A case of atypical diffuse feline fibrotic lung disease

Abstract

An 11-year-old cat presented for respiratory distress and weight loss. Thoracic radiographs were interpreted as a diffuse bronchointerstitial pattern with bronchiectasis and a mild ventral alveolar pattern on the lateral views. Computed tomography revealed a severe diffuse reticular pattern, relatively hyperattenuating in subpleural regions, with diffuse traction bronchiectasis and some degree of honeycombing. Despite the absence of basal predominance, this pattern was considered to be suggestive of usual interstitial pneumonia (UIP). Other differentials (other types of interstitial lung disease, infectious pneumonitis, neoplasia, or early edema or hemorrhage) were considered less likely based on history and other test results. The cat was discharged without any treatment, and euthanased 5 months later. Post-mortem histological analysis of the lung revealed end-stage lung, with extensive fibrosis that was more severe in subpleural regions, fibroblastic foci and honeycombing, suggestive of UIP. A probable diagnosis of idiopathic pulmonary fibrosis (IPF) was made. The diffuse distribution of the lesions was atypical compared with previous tomographic and histologic descriptions of IPF in cats. This case report suggests a heterogeneity of the pulmonary fibrotic disorders in cats that warrants further investigation for better characterization and classification.

Case Report

Interstitial lung diseases are a heterogeneous group of respiratory tract disorders that implies inflammatory–fibrotic infiltration of the interstitium. ¹ In human medicine, they are classified into five large groups, including connective tissue diseases, treatment-related/drug-induced diseases, primary unclassified diseases, occupational/environmental diseases and idiopathic fibrotic disorders. ¹ Idiopathic pulmonary fibrosis (IPF) is a common idiopathic fibrotic disorder in humans, and has recently been recognized in cats.^1–3 It is characterized by a histopathological pattern of usual interstitial pneumonia (UIP). ⁴ UIP is recognized by a patchy distribution of marked fibrosis, leading to honeycombing, predominantly in a subpleural/paraseptal location, and with presence of fibroblastic foci. ⁵ Spatial and temporal heterogeneity of the fibrotic lesions is characteristic for the disease. ⁴ This pattern, mixing normal parenchyma and affected tissue, was seen in the previously published histological descriptions of IPF in cats. ² The previous computed tomographic (CT) description in two cats showed focally increased soft tissue attenuation, masses and ventral consolidation. ⁶ Here, we report the atypical case of pulmonary fibrosis in a cat that showed a diffuse (but still heterogeneous) distribution of lesions. This case suggests heterogeneity in the tomographic and histological appearance of pulmonary fibrosis in cats.

An 11-year-old, spayed female, indoor-only domestic shorthair cat was referred for an 8 month history of weight loss, inconsistent coughing, decreased activity levels and a 1 week history of rapid, shallow breathing. Upon presentation, the cat was in mild respiratory distress, with diffuse harsh lung sounds and a respiratory rate of 60 breaths per minute. Complete blood count, chemistry profile and blood pressure were within reference intervals. The cat was stable enough to allow thoracic radiographs and whole body computed tomography (CT) under sedation (butorphanol 0.4 mg/kg IM and midazolam 0.4 mg/kg IM), with supplemented oxygenation provided. No incidents related to sedation were reported.

Thoracic radiographs were interpreted as a severe and diffuse bronchointerstitial pattern with peribronchial cuffing, evidence of bronchiectasis in the peripheral lung fields and a ventral alveolar pattern superimposed with the heart on lateral views (Figure 1).

Figure 1

(a) Right lateral and (b) ventrodorsal thoracic radiographs of the patient. The ventral aspect of the middle lung region has been magnified (c). A severe mixed pulmonary pattern with alveolar (dashed line) and bronchointerstitial opacities can be seen. Note the peribronchial cuffing (open arrows) and peripheral bronchiectasis (plain arrow) on the magnified view

