| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Vascular and Interventional Radiology |
1 From the Far Eastern Memorial Hospital and College of Medicine, National Taiwan University (W.Y.L.), and Division of Chest Medicine, Department of Internal Medicine, National Taiwan University Hospital, 7 Chung-Shan S Rd, Taipei 100, Taiwan (W.Y.L., M.Z.C., Y.L.C., H.D.W., C.J.Y., P.H.K., P.C.Y.). Received December 30, 1999; revision requested February 12, 2000; revision received March 13; accepted April 4. Address correspondence to P.C.Y. (e-mail: pcyang@ha.mc.ntu.edu.tw).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Fifty consecutive patients with peripheral thoracic lesions less than 3 cm in diameter underwent US-guided percutaneous transthoracic cutting biopsy with a modified technique. Fifty lesions (43 parenchymal lung, two pleural, two chest wall, and three anterior mediastinal lesions) were sampled for biopsy. The final diagnosis was based on histopathologic analysis of surgical specimens (n = 18) or clinical follow-up (n = 32).
RESULTS: The histology recovery rate was 98% (49 lesions), and the correct diagnosis was obtained in 48 lesions (96%). Twenty-four (48%) lesions were malignant, and 26 (52%) were benign. The diagnostic accuracy for malignant lesions was 92% (22 of 24 lesions). A specific benign diagnosis was made in 17 (65%) of the 26 benign lesions, and the negative predictive value for malignancy was 93% (26 of 28 lesions). Only two patients (4%) developed postbiopsy pneumothorax, and three (6%) developed postbiopsy hemoptysis. Biopsy helped prevent surgery or thoracoscopy in 32 patients (64%): 18 patients with benign disease and 14 with multiple metastases or inoperable cancer.
CONCLUSION: US-guided transthoracic cutting biopsy appears to be a safe and effective method for diagnosing peripheral thoracic lesions less than 3 cm in diameter. The high diagnostic accuracy for benign lesions and metastatic lung cancer can help prevent surgery in many cases.
Index terms: Biopsies, technology, 60.1269 • Neoplasms, diagnosis, 60.1269, 60.12985 • Thorax, biopsy, 60.1269 • Thorax, neoplasms, 60.32 • Ultrasound (US), guidance, 60.12985
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Imaging-guided transthoracic biopsy is generally performed with fine needles, and the materials obtained by means of aspiration are usually suitable only for cytologic examination (1–3). The dependence on cytologic analysis remains a limitation of this technique in certain clinical settings, and a negative result does not exclude malignancy. Although the reported sensitivity of transthoracic fine-needle aspiration biopsy detection of lung cancer exceeds 90% (3), the false-negative diagnostic rate for malignancy ranges up to 29% (13). Suboptimal results are often obtained in cases of lymphoma and thymoma (8). In addition, transthoracic fine-needle aspiration biopsy does not allow adequate subtyping of carcinoma and seldom yields a specific pathologic diagnosis in cases of benign disease (1–3,7–9,13,14). Transthoracic cutting biopsy, which provides a tissue core for histologic examination, may improve the diagnostic yield and increase the chances of obtaining a specific diagnosis, especially in patients with benign thoracic lesions (15–19).
For tissue diagnosis, the yield of fiberoptic bronchoscopy is lower for peripheral lung cancers than for central ones, and lower still for small (<2-cm) lesions (20). Despite the proved effectiveness of transthoracic needle biopsy with image guidance for diagnosis of peripheral lung cancer, the diagnostic accuracy for small pulmonary nodules is not as well established (21–28). Some investigators have reported that the diagnostic accuracy decreases with small lesions (21,26), although others have reported little or no difference (22,23). Furthermore, transthoracic cutting biopsy is thought to be more technically difficult for small thoracic lesions and may also produce greater trauma in normal lung tissue (25).
