Introduction

Mediastinal lymph node enlargement can occur in several benign diseases such as sarcoidosis and tuberculosis and also in malignant diseases such as lung cancer, lymphoma, or metastases of other primary origin. Lymph node staging is important with regards to the therapeutic and prognostic implications this carries for the patient.

Computer tomography is often the first staging tool when lung cancer is suspected and lymph nodes at a size of over 1 cm are deemed suspicious for metastases. The sensitivity and specificity of using CT alone is low especially in lymph nodes under 2 cm. The prognostic accuracy increases with the use of PET-CT but even then it is important to sample enlarged lymph nodes and obtain a tissue diagnosis before proceeding to potentially curative surgery. For many years mediastinoscopy has been the gold standard for the investigation and staging of the mediastinum. But it is an invasive and expensive procedure with associated morbidity and mortality. Over the last few years endosonography has been increasingly used in the staging of lung cancer as well as the diagnosis of unexplained mediastinal lymphadenopathy.

Gastroenterologists have been using endosonography for many years in the staging of oesphageal carcinoma and in the 1990s the same technique was adapted as a diagnostic tool in the airways. EBUS and EUS (endoesophageal ultrasound) are these days routinely used and allow for the visualization and sampling of mediastinal lymph nodes that lie adjacent to the trachea or oesophagus respectively. They are not only used in the diagnosis of benign diseases such as sarcoidosis, tuberculosis, or inflammatory processes but also in malignant diseases. With EUS the left adrenal gland and lesions in the left lobe of the liver can also be sampled to allow for metastatic staging.

EBUS Scope

An EBUS (endobronchial ultrasound) scope is similar to a regular scope and uses video and fiberoptic technologies to obtain a picture. In addition to a regular bronchoscope a curved linear array 7.5–12 MHz ultrasonic transducer is mounted at its distal end (Figure 13.1). Models are available through different companies and currently the smallest available diameter is 6.3 mm with a 2.2-mm working channel.

Figure 13.1 Distal tip of a EBUS bronchoscope.

The endoscopic picture is supplied by a 30° angle optic and hence introducing this bronchoscope past the vocal cords can be a little more challenging than a normal videobronchoscope at the beginning. The EBUS scope is wider than a regular bronchoscope and access to the more distal airways can be difficult. The distal ultrasonic tip can be fitted with a water inflatable balloon to improve contact with the mucosal wall and visualization of the structures lying beyond. This can be useful when mucosal contact is otherwise poor or cartilage is overlying the targeted lymph node. Use of the balloon can be particularly useful in the trachea when the procedure is performed under light sedation only to avoid coughing.

By pushing the distal tip against the mucosal wall the ultrasonic picture can be obtained. The ultrasound image is obtained with a dedicated ultrasound processor and color Doppler mode can also be displayed in order to differentiate vascular structure from solid tissue (Figure 13.2).

Figure 13.2 Sonographic picture with Doppler mode activated.

Identified structures can be visually enlarged and measured for documentation during the procedure. The EBUS scope allows for the visualization of lymph nodes but also for tumour infiltration or imaging of other structures adjacent to the trachea or main bronchi.

When a lymph node is identified it should be measured and its echogeneity should be documented. Lymph nodes of under 1 cm, of oval shape with visible presence of the hilar structure are more likely to be benign whereas lymph nodes over 1 cm with low echogeneity and necrosis tend to be malignant nodes. It is important to remember however that echo features are not absolute measures and can only provide guidance with regards to which node to biopsy.

Once a structure has been identified it can be punctured with a dedicated 21 or 22 gauge needle under continued direct visualization. The needle is passed into the working channel and secured in place. It is important to adjust the sheath containing the needle for it to be just visible in the endoscopic picture. This has to be performed with the distal tip in a straightened position in order to avoid damage to the distal bronchoscope. Once the needle is advanced into the working channel the distal tip becomes stiffer and may be more difficult to flex. The lymph node is punctured under direct visualization and can be seen on the endoscopic as well as the ultrasound picture (Figure 13.3). A stylet within the needle avoids sampling of bronchial mucosa during the puncture and has to be removed after puncturing the lymph node.

Figure 13.3 Needle within lymph node.

Suction is applied via a 20 ml vacutainer syringe. Aspirations can also be performed by simply withdrawing the stylet a few centimeters and aspirating with the use of capillary forces within the needle. The use of the vacutainer syringe in comparison to not using suction has been studied by Casal et al and no difference in sample yield or quality has been found.

During the aspiration it is recommended to advance and withdraw the needle within the lymph node several times in order to obtain an adequate sample. No data has been published as to the recommended amount of times the lymph node should be traversed, but experts recommend approximately 8–20 times. Once this has been performed, the suction is removed, the needle withdrawn back into the sheath and secured, and the sheath and needle are removed from the working channel. The samples are then transferred onto glass slides or into formalin for a cell block.

