- Email: [email protected]
Percutaneous core needle biopsy in the diagnosis of mediastinal tumors
Lung Cancer 25 (1999) 169 – 173 www.elsevier.nl/locate/lungcan
Percutaneous core needle biopsy in the diagnosis of mediastinal tumors Joel Greif a,*, Alexander N. Staroselsky a, Moshe Gernjac d, Yehuda Schwarz a, Silvia Marmur b, Marina Perlsman e, Alon Yellin c a
Department of Pulmonary Medicine, Tel-A6i6 Sourasky Medical Center, 6 Weizman Street, Tel-A6i6 64239, Israel b Department of Pathology, Tel-A6i6 Sourasky Medical Center, 6 Weizman Street, Tel-A6i6 64239, Israel 1 c Department of Thoracic Surgery, Chaim Sheba Medical Center, Tel-Hashomer, Israel d Department of Radiology, Chaim Sheba Medical Center, Tel-Hashomer, Israel e Department of Pathology, Chaim Sheba Medical Center, Tel-Hashomer, Israel 1 Received 18 January 1999; received in revised form 27 April 1999; accepted 29 April 1999
Abstract Objective: to determine the contribution of percutaneous core cutting needle biopsy (PCNB) in the diagnosis of mediastinal tumors. Design: retrospective review of 70 patients with mediastinal lesions who underwent CT-guided PCNB between 1988 and 1996. Results: PCNB provided adequate material in 62/70 cases, giving a total sample rate of 88.6%. Of these 62 patients, 57 were diagnosed correctly by PCNB whereas 5/62 were misdiagnosed as nonspecific inflammation, providing an overall sensitivity of 91.9%. PCNB established a specific histologic diagnosis in 90.3% of the patients, mainly in cases of lymphoma, bronchogenic carcinoma, and thymoma. Pneumothorax was the most commonly encountered complication (11%). Hemoptysis (30 – 50ml) occurred in only one (1.6%) of the patients. Conclusion: CT guided PCNB is an easy and safe procedure which can provide a precise diagnosis in the majority of mediastinal tumors and can obviate the need for exploratory thoracic surgery in cases which are medically treatable or non-resectable. © 1999 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Mediastinal lesions; Percutaneous cutting needle biopsy; Pneumothorax
1. Introduction Radiologically guided transthoracic needle biopsy has become an accepted technique in the * Corresponding author. Tel.: + 972-3-6973592; fax: + 9723-6974601. 1 Affiliated with the Sackler Faculty of Medicine, Tel-Aviv University, Tel-Aviv, Israel.
assessment of mediastinal lesions. This method may highlight the need for more invasive diagnostic procedures, such as mediastinoscopy, thoracoscopy, or exploratory thoracotomy [1–3]. Most clinicians use fine needle aspiration (FNA) to provide a sample which is usually adequate for cytological and microbiological examination, but not for a histological specimens for which corecutting needles are used [4]. A larger volume of
0169-5002/99/$ - see front matter © 1999 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 1 6 9 - 5 0 0 2 ( 9 9 ) 0 0 0 5 3 - 7
170
J. Greif et al. / Lung Cancer 25 (1999) 169–173
tissue obtained by percutaneous core needle biopsy (PCNB) allows for more sophisticated laboratory analysis, such as electron microscopy, flow cytometry, immunocytochemistry, and surface tumor markers, all of which increase diagnostic specificity [5–7]. In this study, we report the contribution of the percutaneous core needle biopsy as the initial invasive diagnostic procedure in the management of patients with mediastinal tumors.
