Keywords
bronchoscopy - EBUS - EUS - FNA - lung
Introduction
Endoscopic ultrasound (EUS) has evolved as a diagnostic and staging tool for gastrointestinal and peri-luminal structures like lymph nodes, pancreas, adrenal gland, gallbladder, bile duct, liver, kidney, lung, etc.[1]
[2]
[3] EUS fine needle aspiration (FNA) with its high diagnostic accuracy can be the modality of choice in sampling peri-luminal suspicious lesions.[1] The diagnostic approach for lung lesions after initial imaging by either computed tomography (CT) or positron emission tomography (PET) is to acquire sample by bronchoscopy, transthoracic needle aspiration lung biopsy, endosonography-guided FNA (EUS-FNA), endobronchial ultrasound-guided FNA (EBUS-FNA), thoracenteses, or by surgery. To diagnose and stage suspicious lung parenchymal lesions and to detect mediastinal lymph nodal metastasis, combination of EBUS and EUS is a highly effective modality.[4] EBUS and EUS can sample lymph node stations 2 R, 4 R, 10, 11 and stations 2 L, 3 P, 4 L, 5, 7, 8, 9 along with left lobe of liver, celiac axis, and left adrenal gland, respectively.[4] Both EBUS and EUS can be used to sample lung parenchymal lesions.[4] Transesophageal EUS-guided FNA can be used for obtaining tissue from centrally placed lung lesions.[5] The earlier studies have described the utility of EUS and EBUS-guided FNA/fine needle biopsy (FNB) in diagnosis of lung lesions. There is paucity of data in the diagnosis of failed EBUS or bronchoscopy-guided biopsy of lung lesions.
The aim of the present study was to evaluate the feasibility and safety of EUS-FNA of lung parenchymal lesions in nondiagnostic bronchoscopy or EBUS-FNA.
Methods
All the consecutive subjects undergoing mediastinal EUS between January 2021 and January 2024 were retrospectively analyzed from prospectively maintained database at a tertiary care university hospital. Contrast-enhanced CT (CECT) or PET-CT was used to determine the location of the lesion and the relation to the esophagus. Patients who underwent EUS-FNA from solid central lung parenchymal lesions or lung with adrenal lesions after a failed EBUS or bronchoscopic biopsy with available histology were included. Central mediastinal lung lesions were defined as those with the closest margin <2 cm in close proximity to the esophageal wall based on pre-EUS-CT or PET-CT to potentially permit transesophageal EUS-FNA.[6] EUS-FNA from mediastinal/abdominal nodes and isolated left lobe liver lesions were excluded from the study. The etiological, clinical, and investigation details were recorded on uniform structured data forms. The general prerequisites for FNA were similar to those for any therapeutic endoscopic procedure.[7]
Aspirin was continued and clopidogrel was stopped for 5 days prior to the procedure. After obtaining written informed consent, EUS was performed using either an Olympus linear EUS GF-UCT 180 scope (2011) connected to Olympus EUS machine EU- ME3 (2023) or using a Pentax curvilinear array echo endoscope (EUS) EG-3870 UTK connected to a Hitachi Avius estiva ultrasound machine (Pentax, Tokyo, Japan, 2012). We used 22G EUS-FNA needles (Wilson-Cook Corporation, Winston-Salem, North Carolina, United States) with removable stylet.[3]
Written informed consent was obtained from all the patients after explaining the disease and the need for EUS-FNA and its complications. The study was performed conforming to the Helsinki Declaration of 1975, as revised in 2000 and 2008 concerning human rights.
EUS Lung and Node Station
EUS Lung and Node Station
EUS-FNA was done under conscious sedation with the assistance of an anesthesiologist by a single echoendoscopist. The linear EUS scope was passed into the stomach and imaging was done of the liver, intra-abdominal nodes, and left adrenal gland, and the scope was withdrawn for evaluation of posterior mediastinum.
EUS Adrenal Station
Linear EUS was used for imaging the left adrenal gland by trans-gastric and right adrenal gland by trans-duodenal route. Adrenals were identified as seagull-like structures above the kidney. After celiac take off from the intra-abdominal descending aorta, a clockwise torque was used to identify the left adrenal gland. From the second part of duodenum, gentle withdrawal and counterclockwise torque would identify the right adrenal gland anterior to the right kidney. If lung and adrenal mass were noted, two different needles were used for sampling.[8]
Tissue Sampling
Once the mass was identified and measured in maximal cross-sectional diameter, a 22G needle with stylet was advanced into the lesion without any intervening lung tissue. Color Doppler was used to avoid intervening blood vessels in the anticipated needle path. After removing the stylet and using negative suction, 10 to and fro motions were performed within the lesion. The amount of subsequent suction used was either 5 to 10 mL depending on the initial aspirate, whether it is bloody or sparse. The entire collected material was put in a formalin bottle and a maximum of three passes were performed and representative material was sent to histopathologist. In the pathology laboratory, the tissue cores were fixed in 10% buffered formalin and processed in the usual way prior to embedding in paraffin wax. Serial sections were cut at 3 µm thickness and stained with hematoxylin and eosin. When necessary, deeper serial sections were cut. Special stains such as Ziehl-Neelson stain were performed as required. Rapid on-site evaluation was not done in the study. A combination of cytology and histology has been shown to improve diagnostic yield of EUS-FNA in different centers.
