Bilateral lung lobe torsions in a cat

Abstract

A 13-year-old spayed female domestic longhair cat was presented for tachypnea and was identified to have reduced lung sounds over the left hemithorax. Thoracic ultrasound examination and computed tomography identified changes consistent with bilateral lung lobe torsion. A median sternotomy confirmed torsion of both the cranial portion of the left cranial lung lobe and the right middle lung lobe. The affected lobes were resected. Pleural fluid analysis was indicative of a modified transudate and histopathology was consistent with a subacute to chronic torsion with no evidence of neoplasia or infection. The patient recovered without complication. Lung lobe torsion is an uncommon presentation across all species and is especially rare in cats. To the authors’ knowledge, bilateral lung lobe torsion has not been previously reported in small animals.

Lung lobe torsion (LLT) is rotation of a lung lobe on its longitudinal axis, typically occurring at the level of the hilus. Lung lobe torsion leads to occlusion of the airways, lymphatics, veins, and arteries, causing secondary pulmonary congestion, edema, and necrosis. Lung lobe torsion is considered infrequent in dogs and rare in cats, and is possibly fatal without surgical intervention (1). Bilateral lung lobe torsion has not previously been reported in small animals. This article describes the presentation and management of a cat with bilateral LLTs.

Case description

A 4 kg, 13-year-old, spayed female domestic longhair cat was evaluated by her primary care veterinarian because of a 3-day history of lethargy, poor appetite, and rapid, shallow breathing noted by the owners. The patient was an indoor-only cat with an unremarkable prior medical history. Rectal temperature was normal (39.5°C), heart rate was within reference intervals (180 beats/min), and respiratory rate was elevated (72 breaths/min). The cat was lethargic with rapid, shallow breathing. A complete blood (cell) count (CBC) revealed a non-regenerative anemia [hematocrit 0.22 L/L, reference interval (RI): 0.30 to 0.45 L/L] and thrombocytopenia (104 × 10⁹/L; RI: 175 to 600 × 10⁹/L) with platelet aggregates detected. Serum biochemistry showed a low blood urea nitrogen (BUN, 4.6 mmol/L, RI: 5.7 to 12.9 mmol/L), low alkaline phosphatase (12 U/L, RI: 14 to 11 U/L), and mild hyperglycemia (9.36 mmol/L, RI: 3.95 to 8.84 mmol/L). A urinalysis revealed a urine specific gravity of 1.050 with 1+ proteinuria. Abdominal radiographs showed cystic calculi with no other abnormalities. The patient was administered 0.9% sodium chloride, 50 mL/kg body weight (BW), SC, maropitant (Cerenia, Zoetis, Parsippany, New Jersey, USA), 1 mg/kg BW, SC, and buprenorphine (Vetergesic, Champion Alstoe Animal Health, Sheriff Hutton, York, UK), 0.02 mg/kg BW, buccal, q8h, and discharged. The owner reported no response to treatment and the patient was referred 24 h later to an emergency and referral hospital for further work-up and care.

On presentation, physical examination identified mild dehydration, an elevated respiratory rate (40 breaths/min) with normal respiratory effort and mildly reduced lung sounds over the left hemithorax. Other vital parameters were within normal limits and the urinary bladder was noted to be mid-sized and easily compressible. Blood was drawn for a venous blood gas and minimum database which revealed anemia (PCV 25%), normal total solids (76 g/L), low BUN (5 mmol/L, RI: 5.36 to 12.14 mmol/L), a mild hyperkalemia (4.7 mmol/L, RI: 2.9 to 4.2 mmol/L), and a urine specific gravity of 1.050. A CBC revealed a mild eosinophilia (1.5 × 10⁹/L, RI: 0 to 1 × 10⁹/L), and thrombocytopenia (69 × 10⁹/L, RI: 300 to 800 × 10⁹/L) with no other abnormalities. Point-of-care thoracic ultrasound examination identified low volume bilateral pleural effusion. Abdominal bedside ultrasound examination was negative for free peritoneal fluid. An intravenous catheter was placed, and the patient was sedated with butorphanol (Torbugesic, Zoetis, Parsippany, New Jersey, USA), 0.3 mg/kg BW. Diagnostic thoracentesis yielded 8 mL of serosanguinous, turbid fluid. Pleural fluid analysis was consistent with a modified transudate with no microorganisms or neoplastic cells seen. Fluid protein content was 420 g/L, white blood cell count was 3580 × 10⁶/L, and red blood cell count was 460 000 × 10⁶/L. Thoracic radiographs (Figure 1), taken following thoracentesis, revealed consolidation of the cranial segment of the left cranial lung lobe, consolidation of the right middle lung lobe, and a low-volume pleural effusion. The patient was admitted to the hospital and managed supportively in the ICU for 72 h while awaiting pleural fluid cytology results, per owner preference. The patient remained eupneic with persistent low-volume pleural effusion on daily point-of-care thoracic ultrasound examination. A grade I/VI parasternal systolic heart murmur was noted during repeat physical examination. The patient was treated with buffered, isotonic, crystalloid fluids (2 mL/kg BW per hour, IV) and buprenorphine (0.03 mg/kg BW, IV, q8h).