2. Material and methods We performed a retrospective analysis of CTguided consecutive biopsies performed for initial evaluation of mediastinal tumors in 70 patients between 1988 and 1996. The age of the patients ranged from 25 to 82 years (mean 55). There were 41 females and 29 males. The lesions presented roentgenologically as solid homogenous masses in 65 (92%) of the cases and as nonhomogeneous infiltrations in 7 (8%). Their diameters ranged from 1 to 10 cm, and most of them (67.9%) were larger than 4 cm. The majority of the lesions (65.4%) were located at a depth of B 2 cm from the skin; the lesions of ‘intermediate’ depth (2 – 6 cm) accounted for 34.6%. Almost all lesions were localized in the anterior mediastinum (95.8%), one (1.4%) was in the middle mediastinum, and two (2.8%) were in the posterior mediastinum. PCNBs were performed using a 18-gauge Temno TruCut automatic cutting needle (Bauer Medical) under CT guidance. This needle is designed to obtain core specimens for histologic analysis through a spring-activated mechanism that fires a cutting cannula which snares a biopsy specimen adequate for histological examination into a 1.7 cm stylet [11]. An average of two PCNB passes were performed and additional specimens were obtained when necessary. After the biopsy, a CT slice through the area of the biopsy was performed to exclude the presence of significant bleeding or pneumothorax. The results of the biopsies were classified as malignant or nonmalignant on the basis of the
presence or absence of malignant cells in the PCNB samples. The biopsy results were later compared with the final diagnosis which was determined by surgical histopathological examination or clinico-radiologic follow-up. Biopsies were considered inadequate if they did not provide sufficient pathological material for diagnostic evaluation (‘necrotic debris’, ‘blood only’), and they were excluded from the final analysis. If a pneumothorax developed after PCNB, it was aspirated with an Inthracath cannula which was replaced by a small chest tube connected to a Heimlich valve when deemed necessary. The pneumothorax usually resolved within a few hours. If the pneumothorax was large or persisted for more than 12 h, a chest tube was connected to a suction device for a few days.
3. Results A total of 70 patients with mediastinal tumors underwent PCNB during the study period. Of these patients, eight (three lymphomas, two thymomas, and three teratomas) were excluded because of inadequate specimen material, leaving a study group of 62 patients (an effective sample rate of 88.6%).The final diagnoses were as follows: lymphoma (n= 30), thymoma (n = 6), bronchogenic carcinoma (n= 16), metastatic (non bronchogenic) carcinoma (n= 7), and one each of teratoma, germ cell tumor, and lymphoepithelial cancer. Regarding malignancy, 57/62 patients were correctly diagnosed by PCNB, giving an overall sensitivity of 92% whereas 5/62 (8%) biopsies were misdiagnosed as nonspecific inflammation. PCNB established the specific histopathological diagnosis in 54/62 (87%) cases. The results of PCNB are presented in Table 1. The final diagnosis was confirmed by surgery in 19 patients (seven patients with lymphoma, four with thymoma, four with teratoma, one bronchogenic cancer, and three miscellaneous). In the other 43 cases, the histopathological diagnosis was accepted as the final one by long-term clinical follow up (from 2 to 6 years) and response to treatment.
J. Greif et al. / Lung Cancer 25 (1999) 169–173
Non Hodgkin’s Lymphoma (NHL); from 21 adequate biopsies, 2 PCNBs were misdiagnosed as being non-specific inflammation (two false negatives). Hodgkin’s Lymphoma (HL); from nine cases examined, one was misdiagnosed as nonspecific inflammation. In another case, the PCNB failed to discriminate between HL and NHL. Subsequent surgical biopsy identified the lesion as HL, nodular sclerosis. Invasive bronchogenic carcinoma; the PCNB provided the correct diagnosis in 15/16 cases of bronchogenic cancer. One case (anaplastic carcinoma) was misdiagnosed by PCNB as being nonspecific inflammation. Metastatic carcinoma; the seven cases of metastatic lesions included two breast, one colon, and one stomach cancer. Three metastatic tumors were of unknown origin. In one case of adenocarcinoma of the stomach, the PCNB showed only fibrotic material (false negative). Thymoma; from six cases with thymoma, one was misdiagnosed by PCNB as lymphoma. A subsequent surgical biopsy provided the correct diagnosis. Teratoma; due to the extensive necrosis that is associated with this tumor, the PCNB did not provide sufficient material for diagnostic examination in three of four cases, and the final diagnosis in these patients was established by surgery.
171
Germ cell tumor; this one case was misdiagnosed as a malignant epithelial tumor. Pneumothorax was the most commonly encountered complication in our study, occurring in seven patients (11.3%); of whom only two (3.2%) required placement of a chest tube and hospitalization after the biopsy. Moderate hemoptysis (30–50ml) occurred in one patient. Only two patients (3.2%) suffered from persistent chest pain after the biopsy procedure and were satisfactorily treated with analgesics.
4. Discussion Mediastinal tumors are traditionally diagnosed by mediastinoscopy, thoracoscopy or exploratory thoracic surgery. These procedures require anesthesia and hospitalization. The purpose of this study was to evaluate the clinical utility of PCNB in the diagnosis of these tumors and to determine if it should be the initial diagnostic procedure in these cases. Between 1988 and 1996, we examined a total of 70 mediastinal tumors. Most of the tumors (89%) were located in the anterior compartment. The most common pathological entities encountered were lymphoprolipherative neoplasms (47%), epithelial cancer (31%), and thymomas (11.4%). These findings are similar to those reported previously in the literature [3–8].
