Acquired Bilateral Laryngeal Paralysis Associated with Systemic Lupus Erythematosus in a Dog

Introduction

Systemic lupus erythematosus (SLE) is a rare multisystemic immunologic disease.^1–5 Common presenting complaints in affected dogs include fever, lameness, and dermatologic lesions. Diagnostic evaluation may demonstrate nonerosive neutrophilic or pleomorphic polyarthritis, glomerulonephritis, and hematologic abnormalities.^1–5 Although neurologic disorders such as seizures, altered mentation, and polyneuritis have been reported in dogs, these manifestations are considered infrequent.¹

Laryngeal involvement resulting in laryngeal edema, vocal cord paralysis, and life threatening airway obstruction has been documented as a presenting sign or a disease complication in humans with SLE.^6–8 The first documented case was published in 1959.⁶ To the authors’ knowledge, this is the first report of a canine patient with laryngeal paralysis associated with SLE.

Case Report

A 4 yr old female spayed Labrador retriever weighing 40 kg was presented to the author's hospital during the fall of 2009 for respiratory distress. The owner reported a 2 day history of exercise intolerance with a voice change, 1 day of wheezing, difficulty breathing, and mucoid expectoration. In the previous 2 mo, the patient had experienced an intermittent shifting lameness that was unresponsive to nonsteroidal anti-inflammatory therapy, episodic vomiting, and one episode of facial edema accompanied by swollen and crusted pinnae. The dog spent most days inside with free access to a fenced back yard; therefore, toxin exposure was considered unlikely. The referring veterinarian treated the dog with diphenhydramine^a (1.25 mg/kg per os [PO] q 8 hr), cimetidine^b (10 mg/kg PO q 12 hr), firocoxib^c (5.7 mg/kg PO q 24 hr), and aminopentamide^d (0.01 mg/kg PO q 12 hr) prior to presentation.

On physical examination, the dog was dyspneic and tachypneic with increased abdominal effort accompanied by respiratory stridor. The respiratory rate was 70 breaths/min, but the heart rate and pulse were difficult to assess due to motion of the thoracic wall. Oxygen saturation was 91% while the dog was receiving flow-by oxygen. Systolic blood pressure was 170 mm Hg, rectal temperature was 40°C, and the body condition score was 4/9. Additionally, there was ulceration of the buccal mucosa; multiple lingual ulcers; a hard palate ulcer measuring approximately 3 cm × 2 cm; enlarged popliteal and superficial cervical lymph nodes; effusion of the stifles, carpi, and tarsi; and diffuse dry seborrhea. The pinnae appeared unremarkable. Mentation was deemed appropriate, and examination of the cranial nerves was unremarkable; however, a complete neurologic examination was not performed because of the unstable condition of the patient.

Packed cell volume, total protein, and blood glucose were within normal limits. Blood lactate was mildly elevated (2.5 mmol/L; reference range, <0.8–2.1 mmol/L). Abnormalities on complete blood count and serum biochemical profile included mild hyperkalemia (5.4 mmol/L; reference range, 3.5–5.0 mmol/L), hyperglobulinemia (4.1 g/dL; reference range, 1.7–3.8 g/dL), and a mild increase in serum amylase activity (1,064 IU/L; reference range, 378–1,033 IU/L).

A presumptive diagnosis of laryngeal paralysis was made on the basis of the physical examination. IV fluid therapy was initiated^e, and the dog was sedated with 0.02 mg/kg acepromazine^f and 0.1 mg/kg butorphanol^g IV. Severe dyspnea continued despite sedation, so a light plane of general anesthesia was induced with propofol^h (4 mg/kg IV to effect), and an endotracheal tube was placed to gain control of the airway. At the time of induction and intubation, the laryngeal apparatus was judged by a board-certified criticalist (T.J.) to be bilaterally paralyzed and edematous. Anesthesia was maintained with 1.5% isofluraneⁱ. A large volume of fetid brown liquid was regurgitated during intubation and was removed with suction. While under general anesthesia, radiographs of the cervical region, thorax, abdomen, right carpus, right stifle, and right tarsus were performed as well as arthrocentesis of the right tarsal, left carpal, and left stifle joints. Fine-needle aspiration of the left popliteal lymph node and an excisional biopsy of one lingual ulcer were performed. A temporary tracheostomy tube was placed to bypass the laryngeal area and maintain control of the upper airway. The patient recovered uneventfully, and no further oxygen supplementation was necessary.

