Malignant Pleural Mesothelioma Caused by Environmental Exposure to Asbestos in the Southeast of Turkey: CT Findings in 117 Patients

Clinical Investigations Respiration 2000;67:615 622 Received: August 6, 1999 Accepted after revision: June 27, 2000 Malignant Pleural Mesothelioma Caused by Environmental Exposure to Asbestos in the Southeast of Turkey: CT Findings in 117 Patients A. Şenyı ğı t a H. Bayram a C. Babayı ğı t a F. Topçu a H. Nazaroğlu b A. Bı lı cı a İ.H. Leblebı cı a a Department of Chest Diseases and b Department of Radiodiagnostics, Faculty of Medicine, University of Dicle, Diyarbakır, Turkey For editorial comment see p. 608. Key Words Mesothelioma W Computed tomography W Environmental asbestos exposure Abstract Background and Objectives: Malignant pleural mesothelioma (MPM) is reported to be common in the southeast of Turkey, as a result of environmental asbestos exposure. The aim of this study was to evaluate the computed tomography (CT) features of MPM in patients with a history of asbestos exposure. Methods: The CT scans of 117 patients who had a diagnosis of MPM were retrospectively evaluated. Additionally, CT findings of histologic subtypes were compared. Results: The most common CT findings included pleural effusion (n = 104, 89%), pleural thickening (n = 96, 82%), mediastinal pleural involvement (n = 77, 66%) and interlobar fissural involvement (n = 62, 53%). Histologic subtype analysis was performed in 89 patients; of these, epithelial, sarcomatous and mixed types were identified in 46, 23 and 20 patients, respectively. An analysis of CT findings demonstrated that the involvement of mediastinal pleural (91%), interlobar fissure (87%) and lung parenchyma (48%) was significantly more frequent in sarcomatous type, as compared to epithelial (61% and p! 0.01; 35 and 4%, p! 0.0001, respectively) and mixed types (65% and p! 0.05; 10% and p! 0.0001; 10% and p! 0.01, respectively). Furthermore, there was a significant correlation between pericardial involvement and chest wall involvement (r = 0.42, p! 0.05) in sarcomatous type. Similarly, lymphadenopathy and parenchymal involvement (r = 0.23, p! 0.02), pericardial and chest wall involvement (r = 0.25, p! 0.01), chest wall and interlobar fissural involvement (r = 0.25, p! 0.01) were significantly correlated, when CT findings of all histologic subtypes were combined. Conclusions: These results suggest that although CT findings of MPM vary, they may provide valuable clues to the diagnosis, at least in patients with a history of asbestos exposure. In addition, the presence of extensive lesions may suggest MPM of sarcomatous subtype. Copyright 2000 S. Karger AG, Basel ABC Fax + 41 61 306 12 34 E-Mail karger@karger.ch www.karger.com 2000 S. Karger AG, Basel 0025 7931/00/0676 0615$17.50/0 Accessible online at: www.karger.com/journals/res Abdurrahman Şenyı ğı t Tıp Fakültesi Göğüs Hastalıkları, Diyarbakır (Turkey) Tel. +90 412 2488001/4287 or 4270, Fax +90 412 2488440, E-Mail senyigit@dicle.edu.tr

Introduction Malignant pleural mesothelioma (MPM) is a primary pleural tumor of mesodermal origin [1]. In the etiology, exposure to asbestos fibers is the most frequent cause, though fibrous zeolite (erionite) and radiation are also among the etiologically important factors [2 6]. MPM is endemically present in some regions of Turkey including several provinces in the southeast of the country, as a result of exposure to environmental fibrous minerals [7]. Yazıcıoğlu [8] and Yazıcıoğlu et al. [9] previously reported that people born or living in some districts of Diyarbakır, Elazıg and Şanlıurfa are exposed to soil containing serpentine and amphibole asbestos fibers and talc, which have no economic value. The material containing asbestos is quarried from mountains by men for both local use and sale in neighboring towns. It is used as a whitewash material for the walls and floors of the houses, and this procedure is repeated every year. Thus, householders are repeatedly exposed to asbestos fibers in their environment, from an early age [9, 10]. The diagnosis of pleural mesothelioma is difficult and often delayed for 6 8 months after the initial symptoms [11]. Cytological examination of pleural effusion, percutaneous transthoracic needle biopsy and even diagnostic thoracoscopy are sometimes insufficient for diagnosis [12]. In such circumstances