The Clinical Respiratory Journal
ORIGINAL ARTICLE
Pleuroscopy in ‘Idiopathic’ eosinophilic pleural effusions Kostas Archontogeorgis1, Stavros Anevlavis1, Paul Zarogoulidis1, Ajay Jain1, Georgia Karpathiou2, Alexandra Giatromanolaki2, Efthimios Sivridis2, Demosthenes Bouros1 and Marios E. Froudarakis1 1 Department of Pneumonology, Medical School, Democritus University of Thrace, Alexandroupolis, Greece 2 Department Pathology, Medical School, Democritus University of Thrace, Alexandroupolis, Greece
Abstract Background: Idiopathic eosinophilic pleural effusions (IEPEs) comprise the eosinophilic pleural effusions for which a specific aetiology cannot be established. There are no reports investigating IEPE on the basis of a systematically applied pleuroscopy approach and entailing an appropriate patient follow-up till the final outcome is established; existing series rather combine clinical and thoracocentesis criteria to establish the idiopathic character of the diagnosis. Objectives: The aim of our study was to assess the clinical outcome of patients with IEPE, who underwent a systematic diagnostic approach by pleuroscopy. Methods: We studied 10 patients with IEPE among 175 consecutive patients who underwent pleuroscopy for undiagnosed pleural effusion. Pleural biopsies were obtained from observed lesions. All patients were followed up by means of clinical examination and imaging. Results: The diagnosis of IEPE was established in 10 patients (median age was 50.5 years, range 35–91). Macroscopic examination of the pleura showed diffuse thickening with pleural plaques in eight patients, consistent with diffuse pleural eosinophilic inflammation histologically proven. In two patients, macroscopic examination showed scattered nodules associated with non-caseating granulomas histologically. In all 10 patients, a specific aetiology could not be established. Follow-up was available for all patients ranging from 24–102 months (median 60 months). No patient received a specific treatment during the follow-up period. No relapse of a pleural effusion was documented during this period. Conclusion: Pleuroscopy is mandatory in diagnosing IEPE. Negative histology and a long follow-up showed a benign course. These findings suggest that we should call these effusions ‘indeterminate’. Please cite this paper as: Archontogeorgis K, Anevlavis Anevlavis S, S, Zarogoulidis Zarogoulidis P, P, Jain Jain A, A, Karpathiou G, Giatromanolaki A, Sivridis E, Bouros D and Froudarakis Karpathiou Bouros D and Froudarakis ME. ME. Pleuroscopy in ‘Idiopathic’ eosinophilic Pleuroscopy eosinophilic pleural pleural effusions. effusions. Clin Clin Respir Respir JJ 2014; 2015; ••: 9: ••–••. DOI:10.1111/crj.12165. 475–480. DOI:10.1111/crj.12165.
Key words eosinophilic – idiopathic – non-specific – pleural effusion – pleuroscopy – thoracoscopy Correspondence Marios E. Froudarakis, MD, PhD, Department of Pneumonology, Medical School, Democritus University of Thrace, 68100 Alexandroupolis, Greece. Tel: +302551075335 Fax: +302551030343 email: [email protected] Received: 06 January 2013 Revision requested: 02 May 2014 Accepted: 16 May 2014 DOI:10.1111/crj.12165 Authorship and contributorship MF designed the investigation. KA, MF, SA, PZ, AJ performed pleuroscopy procedures and attended the patients. GK, AG and ES examined and reviewed histological specimen. KA, AJ, SA, DB collected patients’ data. MF and DB reviewed data. KA wrote the manuscript, which was then reviewed and approved by all authors. Ethics This retrospective study has been approved by the Internal Review Board of our hospital. Conflict of interest The authors have stated explicitly that there are no conflicts of interest in connection with this article. Dr. Ajay Jain was fellow during 1 month (October 2011) of the Interventional Unit of the Department of Pneumonology, University Hospital of Alexandroupolis, Greece. His current affiliation is: Gujarat Pulmonary and Critical Care Clinic, Navrangpura, Ahmedabad-380009, India
The Clinical Clinical Respiratory Respiratory Journal Journal (2015) (2014) •• ISSN ISSN 1752-6981 1752-6981 C 2014 John Wiley & Sons Ltd V © 2014 John Wiley & Sons Ltd
4751
Idiopathic pleural effusion effusion Idiopathic eosinophilic eosinophilic pleural
Introduction Pleural effusion is a common finding in pulmonary medicine and although a simple radiographic test is often enough to establish its presence, reaching a specific diagnosis can prove difficult (1). A pleural effusion is defined as eosinophilic if it contains 10% or more eosinophils (2). The incidence of eosinophilic pleural effusion (EPE) is approximately 10% (3). Whether EPE should be considered benign per se or because of a non-benign disease is under debate (2, 4, 5). Even though pleural fluid eosinophilia has been associated with benign disorders (5), a recent metaanalysis did not confirm this perception (2). In fact, pleural eosinophilia may be present in various occasions. The presence of air or blood in the pleural space (repeated thoracentesis, hemothorax, pulmonary embolism, pneumothorax) may be the primary cause (6). Malignancies, drugs, asbestos exposure, infections, Churg–Strauss syndrome and rheumatoid arthritis are also causes of EPE (7). When an etiological diagnosis cannot be established, EPE is referred as ‘idiopathic’. The incidence of idiopathic EPE (IEPE) is quite large according to the different studies ranging between 3.8% and 46% (8, 9). This great difference in incidence is due to the fact that generally, these patients do not undergo an invasive procedure to establish tissue diagnosis, and the term IEPE is applied based only on pleural fluid eosinophilia in the absence of evident cause. Pleuroscopy is a useful tool in the diagnosis and treatment of pleural disease by offering a direct visualization of the pleural space to obtain tissue biopsies (10). Diagnostic accuracy of pleuroscopy is high, reaching 95% in patients with malignancy (1, 10). However, in few cases of pleural effusion of indeterminate cause, even after pleuroscopic study, the etiology remains unclear, as histology of pleural biopsies revealing non-specific pleuritis (11, 12). Therefore, histological diagnosis of non-specific pleuritis does not necessarily correspond to a clinical diagnosis of idiopathic pleuritis (12). Thus, the aim of our study is to report a study of patients with IEPE who underwent pleuroscopy as a diagnostic approach, along with macroscopic, microscopic findings and outcome.