Megaesophagus (ME) was identified on the lateral cervical radiograph. The remainder of the thoracic and abdominal radiographs were unremarkable. Limb radiographs were consistent with nonerosive polyarthritis with possible mild osteoarthritis of the stifle and carpus. Joint fluid analysis revealed moderate neutrophilic inflammation of the tarsal joint and mild mononuclear inflammation of the carpal and stifle joints (Table 1). Etiologic agents were not identified in the joint fluid, and microbial culture was not performed. Cytologic examination of the lymph node was consistent with lymphoid reactivity. Lingual histopathology was consistent with an ulcer as infiltration with neutrophils, lymphocytes, plasma cells, macrophages, and few eosinophils was reported, but a primary cause for the ulcer was not apparent.

TABLE 1 Results of Joint Fluid Analysis in a 4 yr old Labrador Retriever with Systemic Lupus Erythematosis and Laryngeal Paralysis

View Fullscreen

After recovery from anesthesia, the dog was treated with metoclopramide^j (2 mg/kg/day IV as a constant rate infusion), famotidine^k (0.5 mg/kg IV q 12 hr), buprenorphine^l (0.01 mg/kg IV q 8 hr), ampicillin, and sulbactam^m (22 mg/kg IV q 8 hr), enrofloxacinⁿ (10 mg/kg IV q 24 hr), and doxycycline^o (10 mg/kg IV q 24 hr). Antibiotics were initiated to address possible aspiration pneumonia secondary to regurgitation and to treat a possible rickettsial-induced polyarthritis. The mouth was rinsed once q 24 hr with 0.12% chlorhexidine gluconate^p. Routine tracheostomy tube care included hourly inspection of the stoma site, instillation of 2 mL of sterile saline followed by sterile suction q 4 hr, and daily replacement of the inner cannula.

On day 2 of hospitalization, persistent joint swelling was noted, although the body temperature was within the normal range. Buprenorphine was discontinued and fentanyl^q was administered IV at a rate of 3 μg/kg/hr. Continued regurgitation was noted. Repeated thoracic radiographs revealed resolution of the ME and an alveolar pattern in the right middle and left caudal subsegment of the cranial lung lobe, consistent with aspiration pneumonia. Omeprazole^r (1 mg/kg PO q 24 hr) and sucralfate^s (1 g PO q 8 hr) were added to the treatment regimen. The dog's clinical condition and treatment regimen remained static through day 3 of hospitalization.

By day 4 of hospitalization, the regurgitation had ceased, the joint effusion had abated, and the dog appeared more comfortable and alert. As the initial laryngeal examination was performed on an emergency basis, a follow-up sedated laryngeal exam was performed by a board-certified surgeon (G.L.) using doxapram^t (2.2 mg/kg IV) and light induction with propofol^h (4 mg/kg IV to effect). This examination confirmed the previously diagnosed laryngeal paralysis. The dog was unable to move air through the larynx during temporary occlusion of the tracheotomy tube while under general anesthesia. Serum samples were submitted for determination of infectious disease titers (Rickettsia rickettsii, Borrelia burgdorferi, Ehrlichia canis, Babesia canis, Anaplasma platys, and Anaplasma phagocytophilum), thyroid hormones (T₄ and thyroid stimulating hormone [TSH]), antinuclear antibody titer (ANA), rheumatoid factor (RF), acetylcholine receptor antibody titer, and blood lead concentration. Infectious disease titers, acetylcholine receptor antibody titers, and blood lead levels were either negative or within normal limits. Baseline T₄ and TSH were 0.779 μg/dL and 0.133 ng/mL, respectively (reference ranges, 1.3–4.0 μg/dL and 0–0.65 ng/mL, respectively). The dog was positive for RF via latex agglutination. An ANA titer^u of >1:640 (reference range, <1:40) supported a diagnosis of SLE. An immunosuppressive dose of dexamethasone^v (0.3 mg/kg/day IV) was initiated. IV antibiotics were discontinued in favor of oral doxycycline^w (5 mg/kg PO q 12 hr), enrofloxacin^x (10 mg/kg PO q 24 hr), and amoxicillin trihydrate/clavulanate potassium^y(13.75 mg/kg PO q 12 hr). Metoclopramide and fentanyl dosages were decreased by 50%, then discontinued after 48 hr. Thoracic radiographs performed on day 5 of hospitalization were consistent with resolving aspiration pneumonia. The urine specific gravity on day 6 (while the dog was receiving fluid therapy) was 1.016, and the remainder of the urinalysis was within normal limits. A urine protein:creatinine ratio test and microalbuminuria testing were not performed.

