Submitted by Dr Arthur CW Lau, ICU, PYNEH, Hong Kong on 7 Sep 2009
A 72-year-old man, a life-long non-smoker, was admitted because of shortness of breath, on and off for one year, with exacerbation in recent one week. He was a retired clerk, and had been coughing up yellow and greenish sputum daily. There was occasional wheeze associated with weather change and which was also worse at night time. He had no symptoms suggestive of cardiac failure. For past health, he had history of pansinusitis. P/E: bilateral crackles. A CXR was taken (figure 1). ECG showed ST elevation in V2 to 3, TnT was raised to 0.29, but he complained of no chest pain. ABG showed type II respiratory failure. His condition later woresened and required ICU admission. Ventilator tracing confirmed mild to moderate airflow obstruction.
Click image to enlarge.
Apart from providing assisted ventilation, giving antibiotics, bronchodilators and steroid as usual, what else would you do?
On closer look, the above CXR shows a hyperinflated chest, as well as many diffuse nodular shadows. Cardiac shadow may well be normal in this AP film. Echocardiogram found no regional wall abnormality. LV ejection fraction was normal, but there was severe RV hypertrophy and paradoxical septal movement compatible with severe pulmonary hypertension.
HRCT thorax was done (figures 2 and 3).
Figure 2. HRCT thorax (click image to enlarge)
Figure 3. HRCT thorax, coronal reconstructions.
The above HRCT thorax shows diffuse tree-in-bud pattern. Sputum for c/st showed H influenzae. Sputum for AFB smear was negative. NPA for influenza and RSV were negative. ESR 52, CRP was raised to 115. Cold agglutinins were raised to 640 (normal <40). ANA -ve. RF slightly increased to 26. Ig G and IgM normal. C3 0.7, C4 0.11. What is the diagnosis?
Impression: ? Diffusion panbronchiolitis (DPB)
Clarithromycin was specifically started to treat this condition. He eventually weaned off the ventilator. Transbronchial biopsy one month after the episode of respiratory failure was non-specific because the sample was inadequate.
Figure 4. HRCT 3 months after Clarithromycin treatment. The previously noted tree-in-bud pattern significantly resolved.
Discussion
Although we could not be absolutely certain that this was a case of diffuse panbronchiolitis, many features in this patient were highly suggestive of this diagnosis. The diagnostic criteria of diffuse panbronchiolitis are:
1. Persistent cough, sputum and exertional dyspnoea
2. History of chronic paranasal sinusitis
3. Bilateral diffuse small nodular shadows on a plain chest radiography film or centrilobular micronodules on chest computed tomography images
4. Coarse crackles
5. FEV1/FVC < 70% and PaO2 < 80 mmHg
6. Titre of cold haemagglutinin >= 64
FEV1: forced expiratory volume in one second; FVC: forced vital capacity; Pa,O2: arterial oxygen tension. Cases definitely established should fulfil criteria 1, 2 and 3, along with at least two of criteria 4, 5 and 6. These parameters are useful for carrying out an epidemiological analysis. In countries in which the disease is very rare, surgical lung biopsy is required to make a diagnosis. Criteria are taken from a working group of the Ministry of Health and Welfare of Japan (ref 7). 1 mmHg = 0.133 kPa.
Diffuse panbronchiolitis is an idiopathic inflammation of the bronchioles. "Diffuse" refers to the lesions which appear throughout both lungs, while "pan-" refers to the inflammation found in all layers of the respiratory bronchioles. DPB is more common in East Asians, including Japanese, Korean, Chinese and Thai than in Caucasians. Haemophilus influenzae or Pseudomonas aeruginosa are common organisms found in this condition. Other features existing in our patient that were compatible with this diagnosis included daily production of purulent sputum, history of chronic pansinusitis, increased cold agglutinin and rheumatoid factor. Our patient also showed response to treatment with macrolide antibiotics, which characteristically exert its effect by immunomodulation (suppression of both neutrophil proliferation and lymphocyte activity) for this condition, instead of by its antimicrobial activity. The condition is not steroid-responsive, and we have also only retained inhaled steroid for him. In macrolide-resistant cases, a recent study found that tiotropium could improve the symptoms of cough, sputum and breathlessness. These beneficial effects were purported to be due to the suppression of airway secretion through the anticholinergic effect of tiotropium on the submucosal gland.
Clinicians must be aware of this disease entity before the diagnosis is even considered. For example, this patient was a non-smoker presenting with airflow obstruction, leading us to think of asthma, though COPD from second hand smoking was possible. The history of pansinusitis should be noted. It is not typical for asthma to have daily production of yellow and greenish sputum. Sputum of COPD is usually not purulent looking. Bronchiectasis is a possibility of purulent sputum, however, CXR usually shows tram-line appearance in fusiform bronchiectasis, or multiple cysts in cystic bronchiectasis. Old TB traction bronchiectasis was absent in our patient. All these atypical features of the above common conditions led us to perform a HRCT thorax for better delineation. The "tree-in-bud pattern" can actually be associated with many differential diagnoses. Follicular bronchiolitis can give similar uniform and diffuse appearance . Respiratory bronchiolitis-interstitial lung disease (RB-ILD) can also give similar appearance, but our patient was a non-smoker. Discussion on the Tree-in-Bud Pattern can be found in the following references.
Reference
1. Lin EC et al. Practical differential diagnosis of CT and MRI. Thieme Medical Publishers, Inc. 2008.
2. Eisenhuber E. The Tree-in-Bud Sign. Radiology 2002; 222: 771-772 (free fulltext here)
3. Kenneth W T Tsang,a Clara G C Ooi,b Mary S M Ip,a Wah-kit Lam,a Henry Ngan,b Eric Y T Chan,c Brian Hawkins,c Chu-shak Ho,a Ryoichi Amitani,d Eisaku Tanaka,d Harumi Itohd Clinical profiles of Chinese patients with diffuse panbronchiolitis. Thorax 1998;53:274-280; doi:10.1136/thx.53.4.274
4. Hye Yun Park,1 Gee Young Suh,1 Man Pyo Chung,1 Hojoong Kim,1 O Jung Kwon,1 Myung Jin Chung,2 Tae Sung Kim,2 Kyung Soo Lee,2 and Won-Jung Koh. Comparison of Clinical and Radiographic Characteristics between Nodular Bronchiectatic Form of Nontuberculous Mycobacterial Lung Disease and Diffuse Panbronchiolitis. J Korean Med Sci. 2009 June; 24(3): 427–432. (free fulltext here)
– Interesting images for comparison available
5. DING Ke, LIU Miao-bing, WU Jin-ling, MA Hui-qing, FANG Xiang-yang, MIAO Guo-bin, ZHANG Lin. Diffuse panbronchiolitis in China: analysis of 45 cases. Chinese Medical Journal, 2007, Vol. 120 No. 22 : 2046-2048 (free fulltext here)
6. V. Poletti1,2, G. Casoni1, M. Chilosi3 and M. Zompatori4. Diffuse panbronchiolitis. Eur Respir J 2006; 28:862-871 (free fulltext here)
7. Nakata K. Revision of clinical guidelines for diffuse panbronchiolitis. Annual report on the study of diffuse lung disease in 1998. Grant-in Aid from the Ministry of Health and Welfare of Japan. Tokyo, 1999; pp. 109–111.