Patients and methods Patients A total of 175 consecutive patients who underwent initial negative evaluation for pleural effusion of 476 2
Archontogeorgis Archontogeorgis et et al. al.
unknown origin were considered for the study from January 2003 until December 2010. All underwent pleuroscopy to establish diagnosis. Pleural effusion was considered eosinophilic when the percentage of eosinophils was >10%. Effusion was classified as IEPE, if no etiology could be established after pleuroscopy. Patients with EPE before pleuroscopy were 21 (12%). From those 21 patients, in 11 (52.4%), a definite etiology was established after pleuroscopy and therefore excluded from the study: nine patients had a malignant cause, and two had a benign one. The remaining 10 undiagnosed after pleuroscopy patients (47.6%) were included in our study. Overall, from the 175 patients with undiagnosed pleural effusion who underwent thoracoscopy for diagnostic purposes, finally 129 (73.7%) were diagnosed with malignancy (including the nine EPE), 36 (20.6%) have shown a true benign disease (including the two EPE) and 10 (5.7%) patients had to be followed for IEPE. This retrospective study has been approved by the Internal Review Board of our hospital.
Methods Initial workup (1) Pleural fluid was obtained by a single-needle thoracentesis from the pleural cavity. A cytological examination was performed to all samples. Biochemical examinations included pH, lactate dehydrogenase (LDH), total protein, albumin and glucose. Pleural fluid red blood cells and nucleated cells counts were performed by an automated method, and the differential nucleated count was established by manual counting. Tests on pleural fluid were performed, including stain and cultures for common bacteria, mycobacteria, opportunistic germs, fungi and parasites. Pleural fluid polymerase chain reaction for Mycobacterium tuberculosis was also performed. Biochemical examinations of peripheral blood including LDH, total protein, albumin and glucose were performed to all patients. Eight patients with a documented risk for malignancy (all heavy smokers) underwent bronchoscopy; microbiology tests (including common bacteria and mycobacteria), cytological examination, as well as total cell count and differential count from bronchoalveolar lavage were obtained. For all patients, the authors did not identify any drug, autoimmune disease, recent chest trauma or intra-abdominal cause, responsible for EPE. No evidence of parasite infection, especially hydatid cyst, was seen to different imaging examinations performed or to the blood serum, and our patients had no history of exposure. Chest X-ray was
The The Clinical Clinical Respiratory Respiratory Journal Journal (2015) (2014) •• ISSN ISSN 1752-6981 1752-6981 C 2014 John Wiley & Sons Ltd V © 2014 John Wiley & Sons Ltd
Archontogeorgis Archontogeorgis et et al. al.
Idiopathic Idiopathic eosinophilic eosinophilic pleural pleural effusion effusion
performed before pleuroscopy along with chest ultrasound, but chest computed tomography (CCT) was performed after evacuation of the pleural fluid. An abdominal CT was also performed to rule out any abdominal causes.
Follow-up All patients were followed at 1 month after pleuroscopy, every 3 months during the first year and then every 6 months by clinical examination, with chest radiography, chest ultrasound and/or CCT.
Pleuroscopy
Results
Pleuroscopy was performed with rigid instruments (Richard Wolf®, Knittlingen, Germany) in the bronchoscopy suite, the patient being in the lateral decubitus position, using single port of entry, under local anesthesia with 1% lignocaine. Patients were under cardiovascular and oxygen saturation monitoring during the procedure. Morphine 0.5 mg was administered as needed. A 7-mm trocar was inserted into the fifth or sixth intercostal space along the midaxillary line. After evacuation of the pleural fluid, an Oo optical telescope was inserted and connected to a light source. The entire pleural cavity was then inspected. Large biopsies were taken from lesions observed from the parietal, the visceral and the diaphragmatic pleura, with optical forceps through the single port of entry, under direct visualization. In presence of pleural plaques, biopsies were obtained from the vicinity of the plaques. Cultures from biopsy specimens for Mycobacterium tuberculosis and other microorganisms were also obtained. Closed intercostal drainage was carried out through a chest tube attached to a water-sealed system, and total drainage was recorded. The chest tube was removed after the procedure by verifying the lung re-expansion by chest X-ray. All complications were recorded, as well as duration of drainage and length of hospital stay.
Patients Among 10 IEPE patients (median age was 50.5 years ranging from 35 to 91), five presented with a left and five with a right pleural effusion. Demographic characteristics are shown in Table 1. None of our patients was immunocompromised. All pleural effusions were characterized as exudates according to Light’s criteria, with a lymphocytic predominance accompanying the eosinophilia. Eosinophilic count ranged from 10% to 59% with a median of 19% (Table 2).
Initial workup Cytological examination of pleural fluid was negative for all patients. Pleural fluid cultures for Mycobacterium tuberculosis, other common microorganisms and fungal infections were negative. Polymerase chain reaction for Mycobacterium tuberculosis of the pleural fluid was also negative for all patients. Common cultures from pleural biopsies were negative. None of the patients presented with peripheral blood eosinophilia. Chest X-ray upon admission showed no additional findings. Chest ultrasound demonstrated a nonseptated pleural effusion in six patients, while fibrinous septa were observed in four. Bronchoscopy did not
Table 1. Patients’ demographics and characteristics Patients
Age
Sex
Side of PE
Bronchoscopy
1 2 3 4 5 6 7 8 9 10 Median
35 39 41 59 84 91 44 38 69 57 50.5
Male Male Male Female Male Female Female Male Female Male
Left Right Right Left Left Right Right Left Left Right
Yes Yes Yes No Yes No Yes Yes Yes Yes
Follow-up (months)
Days of drainage
Total fluid
Length of stay
102 96 72 90 48 24 27 72 24 26 60
1 1 1 1 3 4 1 2 2 3 1.5
1500 700 1200 1000 850 1300 750 2000 800 2000 1100
4 12 1 2 6 7 2 5 2 3 3.5
PE, pleural effusion.
The The Clinical Clinical Respiratory Respiratory Journal Journal (2015) (2014) •• ISSN ISSN 1752-6981 1752-6981
C 2014 John Wiley & Sons Ltd V © 2014 John Wiley & Sons Ltd
4773
Idiopathic pleural effusion effusion Idiopathic eosinophilic eosinophilic pleural
Archontogeorgis Archontogeorgis et et al. al.
Table 2. Pleural fluid characteristics Patients
pH
Glucose
Proteins
Albumin
LDH
Total cells
Neutrophils %
Eosinophils %
Lymphocytes %
1 2 3 4 5 6 7 8 9 10 Median
7.37 7.37 7.38 7.30 7.31 7.38 7.34 7.39 7.29 7.41 7.37
86 31 71 98 206 94 107 145 20 83 90
4.6 4.2 4.2 5.3 4.7 3.4 5.5 4.1 5.0 5.0 4.65
3.3 2.2 2.8 3.1 2.7 3.2 2.9 3.0 3.1 3.2 3.1
266 370 415 327 168 329 522 433 1509 521 392.5
4900 1500 1600 3100 600 600 2300 1700 1700 4500 1700
8.8 21 10 38 15 2.7 30 15 18 5 15
59 15 48 10 10 16 11 23 45 40 19
22 54 34 52 60 81 45 61 35 45 48.5
LDH, lactate dehydrogenase.
reveal endobronchial lesions in any of the patients and the cytological examination and cultures of bronchial lavage for common bacteria; M. tuberculosis and mycosis were negative. Total cell count and differential cell count of bronchoalveolar fluid did not reveal any specific pattern. Smear cultures for M. tuberculosis, common microorganisms and cytological examination were negative. None of the patients presented with acute febrile illness.
Pleuroscopy During pleuroscopy, few adhesions were present in seven patients, while three had multiple adhesions. However, we were able to explore the entire pleural cavity after cutting the adhesions, in all patients. Macroscopic examination of the pleura demonstrated diffuse thickening associated to asbestos pleural plaques in eight patients (Fig. 1A) and diffuse multiple nodules in two (Fig. 1B). Patients had a median of 19 biopsies (range 15–26) on parietal pleura, while four patients (40%) had a median of 12 biopsies (range
10–15) on lesions identified on visceral pleura. Two patients had lesions on diaphragmatic pleura that have been biopsied (one patient had three biopsies, the second had five). Microscopical examination of pleural biopsies documented non-specific inflammation with eosinophilic predominance (Fig. 1C) in eight patients and non-caseating granulomas in two. Specifically, an association between pleural plaques, pleural thickening and the presence of diffuse eosinophilic inflammation was noted, as well as between scattered nodules and non-caseating granulomas. Macroscopic and microscopic features are summarized in Table 3. The median quantity of pleural fluid drained was 1100 cc (range 700–2000 cc). The chest tube was left in drainage for a median time of 1.5 days ranging from 1 to 4 days (Table 1). Complications were not observed during or after the procedure, except for a mild temperature rise (up to 37.5oC) in two patients and mild pain during biopsies in three patients. The chest tube was removed, and patients were dismissed the same day.
Figure 1. (A) Pleural plaque with pleural thickening in patient #5. (B) Scattered nodules in patient #2. (C) Biopsy of the parietal pleura showing dense infiltrates with eosinophils (magnification ×20).
478 4
The The Clinical Clinical Respiratory Respiratory Journal Journal (2015) (2014) •• ISSN ISSN 1752-6981 1752-6981 C 2014 John Wiley & Sons Ltd V © 2014 John Wiley & Sons Ltd
Archontogeorgis Archontogeorgis et et al. al.
Idiopathic Idiopathic eosinophilic eosinophilic pleural pleural effusion effusion
Table 3. Pathological diagnosis of pleural biopsies Patients
Macroscopic
Microscopic
Adhesions
1 2 3 4 5 6 7 8 9 10
PP, PT PP, PT PP, PT PP, PT DN PP, PT DN PP, PT PP, PT PP, PT
DEI DEI DEI DEI NCG DEI NCG DEI DEI DEI
Yes Yes No Yes No Yes No Yes Yes Yes
PP, pleural plaques; PT, pleural thickening; DEI, diffuse eosinophilic inflammation; DN, diffuse nodule; NCG, non-caseating granulomas.
CCT was performed in all patients after pleuroscopy showing pleural thickening in eight patients, atelectasis of the lower lobe in three patients and absence of specific lung parenchyma lesions.
Follow-up The median follow-up period was 60 months (range 24–102 months). All patients showed total regression of the pleural effusion within the first year of follow-up (range 3–12 months), with no additional treatment.
Discussion We report features of 10 patients presenting with IEPE. After an initial meticulous evaluation by means of the classic non-invasive methods, our patients underwent pleuroscopy for diagnostic purpose, yet their pleural effusion remained unspecified. Our patients had histological confirmation of non-specific cause of their effusion with pleuroscopic biopsies and a very long follow-up period with resolution of their effusion without treatment. These parameters enable us to draw firm conclusion about the benign character of the effusion of our patients. Generally, it was believed that EPEs were associated with benign disease (2, 5), but a recent meta-analysis (3) demonstrated that malignancies were in fact the most frequent cause of EPEs (26%). In our 10 patients with IEPE, we have excluded the possibility of malignancy (13) by obtaining large biopsies from parietal, visceral and diaphragmatic pleura. Also, the hypothesis that IEPE occurs in young patients by some authors (7) is not supported by our findings (Table 1). Discrepancies between the different studies are due to the lack of use of invasive procedures and to the short period of The The Clinical Clinical Respiratory Respiratory Journal Journal (2015) (2014) •• ISSN ISSN 1752-6981 1752-6981
C 2014 John Wiley & Sons Ltd V © 2014 John Wiley & Sons Ltd
follow-up (3). Indeed, most of the authors reported results after initial non-invasive negative workup. Only one study (6) mentioned the use of video-assisted thoracoscopic surgery in four patients, yet with no details reported. In patients with undiagnosed pleural effusion, nonspecific pleuritis is an issue, although pleuroscopy offers 91% sensitivity and 100% specificity, with a positive predictive value of 100% and a negative predictive value of 92% (11). Indeed, diagnostic errors in patients with non-specific pleuritis are the result of inadequate exploration of the pleural cavity because of significant adhesions and/or symphysis, lead to sampling errors (11). In such cases, difficulties in exploring the pleural cavity, together with asbestos exposure and suspicion of malignancy may lead to repeat thoracoscopy and/or refer the patient to a thoracic surgeon, if the effusion remains unresolved during the follow-up period (11). In case of unfit patient, or patient’s denial for further investigation, clinical and radiological follow-up is mandatory (11, 12). In our series, we were able to explore the whole of the pleural cavity and to take large biopsies after cutting the adhesions. However, non-specific pleuritis after pleuroscopy assessment does not necessarily mean ‘idiopathic’ pleuritis (12), especially in patients with asbestos exposure, in which mesothelioma must be ruled out (11) (12). Considering that IEPE is in fact a non-specific pleuritis with the presence of eosinophils, it is evident that pleural puncture together with follow-up are not sufficient for final diagnosis. Therefore, pleuroscopy is mandatory in this patient population, and the term of ‘idiopathic’ should be replaced by ‘indeterminate’, as in case of non-specific pleuritis (12). In some patients, macroscopic findings of pleural plaques were associated with pleural thickening. Therefore, we took several and large biopsies from the pleura neighboring the plaques to diagnose mesothelioma, as suggested by Boutin and Rey (14). We may rule out mesothelioma because histology was negative for mesothelioma and the long follow-up period showed no relapse of the effusions. The hallmark of asbestos exposure is pleural plaques. Asbestos is a wellrecognized cause of benign, self-limiting pleural effusion and the pleural histology in our case might be consistent with this condition (14). The association of macroscopic nodules to microscopic non-caseating granulomas (two patients) may be consistent with many granulomatous disorders including tuberculosis (15) or sarcoidosis (16). Tuberculosis was excluded not only with our initial workup (17), but also with pleuroscopy, because its diagnostic 4795
Idiopathic pleural effusion effusion Idiopathic eosinophilic eosinophilic pleural
accuracy for tuberculosis is 100% (17, 18). Pleural involvement in sarcoidosis accounts for less than 5% (19). High-resolution computed tomography of the chest showed no parenchymal involvement or other elements sustaining the diagnosis of sarcoidosis or other granulomatous disorders. Blood tests including serum angiotensin-converting enzyme and calcium levels were unremarkable, and bronchoalveolar lavage CD4+ to CD8+ ratio (20) was non-specific. To date, diagnostic criteria for sarcoidosis are not fulfilled for both patients. The presence of ‘sarcoid-like’ reactions in such patients may indicate a local host tissue response to a hypothetical causal agent manifested as a T-cell-mediated immune response (21). Few data exist regarding follow-up in IEPE: only four studies (4, 6, 8, 22) reported a much shorter period of follow-up ranging from 6 (22) to 21 (8) months. The median follow-up period after pleuroscopy in our series was 60 months (Table 1), which is the longest ever reported, giving the possibility to draw more reliable conclusions. The spontaneous fluid absorption, with no relapse during the follow-up period, together with the histology findings from pleuroscopic biopsies are suggestive of a benign cause. To conclude, EPEs with negative initial workup should undergo pleuroscopy with thorough investigation of the pleural cavity and large biopsies to exclude a specific cause before being characterized as ‘idiopathic’. These effusions may occur in any age and may follow a benign course with spontaneous resolution within a year. According to the findings of our study, we believe that these effusions should be better called ‘indeterminate’ than ‘idiopathic’.
References 1. Froudarakis ME. Diagnostic work-up of pleural effusions. Respiration. 2008;75: 4–13. 2. Rubins JB, Rubins HB. Etiology and prognostic significance of eosinophilic pleural effusions. A prospective study. Chest. 1996;110: 1271–4. 3. Oba Y, Abu-Salah T. The prevalence and diagnostic significance of eosinophilic pleural effusions: a metaanalysis and systematic review. Respiration. 2012;83: 198–208. 4. Martinez-Garcia MA, Cases-Viedma E, Cordero-Rodriguez PJ, et al. Diagnostic utility of eosinophils in the pleural fluid. Eur Respir J. 2000;15: 166–9.
480 6
Archontogeorgis Archontogeorgis et et al. al.
5. Adelman M, Albelda SM, Gottlieb J, Haponik EF. Diagnostic utility of pleural fluid eosinophilia. Am J Med. 1984;77: 915–20. 6. Krenke R, Nasilowski J, Korczynski P, et al. Incidence and aetiology of eosinophilic pleural effusion. Eur Respir J. 2009;34: 1111–17. 7. Kalomenidis I, Light RW. Eosinophilic pleural effusions. Curr Opin Pulm Med. 2003;9: 254–60. 8. Matthai SM, Kini U. Diagnostic value of eosinophils in pleural effusion: a prospective study of 26 cases. Diagn Cytopathol. 2003;28: 96–9. 9. Pettersson T, Riska H. Diagnostic value of total and differential leukocyte counts in pleural effusions. Acta Med Scand. 1981;210: 129–35. 10. Rodriguez-Panadero F. Medical thoracoscopy. Respiration. 2008;76: 363–72. 11. Janssen JP, Ramlal S, Mravunac M. The long-term follow up of exudative pleural effusion after nondiagnostic thoracoscopy. J Bronchol. 2004;11: 169–74. 12. Venekamp LN, Velkeniers B, Noppen M. Does ‘idiopathic pleuritis’ exist? Natural history of non-specific pleuritis diagnosed after thoracoscopy. Respiration. 2005;72: 74–8. 13. Kuhn M, Fitting JW, Leuenberger P. Probability of malignancy in pleural fluid eosinophilia. Chest. 1989;96: 992–4. 14. Boutin C, Rey F. Thoracoscopy in pleural malignant mesothelioma: a prospective study of 188 consecutive patients. Part 1: diagnosis. Cancer. 1993;72: 389–93. 15. Ozkara SK, Turan G, Basyigit I. Clinicopathologic significance of eosinophilic pleural effusions in a population with a high prevalence of tuberculosis and cancer. Acta Cytol. 2007;51: 773–81. 16. Myers JL, Tazelaar HD. Challenges in pulmonary fibrosis: 6 – problematic granulomatous lung disease. Thorax. 2008;63: 78–84. 17. Porcel JM. Tuberculous pleural effusion. Lung. 2009;187: 263–70. 18. Diacon AH, Van de Wal BW, Wyser C, et al. Diagnostic tools in tuberculous pleurisy: a direct comparative study. Eur Respir J. 2003;22: 589–91. 19. Huggins JT, Doelken P, Sahn SA, King L, Judson MA. Pleural effusions in a series of 181 outpatients with sarcoidosis. Chest. 2006;129: 1599–604. 20. Winterbauer RH, Lammert J, Selland M, et al. Bronchoalveolar lavage cell populations in the diagnosis of sarcoidosis. Chest. 1993;104: 352–61. 21. Agostini C, Facco M, Chilosi M, Semenzato G. Alveolar macrophage-T cell interactions during Th1-type sarcoid inflammation. Microsc Res Tech. 2001;53: 278–87. 22. Riantawan P, Bangpattanasiri K, Chaowalit P, Sangsayan P. Etiology and clinical implications of eosinophilic pleural effusions. Southeast Asian J Trop Med Public Health. 1998;29: 655–9.