On day 7 of hospitalization, the dog was deemed stable for general anesthesia and an arytenoid lateralization surgery was planned. After induction, electromyography (EMG) and motor nerve conduction velocities (NCV) were performed in all four limbs as well as the laryngeal and pharyngeal muscles. Motor NCV of the right tibial nerve at the hock (65.0 m/s), stifle (84.8 m/s), and hip (73.5 m/s) were normal (reference range, >55 m/s). There was no spontaneous activity in the muscles of the limbs or the arytenoid/laryngeal muscles. This finding indicated that denervation to these structures was unlikely. Because EMG was normal and airflow through the nares was documented after occlusion of the tracheotomy tube, the planned arytenoid lateralization was cancelled. The tracheotomy tube was removed, and the dog recovered uneventfully. During days 8 and 9 of hospitalization, oral antibiotic therapy and IV fluid therapy were discontinued. At the time of discharge on day 10, the tracheotomy site was healing well, and the dog was breathing comfortably through the nares without obvious stertor or signs of respiratory distress. Medications dispensed at the time of discharge included prednisone^z (2 mg/kg PO q 24 hr) and omeprazole^r (1 mg/kg PO q 24 hr). Three weeks later, the dog was examined by the referring veterinarian and multiple swollen joints were noted. Azathioprine^aa (1.5 mg/kg PO q 24 hr) was prescribed for additional immunosuppression.

Six weeks later (4 wk after starting treatment with azathioprine), a scheduled follow-up examination revealed no joint effusion or pain. Several small buccal ulcers were present, and there was a very small (1 mm) area of depigmentation on the left upper lip. All other mucocutaneous junctions were unremarkable. The tracheotomy was healing well, and the remainder of the physical examination was within normal limits. Sedated laryngeal examination revealed normal laryngeal function. There were no cytologic abnormalities in fluid collected from the stifles, right carpus, and right hock. A complete blood count revealed a mild anemia with a hematocrit of 34.7% (reference range, 37–55%), a mature neutrophilia (17.6×10³/μL; reference range, 6–17×10³/μL), and adequate platelets (226,000/μL, reference range, 200,000–900,000/μL). The buccal ulcers were not biopsied as they were considered consistent with inadequate control of the SLE. The immunosuppressive therapy was modified by changing from prednisone to methylprednisolone^aa (1.5 mg/kg PO q 24 hr) and increasing the azathioprine^bb dose to 2.5 mg/kg PO q 24 hr. The dog was prescribed cefpodoxime^cc (10mg/kg PO q 12 hr) for suspected secondary bacterial infection of the ulcerated mucosa.

Ten weeks later (three months after initial presentation), the dog was presented for respiratory distress. This appeared to be due to severe pulmonary compromise and was not consistent with laryngeal paralysis. Physical examination revealed severe generalized muscle atrophy and continued ulceration of multiple mucocutaneous junctions. At the request of the owners, the dog was euthanized. Necropsy revealed massive pulmonary arterial thrombosis (PTE) with pulmonary necrosis. The PTE was considered the cause of the respiratory distress. Other significant findings included generalized muscle wasting, vacuolar hepatopathy, and adrenal cortical atrophy. Each was considered to be associated with chronic glucocorticoid administration.

