| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Vascular and Interventional Radiology |
1 From the Department of Radiology, Churchill Hospital, Oxford Radcliffe Hospitals, Old Rd, Headington, Oxford OX3 7LJ, England. From the 2000 RSNA scientific assembly. Received April 17, 2000; revision requested June 7; revision received August 22; accepted October 2. Address correspondence to F.V.G. (e-mail: fergus.gleeson@radiology.oxford.ac.uk).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Thirty-three adult patients with diffuse or focal pleural thickening (median, 1.0 cm; range, 0.2–6.0 cm), pleural effusion, and suspected pleural malignancy underwent percutaneous image-guided cutting-needle biopsy. Biopsy guidance was performed with computed tomography in 24 patients and ultrasonography in nine patients. A final diagnosis of benign or malignant disease was established with radiologic and clinical follow-up findings and with other histologic or cytologic findings, when available.
RESULTS: A correct histologic diagnosis of malignant disease was made in 21 of 24 patients (sensitivity, 88%; specificity, 100%), including 13 of 14 patients with mesothelioma (sensitivity, 93%). A correct histologic diagnosis of benign pleural disease was made in nine patients. Positive and negative predictive values for malignant disease were 100% and 75%, respectively. The overall accuracy was 91%. Complications comprised a chest wall hematoma in one patient.
CONCLUSION: Image-guided percutaneous cutting-needle biopsy of pleural thickening in the presence of a pleural effusion is a safe procedure, with an overall accuracy of 91% in the diagnosis of malignancy.
Index terms: Computed tomography (CT), guidance, 66.1263 • Mesothelioma, 66.317 • Pleura, biopsy, 66.1263 • Pleura, fluid, 66.76 • Pleura, neoplasms, 66.317, 66.33 • Ultrasound (US), guidance, 66.12985
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Computed tomography (CT) has been shown (12,13) to be of value in the assessment of malignant pleural disease, with the demonstration of pleural thickening that is nodular, is circumferential, crosses the mediastinum, or is more than 1 cm thick; thus, CT permits the diagnosis of malignancy, with a sensitivity of 72%–97% and a specificity of 87%. In addition, the use of pleural biopsy in patients with CT-demonstrated pleural thickening in the absence of a pleural effusion has been shown (13) to increase the sensitivity and specificity. However, to our knowledge, the diagnostic accuracy of image-guided pleural biopsy in patients with pleural effusions has not been previously reported. The aim of this study was to evaluate the diagnostic accuracy of percutaneous image-guided cutting-needle biopsy (CNB) of pleural thickening in the presence of a suspected malignant pleural effusion.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
CNB of the pleura is considered a routine procedure in our institution, and it is performed after informed consent is obtained. Biopsy was performed by one radiologist (F.V.G.) with more than 7 years of experience at a single institution. Biopsy guidance was performed with CT in 24 patients and with ultrasonography (US) in nine patients.
The optimal biopsy site was selected from the diagnostic CT findings. All patients were placed in the prone position for US-guided biopsy and in either the supine or prone position for CT-guided biopsy. For focal pleural masses, the biopsy needle was advanced along the maximal diameter of the lesion. For diffuse pleural thickening, the cutting needle was advanced along the plane of maximal pleural thickness to allow the majority of the biopsy specimen notch to lie within the abnormal pleura, as shown in Figure 1. The number of CNB passes was determined on the basis of the macroscopic adequacy of the core specimen and the ease of biopsy. A single CNB pass was made if an adequate specimen was obtained. In some patients, fine-needle aspiration (FNA) of the pleura was performed with a 22-gauge needle prior to CNB, in the absence of a cytopathologist, and only one pass was made if the sample appeared macroscopically adequate. The patients had at least 4 hours of bed rest after the procedure; routine postbiopsy observations (pulse rate, blood pressure, and respiratory rate) were made. Chest radiographs were obtained 1 and 4 hours after the procedure to exclude pneumothorax.
|
All patients underwent contrast-enhanced diagnostic CT. In those patients undergoing CT-guided biopsy, the pleural thickness at the site of biopsy was recorded from the biopsy CT scan. In the nine patients undergoing US-guided biopsy, diagnostic CT was performed within 23 days (within 6 days in six patients), and the pleural thickness at the US biopsy site was recorded from the CT scan. By using the CT scans, we recorded whether the pleural disease was focal or diffuse. Focal disease was defined as one or more pleural masses adjacent to apparently normal pleura (Fig 2) requiring specific CNB targeting. Diffuse disease was defined as continuous pleural thickening (Fig 3). CT observations were made by both authors in consensus.
|
|
All CNB specimens were reported by a single histopathologist with a special interest in thoracic disease. Two of the samples (a case of sarcoma and a case of malignancy that could not be typed) were both examined by two additional pathologists. FNA and CNB findings were reported independently.