Figure 1.

Ventrodorsal (A) and right lateral (B) thoracic radiographs taken following thoracocentesis demonstrating consolidation of the cranial segment of the left lung lobe and the right middle lung lobe. The cardiac silhouette is shifted to the right and there is a small amount of remaining pleural effusion.

On the third day in hospital, once the pleural fluid cytology results were available, the owner consented to thoracic ultrasound examination. Thoracic ultrasound revealed a swollen (up to 3 cm in diameter), consolidated lung lobe in the right hemithorax with a homogenous echotexture and overall hepatized appearance (Figure 2). Color-flow doppler showed lack of flow within the lobe, suggestive of a right LLT. The left cranial lung lobe was partially consolidated with a moderate volume of gas trapped within the lobe. Color-flow doppler identified flow within the lobe, suggestive of a partial or early LLT. Computed tomography (CT) scan was recommended to further evaluate the lungs (Figure 3) and to help guide surgical approach (lateral thoracotomy versus median sternotomy). General anesthesia was induced with hydromorphone (Hydromorphone; Sandoz, Princeton, New Jersey, USA), 0.1 mg/kg BW, IV, dexmedetomidine (Dexdomitor, Zoetis, Parsippany, New Jersey, USA), 3 μg/kg BW, IV, and alfaxalone (Alfaxan, Jurox, Brampton, Ontario), 1 mg/kg BW, IV. Following orotracheal intubation, the patient was maintained on fentanyl (Fentanyl, Sandoz), 5 μg/kg BW per hour, CRI, ketamine (Narketan, Vétoquinol, Fort Worth, Texas, USA), 0.5 mg/kg BW per hour, CRI, and isoflurane inhalant (Forane; Baxter, Deerfield, Illinois, USA), in 100% oxygen. During anesthesia, electrocardiography, pulse oximetry, esophageal temperature, and oscillometric noninvasive blood pressure were continuously recorded using a multiparameter monitor. Capnography and monitoring of the anesthetic agent were performed using a mainstream analyzer.

Figure 2.

Thoracic ultrasound image of the right middle lung lobe demonstrating a hepatized appearance. No color doppler flow was detected within the lobe.

Figure 3.

Coronal CT image demonstrating a completely consolidated and swollen right middle lung lobe, with the lobar bronchus discontinued at the hilus. There is a resultant leftward mediastinal shift. The left cranial lung lobe is almost completely consolidated with a vesicular pattern, with the associated bronchus ending bluntly at the hilus. There is persistent arterial supply and occluded venous supply from the left cranial lung lobe. The caudal portion of the left cranial lobe is reduced in size and partially infiltrated with soft tissue attenuation, with a normally positioned lobar bronchus.

Exploratory thoracotomy via median sternotomy was pursued and confirmed complete torsion of both the cranial portion of the left cranial lung lobe and the right middle lung lobe. The cranial portion of the left cranial lobe and right middle lobe were completely consolidated, swollen, and dark purple causing a leftward mediastinal shift. Each torsed lobe was stapled at the hilus using a thoracoabdominal stapler (TA-30 V3; Medtronic of Canada, Brampton, Ontario) then transected and removed. Saline was placed in the thoracic cavity to evaluate for any potential leakage from the resection sites and none was noted. Bilateral 12 Fr trocar chest tubes were placed, and each was secured using 2-0 Ethilon (Ethicon, Somerville, New Jersey, USA) in a Chinese fingertrap. The thorax was copiously lavaged with 0.9% sodium chloride before closure. The sternebrae were wired together using 24-gauge orthopedic wire. Muscles and fascia were apposed with 3-0 polydioxanone (Ethicon). A subcutaneous wound diffusion catheter was placed into the incision exiting caudally. Then subcutaneous tissues and skin were apposed routinely. A 14 Fr esophagostomy feeding tube was placed into the left side of the neck to facilitate post-operative nutrition as the patient had been anorexic for approximately 7 d.

