Tussive syncope in a pug with lung-lobe torsion

Abstract

The most common presenting clinical signs of lung-lobe torsion include dyspnea, tachypnea, lethargy, and anorexia. Tussive syncope secondary to lung-lobe torsion has not been documented. This article describes the presentation, diagnosis, management, and outcome of a pug with tussive syncope secondary to lung-lobe torsion.

The torsion of a lung lobe at its hilus is a rare but life-threatening condition in dogs and cats. Most commonly lung-lobe torsion is seen in large deep chested breeds; however, despite their small size pugs appear to be predisposed to the condition (1–4). The presenting clinical symptoms most commonly seen with lung-lobe torsion include dyspnea, tachypnea, lethargy, anorexia, pyrexia, cough, and vomiting (1–4). Less commonly pale mucous membranes, cyanosis, pain on abdominal palpation, and diarrhea are seen (1–4). While a single instance of acute collapse secondary to lung-lobe torsion has been reported in a dog (2), to the authors’ knowledge, tussive syncope secondary to lung-lobe torsion has not been previously described.

Case description

A 6.7-kg, 3-year-old, spayed female pug was presented to the Western College of Veterinary Medicine Small Animal Clinic with a 3-day history of multiple episodes of acute collapse. The dog had collapsed 8 times in the 48-hour period preceding presentation with 3 of the episodes occurring within 3 h of presentation. Each episode of collapse lasted approximately 5 to 10 s and was always preceded by a honking cough and gagging fit. After each event the owner reported that the dog seemed weak but otherwise normal. No urination, defecation or limb paddling occurred during the episodes of collapse. Based on the history, the short duration of the events, and the lack of post-event clinical signs, syncope was suspected. Other causes of collapse such as seizures, cataplexy, and narcolepsy were deemed less likely.

The dog had no prior history of trauma, cardiac disease, or respiratory problems. The patient was not on any medications and did not receive any supplements. According to client records, the dog was up-to-date on her routine vaccinations. She was the only dog in the household and was fed a hypoallergenic diet. Her activity level and appetite were both reportedly normal prior to the 72 h history of collapsing episodes.

On physical examination the patient was bright, alert, responsive, and in good body condition. Respiratory distress was not observed and her temperature was within normal limits. The patient was tachypneic (72 breaths/min; reference range: 20 to 40 breaths/min) and tachycardic [172 beats per min (bpm); reference range: 60 to 160 bpm]. No cardiac murmurs, arrhythmias, or pulse deficits were detected. Spontaneous coughing was observed during examination; however, it was difficult to elicit a cough on laryngeal or tracheal palpation. After coughing the patient would retch and swallow. Thoracic auscultation was difficult due to referred upper respiratory sounds attributable to her brachycephalic conformation but no crackles or wheezes were auscultated. However, the lung sounds on the right were diminished compared to the left. The remainder of the physical examination was unremarkable.

An emergency panel [packed cell volume (PCV), total protein (TP), blood urea nitrogen (BUN), blood glucose] was performed and the dog’s oxygen saturation was assessed using a portable pulse oximeter. No abnormalities were noted. A venous blood gas sample was normal with the exception of a mild hypokalemia [3.59 mmol/L; reference interval (RI): 3.8 to 5.6 mmol/L].

A 6-lead ECG showed a normal sinus rhythm with mild tachycardia (160 bpm). There was a right cranial deviation of the mean electrical axis suggesting a possible right heart enlargement. A heartworm antigen test was not performed as the dog had not traveled and heartworm is not endemic in this region.

A complete blood cell count was unremarkable except for a moderate leukocytosis (30 × 10⁹/L; RI: 4.8 to 13.9 × 10⁹/L) characterized by a moderate neutrophilia (24.9 × 10 9/L; RI: 3.0 to 10.0 × 10⁹/L) with slight toxic changes, and a mild monocytosis (2.7 × 10⁹/L; RI: 0.08 to 1.0 × 10⁹/L). These findings suggested a focus of established inflammation. A serum biochemistry profile was unremarkable except for a mild hypokalemia (3.6 mmol/L; RI: 3.8 to 5.6 mmol/L), mild hypercholesterolemia (6.7 mmol/L; RI: 2.7 to 5.94 mmol/L), mild hyperbilirubinemia (7 μmol/L; RI: 1.0 to 4.0 μmol/L), a mild elevation in alkaline phosphatatse (ALP) (332 U/L; RI: 9 to 90 U/L) and a mild hypoalbuminemia (27 g/L; RI: 28 to 38 g/L). The elevations in cholesterol, bilirubin, and ALP were attributed to mild cholestasis and the mild hypoalbuminemia was attributed to a negative acute phase protein response secondary to established inflammation.

