Abstract
CASE DESCRIPTION A 7-year-old castrated male Italian Greyhound (dog 1) and an approximately 1-year-old female Labrador Retriever (dog 2) were evaluated because of respiratory distress 8 and 10 days, respectively, after a tornado.
CLINICAL FINDINGS No obvious external injuries were identified auscultation revealed decreased bronchovesicular sounds in the affected hemithorax of both dogs. Clinicopathologic changes were mild, with evidence of inflammation in both dogs. Thoracic radiography of both dogs revealed pneumothorax and pleural effusion with effacement of the diaphragm; findings on CT included severe pulmonary atelectasis of affected lung lobes with normal bronchial tree configurtion and no evidence of diaphragmatic hernia.
TREATMENT AND OUTCOME Exploratory thoracotomy of both dogs confirmed CT findings Pulmonary parenchymal damage consistent with a large rupture was found in both patients. A large hematoma was adhered to the ruptured lung lobe of dog 1. Grossly affected lung tissue was removed; histologic examination revealed atelectasis, pulmonary fib osis, thrombosis, and minimal (dog 1) to marked (dog 2) inflammation Microbial culture of lung tissue yielded no growth for dog 1 and Streptococcus spp and Escherichia coli susceptible to amoxicillin-clavulanic acid for dog 2. Dog 1 had a recurrence of pneumothorax treated by drainage with a thoracostomy tube 1 month after surgery. Eighteen months after surgery, both dogs were reportedly doing well.
CLINICAL RELEVANCE Development of clinical signs after a tornado, together with clinical, diagnostic imaging, surgical, and histologic findings led to a presumptive diagnosis of pulmonary barotrauma for both dogs. Long-term outcome for these dogs, treated at a referral hospital, was good.
A 7-year-old 8.8-kg (19.4-lb) castrated male Italian Greyhound (dog 1) was referred to the Oklahoma State Veterinary Medical Hospital for evaluation of excessive respiratory effort, coughing, and lethargy. Eight days prior to the referral examination, a tornado classified as grade 5 on the enhanced Fujita scale had developed in the area where the dog resided. According to the US National Weather Service, a tornado of this grade has winds ranging from 116 to 142 m/s (261 to 318 miles/h).1 One of the dog's owners reported having moved with the dog to an underground shelter during the tornado. In addition, the owner had experienced pressure-related middle-ear changes (ie, popping of the ears) several times as the tornado passed overhead. The dog's behavior was reportedly normal, and no health problems were evident prior to the tornado; however, the owner reported noticing a change in the dog's breathing pattern to short, rapid breaths immediately after the tornado. The owners initially elected to monitor the dog at home because they attributed the change to anxiety resulting from the storm. They subsequently became concerned when they observed that the dog was reluctant to move, would frequently take a so-called praying-dog stance (with the forelimbs extended, chest and elbow joints near the floo, and hindquarters raised in a standing position), and was anorectic. The referring veterinarian suspected the clinical signs were consistent with a possible diaphragmatic hernia and lung trauma. Radiographs were obtained, and the patient was referred to the veterinary teaching hospital for further evaluation 8 days after the storm.
On examination at the referral facility, the patient had a shallow breathing pattern (48 breaths/min) and intermittently adopted the described praying-dog stance. A prolonged capillary refill time (> 3 seconds) was observed. Thoracic auscultation revealed decreased bronchovesicular lung sounds in the left hemithorax. The remainder of the physical examination was unremarkable. There were no bruises or external wounds on examination and no findings or events reported by the owner to suggest trauma as a cause for the patient's clinical signs.
Results of a serum biochemical analysis revealed moderately high aspartate aminotransferase activity (359 U/L; reference range, 15 to 66 U/L), BUN concentration (35 mg/dL; reference range, 6 to 31 mg/dL), phosphorus concentration (30 mg/dL; reference range, 4 to 27 mg/dL), amylase activity (3,861 U/L; reference range, 290 to 1,125 U/L), and lipase activity (1,407 U/L; reference range, 77 to 695 U/L). A CBC revealed mild, acute inflammation with leukocytosis (16.6 × 103 WBCs/μL; reference range, 4.0 × 103 WBCs/μL to 15.5 × 103 WBCs/µL), neutrophilia (15.77 × 103 neutrophils/µL; reference range, 2.06 × 103 neutrophils/μL to 10.60 × 103 neutrophils/µL), and a mildly high band neutrophil concentration (332 band cells/µL; reference range, 0 to 300 band cells/µL). Prothrombin and partial thromboplastin times were within the respective reference ranges.
