Abstract
Management of feline chronic lower airway disease focuses on controlling clinical signs and decreasing airway inflammation. This retrospective study evaluated the correlation between the resolution of clinical signs in cats with lower airway disease receiving oral glucocorticoids with the resolution of inflammation based on bronchoalveolar lavage fluid (BALF) cytology. Ten cats diagnosed with lower airway disease based on characteristic clinical signs and inflammatory BALF cytology received oral glucocorticoids for at least 3 weeks. They were required to have resolution of clinical signs and BALF collected while asymptomatic and still receiving glucocorticoids. Cats received prednisolone or prednisone (average dose of 1.8±0.2 mg/kg daily) for 35.7±5.5 days. Three cats had resolution of clinical signs and lacked inflammatory BALF cytology; seven had persistent inflammatory BALF cytology despite resolution of clinical signs. Given that subclinical inflammation during high-dose glucocorticoid treatment was common, current recommendations to taper therapy based on resolution of clinical signs should be re-evaluated.
Feline asthma and chronic bronchitis are chronic, diffuse inflammatory diseases of the lower airways that have been estimated to affect 1% of the pet cat population. 1 Inflammation in the airways can lead to irreversible damage termed airway remodeling. 2–5 Airway remodeling in turn can cause a decline in pulmonary function leading to substantial morbidity and occasional mortality. Bronchoalveolar lavage fluid (BALF) analysis is the only diagnostic test readily available in a clinical setting to determine the inflammatory cell type present in the lower airways of cats. 6 In health, alveolar macrophages are the predominant cell type in the lower airway, but during inflammation eosinophils, neutrophils, or both infiltrate into the airways. 7–9 Analysis of BALF is commonly used to diagnose asthma and chronic bronchitis in cats with clinical features suggestive of the disease.
Management of chronic lower airway disease focuses on controlling clinical signs and decreasing airway inflammation with the use of glucocorticoids. 8–11 Response to therapy may be evaluated via clinical signs, thoracic radiographs, and/or BALF cytologic analysis. Traditionally, response to therapy is monitored non-invasively by the assessment of clinical signs and radiographs alone. 7–10 In humans with asthma, clinical signs and radiographs are poor indicators of airway inflammation resolution. 12–15 Despite the role inflammation plays in progression of airway remodeling, no studies in cats have determined if absence of clinical signs correlates with absence of airway inflammation. Furthermore, up to 23% of cats with lower airway disease have normal thoracic radiographs making the utility of this diagnostic test for documenting resolution of airway inflammation questionable. 16
Glucocorticoid therapy for feline asthma and chronic bronchitis is ideally tapered to the lowest effective dose. The decision to taper glucocorticoids is often based on resolution of clinical signs. 7–10 To date, the correlation between improvement of clinical signs and resolution of airway inflammation has not been documented in cats receiving glucocorticoid therapy for asthma and chronic bronchitis. As clinical signs are often the driving force behind recommendations to alter therapy in cats with lower airway disease, it is essential to determine if clinical signs can reliably predict the presence of airway inflammation in cats. Starting in January 2007, the Internal Medicine service at the University of Missouri Veterinary Medical Teaching Hospital recommended using BALF cytology to monitor the efficacy of therapy in pet cats with lower airway disease. This decision was based on information derived from the human literature as well as experiences with an experimental model of feline asthma (unpublished data). We hypothesized that lower airway inflammation could be documented in asymptomatic pet cats diagnosed with asthma or chronic bronchitis being treated with high-dose oral glucocorticoids, and caution must be exercised in equating the absence of clinical signs with the absence of airway inflammation.
Material and methods
Case selection
The electronic medical database of the Veterinary Medical Teaching Hospital, University of Missouri, Columbia was searched for cats diagnosed with asthma or chronic bronchitis based on BALF cytology between January 2007 and December 2009. Five criteria had to be met for inclusion in the study. First, a cat had to have at least one of the following clinical signs: cough, wheeze, or episodes of expiratory respiratory distress. In addition, the clinical signs had to be chronic in duration which was defined as ≥2 months. Second, all cats had to have thoracic radiographs performed to rule out other cardiac or pulmonary diseases (eg, heart failure, neoplasia, pneumonia, etc) that might have similar clinical signs. Third, BALF cytology had to demonstrate ≥17% eosinophils or ≥7% neutrophils. 17,18 Fourth, there had to be no evidence of bacterial infection in the lower airways based on BALF cytology, culture, or both. Finally, after an initial diagnosis of asthma or chronic bronchitis, cats had to be treated with daily high-dose oral glucocorticoid therapy for a minimum of 3 weeks, have a recheck BAL while still receiving daily oral glucocorticoids and have complete resolution of clinical signs at the time of re-evaluation. The minimum requirement of 3 weeks of treatment was based on the finding that airway inflammation resolves after 2 weeks of glucocorticoid therapy in cats with experimentally induced asthma. 11 Cats meeting the entry criteria were subsequently divided into two groups: cats that were asymptomatic and had cytologic resolution of inflammation (group 1) and cats that were asymptomatic but still had cytologic evidence of inflammation (group 2).
