Presumptive acute lung injury following multiple surgeries in a cat

Abstract

A 12-year-old, 3.5-kg spayed female domestic shorthair cat had a tracheal mass identified as malignant B-cell lymphoma. The cat had tracheal resection and subsequently developed laryngeal paralysis. Due to multiple episodes of respiratory distress the cat subsequently had tracheal surgeries. Finally, the cat had a sudden onset of severe respiratory distress and collapsed. Computed tomography imaging and arterial blood gas analysis supported a diagnosis of acute lung injury.

Acute lung injury (ALI) and acute respiratory distress syndrome (ARDS) are syndromes of acute respiratory failure that lead to bilateral diffuse lung inflammation and edema and result from various clinical states. The major difference between ALI and ARDS is the degree of hypoxemia. In humans, ALI/ARDS has been associated with various risk factors (1) and an effective therapy has not been established. The American-European Consensus Conference (AECC) defined simple criteria for clinical diagnosis of ALI/ARDS in humans in 1992 (2). Although Wilkins et al (3) advocated similar criteria for use in veterinary medicine in 2007, ALI/ARDS is currently a poorly understood condition and may be underdiagnosed. Walker et al (4) reported a dog with ALI caused by massive bee envenomation, which was diagnosed mainly by computed tomography (CT) images and pulmonary function tests. To our knowledge, there are no reports describing ALI/ARDS in a feline patient. In this report, we present a cat presumptively diagnosed with ALI following multiple surgeries.

Case description

A 12-year-old, 3.5-kg spayed female domestic shorthair cat was presented to the Iwate University Veterinary Teaching Hospital with sudden onset of labored open-mouth breathing, cyanosis, and stridor. Physical examination revealed labored breathing associated with an inspiratory noise, which was loudest over the cervical trachea. The heart rate was 140 beats/min, and the respiratory rate was estimated at 70 breaths/min. The capillary refill time was < 1 s and the mucous membrane was pink. No heart abnormalities, such as cardiac murmur or arrhythmia, were detected on auscultation. Complete blood (cell) count, electrolytes, and biochemical profile showed no abnormalities. Serological tests for feline leukemia virus antigen and feline immunodeficiency virus antibodies were both negative. Thoracic radiographs revealed an intratracheal mass at the thoracic inlet; no abnormalities were found in the rest of the thoracic structures (Figure 1). Because the cat was worsening even in an oxygen cage [fraction of inspired oxygen (FiO₂) was estimated at approximately 0.4], surgical removal of the affected trachea (10 rings) was performed on day 2. At the surgery, cefmetazole (Cefmetazon; Daiichi Sankyo, Tokyo, Japan), 25 mg/kg body weight (BW), IV, q8h, was started. Prednisolone (Prednisolone injection solution KS^®; Kyoritsu, Tokyo, Japan), 1 mg/kg BW, SC, q24h, was started to reduce tracheal inflammation. Butorphanol (Vetorphale; Meiji Seika, Tokyo, Japan), 0.4 mg/kg BW, IV, q6h, was administered for analgesia. The cat was also started on a balanced electrolyte solution (lactated Ringer’s solution; 5–10 mL/kg/h). Histopathology on the mass revealed a malignant B-cell lymphoma.

Figure 1.

Lateral (A) and dorsoventral (B) thoracic radiographs obtained before surgeries. There is no evidence of left-sided heart failure. There is an intra-tracheal mass at the thoracic inlet (arrows).