Discussion

Establishing a diagnosis of SLE may be challenging, and multiple diagnostic schemes exist. One traditional diagnostic scheme has been adapted from the 1982 American Rheumatism Association diagnostic criteria for humans and relies on the presence of at least 4/11 specific criteria.² Other clinicians require the presence of two major signs or one major and two minor signs (Table 2) with positive serology, or rely on the presence of at least two manifestations of autoimmune disease and a high serum ANA titer.^3,4SLE occurs most commonly in middle-aged medium- to large-breed dogs of either sex.^1,3,5,9 In the present case, clinical signs included lameness, joint swelling, oral ulcers, peripheral lymphadenopathy, and neurologic manifestations. Despite the absence of some of the common manifestations of canine SLE, such as renal and hematologic disorders, this patient's clinical signs of fever, polyarthritis, and lymphadenopathy, combined with an ANA titer >1:640 and a clinical response to immunosuppressive therapy, were consistent with a diagnosis of SLE. The presence of a positive latex agglutination test for RF, although nonspecific, is supportive of the diagnosis as up to 20% of canine patients with SLE are RF positive.¹ Although bacterial culture of the joint fluid was not performed, bacterial polyarthritis was considered unlikely based on the joint fluid cytology and the clinical presentation. Buccal histopathology did not support the diagnosis of SLE; however, oral biopsies are rarely beneficial in the diagnosis of SLE because ulcers are common in this region and an intact epidermis is needed to make a diagnosis.⁴

TABLE 2 Major and Minor Signs of Systemic Lupus Erythematosus

Adapted from CVT Update: Diagnosis and Treatment of Systemic Lupus Erythematosus.³

View Fullscreen

To the authors’ knowledge, this is the first reported case of laryngeal paralysis in association with SLE in a canine patient. Clinical signs of laryngeal paralysis in dogs include voice change, gagging, coughing, vomiting, dyspnea, and stridor. Most dogs present with laryngeal paralysis in late spring or early summer as elevated ambient temperature can exacerbate clinical signs.¹⁴ Although this dog presented in the fall, the ambient temperature that day was 26.6°C. A diagnosis of laryngeal paralysis is routinely confirmed by laryngeal examination performed in a lightly anesthetized patient. Alternative or confirmatory diagnostics may include EMG, histopathology of muscle biopsies, nerve transmission studies, echolaryngoscopy, and transnasal laryngoscopy.^11–14 Echolaryngoscopy has recently been described as a useful and safe technique for the diagnosis of laryngeal paralysis in dogs. Benefits include the short examination time and the ability to perform the procedure in unsedated animals, thereby minimizing the impact of sedation and anesthesia on laryngeal motion.¹² Transnasal laryngoscopy has also recently been adapted for use in dogs and has been described as a successful means to evaluate laryngeal function in sedated dogs.¹⁴ The reported sensitivity and specificity of echolaryngoscopy (60% for each) is less than those reported for transnasal laryngoscopy (100% and 70%, respectively) compared with findings of oral laryngoscopy, which remains the gold standard.¹³ The presence of stridor in this dog was consistent with laryngeal paralysis, which was confirmed by direct visualization on two separate occasions.