computed tomography (CT), as a noninvasive procedure, may contribute to the diagnosis, although none of the findings are pathognomonic [13]. In this study, we have evaluated CT findings in 117 cases of MPM caused by environmental exposure to asbestos fibers. We have also investigated the frequency of CT findings in patients whose histologic subtype of disease was identified. diagnoses of MPM were pathologically confirmed in all cases, and the histologic subtype, where noted, was included. Thorax CT was performed in all cases after hospitalization, using a Toshiba TCT 600S CT device and Toshiba X vision GX (Otowara City, Naso, Japan). Evaluation was carried out in chest images obtained in 10-mm slices from the apices of the lungs to costophrenic angles. All of the sections were taken in the supine position at the end of the inspirium. Both mediastinal and parenchymal window settings were used. Intravenous iodinated contrast medium was given to the patients with normal respiratory function to determine mediastinal lymph node enlargement and/or the relation of the lesions to adjacent vascular structures. The CT scans were evaluated by a team of 3 pneumologists and 4 radiologists. A conclusion was reached by consensus. On CT scans, pleural thickening was classified as diffuse, mass type and nodular, and the localization of pleural effusion as ipsilateral and bilateral. In cases of different types of pleural thickening in the same individual, each type was noted separately. Dislocation of the mediastinal structures due to pleural lesions was defined as mediastinal shift. Contraction of the involved hemithorax or mediastinal displacement, distortion of the bronchovascular structures, elevation of the ipsilateral hemidiaphragm and compensatory hyperinflation of contralateral lung were evaluated as volume contraction. Thickening of the pleura of 10 mm or less was defined as pleural thickening, and 10 30 mm focal pleural thickening as pleural nodule. Pleura-based soft tissue mass with a width of 30 mm or more was referred to as pleural mass. Transversal or craniocaudal pleural thickening less than 50 mm in length was noted as pleural plaque. Mediastinal lymph nodes were considered pathologically as enlarged, if they were greater than 10 mm in short-axis diameter in the transverse plane. Involvement of interlobar fissures and mediastinal pleura were also noted. Furthermore, an association between CT findings and histologic subtypes (i.e. epithelial, sarcomatous and mixed) was investigated in cases whose histologic subtype of disease was identified. Statistical Analysis Student s t test was applied to determine the significance of any differences between CT findings in subgroups of patients with epithelial, sarcomatous or mixed histologic subtypes. Furthermore, the association between CT findings within histologic subtypes was tested using Pearson s correlation coefficient, and all values of p! 0.05 were considered to be significant. Materials and Methods The CT scans of 117 patients (67 men and 50 women) who were diagnosed as having MPM at the Departments of Chest Diseases and Tuberculosis, and Chest Surgery, Dicle University Research Hospital, Diyarbakır, Turkey, between January 1988 and March 1999 were retrospectively evaluated. All patients had a history of environmental asbestos exposure, and their mean age was 53.2 years with a range of 21 79 years. The diagnosis was made on the basis of pleural tissue obtained using Ramel needle biopsy, cytologic examination and by videoassisted thoracoscopic surgery (VATS) in 96 (82%), 8 (7%), and 13 cases (11%), respectively. Thirteen patients had no pleural effusion; of these, 4 were diagnosed by VATS, 3 by transthoracic needle biopsy, and 6 by CT or ultrasonography-guided Ramel needle biopsy. The Results The evaluation of CT scans obtained from 117 patients demonstrated that the right hemithorax was affected more frequently in 82 patients (70%) as compared to the left hemithorax in 30 patients (26%), and bilateral involvement was found in 5 patients (4%). Most of the patients had pleural effusion (89%), and the effusion was encapsulated in 6 patients. Pleural thickening was present in 96 patients (fig. 1), and mediastinal pleural involvement in 77 patients. Additionally, pleural plaques were present in 21% of our cases; of 18% with hyaline and 3% 616 Respiration 2000;67:615 622 Şenyı ğı t/bayram/babayı ğı t/topçu/ Nazaroğlu/Bı lı cı /Leblebı cı