Information obtained from the medical records included breed; sex; age at initial diagnosis; current medications; clinical signs; physical exam findings; results of diagnostic tests (thoracic radiographs, BALF cytology, and BALF culture); treatment and all follow-up information. All thoracic radiographs were interpreted by a board certified radiologist. Treatment consisted of glucocorticoids (prednisone or prednisolone) and antibiotics or bronchodilators at the attending clinician's preference.
Collection and analysis of BALF
The anesthetic protocol for BALF collection was determined for each cat on an individual basis by a board certified anesthesiologist. Three of the 10 cats were administered subcutaneous terbutaline 30 min before induction. The decision to give terbutaline was clinician dependent and was often based on a history of the patient having episodes of respiratory distress. Bronchoalveolar lavage fluid was collected through a sterile bronchoscope or a sterile 8 French red rubber catheter blindly wedged in a lower airway. 17 Usually, 20 ml of sterile saline were used for the procedure, however, this amount could vary between 10 and 30 ml based on clinician preference. BALF cytology was prepared by cytospin and was interpreted by a board certified clinical pathologist. Airway inflammation was defined as ≥17% eosinophils or ≥7% neutrophils. 17,18 At the clinician's discretion, an aliquot of the BALF sample was submitted for culture to the Veterinary Medical Diagnostic Laboratory at the University of Missouri.
Results
Forty-three cats had BALF collected between January 2007 and December 2009. Thirty-two (74%) of those cats were diagnosed with asthma or chronic bronchitis. Of those 32 cats, 15 (47%) returned for recheck collection of BALF. Five of these cats were excluded because of persistent clinical signs; concurrent persistent airway inflammation on BALF cytology was confirmed in all five cats. Of the remaining 10 cats, three (30%) had resolution of their clinical signs and no evidence of inflammation on BALF cytology (group 1), and seven (70%) had resolution of their clinical signs but had persistent evidence of airway inflammation based on BALF cytology (group 2).
Group 1
The age of the three cats in group 1 at the time of diagnosis was 2, 3 and 14 years of age. There were two spayed females and one castrated male cat. Breeds included domestic longhair (two) and domestic shorthair (one). Clinical signs related to lower airway disease at the time of presentation were coughing (two) and intermittent respiratory distress (one). The physical exam findings related to lower airway disease included increased bronchovesicular sounds (two), wheeze (one), and tachypnea (one). Only one of the cats had been treated for a cough prior to initial presentation. This cat was on prednisone 0.6 mg/kg, PO, q 24 h at the time of initial presentation and did not have a cough but was having episodes of respiratory distress.
All cats had three view thoracic radiographs which revealed a bronchointerstitial pattern (two), diaphragm flattening (two), bronchial pattern (one), bronchial mineralization (one), and enlarged pulmonary artery (one). Collection of BALF was performed using bronchoscopy (two) or in a blind fashion (one). Bacterial culture was performed on all samples and they all revealed no growth. Cytology was performed on all samples and the results of cytology are listed in Table 1
Bronchoalveolar lavage fluid eosinophil and neutrophil percentage from 10 cats at initial diagnosis and after resolution of clinical signs during treatment. Group 1 cats had complete resolution of airway inflammation while group 2 cats had persistent evidence of airway inflammation on BALF cytology despite resolution of clinical signs.
FS=female spayed, MC=male castrated, DSH=domestic shorthair, DLH=domestic longhair, Eos=eosinophil, Neut=neutrophils.