After extubation, the cat again showed labored breathing upon inspiration. Radiographs showed no stenosis at the tracheal anastomosis site. With a tentative diagnosis of postsurgical irritation of the trachea, the cat was supported with supplemental oxygen. On day 3, the cat deteriorated and was induced with propofol (Rapinovet; Takeda Schering-Plough, Tokyo, Japan), 13 mg/kg BW, IV, intubated, and maintained with 100% oxygen and sevoflurane. At intubation, the movement of the arytenoid cartilages was evaluated because laryngeal paralysis was suspected. Laryngeal paralysis was confirmed by direct observation of impaired abduction of both arytenoid cartilages during multiple respiratory cycles. Nerve damage inflicted by tracheal resection was suspected as the cause of this laryngeal paralysis. On day 4, the cat underwent temporary tracheostomy with a 3-mm ID tracheostomy tube modified by the authors to maintain the airway. A percutaneous endoscopic gastrostomy (PEG) tube was simultaneously placed for nutritional support. The tracheostomy tube was fashioned from a standard 3-mm ID endotracheal tube by first removing the connector, then making 2 longitudinal cuts down the length of the tube approximately 180° apart. These cuts were extended about one-third of the way along the tube from the end of the connector to the point where the cuff-inflator mechanism attaches to the tube. The connector was then firmly re-inserted between the 2 wings that had been created. The cat was managed in an oxygen cage with nebulization of tyloxapol inhalation solution (Alevaire; Alfresa, Osaka, Japan). Mucous secretions were removed by suction through the tracheostomy tube every 3 to 5 h to prevent occlusion of the tube. The cat voluntarily started to eat a small amount of food. The cat was administered a maintenance fluid (Soldem 3A; Terumo, Tokyo, Japan) at 3 mL/kg BW/h in addition to nutritional support (a/d; Hill’s-Colgate, Tokyo, Japan) through the PEG tube. On day 8, the cat was depressed and showed labored breathing, vomiting, and sialorrhea. Tracheostomy tube occlusion was confirmed by difficult insertion of the suction tube. Ileus was suspected based on abdominal palpation and abdominal radiographs. Then left-sided arytenoid cartilage lateralization was done and the tracheostomy tube was removed. An exploratory laparotomy was simultaneously performed, and a hairball obstructing the ascending colon was removed by colotomy.

Over the next 3 d the cat was clinically stable. The cat voluntarily started to eat a small amount of food. During this period, the maintenance fluid (Soldem 3A) at 3 mL/kg BW/h and the nutritional support (a/d; Hill’s Colgate) through the PEG tube were continued. On day 12, the cat had a sudden onset of severe respiratory distress and sialorrhea and finally collapsed. The cat was intubated and managed by machine ventilation with 100% oxygen and sevoflurane. Computed tomography (CT) images identified diffuse, bilateral, homogenous opacification and consolidation in all lung lobes (Figure 2). Hounsfield unit (HU) measurements ranged from −10 to −700. These values corresponded to the visibly higher and relatively lower attenuating regions of the lung, respectively. Although the increased opacity was slightly more apparent in the ventral aspect of the lung, there was no evidence of consolidation suggesting aspiration pneumonia. Enlargement of pulmonary vessels was not observed. Although an arteriogram was not performed, pulmonary embolism was considered unlikely based on the distribution of lung changes. Methylprednisolone (Solu-medrol; Pfizer, Tokyo, Japan), 15 mg/kg BW, IV, q12h, and furosemide (Lasix; Sanofi-Aventis, Tokyo, Japan), 2 mg/kg BW, IV, q6h, were administered. The cat was given a balanced electrolyte solution (lactated Ringer’s solution; 5 to 10 mL/kg BW/h) to maintain the urine output at a minimum of 1 to 2 mL/kg BW/h. To keep SpO₂ at > 95%, we used 100% oxygenation by machine ventilation with positive end-expiratory pressure (PEEP, 8 cm H₂O), respiratory rate (12/min), tidal volume (20 mL/kg BW) and plateau pressure titrated at 25 cm H₂O.

Figure 2.

Computed tomography images on day 12. A) Axial image at the level of the 10th thoracic vertebra. B) Coronal image at the middle of the thoracic cavity. C) Sagittal image of the right lung lobes. D) Sagittal image of the left lung lobes. There was severe diffuse, bilateral opacification and consolidation in all lung lobes, and air bronchograms were present. Bronchovascular margins were obscured. Hounsfield unit measurements ranged from −10 in the visibly higher attenuating regions to −700 in the relatively low attenuating regions. These observations were consistent with acute lung injury/acute respiratory distress syndrome.

On day 13, under the mechanical ventilation settings used, arterial blood gas analysis at a FiO₂ of 1.0 showed partial pressure of oxygen in arterial blood (PaO₂), 208 mmHg; partial pressure of carbon dioxide in arterial blood (PaCO₂), 55 mmHg; pH, 7.48; HCO₃, 37.3 mmol/L and base excess (BE), 12.5 mmol/L, indicating impaired oxygenation and metabolic derangement. Although the exact cause of this metabolic alkalosis was not identified, furosemide might have partially contributed. The ratio of PaO₂ to FiO₂ was 208. Based on the combination of the acute onset of respiratory distress, presence of bilateral pulmonary infiltrates on CT images, no clinical evidence of cardiogenic pulmonary edema, suspected risk factors caused by multiple surgeries, and decreased PaO₂:FiO₂ ratio, ALI was clinically diagnosed. The blood examination showed no abnormalities except hypoalbuminemia (24 g/L) and hypo-proteinemia (51 g/L). More than 24 h following intubation, the cat still could not be withdrawn from the ventilator, and its pulmonary status on CT images had not improved. Euthanasia was finally elected by the owner on day 13 because of the poor prognosis. Histopathological findings in the affected lungs were not obtained because the owner declined a necropsy.