CT images were acquired with a 16-slice helical scanner (GE Healthcare) with the patient restrained in dorsal recumbency using Velcro straps. The images were acquired using contiguous, 0.625 mm thick transverse images of the thorax, and reconstructed using soft tissue and lung algorithms. Post-contrast CT images were acquired after intravenous injection of contrast material (Iohexol, Omnipaque; GE Healthcare). Overall, the lung lobes had a similar appearance of a diffuse, bilateral and severe irregular reticular pattern. No basal predominance was noted. The averaged Hounsfield units (HU) from nine random measurements of nondependent parts of all lung lobes was −632, with values ranging from −402 to −775. The expected averaged HU of an unaffected feline lung should be approximately −715, with values ranging from −630 to −800. ⁷ Although the reticular pattern was present in all lung fields, it was especially hyperattenuating in subpleural areas. The extensive reticular pattern was associated with some degree of cystic changes, suggestive of honeycombing and extensive traction bronchiectasis involving the most peripheral airways (Figures 2 and 3). The pulmonary lesions were not contrast-enhanced. Neither enlarged lymph nodes nor pleural disease were seen. Cardiovascular structures appeared normal. Tomographic study of the abdomen, also performed, revealed no abnormalities.

Figure 2

Reformatted dorsal plane computed tomographic images at (a) ventral thorax and (b) mid-thoracic level. Note the diffuse, severe, reticular lung pattern with areas of honeycombing (dashed line). The lesions are more hyperattenuating subpleurally (plain arrows). Traction bronchiectasis is also present (open arrows)

Figure 3

Transverse computed tomographic image of the thorax at the level of the (a) ninth thoracic vertebra and (b) eighth thoracic vertebra. Note the severe reticular pattern throughout the lung parenchyma, more hyperattenuating in subpleural areas (plain arrows). There are also areas of honeycombing (dashed line) and traction bronchiectasis (open arrows)

Differentials for the pulmonary findings initially included pulmonary fibrosis (especially IPF, non-specific interstitial pneumonia or other types of interstitial lung disease), infectious pneumonitis (eg, viral, fungal, protozoal, parasitic or secondary to atypical bacteria), primary or metastatic neoplasia, or early edema or hemorrhage. ⁸ The absence of a primary tumor on physical examination and abdominal CT made metastatic neoplasia unlikely. Chronicity of the clinical signs and restriction of the disease to the lungs were not suggestive of viral pneumonitis or toxoplasmosis. Fungal pneumonia was considered unlikely based on the low probability of exposure, and absence of organized granuloma and enlarged tracheobronchial lymph nodes. Urinary Histoplasma antigen test was declined by the owner, as were fine-needle aspirates of the lungs, which were offered to screen for pulmonary lymphoma. Based on the high suspicion of pulmonary fibrosis, further investigations were declined. The patient was discharged without treatment.

The cat was presented 5 months later for severe respiratory distress and anorexia. The cat’s breathing had progressively worsened to a point where its quality of life was significantly affected. On physical examination, the cat was open-mouth breathing, with severe respiratory distress, and a respiratory rate of 80 breaths per minute. The cat’s gums were cyanotic, and it was cachectic (body condition score 2/9). The owner elected humane euthanasia.

Gross necropsy revealed that the lungs were moderately heavy and firm. They failed to completely collapse when the thorax was opened, and occupied approximately 40% of the chest. Diffusely, the lungs were mottled tan-to-red, and had a cobblestone pleural surface with rib impression.

Microscopically, approximately 85% of the lung parenchyma showed multiple areas of fibrosis characterized by irregularly distributed thick collagen deposits in the interstitium. Trichrome staining confirmed severe fibrosis, which was more prevalent subpleurally, and often associated with pleural retraction (Figure 4). Fibrosis occasionally distorted the subpleural alveolar architecture, resulting in honeycombing (Figure 5). A few fibroblastic foci were observed at the interface between mature fibrotic areas and more normal parenchyma (Figure 6). The alveoli were frequently atelectatic and lined with hyperplastic type II pneumocytes. There was marked smooth muscle hyperplasia. The bronchi and bronchioles were often ectatic and lined with hyperplastic goblet cells. Minimal lymphoplasmacytic inflammation was observed throughout the alveolar interstitium, with no obvious airway involvement. These findings supported a diagnosis of end-stage lung secondary to probable UIP.

