Criteria for Selection of Cases

Medical records of all dogs with LLT examined at the Ontario Veterinary College from January 1991 through December 2004 were reviewed. Dogs were included if LLT was confirmed surgically or during necropsy.

Procedures

Cases were allocated among 3 groups by breed: Pugs, all other small-breed dogs, and large-breed dogs. The distinction between large- and small-breed dogs was made according to generally accepted breed classifications or a body weight of ≤ 15 kg (33 lb) in mixedbreed dogs.

Data collected included details of signalment (eg, age, sex, and breed), weight, preexisting medical conditions, lung lobe affected, duration of clinical signs, clinical findings, results of CBC and serum biochemical analyses, diagnostic imaging, pleural fluid analyses, intra- and postoperative complications, pathologic findings, and overall outcome. Data for Pugs were compared with data for other dogs.

Statistical analysis—Odds ratios and exact 95% CIs were calculated for prevalence data. The prevalence of LLT in Pugs was compared with prevalence in other breeds by use of the Fisher exact test. Values of P < 0.05 were considered significant.

Results

During the period studied, 44,776 dogs were admitted to the Ontario Veterinary College teaching hospital as new admissions, 274 of which were Pugs. Twenty-three dogs met the criteria for diagnosis of LLT and were included in the study. Ten of those dogs were large-breed dogs (purebred and mixed-breed dogs), and 13 were small-breed dogs. Seven of the 13 small-breed dogs were Pugs. Other small breeds represented included Shih Tzu (n = 1), Miniature Schnauzer (1), Pomeranian (1), Jack Russell Terrier (1), Bichon Frise (1), and mixed breed (1). The large breeds included Afghan Hound (n = 3), mixed breed (4), Malamute (1), Briard (1), and English Setter (1). Pugs with LLT were significantly (OR, 72.9; P < 0.001; CI, 27.9 to 185.2) overrepresented among affected dogs, compared with the general hospital population.

Age of Pugs with LLT ranged from 4.5 months to 4 years (median, 1.5 years; Table 1). Six were male, and of those, 4 were sexually intact and two were castrated. Age of the other 6 small-breed dogs ranged from 1 to 10 years (median, 7 years) and included 3 males (2 castrated) and 3 spayed females. Age of large-breed dogs ranged from 2 to 7 years (median, 3.5 years) and included 6 males (3 castrated) and 4 females (3 spayed).

Six of the 7 Pugs had no medical history of potential predisposing conditions. One had a history of thoracic trauma 2 years prior to diagnosis of LLT. Two of 6 small-breed dogs had preexisting intrathoracic disease (chronic mesothelioma and pneumothorax) at the time of diagnosis. Two of the large-breed dogs had potential predisposing conditions (pulmonary parenchymal disease with megaesophagus and myasthenia gravis, and bronchoalveolar carcinoma).

In Pugs, the duration of clinical signs prior to examination at the hospital ranged from 1 to 9 days (median, 3 days). In the other small- and large-breed dogs, duration of clinical signs ranged from 1 to 21 days (median, 3.5 days) and from 2 to 120 days (median, 4.5 days), respectively.

Clinical signs in Pugs included dyspnea or tachypnea (7 dogs), anorexia and lethargy (5), cough (3), and vomiting (1). Findings on physical examination included pale mucous membranes (3 dogs), signs of pain associated with abdominal palpation (2), and pyrexia (2). Clinical signs in the other small-breed dogs included dyspnea or tachypnea (5 dogs), lethargy (5), anorexia (3), cough (2), and vomiting (1). Physical examination in those dogs revealed pale mucous membranes, signs of pain on abdominal palpation, and pyrexia (2 dogs each). Clinical signs in the large-breed dogs included lethargy (10 dogs), dyspnea or tachypnea (8), cough (7), anorexia (6), and vomiting (3). On physical examination in large-breed dogs, findings included pyrexia (6 dogs), signs of pain on abdominal palpation (2), pale mucous membranes (1), and cyanosis (1).

Thoracic auscultation findings were recorded for 6 Pugs. Five had decreased cardiopulmonary sounds in the left hemithorax, and 1 had decreased cardiopulmonary sounds in the right cranial portion of the thorax. In 4 of 6 other small-breed dogs for which thoracic auscultation findings were recorded, decreased cardiopulmonary sounds were reported in 3 and wheezes were reported in 1. In the 10 large-breed dogs, decreased cardiopulmonary sounds were recorded in 8 dogs, increased pulmonary sounds with harshness were recorded in 1 dog, and no abnormal findings were recorded in 1 dog.

