| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Institute of Diagnostic Radiology, Interventional Radiology and Nuclear
Medicine, BG Clinics "Bergmannsheil," Buerkle-de-la Camp Platz 1,
Ruhr-University of Bochum, Bochum D-44791, Germany.
2 Department of Pneumology, Allergology, and Sleep Medicine, Medical Clinic III,
BG Clinics "Bergmannsheil," Ruhr-University of Bochum, Bochum,
Germany.
3 Institute of Pathology, BG Clinics "Bergmannsheil,"
Ruhr-University of Bochum, Bochum, Germany.
Received July 1, 2005;
accepted after revision August 17, 2005.
Address correspondence to C. M. Heyer
(christoph.heyer{at}rub.de).
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. We included 33 consecutive patients (25 men; mean age ± SD, 64 ± 9.6 years) with solitary pulmonary nodules at different sites and with a lesion-to-pleura distance of at least 2 cm who previously underwent conventional bronchoscopy that did not result in histologic diagnosis. All patients were prospectively investigated with transbronchial bronchoscopic biopsy under MDCT guidance. Examinations were performed with the patient in conscious sedation using a low-dose protocol (80 kV, 20 mAs, 5-mm collimation, 10-mm slices). The position of the tip of the biopsy device was confirmed and documented before biopsies were performed. All specimens were examined by standard histopathologic techniques. The effective radiation dose was calculated for every patient.
RESULTS. The diagnostic yield was 24 in 33 selected patients (overall accuracy, 72.7%): 13 (54%) had primary lung cancer and 11 (46%) had benign diagnoses. The formal operative characteristics were sensitivity, 59%; specificity, 100%; positive predictive value, 100%; and negative predictive value, 55%. The final diagnoses of the remaining nine patients in whom transbronchial bronchoscopic biopsy was not diagnostic were non-small cell lung cancer (n = 3); small cell lung cancer (n = 3); and alveolar carcinoma, carcinoid tumor, and hemorrhaged bulla (n = 1 each). All nonmalignant diagnoses were confirmed by 6 months radiographic and clinical follow-up. The mean duration of the procedure was 39 ± 15 minutes, and the average effective dose was 0.7 mSv (range, 0.5-1.1 mSv). One case of pulmonary hemorrhage (3%) occurred after the procedure.
CONCLUSION. MDCT-guided transbronchial bronchoscopic biopsy is a promising and safe tool for the diagnostic pathway of solitary pulmonary nodules in previously undiagnosed patients. Image quality was sufficient with low-dose protocols, which resulted in low radiation exposure for patients and personnel.
Keywords: biopsy • bronchoscopy • CT guidance • lung cancer • MDCT • pulmonary nodules • radiation dose • tracheobronchial tree
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Bronchoscopy guided by conventional fluoroscopy is well established and is the traditional method for the assessment of centrally located solitary pulmonary nodules. The more peripheral an intrapulmonary lesion is located, the more difficult are visualization and successful biopsy [4, 5]. Moreover, routinely used bronchoscopes usually do not offer the opportunity to enter bronchial structures far beyond segmental bronchi. Transbronchial biopsy under fluoroscopy guidance is therefore often performed without definite visualization of the target lesion [6, 7].
Percutaneous transthoracic lung biopsy with CT guidance is an alternative procedure with which to establish histologic diagnosis in solitary pulmonary nodules. This tool is useful for reaching intraparenchymal lesions within the lung periphery, whereas complications increase when applying the procedure to lesions located centrally within the lungs.
|
|
The goal of this study was to establish efficacy and safety of MDCT-guided transbronchial bronchoscopic biopsy using a standard low-dose protocol in patients with solitary pulmonary nodules and suspected lung cancer who had undergone conventional bronchoscopic procedures that were not diagnostic.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
History and the clinical signs and symptoms justifying the suspicion of lung cancer were involuntary weight loss, fever not explained by infectious origin (e.g., pneumonia), night sweats, smoking, or occupational exposure associated with a higher incidence of malignant pulmonary diseases. All patients had undergone standard flexible bronchoscopy without definite diagnosis from the suspicious lesion and transesophageal sonography and biopsy for nodal staging. All biopsy samples obtained during these procedures had to be nondiagnostic for inclusion in the study.