Histopathology of the lung lobes was supportive of subacute to chronic torsion with no evidence of neoplasia or infectious organisms. Following surgery, the patient was maintained in the ICU on buffered, isotonic, crystalloid fluids (2 mL/kg BW per hour, IV), fentanyl (3 μg/kg BW per hour, IV, CRI), bupivacaine (q6h via diffusion catheter), bolus feedings via esophagostomy tube, and chest tube maintenance. The chest tubes were removed 3 d after surgery, and the patient was discharged from hospital 5 d after surgery. There were no complications during the patient’s recovery. At a 3-week and a 7-month follow-up, the patient was doing well with no concerns. Thoracic radiographs performed 7 mo after surgery revealed that 1 of the wires around the sternebrae had failed but the patient was experiencing no clinical signs. This was discussed with the owner and removal was declined. The pulmonary fields were normal and well-inflated, and the staple lines were visible within the thoracic cavity.

Discussion

There are 18 cases of feline LLT reported in the literature (2–12). The cats described range in age from 10 mo to 19 y with a median age of 10 y. Domestic shorthairs account for 11 of these cases, with the remaining cases represented by 2 Persians, and 1 each of domestic longhair, Minskin, Chinchilla, Siamese, and Himalayan. There is not a clear sex predilection. The most commonly reported clinical signs associated with lung lobe torsion are inappetence, lethargy, tachypnea, and dyspnea. Hemoptysis and epistaxis have also been reported (12). Harsh or dull lung sounds may be noted on examination. Nine of these cases had preexisting or concurrent thoracic disease including chylothorax (3,7,9), pyothorax (9), fibrosing pleuritis (2), mediastinal lymphoma (2,9), pulmonary carcinoma (6), cardiac disease (8), chronic traumatic diaphragmatic hernia (10), peritoneopericardial diaphragmatic hernia (7), and chronic asthma (11). In these cases, it is assumed that the preexisting thoracic disease created an environment suitable for LLT to develop. Alternatively, in some circumstances, the thoracic disease may be concurrent and secondary to a spontaneous torsion, e.g., pyothorax, chylothorax, or fibrosing pleuritis. Pleural effusion will likely develop following the LLT due to pulmonary congestion and necrosis; however, pleural effusion is also potentially a risk factor for LLT. The cat herein had no preexisting thoracic disease to explain her LLTs. The low volume pleural effusion identified on presentation was therefore suspected to be secondary to an existing congested lung lobe.

Lung lobe torsion has been reported numerous times in dogs. Although a wide variety of breeds are described in the literature, Afghan hounds and pugs are overrepresented, accounting for up to 23% of cases and 22% to 38% of cases, respectively (6,13–16). The higher incidence of LLT seen in both deep, narrow-chested dogs as well as pugs suggests an anatomic or genetic predisposition. Dogs with LLT may present with an acute, chronic, or an acute-on-chronic history. Clinical signs are similar to those of cats, including lethargy, anorexia, weight loss, dyspnea, tachypnea, cough, vomiting, exercise intolerance, hemoptysis, and collapse. Physical examination findings may include decreased or harsh lung sounds, crackles, or wheezes and fever. Similar to cats, an underlying cause of LLT is identified in many dogs, but some cases remain idiopathic or spontaneous. Documented causes of LLT include pleural effusion, neoplasia, chronic bronchopneumonia, trauma, congenital peritoneopericardial diaphragmatic hernia, and traumatic diaphragmatic hernia (6,14,17). Recurrence of LLT has been described in several dogs, occurring from days to up to 2 y following lung lobectomy (14,18,19).

Diagnosis of LLT may be aided by thoracic radiographs, thoracic ultrasound, CT, and/or bronchoscopy. A review of the radiographic diagnosis of 13 dogs and 2 cats with LLT identified that all cases had pleural effusion and increased lobar opacity on radiographs. Air bronchograms may be visible in the early stages of LLT; however, over a few days the air will be resorbed and replaced by fluid. Although less frequently observed, radiographic identification of a lobar vesicular gas pattern and/or a displaced or proximally narrowed bronchus are the most specific radiographic findings for LLT (5). It is worth noting that LLT has been described in dogs that did not have pleural effusion, thus the absence of pleural effusion should not rule out LLT (6). A study comparing radiographic versus CT findings of LLT in dogs and cats concluded that CT findings are similar to those of radiographic findings, but the diagnosis is easier and more commonly reached with the use of CT (4). Bronchoscopy may identify folding or pinching of a bronchus and inability to advance the scope into the affected lung (6). Thoracic ultrasound findings will be dependent on the chronicity of the torsion, and thus may or may not be diagnostic. The affected lobe may be hyperechoic and filled with gas (consistent with a vesicular gas pattern on radiographs), have rounded borders, have lack of blood flow on color doppler, or may be atelectatic, potentially giving a hepatized appearance to the lobe (5). Definitive diagnosis of LLT is ultimately made through thoracoscopy or exploratory thoracotomy.