The purpose of our study was to evaluate the safety and diagnostic accuracy of US-guided biopsy with an 18-gauge cutting needle for thoracic lesions less than 3 cm in diameter by using a modified technique.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
All patients were examined with real-time, linear, convex, and sector US units with 3.5- and 5.0-MHz transducers (model SSD 630; Aloka, Tokyo, Japan; model 100A, Toshiba, Tokyo, Japan). The US units were also equipped with Doppler US, which could be used to detect great vessels and blood flow. The patients were scanned in the supine or prone position with an intercostal approach. The size and location of the lesions were recorded, and the margins of the lesions, which were in contact with the aerated lung, were demarcated (recorded by W.Y.L. and M.Z.C.).
After written informed consent was obtained, the lesion was subjected to percutaneous transthoracic cutting biopsy with an 18-gauge Urocut needle through a sterile puncture transducer (model UST-507 BP; Aloka), which has a guiding channel and an angle selection. The Urocut needle, which is structurally similar to the Tru-Cut needle (Baxter Healthcare, Valencia, Calif), has an outer cannula and an inner obturator with a 20-mm specimen notch at the tip. When the puncture probe reached the lesion, the cutting needle was passed through the guiding channel and introduced into the margin of the lesion. If the lesion was less than 20 mm in diameter, the tip of the needle was placed at least 20 mm away from the posterior margin of the lesion-lung interface. While the cannula was held firmly, the obturator was advanced to place the specimen notch inside the lesion. The outer cannula was then advanced rapidly to cut off the tumor tissue in the specimen notch. If the lesion was less than 20 mm in diameter, the specimen contained not only the target lesion but also a tissue core from the pleura and chest wall. Subsequently, the entire unit was withdrawn. Figure 1 shows the exact technique of modified US-guided cutting biopsy for lesions less than 20 mm in diameter.
|
Patients who underwent the biopsy procedure were then treated by their attending physicians at their department. The subsequent management regarding surgical intervention or follow-up of the patients was decided by the attending physicians. If a patient went to surgery, the pathology report of the surgical specimen was obtained. If a patient was observed without surgical intervention, he or she was followed up at an outpatient clinic and serial chest radiographs were obtained. The biopsy procedure was performed on an inpatient basis. Thus, all patients were admitted overnight.
The definite diagnosis of malignant thoracic lesions was based on (a) the histopathologic analysis of the surgical specimen or (b) the histopathologic analysis of the cutting biopsy specimen and a subsequent clinical course showing progressive disease and/or metastatic disease that were consistent with cancer. The definite diagnosis of benign thoracic lesions was based on (a) surgical confirmation, (b) subsequent disappearance of the lesion or decrease in its size, or (c) follow-up chest radiographs or CT scans showing that the lesion remained stable for at least 2 years.
A true-positive result for malignancy indicates that histopathologic analysis of the biopsy specimen was diagnostic of a malignant thoracic lesion. A false-negative result for malignancy indicates that histopathologic analysis of the biopsy specimen showed the thoracic lesion to be benign, but surgery or the subsequent clinical course showed the lesion to be malignant. A specific benign diagnosis indicates that one of the following was identified at histologic examination of the biopsy sample: (a) granulomatous inflammation, (b) an identifiable microorganism, (c) a core of fibrosis, or (d) a specific benign tumor, such as a neurogenic tumor. The following were considered to be nonspecific benign diagnoses: (a) chronic inflammation, (b) an inflammatory cell aggregate, and (c) necrosis. If only skeletal muscle or mesothelial cells were found on the histology section, or if the pathologist reported only minimal histologic change, the tissue sample was considered to be inadequate. The histology recovery rate was calculated as the number of specimens that were adequate for histologic diagnosis divided by the total number of specimens times 100%.
