Chapter 9

Minimally invasive endosonographic techniques: combined EBUS and EUS

Pravachan V.C. Hegde¹ and Moishe Liberman²

¹Advanced Interventional Thoracic Endoscopy/Interventional Pulmonology, Division of Pulmonary Critical Care Medicine, University of California San Francisco-Fresno, Fresno, CA, USA. ²Division of Thoracic Surgery, Dept of Surgery, CHUM Endoscopic Tracheobronchial Center (CETOC), Centre Hospitalier de l’Universite de Montreal, University of Montreal, Montreal, QC, Canada.

Correspondence: Pravachan V.C. Hegde, Advanced Interventional Thoracic Endoscopy/Interventional Pulmonology, Division of Pulmonary Critical Care Medicine, University of California San Francisco-Fresno, 2335 East Kashian Lane, Suite 260, Fresno, CA 93701, USA. E-mail: pv.pulm@gmail.com

Minimally invasive endosonographic techniques include both EBUS- and EUS-guided fine-needle aspiration (FNA). These are minimally invasive ultrasound-guided FNA techniques either through the endobronchial route (EBUS) or endo-oesophageal route (EUS). They are predominantly used in the diagnosis and staging of lung cancers. In addition, they are also used in diagnosing mediastinal adenopathy, infection and granulomatous lesions of the thoracic cavity accessible by these techniques. These techniques have replaced invasive surgical techniques in the evaluation of the mediastinum in many centres. In addition, combined EBUS/EUS/EUS-B is superior to standard mediastinoscopy in the initial staging of lung cancer because it allows for biopsy of lymph nodes and metastases not attainable with mediastinoscopy. In comparison with conventional mediastinoscopy, endosonographic techniques are less invasive and are carried out as an outpatient day case under conscious sedation, with less morbidity, considerable cost savings and are also well tolerated by patients. The ability of the combined EBUS/EUS/EUS-B techniques to sample multiple lymph node stations and distant metastases, including structures below the diaphragm, makes the use of such techniques the preferred test in the initial staging of lung cancer. Current European Society of Gastrointestinal Endoscopy (ESGE)/European Respiratory Society (ERS)/European Society of Thoracic Surgeons (ESTS) guidelines recommend a combination of EBUS/EUS over either test alone in the mediastinal staging of lung cancer [1]. Current American College of Chest Physicians (ACCP) lung cancer guidelines recommend EBUS-FNA, EUS-FNA or the combination of EBUS/EUS-FNA over surgical staging as the initial test for the staging of the mediastinum in nonsmall cell lung cancer (NSCLC) [2].

Accurate staging of the mediastinum is extremely important in determining the treatment plan in NSCLC. Testing with minimally invasive or noninvasive tests can prevent surgery in patients with advanced, benign or medically treatable diseases. Noninvasive radiological tests, including CT and PET-CT scans, are very valuable in procedure planning, but these techniques cannot provide tissue diagnosis. Therefore, it is vital to prove positivity or negativity of lymph node stations by obtaining a definitive tissue diagnosis in certain subgroups of patients.

In this chapter, we discuss the utility of combined EBUS/EUS/EUS-B in daily practice and highlight some technical aspects of the procedure. The traditional gold standard test in mediastinal staging has been cervical mediastinoscopy. We also discuss why mediastinoscopy may no longer be the gold standard for invasive mediastinal staging and why it is time to put an end to the era of a traditional gold standard procedure.

Combined endosonographic techniques are superior to conventional mediastinoscopy and should be the new gold standard in the initial staging in NSCLC. Our aim in this chapter is to discuss the current role of endosonographic techniques for the evaluation of mediastinal adenopathy and staging of lung cancer. A special emphasis will be on staging of NSCLC as it is the most common indication for using endosonographic techniques in daily practice. In addition, endosonographic FNA has become the first procedural test in cases where the clinical and imaging findings suggest an infectious and granulomatous lesion accessible by these techniques.

Endosonographic ultrasound (combined EBUS/EUS/EUS-B)

What are EBUS, EUS and EUS-B?

EBUS-FNA and EUS-FNA are minimally invasive ultrasound-guided FNA techniques. When a linear EBUS ultrasound scope is used to sample tissue through the transoesophageal route, the terminology EUS-B-FNA is used.

Indications and contraindications

The most common indication in daily practice is diagnosis and staging of lung cancer. In addition, they are also used in diagnosing mediastinal adenopathy, infection and granulomatous lesions of the thoracic cavity.

Contraindications include patient-related clinical conditions such as unstable angina, refractory hypoxia, haemodynamic instability, bleeding diathesis and coagulopathy. Other operator-dependent contraindications include inadequate expertise, facilities and equipment to handle complications.

Brief technical aspects and protocol of systematic FNA

A detailed description of the endosonographic anatomy and procedural technique is beyond the scope of this chapter. Here, we briefly describe some basic technical aspects.

