Patients
This study retrospectively analyzed medical data of consecutive patients with enlarged mediastinal and/or hilar lymphadenopathy and tumors reachable by EBUS-TBNA at Kasukabe Medical Center between July 2016 and December 2017. This study was approved by the institution review board (IRB no: 2017-013), and all participants provided written informed consent.
Lymphadenopathy was defined as an enlargement (>5 mm in short axis diameter) on chest computed tomography (CT) or increased [18F]-fluorodeoxyglucose uptake on positron emission tomography-CT (maximum standardized uptake value >2.5). Prior to EBUS-TBNA, each patient underwent a 1-mm slice chest CT scan, and EBUS-TBNA was performed using 22-gauge SS needles (Vizishot® NA-201SX-4022, Olympus, Tokyo, Japan, Fig. 1A and C) used on Mondays and Thursdays (the SS group), and 22-gauge CC needles (ExpectTM Pulmonary E00558220, Boston Scientific Corporation, Natick, MA, USA, Fig. 1B and D) used on Tuesdays, Wednesdays and Fridays (the CC group). Patient clinical characteristics, target lesions, procedure times, complications, diagnostic yield per lesion (including histology and cytology) and successful achievement of diagnostic histological specimens per puncture were compared between these two groups. Diagnostic yields of histological findings of the specimens in these two groups were also compared.
EBUS-TBNA procedure
Needles
A SS needle tip, available as 21- or 22-gauge (Vizishot®, Fig. 1A and C) and a CC needle tip, available as 25- or 22-gauge (ExpectTM Pulmonary, Fig. 1B and D) were used.
Procedure
EBUS-TBNA were performed using a convex probe ultrasound bronchoscope (CP-EBUS BF-UC260FW; Olympus), and only one needle type (SS or CC) was used for each patient. All the bronchoscopic procedures were performed by bronchoscopists with more than 5 years of bronchoscopic experience, including extensive experience of EBUS-TBNA, similar to the previous reports [4,7]. Under local anesthesia with midazolam and opioids during the procedure, bronchoscope was orally inserted without intratracheal intubation, and pharynx was initially locally anaesthesized by spraying 4% lidocaine (5 ml), followed by intermittent application of additional 2% lidocaine to the vocal cords, trachea and bronchi via the bronchoscope.
A convex probe transducer of bronchoscope can scan target lesions parallel to the direction of the bronchoscope insertion, and the ultrasound images were evaluated using a dedicated ultrasound processor (EU-ME2 PREMIER, Olympus) in B-mode, with the vascular pattern assessed using power Doppler ultrasound. An internal stylet was used and slightly pulled out to facilitate the punctures. Under real-time EBUS guidance, the target lesion was punctured by the needle through the tracheobronchial wall; the stylet was then deeply pushed into the needle and removed to prevent contamination of the airway cells, bronchial walls and tracheobronchial cartilage [16]. The needle was moved back and forth about 20 to 30 times within the target lesion while applying a negative pressure with a disposable 20-ml syringe. After sampling the tissue, the suction was released and the needle was removed from the bronchoscope. The time from insertion of the bronchoscope into the vocal cords to removal was recorded as the procedure time. The procedure involved three punctures or the collection of two core tissue samples per lesion. After each puncture, the stylet was inserted into the needle and the collected specimen was pushed onto a glass slide and fixed with 10% neutral buffered formalin for histopathological evaluation. The remaining needle aspirate was then blown onto the glass slide and smeared for cytological evaluation. The needle was rinsed with saline and the rinsed fluid was submitted for bacteriological examination. The final diagnosis for each patient was established from the pathological evidence obtained by EBUS-TBNA, surgery, microbiological analysis or clinical follow-up for at least 6 months. Rapid on-site cytology evaluations were not available for all patients. EBUS-TBNA-related complications defined as conditions that required hospital admission, including pneumothorax, hemorrhage, infection and air embolism were documented.
Pathological evaluation
Pathology was evaluated by independent cytologist and pathologist who were blinded to the needle type. Using a previously reported histological classification [12,13], the histological specimens obtained from the target lesions were categorized into three groups: Ⅰ, diagnostic (Fig. 2A); Ⅱ, non-diagnostic (e.g. blood clot (Fig. 2B) or cartilage (Fig. 2C)); Ⅲ, no specimens (Fig. 2D). The numbers of specimens classified as Ⅰ, Ⅱ and Ⅲ were used to calculate the diagnostic yield of histological specimens using the formula (Ⅰ / [Ⅰ + Ⅱ + Ⅲ]). Blood contamination and cartilage in the specimens were investigated in the present study; the numbers of specimens that contained some cartilage and the numbers of specimens that only contained blood clot or cartilage were compared between the two groups.
Statistical analysis
Descriptive statistics are presented as frequency with percentage or median (range). Clinical characteristics, procedure time, and numbers of complications were recorded per patient, and the diagnostic yield was calculated per lesion. Differences between the two groups were examined with Fisher’s exact test or the Mann–Whitney U test, based on a two-sided hypothesis. A P-value <0.05 was considered to be statistically significant. The statistical analyses were performed with EZR (Saitama Medical Center, Jichi Medical University, Saitama, Japan), which is a graphical user interface for R, version 2.13.0 (The R Foundation for Statistical Computing, Vienna, Austria), and a modified version of R commander (version 1.8-4).