CSF analyses were excluded due to severe blood contamination preventing analysis. One CSF analysis was excluded because lumbosacral abnormality was diagnosed at necropsy. Reference limits were: total nucleated cell count (TNCC) < 3 cells/μL and total protein 0.11 g/L — 0.33 g/L (11,12). Results were classified as mononuclear pleocytosis (n = 6) and mixed pleocytosis (n = 1) (Table 1).

Table 1.

Results of cerebrospinal fluid analysis in 10 calves

Case number	TNCC (cells/μL)a,b	Red blood cells (cells/μL)	Total protein (g/L)a	Neutrophils (%)	Lymphocytes (%)	Monocytes (%)	Macrophage (%)	Conclusion
6	5.5	7.7	0.22	7	0	0	93	Mononuclear pleocytosis
8	13.8	1.1	NA	2	38	53	7	Mononuclear pleocytosis
9	9.9	19.2	0.25	0	3	8.5	88.5	Mononuclear pleocytosis
12	13.8	0	0.33	3	19	61	17	Mononuclear pleocytosisc
15	11.55	7.7	0.16	0	3	9	88	Mononuclear pleocytosis
24	3.85	0	0.34	0	16	61	23	Mononuclear pleocytosis
26	52.25	336.6	0.25	24	8	3	65	Mixed pleocytosis
Mean	88.92		0.36
Median	12.68		0.25

^aReference interval for total protein is 0.11 to 0.33 g/L and for total nucleated cell count is 0 to 3 cells/μL, composed of lymphocytes and monocytes in approximately equal proportion.

^bTNCC — total nucleated cell count.

^cStructures suggestive of Mycoplasma spp. were observed.

NA — not available.

Cultures of the aural purulent discharges were done in 6 cases. In 2 cases, culture was done for both ears. Swabs of the bullae taken during necropsy from 6 additional cases were submitted. Five of these 12 cases also had transtracheal wash submitted, and during necropsy 3 had lung samples submitted and 2 had joint samples submitted. All samples were submitted for Mycoplasma culture, with or without aerobic and anaerobic bacteriologic cultures.

Mycoplasma spp. were cultured in the ear or tympanic bullae samples of all aforementioned cases (n = 12), 10 of which were M. bovis and 1 was M. arginini. Aerobic culture yielded Arcanobacterium pyogenes in 2 cases, Pasteurella multocida in 1 case, and was negative in 7 of 14 samples. Two transtracheal wash samples and 2 septic joint samples were positive for M. bovis (Table 2).

Table 2.

Chronicity, clinical improvement, and bacteriologic culture results

Case	X-ray or necropsy chronic lesions	Improvement	Ear discharge/tympanic bullae		Transtracheal wash/lung/others
			Aerobic/anaerobic	Mycoplasma	Aerobic/anaerobic	Mycoplasma
27	Yes	Yes	NA	M. arginini	NA	NA
7	Yes	Discharged, no follow-up	Negative	M. bovis	Pasteurella spp., Streptococcus dysgalactiae	Mycoplasma spp.a
			Negative (second ear)	Mycoplasma spp.a
8	Yes	Euthanized	Negative	M. bovis	Negative	Negative
1	Yes	Euthanized	Negative	M. bovis	NA	NA
					carpus: negative	carpus: M. bovis
22	Yes	Euthanized	Negative	M. bovis	Enterococcus spp., coliforms	NA
					elbow: negative	elbow: M. bovis
18	No	Yes	Negative	Mycoplasma spp.a	alpha-hemolytic Streptococcus	M. bovis
			Staphylococcus (second ear)	Mycoplasma spp.a
12	Yes	Yes	Staphylococcus	M. bovis	Gram-negative	Mycoplasma spp.a
					non-fermentive rods
		Necropsy	Negative	M. bovis	Negative	Negative
26	Yes	Yes	Staphylococcus	M. bovis	Negative	M. bovis
15	No	Yes	A. pyogenes, Bacillus spp.,	M. bovis	NA	NA
			Veillonella parvula
25	Yes	Euthanized	A. pyogenes, P. multocida	M. bovis	P. multocida	Negative
9	Yes	Died	Contaminants	M. bovis	Enterococcus spp.	Negative
13	Yes	Euthanized,	NA	NA	Negative	Negative
		necropsy	Lactobacillus spp., E. coli	M. bovis	brain: negative	brain: NA

^aIdentification of the Mycoplasma spp. was done on another sample.