Figure 4

Histopathology of lung tissue acquired following necropsy. Magnified image with Masson’s trichrome staining demonstrating irregularly distributed thick bands of brightly blue collagenous connective tissue disrupting the lung parenchyma (× 40). Fibrosis was more prevalent subpleurally, and associated with pleural retraction (arrows). Distortion of the subpleural alveolar architecture resulted in honeycombing (dashed lines)

Figure 5

Histopathology of lung tissue acquired following necropsy. Magnified image with hematoxylin and eosin staining demonstrating subpleural honeycombing (× 100)

Figure 6

Histopathology of lung tissue acquired following necropsy. Magnified image with hematoxylin and eosin staining demonstrating fibroblastic foci (dashed lines) (× 200)

IPF is the most prevalent of the idiopathic interstitial pneumonias in humans. ⁹ It is characterized by a UIP pattern on histology and high-resolution CT (HRCT). ⁵ Recognition of a UIP pattern on HRCT is based on four typical features (Table 1). ⁵ Despite, in our case, a narrow CT slice width (0.625 mm) and the use of a sharping filter, these were not HRCT images. HRCT images would have required a high spatial frequency algorithm, scanning settings for optimal resolution and breath-holding. ¹⁰ Also, as the cat was awake and mildly tachypneic during the CT, motion artifact may be present on the images. These limitations could have affected the interpretation of the CT images. For example, in this case, it was difficult to assess the presence of ground glass opacity. Some of the blurriness could be attributed to mild motion artifact. If some degree of ground glass opacity was present, its extent was considered less than the reticular pattern extent, which is still consistent with UIP. ¹¹ Thick-walled, rounded cysts, arranged in several layers, and especially peripherally distributed, were interpreted as honeycombing, which is a key feature of UIP. ⁴ There was no basal predominance, as the reticulation was present throughout the lungs. However, the lung disease may have been too advanced at the time of the CT to appreciate a basal predominance. Furthermore, no upper or mid-lung predominance was observed, which would have been inconsistent with UIP. ⁵ More hyperattenuating areas of reticulation were found under the pleura, which might suggest that the disease started subpleurally. ¹² This temporal heterogeneity is fundamental in diagnosing UIP. ⁴ For these reasons and despite the diffuse distribution of the lesions, the CT images were interpreted to be most consistent with a pattern of UIP.

Table 1

High-resolution computed tomography criteria for usual interstitial pneumonia (UIP)/idiopathic pulmonary fibrosis in people, according to the official ATS/ERS/JRS/ALAT statement ⁵

UIP pattern*	Possible UIP pattern †	Inconsistent with UIP pattern ‡
Subpleural, basal predominance Reticular abnormality Honeycombing with or without traction bronchiectasis Absence of features listed as inconsistent with UIP pattern	Subpleural, basal predominance Reticular abnormality Absence of features listed as inconsistent with UIP pattern	Upper or mid-lung predominance Peribronchovascular predominance Extensive ground-glass abnormality (extent >reticular abnormality) Profuse micronodules (bilateral, predominantly upper lobes) Discrete cysts (multiple, bilateral, away from areas of honeycombing) Diffuse mosaic attenuation/air-trapping (bilateral, in ⩾three lobes) Consolidation in bronchopulmonary segment(s)/lobe(s)

*

All four features

†

All three features

‡

Any of the seven features

ATS = American Thoracic Society; ERS = European Respiratory Society; JRS = Japanese Respiratory Society; ALAT = Latin American Thoracic Association

Four histological criteria are recognized to define UIP in humans (Table 2). ⁵ Predominantly subpleural marked fibrosis resulting in architectural distortion and honeycombing was retrieved in our case. Fibroblastic foci were present, and no evidence of inflammation was found in the alveoli or airways. These features support UIP. ⁵ However, the lesions were so extensive and advanced that definitive histological diagnosis of UIP was difficult to achieve. In humans, other interstitial lung diseases, such as desquamative interstitial pneumonia, may look exactly like UIP at their end stage. ¹³ Also, histological distinction between end-stage UIP and end-stage non-specific interstitial pneumonia is associated with poor reproductibility. ¹⁴ Some authors recommend not trying to assign a particular histological pattern in this situation, and classify the changes as end-stage lung. ¹⁴

