|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From C.U.B.E. Department of Cardiovascular and Respiratory Sciences (Drs. Paone, Lucantoni, and DAngeli), University of Rome "La Sapienza," Forlanini Hospital and Don Gnocchi Foundation; Department of Epidemiology (Dr. Nicastri), National Institute for Infectious Diseases, IRCCS Lazzaro Spallanzani; and Center of Thoracic Endoscopy (Drs. Dello Iacono, Battistoni, Lucatoni, DAngeli, and Galluccio), Forlanini Hospital, Rome, Italy.
Correspondence to: Gregorino Paone, MD, Via Umberto Saba 72-c, 00144 Roma, Italy; e-mail: zpaone1023{at}yahoo.com.it
| Abstract |
|---|
|
|
|---|
Design: Prospective, randomized, blinded study.
Setting: University Hospital of Rome, Italy.
Patients and methods: We examined 799 patients with peripheral lung lesions using bronchoscopy. Patients who could undergo a complete clinical diagnostic follow-up (n = 293) were enrolled in the study and randomly assigned to EBUS-TBB or TBB. We performed these two procedures on 221 patients (97 EBUS-TBB and 124 TBB). Patients in whom biopsies were not diagnostic underwent more invasive procedures to obtain a final diagnosis, and a complete follow-up was possible in 206 patients (87 EBUS-TBB and 119 TBB).
Results: Lung cancer was diagnosed in 61 patients in the EBUS-TBB group and in 83 patients in the TBB group. Pulmonary diseases other than cancer were diagnosed in 26 patients and 36 patients, respectively. For patients with lung cancer, sensitivity was 0.79 in the EBUS group and 0.55 in the TBB group (p = 0.004), and accuracy was 0.85 and 0.69, respectively (p = 0.007). The analysis of a subset of patients with lesions > 3 cm showed no significant difference in diagnostic ability between the two procedures. In lesions < 3 cm, we found a considerable decline in TBB sensitivity and accuracy (0.31 and 0.50) while EBUS-TBB maintained their diagnostic yield (0.75 and 0.83) [p = 0.0002 and p = 0.001, respectively]. A similar difference was observed when we compared the sensitivity of the two procedures in lesions < 2 cm (0.23 vs 0.71, p < 0.001).
Conclusions: EBUS-TBB can be an important option in the early diagnosis of peripheral lung cancer, especially in small-sized lesions and in patients who are not eligible for surgery.
Key Words: CT scan endobronchial ultrasound-driven transbronchial biopsy lung cancer peripheral lesions sensitivity specificity transbronchial biopsy
| Introduction |
|---|
|
|
|---|
Early histologic diagnosis is crucial for lung cancer treatment because histopathologic types correlate with tumor behavior and management. Diagnosis affects the choice of treatment, as surgery is considered the main therapy for non-small cell lung cancer (NSCLC), and chemotherapy/radiotherapy combination is the most important therapeutic option for small cell lung cancer (SCLC).1. The diagnosis of lung cancer can be determined either by noninvasive or invasive approaches. However, while some of the noninvasive tests (eg, CT, positron emission tomography, MRI) may identify lesions suspicious for cancer, they do not provide a definitive histologic diagnosis and are often not sufficient for the planning of correct cancer management.145
The use of other noninvasive methodologies (eg, tumor markers), either alone or in combination, lacks a consensus.6789 Peripheral lesions (25 to 30% of lung cancers) usually require more aggressive measures.10 Among these methodologies, endobronchial ultrasound (EBUS) is a recent and relatively noninvasive procedure that can be used during bronchoscopy.11 We designed a prospective, randomized study to compare the diagnostic sensitivity, specificity, accuracy, and predictive values of EBUS-driven transbronchial biopsy (EBUS-TBB) vs transbronchial biopsy (TBB) in peripheral lung cancers, in particular for patients with lesions < 3 cm (T1) in diameter.
| Materials and Methods |
|---|
|
|
|---|
We screened 799 patients with peripheral lung lesions; 506 patients were ineligible because they did not satisfy the inclusion criteria (Fig 1 ). Among the ineligible patients, we excluded 386 outpatients because of previously experienced low compliance of this subset of patients to accomplish a complete follow-up algorithm. We randomly assigned the 293 eligible patients to EBUS-TBB or TBB (144 patients and 149 patients, respectively) using random numbers with a 1:1 allocation ratio.
|
Chest CT
CT imaging with a contrast medium (total volume, 130 mL at 2.5 mL/min) was performed using a multisliced CT (Somatom Volume Zoom; Siemens; Erlangen, Germany) with 5-mm slices. Images were reconstructed at 1 mm. The chest CT was reviewed by two independent radiologists who classified the characteristics of the lesions. The size of the lesions was determined by measuring the greatest diameter on the CT scan. A lesion was defined as peripheral when it was beyond the segmental bronchus.10 Hard copies of the CT scans were made available to the bronchoscopists during the endoscopic procedures.
