Spontaneous Feline Pneumothorax Caused by Ruptured Pulmonary Bullae Associated With Possible Bronchopulmonary Dysplasia

Introduction

Spontaneous pneumothorax is the accumulation of air within the pleural space without underlying trauma. In the dog, the most common cause of spontaneous pneumothorax is rupture of emphysematous bullae;^1–3 however, the underlying cause for bulla formation usually is not reported. Spontaneous pneumothorax is less commonly reported in the cat. Implicated diseases are parasitic infection,^4,5 feline asthma,⁶ and eosinophilic small airway inflammation.⁷ A diagnosis of pneumothorax is readily made on thoracic radiography. The lungs should be evaluated for underlying pulmonary parenchymal disease. Pulmonary bullae appear as circular, lucent cavities with thin walls; however, other differentials such as pneumatoceles or bronchial cysts should also be considered. Smaller blebs are not often seen. Radiography is known to be unreliable in the diagnosis of pulmonary bullae or blebs. In humans, computed tomography (CT) is considered the imaging modality of choice for the detection of pulmonary bullae and blebs,⁸ and a recent publication has supported similar findings in dogs.⁹ In addition, CT provides valuable information to assist surgical planning.

To our knowledge, this is the first description of the use of CT to diagnose pulmonary bullae in a cat with recurrent spontaneous pneumothorax. This is also the first reported case of possible bronchopulmonary dysplasia (BPD) as an underlying cause of feline pulmonary bulla formation.

Case Report

A 3-year-old, neutered male, domestic longhaired outdoor cat was presented to the referring veterinarian with sudden-onset respiratory distress. Thoracic radiographs revealed a pneumothorax, and thoracocentesis drained 70 mL of air. The cat was referred to an emergency center for further care.

On presentation to the emergency care center, the cat had a heart rate of 204 beats per minute with a respiratory rate of 120 breaths per minute. Mucous membranes were pink with a normal capillary refill time, and the temperature was 38.9°C. Thoracic auscultation identified reduced breath sounds on the right side. No evidence of external trauma was seen. The cat was sedated with butorphanol^a (0.22 mg/kg) and diazepam^b (0.22 mg/kg) administered intravenously (IV), and percutaneous needle thoracocentesis removed 250 mL of air from the right hemithorax and only a minimal volume of air from the left hemithorax. The cat was placed on oxygen supplementation, and once it was stable, thoracic radiography^c was performed. Radiographs revealed alveolar changes in the left cranial and right caudal lung lobes, suggestive of pulmonary hemorrhage or contusion. Traumatic pneumothorax was suspected despite the absence of a history and any physical evidence of trauma. The cat remained stable and was discharged from the hospital 1 day after presentation.

The cat was presented again to the emergency center for recurrent respiratory distress 1 day after discharge. Respiratory rate was 80 breaths per minute with increased respiratory effort and reduced respiratory sounds on the right side. Repeat thoracic radiographs confirmed reoccurrence of the pneumothorax, but this time it was bilateral with alveolar changes and collapse of all lobes, attributed to atelectasis. Thoracocentesis removed 200 mL of air from the right side of the thorax and 50 mL from the left. The cat was discharged after overnight observation, and owners were instructed to restrict the cat’s activity and have thoracic radiography repeated after 48 hours.

Ten days later, owners presented the cat with dyspnea and a respiratory rate of 80 breaths per minute. Thoracic radiography identified elevation of the heart from the sternum and collapse of lung lobes, with scalloping of the diaphragm, which indicated a tension pneumothorax. Thoracocentesis removed 150 mL of air from the right side and 75 mL from the left side. At this stage, spontaneous pneumothorax was suspected, and thoracic CT was performed to further investigate an underlying cause.

The cat was premedicated with buprenorphine^d (0.015 mg/kg subcutaneously [SC]), and anesthesia was induced with propofol^e (5 mg/kg IV) and maintained with isoflurane^f in oxygen, delivered by a Bain nonrebreathing system. The CT images of the thorax^g were acquired as transverse, 3-mm, contiguous slices that were reconstructed using a lung and soft tissue algorithm with a 3-mm interslice interval. The CT image revealed the presence of a moderate-volume bilateral pneumothorax. A round, thin-walled, air-filled structure consistent with a pulmonary bulla was identified in the apex of the right middle lung lobe [Figure 1]. Irregular, small, subpleural cystic cavities were identified at the apex of the right cranial lung lobe and the caudal portion of the left cranial lung lobe.

Following CT, the cat was recovered from anesthesia and monitored overnight in the emergency center. Thoracotomy was planned for the following day. The cat was premedicated with acepromazine^h (0.02 mg/kg) and morphineⁱ (0.3 mg/kg) SC. Anesthesia was induced with alfaxalone^j (1.5 mg/kg IV) and maintained with isoflurane^f in oxygen, delivered by a Bain nonrebreathing system with intermittent positive-pressure ventilation. A median sternotomy was performed from and including the second sternebra to the xiphoid process, using a pneumatic oscillating saw. A large pulmonary bulla was readily identified at the periphery of the right middle lung lobe [Figure 2]. Smaller blebs were identified at the caudal aspect of the left cranial lung lobe and at the caudal aspect of the right cranial lung lobe.

Partial lobectomies of the right middle, right cranial, and the caudal portion of the left cranial lung lobe were performed. The right middle lung lobe was stapled with a TA 55 linear stapler,^k while the right cranial and left cranial lung lobes were closed with a continuous overlapping suture of 4-0 polydioxanone followed by a simple continuous suture of 4-0 polydioxanone. No leaks were detected from either the staple or suture lines following saline infusion of the thoracic cavity. Saline was suctioned from the thoracic cavity, and a 10-French thoracostomy tube was placed in the right hemithorax. The sternum was closed with an overlapping figure-of-eight pattern using 1 polydioxanone sutures. The ventral thoracic musculature was closed with a continuous suture pattern of 2-0 polydioxanone. Subcutaneous tissue was closed with 3-0 continuous suture, and the skin was closed with 3-0 nylon interrupted sutures.

Lung samples were submitted for histopathological evaluation and showed abnormal alveolar structures at the periphery of the lobes, with clusters and bilayered cords of plump cells generally resembling type II pneumocytes. Cells had poorly defined borders and variably sized nuclei [Figure 3]. Occasional binucleated cells were seen. Clusters of these pneumocyte-like cells were sometimes associated with a collagenous core, but Masson’s trichrome and reticulin staining confirmed that, in general, the collagenous and elastic elements of the alveolar septa were absent [Figures 4A, 4B]. The spaces between clusters of pneumocyte-like cells were variably sized, ranging from channels reminiscent of respiratory bronchioles that have not produced alveoli to large and irregular bullae. Focal scarring of the pleural surface was present in the region of the largest bullae, and acute hemorrhage was seen on the pleural surface and within the parenchyma at the tip of one lung lobe. Some alveolar cells contained pale golden-brown pigment, confirmed by Perl’s Prussian blue staining to be hemosiderin, which indicated previous episodes of hemorrhage. Larger bronchioles at the margins of the more normal tissue were themselves normal in structure, with no evidence of mucostasis or other obstruction, and no abnormalities were apparent in the local blood vessels. A few focal accumulations of lymphocytes were in the parenchyma, including intraluminal alveolar macrophages and small numbers of mixed inflammatory cells (neutrophils and lymphocytes) in the air spaces.

Postoperative medical care consisted of supportive therapy including IV fluids, analgesia with bupivacaine^l (1.1 mg/kg) instilled through the chest drain q 4 hours, and morphineⁱ (0.3 mg/kg SC) as required. The bupivacaine and morphine were later replaced by buprenorphine^d (0.01 mg/kg SC q 8 hours) starting 36 hours postsurgically. The chest drain was removed approximately 36 hours after surgery. The cat remained stable and was discharged from the hospital on the third day. A postoperative examination 10 days after surgery showed a normal breathing pattern and thoracic auscultation. At a follow-up examination 2 months after surgery, the cat was well and had no signs of respiratory difficulties.

Discussion

In humans, spontaneous pneumothorax is classified as either primary or secondary. Primary spontaneous pneumothorax occurs in patients who are otherwise healthy with no microscopic evidence of pulmonary disease; patients are most commonly young, ectomorphic men. Spontaneous pneumothorax may result from a pulmonary collagen defect leading to formation of small, subpleural blebs or bullae that rupture.¹⁰ Cigarette smoking has been identified as a risk factor for primary spontaneous pneumothorax in people.¹¹

In two reported case series of spontaneous pneumothorax in dogs, 38% to 68% of these cases were caused by rupture of pulmonary bullae or blebs for which an underlying cause was not identified.^1,3 In addition to rupture of pulmonary bullae or blebs, other causes of spontaneous pneumothorax reported in dogs included neoplasia, migration of plant awns, bacterial pneumonia, parasitic granulomas, mycotic granuloma, Dirofilaria immitis infection, lung abscesses, and canine asthma. Although less widely reported in cats, spontaneous pneumothorax appears to be associated with parasitic infection, feline asthma, and eosinophilic small airway inflammation. In these cases, histological evidence is sparse, and feline spontaneous pneumothorax associated with rupture of pulmonary bullae does not appear to have been reported.^4–7

