An accurate understanding of the position of the target lesion is essential to improve the diagnostic rate of TBNA, because failure to place the needle within the lesion is the leading cause of a low biopsy yield. The advent of new technologies such as EBUS and CT fluoroscopy has led to the concept of integration with TBNA to improve the diagnostic yield. Although EBUS images show target lesions beyond the airway, needle penetration of the target lesion could not be proved by EBUS using a single-channel bronchoscope. To confirm whether the target lesion was aspirated, rapid on-site cytopa-thology was needed immediately after aspiration. In this study, high diagnostic rates were established by EBUS-D without requiring rapid on-site cytopathology because the real-time EBUS image confirmed that the TBNA needle was within the lesion, in which it offered as a hyperechoic point. An advantage of EBUS-D is that if the TBNA needle is not placed correctly on the first penetration, relocation of the penetration site can be easily performed without replacing the TBNA catheter and EBUS probe through the working channel.
Factors reported to influence the diagnostic rate of TBNA were the site of the lymphadenopathy sampled, use of a histology needle, broncho-scopic findings at the penetration site such as carinal widening, extrinsic airway compression, submucosal invasion, and the increasing numbers of TBNA trials.’ In the present study, there were no significant differences for the EBUS-D and EBUS-S groups in the above conditions in each group. As rapid on-site cytopathology also contributes to the improvement of the diagnostic rate of TBNA, the use of rapid on-site cytopathology in our EBUS-S group might have led to a higher diagnostic rate. Several investigations recommended that the number of aspirations per lesion with a cytology needle should be four to seven, or at least three for each target lymph node in the conventional TBNA. Shannon et al reported that EBUS-guided TBNA exhibited a similarly high diagnostic rate to conventional TBNA, and decreased the number of aspirations for paratracheal lymph node sampling (mean [±SD], 2.00 ± 0.20 aspirations vs 2.91 ± 0.34 aspirations, respectively; p = 0.03). In the present study, the mean number of aspirations to establish diagnosis was 1.24 in the EBUS-D group. This means that EBUS guidance is useful in decreasing the number of aspirations when rapid on-cite cytopathology is not available, and that EBUS guidance spares patients from risks such as hemorrhage from unnecessary additional TBNA puncture.
Another adjunctive imaging technique to visualize the TBNA needle within the lesion, CT fluoroscopy guidance, has been reported. The diagnostic rate of CT fluoroscopy-guided TBNA (74.1% as reported by White et al, and 68.7% as reported by Garpestad et al) was lower than that of EBUS-guided TBNA reported by Shannon et al (82.6%), Herth et al (86%), and also our EBUS-D results. Moreover, the three to four penetrations per lesion that were required with CT fluoroscopy-guided TBNA and CT fluoroscopy has the disadvantage of radiation exposure to the patients and the operators. According to Goldberg et al, CT fluoroscopy confirmed that only 6 of the initial 18 needle passes (33%) were positioned properly within the target lesion. With CT fluoroscopy guidance, the rate of subsequent successful passes increased to 62%. In contrast, needle confirmation in the initial trial of EBUS-D was higher (75.9%), and 100% confirmation of needle placement within the lesion was achieved in a subsequent trial (data were shown). Additionally, although six penetrations (5.2%) punctured great vessels with CT fluoroscopy guidance, there was no accidental puncture in the EBUS-D group.
EBUS guidance using the double-channel bronchoscope also has its limitations. The double-channel bronchoscope (7.2 mm in diameter) may not be well tolerated by all patients. However, in our experience, transoral insertion encountered no serious problems, even though Japanese people are usually of slighter stature than many Westerners. No endotracheal tube was used, and this scope is used exclusively for EBUS-guided TBNA. Furthermore, since our transducer was of the radial type, we could not see the entire course of the needle within the lesion. We are currently performing TBNA using a convex type ultrasonic bronchoscope, which shows the image of the needle course within the lesion. Its outer diameter is 6.9 mm, and it has a small curved array transducer located in front of a 30° oblique forward viewing optic lens and a biopsy channel of 2 mm. It was reported to be useful in the diagnosis of thoracic diseases. It may be easier to penetrate the target lymph node using the ultrasonic bronchoscope than with the EBUS-D because the entire puncture procedure of the target lesion is performed under direct ultrasound guidance. However, it is difficult to obtain a histologic diagnosis with the convex type ultrasonic bronchoscope because only a 22-gauge needle can be used for sampling. The 22-gauge needle has been reported to be inferior to the 19-gauge needle in the diagnostic rate. Therefore, in cases requiring cytology such as staging of lung cancer and diagnosis for suspected malignancy, the ultrasound bronchoscope will be useful. Otherwise, EBUS-guided TBNA in this study could easily enable histologic diagnosis with the 19-gauge histology needle, and is able to establish the diagnosis of benign diseases such as sarcoidosis, silicosis, and tuberculosis.
Transesophageal lymph node sampling under endoscopic ultrasound guidance fine-needle aspiration can also generate the image of course of the needle. Endoscopic ultrasound through the esophagus gives access to the subcarinal, aortopulmonary, and posterior mediastinum lymph nodes. However, the images of paratracheal and anterior mediastinal lesions are limited by distortion caused by the intratracheal airspace, and the hilar lymph nodes cannot be accessed from the esophagus. In contrast, EBUS guidance allows access to such limited locations with under endoscopic ultrasound guidance fine-needle aspiration. Furthermore, the target lymph node can be detected more readily through the tracheobronchial tree because it has many points of orientation.
The weak point of our study was that the locations of the penetrated lymph nodes were not exactly equalized. If many patients with lymphadenopathy in the left paratracheal/aortopulmonary window had been included in this study, the diagnostic rate might be decreased because TBNA in this location is more difficult than in any other site. By visualizing TBNA needle placement, operators will have more confidence in attempting biopsies as well as less fear of puncturing major vessels, and the specimen obtained by use of the TBNA needle placement with real-time confirmation will yield an accurate diagnosis with a decreasing number of penetrations.