Bronchoscopy
Two experienced, independent bronchoscopists (G.G., R.D.I.) randomly performed all the bronchoscopy-driven biopsies using a flexible bronchoscope (Olympus BF B3 or BF T20; Olympus; Tokyo, Japan) under local anesthesia, as previously reported.1213 In patients undergoing EBUS-TBB, ultrasound analysis was performed using an endoscopic ultrasound system (EU-M 30; Olympus), equipped with a 20-MHz flexible probe, following careful examination of the upper airways and bronchial tree. The bronchial wall and its layers were visualized with a 360° view. The small diameter of the peripheral branches permitted perfect adhesion with the inner surface of the airways wall, making the use of the water-filled balloon unnecessary. In some cases, we preinstilled 10 mL of saline solution to improve ultrasound imaging. The flexibility of the peripheral bronchi allowed us to guide the probe into the subpleural regions.
After localization of the target lesion, the probe was removed and five biopsy samples were taken in the same place indicated by the probe, using flexible transbronchial biopsy forceps. In patients who underwent TBB, an equal number of biopsies was performed using identical forceps that were inserted through the working channel of the bronchoscope and moved within the lesion into the corresponding segmental bronchus previously localized with the CT scan. The material sampled was immersed in 10% formalin and analyzed by two study-blinded pathologists for the definitive histologic assessment. The overall mean time (including instrument set-up and interpretation of the findings) required to perform the biopsies was 9.8 min (range, 7 to 14 min) for EBUS-TBB and 8.1 min (range, 6 to 12 min) for TBB.
Statistical Analysis
To compare the diagnostic yield of both procedures in malignant peripheral lesions, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy were calculated by using two-by-two tables with their respective 95% confidence intervals (CIs). The characteristics of the two procedures were compared using the
2 test. All analyses were two tailed. Statistical software (Statistical Package for the Social Sciences for Windows, version 11.0.1; SPSS; Chicago, IL) was used for statistical analysis, and p < 0.05 was considered statistically significant.
Sample size and power of the study were calculated as follows: considering a sensitivity difference of at least 0.15 between the two bronchoscopic procedures aimed to detect lung cancer among all enrolled patients, the double-side sample size was calculated using a statistical power level of 0.80 and suggested that a minimum number of 139 persons was enough to show an
level of 0.05. Considering a sensitivity difference of at least 0.25 between the two bronchoscopic procedures aimed to detect lung cancer among the subset of patients with peripheral lesion < 3 cm, the double-side sample size was calculated using a statistical power level of 0.80 and suggested that a minimum number of 61 persons was sufficient to show an
level of 0.05.
| Results |
|---|
|
|
|---|
|
|
We compared sensitivity, specificity, accuracy, and predictive values between the two groups of patients with malignant peripheral lesions. Significantly higher sensitivity and accuracy rates were observed in the study group (p = 0.004 and p = 0.007, respectively). The results of the analysis of diagnostic yield for the two groups are shown in Table 2 .
|
|
|
|
| Discussion |
|---|
|
|
|---|
While the role of EBUS-driven biopsies in central malignancy diagnosis or in the evaluation of lymph nodes involvement is well established, there are few studies investigating its role in the diagnosis of peripheral lesions. Although those studies141516 have pointed out a role of EBUS in increasing the diagnostic yield of TBBs, some of them had limitations: they were retrospective or crossover studies, or they analyzed the role of EBUS in improving the diagnostic efficacy of other procedures.
The aim of our study was to answer the question of whether EBUS can improve diagnostic yield of TBBs performed for malignant peripheral pulmonary lesions diagnosis. To accomplish this, we compared EBUS-TBB vs standard CT-driven TBB by analyzing homogeneous groups of individuals. For this purpose, we designed a longitudinal prospective study in which we screened 799 consecutive patients who were referred to our units for peripheral lesions.
Lung cancer is a potentially lethal neoplasm in which the prognosis depends on the early diagnosis of the lesion, as the 5-year global survival rate in patients in the early stages of disease is 38 to 67%, and in later stages 1 to 8%.17 The management of patients with small peripheral pulmonary lesions is controversial, and patients with lesions highly suspicious for cancer and eligible for surgery should undergo excision biopsy and lobectomy if cancer is confirmed. However, there are subsets of patients in whom histologic diagnosis is recommended before deciding on any further approach; individuals not eligible for surgery and patients with metastatic disease would benefit from a rapid histotype diagnosis.18
Recently, Vilmann and Larsen19 suggested performing endoscopic ultrasound-guided biopsies in positron emission tomography-positive patients due to the high false-positive rate observed. A previous study20 using a panel of standard bronchoscopic procedures obtained a good rate of diagnosis in lesions > 3 cm and a low diagnostic yield in lesions < 3 cm in diameter. In our study, we analyzed patients with lesions > 3 cm, < 3 cm, and < 2 cm.