Pleural fluid analysis is generally non-diagnostic of LLT as it may only be indicative of underlying intrathoracic disease or be otherwise non-specific. In both canine and feline patients, modified transudates, transudates, hemorrhagic, septic, chylous, and neoplastic effusions have been described. Analysis of pleural fluid of the patient described in this report indicated a modified transudate consisting of mixed inflammatory cells with no microorganisms or neoplastic cells. Given that small lymphocytes were not the predominant cell type, preexisting chylothorax was not prioritized in this case; however, contemporaneous measurement of pleural effusion and peripheral blood triglyceride content could have been considered to rule this out. Bacterial culture is positive in 6% to 60% of cases (13,15). Complete blood (cell) count and biochemistry findings are also non-specific and highly variable.

Certain lung lobes appear to be more commonly affected than others. In cats, the right middle lung lobe is the most commonly affected (7/19 cases); however, the left cranial, left caudal, left middle, and right cranial are also reported. In dogs, the right middle (33% to 46% of cases) and the left cranial (42% to 44% of cases) are the most commonly torsed lobes (6,14,15). There is a handful of cases describing concurrent right cranial and right middle lung lobe torsion in dogs, but bilateral LLT has not been described (5,14,20,21). Interestingly, torsion of the right middle lung lobe is the most common torsion in large, deep-chested dogs, whereas most pugs experience left cranial lung lobe torsion (6,14). These lung lobe predispositions may be due to the fact that the right middle lung lobe is relatively narrow with weak attachments to the chest wall and cranial lobe, and similarly, the cranial lung lobes in general lack supportive pulmonary ligaments. Typically, when the left cranial lobe torses, both the cranial and caudal segments will be involved as they share a primary bronchus.

Treatment of patients with LLT involves initial stabilization of the patient, including supplemental oxygen and likely thoracentesis, followed by thoracotomy for lung lobe resection. Thoracostomy tubes are typically placed after surgery. Detorsion of the affected lung lobe before removal is avoided due to concern for possible reperfusion injury (1).

Prognosis following lung lobectomy for treatment of a LLT in all species appears to be good. All treated cats reported in the veterinary literature survived to follow-up. One cat died before surgery due to a tension pneumothorax following thoracocentesis (6). One cat had a generalized seizure after surgery (11), while another had recurrent postoperative chylothorax that required repeated thoracotomy for thoracic duct ligation 7 d after surgery (7). Although older studies suggested higher mortality rates, more recent larger retrospective studies in dogs indicate 86% to 92% of dogs survived to discharge (6,14,15). Post-operative complications described in dogs include aspiration pneumonia, pneumothorax, development of chylothorax, persistent non-chylous pleural effusion, hemorrhage, complications associated with a pre-existing condition, or recurrence of lung lobe torsion (6,14,15). In Park et al (14), dogs with concurrent torsion of the right cranial and middle lung lobes were less likely to survive than those with torsion of the left cranial lung lobe; however, this was demonstrated in only a small number of cases. A recent retrospective study investigating risk factors for mortality identified that dogs with a depressed mentation at presentation were 21 times more likely to die than dogs with a normal mentation (15). No other risk factors for mortality have been identified.

Lung lobe torsion is also uncommon in humans, with an incidence of 0.089% to 0.4% and a mortality rate of 8.3%. It is typically seen as a postoperative complication after lung resection, lung transplantation, or other thoracic surgery (62.4% of cases). It is suspected that torsion results from increased perilobar space, as well as possible postoperative atelectasis and pleural effusion. Spontaneous and post-trauma are the next most common causes of LLT in humans. The right middle lung lobe is the most likely to torse in post-operative patients, and the right upper lobe is the most likely to spontaneously torse. Similar to veterinary recommendations, treatment of LLT in humans is typically resection of the affected lobe without detorsion. However, in some cases, if the lung lobe still appears viable, it will instead be repositioned (33% of cases). A variety of techniques may be used to reduce the risk of future repeated torsion, including suturing of lung to adjacent pericardial fat or pleura, suturing to adjacent lung lobe(s), or using adhesive. Repositioning of the affected lung lobe rather than resection allows for retention of pulmonary function. However, risks of repositioning include the possibility for venous embolism and reperfusion injury. Generally, the decision to reposition rather than resect is made based on whether or not the lung lobe appears infarcted (markedly swollen, atelectatic, hemorrhagic) and the delay in time from torsion to surgery. No survival benefit has been associated with either resection or repositioning. Occasionally in humans, the affected lung lobe will be detorsed and then resected to facilitate ease of surgery. There is concern for reperfusion injury and embolism with this approach (22).

Lung lobe torsions are rarely reported in cats. To the authors’ knowledge, this is the first reported case of bilateral LLT in any veterinary species.