Table 1 PCNB/mediastinal lesions Final diagnosis
Sensitivitya
Histospecific diagnosisb
False-negative diagnosis
Non-Hodgkin’s lymphoma (n=21) Hodgkin’s lymphoma (n =9) Bronchogenic carcinoma (n=16) Thymoma (n = 6) Metastatic carcinoma (n =7) Teratoma (n = 1) Germ cell tumor (n =1) Lymphoepithelial cancer (n= 1)
95.2% 88.9% 93.7% 100% 85.7% 100% 100% 100%
90.2% 77.8% 100% 83.3% 85.7% 100% 100% 100%
2/21 (9.5%) 1/19 (11.1%) 1/16 (6.3%) 1/6 (16.7%) 1/7 (14.3%) 0/1 (0%) 0/1 (0%) 0/1 (0%)
91.9%
90.3%
Total (n = 62) a b
Sensitivity, true positive test results /all patients-regarding malignancy Histospecific diagnosis, correct histological diagnosis by PCNB
9.7%
172
J. Greif et al. / Lung Cancer 25 (1999) 169–173
The knowledge of the nature of mediastinal tumors is of critical importance in order to reach the correct therapeutic decisions. For example, thymomas are primarily treated by surgical resection, whereas lymphomas are treated by irradiation and/or chemotherapy. Patients with metastatic disease need to have their diagnosis established with certainty so that further treatment (if any) can be planned accordingly. Among the 62 adequate biopsies described in this study, PCNB demonstrated a high degree of histospecificity, as we were able to obtain an accurate histological diagnosis in 87% of these patients. PCNB showed a very high specificity in patients with lymphoma (92.5%), Hodgkin’s lymphoma (88.9%), and metastatic disease (85.7%). There were no false positive and only two false negative results in the latter group. Conversely, the diagnostic yield of tumors that tend to become necrotic, such as teratoma, is much lower and the tissue obtained by PCNB was considered to be inadequate for diagnosis in three-quarters of them. Similar results to ours were obtained in smaller series by Morrissey et al. [8] Noppen et al. [9], and Heilo [10]. Pneumothorax was the most frequent complication resulting from PCNB, with an incidence of 11.3%. This rate is similar to the range between 4.6% and 41% reported for pneumothorax induced by transthoracic fine needle biopsies [11,12]. The relative low rate of pneumothorax in spite of the use of a large cutting needle can be explained by the superficial location of 2/3 of the biopsied lesions ( B 2 cm from the chest surface). Other complications such as hemoptysis and pain (1.6 and 3.2%, respectively) are also comparable to reported rates using FNA [13,14]. In six of the 19 operated patients (31%) of whom four had lymphoma and two had adenocarcinoma, thoracic surgery (performed on the early phase of the study) only confirmed the PCNB diagnosis and could have therefore been avoided. In light of the large number of mediastinal entities that are medically treatable, such as lymphomas, or that are clearly non-resectable, such as metastatic carcinoma, it is clear that a
precise histopathological diagnosis is essential. On the basis of our experience, we believe that radiologically guided PCNB is an easy and safe procedure which can provide accurate diagnostic information in the majority of malignant mediastinal tumors and can in many cases highlight the need of more extensive diagnostic surgical procedures such as thoracoscopy or thoracotomy, which require general anesthesia and are associated with higher morbidity and cost. In resectable cases, a definite preoperative diagnosis by transthoracic needle biopsy allows for better surgical planning, especially in patients with a high surgical risk or those requiring induction therapy. In conclusion, we recommend the performance of transthoracic core needle biopsy as the initial diagnostic procedure in the evaluation of mediastinal tumors.
References [1] Adler OB, Rosenberger A, Peleg H. Fine needle aspiration biopsy of mediastinal masses: evaluation of 136 experiences. Am J Radiol 1983;140:893 – 6. [2] Weisbrod GL, Lyons DJ, Tao LC, Charberlain DW. Percutaneous fine needle aspiration biopsy of mediastinal lesions. Am J Radiol 1984;143:525 – 9. [3] Herman SJ, Holub RV, Weisbrod GL, Charberlain DW. Anterior mediastinal masses: utility of thoracic needle biopsy. Radiology 1991;180:167 – 70. [4] Quinn SF, Demlow T, Dunkley B. Temno biopsy needle: evaluation of efficacy and safety in 165 biopsy procedures. Am J Radiol 1992;158:641 – 3. [5] Bocking A, Klose K-C, Kyll HJ, Hauptmann S. Cytologic versus histologic evaluation of needle biopsy of the lung, hilum, and mediastinum. Acta Cytol 1995;35:463 – 71. [6] Ben Yehuda D, Polliack A, Dkon E, Sherman Y, Bebenshart P, Lotan H, Libson E. Image-guided core needle biopsy in malignant lymphoma-experience with 100 patients that suggests the technique is reliable. J Clin Oncol 1996;14:2431 – 4. [7] Klein JS, Salomon G, Stewart EA. Transthoracic needle biopsy with a coaxially placed 20-gauge automated cutting needle: results in 122 patients. Klein Radiol 1996;198:715 – 20. [8] Morrissey B, Adams H, Gibbs AR, Crane MD. Percutaneous needle biopsy of the mediastinum. Thorax 1993;48:632 – 7. [9] Noppen MP, DeMey JD, Meysman M, Opdebeek K, Vincken WG, Osteaux M. Percutaneous needle biopsy of localized pulmonary, mediastinal, and pleural disease tis-
J. Greif et al. / Lung Cancer 25 (1999) 169–173
173
the lungs: its safety and utility. Radiology 1984;151:301 – 3. [13] Lalli AF, McCormack LJ, Zelch M, Reich NE, Belovich D. Aspiration biopsy of chest lesions. Radiology 1978;127:35 – 40. [14] Crosby JH, Hager B, Hoeg K. Transthoracic fine needle aspiration. Cancer 1985;56:3504 – 7.
sue with an automatic disposable guillotine soft-tissue needle. Chest 1995;107:1615–20. [10] Heilo A. Tumors of the mediastinum: US-guided histologic core-needle biopsy. Radiology 1993;189:143–6. [11] Conces DS, Schwenk GR, Doering PR, Glant MD. Thoracic needle biopsy. Chest 1987;91:813–6. [12] Stevens GM, Jackman RJ. Outpatient needle biopsy of
.
Percutaneous core needle biopsy in the diagnosis of mediastinal tumors Joel Greif a,*, Alexander N. Staroselsky a, Moshe Gernjac d, Yehuda Schwarz a, Silvia Marmur b, Marina Perlsman e, Alon Yellin c a
Department of Pulmonary Medicine, Tel-A6i6 Sourasky Medical Center, 6 Weizman Street, Tel-A6i6 64239, Israel b Department of Pathology, Tel-A6i6 Sourasky Medical Center, 6 Weizman Street, Tel-A6i6 64239, Israel 1 c Department of Thoracic Surgery, Chaim Sheba Medical Center, Tel-Hashomer, Israel d Department of Radiology, Chaim Sheba Medical Center, Tel-Hashomer, Israel e Department of Pathology, Chaim Sheba Medical Center, Tel-Hashomer, Israel 1 Received 18 January 1999; received in revised form 27 April 1999; accepted 29 April 1999
Abstract Objective: to determine the contribution of percutaneous core cutting needle biopsy (PCNB) in the diagnosis of mediastinal tumors. Design: retrospective review of 70 patients with mediastinal lesions who underwent CT-guided PCNB between 1988 and 1996. Results: PCNB provided adequate material in 62/70 cases, giving a total sample rate of 88.6%. Of these 62 patients, 57 were diagnosed correctly by PCNB whereas 5/62 were misdiagnosed as nonspecific inflammation, providing an overall sensitivity of 91.9%. PCNB established a specific histologic diagnosis in 90.3% of the patients, mainly in cases of lymphoma, bronchogenic carcinoma, and thymoma. Pneumothorax was the most commonly encountered complication (11%). Hemoptysis (30 – 50ml) occurred in only one (1.6%) of the patients. Conclusion: CT guided PCNB is an easy and safe procedure which can provide a precise diagnosis in the majority of mediastinal tumors and can obviate the need for exploratory thoracic surgery in cases which are medically treatable or non-resectable. © 1999 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Mediastinal lesions; Percutaneous cutting needle biopsy; Pneumothorax
1. Introduction Radiologically guided transthoracic needle biopsy has become an accepted technique in the * Corresponding author. Tel.: + 972-3-6973592; fax: + 9723-6974601. 1 Affiliated with the Sackler Faculty of Medicine, Tel-Aviv University, Tel-Aviv, Israel.
assessment of mediastinal lesions. This method may highlight the need for more invasive diagnostic procedures, such as mediastinoscopy, thoracoscopy, or exploratory thoracotomy [1–3]. Most clinicians use fine needle aspiration (FNA) to provide a sample which is usually adequate for cytological and microbiological examination, but not for a histological specimens for which corecutting needles are used [4]. A larger volume of
0169-5002/99/$ - see front matter © 1999 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 1 6 9 - 5 0 0 2 ( 9 9 ) 0 0 0 5 3 - 7
170
J. Greif et al. / Lung Cancer 25 (1999) 169–173
tissue obtained by percutaneous core needle biopsy (PCNB) allows for more sophisticated laboratory analysis, such as electron microscopy, flow cytometry, immunocytochemistry, and surface tumor markers, all of which increase diagnostic specificity [5–7]. In this study, we report the contribution of the percutaneous core needle biopsy as the initial invasive diagnostic procedure in the management of patients with mediastinal tumors.
2. Material and methods We performed a retrospective analysis of CTguided consecutive biopsies performed for initial evaluation of mediastinal tumors in 70 patients between 1988 and 1996. The age of the patients ranged from 25 to 82 years (mean 55). There were 41 females and 29 males. The lesions presented roentgenologically as solid homogenous masses in 65 (92%) of the cases and as nonhomogeneous infiltrations in 7 (8%). Their diameters ranged from 1 to 10 cm, and most of them (67.9%) were larger than 4 cm. The majority of the lesions (65.4%) were located at a depth of B 2 cm from the skin; the lesions of ‘intermediate’ depth (2 – 6 cm) accounted for 34.6%. Almost all lesions were localized in the anterior mediastinum (95.8%), one (1.4%) was in the middle mediastinum, and two (2.8%) were in the posterior mediastinum. PCNBs were performed using a 18-gauge Temno TruCut automatic cutting needle (Bauer Medical) under CT guidance. This needle is designed to obtain core specimens for histologic analysis through a spring-activated mechanism that fires a cutting cannula which snares a biopsy specimen adequate for histological examination into a 1.7 cm stylet [11]. An average of two PCNB passes were performed and additional specimens were obtained when necessary. After the biopsy, a CT slice through the area of the biopsy was performed to exclude the presence of significant bleeding or pneumothorax. The results of the biopsies were classified as malignant or nonmalignant on the basis of the
presence or absence of malignant cells in the PCNB samples. The biopsy results were later compared with the final diagnosis which was determined by surgical histopathological examination or clinico-radiologic follow-up. Biopsies were considered inadequate if they did not provide sufficient pathological material for diagnostic evaluation (‘necrotic debris’, ‘blood only’), and they were excluded from the final analysis. If a pneumothorax developed after PCNB, it was aspirated with an Inthracath cannula which was replaced by a small chest tube connected to a Heimlich valve when deemed necessary. The pneumothorax usually resolved within a few hours. If the pneumothorax was large or persisted for more than 12 h, a chest tube was connected to a suction device for a few days.
3. Results A total of 70 patients with mediastinal tumors underwent PCNB during the study period. Of these patients, eight (three lymphomas, two thymomas, and three teratomas) were excluded because of inadequate specimen material, leaving a study group of 62 patients (an effective sample rate of 88.6%).The final diagnoses were as follows: lymphoma (n= 30), thymoma (n = 6), bronchogenic carcinoma (n= 16), metastatic (non bronchogenic) carcinoma (n= 7), and one each of teratoma, germ cell tumor, and lymphoepithelial cancer. Regarding malignancy, 57/62 patients were correctly diagnosed by PCNB, giving an overall sensitivity of 92% whereas 5/62 (8%) biopsies were misdiagnosed as nonspecific inflammation. PCNB established the specific histopathological diagnosis in 54/62 (87%) cases. The results of PCNB are presented in Table 1. The final diagnosis was confirmed by surgery in 19 patients (seven patients with lymphoma, four with thymoma, four with teratoma, one bronchogenic cancer, and three miscellaneous). In the other 43 cases, the histopathological diagnosis was accepted as the final one by long-term clinical follow up (from 2 to 6 years) and response to treatment.
J. Greif et al. / Lung Cancer 25 (1999) 169–173
Non Hodgkin’s Lymphoma (NHL); from 21 adequate biopsies, 2 PCNBs were misdiagnosed as being non-specific inflammation (two false negatives). Hodgkin’s Lymphoma (HL); from nine cases examined, one was misdiagnosed as nonspecific inflammation. In another case, the PCNB failed to discriminate between HL and NHL. Subsequent surgical biopsy identified the lesion as HL, nodular sclerosis. Invasive bronchogenic carcinoma; the PCNB provided the correct diagnosis in 15/16 cases of bronchogenic cancer. One case (anaplastic carcinoma) was misdiagnosed by PCNB as being nonspecific inflammation. Metastatic carcinoma; the seven cases of metastatic lesions included two breast, one colon, and one stomach cancer. Three metastatic tumors were of unknown origin. In one case of adenocarcinoma of the stomach, the PCNB showed only fibrotic material (false negative). Thymoma; from six cases with thymoma, one was misdiagnosed by PCNB as lymphoma. A subsequent surgical biopsy provided the correct diagnosis. Teratoma; due to the extensive necrosis that is associated with this tumor, the PCNB did not provide sufficient material for diagnostic examination in three of four cases, and the final diagnosis in these patients was established by surgery.
171
Germ cell tumor; this one case was misdiagnosed as a malignant epithelial tumor. Pneumothorax was the most commonly encountered complication in our study, occurring in seven patients (11.3%); of whom only two (3.2%) required placement of a chest tube and hospitalization after the biopsy. Moderate hemoptysis (30–50ml) occurred in one patient. Only two patients (3.2%) suffered from persistent chest pain after the biopsy procedure and were satisfactorily treated with analgesics.
4. Discussion Mediastinal tumors are traditionally diagnosed by mediastinoscopy, thoracoscopy or exploratory thoracic surgery. These procedures require anesthesia and hospitalization. The purpose of this study was to evaluate the clinical utility of PCNB in the diagnosis of these tumors and to determine if it should be the initial diagnostic procedure in these cases. Between 1988 and 1996, we examined a total of 70 mediastinal tumors. Most of the tumors (89%) were located in the anterior compartment. The most common pathological entities encountered were lymphoprolipherative neoplasms (47%), epithelial cancer (31%), and thymomas (11.4%). These findings are similar to those reported previously in the literature [3–8].
Table 1 PCNB/mediastinal lesions Final diagnosis
Sensitivitya
Histospecific diagnosisb
False-negative diagnosis
Non-Hodgkin’s lymphoma (n=21) Hodgkin’s lymphoma (n =9) Bronchogenic carcinoma (n=16) Thymoma (n = 6) Metastatic carcinoma (n =7) Teratoma (n = 1) Germ cell tumor (n =1) Lymphoepithelial cancer (n= 1)
95.2% 88.9% 93.7% 100% 85.7% 100% 100% 100%
90.2% 77.8% 100% 83.3% 85.7% 100% 100% 100%
2/21 (9.5%) 1/19 (11.1%) 1/16 (6.3%) 1/6 (16.7%) 1/7 (14.3%) 0/1 (0%) 0/1 (0%) 0/1 (0%)
91.9%
90.3%
Total (n = 62) a b
Sensitivity, true positive test results /all patients-regarding malignancy Histospecific diagnosis, correct histological diagnosis by PCNB
9.7%
172
J. Greif et al. / Lung Cancer 25 (1999) 169–173
The knowledge of the nature of mediastinal tumors is of critical importance in order to reach the correct therapeutic decisions. For example, thymomas are primarily treated by surgical resection, whereas lymphomas are treated by irradiation and/or chemotherapy. Patients with metastatic disease need to have their diagnosis established with certainty so that further treatment (if any) can be planned accordingly. Among the 62 adequate biopsies described in this study, PCNB demonstrated a high degree of histospecificity, as we were able to obtain an accurate histological diagnosis in 87% of these patients. PCNB showed a very high specificity in patients with lymphoma (92.5%), Hodgkin’s lymphoma (88.9%), and metastatic disease (85.7%). There were no false positive and only two false negative results in the latter group. Conversely, the diagnostic yield of tumors that tend to become necrotic, such as teratoma, is much lower and the tissue