Table 2

Histopathological criteria for usual interstitial pneumonia (UIP)/idiopathic pulmonary fibrosis in people, according to the official ATS/ERS/JRS/ALAT statement ⁵

UIP pattern*	Probable UIP pattern	Possible UIP pattern †	Inconsistent with UIP pattern ‡
Evidence of marked fibrosis/architectural distortion ± honeycombing in a predominantly subpleural/paraseptal distribution Presence of patchy involvement of lung parenchyma by fibrosis Presence of fibroblast foci Absence of features against a diagnosis of UIP, suggesting an alternative diagnosis	Evidence of marked fibrosis/architectural distortion ± honeycombing Absence of either patchy involvement or fibroblastic foci, but not both Absence of features against a diagnosis of UIP, suggesting an alternative diagnosis OR Honeycomb changes only	Patchy or diffuse involvement of lung parenchyma by fibrosis, with or without interstitial inflammation Absence of other criteria for UIP pattern Absence of features against a diagnosis of UIP, suggesting an alternative diagnosis	Hyaline membranes Organizing pneumonia Granulomas Marked interstitial inflammatory cell infiltrate away from honeycombing Predominant airway-centered changes Other features suggestive of an alternative diagnosis

*

All four criteria

†

All three criteria

‡

Any of the six criteria

ATS = American Thoracic Society; ERS = European Respiratory Society; JRS = Japanese Respiratory Society; ALAT = Latin American Thoracic Association

HRCT and histological findings are important in the diagnosis of IPF in humans, but as well as clinical features and the absence of identifiable causes for interstitial lung disease. Indeed, dissimilar clinical entities may result in similar tomographic and histological appearances. ¹ The clinical presentation of our cat was similar to previous descriptions of the disease.^2,3 No known history of toxic exposure (eg, nitrosoureas) that could have caused pulmonary fibrosis was present in our cat. ¹⁵ Diffuse pulmonary metastases of carcinoma can sometimes present as an unstructured interstitial pattern. ¹⁶ Neoplasia was not identified histologically in our cat. An unstructured interstitial pulmonary pattern can also be observed in fungal pneumonia. ¹⁷ Based on low clinical suspicion, fungal urinary antigen testing was not performed, although histoplasmosis has been described in indoor-only cats. ¹⁸ Histologically, no evidence of infection was found in our case.

Despite generalized traction bronchiectasis and coarse reticular opacity, which represents the most severe indicator of fibrosis in humans, the cat lived for 5 months after CT. ¹⁹ This survival time is consistent with previous reports of IPF in cats (average duration of signs before death/euthanasia of 5.5 months). ² In the largest study on feline IPF, only 4/23 cats lived for 1 year or more after initial veterinary evaluation of respiratory signs, despite various treatments, including glucocorticoids, antimicrobials, bronchodilators, diuretics, anthelmintics, cyproheptadine and cyclophosphamide. ³ According to new human guidelines, patients with IPF should not be treated with steroids, colchicine, cyclosporine, interferon-γ, bosetan, etanercept, anticoagulants or pirfenidone. ⁵ For this reason, and based on a previously reported poor response to steroids in cats, our cat was not discharged with any treatment. ³ Some human patients may benefit from acetylcysteine, long-term oxygen therapy or lung transplantation. ⁵ As cats with IPF often deteriorate markedly after invasive respiratory diagnostic procedures, awake or sedated CT should be preferred over CT under general anesthesia.^3,20

The CT findings in this case were different from a previous tomographic description of pulmonary fibrosis in cats, which included focally increased soft tissue attenuation, masses and ventral consolidation. ⁶ Despite their atypical distribution, in our case the lung lesions were more completely defined on the CT scans as opposed to conventional radiographs. Alveolar and bronchial diseases were suspected based on survey radiographs. Therefore, CT is likely a better imaging tool in cats with suspected IPF.

Conclusions

This case report illustrates the likely heterogeneity of feline pulmonary fibrotic disorders. Interstitial lung disease warrants further investigation for better characterization and classification in cats.