NA — not available.

Medical imaging techniques used to confirm the clinical diagnosis included radiographs, CT, and ultrasonography. Only 1 medical imaging technique was used in 23 calves and 2 techniques were used in 4 calves. No medical imaging technique was used in 2 cases (confirmed at necropsy). Radiographs of the tympanic bullae were taken in 24 calves, CT in 3 calves, and ultrasound examination in 4 calves. Radiographic images were considered normal in 6 calves and abnormal in 17 calves (unilateral lesion, n = 5; bilateral lesions, n = 12). Among the 29 affected bullae, radiographic abnormalities included increased opacity of the space confined by the tympanic bulla, thickened wall of the bulla, lysis of the trabeculae and of the wall, increased size of the bulla and irregularity of the hyoid bone (1 bulla). The CT study in 1 case showed an accumulation of material of soft tissue density in the most ventral part of both tympanic bullae, with thickening, irregularities or lysis of the bony partitions, and thickening of the bulla wall. On ultrasound examination, abnormalities ranged from anechoic fluid accumulation in the bulla with intact trabeculae, to heterogenic hyperechoic material accumulation with lysis of the trabeculae, edema of the mucosa and irregularities, thinning and rupture of the bulla wall, leading to leakage of exudate into the subcutaneous tissues. Three calves had both ultrasound and radiograph results available. Ultrasound findings were consistent with radiographic findings, except in 1 case in which fluid and trabeculae were visible on ultrasound but no abnormality was identified on radiographs. In 3 calves, ultrasound examination was repeated during hospitalization to evaluate the response to antimicrobial therapy and to correlate to clinical progress. Two cases had initial worsening of the infection based on marked bulla wall distortion and discontinuation on ultrasound, and then remained stable. In the third case, there was resolution of the purulent material previously observed in the right bulla and regression of the amount of exudate surrounding the ruptured left bulla. The 3 calves showed clinical improvement during hospitalization. Sixty-nine percent of the cases were likely chronic according to the presence of osteolysis on radiographs, CT, or at necropsy.

Antimicrobials received prior to referral were known for 27 cases, although detailed regimens were not available. Twenty-four cases were treated with at least 1 antimicrobial drug effective against Mycoplasma spp. Antimicrobials used during hospitalization included enrofloxacin (n = 19), penicillin (n = 8), spectinomycin (n = 7), ampicillin (n = 5), oxytetracycline (n = 3), tulathromycin (n = 3), isoniazid (n = 2), trimethoprim-sulfa (n = 1) and ceftiofur (n = 1). No treatment was attempted in 2 calves which were euthanized shortly after admission because of septic arthritis and lumbosacral pathology. Tulathromycin in 2 of 3 cases was given at an interval of 7 d, as used in a study to reduce the incidence of clinical otitis media (13). For the calculation of total treatment duration, each dose was counted as 7 d in duration, although its half-life in tympanic bullae is unknown. Five calves died or were euthanized 3 to 11 d after initiation of treatment. Duration of treatment, among 22 cases (after exclusion of these 5 cases) ranged from 6 to 50 d, with a mean of 23.3 d and a median of 21 d. Eleven calves received 1 antimicrobial. The remaining treated calves received 2 (n = 12) or more (n = 4) antimicrobials, not simultaneously. Among these cases, therapy was changed in 5 cases following 3 to 5 d of treatment without clinical improvement. For other cases, therapy was changed for convenience. Average duration of hospitalization was 10 d, ranging from 2 to 29 d. Analysis using a t-test revealed no difference in the duration of treatment between this study and the previous study conducted at the CHUV (3).