The size, number, and location of the lesions, the pathology reports of the biopsy specimens, the final diagnoses of the lesions, and the demographic features and clinical outcomes of the patients were recorded (W.Y.L., M.Z.C.).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
According to the final diagnoses of the 50 lesions, 24 lesions (48%) were malignant and 26 lesions (52%) were benign. The mean diameter of the 50 lesions was 2.3 cm ± 0.6 (SD; range, 1–3 cm). Twenty-nine lesions (16 malignant, 13 benign) were 21–30 mm, 20 (eight malignant, 12 benign) were 11–20 mm, and one (benign) was 10 mm in diameter. In the 50 biopsy specimens, the histology recovery rate was 98% (49 of 50 lesions). Histologic examination yielded the correct diagnosis for 48 (96%) of the tissue samples. The relationship between lesion size and diagnostic accuracy is demonstrated in the Table.
|
The remaining 15 lesions were diagnosed on the basis of histopathologic evaluation of the biopsy specimens and subsequent clinical courses. One patient in whom the diagnosis of lymphoma was made by means of transthoracic biopsy responded to systemic chemotherapy and remained in complete remission after 2 years of follow-up. The other 14 patients were followed up for a mean of 4 months and had progressive disease; three of them died of cancer progression. The diagnostic accuracy for malignant lesions was 92% (22 of 24 lesions); there was one false-negative result and one inadequate specimen. There were no false-positive results; the positive predictive value for malignancy was 100% (22 of 22 lesions).
Of the 24 patients with histologically proved malignant thoracic lesions, 12 had newly diagnosed lung carcinoma, one had recurrent lung cancer, two had lymphoma, one had thymic carcinoma, and eight had metastatic cancers (two hepatocellular carcinomas, one brain hemangiopericytoma, one squamous cell carcinoma of the tongue, one transitional cell carcinoma of the bladder, one leiomyosarcoma of the uterus, one adenocarcinoma of rectum, and one infiltrating ductal carcinoma of the breast). In seven of the eight patients with metastatic cancers, histologic analysis of the biopsy specimens showed features consistent with those of specimens resected during the previous surgery. In the other one patient, who had multiple hepatic and lung tumors, hepatocellular carcinoma was proved with US-guided transthoracic cutting biopsy (Fig 2).
|
|
|
|
Five of the nine patients in whom US-guided transthoracic cutting biopsy yielded a nonspecific benign diagnosis underwent surgical resection, which revealed tuberculosis in four patients and focal organizing pneumonia in the other patient. The remaining four patients were followed up closely without surgical intervention because of a low probability of malignancy. The mean follow-up period was 9 months (range, 5–18 months) for these four patients, and the lesions resolved during the follow-up period. Figure 3 demonstrates the imaging studies and the histopathologic examination findings of the biopsy specimen in a patient with a pulmonary nodule.
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Peripheral thoracic lesions are usually not assessable with fiberoptic bronchoscopy, and the diagnostic yield of bronchoscopic biopsy is low (20). Some authors have proposed aggressive diagnostic procedures such as thoracotomy or thoracoscopic biopsy for small peripheral thoracic lesions (32,33); however, with this strategy, patients with benign lesions and those with metastatic tumors would undergo unnecessary surgical procedures. Furthermore, a minority of patients are not surgical candidates because of poor medical condition or because the underlying malignancy is not suitable for surgery, as is the case with small cell lung cancer and with lymphoma.
With recent improvements in the resolution of US, several authors have demonstrated that US can be as effective as fluoroscopy or CT for guidance of transthoracic biopsy for peripheral thoracic lesions (4–10). US has a number of advantages over CT and fluoroscopy, including bedside approach, lower cost, and no radiation exposure, which led to our preference to perform US-guided biopsy of peripheral lesions larger than 2 cm. With real-time monitoring by means of US guidance, biopsy can be conducted in the respiratory phase during which the nodule is most accessible. With experienced technicians and good patient cooperation in terms of breath holding, US-guided needle biopsy can be as safe and effective for small peripheral pulmonary lesions as for large ones. In our study, specimens adequate for histologic examination were obtained from 49 (98%) of the 50 lesions, and correct diagnosis was made in 48 of these (96%).
