CT-guided transbronchial biopsy using an ultrathin bronchoscope with virtual bronchoscopic navigation

Abstract

Chest 2004;125:1138–1143. Shinagawa N, Yamazaki K, Onodera Y, et al. Study Summary: In this study, the authors report their experience of performing computed tomography (CT)-guided transbronchial biopsy (TBB) with an ultrathin bronchoscope with navigation by virtual bronchoscopy in 25 patients with 26 small peripheral lung nodules. The average diameter of lung nodules was 13.2 mm. Nineteen lesions were not seen on the chest radiograph. CT examinations were performed using a multidetector CT scanner and virtual bronchoscopy images were reconstructed to fifth-order bronchi. The authors used an ultrathin bronchoscope with an external diameter of 2.8 mm and working channel of 1.2 mm. Information from virtual bronchoscopy was used to guide the ultrathin bronchoscope to the lung nodule. The biopsy forceps was introduced and its position was adjusted under real-time CT scan. Diagnostic tissue was obtained from 17 of 26 lesions (65%). The diagnoses were lung cancer (13 patients), atypical adenomatous hyperplasia (1 patient), and benign diagnosis (2 patients). Of the remaining 9 lesions, the specimen was inadequate in 6 cases and the lesion was inaccessible in 3 cases. Five of the 9 patients with nondiagnostic bronchoscopy had surgical resection. Lung cancer was found in 4 cases, and atypical adenomatous hyperplasia was found in 1 patient. The sensitivity, specificity, negative predictive value, positive predictive value, and accuracy of CT-guided TBB were 75%, 100%, 29%, 100%, and 77%, respectively. The average duration of bronchoscopy was approximately 30 minutes. The procedure was safe in all 25 patients. Virtual bronchoscopy guidance was felt to be accurate during actual bronchoscopy. Comments: Physicians and patients face difficult choices when a small lung nodule is detected incidentally or as part of a screening process. Although the majority of lung nodules detected during the screening process are benign, 1.3% to 3.1% of them represent early lung cancer (Radiology 2002;222:773–781, Lancet 1999;354:99–105, Am J Respir Crit Care Med 2002;165:508–513). Studies have shown that 5% to 15% of patients undergoing a screening process are referred for biopsy, and the majority of these patients receive a diagnosis of lung cancer (Ann Intern Med 2004;140:740–753). Although surgical resection is the gold standard for diagnosis, it is invasive and is not suitable for many patients with limited pulmonary reserve. In a recent study, surgery was unnecessary in as many as 20% patients (Am J Respir Crit Care Med 2003;168:1167–1173). A less invasive approach to management of small peripheral lung nodules is, therefore, highly desirable. Traditional bronchoscopy has a low diagnosis yield for solitary lung nodules, ranging from 10% to 33% for lesions smaller than 2 cm in diameter (Cleve Clin J Med 1998;65:315–326). An important reason for low yield of bronchoscopy in this circumstance is inability to visualize some of these lesions on fluoroscopy. Biopsy under CT fluoroscopy is more successful in such cases (Chest 2000;118:1630–1638). Another reason is failure to navigate the bronchoscope and the biopsy forceps to the intended target. A recent study found information from multiplanar CT reconstruction to be useful for more accurate navigation of biopsy instrument to the lesion (Chest 2003;124:1985–1992). Recent data also suggest a potential role of ultrathin bronchoscope in improving diagnostic yield from peripheral lung nodules. For instance, a recent study reported 70% diagnostic yield for lesions <3 cm when an ultrathin bronchoscope was use as an adjunct to standard bronchoscope (Respiration 2002;69:63–68). In the current study, virtual bronchoscopy, ultrathin bronchoscopy, and CT-fluoroscopy were used together, and the authors were able to obtain tissue diagnosis from 65% of lung nodules. This was impressive considering the size of the lesions biopsied. Clearly, this approach is safer than transbronchial needle aspiration or surgical biopsy. Some limitations of this study should be noted. The negative predictive value of bronchoscopy was only 30%. Therefore, negative bronchoscopy results have no clinical value. These patients will either need a more definitive biopsy or a close radiologic follow up. The other major issue is the risk of excessive radiation exposure to both the patient and the operator. Depending on the case volume, there is risk for considerable cumulative exposure, especially if appropriate radiation protection guidelines are not followed (Clin Chest Med 1999;20:33–38). The third major problem is logistic difficulty in scheduling and performing bronchoscopy in the CT suite. Undiagnosed lung nodules cause considerable patient anxiety. A firm diagnosis allows patients to make an informed decision, and it opens options for alternative treatments such as radiation therapy for nonsurgical candidates. In this context, the method used in this study has a better diagnostic yield than conventional bronchoscopy and is less invasive than surgical biopsy.