All cats were treated with prednisolone at a mean±standard deviation (SD) dose of 2.0±0.1 mg/kg, PO, q 24 h (range 1.9–2.2 mg/kg) for an average of 31.3±6.5 days (range 21–43 days). Two cats were treated with doxycycline 5 mg/kg, PO, q 12 h for 2 weeks. One cat was also treated with theophylline 23 mg/kg, PO, q 24 h and with an albuterol inhaler, two puffs as needed for rescue, until the recheck. At the time of recheck all previous clinical signs had completely resolved. One cat still had a slight increase in bronchovesicular sounds on thoracic auscultation but an otherwise normal physical examination; physical examination was unremarkable for the other two cats. All cats had BALF collected blindly and cytology was performed on all samples. Cytologic resolution of airway inflammation was confirmed in all cats (Table 1).
Group 2
The median age of the seven cats in group 2 at the time of diagnosis was 7 years of age (range 2–10 years). There were five spayed female and two castrated male cats. Breeds included domestic shorthair (five) and domestic longhair (two). Clinical signs related to lower airway disease at the time of presentation were coughing (seven) and episodes of respiratory distress (one). The physical exam findings related to lower airway disease included increased bronchovesicular sounds (one), inducible cough (one), and increased respiratory effort (one). Four of the cats had been treated for a cough prior to presentation with antibiotics (three) or glucocorticoids (one). Only two cats were receiving treatment at the time of initial presentation, one with cefovecin (Convenia; Pfizer), 8 mg/kg SC once, and the other enrofloxacin (Baytril; Bayer) 6 mg/kg, PO, q 24 h.
All cats had three view thoracic radiographs which revealed a bronchointerstitial pattern (four), peribronchial thickening/cuffing (three), cardiomegaly (three), aerophagia (two), bronchial pattern (one), pulmonary artery enlargement (one), and diaphragm flattening (one). Two cats had thoracic radiographs interpreted as unremarkable. Collection of BALF was performed in a blind fashion (four) or using bronchoscopy (three). Bacterial culture was performed on six samples and all but one culture revealed either no growth or growth of <2×103 cfu/ml consistent with findings in healthy cats. 7 The cat that had a positive culture (Pseudomonas species) did not have evidence of infection on BALF cytology based on the absence of degenerative neutrophils and intra-or extracellular rods, and so was most likely an upper airway contaminant. Cytology was performed on all samples. The results of cytology are listed in Table 1.
Six cats were treated with prednisolone and one with prednisone at a dose of 1.8±0.2 mg/kg, PO, q 24 h (range 0.8–2.6 mg/kg) for an average of 37.6±7.8 days (range 21–66 days). Five cats were treated with antibiotics: three with doxycycline 6–8 mg/kg, PO, q 12 h for 2 weeks; one with enrofloxacin 6 mg/kg, PO, q 24 h for 2 weeks; and one with azithromycin 5 mg/kg, PO, q 24 h for 5 days. In addition, one cat was prescribed fenbendazole 49 mg/kg, PO, q 24 h for 10 days and theophylline 3 mg/kg, PO, q 24 h until the recheck. Two cats were prescribed an albuterol inhaler, 1–2 puffs, and one terbutaline, 0.01 mg/kg SC, both to be given as needed for rescue. At the time of recheck all previous clinical signs reported by the owners had completely resolved. On physical examination, two of the cats had an increased respiratory rate which was attributed to stress as they both had normal effort and did not have an increased respiratory rate at home. All cats had recheck BALF collected blindly and cytology was performed on all samples. In these cats, who no longer exhibited clinical signs of asthma or chronic bronchitis, cytologic persistence of airway inflammation was confirmed (Table 1).