Despite its high diagnostic yield for malignant lesions, fine-needle transthoracic biopsy has several disadvantages. First, a specific benign diagnosis, which can increase the dependability of a cancer-negative biopsy result, is difficult to obtain with this method (14). The reported accuracy of a specific benign diagnosis made on the basis of transthoracic fine-needle aspiration biopsy ranges from 12% to 57% and is more commonly about 20%–30% (34). In the study by Calhoun et al (13), transthoracic fine-needle aspiration biopsy yielded a specific benign diagnosis in only 16 (12%) of 132 cancer-negative biopsy specimens. Thirty-nine (29%) of the 132 patients were ultimately found to have cancer; a nonspecific benign diagnosis cannot exclude malignancy.
Second, the reported agreement between results of transthoracic fine-needle aspiration cytology and final histology varies from approximately 60% to over 90% (3). Although fine-needle aspiration cytology may be sufficient for diagnosis in patients with primary lung cancer, it is not reliable for diagnosing metastatic lung cancer or mediastinal tumors. In a previous report by Yang et al in 1992 (7), the diagnostic accuracy of transthoracic fine-needle aspiration cytology for metastatic lung cancer was only 33%. In addition, the diagnostic accuracy of fine-needle aspiration cytology for mediastinal tumors, such as lymphomas, thymomas, and germ cell tumors, is lower than that for lung cancer (8).
Various maneuvers to improve fine-needle biopsy results have been attempted. Repeated biopsy and having a cytologist present during biopsy have been advocated to reduce the false-negative rate (14,35), but the likelihood of complications increases with repeated needle passes. Greene et al (36), using supplement tissue core histologic analysis, reported a specific benign diagnosis rate of 44% (12 of 27 lesions), but adequate tissue was obtained from only 72% (108 of 150) of the lesions. In recent years, transthoracic cutting biopsy has been reported to be superior to fine-needle aspiration for specific diagnosis of benign thoracic lesions and mediastinal tumors, determination of cancer cell type, and prediction of cancer-negative findings. The reported accuracy of a specific benign diagnosis with transthoracic cutting biopsy ranges from 52% to 91% (7–9,15–19). In our study, a specific benign diagnosis was obtained in 17 (65%) of the 26 benign lesions. Comparison with the histologic diagnosis of the previous surgical specimens confirmed the diagnosis reached by means of histologic examination of the cutting biopsy samples in seven of the eight cases of metastatic lung cancer. Correct histologic diagnoses were obtained for all three patients with anterior mediastinal tumors (one lymphoma, one thymic carcinoma, and one lymphangiomyoma).
Previous reports have conflicted with one another regarding the accuracy of transthoracic needle biopsy for the diagnosis of small and large thoracic lesions (21–28). Hayata et al (21) reported that the diagnostic accuracy was higher for large (>2-cm) peripheral pulmonary carcinomas (90% [60 of 67]) than for small ones (75% [12 of 16]). Li et al (26) also found that the diagnostic yield was lower for small (<1.5 cm in diameter) peripheral pulmonary nodules (74% [20 of 27]) than for large ones (96% [67 of 70]). However, equal diagnostic accuracies for large and small thoracic lesions have been reported in other studies (22,23). None of these four studies reported a difference in the rate of pneumothorax between patients with small and those with large thoracic lesions. Hayashi et al (28) reported a histology recovery rate of 90% when CT-guided transthoracic cutting biopsy was performed for pulmonary nodules less than 3 cm in diameter. In their study, the diagnostic accuracy for malignant lesions was 100%, and a specific benign diagnosis was obtained in 65% of cases with a benign thoracic lesion.
Although a previous report by Yuan et al (24) demonstrated that US-guided transthoracic aspiration biopsy is an accurate and safe approach for diagnosing small peripheral pulmonary nodules less than 3 cm, the majority of the pulmonary nodules were malignant (80% [24 of 30]), and they used only fine-needle aspiration for cytologic examination. In the present series, we chose 3 cm as the cutoff point to include a lesion, because the technical difficulty of cutting biopsy is greater in thoracic lesions less than 3 cm in diameter. In patients who had thoracic lesions less than 2 cm in diameter in our study, the histology recovery rate (95%; 20 of 21) as well as diagnostic accuracy (95%; 20 of 21) were compatible with those in previous studies (7–9,15–19), which were not restricted to small thoracic lesions.