Thoracic radiographs showed a diffuse and virtually complete opacification of the cranial and caudal portions of the left cranial lung lobe. A small degree of aeration in the dorsal aspect of the left cranial lobe was evident on the right lateral view (Figures 1,2). There were no air bronchograms or evidence of pleural effusion. The trachea was slightly deviated to the right on the ventrodorsal view suggesting a mass effect, but the right border of the cranial mediastinum was normal in profile making the existence of a mediastinal mass doubtful.

Figure 1.

Right lateral radiograph showing aeration in the dorsal portion of the consolidated left cranial lung lobe.

Figure 2.

Ventro-dorsal radiograph showing consolidated left cranial lung lobe with deviation of the trachea to the right.

Differential diagnoses for the opacification of the left cranial lung lobe included a lung-lobe torsion, atypical lobar pneumonia, intrapulmonary hemorrhage, or pulmonary neoplasia. Torsion of the left cranial lung lobe was strongly suspected based on the high incidence of this condition in the breed; however, the lack of pleural effusion was atypical for a lung-lobe torsion.

Ultrasound of the affected left cranial lung lobe was used to evaluate these possibilities. Hypoperfusion was documented in the left cranial lung lobe with the use of Doppler ultrasound (Figure 3). Most of the consolidated lobe was hyperechoic and swollen with small irregular areas of cavitation. These findings supported a tentative diagnosis of a left cranial lung-lobe torsion. No pleural fluid was present, nor was there evidence of a concurrent mediastinal mass.

Figure 3.

Ultrasound image showing consolidated and swollen left cranial lung lobe with a liver-like appearance.

Based on the presenting signs and diagnostic findings, tussive syncope secondary to torsion of the left cranial lung lobe was diagnosed.

Prior to surgery the patient was blood typed and clotting times [prothrombin time (PT), partial thromboplastin time (PTT)] were assessed to screen for possible disseminated intra-vascular coagulation. The dog was DEA 1.1 positive, the PT was normal, and there was a mild, clinically insignificant, prolongation of PTT (15.9 s; RI: 9.6 to 13.8 s). The dog was monitored overnight with periodic assessment of her saturation pressure of peripheral oxygen (SpO₂) and hourly assessment of her respiratory rate. No further tussive syncopal episodes occurred prior to surgery despite her experiencing multiple coughing episodes.

The following morning the dog was premedicated with hydromorphone (Sandoz Canada, Burlington, Ontario), 0.05 mg/kg body weight (BW), IV, and induced with propofol (Novopharm, Toronto, Ontario), 6 mg/kg BW, IV. She was intubated carefully to avoid endobronchial intubation and anesthesia was maintained with inhaled isoflurane (Abbott Laboratories, Montreal, Quebec) and oxygen. Cefazolin (Sandoz, Boucherville, Quebec) 22 mg/kg BW, IV was administered after induction, and continued q90 min during surgery. A left 6th intercostal thoracotomy was performed. The cranial and caudal portions of the left cranial lung lobe appeared grossly congested and were torsed around the hilus. Bupivacaine (Hospira Healthcare Corporation, Montreal, Quebec) was injected (1.2 mL total) into the dorsal aspects of intercostal spaces 3, 4, 5, 6, and 7 to block the costal nerves and the pleura. A thoracoabdominal stapler (Tyco, St. Laurent, Quebec) with a TA 30-3.5 cartridge (Tyco) was used to ligate the torsed left cranial lung lobe at its hilus and the lobe was resected. Prior to closure the remaining lung lobes were inspected for inflation; the chest cavity was filled with sterile saline and the lungs were inflated to a pressure of 20 cm H₂O. No leaks were identified. A thoracostomy tube was placed through the 8th intercostal space, and secured in the skin using a Chinese finger trap suture pattern at the 10th intercostal space. Routine closure of the thoracic cavity was performed.

Recovery from anesthesia in intensive care was uneventful. Analgesia was maintained for the first 24 h after surgery with hydromorphone, 0.05 mg/kg BW, IM, q4h, followed by a combination of oral tramadol (Wiler Fine Chemicals, London, Ontario), 3 mg/kg BW, PO, q8h, and meloxicam (Boehringer Ingelheim, Burlington, Ontario), 0.1 mg/kg BW, PO, q24h. During the first 36 h after surgery minimal amounts of air were recovered from the thoracic cavity, a normal respiratory between rate was observed, and the patient maintained a SpO₂ 96% to 99%. The thoracostomy tube was removed after 36 h and the patient was discharged the following day. No episodes of coughing or collapse were noted after surgery.