Before the study, all patients underwent a complete clinical evaluation including history, physical examination, standard and symptom-targeted laboratory workup (including prothrombin time, partial thromboplastin time, and total bleeding time), ECG, echocardiography, and lung function test. Written informed consent was given at least 24 hours before the procedure by each patient after the nature of the procedure had been fully explained. All patients with negative biopsy results or a diagnosis other than lung cancer were followed up for at least 6 months to either establish that diagnosis or verify an alternate diagnosis. The study was approved by the Ethics Committee of the Ruhr-University of Bochum, Bochum, Germany.
|
|
|
After intubation, the correct position of the endotracheal tube was ensured bronchoscopically (Olympus), and a standard inspection of the tracheobronchial tree was performed to confirm the exclusion of intrabronchial masses. In the absence of such lesions, the tip of the bronchoscope was guided to the suspected segmental bronchus and a biopsy forceps (FB-19CR-1, Olympus) was positioned distally from the tip of the bronchoscope (Fig. 1). The position was then ensured with consecutive CT scans, and three to five biopsies were performed when imaging suggested an optimal position within the target lesion and acceptable distance to surrounding structures (e.g., vessels). Specimens were immediately fixed in formalin to avoid drying artifacts and were processed by standard histopathologic techniques.
Radiologic Procedure
All procedures were performed at the Institute of Diagnostic Radiology,
Interventional Radiology and Nuclear Medicine, BG Clinics
"Bergmannsheil," Ruhr-University of Bochum on an MDCT scanner. In
19 patients (patients 1-19), a 4-MDCT scanner (Somatom Volume Zoom, Siemens
Medical Solutions) was used; in 14 patients (patients 20-33), examinations
were performed on a 16-MDCT scanner (Sensation 16, Siemens). The patients were
positioned on the CT table in the supine position before endotracheal
intubation. After acquisition of a CT topogram, a complete scan of the thorax
using a low-dose protocol was performed (120 kV, 20 mAs, 2.5- or 1.5-mm
collimation, 6-mm slice thickness). During the biopsy procedure, the position
of the bronchoscope and the biopsy forceps was ensured by applying consecutive
singular scans using a standard low-dose protocol (120 kV, 20 mAs, 5-mm
collimation, 10-mm slice thickness) in each patient. These slices were
obtained following the bronchoscope; once the target area was reached, slices
were obtained at the level of the pulmonary nodule. Images were displayed in
lung window settings on a separate screen in the examination room to ensure
detailed visualization of the tracheobronchial tree
(Fig. 2). Application of each
CT slice was started by the attending radiologist in the examination room by
using a remote control pedal. Ten to 21 slices were necessary per patient to
ensure correct positioning of the biopsy device within (or within the vicinity
of) the target lesion (Figs. 3
and 4). After each procedure,
three single-detector CT slices with the parameters mentioned earlier were
obtained at the level of the lesion and 3 cm below and above to exclude
pneumothorax and pulmonary hemorrhage.
For each patient, effective radiation dose was calculated with a computer program (CT-Expo version 1.3, G. Stamm and H.-D. Nagel) considering the weighted CT dose index (CTDIw) and dose-length product, as displayed on the workstation [8].
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The average effective dose administered was 0.7 mSv (range, 0.5-1.1 mSv). Of 33 biopsies, 24 (72.7%) were diagnostic for the cause of the pulmonary lesion. Among those, the most frequent diagnoses were non-small cell lung cancer (NSCLC) (11/24, 46%), pneumonia (8/24, 33%), coal workers' pneumoconiosis (2/24, 8%), and small cell lung cancer (SCLC) (2/24, 8%). Histology was not explanative for the pulmonary lesion in nine (27%) of 33 biopsies and showed normal lung parenchyma in five (56%) of nine biopsies, chronic bronchitis in three (33%), and coal workers' pneumoconiosis in one (11%).