The importance of EBUS in the diagnosis of peripheral lesions was observed for diameter sizes < 3 cm and even more for lesions < 2 cm, for which, according to the literature, this procedure maintained an elevated efficacy (75% and 71%), while TBB showed a further decrease in yield (30.7 and 23.3%, respectively) [p < 0.001].21 More recently, imaging guidance techniques such as CT fluoroscopy have been utilized with good results in order to improve the diagnostic yield of traditional fibrobronchoscopic biopsies, which are fairly blinded techniques.222324 However, this approach is time consuming and costly, exposes both personnel and patients to radiation, and requires several safety procedures to avoid radiation effects.25
Transthoracic needle aspiration (TTNA) has shown a high sensitivity in peripheral lung lesions and its use is suggested as a diagnostic procedure26. On the other hand, fiberbronchoscopy is an important step in the diagnostic algorithm of lung cancer and its use allows the performance of several diagnostic procedures at the same time (eg, brushing, BAL, transbronchial needle aspiration, TBB),4 while TTNA does not provide information about a possible central involvement. We believe that, especially for lesions < 2 cm, EBUS-TBB may have a diagnostic yield rivaling that of TTNA and could be helpful to shorten diagnostic time in patients with suspected lung cancer. Although an elevated diagnostic yield may be obtained by using more aggressive approaches, these procedures have the disadvantages of a higher percentage of complications.2728293031
Our study had some limitations. First, it was a longitudinal randomized clinical trial performed in a single center of thoracic endoscopy. We believe that further prospective multicenter studies may be important to confirm our data. Second, because of the overwhelming importance of the early diagnosis of lung cancer, we focused our study on the diagnostic yield of EBUS-TBB or TBB in the subset of patients with malignant lesions. However, we found similar results comparing the two procedures on the diagnosis of benign lesions (data not shown). Third, although a 1:1 randomization protocol was designed, a higher number of patients undergoing EBUS-TBB were unavailable or did not complete the follow-up period. However, a similar number of procedures was performed by the two bronchoscopists with a similar rate of sensitivity, specificity, and accuracy (data not shown). Furthermore, the analysis of the measured characteristics between the study group and the control group did not show any difference (data not shown).
Our study is, to date, the largest prospective randomized study comparing a standard bronchoscopic procedure and EBUS, and it shows significantly higher sensitivity, accuracy, and NPVs for EBUS-driven biopsies. In conclusion, our study suggests that, at least in selected subsets of patients, EBUS-TBB is useful, cost beneficial, safe, minimally invasive, and accurate. EBUS is a powerful tool that could be important in the early diagnosis and staging of small-sized peripheral lung cancers.
| Acknowledgements |
|---|
| Footnotes |
|---|
Received for publication August 12, 2004. Accepted for publication June 21, 2005.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
F. D. Sheski and P. N. Mathur Endobronchial Ultrasound Chest, January 1, 2008; 133(1): 264 - 270. [Abstract] [Full Text] [PDF] |
||||
![]() |
N. Yamada, K. Yamazaki, N. Kurimoto, H. Asahina, E. Kikuchi, N. Shinagawa, S. Oizumi, and M. Nishimura Factors Related to Diagnostic Yield of Transbronchial Biopsy Using Endobronchial Ultrasonography With a Guide Sheath in Small Peripheral Pulmonary Lesions Chest, August 1, 2007; 132(2): 603 - 608. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Yoshikawa, N. Sukoh, K. Yamazaki, K. Kanazawa, S.-i. Fukumoto, M. Harada, E. Kikuchi, M. Munakata, M. Nishimura, and H. Isobe Diagnostic Value of Endobronchial Ultrasonography With a Guide Sheath for Peripheral Pulmonary Lesions Without X-Ray Fluoroscopy Chest, June 1, 2007; 131(6): 1788 - 1793. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Eberhardt, D. Anantham, F. Herth, D. Feller-Kopman, and A. Ernst Electromagnetic Navigation Diagnostic Bronchoscopy in Peripheral Lung Lesions Chest, June 1, 2007; 131(6): 1800 - 1805. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. R. Gildea, P. J. Mazzone, D. Karnak, M. Meziane, and A. C. Mehta Electromagnetic Navigation Diagnostic Bronchoscopy: A Prospective Study Am. J. Respir. Crit. Care Med., November 1, 2006; 174(9): 982 - 989. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |