Copyright ©ERS Journals Ltd 2002 Steroid treatment based on the findings of transbronchial biopsy in idiopathic interstitial pneumonia1 Division of Respiratory Medicine, Dept of Internal Medicine, Fukuoka University School of Medicine, Fukuoka, 2 Dept of Internal Medicine, Hamanomachi Hospital, Fukuoka, 3 Dept of Respiratory Medicine, Fukuoka Higashi Hospital, Fukuoka, 4 Dept of Internal Medicine, Kyushu Kosei-Nenkin Hospital, Kitakyushu, 5 Dept of Internal Medicine, Fukuoka University Chikushi Hospital, Fukuoka and 6 Saitama Cardiovascular and Respiratory Centre, Division of Pathology, Saitama, Japan CORRESPONDENCE: K. Watanabe, Division of Respiratory Medicine, Dept of Internal Medicine, Fukuoka University School of Medicine, Fukuoka, 814-0180, Japan. Fax: 81 928738008. E-mail: watanabe@fukuoka-u.ac.jp Keywords: idiopathic interstitial pneumonia, nonspecific interstitial pneumonia, steroid, transbronchial lung biopsy, vital capacity
Received: September 9, 2001
This study was performed to find the rationale for administering steroids to patients with idiopathic interstitial pneumonia (IIP), which was unlikely to be usual interstitial pneumonia (UIP) but was not surgically biopsied. Among IIP patients in the file of the departments, nine patients who met the following criteria were selected for this study ("non-UIP" group): 1) transbronchial lung biopsy showed dense mononuclear cell infiltration in thickened alveolar septa; 2) chest radiograph and computed tomography showed irregular linear, reticular or ground-glass opacities with alveolar consolidation without honeycombing in the lung base; and 3) spirometry was performed before and after steroid therapy. Ten patients with pathologically confirmed nonspecific interstitial pneumonia ("NSIP" group) were also selected for the comparison. Baseline values and percentage increase of vital capacity (VC) after steroid therapy were plotted. Steroids improved VC in both groups of patients. After 1 yr of steroid therapy, percentage increase of VC in "non-UIP" was 28.8±7.7%, which was not significantly different from that in NSIP (30.0±11.7%). One "non-UIP" patient and one NSIP patient died after 6.4 and 4.3 yrs of follow-up, respectively. Patients with idiopathic interstitial pneumonia presenting cellular interstitial pneumonia in transbronchial lung biopsy, in addition to radiographic findings not typical for usual interstitial pneumonia, could expect a beneficial effect of steroids without undergoing surgical biopsy. Idiopathic interstitial pneumonias (IIPs) are interstitial pneumonias of unknown aetiology including several different pathological entities. These are usual interstitial pneumonia (UIP), desquamative interstitial pneumonia (DIP), respiratory bronchiolitis interstitial lung disease, cryptogenic organising pneumonia or bronchiolitis obliterans organising pneumonia (BOOP), acute interstitial pneumonia, nonspecific interstitial pneumonia (NSIP), and lymphoid interstitial pneumonia (LIP) 1.
Idiopathic pulmonary fibrosis (IPF) is defined as a specific form of chronic fibrosing interstitial pneumonia limited to the lung and associated with the histological appearance of UIP on surgical biopsy 2, 3. Improvements of pulmonary function and chest radiographic and computed tomography (CT) findings in IPF by steroid treatment are minimal and the prognosis is poor. However, patients with a more cellular histology can expect longer survival 4. In addition, less parenchymal abnormality on chest radiography, presence of ground-glass and reticular opacities on CT, younger age, female sex, shorter symptomatic period ( In the present study, the authors examined the rationale of steroid therapy for patients with IIP, which was unlikely to be UIP but not surgically biopsied. Respiratory functions were retrospectively examined, including the change of vital capacity (VC) after the start of steroid treatment in patients whose transbronchial lung biopsy (TBLB) showed cellular interstitial pneumonia, and whose chest radiograph and CT showed irregular linear, reticular or ground-glass opacities with alveolar consolidation without honeycombing in the bilateral lung base. Respiratory function data of these patients before and after steroid therapy were compared with those of pathologically confirmed NSIP.
Patient population All medical files of patients admitted to the Dept of Respiratory Medicine at Fukuoka University Hospital and the Dept of Internal Medicine at Fukuoka University Chikushi Hospital from 19882001 were reviewed. Ninety-seven cases with IIP were found. From the 97 patients with IIP, nine patients who met the following criteria were selected for the present study ("non-UIP" group): 1) although pulmonary lesions were not surgically biopsied, TBLB was performed showing dense mononuclear cell infiltration in thickened alveolar septa in the absence of granuloma, tumour, organising pneumonia or neutrophilic infiltration; 2) dominant patterns of chest radiographs or CT were irregular linear, reticular or ground-glass opacities with alveolar consolidation, but without honeycombing in the bilateral lung base; 3) restrictive ventilatory disturbance (decreased VC and/or decreased total lung capacity (TLC)) and decreased diffusion capacity of carbon monoxide (DL,CO) were found; 4) steroids were administered after diagnosis; 5) spirometry was performed before the start of steroids and it was performed three times or more during the follow-up period; and 6) known causes of pulmonary fibrosis, such as collagen vascular disease, hypersensitivity pneumonitis, and drug-induced lung disease were absent. In the medical files of Fukuoka University Hospital, Hamanomachi Hospital, Fukuoka Higashi Hospital and Kyushu Kosei-Nenkin Hospital between 19882001, 10 patients with idiopathic NSIP were found, seven with fibrosing pattern 9, Group II 10 and three with cellular pattern 9, Group I 10, who had been administered steroids and had performed a respiratory function test before and after steroid therapy. Their diagnosis had been pathologically confirmed by open lung biopsy (OLB) or video-assisted thoracoscopic biopsy (VATS). These patients were also selected for the comparison with the respiratory function data of "non-UIP" patients. Length of follow-up periods was calculated in years as the interval from the start of steroid therapy to the latest clinic visit or time of death.
