To the Editors:
Herein we report the progression of irreversible airflow obstruction in a patient with diffuse idiopathic pulmonary neuroendocrine cell hyperplasia (DIPNECH). Following an episode of chronic eosinophilic pneumonia associated with raised sputum substance P and infiltration of the interstitum and alveolar spaces with eosinophils, the fixed airflow limitation became variable (i.e. asthma), requiring treatment with inhaled corticosteroids and long-acting β-agonists.
A number of airway diseases, such as smoker's bronchitis, emphysema, asthma, bronchiolitis and bronchiectasis, can lead to chronic airflow limitation 1. However, variability and reversibility of airflow limitation is only observed in asthma. Inflammatory cell infiltration of the airway, particularly with eosinophils, may be one of the mechanisms of variable airflow limitation 2. Although the natural history of eosinophilic bronchitis is not well known, it is believed that it may lead to chronic irreversible airflow limitation 3. However, it is unusual for chronic irreversible airflow limitation to show variability and improvement in airway calibre following treatment with corticosteroids. This case report describes the changes in lung function and response to treatment of a patient with chronic airflow limitation due to smoker's bronchitis and DIPNECH following eosinophilic inflammation of the airway and the interstitium. The case also highlights the importance of measuring airway inflammation in the management of airway diseases.
A 65-yr-old female with a history of hypercholesterolemia, vertebral osteoarthritis, gall stone disease and endometriosis but no previous respiratory illness, except a mild intermittent cough for the past 4 yrs, presented with fever, night sweats, worsening cough and weight loss of 6.8 kg (15 lbs). She did not have muscle or joint aches, a skin rash or bowel symptoms. She appeared pale and had inspiratory crackles in her left axillary region. After accumulating a 12 pack-yr smoking history, she had quit smoking 15 yrs earlier. She had immigrated to Canada from Germany in 1961. She did not keep any pets at home, had not recently travelled outside of Ontario (Canada) and did not have any known risk factors for HIV infection. She was not on any medications that could cause adverse effects on the lung.
A blood count showed anaemia, lymphopenia, mild eosinophilia (1.0 ×109·L−1) and thrombocytosis. An auto-antibody profile and fungal and avian precipitins were negative. Skin prick tests with common aeroallergens including aspergillus were also negative. A tuberculin test showed a 20-mm induration. A chest radiograph performed at the onset of her illness showed soft fluffy infiltrates of both upper zones which became extensive 1 week later. At the time of assessment in our clinic (Firestone Institute for Respiratory Health, Hamilton, Ontario), a chest radiograph showed almost complete resolution of the right upper-zone infiltrate and further extension of the infiltrate in the left lung to the mid-zones. A high-resolution computed tomography (HRCT) scan of the chest showed ground-glass opacities in both upper lobes and extensive infiltrates with air bronchogram in the left lower lobe extending to the periphery of the lung. Sputum and bronchoalveolar lavage fluid examination did not grow any pathogenic organisms. Substance P in sputum supernatant (measured by radioimmunoassay using antisubstance P antibody (Peninsula Laboratories, San Carlos, CA, USA), as previously described 4) was elevated at 1,210 pg·mL−1 (normal <400 pg·mL−1). Spirometry showed a mixed obstructive and nonobstructive ventilatory pattern without variable airflow obstruction (table 1⇓).
Histopathological examination of a video-assisted thorascopic biopsy from the left lower lobe showed two distinct disease processes. The alveoli were filled with abundant eosinophils and some fibrinous exudate, which extended into the related respiratory bronchioles and showed evidence of organisation. The muscular pulmonary arteries and the veins of the affected areas showed low-grade non-necrotising vasculitis and scant infiltration by eosinophils and lymphocytes. Examination for fungi by Grocott staining was negative. These findings were consistent with chronic eosinophilic pneumonia (fig. 1a⇓). The other finding was proliferation of neuroendocrine cells, which stained positive for chromogranin A and stomatophysin in relation to membranous bronchioles, some of which showed fibrosis with narrowing and obliteration. This was consistent with a diagnosis of DIPNECH (fig. 1b⇓).