The The Clinical Clinical Respiratory Respiratory Journal Journal (2015) (2014) •• ISSN ISSN 1752-6981 1752-6981 C 2014 John Wiley & Sons Ltd V © 2014 John Wiley & Sons Ltd
ORIGINAL ARTICLE
Pleuroscopy in ‘Idiopathic’ eosinophilic pleural effusions Kostas Archontogeorgis1, Stavros Anevlavis1, Paul Zarogoulidis1, Ajay Jain1, Georgia Karpathiou2, Alexandra Giatromanolaki2, Efthimios Sivridis2, Demosthenes Bouros1 and Marios E. Froudarakis1 1 Department of Pneumonology, Medical School, Democritus University of Thrace, Alexandroupolis, Greece 2 Department Pathology, Medical School, Democritus University of Thrace, Alexandroupolis, Greece
Abstract Background: Idiopathic eosinophilic pleural effusions (IEPEs) comprise the eosinophilic pleural effusions for which a specific aetiology cannot be established. There are no reports investigating IEPE on the basis of a systematically applied pleuroscopy approach and entailing an appropriate patient follow-up till the final outcome is established; existing series rather combine clinical and thoracocentesis criteria to establish the idiopathic character of the diagnosis. Objectives: The aim of our study was to assess the clinical outcome of patients with IEPE, who underwent a systematic diagnostic approach by pleuroscopy. Methods: We studied 10 patients with IEPE among 175 consecutive patients who underwent pleuroscopy for undiagnosed pleural effusion. Pleural biopsies were obtained from observed lesions. All patients were followed up by means of clinical examination and imaging. Results: The diagnosis of IEPE was established in 10 patients (median age was 50.5 years, range 35–91). Macroscopic examination of the pleura showed diffuse thickening with pleural plaques in eight patients, consistent with diffuse pleural eosinophilic inflammation histologically proven. In two patients, macroscopic examination showed scattered nodules associated with non-caseating granulomas histologically. In all 10 patients, a specific aetiology could not be established. Follow-up was available for all patients ranging from 24–102 months (median 60 months). No patient received a specific treatment during the follow-up period. No relapse of a pleural effusion was documented during this period. Conclusion: Pleuroscopy is mandatory in diagnosing IEPE. Negative histology and a long follow-up showed a benign course. These findings suggest that we should call these effusions ‘indeterminate’. Please cite this paper as: Archontogeorgis K, Anevlavis Anevlavis S, S, Zarogoulidis Zarogoulidis P, P, Jain Jain A, A, Karpathiou G, Giatromanolaki A, Sivridis E, Bouros D and Froudarakis Karpathiou Bouros D and Froudarakis ME. ME. Pleuroscopy in ‘Idiopathic’ eosinophilic Pleuroscopy eosinophilic pleural pleural effusions. effusions. Clin Clin Respir Respir JJ 2014; 2015; ••: 9: ••–••. DOI:10.1111/crj.12165. 475–480. DOI:10.1111/crj.12165.
Key words eosinophilic – idiopathic – non-specific – pleural effusion – pleuroscopy – thoracoscopy Correspondence Marios E. Froudarakis, MD, PhD, Department of Pneumonology, Medical School, Democritus University of Thrace, 68100 Alexandroupolis, Greece. Tel: +302551075335 Fax: +302551030343 email: [email protected] Received: 06 January 2013 Revision requested: 02 May 2014 Accepted: 16 May 2014 DOI:10.1111/crj.12165 Authorship and contributorship MF designed the investigation. KA, MF, SA, PZ, AJ performed pleuroscopy procedures and attended the patients. GK, AG and ES examined and reviewed histological specimen. KA, AJ, SA, DB collected patients’ data. MF and DB reviewed data. KA wrote the manuscript, which was then reviewed and approved by all authors. Ethics This retrospective study has been approved by the Internal Review Board of our hospital. Conflict of interest The authors have stated explicitly that there are no conflicts of interest in connection with this article. Dr. Ajay Jain was fellow during 1 month (October 2011) of the Interventional Unit of the Department of Pneumonology, University Hospital of Alexandroupolis, Greece. His current affiliation is: Gujarat Pulmonary and Critical Care Clinic, Navrangpura, Ahmedabad-380009, India
The Clinical Clinical Respiratory Respiratory Journal Journal (2015) (2014) •• ISSN ISSN 1752-6981 1752-6981 C 2014 John Wiley & Sons Ltd V © 2014 John Wiley & Sons Ltd
4751
Idiopathic pleural effusion effusion Idiopathic eosinophilic eosinophilic pleural
Introduction Pleural effusion is a common finding in pulmonary medicine and although a simple radiographic test is often enough to establish its presence, reaching a specific diagnosis can prove difficult (1). A pleural effusion is defined as eosinophilic if it contains 10% or more eosinophils (2). The incidence of eosinophilic pleural effusion (EPE) is approximately 10% (3). Whether EPE should be considered benign per se or because of a non-benign disease is under debate (2, 4, 5). Even though pleural fluid eosinophilia has been associated with benign disorders (5), a recent metaanalysis did not confirm this perception (2). In fact, pleural eosinophilia may be present in various occasions. The presence of air or blood in the pleural space (repeated thoracentesis, hemothorax, pulmonary embolism, pneumothorax) may be the primary cause (6). Malignancies, drugs, asbestos exposure, infections, Churg–Strauss syndrome and rheumatoid arthritis are also causes of EPE (7). When an etiological diagnosis cannot be established, EPE is referred as ‘idiopathic’. The incidence of idiopathic EPE (IEPE) is quite large according to the different studies ranging between 3.8% and 46% (8, 9). This great difference in incidence is due to the fact that generally, these patients do not undergo an invasive procedure to establish tissue diagnosis, and the term IEPE is applied based only on pleural fluid eosinophilia in the absence of evident cause. Pleuroscopy is a useful tool in the diagnosis and treatment of pleural disease by offering a direct visualization of the pleural space to obtain tissue biopsies (10). Diagnostic accuracy of pleuroscopy is high, reaching 95% in patients with malignancy (1, 10). However, in few cases of pleural effusion of indeterminate cause, even after pleuroscopic study, the etiology remains unclear, as histology of pleural biopsies revealing non-specific pleuritis (11, 12). Therefore, histological diagnosis of non-specific pleuritis does not necessarily correspond to a clinical diagnosis of idiopathic pleuritis (12). Thus, the aim of our study is to report a study of patients with IEPE who underwent pleuroscopy as a diagnostic approach, along with macroscopic, microscopic findings and outcome.
Patients and methods Patients A total of 175 consecutive patients who underwent initial negative evaluation for pleural effusion of 476 2
Archontogeorgis Archontogeorgis et et al. al.
unknown origin were considered for the study from January 2003 until December 2010. All underwent pleuroscopy to establish diagnosis. Pleural effusion was considered eosinophilic when the percentage of eosinophils was >10%. Effusion was classified as IEPE, if no etiology could be established after pleuroscopy. Patients with EPE before pleuroscopy were 21 (12%). From those 21 patients, in 11 (52.4%), a definite etiology was established after pleuroscopy and therefore excluded from the study: nine patients had a malignant cause, and two had a benign one. The remaining 10 undiagnosed after pleuroscopy patients (47.6%) were included in our study. Overall, from the 175 patients with undiagnosed pleural effusion who underwent thoracoscopy for diagnostic purposes, finally 129 (73.7%) were diagnosed with malignancy (including the nine EPE), 36 (20.6%) have shown a true benign disease (including the two EPE) and 10 (5.7%) patients had to be followed for IEPE. This retrospective study has been approved by the Internal Review Board of our hospital.
Methods Initial workup (1) Pleural fluid was obtained by a single-needle thoracentesis from the pleural cavity. A cytological examination was performed to all samples. Biochemical examinations included pH, lactate dehydrogenase (LDH), total protein, albumin and glucose. Pleural fluid red blood cells and nucleated cells counts were performed by an automated method, and the differential nucleated count was established by manual counting. Tests on pleural fluid were performed, including stain and cultures for common bacteria, mycobacteria, opportunistic germs, fungi and parasites. Pleural fluid polymerase chain reaction for Mycobacterium tuberculosis was also performed. Biochemical examinations of peripheral blood including LDH, total protein, albumin and glucose were performed to all patients. Eight patients with a documented risk for malignancy (all heavy smokers) underwent bronchoscopy; microbiology tests (including common bacteria and mycobacteria), cytological examination, as well as total cell count and differential count from bronchoalveolar lavage were obtained. For all patients, the authors did not identify any drug, autoimmune disease, recent chest trauma or intra-abdominal cause, responsible for EPE. No evidence of parasite infection, especially hydatid cyst, was seen to different imaging examinations performed or to the blood serum, and our patients had no history of exposure. Chest X-ray was
The The Clinical Clinical Respiratory Respiratory Journal Journal (2015) (2014) •• ISSN ISSN 1752-6981 1752-6981 C 2014 John Wiley & Sons Ltd V © 2014 John Wiley & Sons Ltd
Archontogeorgis Archontogeorgis et et al. al.
Idiopathic Idiopathic eosinophilic eosinophilic pleural pleural effusion effusion
performed before pleuroscopy along with chest ultrasound, but chest computed tomography (CCT) was performed after evacuation of the pleural fluid. An abdominal CT was also performed to rule out any abdominal causes.
Follow-up All patients were followed at 1 month after pleuroscopy, every 3 months during the first year and then every 6 months by clinical examination, with chest radiography, chest ultrasound and/or CCT.
Pleuroscopy
Results
Pleuroscopy was performed with rigid instruments (Richard Wolf®, Knittlingen, Germany) in the bronchoscopy suite, the patient being in the lateral decubitus position, using single port of entry, under local anesthesia with 1% lignocaine. Patients were under cardiovascular and oxygen saturation monitoring during the procedure. Morphine 0.5 mg was administered as needed. A 7-mm trocar was inserted into the fifth or sixth intercostal space along the midaxillary line. After evacuation of the pleural fluid, an Oo optical telescope was inserted and connected to a light source. The entire pleural cavity was then inspected. Large biopsies were taken from lesions observed from the parietal, the visceral and the diaphragmatic pleura, with optical forceps through the single port of entry, under direct visualization. In presence of pleural plaques, biopsies were obtained from the vicinity of the plaques. Cultures from biopsy specimens for Mycobacterium tuberculosis and other microorganisms were also obtained. Closed intercostal drainage was carried out through a chest tube attached to a water-sealed system, and total drainage was recorded. The chest tube was removed after the procedure by verifying the lung re-expansion by chest X-ray. All complications were recorded, as well as duration of drainage and length of hospital stay.
Patients Among 10 IEPE patients (median age was 50.5 years ranging from 35 to 91), five presented with a left and five with a right pleural effusion. Demographic characteristics are shown in Table 1. None of our patients was immunocompromised. All pleural effusions were characterized as exudates according to Light’s criteria, with a lymphocytic predominance accompanying the eosinophilia. Eosinophilic count ranged from 10% to 59% with a median of 19% (Table 2).
Initial workup Cytological examination of pleural fluid was negative for all patients. Pleural fluid cultures for Mycobacterium tuberculosis, other common microorganisms and fungal infections were negative. Polymerase chain reaction for Mycobacterium tuberculosis of the pleural fluid was also negative for all patients. Common cultures from pleural biopsies were negative. None of the patients presented with peripheral blood eosinophilia. Chest X-ray upon admission showed no additional findings. Chest ultrasound demonstrated a nonseptated pleural effusion in six patients, while fibrinous septa were observed in four. Bronchoscopy did not
Table 1. Patients’ demographics and characteristics Patients
Age
Sex
Side of PE
Bronchoscopy
1 2 3 4 5 6 7 8 9 10 Median
35 39 41 59 84 91 44 38 69 57 50.5
Male Male Male Female Male Female Female Male Female Male
Left Right Right Left Left Right Right Left Left Right
Yes Yes Yes No Yes No Yes Yes Yes Yes
Follow-up (months)
Days of drainage
Total fluid
Length of stay
102 96 72 90 48 24 27 72 24 26 60
1 1 1 1 3 4 1 2 2 3 1.5
1500 700 1200 1000 850 1300 750 2000 800 2000 1100
4 12 1 2 6 7 2 5 2 3 3.5
PE, pleural effusion.
The The Clinical Clinical Respiratory Respiratory Journal Journal (2015) (2014) •• ISSN ISSN 1752-6981 1752-6981
C 2014 John Wiley & Sons Ltd V © 2014 John Wiley & Sons Ltd
4773
Idiopathic pleural effusion effusion Idiopathic eosinophilic eosinophilic pleural
Archontogeorgis Archontogeorgis et et al. al.
Table 2. Pleural fluid characteristics Patients
pH
Glucose
Proteins
Albumin
LDH
Total cells
Neutrophils %
Eosinophils %
Lymphocytes %
1 2 3 4 5 6 7 8 9 10 Median
7.37 7.37 7.38 7.30 7.31 7.38 7.34 7.39 7.29 7.41 7.37
86 31 71 98 206 94 107 145 20 83 90
4.6 4.2 4.2 5.3 4.7 3.4 5.5 4.1 5.0 5.0 4.65
3.3 2.2 2.8 3.1 2.7 3.2 2.9 3.0 3.1 3.2 3.1
266 370 415 327 168 329 522 433 1509 521 392.5
4900 1500 1600 3100 600 600 2300 1700 1700 4500 1700
8.8 21 10 38 15 2.7 30 15 18 5 15
59 15 48 10 10 16 11 23 45 40 19
22 54 34 52 60 81 45 61 35 45 48.5
LDH, lactate dehydrogenase.
reveal endobronchial lesions in any of the patients and the cytological examination and cultures of bronchial lavage for common bacteria; M. tuberculosis and mycosis were negative. Total cell count and differential cell count of bronchoalveolar fluid did not reveal any specific pattern. Smear cultures for M. tuberculosis, common microorganisms and cytological examination were negative. None of the patients presented with acute febrile illness.
Pleuroscopy During pleuroscopy, few adhesions were present in seven patients, while three had multiple adhesions. However, we were able to explore the entire pleural cavity after cutting the adhesions, in all patients. Macroscopic examination of the pleura demonstrated diffuse thickening associated to asbestos pleural plaques in eight patients (Fig. 1A) and diffuse multiple nodules in two (Fig. 1B). Patients had a median of 19 biopsies (range 15–26) on parietal pleura, while four patients (40%) had a median of 12 biopsies (range
10–15) on lesions identified on visceral pleura. Two patients had lesions on diaphragmatic pleura that have been biopsied (one patient had three biopsies, the second had five). Microscopical examination of pleural biopsies documented non-specific inflammation with eosinophilic predominance (Fig. 1C) in eight patients and non-caseating granulomas in two. Specifically, an association between pleural plaques, pleural thickening and the presence of diffuse eosinophilic inflammation was noted, as well as between scattered nodules and non-caseating granulomas. Macroscopic and microscopic features are summarized in Table 3. The median quantity of pleural fluid drained was 1100 cc (range 700–2000 cc). The chest tube was left in drainage for a median time of 1.5 days ranging from 1 to 4 days (Table 1). Complications were not observed during or after the procedure, except for a mild temperature rise (up to 37.5oC) in two patients and mild pain during biopsies in three patients. The chest tube was removed, and patients were dismissed the same day.
Figure 1. (A) Pleural plaque with pleural thickening in patient #5. (B) Scattered nodules in patient #2. (C) Biopsy of the parietal pleura showing dense infiltrates with eosinophils (magnification ×20).
478 4
The The Clinical Clinical Respiratory Respiratory Journal Journal (2015) (2014) •• ISSN ISSN 1752-6981 1752-6981 C 2014 John Wiley & Sons Ltd V © 2014 John Wiley & Sons Ltd
Archontogeorgis Archontogeorgis et et al. al.
Idiopathic Idiopathic eosinophilic eosinophilic pleural pleural effusion effusion
Table 3. Pathological diagnosis of pleural biopsies Patients
Macroscopic
Microscopic
Adhesions
1 2 3 4 5 6 7 8 9 10
PP, PT PP, PT PP, PT PP, PT DN PP, PT DN PP, PT PP, PT PP, PT
DEI DEI DEI DEI NCG DEI NCG DEI DEI DEI
Yes Yes No Yes No Yes No Yes Yes Yes
PP, pleural plaques; PT, pleural thickening; DEI, diffuse eosinophilic inflammation; DN, diffuse nodule; NCG, non-caseating granulomas.
CCT was performed in all patients after pleuroscopy showing pleural thickening in eight patients, atelectasis of the lower lobe in three patients and absence of specific lung parenchyma lesions.
Follow-up The median follow-up period was 60 months (range 24–102 months). All patients showed total regression of the pleural effusion within the first year of follow-up (range 3–12 months), with no additional treatment.
Discussion We report features of 10 patients presenting with IEPE. After an initial meticulous evaluation by means of the classic non-invasive methods, our patients underwent pleuroscopy for diagnostic purpose, yet their pleural effusion remained unspecified. Our patients had histological confirmation of non-specific cause of their effusion with pleuroscopic biopsies and a very long follow-up period with resolution of their effusion without treatment. These parameters enable us to draw firm conclusion about the benign character of the effusion of our patients. Generally, it was believed that EPEs were associated with benign disease (2, 5), but a recent meta-analysis (3) demonstrated that malignancies were in fact the most frequent cause of EPEs (26%). In our 10 patients with IEPE, we have excluded the possibility of malignancy (13) by obtaining large biopsies from parietal, visceral and diaphragmatic pleura. Also, the hypothesis that IEPE occurs in young patients by some authors (7) is not supported by our findings (Table 1). Discrepancies between the different studies are due to the lack of use of invasive procedures and to the short period of The The Clinical Clinical Respiratory Respiratory Journal Journal (2015) (2014) •• ISSN ISSN 1752-6981 1752-6981
C 2014 John Wiley & Sons Ltd V © 2014 John Wiley & Sons Ltd
follow-up (3). Indeed, most of the authors reported results after initial non-invasive negative workup. Only one study (6) mentioned the use of video-assisted thoracoscopic surgery in four patients, yet with no details reported. In patients with undiagnosed pleural effusion, nonspecific pleuritis is an issue, although pleuroscopy offers 91% sensitivity and 100% specificity, with a positive predictive value of 100% and a negative predictive value of 92% (11). Indeed, diagnostic errors in patients with non-specific pleuritis are the result of inadequate exploration of the pleural cavity because of significant adhesions and/or symphysis, lead to sampling errors (11). In such cases, difficulties in exploring the pleural cavity, together with asbestos exposure and suspicion of malignancy may lead to repeat thoracoscopy and/or refer the patient to a thoracic surgeon, if the effusion remains unresolved during the follow-up period (11). In case of unfit patient, or patient’s denial for further investigation, clinical and radiological follow-up is mandatory (11, 12). In our series, we were able to explore the whole of the pleural cavity and to take large biopsies after cutting the adhesions. However, non-specific pleuritis after pleuroscopy assessment