Discussion

A syndrome similar to ALI/ARDS in humans has been clinically recognized in small animals. Wilkins et al (3) recently reported criteria in veterinary medicine similar to the AECC criteria in humans for identification of ALI/ARDS. They include an acute onset of respiratory distress, presence of bilateral pulmonary infiltrates on thoracic radiographs, a pulmonary capillary wedge pressure < 18 mmHg or absence of clinical evidence of cardiogenic pulmonary edema, presence of risk factors, and a decreased PaO₂:FiO₂ ratio. The normal PaO₂:FiO₂ ratio is > 500, while a value of < 300 is compatible with ALI and a value of < 200 is compatible with ARDS (2). Because a nonspecific indicator such as PaO₂:FiO₂ is mainly used in these criteria, it may be considered that diagnostic sensitivity will be enhanced while specificity will be reduced. To confirm ALI/ARDS with precision, biopsy of the diseased lung should be considered as the most reliable method. In humans, it was reported that prognosis was improved by biopsy of the lung (5). In most canine patients, ALI/ARDS have only been diagnosed by histopathology following necropsy (6,7). However, lung biopsy may be invasive and has a relatively high complication rate. Furthermore, an effective treatment protocol has not been established in veterinary medicine. Histopathological findings of lung biopsies may not necessarily lead to successful treatment of ALI/ARDS in veterinary patients. Therefore, it is considered that lung biopsy should not become an appropriate diagnostic tool for ALI and ARDS in small animal patients. Bronchoalveolar lavage fluid (BALF) analysis may be another diagnostic tool for ALI/ARDS (8). In patients with ALI/ARDS, BALF analysis may indicate increased protein concentration and the presence of suppurative inflammation (9–11). In our case, we did not use this technique because this procedure may worsen the respiratory condition and induce further progression of hypoxemia. As an alternative, cytology and/or cultures of the sputum may support the diagnosis. Currently, definitive diagnostic criteria for ALI/ARDS in veterinary medicine do not include cytological evaluation of BALF/transtracheal wash samples, which is optional (3). Moss et al (12) reported that the simple AECC criteria in humans have both high sensitivity and high specificity when adapted to patients with a determined risk factor. Therefore, criteria that do not contain histopathological and/or cytological evaluations of the lung may be useful tools for clinical diagnosis of ALI/ARDS in veterinary medicine when used appropriately.

The criteria of ALI/ARDS include the exception of the cardiogenic pulmonary edema confirmed by a pulmonary artery pressure of < 18 mmHg and/or no clinical evidence of left heart failure. In small animal medicine, it is rare to perform pulmonary artery catheterization, and it is not practical to measure the pulmonary artery pressure. In our case, we estimated the cardiopulmonary condition by clinical evidence such as the lack of a cardiac murmur, enlargement of pulmonary vessels on CT images, distension of the jugular vein, and left-sided heart enlargement. Ideally, echocardiography may also be performed to support this clinical evidence.