Fig. 1. Nodular pleural thickening in the lateral right hemithorax. Fig. 2. Minimal pleural thickening with calcification on the medial side of the left hemithorax, and pleural effusion together with calcified pleural plaques in the right hemithorax. Although malignity was not detected in the left pleura, MPM was diagnosed by biopsy obtained using thoracoscopy on the right side. Fig. 3. Irregular pleural thickening, mediastinal pleural involvement, and volume contraction accompanied by posterior costal invasion indicating chest wall involvement in the right hemithorax. Fig. 4. A heterogeneous mass filling the right hemithorax completely with localized anterior-lateral chest wall involvement. The mediastinum is shifted to the left because of the mass. Sarcomatous subtype MPM was diagnosed using closed pleural needle biopsy. with calcified pleural plaques (fig. 2). Atelectasis was detected in 30 cases, and it was rounded in 4 patients and compressional in 26 patients due to pleural effusion. Pericardial involvement was found in 7 cases, while 3 had no pericardial effusion. Ten patients had chest wall involvement; 8 patients demonstrated costal involvement (fig. 3) and 2 vertebral destruction. In 2 patients with chest wall invasion, the hemithorax was found to be completely filled with large tumor masses on the same side (fig. 4). However, CT did not indicate any pleural pathology, but pleural effusion in 3 cases, who subsequently underwent VATS, and the diagnosis was made by biopsies taken from tiny nodular lesions on visceral pleura (fig. 5). CT findings of patients are presented in table 1. Mesothelioma and CT Findings Respiration 2000;67:615 622 617

Fig. 5. A pleural effusion in the right hemithorax without pleural thickening or mass. The diagnosis was made by biopsies taken from tiny nodular lesions on visceral pleura under thoracoscopy. Fig. 6. In the left hemithorax, there was an indication of parenchymal involvement accompanied by nodular pleural thickening in a patient diagnosed as sarcomatous type MPM. Parenchymal involvement was confirmed by thoracotomy. Table 1. CT findings in 117 patients with MPM Radiological findings Patients Pleural effusion 104 89 Ipsilateral 99 85 Bilateral 5 4 Pleural thickening 96 82 Diffuse 63 72 Nodular 24 21 Mass type 9 8 Mediastinal pleural involvement 77 66 Interlobar fissural involvement 62 53 Volume contraction 60 51 Mediastinal shift 51 44 Atelectasis 30 26 Lymphadenopathy 15 13 Pleural plaques Calcified 4 3 Hyaline 21 18 No effusion 13 11 Involvement of lung parenchyma 12 10 Chest wall invasion 10 9 Pericardial involvement 7 6 Normal pleural appearance (only effusion) 3 3 Involvement of Diaphragmatic pleura 3 3 Pneumothorax 1 1 Table 2. The chest X-ray findings in environmental asbestos-related MPM Radiological finding Patients Pleural effusion 101 86 Pleural thickening 39 33 Volume contraction 38 32 Mediastinal shift 34 29 Mediastinal pleural thickening 24 21 Pleural calcification 18 15 Pleural nodule 8 7 Fissural thickening 7 6 Cardiomegaly 3 3 Paratracheal nodule 2 2 Pneumothorax 1 1 The most common chest X-ray findings included pleural effusion (86%), pleural thickening (33%), volume contraction (32%) and mediastinal shift (29%). Additionally, X-ray indicated mediastinal pleural thickening in 21% of the cases, however, this was detected in 9% of cases with massive pleural effusion (table 2). Histologic subtype analysis was performed in 89 patients in total. Of these, epithelial, sarcomatous and mixed types were identified in 46, 23 and 20 patients, respectively. The statistical analysis of CT findings seen in histologic subtypes demonstrated that the mediastinal 618 Respiration 2000;67:615 622 Şenyı ğı t/bayram/babayı ğı t/topçu/ Nazaroğlu/Bı lı cı /Leblebı cı

