Abstract
Thirty-five cases of spontaneous pneumothorax were reviewed. In contrast to dogs, cats with an established etiology all had spontaneous pneumothorax associated with lung disease. Underlying diseases identified in affected cats included inflammatory airway disease, neoplasia, heartworm infection, pulmonary abscess and lungworm infection. Many cats were managed successfully with observation alone or needle thoracocentesis and specific therapy for their primary lung disease. Cats who present with spontaneous pneumothorax may be treated successfully with non-surgical therapies and appear to have a better prognosis than previously extrapolated from canine studies.
Introduction
Pneumothorax, or free air in the pleural space, is defined as either traumatic or spontaneous. Traumatic pneumothorax is more common and occurs as a result of damage to the lung parenchyma and/or chest wall. Treatment of traumatic pneumothorax is typically supportive, with intermittent or continuous aspiration of air, oxygen supplementation and resolution in 2–3 days time. Iatrogenic pulmonary damage, such as associated with thoracocentesis or positive pressure ventilation, may also result in pneumothorax.
Spontaneous pneumothorax (SP) occurs independently of a known, or suspected, traumatic event and can be classified as primary or secondary. Both primary and secondary SP are well-reported in humans and dogs.
Primary SP occurs as a result of rupture of apical blebs and bullae in the absence of underlying lung disease. 1 Primary SP accounts for up to 68% of SP cases in dogs; 2 many clinicians perceive lung lobectomy to be the treatment of choice, producing the most rapid resolution of signs and best long-term prognosis. Primary SP has not been reported in cats.
Secondary SP occurs in association with underlying lung disease that results in air leakage. Reported underlying pulmonary diseases associated with secondary SP in dogs include neoplasia, 3 heartworm (Dirofilaria immitis) infection, 4–7 lung abscess, 8 pulmonary thromboembolism, 9 congenital lobar emphysema,10,11 grass awn migration,12,13 bacterial pneumonia, 14 parasitic granuloma, 15 mycotic granuloma 3 and uremic pneumonitis. 16 Treatment of secondary SP in dogs is dependent upon the cause, with focal lesions (eg, tumor, abscess, foreign body) treated with lung lobectomy and diffuse disease treated supportively. Spontaneous pneumothorax is well-reported in humans, with both primary and secondary SP reported. Treatment of affected people is reflective of the underlying disease, and may include observation, thoracostomy tube placement, video-assisted thoracoscopy (VATS) or open thoracotomy. 17
There are few reports of SP in cats, and all had secondary SP with underlying diseases including inflammatory airway disease,18,19 heartworm infection, 20 lungworm (Aelurostrongylus abstrusus) infection 21 and bronchopulmonary dysplasia. 22 The ideal management approach to SP in cats is not known.
The objective of this study was to identify causes, evaluate treatment options and report the outcome of SP in cats. Owing to the retrospective nature of this study, it was not intended to address the role of surgical intervention.
Materials and methods
Case enrolment
The patient database for the Foster Hospital for Small Animals at Tufts Cummings School of Veterinary Medicine (North Grafton, MA, USA) and Tufts Veterinary Treatment and Specialties (Walpole, MA, USA) were searched using the key words ‘pneumothorax’ and ‘cats’. The medical records and radiographic reports were reviewed to exclude cats with known or suspected (ie, outside cats) trauma, and those that underwent thoracocentesis prior to diagnosis of pneumothorax. All diagnostic images [radiographs and computed tomography (CT)] were reviewed by a board-certified radiologist (RK) to confirm the presence of pneumothorax.
Classifying the pneumothorax
Pneumothorax was recorded as bilateral or unilateral, and the volume of free air as small, moderate or large based on the degree of retraction of the lung parenchyma. A pneumothorax was classified as primary spontaneous if it was associated with a bulla or bleb in the absence of parenchymal lung disease and secondary spontaneous if it was associated with airway or parenchymal lung disease.
A diagnosis of inflammatory airway disease (IAD) was made if at least two of the following were present: a bronchial pattern on thoracic radiographs, response to glucocorticoids and airway cytology consistent with inflammation without evidence of infection. Inflammatory airway infection was diagnosed if intracellular bacteria were identified on airway cytology or if there was a positive culture and a response to antibiotic therapy. Neoplasia and pulmonary abscess were identified by cytology and/or histopathology. Heartworm infection was identified by positive antigen serology and/or echocardiography. Lungworm infection was identified on histopathology or Baermann fecal sedimentation. Diagnosis was recorded as ‘inconclusive’ if test results were not definitive or ‘unknown’ if further tests were declined by the owner.
‘Survivors’ were cats that survived to discharge. ‘Non-survivors’ included cats that died in the hospital or were euthanased because of severity of disease or perceived poor prognosis without further diagnostics.
Management
The ultimate method by which the pneumothorax was managed was recorded as either observation/benign neglect, thoracocentesis (single or multiple), tube thoracostomy or thoracotomy/lung lobectomy. All cats received ancillary therapy for their underlying disease, including oxygen supplementation and drugs such as bronchodilators, glucocorticoids and antibiotics.
Results
Three-hundred-and-three cases of feline pneumothorax were identified over the study period. One-hundred-and-fifty cats had experienced a traumatic event; the pneumothorax was believed to be a result of trauma. There were 118 cases where the pneumothorax was identified after thoracocentesis and could not be definitively classified as spontaneous or iatrogenic. Thirty-five cases met the criteria for study enrolment (Table 1). The median age was 8 years (range 0.3–16 years) with 26 males (two entire) and nine females (one entire). Thirty cats were of mixed lineage and five cats were reported as purebreds.
Clinical characteristics of 35 cats with spontaneous pneumothorax
Months
SF = spayed female, CM = castrated male, DSH = domestic shorthair, DLH = domestic longhair, AMA = against medical advice, PTE = pulmonary thromboembolism
The most common presenting complaint was an acute onset of respiratory distress (24 cats; 68%). Ten cats (29%) had a history of cough and four were found collapsed. Three cats were lethargic and anorexic, two were vomiting and one was hiding.
All cats had an increased respiratory rate and effort, and 22 (63%) had abnormal lung sounds. Nine were described as harsh or increased bronchovesicular sounds; lung sounds were absent bilaterally in five cats, unilaterally in three cats and dorsally in one cat. Crackles were ausculted in three cats and wheezes in one cat. The median heart rate was 180 beats/min (range 80–300). The median rectal temperature was 99.3°F (37.4°C) with a range of <90–103.9°F (<32.2–39.9°C).
Per the study design, all 35 cats had pneumothorax identified via thoracic radiographs (Figures 1–3). The pneumothorax was bilateral in 23 cats, with five considered large volume, 10 moderate volume and eight small volume. The pneumothorax was unilateral in seven cats, with large volume in one, moderate volume in one and small volume in the remaining five. Five cats only had a single lateral radiograph performed such that it could not be determined if the pneumothorax was unilateral or bilateral.
This lateral thoracic radiograph of a cat with inflammatory airway disease is an example of a large volume pneumothorax. The radiograph is obliqued because of the cat’s respiratory distress
This lateral thoracic radiograph demonstrates an example of a small volume pneumothorax. In this cat, heartworm infection was identified
A lateral radiograph of a the cat shown in Figure 2 following 2 months of therapy with prednisolone. Note the resolution of the pneumothorax, as well as the improvement in the parenchymal pattern
Five cats also had a thoracic CT scan. The rationale for this test was not clear from the medical records, but CT scanning is commonly pursued in dogs with spontaneous pneumothorax to further characterize the pulmonary pathology. New information was gained in only one cat, where CT revealed nodules not seen on radiographs; a diagnosis of neoplasia was made via fine needle aspiration and cytology.
Five separate underlying diseases were identified in 21 cats, including IAD (nine), neoplasia (five), heartworm infection (three), pulmonary abscess (three), and lungworm infection (one) (see Table 1). In the remaining 14 cats, the underlying disease remained unclassified owing to lack of diagnostic testing.
Nineteen cats (54%) survived to hospital discharge — two of these were taken home against medical advice. Four cats died and 12 cats were euthanased. In 10 cats, the medical record documented that they were euthanased owing to severity of disease. The other two were euthanased because of perceived poor prognosis (likelihood of neoplasia) without further evaluation.
Sixteen cats were treated with observation; 10 of these cats survived to discharge. One cat died, one was euthanased for perceived poor prognosis and four were euthanased when they deteriorated clinically and their owners declined further therapy.
Twelve cats were managed with thoracocentesis and seven of these cats survived to discharge. Four were euthanased: one following relapse of pneumothorax, two after their condition deteriorated and one for perceived poor prognosis. One cat died for unknown reasons. Thoracocentesis removed between 5 ml and 615 ml of air, with the volume not recorded in two cats. Known diagnoses in cats treated with thoracocentesis included four cats with IAD and two cats with heartworm infection.
Two additional cats had continuous pneumothoraces and were treated with indwelling thoracostomy tubes. One of these cats was identified with neoplasia and was subsequently euthanased; the other cat was treated with supportive care and discharged without a final diagnosis 4 days later.
Five cats underwent exploratory thoracotomy. Three of these cats were suspected to have an isolated lung mass based on thoracic radiographs. This was confirmed at surgery for all three cats; lung lobectomy was performed and histopathology revealed neoplasia in all cases. Only one of these cats survived to discharge. One cat died postoperatively for unknown reasons and the other was euthanased after it was unable to be weaned off 100% oxygen. On necropsy, this cat also had pulmonary thromboembolism.
Records for the other two cats were incomplete so the reason for sending them to surgery could not be ascertained. Following lung lobectomy, only one cat survived to discharge. One cat was euthanased intra-operatively owing to non-resectable, diffuse parenchymal disease, later documented to be pulmonary abscessation. The other died 6 h postoperatively with a final diagnosis of pulmonary abscessation and a caudal vena caval thrombus. The remaining three cats were diagnosed with neoplasia.
Discussion
This is the first large-scale report of cats with SP and suggests differences between SP in cats compared with humans and dogs. In both humans and dogs, the most common form of SP is primary, and, in dogs, lung lobectomy is the treatment of choice. 2 This descriptive study of feline SP did not definitively identify any cases of primary SP and supported some success with medical management only. The underlying cause of the pneumothorax was not clearly identified in many affected cats and this limits the ability to adequately classify the response to therapy and to make treatment recommendations.
Deep-chested, large breed dogs are most commonly affected by primary SP and Huskies and other Northern breeds are over-represented. 2 Similarly to humans, bulla and blebs are more often seen at the apex. Thoracotomy is advised and is usually curative. Dogs also develop secondary SP, most often associated with a ruptured neoplasm, abscess or infarct, often treated successfully with surgery. Conservative (non-surgical) management of SP is not generally advised in dogs, although it may have a role in humans and in cats.
Of the cats with a definitive diagnosis based upon the criteria outlined earlier, inflammatory airway disease was most common. However, owing to limitations with unclassified/unknown cats and subsequent small sample size, the true incidence of many causes of SP is unknown. Nine cats developed secondary SP caused by IAD. Secondary SP has been previously reported in seven cats with IAD, although an iatrogenic pneumothorax could not be excluded in some of those cats because of the presence of a ‘negative’ thoracocentesis prior to radiography.18,19 In the last 10 years, as ultrasonography has become common place in emergency practice, there has presumably been a decline in the number of cats exposed to a ‘diagnostic thoracocentesis’ prior to confirmation of pleural space disease. Severe IAD, including asthma, is occasionally associated with SP in humans. 20
In dogs, the role of CT scanning in SP is actively debated. Some clinicians support the utility of CT in identifying small or multiple lesions and planning the surgical approach, 23 whereas others proceed directly to surgery. With so few cats in this study having CT scans, the utility if this imaging modality in feline SP is unclear.
Prompt surgical thoracotomy and exploration of the chest is advised in dogs with SP, with a high postsurgical cure rate and limited postoperative morbidity. In contrast to dogs, surgery was associated with a poor outcome in 4/5 cats. It is very likely that this reflects the severity and diffuse nature of the underlying disease in these cats, rather than a direct complication of the thoracotomy. As is true in any retrospective study, the decision to pursue surgery, treat medically, or euthanase in cats with SP was occasionally unclear from the medical record. Ongoing air leak is very challenging to manage medically, and surgery with resection of diseased lung or pleurodesis should be considered in these cats.
Limitations of this study include those found in many retrospective studies, including lack of follow-up and sufficient tests to determine an underlying cause in some cases. Additionally, diseases such as heartworm and/or lung worm infection may be hard to definitively exclude given small worm burdens and periodic shedding. Heartworm infection was identified by a positive antigen assay and echocardiography in cats assigned to the heartworm disease group; however, a negative antigen may not exclude heartworm disease in cats.
In humans treated conservatively with secondary SP, up to 1 in 5 recur within the first 2 years; 24 it is not clear from this report the frequency at which discharged cats might relapse. It is prudent to discuss recurrence with clients of affected cats, in particular if a diffuse underlying disease was identified.
Conclusions
Limited diagnostic evaluation in some affected cats in this retrospective study precludes providing specific treatment recommendations. In small volume pneumothoraces, conservative therapy may be associated with a good outcome. Surgical exploration is indicated in severe or persistent pneumothorax.
Footnotes
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Conflict of interest
The authors do not have any potential conflicts of interest to declare.