Thoracic radiographs (right lateral and ventrodorsal views) provided by the referring veterinarian revealed a mild pneumothorax in the left hemithorax, a moderate amount of pleural effusion in the right hemithorax, and severe consolidation of the left caudal lung lobe. Air bubbles were evident within the thoracic cavity, outside of the pulmonary parenchyma, at the fluid-gas interface. The diaphragm was easily visualized on the ventrodorsal view, but effacement with pleural effusion was found on the lateral radiograph. The radiographic findings were inconsistent with a diaphragmatic hernia or lung lobe torsion.
The patient was stabilized overnight by treatment with isotonic fluids delivered IV at a maintenance rate (60 mL/kg/d [27.3 mL/lb/d]) and placement in an oxygen cage at 40% oxygen saturation. The following day, general anesthesia was induced and contrast medium–enhanced CT (with iodixanol, 2.8 mL/kg [1.3 mL/lb], IV) of the thorax was performed by use of a 4-slice CT scanner. The findings included pneumothorax, pneumomediastinum, pleural effusion, and severe diffuse pulmonary disease, which was most pronounced within the left and right caudal lung lobes. Thick-walled cavitary lesions that might have represented pulmonary cysts, bullae, or blebs were also present within the left caudal lung lobe. No evidence of diaphragmatic hernia or lung lobe torsion was present.
On the basis of thoracic CT findings the decision was made to surgically evaluate the left caudal lung lobe. Under the same anesthetic episode as for CT, the patient was positioned in right lateral recumbency, and thoracotomy was performed through the left eighth intercostal space. An antimicrobial was administered intraoperatively (cefazolin, 22 mg/kg [10 mg/lb], IV, q 90 min). Exploration of the left caudal hemithorax revealed the diaphragm to be intact. Interestingly, the left caudal lung lobe appeared to have ruptured or burst from the inside outward, and a large hematoma was present on the surface of the pulmonary parenchyma. Lung tissue immediately adjacent to the hematoma was atelectatic. No evidence of lung lobe torsion was present. A left caudal lung lobectomy was performed, and a grossly abnormal portion of the caudal part of the left cranial lung lobe was also removed by application of a reusable, 40-mm surgical staplera with a 55-mm cartridge. A 20-F thoracostomy tube was placed at the ninth intercostal space during closure to allow for continued removal of air or fluid from the thoracic cavity after surgery.
Microbial culture of the lung tissue yielded no bacterial growth. Histopathologic findings included marked pulmonary atelectasis and hemorrhage with pleural thickening, regional fib osis, and thrombosis. Other findings included the absence of a substantial inflammato y component; only occasional lymphocytes and alveolar macrophages were seen (Figure 1). Taken together, the clinical, imaging, surgical, and histologic findings were considered likely attributable to barotrauma from the tornadic incident.
Photomicrograph of a section of affected lung tissue from a 7-year-old castrated male Italian Greyhound (dog 1) that was evaluated because of signs of respiratory distress 8 days after a tornado. A—Marked pulmonary atelectasis and hemorrhage are evident. H&E stain; bar = 50 μm. B—At higher magnification pulmonary atelectasis is more pronounced. There is moderate alveolar histiocytosis and organizing fib osis, but overall, a substantial inflammato y component is lacking. There is no evidence of neoplasia or foreign material. H&E stain; bar = 20 μm.
Citation: Journal of the American Veterinary Medical Association 248, 11; 10.2460/javma.248.11.1274
The patient was moved to the intensive care unit for recovery after surgery. A constant rate IV infusion of a mixture of fentanyl (5 μg/kg/h [2.3 μg/lb/h]), lidocaine (25 μg/kg/h [11.4 μg/lb/h]), and ketamine (30 μg/kg/h [13.6 μg/lb/h]; all 3 drugs combined in a single syringe) was administered for analgesia after surgery. Cefazolin administration was continued at 22 mg/kg, IV, every 8 hours. A urinary catheter was placed to allow monitoring of urine output for 2 days after surgery. Four days after surgery, IV medications were discontinued and treatment with cephalexin (27 mg/kg [12.3 mg/lb], PO, q 12 h) and tramadol (3.9 mg/kg [1.8 mg/lb], PO, q 8 h) was initiated. The thoracostomy tube was removed 2 days after surgery because fluid drainage had ceased. The dog was discharged from the hospital 6 days after initial evaluation; prognosis for a full recovery was considered good.