Table 3. CT findings in different histologic subtypes of MPM CT finding Epithelial (n = 46) Sarcomatous (n = 23) Mixed (n = 20) Pleural thickening Diffuse 34 74 18 78 13 65 Nodular 10 22 4 17 7 35 Mass type 4 9 2 9 1 5 Mediastinal pleural involvement 28 61 b 21 91 13 65 c No effusion 7 15 3 13 3 15 Normal pleura (only effusion) 2 4 0 0 0 0 Lymphadenopathy 3 7 a 6 26 4 20 Volume contraction 24 52 11 48 11 55 Mediastinal shift 19 41 13 57 7 35 Pericardial involvement 2 4 3 13 1 5 Chest wall invasion 1 2 b 5 22 1 5 Interlobar fissural involvement 16 35 b 20 87 2 10 d Involvement of lung parenchyma 2 4 b, e 11 48 2 10 d Number and percent of cases. a p! 0.05 and b p! 0.01 vs. sarcomatous type, c p! 0.05 and d p! 0.01 vs. sarcomatous type, e p! 0.05 vs. mixed type. pleural involvement of tumor was significantly more frequent in sarcomatous type (91%), as compared to epithelial (61% and p! 0.01) and mixed types (65% and p! 0.05). Similarly, the involvement of interlobar fissure and lung parenchyma (fig. 6) was found to be significantly higher in sarcomatous type (87 and 48%, respectively) than in both epithelial (35 and 4%, respectively; p! 0.0001) and mixed types (10% and p! 0.0001 for interlobar fissural involvement, and 10% and p! 0.01 for involvement of lung parenchyma). Furthermore, lymphadenopathy (26% and p! 0.05) and chest wall invasion (22% and p! 0.01) were significantly increased in sarcomatous type, when compared to epithelial type (7 and 2%, respectively). A comparison between CT findings detected in epithelial and mixed types demonstrated that involvement of lung parenchyma was significantly more frequent in mixed type (10% and p! 0.05) as compared to epithelial type (4%). CT findings in all histologic subtypes are shown in table 3. Looking at correlations between CT findings within subtypes we found that pericardial involvement was significantly correlated with chest wall involvement (r = 0.42; p! 0.05) only in sarcomatous type. However, when CT findings in all subtypes were considered, there was a significant correlation between lymphadenopathy and parenchymal involvement (r = 0.23 and p! 0.02), pericardial involvement and chest wall involvement (r = 0.25 and p! 0.01), and between chest wall involvement and interlobar fissural involvement (r = 0.25 and p! 0.01) in all histologic subtypes. Discussion In the present study, we retrospectively analyzed CT findings of 117 patients with MPM caused by environmental asbestos exposure. Furthermore, we investigated whether there was any difference between the frequency of CT findings seen in patient groups with different histologic subtypes of MPM. The evaluation of CT scans demonstrated that pleural effusion, pleural thickening, mediastinal pleural involvement, interlobar fissural involvement and volume contraction were the most common CT findings. Moreover, we found that the involvement of mediastinal pleura, interlobar fissure and lung parenchyma was more common in cases with sarcomatous type as compared to both epithelial and mixed types. Analysis of CT findings in patients whose histologic subtype was determined demonstrated that lymphadenopathy and parenchymal involvement, involvement of pericardium and chest wall, and involvement of chest wall and interlobar fissure were correlated. It has been reported that chest X-ray findings are usually nonspecific and insufficient for diagnosis of MPM Mesothelioma and CT Findings Respiration 2000;67:615 622 619

[14, 15], and that most common radiographic findings include pleural effusion, diffuse irregular pleural thickening, pleural masses and to a lower extent destruction of costae [12, 16]. Nevertheless, chest X-rays may be important, particularly in the presence of pleural thickening and/or lobulated masses. Volume contraction because of coverage of the lung by the fibrous component of the tumor, depression of the ipsilateral shoulder and findings of fixed chest may also be detected by X-ray. However, CT is known to be more accurate and reliable in diagnosing and determining the extent of mesothelioma [17, 18]. Selçuk et al. [3] reported that the involvement of the mediastinal pleura and interlobar fissure was more detectable in CT scans compared to chest X-rays. Similarly, we found that involvement of interlobar fissure and mediastinal pleura was not visible on chest X-rays of patients with massive pleural effusion, and although chest X-rays showed these pathologies in only 6 and 21% of patients, involvement of interlobar fissure and mediastinal pleura were detected in 53 and 66% of the cases by CT. Furthermore, our finding that pleural thickening was detected in 82% of cases compared with 33% by X-ray is in accordance with the report by Adams et al. [11] suggesting that pleural thickening may not be seen because of pleural effusion. Although pleural effusion is frequently