The radiologic and clinical follow-up of the patients, with other histologic or cytologic findings (when available), were used to establish whether the patients had benign or malignant disease.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Complications comprised one chest wall hematoma (2-cm diameter) that occurred after biopsy was performed with US guidance; this complication did not require active management. No patient experienced a vasovagal reaction or pneumothorax.
Final Diagnoses
In 24 patients, the final diagnosis was malignant pleural disease, which was detected at CNB in 21, at FNA in 12, at thoracoscopy in two, and at postmortem examination in 14. All patients with malignant disease had an appropriate clinical and radiologic course. In one patient, a diagnosis of malignant fibrous tumor of the pleura was established with CNB histologic findings, with the progressive increase in the size of the mass during 15 months, with recurrent uncontrollable pleural effusions, and with the death of the patient from respiratory compromise. Nine patients had benign pleural disease—which was detected at CNB in all patients, at mediastinoscopy in one, and at postmortem examination in one—with follow-up of at least 8 months. The final diagnoses are listed in the Table.
|
In patients with mesothelioma, a correct histologic diagnosis was made at CNB in 13 of 14 (sensitivity, 93%; specificity, 100%). All cases of mesothelioma were confirmed with further histologic examinations—13 at postmortem examination and two at thoracoscopic biopsy. In these latter two patients, thoracoscopy was performed in one patient for talc pleurodesis, but in the other (female patient), CNB diagnosis of mesothelioma was unexpected, and the physician required confirmation at thoracoscopic biopsy. One biopsy finding (single CNB pass producing a 1-cm-long core from 0.3-cm pleural thickening) incorrectly showed only fibrosis and chronic inflammation; similar findings had been made at thoracoscopic biopsy.
False-negative findings occurred in two patients with nonmesothelial malignancy at CNB, but FNA cytologic findings showed carcinoma cells. One biopsy (single CNB pass producing a 1.2-cm core from 3.0-cm pleural thickening) revealed a nondiagnostic histologic finding that was the same as that of the preceding thoracoscopic biopsy, in which the tissue was necrotic. The other biopsy (single CNB pass producing a 0.8-cm core from 1.1-cm pleural thickening) was a technical failure, with no pleural tissue obtained.
A correct diagnosis of benign pleural disease was made at CNB in nine patients (sensitivity, 100%; specificity, 75%). Sarcoidosis was diagnosed in one patient and confirmed at mediastinoscopic lymph node biopsy. Inflammatory thickening after empyema was diagnosed in three patients; in one, a right middle lobe bronchial carcinoma was found at CT and confirmed at bronchoscopic biopsy. This patient received radical radiation therapy to the proximal bronchial mass and died 17 months after pleural biopsy; this patient did not develop clinical or radiologic evidence of malignant pleural disease. One patient died of cardiac causes, and postmortem findings confirmed the benign nature of the pleural thickening. Apart from the two deceased patients described, all other patients with a diagnosis of benign pleural thickening were alive and well when this article was written. The clinical and radiologic follow-up (range, 8 months to 5 years; median, 24 months) findings were concordant with the benign biopsy results, with no evidence of progressive pleural disease. The overall accuracy of CNB in the differentiation of benign from malignant disease was 91%.
FNA Results
FNA was performed in 21 patients who were subsequently found to have malignant (12 patients with mesothelioma) or benign (three patients) disease. FNA findings were suggestive but not diagnostic for mesothelioma in five of 12 patients with mesothelioma. Of nine patients with nonmesotheliomal malignant disease, FNA findings were diagnostic of malignancy in five patients, but we were unable to determine the type in two.