The patient showed remarkable clinical and radiological improvement following treatment with prednisone for a total period of 6 months. The proportion of eosinophils in induced sputum was 1% and the blood eosinophil count had decreased to 0.2 ×109·L−1. Although the post-bronchodilator forced expiratory volume in 1 s (FEV1) improved to 1.6 L, she had mild impairment of gas exchange (diffusing capacity 58% predicted, but normal when adjusted for the alveolar volume) and slightly reduced lung volume (72% pred). The exercise capacity was also in the normal range but it was low. She remained symptom free for 1 yr before she noticed wheezing on exertion and a recurrence of her cough. The FEV1 and vital capacity (VC) were 1.5 L and 2.4 L, respectively, improving to 1.8 L and 2.6 L, respectively, after inhalation of 200 μg of salbutamol. The blood eosinophil count was normal (0.2 ×109·L−1), but the proportion of eosinophils in sputum had risen to 6% (normal <1.1%). A HRCT scan of the chest did not show alveolitis. A few small nodules were seen to be scattered over both lung fields. She was started on a combination of inhaled fluticasone (250 μg) and salmeterol (50 μg) twice daily with remarkable improvement in symptoms and lung function (FEV1/VC 1.7/2.7 L) without further bronchodilator improvement. She has remained symptom free during 4 yrs of follow-up. Sputum substance P was 82 pg·mL−1 (table 1⇑).
This report highlights two interesting observations. First, we describe the first reported association of DIPNECH and chronic eosinophilic pneumonia and speculate that products of neuroendocrine cells, such as substance P, may have contributed to the development of lung eosinophilia. Secondly, we describe an unusual progression of airway physiology; eosinophilic inflammation of the airway and the alveolar spaces. This resulted in marked variability and bronchodilator reversibility in the airflow limitation requiring monitoring of airway inflammation using induced sputum to guide maintenance therapy with an inhaled corticosteroid and a long-acting bronchodilator.
Reactive pulmonary neuroendocrine cell hyperplasia has been observed in people living at high altitudes 5 and individuals living with other pulmonary conditions, particularly bronchiectasis, chronic lung abscesses, chronic obstructive pulmonary disease and pulmonary interstitial fibrosis 6. However, DIPNECH has been defined as a precursor to carcinoid tumours 7, and is mostly described in middle-to-older aged, nonsmoking females 8–10. It is plausible that hyperplasia of neuroendocrine cells may have been triggered in our patient by smoker's bronchitis or it may be idiopathic. It is difficult to speculate if the eosinophilic pneumonia preceded or succeeded the development of DIPNECH. We believe that smoker's bronchitis, diffuse neuroendocrine hyperplasia and eosinophilic pneumonia all contributed to the mixed airway and interstitial disease in this patient (fig. 2⇓).
It is well recognised that eosinophilic bronchitis and eosinophilic pneumonia are responsive to treatment with corticosteroids. Indeed, this is reflected in the clinical, radiological and physiological improvement observed in our patient. The steroid-responsive component of the airway disease resulted in partial improvement in FEV1. There was no significant change in the diffusion coefficient. The subsequent worsening of airflow limitation that showed bronchodilator reversibility, which by definition is asthma 11, is unusual. It is recognised that a large proportion of patients with chronic eosinophilic pneumonia may have asthma preceding the illness 12. However, it is rare for asthma to develop after 6 months of treatment with prednisone for eosinophilic pneumonia, particularly when the patient did not have peripheral blood eosinophilia. It is likely that an eosinophilic inflammation localised to the airway resulted in the variable airflow limitation. This was detected by examining cell counts in cytospins prepared from dithiothreitol dispersed sputum induced with hypertonic saline. This is a reproducible and reliable method to assess airway inflammation and is clinically useful in the management of airway diseases and in assessing responses to treatment 13.
A tantalising speculation for the development of interstitial and airway eosinophilia and asthma is DIPNECH. The secretory products of from these cells, such as bombesin-like peptides, tachykinins and calcitonin gene-related peptides, may cause airway eosinophilia and bronchoconstriction 14–18. Raised levels of substance P in the patient's sputum supernatant allowed us to speculate that the hyperplastic neuroendocrine cells may be responsible for airway eosinophilia 19 and variable airflow limitation.
In summary, we report the first association of DIPNECH and chronic eosinophilic pneumonia and its progression to eosinophilic bronchitis and asthma.
Support statement
P. Nair is supported by a Canada Research Chair in Airway Inflammometry.
Statement of interest
A statement of interest for P.M.O’Byrne can be found at www.erj.ersjournlas.com/misc/statements.dtl
Acknowledgments
We would like to thank M. Kay for the pathology reports and J. Urschel and G. Cox (all McMaster University, Hamilton, ON, Canada) for assisting in the management of the patient.
- © ERS Journals Ltd