Confirmation of Diagnosis
A histologic diagnosis of lung cancer was rated definite, and patients were
treated according to histologic type and clinical stage. In all patients with
pneumonia, antibiotic therapy was initiated and serial chest radiographs were
obtained up to 6 months after the procedure to exclude pneumonia secondary to
lung cancer. At the 6-month follow-up, the pulmonary lesions had resolved in
all cases. Patients with coal workers' pneumoconiosis were followed up at
6-month intervals for 1 year, and the absence of progression of the pulmonary
lesion was considered to be a benign diagnosis.
In patients with nondirective histology results (normal, chronic bronchitis, and emphysema), an individual approach was taken. The final diagnoses and procedures are summarized in Table 1. In patients with satisfactory clinical condition and clinical findings strongly suggesting lung cancer, open lung biopsy (n = 1) or thoracotomy (n = 5) was recommended. Thoracotomy revealed lung cancer in three cases (SCLC, n = 2; NSCLC, n = 1). One patient showed a pulmonary metastasis of a malignant pleural mesothelioma (patient 21), and one patient had a carcinoid tumor (patient 14). Open lung biopsy revealed alveolar carcinoma in patient 9. One diagnosis could be established through endosonography and biopsies of the adrenal gland (patient 2). In patient 22, percutaneous transthoracic lung biopsy guided using CT was performed and showed histology of NSCLC. In patients with comorbidities and a high risk of respiratory failure after surgery or in those who refused to undergo open lung biopsy, 3-month follow-up CT was recommended. One patient died during follow-up, and the diagnosis of SCLC was established at autopsy (patient 18). Patient 28 had received anticoagulation therapy and was not a candidate for explorative thoracotomy because of severe emphysema. Follow-up CT scans showed complete resolution of the initial lesion, and in the absence of clinical signs of infection, a diagnosis of pulmonary hemorrhage was made.
Complications of the Procedure
All biopsies were performed without major complications. One patient
(patient 13) developed circumscribed intraparenchymal hemorrhage with mild
hemoptysis and a pulmonary infiltrate that was treated with antibiotics. No
patient suffered pneumothorax; thus, none of the patients required insertion
of a chest tube after the procedure.
Operative Characteristics of the Procedure
In our study population of patients with suspected lung cancer who had
previously undergone unsuccessful transbronchial biopsy using conventional
bronchoscopy, we established histologic diagnosis of cancer in 13 (39%) of 33
patients, and we were able to exclude cancer as the cause of a radiologically
and clinically suspicious pulmonary lesion in another 11 (33%) of 33 patients.
The overall diagnostic yield of MDCT-guided transbronchial bronchoscopic
biopsy was therefore 72.7%. The value of the formal operative characteristics
is limited for this kind of investigation (e.g., no false-positive results for
malignant histology) but will be reported here to enhance the transparency of
the results: sensitivity, 59%; specificity, 100%; positive predictive value,
100%; and negative predictive value, 55%.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Lung cancer is one of the most common malignant tumors in our society, but specific clinical symptoms associated with an early diagnosis do not exist. Although lung cancer is symptomatic at presentation in most lung cancer patients, approximately 50% of these patients seek medical attention in the advanced stages of metastatic disease [9]. Thus, concepts of early diagnosis of lung cancer are of pivotal importance. Albeit benign diseases may be the cause of pulmonary nodules, 20-50% of randomly discovered pulmonary nodules remain malignant tumors—predominantly lung cancer. Although CT remains the method of choice for assessing pulmonary nodules, their imaging characteristics, and their relation to surrounding structures, CT cannot be used to establish a histopathologic diagnosis of solitary pulmonary nodules.