The convex probe EBUS scope is a flexible scope with a field of vision between 50° and 70° and image parallel to the shaft of the scope. The depth of the image is between 3 and 5 cm. The linear EUS scope has a wider range of imaging compared with the EBUS scope. It provides a 180° view and can obtain images to a depth of 8 cm. EUS can also be performed using a convex probe EBUS scope (EUS-B-FNA). In order to produce good quality images, the transducer must oppose the tracheal or oesophageal wall with the lever pushed down and continuous suction applied throughout while performing needle aspiration. It is recommended to start with assessing the contralateral hilum/mediastinal nodes (N3), followed by N2 mediastinal nodes and then finally ipsilateral hilar nodes (N1) (in appropriate cases). The order of the examinations depends mainly on the side and location of the tumour [3]. The International Association for the Study of Lung Cancer (IASLC) 2009 map of the mediastinal and hilar lymph nodes is shown in Chapter 8 of this Monograph [4].

The classical approach described by JENSSEN et al. [5] is to start by learning the six basic landmarks for EBUS and EUS, and to practice finding them in sequence order. The six EBUS landmarks described in sequence order are: stations 4L, 7, 10L, 10R, azygous vein and 4R. Similar EUS landmarks in sequence order are: liver, aorta, left adrenal gland, stations 7, 4L and 4R.

Lymph node puncture is achieved with a quick forward movement advancing the needle. We prefer to hold the needle with the thumb downward. The thumb controls the needle advancing into the lymph node. In and out movements are performed with the needle inside the lymph node. It is very important to see the needle moving inside the lymph node and not to move the lymph node with the needle. Two samples are taken without suction to avoid blood on the slides. The third sample is collected with negative pressure suction for cell block, histology and molecular analysis. This is just an example of how to approach a node. There is no evidence for the use of suction, no suction or a slow-pull method.

Balloons filled with saline can be used to overcome poor contact between the ultrasound probe and the bronchial wall, and to assist in obtaining a clear ultrasound image. Although a saline-filled balloon can enhance image acquisition, it is unclear if this translates into a better diagnostic yield. Balloons are made of latex and thus cannot be used in patients with latex allergies. Latex-free balloons are available for linear EUS scopes. From a practical perspective, balloons are commonly used for the slightly challenging angle of the left paratracheal lymph node (4L) and hilar stations (10R and 10L). There are no studies comparing the use of a balloon versus no balloon and diagnostic yield with endosonography.

Needle size and number of aspirations

There is no difference in specimen adequacy or yield between 21G and 22G needles in EBUS-TBNA; however, 21G needles are associated with fewer needle passes only if rapid onsite evaluation (ROSE) is available [6]. Better characterisation of benign disease (83% versus 60%) and NSCLC histology subtyping (88% versus 65%) has been shown with 21G needles [7]. Studies evaluating the role of EUS in solid lesions adjacent to the gastrointestinal tract showed no difference in diagnostic yield between EUS-FNAs versus fine-needle biopsy needles [8]. The newer 19G EBUS needles have shown promising results with a greater degree of flexion, safety and diagnostic yield [9]. Prospective randomised controlled trials have shown no benefit in applying suction to needle aspiration. There is no difference in diagnostic yield or adequacy of the specimen [10]. There is insufficient data in studies comparing with and without suction in obtaining molecular markers in NSCLC. Optimal results can be obtained with three passes per lymph node station in EBUS-TBNA for mediastinal staging of potentially operable NSCLC. The increased yield plateaus after three passes [11].

Lymph node characteristics

The lymph node characteristics that are suggestive but not diagnostic of malignancy include round and heterogeneous nodes with a distinct margin and a coagulation necrosis sign [12]. A heterogeneous sign is more predictive of malignancy compared with other signs [13]. However, no lymph node characteristics can exclude or prove malignancy. These signs may be potentially useful if there are multiple lymph nodes in a single station and lymph node size is below the commonly used cut-off of 5 mm [14]. Colour Doppler sonography based on the bronchial artery in-flow sign is a fast, reproducible and effective tool that could help in targeting suspected malignant lymph nodes during EBUS-TBNA. The sensitivity, specificity, positive predictive value, negative predictive value and diagnostic accuracy of the bronchial artery flow sign using colour Doppler were found to be 93%, 64%, 84%, 60% and 81%, respectively, in a prospective study by NOSOTTI et al. [15].

In addition, endosonographic elastography is potentially capable of further differentiating between benign and malignant lymph nodes. Elastography is a noninvasive method in which the relative stiffness of tissues is imaged as a colour map or measured as shear wave velocity. Two techniques are commonly used: shear wave elastography and strain elastography. Most inflammatory processes do not change the elastographic architecture of lymph nodes, whereas metastases cause diffuse hard infiltration of the lymph node. Normal and inflammatory lymph nodes reveal a significantly harder cortex than the medulla and the hilum. Early metastatic infiltration shows a circumscribed localised stiffer neoplastic infiltration. At a later stage, there is diffuse stiff infiltration of the lymph node. A meta-analysis showed a pooled sensitivity of 88% and specificity of 85% with EUS-guided elastography for the discrimination of malignant versus benign lymph nodes [16]. In their prospective study of 56 patients, SUN et al. [17] reported EBUS elastography had a sensitivity of 91% and a negative predictive value of 82% for detecting metastases in lymph nodes.