All three patients in whom CNB failed to demonstrate malignant pleural disease underwent FNA. The FNA finding was also negative in the one case of mesothelioma that was not diagnosed at CNB, but it was diagnostic in the two cases of nonmesothelial malignant disease with nondiagnostic CNB findings, as described earlier. FNA findings in three patients with benign pleural thickening were nondiagnostic. There were no false-positive diagnoses of malignant disease at FNA.
Pleural Thickness and Biopsy Core Size
The pleural thickness at the site of biopsy ranged from 0.2 to 6.0 cm (median, 1.0 cm). For patients with malignant disease, the range was 0.3–6.0 cm (median, 1.4 cm), and for those with benign disease, the range was 0.2–1.0 cm (median, 0.5 cm).
The size of the core obtained with the cutting needle was stated for 31 of 33 patients in the histology report. The core lengths ranged from tiny fragments to 2.4 cm, but they were independent of pleural thickness. Fragments were obtained in five of nine patients with a pleural thickness of 2 cm or more, but they were obtained in only one of 22 patients with a pleural thickness of less than 2 cm. The pleural thicknesses at CNB with nondiagnostic findings were 0.3, 1.1, and 3.0 cm.
In nine patients, CNB was performed with a pleural thickness of 0.2–0.5 cm; findings from only one CNB were nondiagnostic, but they provided a 1-cm-long core. In nine patients, the core size was minute fragments in one, and the rest ranged in length from 0.5 to 1.5 cm (median, 0.8 cm). Four of these nine patients had a final diagnosis of malignant pleural disease, and the minimum pleural thickness with a biopsy diagnosis of malignancy was 0.5 cm.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The conventional investigation in patients suspected of having malignant pleural effusion at aspiration cytologic examination and pleural biopsy is diagnostically unsuccessful in 30%–35% of patients (3,7,16,17). As shown in Figure 2, pleural biopsy without image guidance is likely to be unsuccessful in patients with focal pleural malignancy, with the pleural abnormality not sampled at biopsy.
During the past decade, thoracoscopic biopsy has become a widely accepted means of diagnosis if findings from pleural fluid cytologic examination and blind pleural biopsy are nondiagnostic. The advantages are its sensitivity of 91%–98% for the diagnosis of pleural malignancy (10,18) and its potential therapeutic benefit, such as enabling talc pleurodesis. Thoracoscopy has the disadvantage of requiring general anesthesia or sedation, requiring chest tube drainage and inpatient stay, having the potential of failure due to adhesions preventing pneumothorax, being substantially more expensive than image-guided biopsy, and having a higher complication rate (5,11).
In the study by Harris et al (11), in 182 patients who underwent thoracoscopy for pleural disease, 98% of procedures were performed with general anesthesia, and 15% of patients had major complications. Other investigators report lower complication rates: In the study by Boutin and Rey (5), in 188 patients undergoing diagnostic thoracoscopy for malignant mesothelioma, the procedure was performed with sedation and a 36-hour hospital stay. The complications were fever for 1–2 days in 26 patients, pleural infection in four, hemorrhage in three, and extensive subcutaneous emphysema in one. There were no major complications in our study, with a single minor complication of a small chest wall hematoma that did not require active management. Overall, image-guided pleural CNB is safer than thoracoscopy, requires use of only a local anesthetic, and can be performed as an outpatient procedure.
Approximately 19% of patients with mesothelioma develop malignant seeding along tracks produced by needles, chest tubes, thoracoscopic trocars, and surgical incisions (19). Although patients are usually offered local radiation therapy to these areas if malignant mesothelioma is diagnosed (19), a test likely to produce a diagnosis with minimal potential tumor seeding tracks is desirable.
Prior investigators (13,20) have suggested that image-guided CNB is a safe and accurate means of diagnosing pleural malignancy in patients with diffuse pleural thickening, with a sensitivity of 77%–83% and specificity of 88%–100%. These reports do not specify the presence of pleural effusions. To our knowledge, there have been only three small studies (21–23) in which the value of CNB of the pleura in the presence of a pleural effusion was specifically examined. The studies by McLeod et al (21) and Christopher et al (22) did not involve image guidance, were performed in areas with a high incidence of pleural tuberculosis, and contained few malignant cases. Chang et al (23) compared nonguided biopsy with Abrams biopsy in 24 patients and with US-guided pleural biopsy with a Tru-cut needle in 25 patients in only a small number of patients with malignant pleural disease; the degree of pleural thickening examined at biopsy was not recorded. Nevertheless, CNB findings were positive for malignant disease in seven of 10 patients, but only four of nine patients had positive biopsy findings at Abrams biopsy (23).