Histopathological examination of the resected lung lobe revealed changes consistent with lung-lobe torsion. The normal parenchyma of the lobe was distorted by large areas of hemorrhage and necrosis. In a few cross sections the epithelial lining of the airway was denuded and replaced by inflammatory cells and necrotic debris. In the tissues subjacent to the airways, large bands of moderately pleomorphic fibroblasts, macrophages, and aggregates of plasma cells, lymphocytes, and neutrophils were present. No underlying cause or etiologic agents were identified. While cultures of the resected lobe were not done, there was no diagnostic evidence to support an infectious agent as the cause of this dog’s lung-lobe torsion.

The patient was continuing to do well 9 mo after surgery. No further episodes of tussive syncope or coughing have been observed.

Discussion

Syncope typically occurs as a result of a sudden, severe, and transient decrease in oxygen delivery to the brain. Oxygen delivery is determined by a combination of cardiac output flow, systemic arterial oxygen tension, and hemoglobin concentration. In the vast majority of cases the syncopal episode is related to a sudden decrease in cerebral perfusion. As cerebral perfusion is determined by the difference between mean arterial pressure (MAP) and the intracranial pressure (ICP), a sudden increase in ICP, a decrease in MAP or combination of the two can lead to syncope (5). Tussive syncope is fairly common in small brachycephalic dog breeds (6). Coughing increases intrathoracic pressure and intracranial pressure transiently. The increased intrathoracic pressure results in a decreased venous return to the heart. Additionally, coughing can stimulate vagal afferents in the upper airways causing vagal efferent stimulation, resulting in bradycardia and vasodilation, both of which further reduce venous return. This results in a situation where intracranial pressure is high and cardiac output is low, potentially leading to syncope if the cerebral perfusion pressure drops sufficiently (6).

Tussive syncope in dogs is most often associated with a collapsing trachea, chronic obstructive pulmonary disease, and valvular heart disease leading to atrial enlargement (6). Although an echocardiogram or transtracheal wash was not performed, evidence to support the existence of tussive syncope secondary to atrioventricular valve disease, tracheal collapse, or chronic obstructive pulmonary disease (COPD) was not found. Other causes of syncope (6) including cardiac disorders, severe systemic hypotension, neurologic disorders, and metabolic abnormalities (e.g., hypoglycemia) were not supported by our clinical, laboratory, or imaging findings.

Possible etiologies for the tussive syncope in this dog that could not be ruled out conclusively, included complete heart block and neurocardiogenic syncope. Complete heart block through extreme vagal responsiveness during a coughing episode has rarely been shown to lead to tussive syncope (6). Although an ECG was available, no syncopal episodes occurred in hospital and we were thus unable to definitively rule this out as a possibility.

Neurocardiogenic syncope has been described in a group of related pugs, in which coughing, barking, and gagging often preceded episodes of complete cardiac standstill and concurrent syncope (6). Surface laryngeal electromyography in these dogs indicated that laryngospasm was associated with the observed transitory cardiac standstill. Stimulation of the vagal afferents in the larynx can cause a reflex loop with vagal efferent stimulation resulting in extreme bradycardia and hypotension. It should be noted, however, that in these pugs an abnormal narrowing of the bundle of His was also noted on necropsy. It was proposed that these pugs were unable to generate escape beats to prevent syncope due to abnormal conduction at the bundle of His (6). While neurocardiogenic syncope cannot be conclusively ruled out, it is unlikely as the dog in this case report has not had any syncopal episodes in the 9 mo since surgery. Given that the coughing preceded each episode, and that the cough resolved after resection of the torsed lung lobe, we hypothesize that the lung-lobe torsion was the underlying cause of this dog’s episodes of tussive syncope.

Lung-lobe torsion in dogs has been reported to occur secondary to accidental and surgical trauma, pleural effusion, and pneumothorax (1). Lung-lobe torsions are most commonly found in deep-chested dogs. A predilection to torse the right middle lobe is observed (1). This has been attributed to the narrow shape of the right middle lung lobe, its thin bronchovesicular pedicle, and its poor attachment to surrounding structures (2,3,7). Pugs, with their barrel-chested conformation do not fit this model. A recent retrospective study of 7 cases of lung-lobe torsion in pugs found that in 6 of the 7 cases no predisposing conditions were identified (4).

It has been proposed that bronchial cartilage dysplasia may result in bronchial hilus instability in brachychephalic breeds, making a lobar torsion more likely (2). However, no cartilaginous abnormalities were found on histological examination of the resected lung lobe, nor were any cartilaginous abnormalities found in the retrospective study by Murphy and Brisson (4). Further, Murphy and Brisson (4) noted that if the observed predisposition of pugs to torse lung lobes was related to dysplastic cartilage, lung-lobe torsion in other brachycephalic breeds should be seen more commonly.