Respiratory function data
Spirometry just before steroid therapy was performed for all patients in this study and it was performed at least six times after the start of steroid therapy for both groups of patients except three patients (five times in patient 3, three times in patient 4, and one time in patient 15; tables 1 and 2
Statistical analysis Values are expressed as mean±sem. Percentage increases in VC, TLC and DL,CO and increase in Pa,O2 at a certain time point were compared between groups by an unpaired t-test. p-Values <0.05 were considered significant.
Patient selection Of 97 patients with IIP observed from the medical files, chest CT and radiographs were available for evaluation in 67 patients, and chest radiographs alone (no CT available) in five patients (table 3
Patient characteristics Nine patients from the "non-UIP" group included four males and five females, and 10 with NSIP included two males and eight females. The mean ages of "non-UIP" and NSIP were 58.2±3.3 and 54.8±3.5 yrs, respectively, without significant difference between the groups (table 1 Periods from the onset of symptoms to the first visit to the hospital and follow-up periods after the start of steroids were not significantly different between groups. Of nine patients in "non-UIP", one (patient 8) died of an unknown cause 6.4 yrs after the start of steroids, and one patient with NSIP (patient 10) died of respiratory failure due to rapid exacerbation of interstitial opacities on chest CT after 4.3 yrs.
Total doses of steroids for the first month of therapy were not significantly different between groups (table 1
Chest radiographs and computed tomography
Chest CT showed irregular linear or ground-glass opacities with air space consolidation predominantly in the bilateral lower lobes. Bilateral lower lobes were involved in all patients. However, honeycomb change was not evident in any of the slices from patients in the "non-UIP" group (fig. 1
Lung biopsies Nonusual interstitial pneumonia patients Lung tissues were biopsied from involved areas identified by chest radiographs and CT. At least two samples were obtained from the periphery of lower lobe bronchi. The mean number of samples obtained by TBLB was 4.3±0.5 (table 1
Nonspecific interstitial pneumonia patients Tissue samples were obtained by VATS in six patients, and by OLB in four patients. Fibrosing NSIP (NSIP, Group II) was found in seven patients (patients 10, 11, 13, 14, 16, 18 and 19), and cellular NSIP (NSIP, Group I) 9, 10 was found in three patients (12, 15 and 17).
Respiratory function data
Percentage increases in VC after 3 months, 6 months and 1 yr from the start of steroids were not significantly different between "non-UIP" and NSIP groups (35.4±6.7% versus 28.0±9.2%, p=0.51; 31.7±7.0% versus 33.8±9.8%, p=0.86; and 28.8±7.7% versus 30.0±11.7%, p=0.93; respectively). In a 2-yr follow-up after steroid therapy, percentage increase in VC in "non-UIP" patients was not significantly different from that in NSIP patients (fig. 4
Total lung capacity TLC before steroid treatment in "non-UIP" patients (71.4±4.7%) was not significantly different from those with NSIP (73.6±5.6%, p=0.77). Maximal percentage increase in TLC in the "non-UIP" group (29.6±5.0%), was not significantly different from that in NSIP patients (35.4±21.0%, p=0.76; table 2
Arterial oxygen tension
Percentage DL,CO before steroid treatment in "non-UIP" patients (46.3±6.3%) was significantly lower than that in NSIP patients (69.0±5.0%, p=0.02). However, maximal percentage increases of DL,CO in "non-UIP" and NSIP patients were 54.2±22.9% and 16.5±12.7%, respectively, without any significant difference between groups (p=0.30; table 2
Since Epler et al. 14 proposed the entity of BOOP and Katzenstein et al. 10 proposed NSIP, pathological differentiation between non-UIP and UIP among IIPs has become more important than ever before, because it is now well known that many non-UIP patients are steroid-responsive and more favourable in prognosis. Surgical biopsies such as VATS and OLB are considered, therefore, necessary for confirming pathological diagnosis, although BOOP may be diagnosed by TBLB in addition to clinical features of patients 15. However, many clinicians in general would like to have less invasive approaches for the diagnosis of IIP, hesitating to persuade patients to undergo surgical biopsy under general anaesthesia, although recently there have been increasing numbers of patients who have undergone VATS. Therefore, in many instances, indication of steroids for patients with IIP, who are unlikely to have UIP, has been decided practically by clinical information, such as chest radiographs, CT and TBLB, less invasive approaches to obtain histological specimens. Pulmonary function testing enables noninvasive quantitative measurement of the severity of IPF, and repeated testing to monitor the course of IPF appears useful for evaluation of disease progression 16. Although clinical criteria such as physical examination and chest radiography can be useful for following patients with IPF, these criteria are difficult to quantitate 17. For these reasons, many researchers have attempted to determine the parameters in pulmonary function testing most reliable for predicting the prognosis of IPF.
In order to give the scientific rationale for using TBLB findings in the decision of steroid indication for patients with IIP, unlikely to be UIP, the extent of the improvement of pulmonary function after steroid therapy between those patients and patients with pathologically confirmed NSIP was compared. NSIP is steroid responsive and more favourable in prognosis than UIP 9, 10. In the present study, "non-UIP" patients were defined as patients having dense mononuclear cell infiltrates in thickened alveolar septa in TBLB and irregular linear, reticular or ground-glass opacities with alveolar consolidation, but without honeycombing in bilateral lung base in chest CT and radiographs. It was demonstrated that steroid responsiveness in "non-UIP" patients was not different from that in NSIP patients and appeared better than that in fibrosing NSIP patients when the effect of steroid was estimated by the improvement of VC. van Oortegem et al. 18 demonstrated that improvement in forced vital capacity (FVC) of >10% from baseline after 3 months of treatment by steroids with or without immunosuppressive agents was predictive of subsequent improvement in IPF. Hanson et al. 19 also reported that IPF patients with In the present study, a close relationship was shown between radiographic appearance not typical for UIP and dense mononuclear cell infiltrates in thickened alveolar septa. However, there were three patients who met radiographic criteria but did not meet the histological criteria for "non-UIP" patients. One of these patients died 2.5 yrs after steroid therapy, and the response to steroids in one patient was minimal. As the third patient could not be followed-up after TBLB, evaluation by spirometry was impossible. Radiographic examination is an initial and important step in identifying steroid-responsive IIP patients. However, IIP patients showing nontypical radiographic appearance for UIP (irregular linear, reticular or ground-glass opacities with alveolar consolidation without honeycombing) could expect a further beneficial effect of steroids without undergoing surgical biopsy if cellular interstitial pneumonia is obtained in TBLB. Several differential diagnoses could be considered for TBLB specimens of "non-UIP" patients. Clinical findings supported the diagnosis of interstitial pneumonia of unknown aetiology in all patients in the present study. In addition to CT findings and short periods from initial symptoms to first visit to hospital, dense mononuclear cell infiltration in alveolar septa in TBLB specimens of "non-UIP" patients could raise the possibility that NSIP, including cellular NSIP, appears most probable in diagnosis among several differential diagnoses. Recently, high-resolution CT findings in NSIP have been extensively examined. Major CT findings in NSIP are bilateral patchy areas of ground-glass opacity with or without areas of consolidation in the lung base. Irregular linear opacities, thickening of bronchovascular bundles and bronchial dilatation are also frequently seen 20, 21. CT findings in these cases are compatible with those in NSIP. LIP is another differential diagnosis. However, many cases of LIP are associated with underlying systemic diseases such as collagen vascular diseases and autoimmune disorders, and idiopathic cases are few. In addition, LIP is more cellular than any other IIPs, such as NSIP. DIP can be denied because of lack of densely packed macrophages in the alveoli. Patient 7 is the only patient who had alveolar macrophages in the alveoli, but the infiltration of macrophages is slight, and this patient was a nonsmoker. Hypersensitivity pneumonitis should also be considered as another differential diagnosis. However, distribution of hypersensitivity pneumonitis is usually dominant in the upper lobe, and irregular linear and air space opacities are inconsistent with hypersensitivity pneumonitis. Hospitalisation of these patients did not result in recovery from their diseases. In addition, no granulomatous lesions were found in any of the TBLB specimens of patients in the "non-UIP" group. Mononulear cell infiltration in the thickened alveolar septa is also found in BOOP. However, alveolitis in BOOP is usually limited in areas of organising pneumonia 22, and there were no intraluminal organised polyps in any of the specimens in the present study. Moreover, the dominant CT pattern in BOOP is circumscribed areas of air space consolidation in a multifocal distribution 23. Such a pattern was not found in "non-UIP" patients in this study. Infectious diseases such as Mycoplasma pneumoniae pneumonia were denied by the favourable response to steroids without use of antimicrobial drugs such as macrolides and fluoroquinolones. Typical idiopathic pulmonary fibrosis (histologically usual interstitial pneumonia) could be diagnosed by clinical criteria 3, 24 without surgical biopsy. Transbronchial lung biopsy is considered to be necessary to rule out other diffuse lung diseases such as granulomatous diseases. However, in some instances, transbronchial lung biopsy could be a diagnostic tool in combination with clinical data including chest radiographs and computed tomography. More importantly, clinicians could administer steroids to such patients without surgical biopsy under the prediction of favourable response to steroid therapy.
For editorial comments see page 1069.
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||