It is recommended to puncture lymph nodes at least three times to obtain enough material for smears and a cell block. Routinely, two punctures are performed for cytosmear and one is expelled into formalin for a cytospin and cell block. If adequate samples are obtained these are often sufficient to perform analyses including immunohistochemical stains. If samples are small or dry aspiration is performed it is recommended to sample the node several more times. Close cooperation with the local cytopathologist is recommended in order to supply good samples and achieve the best results.

Some centers may have access to rapid on site evaluation (ROSE) or physicians may perform this themselves. ROSE has been shown to decrease the number of punctures required if core samples were obtained hence decreasing the intervention time. Lee et al were able to show that 75 percent of EBUS TBNA contained core tissue allowing for molecular testing including EGFR and ALK mutation analysis.

EBUS bronchoscopy can be performed under local anesthesia, mild sedation, or general anesthesia. Coughing can make it difficult for beginners to obtain adequate pictures and more sedation may be required. Yarmus et al compared moderate versus deep sedation and found that the diagnostic yield as well as number of lymph nodes sampled was greater in procedures performed under deep rather than only moderate sedation. Moderate sedation for this procedure is generally well tolerated by patients.

Anatomy

When using an EBUS scope it is important to understand the mediastinal anatomy and the areas which can be accessed by the EBUS scope. It is important to apply a systematic approach to the different lymph node stations in order to avoid oversight. Especially in patients with (suspected) lung cancer adequate lymph node staging is important to choose the most appropriate form of therapy for the patient.

The modified Mountain–Dresler lymph node map presents a good starting point to identify the different lymph node stations (Figure 13.4).

Figure 13.4 Mountain–Dresler lymph node map.

EBUS can reach the high mediastinal nodes (station 1), the upper paratracheal nodes (station 2), lower paratracheal nodes (station 4), subcarinal nodes (station 7) as well as hilar nodes (stations 10 and 11). In some patients stations 12 and/or 13 can also be reached. Endo-oesophageal ultrasound can reach positions 2, 4, 7, 8, and 9 as well as the left adrenal gland and the left lobe of the liver.

When performing a lymph node staging in suspected or known cancer patients it is important to initially sample the N3 lymph nodes, followed by N2, and then N1 in order to avoid contamination.

Whilst performing EBUS TBNA it is easiest to use anatomical landmarks visible in the endoscopic picture as well as vascular markings identified by endobronchial ultrasound.

Station 12R: The EBUS scope is advanced into the lower lobe bronchus proximal to branching of the lower basal bronchus. Station 12R is adjacent to the interlobular pulmonary artery.

Station 11R: Withdraw EBUS scope into the intermediate bronchus, turn clockwise, and press the tip to the upper lobe carina. Station 11R can be seen adjacent to the interlobar pulmonary artery.

Station 10R: Move the tip of the EBUS scope into the right upper lobe bronchus, turn clockwise to a 3 o’clock position, and withdraw the bronchoscope slightly.

Lymph node stations 10R and 4R are separated by the azygos vein. The lower border of the azygos vein indicates the top of lymph node station 10R, anything above the lower border is declared as lymph node station 4R

Station 4R: Withdraw the EBUS scope from here further and rotate slightly anti-clockwise. The station can be found just above the level of the main carina.

Station 2R: On withdrawing the bronchoscope further in the trachea keeping it in a 2–3 o’clock position, station 2R can be visualized.

Lymph node station 1R is difficult to visualize.

Station 12L: As on the contralateral side the bronchoscope is advanced into the basal segment and rotated clockwise and anticlockwise slightly. These nodes can be difficult to identify.

Station 11L: On withdrawing the bronchoscope and pushing it against the carina between the upper and lower lobe the lymph node can be visualized next to the pulmonary artery.

Station 10L: Retracting the bronchoscope into the left main bronchus into a 10 o’clock position lymph node 10L is seen in close proximity to the pulmonary artery.

Station 4L: The bronchoscope is withdrawn further in the left main bronchus and turned anticlockwise to a 10 o’clock position at the level of the main carina. Lymph node station 4L lies between the aorta and the left main pulmonary artery.

Station 2L: In the upper trachea turn the bronchoscope anticlockwise into a 11 o’clock position. Lymph node station 2L lies above the aortic arch (Figure 13.5).

Figure 13.5 a–ef Endobronchial equivalent of lymph node stations 2L (A), 2R (B), 4L (C), 4R (D), 10L (E), and 10R (F)

Complications

EBUS TBNA is a very safe procedure and complications observed are very rare. In a meta-analysis by Gu et al an overall complication rate of 0.15 percent was recorded and rare complications such as mediastinitis, lung abscess, pericarditis, and pneumothorax have been reported as case series only. Infectious complications were found when puncturing cystic lesions and antibiotics are recommended should a puncture be performed via EUS. EBUS TBNA was also found to be safe to perform in patients taking clopidogrel although this was a small case series only. As with regular bronchoscopy other complications such as infection, transient fever, haemoptysis, and others can also occur.

