A4-year-old sexually intact male English Pointer weighing 22 kg (48.4 lb) was referred to the Veterinary Teaching Hospital at Ontario Veterinary College for evaluation of a progressive, productive cough of 1 year's duration and exercise intolerance of 1 month's duration. The dog had been examined because of a dry, hacking cough 10 months before referral. Two weeks prior to development of the cough, the owners had removed numerous porcupine quills from the dog. The dog was housed in indoor and outdoor runs with 2 other English Pointers that were free of clinical signs of disease. All of the dogs were field trial competitors in Alberta and Ontario.
Results of diagnostic evaluation during the 10 months prior to referral resulted in treatment of the dog with pyrantel pamoatea for intestinal parasitism and administration of a combination of novobiocin, tetracycline, and prednisoloneb for treatment of the cough. Absence of clinical improvement and suspicion of chronic bronchitis prompted subsequent treatment with prednisonec and theophylline.d Eight months after the initial examination, the dog was again evaluated by the primary care veterinarian for persistence of the cough and progression of clinical signs to intermittent open-mouth breathing and decreased energy. The dog was referred to the veterinary teaching hospital for further evaluation.
The medical history at the time of initial examination at the teaching hospital included frequent coughing, weight loss of 9 kg (19.8 lb) over 9 months, and fatigue after brief periods of activity. The body condition score was 2/5. The dog had a mildly barrel-chested appearance, and the distal portions of the forelimbs were disproportionately thickened for the dog's stature. Inspiratory effort was increased, but the respiratory rate and thoracic sounds on auscultation were considered to be normal. Coughing was not evident during the examination and was not elicited by tracheal palpation. Rectal temperature, heart rate, mucous membrane color, and capillary refill time were within reference ranges. The carpal, tarsal, and metatarsal regions were swollen and firm but not painful on palpation. The mandibular, prescapular, and popliteal lymph nodes were prominent. Results of rectal, cardiac, fundic, and abdominal examinations were considered normal.
A CBC revealed mild leukocytosis (WBCs, 16.7 × 109 cells/L; reference range, 4.9 to 15.4 × 109 cells/L), mild left shift (band neutrophils, 0.50 × 109 cells/L; reference range, 0.0 to 0.3 × 109 cells/L), monocytosis (monocytes, 3.17 × 109 cells/L; reference range, 0.0 to 1.1 × 109 cells/L), and thrombocytosis (platelet count, 611 × 109 platelets/L; reference range, 117 to 418 × 109 platelets/L). Serum biochemical analysis revealed mild hypocalcemia (calcium concentration, 2.27 mmol/L; reference range, 2.30 to 2.80 mmol/L) and hypoalbuminemia (albumin concentration, 23 g/L; reference range, 29 to 43 g/L). The serum calcium concentration was found to be within reference range after correction for hypoalbuminemia. Severe hyperglobulinemia (globulins, 70 g/L; reference range, 21 to 42 g/L) accounted for a substantial portion of the hyperproteinemia (total protein concentration, 93 g/L; reference range, 55 to 74 g/L). Results of serum electrophoresis revealed a broadbased increase in β and γ-globulins and ruled out monocolonal causes of hyperglobulinemia. Serum activities of liver-derived enzymes were mildly high (alkaline phosphatase activity, 174 U/L [reference range, 22 to 143 U/L]; γ-glutamyltransferase, 8 U/L [reference range, 0 to 7 U/L]; and alanine aminotransferase, 303 U/L [reference range, 19 to 107 U/L]). Results of urinalysis were within reference range.
Radiography of the thorax revealed air bronchograms and pulmonary radiopacity consistent with parenchymal consolidation in the caudal portion of the right lung (Figure 1). Radiography of each forelimb revealed severe periosteal reaction involving the humerus, radius, ulna, and metacarpus (Figure 2). Detection of pulmonary disease made a diagnosis of HO the most likely cause of the periosteal reaction, although infectious or immune-mediated causes could not be ruled out.