Approximately 1 month after the initial referral examination and surgery, the patient was reevaluated because of signs of respiratory distress. Tachypnea (68 breaths/min) and dyspnea were observed on physical examination. Thoracic auscultation revealed decreased pulmonary and cardiac sounds in the left hemithorax. Thoracocentesis was performed, and 1,200 mL of air was recovered from the left hemithorax. Evaluation of thoracic radiographs (right and left lateral and ventrodorsal views) revealed moderate to severe pneumothorax, with a possible bulla in the accessory or right caudal lung lobe; however, an artifact from gas and fluid accumulation in the pleural space could not be ruled out. The patient was anesthetized, and a thoracostomy tube was placed at the left ninth intercostal space.
A 3-bottle continuous suction systemb was used to evacuate air from the thoracic cavity. A constant rate infusion of fentanyl (5 μg/kg/h [2.3 μg/lb/h], IV) was administered for analgesia. Continuous suction was discontinued 6 days after hospital admission, and intermittent aspiration via the chest tube was performed until its removal. During that time, the analgesic treatment was changed to tramadol (5 mg/kg [2.3 mg/lb], PO, q 8 h). The thoracostomy tube was removed 2 days later when the amount of evacuated air was negligible. On follow-up by telephone conversation with the owner 18 months after the recurrence of the patient's pneumothorax, the patient was described as healthy with no signs of recurring respiratory distress.
An approximately 1-year-old 29.5-kg (64.9-lb) female Labrador Retriever (dog 2) was referred to the same veterinary teaching hospital as dog 1 because of signs of respiratory distress that were fi st observed approximately 10 days after the previously described tornado. The dog had been found as a stray after the tornado and had been admitted to a disaster relief shelter. No previous medical history was available, but the dog was presumed to have been healthy prior to the incident on the basis of young age and ideal body condition score (5/9). No external wounds or bruising had been identified on evaluation by a veterinarian at admission to the disaster relief shelter. While housed at the shelter, the dog developed signs of respiratory distress. Findings on auscultation of lung fields by an attending veterinarian were considered suggestive of pneumothorax, and the dog was subsequently referred for further evaluation and care.
Examination at the veterinary teaching hospital revealed that the patient was mildly tachypneic (40 breaths/min) with notable abdominal effort on respiration. Thoracic auscultation revealed decreased to absent bronchovesicular lung and cardiac sounds in the right hemithorax. No obvious external injuries such as bruising or puncture wounds were identified and remaining results of the physical examination were considered normal. All results of serum biochemical analysis were within the respective reference ranges. A CBC revealed a mildly high concentration of band neutrophils (676 band cells/µL) and a moderately high monocyte concentration (2,197 monocytes/µL; reference range, 0 to 840 monocytes/µL). Results of a urinalysis were unremarkable. Thoracocentesis of the right hemithorax shortly after the examination resulted in evacuation of 150 mL of serosanguinous fluid Cytologic evaluation of the fluid as not performed.
Thoracic radiographs (right lateral, left lateral, and ventrodorsal views) obtained before and after thoracocentesis revealed bilateral pleural effusion that was marked before the treatment and moderate afterward. Effacement of the diaphragm with pleural effusion was also identified on thoracic radiographs before and after thoracocentesis. A small volume of free air (mild pneumothorax) was evident on the radiographs obtained after thoracocentesis. Differential diagnoses for pneumothorax included a secondary change resulting from a ruptured cyst, bulla, or bleb or from previous thoracocentesis. An abrupt narrowing of the right middle lobar bronchus, which could have been secondary to lung lobe torsion or compression, was also observed.
Dog 2 was stabilized overnight with a bolus of isotonic fluid (22.3 mL/kg [10.1 mL/lb], IV). Once the bolus was complete, isotonic fluids were delivered IV at a maintenance rate (60 mL/kg/d), the patient was placed in an oxygen cage (40% oxygen saturation), and treatment with ampicillin-clavulanic acid (22 mg/kg, IV, q 8 h) was initiated. During the night, the patient became dyspneic, and 550 mL of serosanguinous fluid was removed from the left hemithorax via thoracocentesis.