present in MPM in 72 100% of cases with MPM [15], it is not a specific finding for MPM [9, 11]. We detected pleural effusion in 89% of our cases. However, in some cases, pleural effusion may be the only finding without prominent pleural mass or thickening [3]. Confirming this, Leung et al. [13] reported pleural effusion as a unique finding of neoplastic involvement in 7.6% of their cases. Hence, it has been proposed not to exclude a possible diagnosis of MPM in the absence of pleural thickening [13,19]. For instance, Maffessanti et al [19] reported malignant pleural effusion in 7 of 12 cases with normally appearing pleura, while Yılmaz et al. [7] detected MPM in 1 of 46 cases with MPM having no pleural thickening. In our study, we diagnosed MPM in 3% of cases who presented pleural effusion without pleural thickening or pleural plaques. These 3 cases were subjected to VATS, and the diagnosis was established by pleural biopsies from tiny nodular lesions which were not visible on CT scans. Thus, we think MPM should be considered in the diagnosis of cases with pleural effusion and a history of asbestos exposure, and further investigation should be carried out, although CT appearance of the pleura is rarely normal initially. CT is also thought to allow a precise characterization of the different types of pleural thickening [20]. Diffuse pleural thickening is a manifestation of asbestos exposure, although its presence is usually less frequent than pleural plaques. When nodular pleural thickening is detected alone, it is difficult to distinguish it from MPM [21]. The presence of regular and slightly thickened pleura with irregular nodular masses is the most prominent feature of MPM, though it is not pathognomonic [22]. We detected nodular pleural thickening in 21% of our cases, while Leung et al. [13] reported 63%. It is possible that this discrepancy was a result of the fact that our cases were in advanced stages at which diffuse pleural thickening had already developed. The interlobar fissural involvement and atelectasis together with pleural thickening are reported to be the earliest findings of MPM detectable by CT [2], and although free or localized pleural effusion and pleural thickening resulting from fibrosis can involve the interlobar fissures, these were reported to cause smooth thickening of the fissures rather than the irregular or nodular thickening produced by MPM [11]. It is also thought difficult to differentiate a sole pleural thickening from effusion, especially when nodularity is absent in the fissure [7, 23]. Overall, interlobar fissural involvement was present in 53% of our cases, which is less than the previously reported 65.5 86% [3, 14, 23]. Ipsilateral volume contraction resulting from the involvement of the pleural compartment is reported to be common in patients with MPM [22]. The fixation of the mediastinum and volume contraction at the involved site were suggested as diagnostic features of MPM [23], but they have, in fact, less value since they may also be present in some benign and malignant diseases of the chest [13, 22]. Several studies reported volume contraction in 42 73% [6, 13, 14, 22, 23]. Similarly, we detected this pathology in 51% of our cases. Pleural plaques are the most common radiological manifestation of asbestos exposure, nevertheless, it may sometimes be difficult to detect this pathology by conventional chest radiography. Baker and Greene reported that only 68% of pleural plaques were visible on chest X-rays, and that they were more prominent on oblique scans [21]. Although calcified pleural plaques are usually regarded as benign lesions [24], MPM may also be considered in differential diagnosis, especially when there is unilateral effusion coexisting [3]. It has been reported that there was evidence of pleural plaques in a substantial number of cases with MPM [25]. We detected pleural plaques in 21% of our cases, 18% with hyaline and 3% with calcified pleu- 620 Respiration 2000;67:615 622 Şenyı ğı t/bayram/babayı ğı t/topçu/ Nazaroğlu/Bı lı cı /Leblebı cı