The definitive EUS-FNA histologic diagnosis was based on the sample review and a positive test meant positive for malignancy. All other interpretations like suspicious, negative, and atypical were considered negative for tissue diagnosis and the procedure was repeated again.
All the patients were monitored for 2 hours post-procedure for early complications with symptomatology, pulse rate, blood pressure, and oxygen monitoring. After 5 days at the histology review, standard physical examination and chest radiograph were done for all the patients to identify any late complications.
Continuous variables are given as the mean and standard deviations. Categorical variables were given in total and as percentages.
Results
Sixty-two mediastinal EUS procedures were performed during the study period. The flow chart depicts the final number included in the study ([Fig. 1]). CECT/PET-CT revealed mass lesions adjacent to the esophagus <1 cm. A total of 13 patients underwent EUS-guided lung biopsy from central lung lesions with failed EBUS FNB. Of which, two patients also underwent left adrenal biopsy. The demographic data of the study group are presented in [Table 1]. A confirmed histologic diagnosis was obtained in all 13 patients with only 1 requiring a repeat biopsy as the sample was inadequate. Both the biopsies from adrenal gland were positive for metastatic small-cell carcinoma. All the 12 patients were referred for consultation liaison with medical oncology with personalized chemotherapy after molecular tests. The one patient diagnosed with tuberculosis improved after anti-tubercular therapy and is well after 1 year of follow-up. There were no adverse events with EUS-FNA of lung lesions.
Fig. 1 Study flow chart showing patient recruitment in the study.
Table 1
Demography of the subjects who underwent EUS lung procedure
|
Sl. No.
|
Variable,
N = 13 (100%)
|
Frequency (%)
|
|
1
|
Male:female
|
7 (53.8%):6 (46.2%)
|
|
2
|
Age: mean ± SD (range) in years
|
49.6 ± 8 years (35–75 years)
|
|
3
|
Indication for EUS procedure
|
|
a. Failed EBUS FNA
|
13 (100%)
|
|
b. Solid central lung lesions
|
11 (84.8%)
|
|
c. Adrenal lesion + central lung lesion
|
02 (15.2%)
|
|
d. Mediastinal node + lung lesion
|
10 (76.9%)
|
|
4
|
Comorbidities
|
|
a. Hypertension
|
8 (61.5%)
|
|
b. Diabetes Mellitus
|
7 (53.8%)
|
|
c. Heart disease
|
2 (15.3%)
|
|
5
|
Concurrent medicines
|
|
a. Aspirin
|
3 (23%)
|
|
b. Clopidogrel
|
2 (15.3%)
|
|
6
|
History of smoking
|
8 (61.5%)
|
|
a. Active smoker
|
5 (38.4%)
|
|
b. Nonsmoker
|
5 (38.4%)
|
|
c. Reformed smoker
|
3 (23.0%)
|
|
7
|
Site
|
|
a. Right lung: upper
|
2 (15.3%)
|
|
b. Right lung: lower
|
3 (23%)
|
|
c. Left lung: lower
|
8 (61.5%)
|
|
d. Left adrenal ([Fig. 2])
|
2 (15.3%)
|
|
8
|
Size of the lesion
|
|
a. Lung, mean ± SD in cm
|
3.6 ± 1.1 cm
|
|
b. Adrenal, in cm
|
2.8 cm and 3.2 cm
|
|
c. Mediastinal node, mean ± SD in cm
|
|
|
9
|
Pathology
|
|
a. Squamous cell carcinoma ([Fig. 3])
|
5 (38.4%)
|
|
b. Small-cell carcinoma ([Fig. 4])
|
7 (53.8%)
|
|
c. Granulomatous inflammation ([Fig. 5])
|
1 (7.6%)
|
|
10
|
Complications within 7 days
|
0
|
|
11
|
Repeat biopsy as the sample was inadequate
|
1 Lung biopsy (7.6%)
|
Abbreviations: EBUS, endobronchial ultrasound; EUS, endoscopic ultrasound; FNA, fine needle aspiration; SD, standard deviation.
Fig. 2 (a and b) Right and left adrenal lesions on CECT abdomen; (c) EUS image of left adrenal mass and left kidney beneath; (d) EUS image showing FNA from the left adrenal mass; (e) histopathology image confirming metastatic adenocarcinoma. CECT, contrast-enhanced computed tomography; EUS, endoscopic ultrasound; FNA, fine needle aspiration.
Fig. 3 (a) CECT thorax showing right lower zone mass; (b and c) EBUS bronchoscopy and FNA of lung lesion; (d and e) EUS image of lung lesion and FNA from the lesion; (f) histopathology image showing fragmented keratin pearls suggestive of squamous cell carcinoma. CECT, contrast-enhanced computed tomography; EBUS, endobronchial ultrasound; EUS, endoscopic ultrasound; FNA, fine needle aspiration.