obtained by PCNB was considered to be inadequate for diagnosis in three-quarters of them. Similar results to ours were obtained in smaller series by Morrissey et al. [8] Noppen et al. [9], and Heilo [10]. Pneumothorax was the most frequent complication resulting from PCNB, with an incidence of 11.3%. This rate is similar to the range between 4.6% and 41% reported for pneumothorax induced by transthoracic fine needle biopsies [11,12]. The relative low rate of pneumothorax in spite of the use of a large cutting needle can be explained by the superficial location of 2/3 of the biopsied lesions ( B 2 cm from the chest surface). Other complications such as hemoptysis and pain (1.6 and 3.2%, respectively) are also comparable to reported rates using FNA [13,14]. In six of the 19 operated patients (31%) of whom four had lymphoma and two had adenocarcinoma, thoracic surgery (performed on the early phase of the study) only confirmed the PCNB diagnosis and could have therefore been avoided. In light of the large number of mediastinal entities that are medically treatable, such as lymphomas, or that are clearly non-resectable, such as metastatic carcinoma, it is clear that a
precise histopathological diagnosis is essential. On the basis of our experience, we believe that radiologically guided PCNB is an easy and safe procedure which can provide accurate diagnostic information in the majority of malignant mediastinal tumors and can in many cases highlight the need of more extensive diagnostic surgical procedures such as thoracoscopy or thoracotomy, which require general anesthesia and are associated with higher morbidity and cost. In resectable cases, a definite preoperative diagnosis by transthoracic needle biopsy allows for better surgical planning, especially in patients with a high surgical risk or those requiring induction therapy. In conclusion, we recommend the performance of transthoracic core needle biopsy as the initial diagnostic procedure in the evaluation of mediastinal tumors.
References [1] Adler OB, Rosenberger A, Peleg H. Fine needle aspiration biopsy of mediastinal masses: evaluation of 136 experiences. Am J Radiol 1983;140:893 – 6. [2] Weisbrod GL, Lyons DJ, Tao LC, Charberlain DW. Percutaneous fine needle aspiration biopsy of mediastinal lesions. Am J Radiol 1984;143:525 – 9. [3] Herman SJ, Holub RV, Weisbrod GL, Charberlain DW. Anterior mediastinal masses: utility of thoracic needle biopsy. Radiology 1991;180:167 – 70. [4] Quinn SF, Demlow T, Dunkley B. Temno biopsy needle: evaluation of efficacy and safety in 165 biopsy procedures. Am J Radiol 1992;158:641 – 3. [5] Bocking A, Klose K-C, Kyll HJ, Hauptmann S. Cytologic versus histologic evaluation of needle biopsy of the lung, hilum, and mediastinum. Acta Cytol 1995;35:463 – 71. [6] Ben Yehuda D, Polliack A, Dkon E, Sherman Y, Bebenshart P, Lotan H, Libson E. Image-guided core needle biopsy in malignant lymphoma-experience with 100 patients that suggests the technique is reliable. J Clin Oncol 1996;14:2431 – 4. [7] Klein JS, Salomon G, Stewart EA. Transthoracic needle biopsy with a coaxially placed 20-gauge automated cutting needle: results in 122 patients. Klein Radiol 1996;198:715 – 20. [8] Morrissey B, Adams H, Gibbs AR, Crane MD. Percutaneous needle biopsy of the mediastinum. Thorax 1993;48:632 – 7. [9] Noppen MP, DeMey JD, Meysman M, Opdebeek K, Vincken WG, Osteaux M. Percutaneous needle biopsy of localized pulmonary, mediastinal, and pleural disease tis-
J. Greif et al. / Lung Cancer 25 (1999) 169–173
173
the lungs: its safety and utility. Radiology 1984;151:301 – 3. [13] Lalli AF, McCormack LJ, Zelch M, Reich NE, Belovich D. Aspiration biopsy of chest lesions. Radiology 1978;127:35 – 40. [14] Crosby JH, Hager B, Hoeg K. Transthoracic fine needle aspiration. Cancer 1985;56:3504 – 7.
sue with an automatic disposable guillotine soft-tissue needle. Chest 1995;107:1615–20. [10] Heilo A. Tumors of the mediastinum: US-guided histologic core-needle biopsy. Radiology 1993;189:143–6. [11] Conces DS, Schwenk GR, Doering PR, Glant MD. Thoracic needle biopsy. Chest 1987;91:813–6. [12] Stevens GM, Jackman RJ. Outpatient needle biopsy of
.