Since fiberoptic bronchoscopy was introduced in the late 1960s, it has evolved into the most common procedure for tissue diagnosis of pulmonary disease. The diagnostic yield of bronchoscopy-guided biopsies in pulmonary nodules has been shown to be 18-60% [4]. The diagnostic value of this procedure declines with decreasing lesion size and increasing distance between the intrapulmonary lesion and hilar structures [10]. Lung nodules that are located more distantly are difficult to reach via endoscopic biopsy because conventional fluoroscopy as a guiding tool is limited by suboptimal contrast resolution and superimposition of structures in the mediastinum. Baaklini et al. [4] showed that diagnostic yield in bronchoscopy under guidance with conventional fluoroscopy strongly depends on lesion size and location. In their study, diagnostic yield was 31% when the lesion was located within the inner two thirds of the lungs versus 14% when the lesion was located in the peripheral third. Other studies have confirmed these findings [5, 11].
Considering these major drawbacks, we evaluated a new image-based biopsy procedure for harvesting lung tissue in solitary pulmonary nodules that could not be visualized endoscopically. We included a total of 33 consecutive patients in whom conventional bronchoscopy had failed and either the tumors could not be reached by percutaneous transthoracic lung biopsy or the risk of pneumothorax was considered too high for this procedure. In this population, we established 24 of 33 diagnoses, all of which were confirmed by either a malignant histology or follow-up of at least 6 months.
Long acquisition times, low image quality, and exposure to radiation have previously limited the routine use of CT-guided endobronchial biopsies of pulmonary nodules. Rong and Cui [12] were among the first to show the use of standard CT to image the location of the bronchoscope in mediastinal lymph node biopsies. They reported a 60% sensitivity in 49 patients with malignant mediastinal adenopathy and noted substantial improvement in their diagnostic yield as compared with the sensitivity without CT (20%) [12]. This study, however, showed also the major limitations of standard CT including substantial delays in finding the appropriate slice position and the long image reconstruction time. The new generation of MDCT scanners offers rapid image acquisition and improved image quality making use of MDCT scanners feasible for endoscopically guided biopsies of pulmonary nodules.
The results of our study prove that MDCT-guided biopsy adds diagnostic value when bronchoscopy with conventional fluoroscopy has failed. The method may be an alternative to, for example, percutaneous needle aspiration in some patients, but the sensitivity for malignant lesions was low. Because of the nature of the procedure, we were able to reach only lesions that were at least partially located within the bronchi. This led to positioning failure in two (6%) of 33 patients. We used strict inclusion criteria and thus investigated a highly selected patient population. Inclusion of patients with lesions located closer to the hilus may improve this failure rate even further.
Therefore, percutaneous transthoracic lung biopsy is regarded as the gold standard and is widely accepted to obtain lung biopsies for pathologic examination in most patients. Accuracy for the diagnosis of benign and malignant diseases has been shown to be greater than 80-90% [13-21]. The method can be performed under sonographic, conventional fluoroscopic, or CT guidance. The latter has improved the ability to detect and perform biopsies of small pulmonary nodules, particularly lesions that were not previously visible or accessible with fluoroscopically guided biopsy [22-25]. One disadvantage of percutaneous transthoracic lung biopsy is decreasing diagnostic accuracy and increasing complication rate with growing target distance from the pleura [26]. Pneumothorax is by far the most common complication of percutaneous transthoracic lung biopsy besides rare events, such as coronary artery [27] or cerebral air [28] embolism. The rate of pneumothorax after percutaneous transthoracic lung biopsy with CT guidance ranges from 19% to 60% and is influenced by patient factors (age and sex [26]; lung function parameters [26, 29, 30]; and presence of emphysema [13, 31]), lesion variables (size [13, 32], depth [13, 26], contact with pleura [31], and location [31-34]), and procedure-related issues (duration of biopsy, size of needle used for biopsy, and angle of the needle trajectory [26]). Furthermore, Haramati and Austin [35] showed no pneumothorax occurring if the needle does not traverse aerated lung parenchyma.