Fig. 4 (a and b) CECT thorax arterial and venous image showing right lower zone mass with central necrotic area; (c) PET CT showing metabolically active soft tissue mass and node at right infra-hilar region; (d) EUS image of node and lung mass; (e) histology image showing small cell carcinoma. CECT, contrast-enhanced computed tomography; CT, computed tomography; EUS, endoscopic ultrasound; PET, positron emission tomography.
Fig. 5 (a) CECT arterial-phase image showing soft tissue mass arising from right middle lobe with multiple mediastinal nodes; (b and c) EUS from the soft tissue mass and FNA from the lung mass; (d and e) histopathology image showing caseating granulomatous inflammation. CECT, contrast-enhanced computed tomography; EUS, endoscopic ultrasound; FNA, fine needle aspiration.
Discussion
EUS has an important role in evaluation, diagnosis, and treatment of luminal gastrointestinal lesions and pancreatico-biliary lesions. However, EUS also extends to extra-luminal structures.[2] In this article, trans-esophageal EUS FNA was used to diagnose pulmonary lesions, which were non-diagnostic on bronchoscopy and/or EBUS-FNA.
The diagnostic approach of suspicious lung lesions is determined by location and size of lesion on imaging. Peripheral lesions can be approached by CT-guided transthoracic needle biopsy. Central lesions can be approached by bronchoscopy or EBUS-FNA or EUS- FNA.[9] Endobronchial lesions can be biopsied by bronchoscopy and lung parenchymal lesions by EBUS-FNA. Trans-esophageal EUS-FNA cytology and biopsy are technically feasible for intra-parenchymal lung lesions.[2]
Although EUS and EBUS are complimentary in lung cancer staging, EUS has specific advantages. EUS with 150 to 180° has better ultrasonic angle as compared with EBUS with 50 to 60° angle. EUS is better tolerated with no cough or hypoxemia. The EUS image has better visualization of smaller structures due to higher resolution and close approximation of esophageal mucosa by suction. EUS is a shorter procedure with lesser sedation and lesser oxygen requirement than EBUS. EUS-FNA is also better as there is no intervening cartilage and needle maneuverability is better with the presence of an elevator. In addition, EUS is excellent for sampling sub-diaphragmatic, left and lower para-esophageal structures and concurrent adrenal lesions.[10] The maximum length of needle that can reach the target in EBUS is 4 cm and in EUS it is 8 cm. There was no reported pneumothorax in spite of intervening normal lung tissue.[9]
Central or peri-esophageal lung lesions can be approached better with EUS-FNA in fragile/hypoxic patients, poor candidates of trans-bronchial procedures, larger amount of tissue needed for molecular profiling, and nondiagnostic bronchoscopy or EBUS.[11] The pooled adequacy of EUS tissue acquisition is 95.4% and an accuracy of 93.4% in a meta-analysis of 19 studies with a total of 640 patients.[12] The pooled probability of obtaining an adequate sample in EBUS-FNA was 94.5% in a systematic review of 33 studies and 2,698 patients.[13] In our study, all the 13 patients had diagnostic EUS-FNA in failed EBUS-FNA or bronchoscopy and biopsy.
The overall adverse rate with EUS lung biopsy reported was 5.36%, common being bleeding or pneumothorax, which might require bronchoscopy or chest tube drainage.[11]
With the above advantages, EUS could be considered as a first line for the diagnosis and staging of pulmonary lesions, evaluation of tumor invasion (T4), mediastinal lymph node staging, pleural fluid evaluation (M1a), assessment of adrenal gland (M1b), and restaging after chemo-radiation. The routine frequency used is 7.5 MHz and the frequency can be adjusted if necessary. The sensitivity of EUS-FNA increases if the lung tumor was >25 mm in size and EUS-FNA is contraindicated in mediastinal cyst aspiration due to the risk of mediastinitis.[14]
The adrenal gland is commonly involved in metastatic lung primary cancer and EUS-FNA from the adrenal gland has a diagnostic yield of 100% without any complications.[8] In our study, two cases had diagnostic EUS-guided left adrenal biopsy without any adverse events.
The strength of the study is the extension of EUS services to diagnose pulmonary and adrenal lesions in nondiagnostic EBUS-FNA. The limitations of the study being the retrospective design, small number of patients, single-center study, limited clinical follow-up, and biopsy of only peri-esophageal lesions. The study did not compare diagnostic yield of EUS and EBUS-FNA.
Conclusions
This report suggests EUS-guided FNA of central lung lesions is technically feasible and may not necessarily result in adverse events like bleeding, pneumothorax, or infection. In all the cases, the lung lesion was less than 2 cm from the esophageal wall and not in direct contact, which may not be mandatory for tissue sampling. In failed EBUS or bronchoscopy cases, EUS can be a complimentary alternative for tissue acquisition in lung lesions.