Pneumothorax and minor bleeding are common complications of transthoracic needle biopsy. The reported frequency of biopsy-associated pneumothorax ranges from 8% to as high as 61% (3). Factors associated with a high frequency of pneumothorax include preexistent lung disease, increased lesion depth, and advanced patient age. Transthoracic cutting biopsy with a large bore needle does not increase the rate of pneumothorax, even for small pulmonary lesions (7–9, 15–19, 28). In our series of 50 patients, postbiopsy pneumothorax developed in only two patients (4%) and did not necessitate chest tube insertion. Real-time monitoring helped avoid puncturing the aerated lung, and the fact that many of the lesions were located peripherally also may have contributed to the low rate of pneumothorax among our patients. Furthermore, the modified US-guided cutting biopsy technique for lesions less than 20 mm in diameter used in this study may have enabled us to avoid cutting the lung parenchyma located beneath the lesion and thus reduce the rate of pneumothorax and hemoptysis in our series. Only three (6%) of our patients developed mild hemoptysis; in two of these patients, the lesions were localized pulmonary infections and in the other patient, it was non–small cell lung cancer. In all three cases, the bleeding stopped spontaneously without specific treatment.
Our study had several limitations. First, our patient number was small. Second, all of the patients selected for US-guided transthoracic cutting biopsy had negative study results of transthoracic fine-needle aspiration or bronchoscopic biopsy. This may have influenced the patient population. Third, only lesions that abutted the chest wall and could be visualized with US were selected for US-guided transthoracic cutting biopsy. Therefore, there may have been a selection bias.
In conclusion, US-guided transthoracic biopsy with an 18-gauge cutting needle appears to be a safe and effective method for diagnosing malignant and benign peripheral thoracic lesions less than 3 cm in diameter. The rates of pneumothorax and hemoptysis after the procedure were low. Our results also demonstrated a high rate of specific benign diagnosis among patients with confirmed benign diagnoses. This procedure can help avoid unnecessary surgical procedures in patients with benign thoracic lesions of an infectious nature or metastatic lung tumors.
|
FOOTNOTES |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  V. Bandi, W. Lunn, A. Ernst, R. Eberhardt, H. Hoffmann, and F. J. F. Herth Ultrasound vs CT in Detecting Chest Wall Invasion by Tumor: A Prospective Study Chest, April 1, 2008; 133(4): 881 - 886. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Sartori, P. Tombesi, L. Trevisani, I. Nielsen, D. Tassinari, and V. Abbasciano Accuracy of Transthoracic Sonography in Detection of Pneumothorax After Sonographically Guided Lung Biopsy: Prospective Comparison with Chest Radiography Am. J. Roentgenol., January 1, 2007; 188(1): 37 - 41. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W-Y. Liao, J-S. Jerng, K-Y. Chen, Y-L. Chang, P-C. Yang, and S-H. Kuo Value of imprint cytology for ultrasound-guided transthoracic core biopsy Eur. Respir. J., December 1, 2004; 24(6): 905 - 909. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Nakano, K. Miyauchi, H. Imagawa, and K. Kawachi Immediate localization using ultrasound-guided hookwire marking of peripheral lung tumors in the operating room Interactive CardioVascular and Thoracic Surgery, March 1, 2004; 3(1): 104 - 106. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Sartori, I. Nielsen, L. Trevisani, P. Tombesi, P. Ceccotti, and V. Abbasciano Contrast-Enhanced Sonography as Guidance for Transthoracic Biopsy of a Peripheral Lung Lesion With Large Necrotic Areas J. Ultrasound Med., January 1, 2004; 23(1): 133 - 136. [Full Text] [PDF]
|
|
|
|
|
|
M. Gorguner, F. Misirlioglu, P. Polat, H. Kaynar, L. Saglam, A. Mirici, and S. Suma Color Doppler Sonographically Guided Transthoracic Needle Aspiration of Lung and Mediastinal Masses J. Ultrasound Med., July 1, 2003; 22(7): 703 - 708. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