A CBC and serum biochemical analyses were performed in all 7 Pugs. Six Pugs had neutrophilia (range, 16.5 to 31.5 × 10⁹ cells/L; reference range, 2.9 to 10.6 × 10⁹ cells/L), and 1 had mild neutropenia (2.2 × 10⁹ cells/L). Six Pugs had a left shift (range, 0.8 to 5.7 × 10⁹ band neutrophils/L; reference range, 0.0 to 0.3 × 10⁹ band neutrophils/L), and 6 had anemia (Hct, 0.15 to 0.36 L/L; reference range, 0.39 to 0.56 L/L). Abnormalities on serum biochemical analyses included high cholesterol concentration (3 Pugs) and high activities of alkaline phosphatase (4) and creatine kinase (3).

A CBC and serum biochemical analyses were performed in all 6 of the other small-breed dogs. All had neutrophilia (range, 11.9 to 31.4 × 10⁹ cells/L). Three had a left shift (range, 1.92 to 4.37 × 10⁹ band neutrophils/L), and 2 had mild anemia (Hct, 0.34 to 0.35 L/L). High serum activities of alkaline phosphatase (4 dogs), creatine kinase (3), amylase (2), and lipase (2) and low serum concentrations of total protein and albumin (3) were also detected.

Among the large-breed dogs, 8 had neutrophilia (range, 11.4 to 28.8 × 10⁹ cells/L) and 1 had neutropenia (2.2 × 10⁹ cells/L). Three dogs had a left shift (range, 1.45 to 1.83 × 10⁹ band neutrophils/L), and 4 had anemia (Hct, 0.26 to 0.38 L/L). High serum activities of creatine kinase (6 dogs), alkaline phosphatase (4), and amylase (3) as well as high serum concentrations of total bilirubin (5) and low serum concentrations of total protein and albumin (4) were also detected.

Aerobic and anaerobic bacterial culture and antimicrobial susceptibility testing of specimens of abnormal lung tissue were performed in 6 Pugs, and all results were negative. Bacterial culture of pleural fluid was performed in 1 Pug; no growth was detected. Bacterial culture and antimicrobial susceptibility testing of lung tissue, with or without pleural fluid, were performed in 3 of the 6 other small-breed dogs and in 7 of the 10 large-breed dogs. Results of bacterial culture in all small-breed dogs were negative. Escherichia coli and Staphylococcus intermedius were identified on bacterial culture in 2 of the large-breed dogs.

Thoracic radiography was performed in 7 Pugs. Abnormal radio-opacity in the left cranial lung field was observed in 6 Pugs; in 1 of those dogs, the opacity involved the entire left hemithorax, and in another, the cranial lung field was bilaterally consolidated. One Pug had abnormal radio-opacity in the right cranioventral lung field. There was mild pleural effusion in 5 Pugs and air bronchograms in regions of pulmonary consolidation in 2. Thoracic radiography was performed in all other small- and large-breed dogs. A region of abnormal tissue opacity or consolidation was a consistent finding among all dogs. Nine of 10 large-breed dogs and 5 of 6 small-breed dogs had pleural effusion. Deviation or displacement of a bronchus was observed in 2 large-breed dogs, and air bronchograms were reported in 2. Radiographic evidence of gas bubbles in the consolidated tissue was reported in 1 large-breed dog and 1 small-breed dog.

Thoracic ultrasonography was performed in 4 Pugs, revealing a soft tissue mass in the left cranial thorax and pleural effusion in each instance. Partial aeration of the soft tissue mass was noted in the 2 Pugs with air bronchograms. Thoracic ultrasonography was performed in 4 other small-breed dogs and 5 large-breed dogs. In those dogs, tissue consolidation in the region of the affected lung lobe and pleural effusion were consistent findings. Multiple gas bubbles were observed in the consolidated lung lobe in 2 small-breed dogs.

Bronchoscopy was performed in 2 Pugs. Narrowing of the left cranial bronchus with intraluminal serosanguinous fluid was observed in one, and a blind-ended bronchus leading into the right cranial lung lobe was reported in the other. Bronchoscopy was also performed in 1 large-breed dog and revealed a distorted, collapsed right bronchus.

Intercostal thoracotomy and lung lobectomy were performed in all Pugs. No additional thoracic cavity abnormalities were observed intraoperatively. Four of the other small-breed dogs and 9 large-breed dogs underwent intercostal thoracotomy with resection of the affected lung lobe. Two small-breed dogs died preoperatively, and 1 large-breed dog was euthanatized preoperatively.