Common risk factors for ALI/ARDS in humans include direct injuries such as smoke inhalation, aspiration pneumonia, and near drowning, and indirect injuries such as sepsis, anaphylaxis, poly-trauma, and multiple transfusions (1). To our knowledge, ALI/ARDS has only been reported in dogs in small animal practice except in experimental models (4,6,7). In dogs, microbial pneumonia, sepsis, aspiration pneumonia, and shock were the most common predisposing factors, and almost 80% of dogs with ALI/ARDS had multiple factors (6). Aspiration pneumonia may be considered as one of the predisposing factors of ALI/ARDS in cats as well as in humans and dogs. Cytological evaluation and bacterial culture of tracheal secretions were not performed in this study. Therefore, aspiration pneumonia may be included as a risk factor in our case. Oxygen toxicosis via ventilation with 100% oxygen may be a contributing factor to ALI/ARDS in dogs (6). In canine lungs, functional and physical changes develop within 12 to 24 h of breathing 100% oxygen (13–15). Exposure to high concentrations of oxygen causes lung damage as a result of release of oxygen free radicals (16). In cats as well as dogs, 100% oxygen exposure for 24 h may cause oxygen toxicity. Hypoalbuminemia may be a risk factor for ALI/ARDS. Parent et al (6) reported that 67% of canine patients with ARDS showed hypoalbuminemia (mean 21 g/L). However, the cat in our case did not show severe hypoalbuminemia (24 g/L). In our case, respiratory distress caused by laryngeal paralysis, tracheostomy tube occlusion, and tracheal occlusion were considered as the primary causes of ALI. Laryngeal obstruction or strangulation is a predisposing factor to development of pulmonary lesions (6). Acute upper airway obstruction is associated with postobstructive pulmonary edema (POPE) caused by a sudden increase in negative intrapleural pressure and possibly increased capillary permeability (17,18). It was reported that POPE could induce ALI/ARDS in an orangutan (19). Young athletic men, who have enough chest wall musculature to generate extremely high negative inspiratory pressure, have been known to be susceptible to the condition of POPE (20–23). However it is unknown whether cats have enough chest wall musculature to induce POPE. Tyloxapol is reportedly toxic to epithelial cells and red blood cells (24) and has been associated with causing pulmonary hemorrhage in infants with respiratory distress syndrome. Tyloxapol may be included as a contributing factor in our case. Mechanical ventilation with high tidal volume has been known to induce lung injury (25). In our case, ventilation was performed frequently during a relatively short period, and a lower tidal volume was not applied. Mechanical ventilation may be a risk factor in our case. Malignant lymphoma can worsen abruptly and may result in bilateral diffuse pulmonary infiltrates, but this is not common. In our case, histopathological evaluations of the lung were not performed and therefore, lymphoma might have contributed as direct lung injury predisposing to development of ALI. In addition, malignant lymphoma is widely recognized to be associated with disseminated intravascular coagulation (DIC) in cats (26). ALI/ARDS is an accentuation of inflammatory states in the lung and is triggered by both injury of vascular endothelial cells and DIC. The pathophysiology of ARDS and DIC appears to be linked; therefore, malignant lymphoma may also be included as a possible indirect risk factor. Although 2 of 19 dogs developed DIC in Parent’s study (6), whether DIC directly contributed to ALI was unclear. In our case, DIC was not confirmed because complete blood counts and measurements of coagulation factors such as fibrin degradation products, fibrinogen, prothrombin time, and activated partial thromboplastin time were not performed at the time of diagnosis. To our knowledge, ALI/ARDS has not been previously reported in a cat. Further study will be required to determine the specific risk factors in cats.

In our case, advanced imaging was used to substantiate a clinical diagnosis of ALI/ARDS. In humans, CT is widely used to characterize diffuse pulmonary parenchymal diseases (27). Morphological changes of diffuse alveolar injury found on CT images include patchy or diffuse ground-glass opacification, consolidation, and a reticular pattern (28). X-ray attenuation, expressed in HU, is therefore increased on CT images. Lung attenuation is generally homogeneous at full inspiration and normally ranges from −650 to −800 HU in cats (29). In our cat, HU ranged from −10 to −700 when ventilation was maintained with PEEP of 8 cm H₂O during the CT scan. We considered that the lung attenuation value in this cat was abnormal and indicative of increased lung density.

Although various treatments of ALI/ARDS have been investigated in humans, the gold standard treatment of ALI/ARDS has not been established. The mortality rate has been reported to be as high as 40% to 60% in humans with ALI/ARDS (1). The use of corticosteroids in patients with ARDS is controversial. However, relatively long-term use of methylprednisolone in the early phase of ARDS has been reported to be effective for modification of the organ dysfunction score, lung injury score, oxygenation, and duration of mechanical ventilation (30). Pulmonary accumulation of neutrophils is seen in the early stages of ALI/ARDS, and therefore the neutrophil elastase inhibitor, may effectively prevent the progression of ARDS (31,32). In terms of ventilator settings, a high positive end-expiratory pressure and low tidal volume were reportedly beneficial for pulmonary protection in human patients with ALI/ARDS (25,33). In canine and feline models of ALI/ARDS, numerous medical therapies aimed at attenuating the inflammatory response have been evaluated and showed positive outcomes (10,34–43). However, only a single aspect of the entire syndrome is tested, thus making the information less applicable to clinical veterinary practice. In our case, treatment was primarily supportive and included oxygen, antibiotics, diuretics, fluid therapy, and glucocorticoids. For a definitive assessment of oxygenation status, arterial blood gas analysis should have been repeated frequently to adjust ventilation settings. Although survival in a dog with ALI/ARDS was reported by Walker et al (4), current mortality rates for ALI/ARDS in small animals are close to 100%. Wilkins et al (3) reported diagnostic criteria for ALI/ARDS in veterinary medicine, and the number of animals diagnosed with ALI/ARDS may increase in the future. Further study is needed to understand the complex pathophysiology and investigate effective therapy to prevent progression of ALI/ARDS. CVJ