The owner declined further initial diagnostic tests until results of the initial laboratory tests became available. The dog was treated empirically for pulmonary parasitism (Filaroides spp, Capillaria spp, Paragonimus spp, Crenosoma spp, or Oslerus osleri) with fenbendazolee (50 mg/kg [22.7 mg/lb], PO, q 24 h for 14 days). Bronchoalveolar lavage and ultrasound-guided transthoracic fine-needle aspiration of the consolidated portion of the right lung lobe were subsequently performed. Cytologic assessment of BAL fluid and aspirated lung tissue revealed septic, suppurative inflammation and chronic hemorrhage. Cellular components of the suppurative inflammatory response included lytic neutrophils (70% of cells) and large macrophages (30%). The predominant bacterial populations observed were single cocci, paired cocci, chains of cocci, short rods, and filamentous organisms. An antimicrobial regimen effective against aerobic, anaerobic, gram-positive, and gram-negative organisms was initiated while results of bacterial culture were pending; the treatment regimen included amoxicillinf (22 mg/kg [10 mg/lb], PO, q 8 h) and enrofloxacing (10 mg/kg [4.5 mg/lb], PO, q 24 h). Despite the observation of bacteria cytologically, results of aerobic, anaerobic, and fungal cultures of BAL fluid and lung tissue were negative for microbial growth.
The dog's clinical condition deteriorated at home during the following week. Because chronic reaction to a foreign body (porcupine quill) and neoplasia were among the differential diagnoses, a lung lobectomy procedure was recommended to facilitate diagnosis and possible treatment. Right intercostal thoracotomy revealed consolidation of the right caudal lung lobe and adhesions to the parietal pleura. The cranial, middle, and accessory lung lobes and intrathoracic lymph nodes appeared normal. The affected lung lobe was removed with the aid of an automatic stapling device.h Anesthetic recovery was uncomplicated, and the dog was discharged 48 hours after surgery. The dog was treated with meloxicami (0.1 mg/kg [0.05 mg/lb], PO, q 24 h for 7 days) and treatment with amoxicillin and enrofloxacin was continued.
Multiple sections of the consolidated lung lobe were examined grossly for foreign material, but none was detected. Histologic analysis of pulmonary tissue revealed small regions of normal lung parenchyma surrounded by large areas in which healthy parenchyma was replaced with lytic neutrophils and macrophages. The histologic diagnosis was chronic, severe, pyogranulomatous bronchopneumonia with fibrosis. The lesion distribution was consistent with a focal source of inoculation (vs inhalation exposure). Results of aerobic culture of lung tissue were negative for bacterial growth, but anaerobic culture yielded growth of Eikenella corrodens. Results of silver stains and fungal culture were negative.
Telephone consultation with the owner 4 days after surgery disclosed that the dog was active and eating and drinking well. The dog had a mild cough, but this was attributed to its activity. The owners were encouraged to keep the dog quiet, and a follow-up consultation was planned in 2 days, but the owners called 2 days later to report that the dog had died suddenly the night before.
Right lateral radiographic view of the thorax in a dog examined because of chronic coughing and progressive exercise intolerance despite various treatments. Notice the interstitial and alveolar patterns with air bronchograms in the dorsocaudal portion of the lungs.
Citation: Journal of the American Veterinary Medical Association 228, 9; 10.2460/javma.228.9.1366
Right lateral radiographic view of the thorax in a dog examined because of chronic coughing and progressive exercise intolerance despite various treatments. Notice the interstitial and alveolar patterns with air bronchograms in the dorsocaudal portion of the lungs.
Citation: Journal of the American Veterinary Medical Association 228, 9; 10.2460/javma.228.9.1366
Right lateral radiographic view of the thorax in a dog examined because of chronic coughing and progressive exercise intolerance despite various treatments. Notice the interstitial and alveolar patterns with air bronchograms in the dorsocaudal portion of the lungs.
Citation: Journal of the American Veterinary Medical Association 228, 9; 10.2460/javma.228.9.1366
Dorsopalmer radiographic view of the distal portion of the right forelimb in the same dog as in Figure 1. Notice the marked periosteal reaction on the humerus, radius, ulna, and metacarpus.
Citation: Journal of the American Veterinary Medical Association 228, 9; 10.2460/javma.228.9.1366
Dorsopalmer radiographic view of the distal portion of the right forelimb in the same dog as in Figure 1. Notice the marked periosteal reaction on the humerus, radius, ulna, and metacarpus.