The proposed pathogenesis behind pulmonary bulla formation has changed over time. An earlier report postulated that the formation of pulmonary bullae in the dog was due to obstruction of small airways creating a valvular mechanism that caused air to be trapped and the lung to over-inflate. A pulmonary collagen defect also has been suggested to possibly permit bulla formation.¹² In human patients, a valvular mechanism was not identified. Instead, bullae were proposed to develop after retraction and collapse of surrounding lung away from regions of weakness.¹³ Other theories involve the presence of biochemical and collagen abnormalities,¹⁴ and more recent work correlates pulmonary bulla formation with BPD.

Bronchopulmonary dysplasia is a term that has undergone some change with time. Early definitions describe the condition in human infants associated with prematurity, acute lung injury, and aggressive ventilation. With increased survival of neonates with BPD, the definition has been expanded to encompass chronic lung disease¹⁵ resulting from incomplete or inappropriate repair of inflamed and injured lung tissue. Bronchopulmonary dysplasia results in marked enlargement of distal air spaces, but, unlike chronic obstructive pulmonary disease, BPD results from disruption of normal lung development,¹⁶ not from destruction of established alveoli. Impairment of either alveolar formation or development of pulmonary capillaries can lead to minimal alveolar wall fibroproliferation, and the abnormal development of alveoli leads to the formation of emphysematous bullae and blebs.¹⁷

Long-term pulmonary dysfunction is seen in children who survive neonatal BPD, despite gradual improvement in respiratory function and radiological parameters.¹⁸ Lesions are best visualized with CT and are characterized by multiple areas of hyperaeration, linear opacities underlying triangular subpleural opacities (which are visible on several consecutive sections), and no bronchiectasia.¹⁸ A recent study following 21 survivors of BPD to adulthood found that 84% of patients had emphysema, identified by CT.¹⁹

Clinical reports of BPD-like disease in the veterinary literature are few. Bronchial cartilage dysplasia and lobar emphysema are reported by several authors,^20–23 in addition to one case report of bronchial dysgenesis leading to lobar emphysema in a cat.²⁴ The case of the cat is unusual in that the bronchial dysgenesis affected all lung lobes except the emphysematous lobe.

The histopathological findings for the case presented in this report suggest that emphysematous regions of lung may have been associated with abnormal alveolar development. An alternate, albeit less likely, possibility is that inflammation resulted in type II pneumocyte hyperplasia, atelectasis, and secondary emphysema with bullae formation. In the cat presented herein, two possibilities are proposed: 1) the developing epithelium in the affected lobes failed to provide the required signal for vascular endothelial proliferation, or 2) peripheral vessels in the three affected lung lobes failed to appropriately signal to the epithelium to differentiate into normal alveolar structures. The result has been an apparently local dysplasia, and, while the complete normality of the remaining lung lobes cannot be guaranteed, no clinical or radiographic evidence suggests involvement of the entire organ.

Thoracic CT is extremely valuable in cases of spontaneous pneumothorax, both from a diagnostic standpoint and in assisting with surgical planning. In a study involving 12 dogs, Au et al found that although radiographs were 100% sensitive in detecting pneumothorax, they were poor in revealing the underlying cause.⁹ Computed tomography examination identified bullae or blebs in nine dogs, and the correlation was good between identification of the affected lobes by CT and surgical exploration. The use of CT in the cat presented in this report allowed identification of pulmonary blebs in both the left and right sides of the thorax, which indicated a median sternotomy approach was necessary to remove the affected portions of lung. Spontaneous pneumothorax may be treated medically with thoracocentesis, placement of a thoracostomy tube, and confinement; however, recurrent or persistent pneumothorax, or the presence of bullae or blebs, warrants surgical intervention. Studies have shown that dogs with spontaneous pneumothorax that are managed surgically have lower recurrence and mortality rates compared to dogs managed medically.^1,3

Conclusion

Spontaneous pneumothorax is rare in cats, and most documented cases are associated with underlying chronic pulmonary disease. This is the first case report of recurrent, spontaneous pneumothorax as a result of possible BPD in a cat. Continuing on from the human literature, this report also provides an explanation for the development of emphysematous bullae. In this case, CT proved extremely useful in the diagnosis of pulmonary bullae, and it assisted in surgical planning. Computed tomography should be considered in the investigation of cases of spontaneous pneumothorax, particularly when underlying pulmonary disease is not suspected.