Twenty-two calves were discharged, among which 8 calves showed improvement of neurological signs during hospitalization. One calf died of suppurative bronchopneumonia. One calf was euthanized despite clinical improvement because the owner did not want to keep a M. bovis positive animal in the herd. Two calves were euthanized because of septic arthritis and 2 others because of CNS neurological signs. The reason for euthanasia of the remaining euthanized calf was unspecified in the medical record.

Follow-up information was available for 12 of the 22 animals that were discharged. Time elapsed since discharge ranged from 1 mo to 6 y. Milk production information was available for 5 discharged animals. According to the owner, neurological signs had resolved in 9 cases. Five calves recovered in < 1 mo after discharge and 2 calves recovered 1 to 6 mo after discharge. The time to recovery was unknown for the 2 remaining cases. Among the 3 calves in which neurological signs had not resolved, 2 calves had persistent head tilt but adequate growth and milk production. In the third case, neurological signs worsened. The animal was able to stand but had loss of balance and was euthanized 6 wk after discharge. Two animals were considered by the owner to have abnormal growth and were culled, although only 1 of them had a concomitant disease that could have contributed to the abnormal growth. One animal was considered by the owner to have good milk production despite inferior growth and was still in the herd 3 y after discharge. Two calves received additional antimicrobial therapy after discharge. Six of 10 animals for which the information was available were still in the herd, at ages ranging from 4 mo to 6 y. When owners were asked to put their satisfaction on the outcome of the case on a scale of 1 (worst) to 10 (best), the median was 8.5 (n = 10). Statistical analysis revealed no significant association between age at admission or duration of illness and clinical recovery.

Discussion

Most of the cases of otitis media/interna in this study were chronic. However, otitis media can be subclinical, as 2 calves were diagnosed only at necropsy, and 2 other calves were subclinical upon arrival and developed neurological signs during hospitalization. The discrepancy between the duration of clinical signs reported by the owner and the evidence of osteolysis on radiographs or at necropsy indicates that it is difficult to determine duration based solely on the history. Osteolysis was used as a criterion for long-standing disease since it has been reported at necropsy in chronic cases but not in acute cases (2).

The most common clinical findings were droopy ear, palpebral ptosis, and head tilt, as reported previously (2,3). Interestingly, 1 calf had opisthotonos, stiff neck, and facial hyperesthesia that were presumed to be secondary to meningitis. However, at necropsy, the brain was examined and only otitis media was diagnosed. Nasopharyngeal collapse has been previously reported and a hypothesis of involvement of cranial nerves IX and X was advanced (6). In our study, 1 calf had dyspnea caused by nasopharyngeal collapse and bronchopneumonia; this calf dramatically improved after tracheostomy. Collapse of the secondary upper airways was believed to be a major factor in that particular case. Evidence of laryngeal hemiplegia in 2 other calves, with regurgitation (n = 1) and dysphagia (n = 1), is consistent with involvement of cranial nerves IX, X, and XI. This supports the hypothesis that nasopharyngeal collapse may occur as a secondary neurological problem. Pneumonia is the most common disease associated with otitis media (1,3). We hypothesize that nasopharyngeal collapse may occur in many cases accompanying otitis media, but because the dyspnea is assumed to be related to the concomitant bronchopneumonia, it remains undiagnosed and unaddressed. Careful clinical examination should allow detection of inspiratory dyspnea suggesting an upper airway dysfunction.