Citation: Journal of the American Veterinary Medical Association 228, 9; 10.2460/javma.228.9.1366
Dorsopalmer radiographic view of the distal portion of the right forelimb in the same dog as in Figure 1. Notice the marked periosteal reaction on the humerus, radius, ulna, and metacarpus.
Citation: Journal of the American Veterinary Medical Association 228, 9; 10.2460/javma.228.9.1366
Necropsy revealed no evidence of increased intrathoracic pressure suggestive of tension pneumo-thorax, and secure placement of all surgical clips used to ligate the caudal lung lobe was confirmed. The lungs were diffusely dark red in color and heavy. A 3 × 2-cm area of emphysema was observed in the left cranial lung lobe, but there was no evidence of tracheal or bronchial obstruction or pulmonary embolization. Histologic assessment of postmortem tissue specimens revealed severe diffuse pleuritis. Fibrin and inflammatory infiltrates were observed in many airways and alveolar spaces. The myocardium and heart valves were histologically normal. Results of bacterial cultures of lung tissue and tracheal lymph nodes were negative for growth. The radius and ulna were thickened as a result of periosteal bone production. Other organ systems were considered to be histologically normal. Diagnoses of severe subacute pleuritis, bronchointerstitial pneumonia, hypertrophic osteopathy, and pulmonary E corrodens infection were rendered.
Discussion
Eikenella corrodens is a common component of the flora of the oropharynx and gastrointestinal and urogenital systems in humans.1 The bacterium is a slow-growing, gram-negative, nonmotile, facultative anaerobic bacillus that is difficult to grow in culture,2 and can be identified on blood agar by the characteristic appearance of pitted, greenish-yellow colonies with irregular margins and emittance of a musty bleachlike odor.1 Although no other organisms were isolated in the dog of this report, concurrent infection with other organisms is common.1
In humans, E corrodens has been associated with pleuropulmonary infections, bacterial endocarditis, periodontal infections,2 illicit drug use,3 arthritis, sinusitis, pancreatic abscesses, osteomyelitis, chronic nail biting,1,4 and genital ulcers secondary to bite wounds.5 Although E corrodens was the sole pathogen isolated from a lung abscess in 1 report,6 the bacterium is usually associated with human-inflicted bite-wound infections7–10 and has been detected in 30% of such wounds.10 The organism has also been isolated from bite wounds inflicted on humans by monkeys.11Pasteurella spp is the most common organism isolated in human wounds from dog and cat bites,12,13 although in 1 study,14 E corrodens was detected in 62% of supragingival plaque samples obtained from 30 healthy dogs, suggesting that E corrodens infection from dog bites may be underestimated. Infection with E corrodens is most prevalent in persons older than 44 years, younger than 14 years, or that have underlying illness.15
To our knowledge, pulmonary infections caused by E corrodens have not been reported in veterinary species. Reports of E corrodens infection in animals are limited to 1 case16 in which the organism caused osteomyelitis in the axis of a foal; in that instance, the condition was successfully treated via dorsal laminectomy and administration of antimicrobials.
Treatment recommendations for E corrodens infections in veterinary patients are extrapolated from information in humans. The bacterium is typically susceptible to ampicillin, penicillin G, amoxicillin-clavulanic acid, trimethoprim-sulfamethoxazole, and fluoroquinolones.17 Despite appropriate antimicrobial treatment, the dog in the present report continued to deteriorate. Poor compliance on the part of the owners did not appear to be a problem. Antimicrobial resistance could have explained the treatment failure but could not be verified because susceptibility testing was not performed in the isolate cultured from the affected lung. Contamination of the bacterial culture with another organism was considered to be unlikely. Although not evident during necropsy, it is possible that a pulmonary thromboembolic event or cardiac arrhythmia caused the dog's death.