ral plaques. This is in accordance with the results reported by Kawashima and Libshitz [23] (20%) and Leung et al. [13] (27%), although the latter is slightly higher. We demonstrated that pleural plaques had irregular margins in all patients having pleural plaques, except 1 case. Furthermore, biopsy specimens obtained from the pleural plaques showed malignant degeneration in 2 of 5 cases who underwent thoracotomy. This suggests that the development of irregular plaques in the presence of smooth plaques may not be sufficient evidence for malignant degeneration. In their study comparing the radiological findings of mesothelioma and asbestosis, Rabinowitz et al. [18] reported that irregular plaques alone were not a criterion of malignancy. Furthermore, these authors suggested that a plaque is not a static structure, since active inflammatory changes are associated with pleural asbestosis, and that a plaque may continue to grow and alter its configuration periodically [18]. Calcified pleural plaques were rarely seen, in only 3% of our cases. Similarly, Rabinowitz et al. [18] reported that these plaques were infrequently encountered in mesothelioma either via CT or radiography, and suggested that this could be a result of the absorption of the calcification by the tumor [13]. Diffuse mesothelioma frequently encases the entire lung, although the greatest region of tumor growth is usually the surface of the lower lobes. This tumor generally spreads locally [1], and may be detected as nodular masses which can be classified according to their source of origin as parenchymal extension or intraparenchymal. Parenchymal extension has been reported to contain pleural-based nodular masses extending into lung, whereas intraparenchymal masses were found to be round, sharply demarcated and located within the lung [18]. In our study, we detected lung penetration in 12 patients (10%), and these were pleural-based masses that extended into lung parenchyma. In 11 patients, the attachment to pleura was on a thin pedicle and in 1 case on a broad base. However, we have not detected any intraparenchymal lesions isolated from pleura. The histopathological examination of tissue obtained from these lesions has revealed MPM. Similarly, Rabinowitz et al. [18] reported that the intraparenchymal lesions histologically represented malignant mesothelioma in patients with mesothelioma, and thickened pleura in patients with asbestosis. Although the degree of therapeutic efficacy is reported to be closely related to the histologic structure of the tumor [18], to our knowledge, there are no or limited studies investigating CT features of histologic subtypes of MPM in large series. In the present study, however, the frequency of most CT findings detected in all three subtypes was not different. The only significant differences were that involvement of mediastinal pleura, interlobar fissure and lung parenchyma was more frequent in sarcomatous type compared to both epithelial and mixed types. Additionally, we found that pericardial involvement and lymphadenopathy were less frequent in epithelial type as compared to sarcomatous type, and 2 cases in the epithelial group did not show any pleural pathology. Thus, although the presence of extensive lesions suggested sarcomatous type, there was no significant difference in the localization and characteristic of the tumor (as being diffuse, nodular or mass type). It has been suggested that epithelial mesotheliomas differ from those with sarcomatoid histology with regard to clinical course and presentation. The former are reported to behave similarly to metastatic carcinomas to the pleura in that they present with large pleural effusions and metastases to regional lymph nodes, while the latter are thought to be more similar to pleural sarcomas in presenting with bulky tumor masses, minimal pleural fluid, and blood-borne metastases [25]. Our results have also indicated that there was a correlation between lymphadenopathy and parenchymal involvement, pericardial involvement and chest wall involvement, and between chest wall involvement and interlobar fissural involvement, when CT findings in all histologic subtypes are considered. To our knowledge, there are limited or no studies from the literature reporting such a correlation. Although it is possible that this may be related to the extent of the disease, we think, the significance of such a correlation remains to be determined. In conclusion, our results have demonstrated that although CT findings of MPM vary, they may provide valuable clues to the diagnosis, at least in patients with a history of asbestos exposure. Additionally, the fact that several CT findings such as the involvement of mediastinal pleura, interlobar fissure and lung parenchyma were more frequent in cases with sarcomatous type may be evidence of the invasive character of this tumor. Acknowledgement We thank Dr. Ömer Satıcı from the Biostatistics Department, School of Medicine, University of Dicle for his advice on the statistical analysis of the data. Mesothelioma and CT Findings Respiration 2000;67:615 622 621