The day after initial evaluation and stabilization, general anesthesia was induced and contrast medium–enhanced CT of the thorax (with iodixanol, 2.6 mL/kg [1.2 mL/lb], IV) was performed as described for dog 1. A large volume of air was observed within the dorsal pleural space, with a greater amount present in the right hemithorax than in the left. A moderate amount of non–contrast-enhancing, soft tissue attenuating material was found in the ventral aspect of the pleural space. A focal alveolar pattern was evident in the right caudal lung lobe. Pleural margins of this lobe were irregular, with several defects extending into the pulmonary parenchyma. Atelectasis was observed in the ventrally dependent aspects of the right lung fields and caudal aspect of the left cranial lung lobe. Pleural effusion was also evident. No evidence of diaphragmatic hernia or lung lobe torsion was found. During the same anesthetic episode, a thoracotomy was performed through the right sixth intercostal space. The right caudal lung lobe was atelectatic with an irregular pleural surface and the same ruptured-tissue appearance as in dog 1, and there were no findings consistent with lung lobe torsion or diaphragmatic hernia. A right caudal lung lobectomy was performed, and the lung tissue was submitted for histologic evaluation and microbial culture.
Culture and susceptibility testing of isolates revealed a β-Streptococcus spp and Escherichia coli, which were both susceptible to amoxicillin-clavulanic acid. Histopathologic findings were similar to those for dog 1 with respect to marked pulmonary atelectasis, hemorrhage, vascular thrombi, and regional fib osis. However, features of maturity and chronicity were evident in the sample from dog 2, with a marked amount of granulation tissue affecting the pleural surfaces, severe alveolar histiocytosis, and fib osis of bronchiolar airways in a manner similar to that described for bronchiolitis obliterans (Figure 2).
Photomicrograph of a section of affected lung tissue from an approximately 1-year-old female Labrador Retriever (dog 2) that was admitted to a disaster relief shelter as a stray and developed signs of respiratory distress approximately 10 days after a tornado. A—Similar to the histologic results for dog 1, findings in dog 2 included pulmonary atelectasis, vascular thrombosis, and multifocal presence of fibrin marked regional alveolar histiocytosis was also present. H&E stain; bar = 50 μm. B—At higher magnification alveolar spaces filled with hemorrhage, alveolar histiocytes, rare fibrin plugs, and early organization via fib osis are evident. H&E; bar = 20 μm.
Citation: Journal of the American Veterinary Medical Association 248, 11; 10.2460/javma.248.11.1274
After surgery, dog 2 was moved to the intensive care unit and received maintenance fluids as described as well as a constant rate infusion of fentanyl (5 μg/kg/h, IV). Supplemental oxygen delivery (2 L/min) was provided intranasally for 24 hours. Carprofen administration (2.2 mg/kg [1.0 mg/lb], PO, q 12 h) was started the morning after surgery to supplement the existing analgesic protocol; over the next 2 days, the treatment was transitioned to orally administered antimicrobial (amoxicillin-clavulanic acid, 12.8 mg/kg [5.8 mg/lb], q 12 h) and analgesic (tramadol, 3.4 mg/kg [1.5 mg/lb], q 8 h) medications, and IV fluid therapy was discontinued. The thoracostomy tube was removed after fluid production reached a plateau at 2.3 mL/kg/d [1.0 mL/lb/d]). The patient was discharged from the hospital and adopted by a veterinary student 8 days after the initial examination. Eighteen months after surgery, the patient was reportedly doing well.
Discussion
The pulmonary damage in both dogs of this report was postulated to be caused by a change in barometric pressure during a grade 5 (enhanced Fujita scale) tornado. Neither dog had evidence of external traumatic injury, and 1 dog was known to have been kept in an underground shelter with its owner during the tornado. In human medicine, pulmonary parenchymal damage related to barometric differences occurs when pulmonary pressure is either lower or higher than environmental pressure. Pulmonary overinflation syndrome is most commonly described in human divers who ascend to the surface of the water too quickly while holding their breath2; the syndrome is characterized by the expansion of nitrogen gas trapped within the lung, resulting in arterial gas embolism and, in rare instances, rupture of alveolar air sacs,2,3 which can lead to pneumothorax and pneumomediastinum. Pulmonary barotrauma has also been described in aviators and airline passengers.4 In this situation, the pilot and passengers are exposed to low barometric pressure if the plane cabin pressurization system malfunctions. Rapid decompression of the tissues causes an increase in the partial pressure of the intrapulmonary gases and gas emboli, resulting in damage to the lung parenchyma.3
Pulmonary damage can also occur as a blast injury in people and nonhuman animals. When an explosive is detonated, a brief period of overpressurization secondary to a sudden shock wave occurs. The initial increase in pressure is followed by negative pressure and then a return to normal barometric pressure.5 One of the common findings in people with blast injuries is ruptured tympanic membranes; the owner of dog 1 had experienced so-called ear-popping, which is typically related to pressure changes, during the storm.