Of the 23 dogs in this study, 9 had torsion of the right middle lobe, 11 had torsion of the left cranial lobe, 2 had torsion of the right cranial lobe, and 1 had torsion of the right caudal lobe. In the 7 Pugs, 6 had torsion of the left cranial lobe and 1 had torsion of the right cranial lobe. Three of the 6 other small-breed dogs had torsion of the right middle lobe, 2 had torsion of the left cranial lobe, and 1 had torsion of the right cranial lobe. Six of 10 large-breed dogs had torsion of the right middle lobe, 3 had torsion of the left cranial lobe, and 1 had torsion of the right caudal lobe. Five of the 7 resected lobes from Pugs were examined histologically. There were no abnormal findings other than those consistent with LLT (tissue necrosis, hemorrhage, and thrombosis). The lung lobes resected from all 6 small-breed dogs and 8 of the 10 large-breed dogs were also examined histologically. Similar to findings in Pugs, no lesions other than those consistent with LLT were observed.

Six Pugs survived to discharge; 1 could not be stabilized preoperatively and died intraoperatively because of hemorrhage and presumptive DIC. That dog was neutropenic at the time of evaluation. All Pugs that survived had uncomplicated recoveries and were discharged from the hospital 2 to 7 days postoperatively (median, 2 days). None of the Pugs have been examined at the Ontario Veterinary College for recurrence of thoracic disease. Owners of Pugs were contacted by telephone. Two of the 6 surviving Pugs were lost to follow-up; the remaining 4 were doing well and had no signs of disease related to the respiratory tract 1 to 75 months postoperatively.

Three of the 6 other small-breed dogs and 5 of the 10 large-breed dogs survived to discharge. Two smallbreed dogs died preoperatively, and 1 died postoperatively as a result of DIC and renal failure. One of the dogs that died preoperatively had chronic mesothelioma, and the other had traumatic thoracic injury; both died as a result of cardiac arrest. One large-breed dog was euthanatized preoperatively because of the poor prognosis for survival with concurrent myasthenia gravis and neuropathy. One large-breed dog could not be stabilized preoperatively and developed cardiac arrest at the time of anesthetic induction. Another large-breed dog was euthanatized intraoperatively because of metastatic bronchoalveolar carcinoma. Two large-breed dogs were euthanatized postoperatively, 1 because of development of chylothorax and the other because of concurrent sepsis and persistent pleural effusion (the latter was neutropenic at the time of evaluation). Two of the 5 surviving large-breed dogs were Afghan Hounds, and both subsequently developed chylothorax. One Afghan Hound was discharged without treatment because pleural effusion and associated respiratory signs were mild; that dog was reportedly doing well 2 years postoperatively. The second Afghan Hound underwent pericardectomy and thoracic duct ligation procedures, which did not resolve the chylothorax. Six months after the diagnosis of LLT, that dog was euthanatized because of persistent chylothorax and torsion of the right middle lung lobe.

Discussion

Six of the 7 pugs with LLT in this study had no predisposing conditions. One had been hit by a car 2 years earlier and had sustained rib fractures but had no clinical problems prior to examination and diagnosis of LLT. This history was consistent with those in 2 previous case reports^6,7 describing spontaneous LLT in Pugs.

Pugs in our study ranged in age from 4.5 months to 4 years, with a median age of 1.5 years. In previous case reports,^6,7 affected Pugs were 1.5, 2, and 2.5 years of age, suggesting that LLT occurs most commonly in young Pugs. In contrast, the median age of the other small-breed dogs in the study was 7 years and 2 had preexisting thoracic disease. We speculate that LLT in Pugs develops spontaneously at a young age, whereas in other small-breed dogs, LLT appears to arise more frequently secondary to diseases that are acquired with increasing age.

Six of the 7 Pugs with LLT in this study were male. This finding was consistent with those in previous case reports,^6,7 in which 2 Pugs with LLT were male and 1 was female. Other small-breed dogs in this study were evenly distributed by sex (50% male). In large-breed dogs, the condition was approximately evenly distributed between the sexes, with 6 of 10 affected dogs being male. Those findings were similar to findings from a previous study,¹³ in which 13 of 22 affected dogs were male. The reason for the apparent predilection in males is not known.

Six of the Pugs in our study had torsion of the left cranial lung lobe, and 1 had torsion of the right cranial lobe. In previously reported cases,^6,7 all affected Pugs had involvement of the left cranial lung lobe. Interestingly, in one of those dogs,⁷ spontaneous torsion of the right cranial lobe occurred 2 years after torsion and resection of the left cranial lobe. The factors contributing to spontaneous torsion of the left cranial lung lobe in Pugs are not known. Investigators of 1 study⁶ have suggested that bronchial cartilage dysplasia may account for some hilar instability in brachycephalic dog breeds, but in that study, authors acknowledged that if that effect was an important predisposing factor, the prevalence of LLT would be expected to be higher in all brachycephalic breeds. Moreover, the bronchial cartilage of 1 Pug in that report was examined histologically and determined to have no abnormal changes. Further investigation into thoracic conformation in Pugs, including the anatomic features of individual lung lobes and their attachments, is warranted. In large deep-chested dogs, the tendency for torsion of the right middle lobe is thought to be associated with the shape and anatomy of the lobe.^4,5,11 Results of our study were consistent with those from another study,¹³ in which the right middle lobe was cited as the most commonly affected lobe in large-breed dogs, followed by the left cranial lobe.