Differential diagnoses for acquired laryngeal paralysis include: damage to the recurrent laryngeal nerves from laceration or compression; and diffuse polyneuropathy, polymyopathy, or junctionopathy associated with systemic disorders such as myasthenia gravis, immune-mediated processes, diabetes mellitus, lead toxicity, hypoadrenocorticism, and hypothyroidism.^11,15 Advanced imaging would have been necessary to exclude a compressive lesion within the brainstem or cervical region of this patient; however, such a lesion would be unlikely to result in bilateral paralysis. Other reported causes of laryngeal paralysis were ruled out by the negative acetylcholine receptor antibody titer, normal serum biochemical profile, and undetectable serum blood lead concentration. Hypothyroidism has been reported in association with laryngeal paralysis; however, a causal association between the two diseases has not been proven. Hypothyroidism seems unlikely in this patient, but cannot be completely discounted without TSH response testing and/or measurement of free T₄ concentrations. Although the cause of the mild hyperkalemia is uncertain, hypoventilation-induced respiratory acidosis may have resulted in a shift of potassium ions from the intracellular to the extracellular fluid in exchange for hydrogen ions.¹⁶ Although the creatinine kinase activity was not measured, it is possible that the hyperkalemia was secondary to either a myopathy or myositis. The mild hyperglobulinemia likely resulted from antigenic stimulation and is not an uncommon finding in dogs with SLE (6.7% of patients with SLE have hyperglobulinemia).⁹ The ME and resultant aspiration pneumonia could have been either complications of the laryngeal paralysis or a result of focal esophageal myositis due to SLE; however, they were perhaps most likely a result of aerophagia from severe dyspnea. A paraneoplastic cause of ME was not completely excluded. Fluoroscopy may have provided more information regarding esophageal motility, but was not performed as the regurgitation resolved with appropriate symptomatic and supportive therapy. It is unfortunate that the urinalysis was not performed on initial presentation as mild renal insufficiency may have been masked by fluid therapy. Substantial proteinuria was excluded by dipstick analysis, but microalbuminuria may have been overlooked without specific testing. This may occur prior to the development of overt proteinuria, and transient microalbuminuria has been documented in human patients with SLE in the absence of clinical renal disease.^17,18

SLE-induced laryngeal paralysis has been reported in humans. Many mechanisms have been proposed, including vasculitis of the vasa nervorum, neuritis, thromboembolism to the vasculature supplying the recurrent laryngeal nerves, laryngeal lupus (arytenoiditis and cricoarytenoid arthritis), and compression of the recurrent laryngeal nerve by an enlarged pulmonary artery.⁷ Immunoglobulin and complement deposition (IgM, IgG, and C3) have been documented in the laryngeal basement membrane of one human patient with SLE. Tissue immune complex deposition results in local complement activation, mononuclear and neutrophilic cell infiltration, and subsequent local inflammation. The reason for immune complex deposition to laryngeal tissue in some patients and not others is poorly understood.⁸

Previous studies have evaluated EMG and NCV in dogs with laryngeal paralysis and found consistent EMG abnormalities, including: fibrillation potentials and positive sharp waves in the muscles of the pelvic limbs, soft palate, and tongue. The majority of dogs also had reduced NCV in one or more of the nerves tested. The combination of EMG changes and delayed NCV suggests that some dogs have generalized peripheral neuropathy contributing to laryngeal paralysis.¹⁹ In the dog described here, EMG and NCV testing did not show any evidence of generalized neuromuscular disease. The true significance of the lack of spontaneous activity in the arytenoid/laryngeal musculature is unknown. It may either have been a function of the sensitivity of the EMG examination or may have indicated an acute onset nerve deficit rather than a chronic condition. It is also possible that because other manifestations of SLE had improved at the time of electrodiagnostic testing, a focal neuritis or myositis may have resolved.

Laryngeal paralysis due to SLE in humans is responsive to corticosteroid therapy, and resolution occurs within 2–4 wk of initiating immunosuppressive therapy.^6,8 Once resolution of laryngeal paralysis occurs in humans, the symptoms rarely return, even during periods of SLE recrudescence.⁷ The dog in this case did not achieve complete remission from SLE and was eventually euthanized due to PTE. Thrombosis occurs with higher frequency in patients with SLE and is an important cause of morbidity and mortality.²⁰ Furthermore, high doses of exogenous corticosteroids have been shown to be prothrombotic in dogs.²¹ The most likely explanation for the development of PTE in this dog was the procoagulable effects of SLE in conjunction with glucocorticoid therapy. Despite this outcome, the dog did not demonstrate any signs of laryngeal paralysis from the time of discharge, and full return to clinically normal laryngeal function was documented by sedated laryngeal exam 6 wk after the initial diagnosis. In addition, histopathologic evaluation of the laryngeal apparatus was unremarkable other than muscle atrophy consistent with glucocorticoid administration.

Conclusion

In humans, laryngeal paralysis may be the first clinical sign of SLE or it may develop during the course of active disease.⁶ This report suggests that a similar event may occur in dogs with SLE and that this possibility should be considered in dogs with acquired laryngeal paralysis or in dogs with SLE that develop inspiratory stridor.