The limitations of our study include its retrospective nature and the relatively small number of patients, although we had more cases of malignant disease than the three studies mentioned earlier. The other limitation is the lack of alternative histologic samples in some patients. We addressed this by assessing the full medical evidence in each case to ascertain that the final diagnosis was highly likely to be true.
Our study findings indicate that image-guided pleural CNB is an accurate and safe procedure in the presence of a pleural effusion. In this study, CNB achieved a sensitivity of 88% in the detection of malignant pleural disease in 24 patients and achieved a specificity of 100%. In 14 patients with mesothelioma, we achieved a sensitivity of 93%. CNB also allows the positive identification of benign pleural disease in patients and, in areas where tuberculosis is prevalent, CNB has been shown (23) to be more sensitive than blind hook-needle biopsy.
The complication rate was low (3%) in our series, possibly due to the presence of pleural effusion that makes lung laceration and pneumothorax less likely. CNB of minimal pleural thickening (0.2–0.5 cm) can provide adequate cores for histologic diagnosis. By performing CNB along the line of the pleura, it is possible to produce tissue cores in patients with a pleural thickness of as little as 2 mm. Fragmentation of the specimen was most common in patients with a pleural thickness of greater than 2 cm, and an inadequate specimen was produced in only one patient with a pleural thickness of less than 1 cm.
The FNA results showed that it was more sensitive in the detection of nonmesothelial malignancy than mesothelioma, with a sensitivity of 78% versus 50% (at best), but they confirmed that CNB is more sensitive than FNA overall.
Three (25%) of 12 benign CNB results were false-negative in this series. The addition of FNA to CNB had no effect on the diagnostic outcome in the one undiagnosed case of mesothelioma at CNB, but it did contribute to the two nondiagnostic CNB cases with nonmesothelial malignant disease. The performance of both CNB and FNA increased the sensitivity for the detection of malignant pleural disease from 88% to 97%, leaving one false-negative result (3%). FNA depicted only additional carcinomas and was unhelpful in the diagnosis of benign pleural disease or mesothelioma; these findings were similar to those of previous reports (24,25). FNA is a useful adjunct to CNB and adds little to the length of the procedure. We routinely perform FNA prior to CNB unless there is a high clinical suspicion of mesothelioma. With benign pleural biopsy results, close follow-up is essential. Further biopsy should be considered, whether performed with a cutting needle or with thoracoscopy, if clinical suspicion of malignant disease persists.
Pleural CNB provides a minimally invasive and safe means of obtaining adequate tissue for a histologic diagnosis of the cause of pleural thickening in the presence of a pleural effusion, including mesothelioma (26), with an overall accuracy of 91%, even with minimal pleural thickening. Pleural CNB, rather than blind hook-needle or thoracoscopic biopsy, should be considered as a first-line test for investigating suspected malignant pleural thickening.
|
FOOTNOTES |
|---|
Author contributions: Guarantor of integrity of entire study, F.V.G.; study concepts and design, F.V.G.; literature research, R.F.A.; clinical studies, R.F.A., F.V.G.; data acquisition, R.F.A.; data analysis/interpretation, R.F.A., F.V.G.; statistical analysis, R.F.A.; manuscript preparation, R.F.A.; manuscript definition of intellectual content, F.V.G.; manuscript editing and revision/review, R.F.A., F.V.G.; manuscript final version approval, F.V.G.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  P. P. Agarwal, J. M. Seely, F. R. Matzinger, R. M. MacRae, R. A. Peterson, D. E. Maziak, and C. J. Dennie Pleural Mesothelioma: Sensitivity and Incidence of Needle Track Seeding after Image-guided Biopsy versus Surgical Biopsy Radiology, November 1, 2006; 241(2): 589 - 594. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N A Maskell and R J A Butland BTS guidelines for the investigation of a unilateral pleural effusion in adults Thorax, May 1, 2003; 58(90002): ii8 - 17. [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