The results of our study show that CT-guided bronchoscopy is a safe alternative tool in patients considered not suitable for or those who deny informed consent for percutaneous transthoracic lung biopsy. For example, severe emphysema is a relative contraindication for percutaneous transthoracic lung biopsy [36], and none of our patients developed pneumothorax. Intrapulmonary hemorrhage has been reported to occur in 5-17% of patients after CT-guided percutaneous transthoracic lung biopsy [37, 38], and lesion depth has been identified as the most important risk factor [39, 40]. In our study, only one (3%) of the 33 patients developed circumscribed paralesional hemorrhage with mild hemoptysis, but those complications could be managed without operative intervention. It must, however, be kept in mind that we used strict inclusion criteria and in order to directly compare the side effects of the procedures, a randomized controlled trial would be helpful.
CT fluoroscopy is an alternative for guidance of percutaneous transthoracic lung biopsy. Garpestad et al. [41] investigated 35 patients with transbronchial needle aspiration under real-time CT fluoroscopy focusing on mediastinal lymph nodes and found that adequate tissue could be obtained in 88%. White et al. [42] reported a diagnostic rate of 83% in mediastinal lesions and 67% in lung lesions, whereas another study reported sufficient tissue for analysis in 96% of patients with an overall diagnostic yield of 90% [26]. However, the complication rate was high in the latter study (44%) with pneumothorax as the most common complication (42%) [26]. Although CT fluoroscopy is easy to perform and the duration of the procedure decreases with increasing experience of the radiologist, it tends to be associated with high radiation exposure. In comparison with conventional fluoroscopy, the dose from CT fluoroscopy is at least five times greater [42]. In our study, we consequently used a low-dose protocol enabling us to dramatically decrease radiation exposure to both patient and investigators compared with standard CT. Although image quality is somewhat decreased, this did not affect the diagnostic efficiency because the tip of the bronchoscope, the biopsy device, and the target lesion were clearly visible on all scans.
MDCT-generated virtual bronchoscopy is a new imaging method that noninvasively provides insights into the tracheobronchial system. Some investigators showed that virtual bronchoscopy can add important information in patients with lung cancer and endoluminal tumor growth [43-46]. However, virtual endoscopy has not yet been routinely combined with MDCT-guided transbronchial biopsy. Further studies must be undertaken to evaluate whether this promising 3D imaging tool might significantly improve the diagnostic work flow in patients with lung cancer.
The small sample size of our study is a possible limitation, but we consider the study large enough to establish CT-guided bronchoscopy as an additional diagnostic tool. Onsite pathology, which could have further improved the diagnostic yield of this method, was not available [47, 48]. In primarily nondiagnostic investigations, a final diagnosis was pursued by means of open lung biopsy and other invasive methods. In patients too sick for this approach or those who refused to undergo further diagnostic tests, we followed up clinically for at least 6 months to confirm the diagnosis.
In conclusion, our study shows that biopsies obtained with MDCT-guided bronchoscopic sampling had a high diagnostic yield in patients with radiologically and clinically suspicious pulmonary lesions who had already been examined unsuccessfully with standard bronchoscopy and fluoroscopy. Radiation exposure was low compared with standard and CT fluoroscopy and the complication rate was low. We suggest that MDCT-guided bronchoscopic sampling should be considered as an alternative tool for diagnostic decision making in patients with solitary pulmonary nodules.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  M. C. B. Godoy, D. Ost, B. Geiger, C. Novak, D. Nonaka, I. Vlahos, and D. P. Naidich Utility of Virtual Bronchoscopy-Guided Transbronchial Biopsy for the Diagnosis of Pulmonary Sarcoidosis: Report of Two Cases Chest, September 1, 2008; 134(3): 630 - 636. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Oki, H. Saka, C. Kitagawa, S. Tanaka, T. Shimokata, K. Mori, and S. Kajikawa Novel thin bronchoscope with a 1.7-mm working channel for peripheral pulmonary lesions Eur. Respir. J., August 1, 2008; 32(2): 465 - 471. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