To the authors’ knowledge, pulmonary overinfltion caused by pressure changes during a tornado has not been reported in human or veterinary medical literature. The largest recorded drop in atmospheric pressure within a tornado was 100 millibars (75 mm Hg).6 Some investigators have examined changes in barometric pressures associated with tornadoes; authors of 1 study7 reported a pressure drop of 24 millibars (18 mm Hg) in a tornado of enhanced Fujita grade 3 in Kansas, and another reported a drop of 40 millibars (30 mm Hg) in the core of an enhanced Fujita grade 0 tornado in Texas.8 Winn et al9 documented a drop of 55 millibars (41 mm Hg) and a subsequent rise in pressure of 60 millibars (45 mm Hg) after a tornado had passed; in that study,9 the tornado passed within 660 m of the closest measurement instrument, so it is likely that the areas even closer to the core experienced a much greater variation in pressure. We consider it likely that a rapid change to a lower barometric pressure was the cause of the alveolar damage and rupture observed in the dogs of this report.
Diagnosis and treatment of pulmonary injury associated with barotrauma, or pulmonary overinflatio syndrome, in human patients is made on the basis of patient history, diagnostic imaging, and response to treatment. Supplemental oxygen treatment is provided until thoracic imaging can be completed. In case reports2,10 of underwater loggers and breath-holding divers with this condition, the trauma was medically managed with continued oxygen therapy and thoracostomy tube placement. Diagnoses in these cases were made on the basis of results of thoracic radiography and CT findings In a human patient who underwent thoracic CT, pulmonary bulla, pneumomediastinum, and mediastinal emphysema were present.10 In premature infants with life-threatening pulmonary overinflation resulting from prolonged unilateral mechanical ventilation, emergency pneumonectomy or contralateral selective intubation can be attempted to resolve hypoxemia.11 In our patients, removal of the diseased lung tissue was performed to resolve pneumothorax and pleural effusion. The intraoperative finding of a lung lobe that appeared to have ruptured from the inside out in both patients was not expected. Therefore, it was only after the histologic findings were reported that pulmonary barotrauma was considered as a differential diagnosis. Being unaware of this condition as a differential diagnosis, we did not consider attempting medical treatment prior to surgical intervention; however, despite the postsurgical diagnosis, we suspect that surgery would still have been indicated in these 2 dogs, given the severity of lung lesions present and the likelihood that the condition would have persisted.
The initial differential diagnoses for the signs of respiratory distress in the dogs of this report included lung lobe torsion and diaphragmatic hernia. However, our imaging and intraoperative findings were inconsistent with either of these conditions. Diaphragmatic hernia was ruled out on the basis of an intact diaphragm on thoracic CT. In dogs, the right middle and left cranial lung lobes have been most commonly reported as affected by torsion.12,13 Radiographic findings in patients with lung lobe torsion include increased lobar opacity and pleural effusion. Of 15 small animal patients (13 dogs and 2 cats) with lung lobe torsion, only 6 had lobar bronchi that appeared irregular, narrow, or blunted on radiographs.12 Thoracic CT has become the imaging modality of choice for evaluation of patients suspected to have this condition; finings include pleural effusion and an abruptly ending bronchus.14 Enlargement, consolidation, emphysema, and absence of contrast medium enhancement are also observed in the affected lung lobe.14 We were able to rule out lung lobe torsion in the dogs of this report by CT; both had caudal lung lobe involvement, and thoracic CT did not reveal any abnormalities of the bronchial tree.
It is possible that a pulmonary bleb or bulla existed in these patients prior to the tornado; a thin-walled bulla would have been more susceptible to rupture during acute barometric pressure changes than would structurally normal lung tissue. However, whereas caudal lung lobes were affected in both of our patients, cranial lung lobes are most commonly affected by pulmonary bulla disease in dogs.15 In 1 study,16 histopathologic changes associated with pulmonary bullae or blebs in 12 dogs included the following: dilated alveoli in tissue surrounding the bleb or bulla, peripheral emphysema, smooth muscle hypertrophy of the respiratory ducts, and mild to moderate lymphoplasmacytic inflammation These changes, particularly smooth muscle hypertrophy, may be chronic in nature. Neither of the dogs described in this report had these changes observed on histologic examination, which cast doubt on the putative presence of a pulmonary bleb or bulla prior to the tornado. Barotrauma related to positive-pressure ventilation has been reported in cats,17,18 and the dogs of the present report had similar clinical signs (including tachypnea and increased respiratory effort) after a tornado. The diagnostic imaging, intraoperative, and histologic findings together with the lack of external wounds on either dog, timing relative to the natural disaster, and the known location of 1 dog during that time, led us to consider this phenomenon. It is our suspicion that a behavioral response of breath-holding may have allowed a substantial change in pulmonary pressures to occur. Otherwise, the condition might have been more widespread. Although possibly a rare condition, clinicians should consider pulmonary barotrauma as a differential diagnosis for a patient with a history of recent involvement in a tornado, respiratory distress, and evidence of pleural effusion, pneumothorax, or both on thoracic radiographs. Outcomes were good for both dogs of this report following surgical intervention.
References
1. Marshall TPMcDonald JRForbes GS. The enhanced Fujita (EF) scale, in Proceedings. 22nd Conf Severe Local Storms 2004. Available at: ams.confex.com/ams/11aram22sls/techprogram/paper_81090.htm. Accessed Jul 1, 2014.
2. Toklu ASErelel MArslan A. Pneumomediastinum or long damage in breath-hold divers from different mechanisms: a report of three cases. Diving Hyperb Med 2013;43:232–235.
3. Guyton APHall JE. Physiology of deep-sea diving and other hyperbaric conditions. In: Textbook of medical physiology. 11th ed. Philadelphia: Elsevier Inc, 2006;537–543.
4. Guyton APHall JE. Aviation, high-altitude, and space physiology. In: Textbook of medical physiology. 11th ed. Philadelphia: Elsevier Inc, 2006;545–550.
5. Yeh DDSchecter WP. Primary blast injuries—an updated concise review. World J Surg 2012;36:966–972.
6. Lee JJSamaras TMYoung CR. Pressure measurements at the ground in an F-4 tornado, in Proceedings. 22nd Conf Severe Local Storms 2004. Available at: ams.confex.com/ams/pdfpapers/81700.pdf. Accessed Jul 1, 2014.
7. Samaras TMLee JJ. Pressure measurements within a large tornado, in Proceedings. 84th Am Meteorolog Soc Ann Meet 2003;1–9.
8. Wurman JSamara S. Comparison of in-situ pressure and DOW doppler winds in a tornado and RHI vertical slices through 4 tornadoes during 1996–2004, in Proceedings. 22nd Conf Severe Local Storms 2004. Available at: www.cswr.org/publications/66-Comp_of_In-Situ_Pressure_and_DOW_Doppler_Winds_in_Tornado_and_RHI_Vert_Slices_through_4_Tornadoes_1996-2004.pdf. Accessed Jan 20, 2016.
9. Winn WPHunyady SJAulich GD. Pressure at the ground in a large tornado. J Geophys Res 1999;104:22067–22082.
10. Rozali ASulaiman AModh Zin B, et al. Pulmonary overinfltion syndrome in an underwater logger. Med J Malaysia 2006;61:496–498.
11. Holzki JKellner M. Life threatening unilateral pulmonary overinflation might be more successfully treated by contralateral selective intubation than by emergency pneumonectomy. Paediatr Anaesth 2003;13:432–437.
12. d'Anjou MATidwell ASHecht S. Radiographic diagnosis of lung lobe torsion. Vet Radiol Ultrasound 2005;46:478–484.
13. Neath PJBrockman DJKing LG. Lung lobe torsion in dogs: 22 cases (1981–1999). J Am Vet Med Assoc 2000;217:1041–1044.
14. Seiler GSchwarz TVignoli M, et al. Computed tomographic features of lung lobe torsion. Vet Radiol Ultrasound 2008;49:504–508.
15. Cohn LA. Pulmonary parenchymal disease. In: Ettinger SJ, ed. Textbook of veterinary internal medicine. 7th ed. St Louis: Saunders, 2010;1096–1119.
16. Lipscomb VJHardie RJDubielzig RR. Spontaneous pneumothorax caused by pulmonary blebs and bullae in 12 dogs. J Am Anim Hosp Assoc 2003;39:435–445.
17. Thomas EKSyring RS. Pneumomediastum in cats: 45 cases (2000–2010). J Vet Emerg Crit Care (San Antonio) 2013;23:429–435.
18. Katayama MOkamura YKatayama R, et al. Presumptive acute lung injury following multiple surgeries in a cat. Can Vet J 2013;54:381–386.