Discussion
Seventy percent of cats diagnosed with asthma or chronic bronchitis that had resolution of clinical signs of cough, wheeze or episodic respiratory distress with concurrent high-dose glucocorticoid therapy (for a minimum of 3 weeks) still had persistent eosinophilic or neutrophilic airway inflammation based on BALF cytology. This is the first study in pet cats with naturally developing asthma or chronic bronchitis treated with high-dose oral glucocorticoids and having complete resolution of clinical signs that documents persistence of airway inflammation, ie, ‘subclinical inflammation’. As current recommendations for cats with asthma and chronic bronchitis are to taper the dosage of glucocorticoids based on resolution of clinical signs, this study has far-reaching implications on the currently recommended management protocols for these diseases. 7–10
The idea that patients with chronic inflammatory airway disease can have persistent subclinical inflammation is not new. This phenomenon has been well documented in humans with lower airway disease, particularly asthma. 19 For example, many children with allergic asthma appear to outgrow the condition with resolution of clinical signs during puberty; however, up to 80% of these individuals relapse with clinical signs years to decades later. 20 When asymptomatic adolescents with previously diagnosed childhood asthma were evaluated, ongoing subclinical inflammation was documented. 13,21 Other studies of asthma and chronic obstructive pulmonary disease (COPD) in humans have documented a lack of correlation between clinical signs and the presence of airway inflammation. 15 When management of asthma in humans is based on monitoring airway inflammation instead of symptoms, the mucosal inflammation is better controlled. 15,22 Persistent inflammation in asthma and COPD leads to more severe exacerbations and more frequent hospital admissions. 15 In people, poorly controlled asthma can ultimately lead to a decline in pulmonary function and irreversible airway obstruction. 23 Short-term studies show that effective corticosteroid therapy is associated with improved lung function, reduced airway responsiveness, fewer exacerbations, and reduced mortality. 23 The mechanisms by which corticosteroids lessen clinical signs without reducing inflammation have yet to be determined; perhaps they alter the neurogenic pathways associated with cough. 24
There were several limitations of this study due to its retrospective nature. First, assessment for other causes of lower airway inflammation such as heartworm disease and lungworms was not always performed. However, the rationale behind performing this study was to determine how effective resolution of respiratory clinical signs was in predicting resolution of airway inflammation. Even if the cats had an undiagnosed parasitic infection triggering airway inflammation, it is still important to note that a large percentage of cats treated with oral glucocorticoids had a lack of clinical signs despite persistent airway inflammation. Second, one cat received prednisone instead of prednisolone which is considered less bioavailable in the feline species. 25 Thus it might be speculated that poor control of inflammation in this cat might be due to the reduced efficacy of this glucocorticoid. However, this cat had complete resolution of respiratory clinical signs, underscoring the concept that there was perceived efficacy of the prednisone based on the current recommendations for monitoring therapeutic efficacy (ie, resolution of clinical signs) in cats with asthma or chronic bronchitis. 7–10 A third limitation was the small number of cases included in this study which limited the utility of statistics to descriptive statistics alone. Due to the added risk and cost associated with BALF collection, this test is not frequently repeated after the initial diagnosis. However, the fact that 70% of asymptomatic cats at recheck evaluation (while still receiving high doses of oral glucocorticoids) had evidence of lower airway inflammation on BALF analysis is especially impressive with a small case size.
As cats with asthma and chronic bronchitis treated with oral glucocorticoids can have subclinical inflammation, the current recommendation to taper the dosage of glucocorticoids based on resolution of clinical signs needs to be revisited. Collection of BALF is invasive, and while the blind technique can be performed rapidly (within 1–2 min in experienced hands), repeated collection of BALF in pet cats prior to each attempt to taper the dose of glucocorticoids is likely unrealistic. Therefore, futures studies need to focus on developing less invasive methods for assessment of airway inflammation. In humans, exhaled nitric oxide (NO) and induced sputum eosinophils have proved valuable for this purpose. 26 Exhaled NO currently requires expensive machinery, and collection of induced sputum in cats for cytologic analysis is not feasible. In cats, hydrogen peroxide in exhaled breath condensate (EBC) has been used to detect lower airway inflammation in a research setting. 27 Other novel biomarkers in EBC are currently under investigation in experimentally asthmatic cats. 17,28 If biomarkers could be identified using complex methodologies such as mass spectroscopy or nuclear magnetic resonance spectroscopy, the next step would be development of point-of-care testing. Additional studies are clearly warranted.
This study documented that it is common for cats with asthma or chronic bronchitis treated with high-dose oral glucocorticoids to have resolution of clinical signs despite having persistent lower airway inflammation. These results support the claim that caution must be exercised in equating the absence of clinical signs with the absence of airway inflammation. Airway inflammation contributes both to airway hyperreactivity and airway remodeling. 29,30 The former can exacerbate clinical signs and the latter can contribute to decreased pulmonary function and irreversible airway obstruction. 31 Thus an important goal of therapy should be to control airway inflammation along with clinical signs. Premature tapering of glucocorticoids based on absence of clinical signs in cats with subclinical inflammation is anticipated to be detrimental in the long run. Future studies need to be performed to determine a better approach to the chronic management of airway inflammation in cats with asthma and chronic bronchitis.