References 1 Patz EF Jr, Shaffer K, Piwnica-Worms DR, Jochelson M, Sarin M, Sugarbaker DJ, Pugatch RD: Malignant pleural mesothelioma: Value of CT and MR imaging in predicting resectability. AJR 1992;159:961 966. 2 Aisner J: Current approach to malignant mesothelioma of the pleura. Chest 1995;107 (suppl):332 344. 3 Selçuk ZT, Çöplü L, Emri S, Kalyoncu AF, Şahin AA, Barış Yİ: Malignant pleural mesothelioma due to environmental mineral fiber exposure in Turkey: Analysis of 135 cases. Chest 1992;102:790 796. 4 Nishimura SL, Broaddus VC: Asbestos-induced pleural disease. Clin Chest Med 1998; 19:311 329. 5 Türler A, Mönig SP, Raab M: Problems in diagnosis and therapy of malignant pleural mesothelioma. Med Klin 1997;92:101 105. 6 Şahin AA, Çöplü L, Selçuk ZT, Eryılmaz M, Emri S, Akhan O, Barış Yİ: Malignant pleural mesothelioma caused by environmental exposure to asbestos or erionite in rural Turkey: CT findings in 84 patients. AJR 1993;161:533 537. 7 Yılmaz UM, Utkaner G, Yalnız E, Kumcuoglu Z: Computed tomographic findings of environmental asbestos-related malignant pleural mesothelioma. Respirology 1998;3:33 38. 8 Yazıcıoǧlu S: Pleural calcification associated with exposure to chrysotile asbestos in Southeast Turkey. Chest 1976;70:43 47. 9 Yazıcıoǧlu S, İlçayto R, Balcı K, Saylı BS, Yorulmaz B: Pleural calcification, pleural mesotheliomas, and bronchial cancers caused by tremolite dust. Thorax 1980;35:564 569. 10 Topçu F, Bayram H, Şimşek M, Kaya K, Özcan C, Işık R, Şenyiǧit A: High-resolution computed tomography in cases with environmental exposure to asbestos in Turkey. Respiration 2000;67:139 145. 11 Adams VI, Unni KK, Muhm JR, Jett JR, Ilstrup DM, Bernatz PE: Diffuse malignant mesothelioma of pleura. Diagnosis and survival in 92 cases. Cancer 1986;58:1540 1551. 12 Işık R, Şimşek M, Coşkunsel M, Bükte Y: CT findings in diagnosis of malignant mesothelioma. Tüberküloz ve Toraks 1993;41:45 50. 13 Leung AN, Müller NL, Miller RR: CT in differential diagnosis of diffuse pleural disease. AJR 1990;154:487 492. 14 Erzen C, Eryılmaz M, Kalyoncu F, Bilir N, Şahin A, Barış Yİ: CT findings in malignant pleural mesothelioma related to nonoccupational exposure to asbestos and fibrous zeolite (erionite). J Comput Assist Tomogr 1991;15:256 260. 15 Rusch VW, Godwin JD, Shuman WP: The role of computed tomography scanning in the initial assessment and the follow-up of malignant pleural mesothelioma. J Thorac Cardiovasc Surg 1988;96:171 177. 16 Bilici A, Uyar A, Özateş M: CT features of malignant pleural mesothelioma. Dicle Üniv Tıp Fak Derg 1994;21:35 44. 17 Alexander E, Clark AC, Colley DP, Mitchell SE: CT of malignant pleural mesothelioma. AJR 1981;137:287 291. 18 Rabinowitz JG, Efremidis SC, Cohen B, Dan S, Efremidis A, Chahinian AP, Teirstein AS: A comparative study of mesothelioma and asbestosis using computed tomography and conventional chest radiography. Radiology 1982;144: 453 460. 19 Maffessanti M, Tommasi M, Pellegrini P: Computed tomography of free pleural effusions. Eur J Radiol 1987;7:87 90. 20 Gevenois PA, Trogrlic S: Imaging of the pleura. Rev Prat 1997;47:1304 1307. 21 Miller WT Jr, Gefter WB, Miller WT Sr: Asbestos-related chest diseases: Plain radiographic findings. Semin Roentgenol 1992;27:102 120. 22 Çakmak F, Kuranel BB, Ünsal M, Işık S: Computed tomographic findings in malignant pleural mesothelioma. Solunum Hastalıkları 1993; 4:31 39. 23 Kawashima A, Libshitz HI: Malignant pleural mesothelioma: CT manifestations in 50 cases. AJR 1990;155:965 969. 24 Munden RF, Libshitz HI: Rounded atelectasis and mesothelioma. AJR 1998;170:1519 1522. 25 Albelda SM, Sterman DH, Litzky LA: Malignant mesothelioma and other primary pleural tumors; in Fishman AP (ed): Fishman s Pulmonary Diseases and Disorders. New York, McGraw-Hill, 1998, pp 1453 1466. 622 Respiration 2000;67:615 622 Şenyı ğı t/bayram/babayı ğı t/topçu/ Nazaroğlu/Bı lı cı /Leblebı cı