In this study, CSF was considered abnormal in all animals from which a CSF sample was collected. This result must be interpreted with caution since unusual clinical signs, possibly consistent with meningitis, may have influenced the decision to collect CSF for analysis. Results of CSF analysis from 1 calf with otitis were reported as normal when CSF was collected from the lumbosacral space but abnormal when CSF was collected from the cerebellomedullary cistern (14). Infection can extend to the cranial cavity through osteomyelitis of the petrous part of the temporal bone (15). However, in our study, a few cases had abnormal CSF results without osteolysis observed on radiographs, suggesting extension of infection by a pathway other than osteomyelitis. Therefore, in such cases, migration of the inflammation or infection could occur along the facial nerve through the internal acoustic meatus to reach the medulla. Pleocytosis may then be detected in the CSF, not associated with an abscess or diffuse meningitis (14,15).

Cerebrospinal fluid analysis in 1 calf that had strabismus and hyperesthesia revealed mononuclear pleocytosis with intra- and extra-cellular basophilic structures consistent with Mycoplasma spp. Mycoplasma bovis was isolated from the ear swab during hospitalization and from the tympanic bulla at necropsy. Samples of CSF submitted for aerobic cultures and Mycoplasma isolation were negative. However, the incidence of mycoplasmal otitis-related meningitis could be more important than previously thought (5,16).

The use of ultrasonography in the diagnosis of otitis media in calves is reported here for the first time. Although it was used in combination with radiography in 3 cases, this technique was useful for detailing abnormalities. On radiographs, loss of definition of the trabeculae can be the result of increased opacity by fluid accumulation or by lysis of the trabeculae. On the other hand, ultrasound can differentiate fluid from exudate in the bulla, and can evaluate trabeculae integrity. It was also easily repeatable for the evaluation of progression over time.

Mycoplasma bovis was isolated in all but 1 case, indicating that it is still predominant as an etiological agent in otitis media/interna in calves. M. arginini was isolated from the other ear sample, and has been reported in otitis media cases (4). Mycoplasma spp. was recovered as a single agent in 7 of 14 samples. Other pathogens isolated included P. multocida and A. pyogenes, which have both been reported in otitis media cases but not in healthy calves (17). Most calves in this study had received 1 or more antimicrobial drugs against Mycoplasma spp. before referral. The fact that Mycoplasma spp. were isolated in spite of previous treatment could be due to antimicrobial resistance or ineffective treatment regimens.

Duration of treatment is influenced by chronicity and concurrent disease (4). According to the medical imaging, most cases were chronic. For these cases, prolonged treatment was recommended. Involvement of the surrounding bone in chronic cases (2) justifies long-term antibiotic therapy. Most cases in this study had concurrent pneumonia or septic arthritis varying in severity that may have influenced the treatment time.

Tulathromycin, gamithromycin, and florfenicol are antimicrobial drugs approved for mycoplasmal infections as a single dose injection in calves in North America. However, repeated treatments are needed to increase the duration of treatment. Other antimicrobial drugs used for mycoplasma infections include spectinomycin, enrofloxacin, tylosin, tilmicosin, and tetracycline. Resistance to tilmicosin, tetracycline, and spectinomycin has been reported (3,4). Although beta-lactam antibiotics are avoided to treat a suspected mycoplasmal infection (1), the likely presence of opportunistic agents, such as A. pyogenes or P. multocida, could justify the use of beta-lactam in combination with pharmacologically compatible antibiotics effective against Mycoplasma spp.

The long-term rate of recovery based on growth or milk production in this study was 75%. Among 12 calves for which follow-up information was available, 9 calves had clinically recovered, although 2 animals had persistent head tilt. Thirteen of 22 calves that were discharged had neurological improvement during hospitalization or after discharge, but 2 of them had retarded growth. This prognosis is similar (60%) to one reported in a previous study conducted at the CHUV (3). The time elapsed between discharge and follow-up interview may have introduced an information bias, especially when animals had been removed from the herd.

For a better individual prognosis, cases would benefit from earlier diagnosis through medical imaging that would allow detection of a subclinical phase. Clinical cases benefit from a thorough clinical examination for detection of neurological complications, such as nasopharyngeal collapse and meningitis.