In our study, the clinical appearance of Pugs with LLT was similar to that of large- and small-breed dogs, and that observation corroborates those from other reports.^1,4–7,13 Dyspnea, tachypnea, lethargy, and anorexia were commonly observed, whereas coughing, pale mucous membranes, pyrexia, and vomiting were variably observed. Coughing was a common clinical sign in large-breed dogs in our study, whereas in another retrospective study,¹³ coughing was reported in only 4 of 22 dogs. In our study, signs of pain on palpation of the abdomen were detected in 2 of the 7 Pugs, 2 of the other 6 small-breed dogs, and 2 of the 10 large-breed dogs; thus, signs of abdominal pain should not preclude a diagnosis of thoracic disease. The reason for abdominal pain in some dogs with LLT is not known.

A radiographic pattern of consolidation in the cranial portion of either hemithorax along with mild pleural effusion was a consistent finding. All other small- and large-breed dogs also had radiographic findings of a consolidated lung lobe and pleural effusion. Two Pugs, 1 small-breed dog, and 1 large-breed dog had air bronchograms in the consolidated region of lung. With acute LLT, air bronchograms are initially observed but disappear as the bronchi fill with fluid and blood.^4,5 In our study, thoracic radiography was a reliable diagnostic tool. Adjunct imaging modalities such as thoracic ultrasonography and bronchoscopy were less frequently used but were valuable in confirming the diagnosis of LLT. This is consistent with previously published reports.^4–9,13

Pleural effusion is a common clinical finding in dogs with LLT. Whether there is a cause-and-effect relationship between effusion and LLT has been debated.^1,4,5 Pleural effusion may result from venous congestion and inflammation associated with the affected lobe, or it may be a preexisting condition that facilitates torsion. The character of the pleural effusion that occurs in conjunction with LLT may range from serosanguinous to sanguinous to chylous.^4,5 An association between chylothorax and LLT has been reported.¹³

In 1 study¹³ of dogs with LLT, 59% had neutrophilia, whereas only 23% had a left shift. In our study, 6 of 7 Pugs, 6 of 6 other small-breed dogs, and 8 of 10 largebreed dogs had neutrophilia. In contrast, the Pug that died intraoperatively and 1 large-breed dog that was euthanatized postoperatively because of sepsis were neutropenic. It is not possible to draw conclusions given that only 2 dogs in our study had neutropenia; however, neutropenia may be associated with a poor outcome. Additionally, 6 of 7 Pugs, 3 of 6 other small-breed dogs, and 3 of 10 large-breed dogs had a left shift. Overall, neutrophilia was a more consistent finding in this study than has previously been reported, and a left shift appears to be a more consistent clinicopathologic feature of LLT in Pugs than in other breeds.¹³ The importance of these findings is not known.

In 1 study,¹³ the prognosis for dogs with LLT was described as fair to guarded. In that study,¹³ the investigators reported a survival rate of 50% in a retrospective evaluation of 22 dogs with LLT, a value that was consistent with the survival rate of 48% calculated from analysis of 25 published case reports of LLT. The overall survival rate among dogs in our study was 61% (14/23), with survival defined as the dog having been discharged from the hospital. Excluding Pugs, the survival rate was 50% (8/16), a value that is consistent with that in the former study.¹³ Interestingly, 5 of those 22 dogs in the former study were small-breed dogs and 2 were Pugs; both Pugs had a favorable outcome. Similarly, 6 of 7 Pugs in our study and all 3 Pugs in other case reports^6,7 had a favorable outcome. Evaluation of all reported cases of LLT in Pugs (including the 7 in our study) revealed a favorable outcome in 11 of 12 dogs.^6,7,13 Breed has been reported¹³ to be a preoperative predictor of a negative outcome for Afghan Hounds and Borzois. In our study, only 1 of 3 Afghan Hounds was alive 6 months after surgery. In contrast, breed appears to be positively associated with survival in Pugs.

Although LLT has been considered a disease that affects large deep-chested dogs, our results suggested that Pugs may be predisposed to LLT. Furthermore, LLT in Pugs appeared to be distinct from the condition as it develops in other breeds in terms of signalment, lung lobe affected, etiology, and prognosis. The condition in Pugs was predominantly spontaneous and most often affected the left cranial lung lobe. Lung lobe torsion in Pugs occurred most frequently in young dogs ≤ 4 years old and affected males more commonly than females. When surgical resection of the affected lobe was performed in a stable patient, the prognosis for survival in Pugs was excellent.