Development of HO is associated with pulmonary infections in humans and other animals, but infection with E corrodens has not been reported as an inciting condition. Although pulmonary infection with Mycobacterium fortuitum18 and Corynebacterium spp19 has been associated with HO in dogs, the most common inciting condition in dogs is primary or metastatic pulmonary neoplasia20; in that respect, the etiology of the disease in dogs is different, compared with humans in which the most common cause is congenital cyanotic heart disease.21 Other non-neoplastic intrathoracic conditions that have been associated with HO in dogs are pneumonia, heartworm disease, various forms of cardiac disease, and focal pulmonary atelectasis.22 Extrapulmonary nonmetastatic conditions that have been linked to HO in dogs are rare but include esophageal sarcoma; urinary bladder rhabdomyosarcoma; ovarian, renal, and hepatic tumors; and nephroblastoma.23–25
The mechanism leading to the periosteal changes observed in HO is poorly understood. One hypothesis is that an increase in blood flow to the distal portions of limbs leads to excessive connective tissue production and local osteogenesis.24 Another possibility is that afferent neural reflexes from the thorax to the periosteum and connective tissues of the distal portions of the limbs lead to new bone production.24,26 Regression of HO after surgical treatment of the inciting intrathoracic condition has been reported25–28 in humans and veterinary species. Lung lobectomy and administration of appropriate antimicrobials have successfully been used in treatment of HO.18,25
In 1 report,8E corrodens infection was associated with a migrating fish bone that resulted in a liver abscess. Transthoracic migration of a porcupine quill was possible in the dog of this report, although E corrodens infection has not previously been associated with quill migration. A penetrating wound through the chest wall could cause similar clinical conditions; however, the dog had no history of being bitten or otherwise wounded, and such an incident would likely have been noticed by the owner. The nature of the dog's activity may have predisposed it to inhalation of foreign material. It is also possible that the duration of disease and prior courses of treatment with antimicrobials may have eradicated an original inciting organism. The cause of death in this dog remains unknown, but subacute pleuritis and bronchointerstitial pneumonia, with or without pulmonary embolism, were suspected.
The absence of bacterial growth on culture of BAL fluid and needle-aspirate lung specimens was inconsistent with cytologic findings in the BAL fluid. Sample contamination was possible but seemed unlikely given that a pleomorphic bacterial population was observed cytologically. Insufficient yield of tissue from the needle aspirate procedure may have contributed to the negative culture results, and poor mixing of the BAL sample could have contributed to the absence of bacterial growth on culture of BAL fluid. It is possible that lung biopsy would have yielded a better tissue sample for bacterial culture.
Difficulty in culturing and identifying E corrodens may contribute to the low number of reports describing infections caused by this bacterium. Infection with the organism may result in death, despite appropriate surgical and medical treatment. Eikenella corrodens should be considered a potential pathogen in dogs with HO and pulmonary lesions, bite wounds, or results of bacterial cultures that are suspiciously negative.
ABBREVIATIONS
| HO | Hypertrophic osteopathy |
| BAL | Bronchoalveolar lavage |
Pyr–a-Pam, Pfizer, Kirkland, QC, Canada.
Delta albaplex tablets, Pharmacia and Upjohn, Enfield, Conn.
Apo-Prednisone, Apotex Inc, Toronto, ON, Canada.
Apo-Theo-LA, Apotex Inc, Toronto, ON, Canada.
Panacur, Intervet, Whitby, ON, Canada.
Amoxicillin, Pfizer, Kirkland, QC, Canada.
Baytril, Bayer, Toronto, ON, Canada.
TA 55 stapler, Borderlink Inc, Barrie, ON, Canada.
Metacam, Boehringer Ingelheim, St Joseph, Mo.
References
- 1↑
Paul K, Patel SS. Eikenella corrodens infections in children and adolescents: case reports and review of the literature. Clin Infect Dis 2001;33:54–61.
- 2↑
Mahapatra A, Samal B, Pattnaik D, et al. Antimicrobial susceptibility pattern of clinical isolates of non-fermentative bacteria. Indian J Pathol Microbiol 2003;46:526–527.
- 3↑
Angus BJ, Green ST, McKinley JJ, et al. Eikinella corrodens septicaemia among drug injectors: a possible association with ‘licking wounds.’ J Infect 1994;28:102–103.
- 4
Newfield RS, Vargas I, Huma Z. Eikenella corrodens infections. Case report in two adolescent females with IDDM. Diabetes Care 1996;19:1011–1013.
- 5↑
Rosen T, Conrad N. Genital ulcer caused by human bite to the penis. Sex Transm Dis 1999;26:527–530.
- 6↑
Hsu CY, Liu LL, Luh KT. Lung abscess caused by Eikenella corrodens: report of a case. Taiwan Yi Xue Hui Za Zhi 1989;88:828–831.