In this case the patient’s left cranial lung lobe was torsed. A review of the literature reveals a predilection in pugs to torsion of the left cranial lung lobe (1–4,8,9). In 12 previously reported cases of lobar torsion in pugs where the affected lobe was identified, 11 involved the left cranial lobe. The remaining pug torsed the right cranial lobe (4) and 1 pug subsequently went on to torse the right cranial lung lobe 2 y after surgical treatment for a left cranial lung-lobe torsion (3).

The fact that most torsions in pugs involve the left cranial lobe may be due in part to the anatomic differences between the right and left cranial lobes. Previous studies have hypothesized that both portions of the left cranial lobe torse together because they share a common hilus (2,7). However, when compared to the right cranial lobe, the left cranial lobe has a longer, more pointed shape (10) that is similar to that of the right middle lobe in large deep-chested dogs and may allow easier rotation around the hilus.

While survey radiographs, angiography, computed tomography, ultrasonography, and bronchoscopy have all been used in the diagnosis of lung-lobe torsion, the imaging technique most commonly used to make a diagnosis is survey thoracic radiographs (1,8). In a recent study of 13 dogs and 2 cats with a lung-lobe torsion, pleural effusion and increased lobar opacity were found in all cases. A lobar vesicular gas pattern and a mediastinal shift were commonly found and in approximately half of the cases, a displaced lobe, a displaced trachea, and axial rotation of the carina were observed. Previous reports have attributed the lobar vesicular gas pattern to air trapping in alveoli after incomplete lobar consolidation following acute torsion, vesicular emphysema as a result of bronchial tears incurred during torsion, and proliferation of gas-forming bacteria in the torsed lobe (8). Where sequential radiographs were taken, progressive bronchial opacification was observed (8).

Common ultrasonographic findings of a lung-lobe torsion include a liver-like consolidated lung lobe (3), a central portion filled with scattered reverberating foci consistent with gas, a hyperechoic periphery, pleural effusion, and bronchi filled with echogenic fluid. Rarely, an arterial pulse may be appreciated (8).

Interestingly, no pleural effusion was visualized on plain radiographs or thoracic ultrasound in this case. However the presence of pleural effusion is a common finding in dogs with a lung-lobe torsion (1–4,8,9). Torsion of the bronchovascular pedicle causes pulmonary venous hypertension and decreased lymphatic drainage as the pulmonary veins and lymphatics are more easily occluded than the thicker, higher pressure pulmonary artery. This results in consolidation of the lung lobe, vascular stasis, necrosis of the lung tissues and subsequent development of pulmonary effusion as fluid leaks across the compromised lobar epithelium (8,9). While other causes of pleural effusion are possible (e.g., lung tumors, chylothorax), in cases with no underlying disease the removal of the torsed lung resolves the pleural effusion (3). It was suspected in this case that the lung-lobe torsion was relatively acute and tissue damage significant enough to cause pleural effusion had not yet occurred. This is supported by the short duration of clinical signs, appearance of the lung lobe on resection and the degree of necrosis seen on histological evaluation. As most pugs with a lung-lobe torsion and pleural effusion do not have evidence of underlying disease (4), this suggests that when pleural effusion is observed in a pug with a lung-lobe torsion, the torsion is likely chronic.

Computed tomography (CT) imaging was not required in this case. Computed tomography, however, is considered the diagnostic imaging modality of choice in human medicine, and is being used with increasing frequency in veterinary patients (9). It has the potential to diagnose a torsion prior to the development of significant lung and thoracic pathology as it does not rely on secondary changes such as lung lobe consolidation and pleural effusion for detection. Instead, abrupt ending bronchi, abnormal lung lobe position, and lack of contrast enhancement in an entire enlarged lung lobe are common CT findings (9). A CT scan is highly recommended in cases where a diagnosis of a lung-lobe torsion is suspected but cannot be made definitively on radiographs and ultrasound. One of the major benefits of CT is that pleural effusion does not make images difficult to interpret as seen with survey radiographs (9). Additionally, bronchoscopy may be used to diagnose a lung-lobe torsion. Partial or complete occlusion of the affected bronchus and folding or twisting of the bronchial mucosa are considered diagnostic for the condition (3).

This report documents the novel presentation of tussive syncope in a pug secondary to a lung-lobe torsion. To the knowledge of the authors, this has not been previously described. Therefore when presented with a pug having syncopal episodes after coughing, a lung-lobe torsion should be considered. CVJ