Damage to the bronchoscope by erroneous use of the needle or other handling mistakes can be a very expensive complication and regular training by the bronchoscope provider for the whole team involved with handling this device is recommended.

Learning Curve

EBUS TBNA takes time to learn under experienced guidance in order to fully understand the ultrasonographic pictures not seen in regular bronchoscopy. There exists no consensus in the national and international guidelines as to how best to train physicians in this technique. Learning curves have been studied and one Australian study reported that the diagnostic yield improved significantly by 20 procedures but 50 procedures were required to reach the peak of diagnostic yield. Another trial recommended performing 13 procedures under supervision to be able to perform the procedure safely and obtain an adequate yield. Different learning models have been proposed and Konge et al published data on the use of virtual reality bronchoscopy simulators to assess the performance in order to set a standard training prior to “real-life” bronchoscopy. Skills learned on simulators appear to be transferrable to a clinical setting. Larger trials are necessary here but more training opportunities need to be implemented since the technique is becoming more ubiquitous.

Literature Review

First published use and results for EBUS TBNA was in 2003 by Mark Krasnik and 2004 by Yasufuku. Since then the technique and its use has spread and multiple studies have been published and meta-analyses summarized the results. Gu et al published a meta-analysis with 1299 patients with lung cancer showing a pooled sensitivity of 0.93 and a pooled specificity of 1.00. A sensitivity of 0.94 was shown for patients who had an EBUS-TBNA based on CT or positron emission tomography (PET)-CT positive results. The sensitivity and accuracy of EBUS-TBNA for the prediction of lymph node involvement was superior to using CT or PET. Even when comparing EBUS TBNA staging to staging with mediastinoscopy its sensitivity and accuracy is very similar prompting national and international guidelines to change their diagnostic pathway to make use of EBUS TBNA prior to subjecting a patient to the more invasive procedure of a mediastinoscopy. EBUS TBNA has fewer complications than mediastinoscopy and can be performed as a day case procedure.

A complete mediastinal staging however cannot be accomplished by the use of EBUS alone since not all lymph nodes are accessible via the bronchial tree. EBUS needs to be combined with EUS in order to cover most mediastinal lymph node stations as well as the left adrenal gland. Stations 5 and 6 however are only accessed by anterior mediastinoscopy. Annema et al were able to show that combined endosonography with surgical staging had a greater sensitivity for mediastinal lymph node metastases than surgical staging alone.

The EBUS scope can be used endobronchially as well as via the esophagus and this “one scope approach” has been shown to be feasible and to provide a better diagnostic accuracy than either procedure alone. Depending on the size of the patient the left adrenal gland can be reached with the regular EBUS scope as well.

When comparing the 21 and 22 gauge needle to perform EBUS TBNA no difference was found in a retrospective study with regards to the diagnostic yield. A transbronchial needle forceps has also been evaluated and found a higher diagnostic yield compared to a 21G needle. These may be useful in diagnosing benign diseases or in the diagnosis of lymphoma but more data is needed here. 19G needles as well as “tru-cut” needles are being produced by different manufacturers to aid diagnosis in benign disease and lymphoma. No trials have been published as yet comparing these to the standard available 21 and 22G needles.

In the diagnosis of benign mediastinal lymphadenopathy due to suspected sarcoidosis a prospective randomized trial compared the use of EBUS TBNA with transbronchial biopsies. It found a diagnostic yield for the detection of granulomas of 80 percent with EBUS TBNA and only 53 percent with transbronchial biopsies and this technique should be employed before resorting to mediastinoscopy. The use of EBUS in suspected tuberculosis is less well established but one multi-center case series describing 156 patients with a diagnosis of TB associated lymphadenopathy reports a sensitivity and accuracy of 94 percent.

In the diagnosis of lymphoma larger specimens are often required to diagnose the exact subtype and grading. A study by Steinfort et al evaluated EBUS TBNA in lymphoma and found a diagnostic sensitivity of 76 percent but some of these patients required further biopsy reducing the sensitivity to 57 percent. EBUS TBNA should not be considered the first diagnostic approach in suspected lymphoma but can identify a proportion of patients with the disease.

Of course EBUS TBNA samples are smaller than samples obtained by mediastinoscopy, potentially resulting in more false negative samples. Hence current guidelines recommend mediastinoscopy following a negative EBUS TBNA result should there still be concern about the lymph node stations. Performing EBUS TBNA prior to mediastinoscopy reduces the need for more invasive procedures and unnecessary thoracotomies.

Summary

EBUS provides a real time assessment and sampling of enlarged mediastinal lymph nodes and is a safe tool in the diagnosis of benign and malignant diseases. It is becoming increasingly important in the diagnosis and staging of lung cancer as well as the diagnosis of benign diseases. Published trials confirm EBUS TBNA as the first line investigation of choice accepting that there may still be a need for more invasive testing if the results are negative. More randomized trials are required particularly in the field of benign disease and lymphoma.