does not necessarily mean ‘idiopathic’ pleuritis (12), especially in patients with asbestos exposure, in which mesothelioma must be ruled out (11) (12). Considering that IEPE is in fact a non-specific pleuritis with the presence of eosinophils, it is evident that pleural puncture together with follow-up are not sufficient for final diagnosis. Therefore, pleuroscopy is mandatory in this patient population, and the term of ‘idiopathic’ should be replaced by ‘indeterminate’, as in case of non-specific pleuritis (12). In some patients, macroscopic findings of pleural plaques were associated with pleural thickening. Therefore, we took several and large biopsies from the pleura neighboring the plaques to diagnose mesothelioma, as suggested by Boutin and Rey (14). We may rule out mesothelioma because histology was negative for mesothelioma and the long follow-up period showed no relapse of the effusions. The hallmark of asbestos exposure is pleural plaques. Asbestos is a wellrecognized cause of benign, self-limiting pleural effusion and the pleural histology in our case might be consistent with this condition (14). The association of macroscopic nodules to microscopic non-caseating granulomas (two patients) may be consistent with many granulomatous disorders including tuberculosis (15) or sarcoidosis (16). Tuberculosis was excluded not only with our initial workup (17), but also with pleuroscopy, because its diagnostic 4795
Idiopathic pleural effusion effusion Idiopathic eosinophilic eosinophilic pleural
accuracy for tuberculosis is 100% (17, 18). Pleural involvement in sarcoidosis accounts for less than 5% (19). High-resolution computed tomography of the chest showed no parenchymal involvement or other elements sustaining the diagnosis of sarcoidosis or other granulomatous disorders. Blood tests including serum angiotensin-converting enzyme and calcium levels were unremarkable, and bronchoalveolar lavage CD4+ to CD8+ ratio (20) was non-specific. To date, diagnostic criteria for sarcoidosis are not fulfilled for both patients. The presence of ‘sarcoid-like’ reactions in such patients may indicate a local host tissue response to a hypothetical causal agent manifested as a T-cell-mediated immune response (21). Few data exist regarding follow-up in IEPE: only four studies (4, 6, 8, 22) reported a much shorter period of follow-up ranging from 6 (22) to 21 (8) months. The median follow-up period after pleuroscopy in our series was 60 months (Table 1), which is the longest ever reported, giving the possibility to draw more reliable conclusions. The spontaneous fluid absorption, with no relapse during the follow-up period, together with the histology findings from pleuroscopic biopsies are suggestive of a benign cause. To conclude, EPEs with negative initial workup should undergo pleuroscopy with thorough investigation of the pleural cavity and large biopsies to exclude a specific cause before being characterized as ‘idiopathic’. These effusions may occur in any age and may follow a benign course with spontaneous resolution within a year. According to the findings of our study, we believe that these effusions should be better called ‘indeterminate’ than ‘idiopathic’.
References 1. Froudarakis ME. Diagnostic work-up of pleural effusions. Respiration. 2008;75: 4–13. 2. Rubins JB, Rubins HB. Etiology and prognostic significance of eosinophilic pleural effusions. A prospective study. Chest. 1996;110: 1271–4. 3. Oba Y, Abu-Salah T. The prevalence and diagnostic significance of eosinophilic pleural effusions: a metaanalysis and systematic review. Respiration. 2012;83: 198–208. 4. Martinez-Garcia MA, Cases-Viedma E, Cordero-Rodriguez PJ, et al. Diagnostic utility of eosinophils in the pleural fluid. Eur Respir J. 2000;15: 166–9.
480 6
Archontogeorgis Archontogeorgis et et al. al.
5. Adelman M, Albelda SM, Gottlieb J, Haponik EF. Diagnostic utility of pleural fluid eosinophilia. Am J Med. 1984;77: 915–20. 6. Krenke R, Nasilowski J, Korczynski P, et al. Incidence and aetiology of eosinophilic pleural effusion. Eur Respir J. 2009;34: 1111–17. 7. Kalomenidis I, Light RW. Eosinophilic pleural effusions. Curr Opin Pulm Med. 2003;9: 254–60. 8. Matthai SM, Kini U. Diagnostic value of eosinophils in pleural effusion: a prospective study of 26 cases. Diagn Cytopathol. 2003;28: 96–9. 9. Pettersson T, Riska H. Diagnostic value of total and differential leukocyte counts in pleural effusions. Acta Med Scand. 1981;210: 129–35. 10. Rodriguez-Panadero F. Medical thoracoscopy. Respiration. 2008;76: 363–72. 11. Janssen JP, Ramlal S, Mravunac M. The long-term follow up of exudative pleural effusion after nondiagnostic thoracoscopy. J Bronchol. 2004;11: 169–74. 12. Venekamp LN, Velkeniers B, Noppen M. Does ‘idiopathic pleuritis’ exist? Natural history of non-specific pleuritis diagnosed after thoracoscopy. Respiration. 2005;72: 74–8. 13. Kuhn M, Fitting JW, Leuenberger P. Probability of malignancy in pleural fluid eosinophilia. Chest. 1989;96: 992–4. 14. Boutin C, Rey F. Thoracoscopy in pleural malignant mesothelioma: a prospective study of 188 consecutive patients. Part 1: diagnosis. Cancer. 1993;72: 389–93. 15. Ozkara SK, Turan G, Basyigit I. Clinicopathologic significance of eosinophilic pleural effusions in a population with a high prevalence of tuberculosis and cancer. Acta Cytol. 2007;51: 773–81. 16. Myers JL, Tazelaar HD. Challenges in pulmonary fibrosis: 6 – problematic granulomatous lung disease. Thorax. 2008;63: 78–84. 17. Porcel JM. Tuberculous pleural effusion. Lung. 2009;187: 263–70. 18. Diacon AH, Van de Wal BW, Wyser C, et al. Diagnostic tools in tuberculous pleurisy: a direct comparative study. Eur Respir J. 2003;22: 589–91. 19. Huggins JT, Doelken P, Sahn SA, King L, Judson MA. Pleural effusions in a series of 181 outpatients with sarcoidosis. Chest. 2006;129: 1599–604. 20. Winterbauer RH, Lammert J, Selland M, et al. Bronchoalveolar lavage cell populations in the diagnosis of sarcoidosis. Chest. 1993;104: 352–61. 21. Agostini C, Facco M, Chilosi M, Semenzato G. Alveolar macrophage-T cell interactions during Th1-type sarcoid inflammation. Microsc Res Tech. 2001;53: 278–87. 22. Riantawan P, Bangpattanasiri K, Chaowalit P, Sangsayan P. Etiology and clinical implications of eosinophilic pleural effusions. Southeast Asian J Trop Med Public Health. 1998;29: 655–9.
The The Clinical Clinical Respiratory Respiratory Journal Journal (2015) (2014) •• ISSN ISSN 1752-6981 1752-6981 C 2014 John Wiley & Sons Ltd V © 2014 John Wiley & Sons Ltd