- 7
Goldstein EJ. Bite wounds and infection. Clin Infect Dis 1992;14:633–638.
- 8↑
Kessler AT, Kourtis AP. Images in clinical medicine. Liver abscess due to Eikenella corrodens from a fishbone. N Engl J Med 2001;345:e5.
- 9
Mollitt DL. Infection control: avoiding the inevitable. Surg Clin North Am 2002;82:365–378.
- 10↑
Talan DA, Abrahamian FM, Moran GJ, et al. Clinical presentation and bacteriologic analysis of infected human bites in patients presenting to emergency departments. Clin Infect Dis 2003;37:1481–1489.
- 11↑
Goldstein EJ, Pryor EP 3rd, Citron DM. Simian bites and bacterial infection. Clin Infect Dis 1995;20:1551–1552.
- 12
Talan DA, Citron DM, Abrahamian FM, et al. Bacteriologic analysis of infected dog and cat bites. Emergency Medicine Animal Bite Infection Study Group. N Engl J Med 1999;340:85–92.
- 13
Gasser I, Osset J, Olarte I, et al. Bite wound infections: study of 22 hospitalized patients [Spanish]. Enferm Infecc Microbiol Clin 1993;11:482–486.
- 14↑
Allaker RP, Young KA, Langlois T, et al. Dental plaque flora of the dog with reference to fastidious and anaerobic bacteria associated with bites. J Vet Dent 1997;14:127–130.
- 15↑
Joshi N, O'Bryan T, Appelbaum PC. Pleuropulmonary infections caused by Eikenella corrodens. Rev Infect Dis 1991;13:1207–1212.
- 16↑
Richardson DW. Eikenella corrodens osteomyelitis of the axis in a foal. J Am Vet Med Assoc 1986;188:298–299.
- 17↑
Tami TA, Parker GS. Eikenella corrodens. An emerging pathogen in head and neck infections. Arch Otolaryngol 1984;110:752–754.
- 18↑
Wylie KB, Lewis DD, Pechman RD, et al. Hypertrophic osteopathy associated with Mycobacterium fortuitum pneumonia in a dog. J Am Vet Med Assoc 1993;202:1986–1988.
- 19↑
Hesselink JW, van den Tweel JG. Hypertrophic osteopathy in a dog with a chronic lung abscess. J Am Vet Med Assoc 1990;196:760–762.
- 20↑
Watrous BJ, Blumenfeld B. Congenital megaesophagus with hypertrophic osteopathy in a 6-year-old dog. Vet Radiol Ultrasound 2002;43:545–549.
- 21↑
Anderson TP, Walker MC, Goring RL. Cardiogenic hypertrophic osteopathy in a dog with a right-to-left shunting patent ductus arteriosus. J Am Vet Med Assoc 2004;224:1464–1466.
- 22↑
Johnson KA, Watson ADJ. Skeletal diseases. In: Ettinger SJ, Feldman EC, eds. Textbook of veterinary internal medicine. 5th ed. Philadelphia: WB Saunders Co, 2000;1903–1904.
- 23
Seaman RL, Patton CS. Treatment of renal nephroblastoma in an adult dog. J Am Anim Hosp Assoc 2003;39:76–79.
- 24↑
Peeters D, Clercx C, Thiry A, et al. Resolution of paraneoplastic leukocytosis and hypertrophic osteopathy after resection of a renal transitional cell carcinoma producing granulocyte-macrophage colony-stimulating factor in a young Bull Terrier. J Vet Intern Med 2001;15:407–411.
- 25
Caywood DD, Kramek BA, Feeney DA, et al. Hypertrophic osteopathy associated with a bronchial foreign body and lobar pneumonia in a dog. J Am Vet Med Assoc 1985;186:698–700.
- 26
Schroeder H, Berry WL. Salivary gland necrosis in dogs: a retrospective study of 19 cases. J Small Anim Pract 1998;39:121–125.
- 27
Becker TJ, Perry RL, Watson GL. Regression of hypertrophic osteopathy in a cat after surgical excision of an adrenocortical carcinoma. J Am Anim Hosp Assoc 1999;35:499–505.
- 28
Shih WJ. Pulmonary hypertrophic osteoarthropathy and its resolution. Semin Nucl Med 2004;34:159–163.