Role of rapid onsite cytology, slides and cell blocks

Lymph node sampling should yield abundant lymphocytes in order to be considered as an adequate sample [18]. ROSE has not shown any benefit in diagnostic yield [19, 20]. However, if available, it results in fewer needle passes and number of slides prepared during the procedure [21]. ROSE may help to improve technique and reduce the number of procedures. A meta-analysis indicated that ROSE does not increase the diagnostic yield and there is no difference in specimen preparation (slide versus core versus cell block). The adequacy of specimen preparation depends mainly on the available expertise [22].

There is evidence from a single randomised trial that ROSE may be helpful in the diagnosis of central lung tumours when conventional FNA is used. In a prospective randomised trial of 125 patients by MONDONI et al. [23], conventional FNA guided by ROSE had a significantly higher sensitivity (96% versus 76%) compared with FNA alone. Training pulmonologists in basic onsite cytological evaluation may help reduce costs and also help make a diagnosis without involving a cytopathologist during the procedure. A prospective study of conventional TBNA in 84 patients with mediastinal adenopathy evaluated the role of the pulmonologist to assess the adequacy of cytological smears onsite. The accuracy of onsite assessment performed by a pulmonologist was not statistically different from that provided by a cytopathologist. There was 81% agreement between the two observers [24]. However, the final cytological diagnosis and report must always be the responsibility of the pathologist. It should be kept in mind that these two studies were performed in patients undergoing conventional TBNA and not endosonography-guided FNA.

In a single-centre retrospective study, cytology slides and core tissue preparations demonstrated high and similar diagnostic performance when comparing cytology slides versus cell blocks. Cytology slides combined with core tissue or cell blocks showed the highest performance; however, these combination methods were more resource-consuming. Diagnostic yield and accuracy were: cytology slides 81% and 80%, cell block 48% and 33%, core tissue 87% and 99%, cytology slides plus core tissue 80% and 100%, and cytology slides plus cell block 86% and 100%, respectively [25].

Diagnostic yield in lung cancer staging and special considerations

The most common indication for using endosonography in daily practice is staging of NSCLC. CT and PET-CT scans have improved the radiological staging of lung cancer; however, these techniques cannot provide tissue diagnosis and are associated with high false-positive and -negative rates, and low sensitivities and specificities [2, 26–35]. Stage dictates therapy and prognosis in lung cancer. The IASLC 2009 map mentioned earlier can be used to locate mediastinal and hilar lymph nodes. Staging helps to identify N2/N3 lymph nodes and distant metastases, which can prevent futile surgery and identify patients for neoadjuvant treatment. Staging is also important in order to identify N1 lymph node metastases in candidates with poor lung function before planning SBRT or sublobar resection. The ESGE/ERS/ESTS and ACCP guidelines recommend staging in all central tumours, peripheral tumours >3 cm, CT scan demonstrating lymph nodes >1 cm, N1 lymph node involvement on PET-CT and PET with standardised uptake values in the primary tumour of <2. Patients with peripheral tumour size <3 cm with no lymph node involvement on CT/PET-CT do not require invasive mediastinal staging [1, 2]. The eighth edition of the IASLC tumour (T), node (N) and metastasis (M) staging system is shown in table 1 [36].

Mediastinal staging can either be performed minimally invasively with EBUS, EUS or the combination of both, or invasively with standard cervical mediastinoscopy, transcervical extended cervical lymphadenectomy (TEMLA) and VATS. Testing with minimally invasive techniques can prevent surgery and pulmonary resection in patients with advanced stage lung cancer.

Why combined EBUS/EUS/EUS-B?

Standard mediastinoscopy allows access to stations 2R, 2L, 4R, 4L and 7, and the hilum (table 2). Access to the posterior and inferior mediastinum is limited. The lymph node stations accessible by EBUS include stations 1, 2, 4, 7, 10, 11 and 12. Stations 8 and 9 cannot be accessed with EBUS. The aortopulmonary window and para-aortic lymph nodes (stations 5 and 6) cannot be accessed with EBUS in most cases without transvascular puncture. In addition, smaller station 4L lymph nodes are occasionally difficult to access easily using EBUS. There is a 1:30 chance of inferior mediastinal nodal involvement without involvement of upper mediastinal nodes in patients with NSCLC [38].

Table 2. Lymph node stations accessible by different techniques

Figure 1. Station 6 access with EUS, a new technique. Reproduced and modified from [37] with permission.

EUS is helpful in accessing the lymph nodes in the inferior mediastinal stations 8 and 9, coeliac axis, left lobe of the liver, and bilateral adrenal glands (table 2). EUS can also access stations 2, 3p, 4L, 5, 6 and 7. Station 4R can be reached if the lymph node is large enough. Smaller station 4L lymph nodes are easily accessible by EUS/EUS-B compared with EBUS. In a study of 138 patients who underwent combined EBUS/EUS, EUS better identified malignant disease in lymph node stations 5, 6 and 7 [39]. Access to stations 2R and 4R is limited due to the intervening trachea. Interestingly, HWANGBO et al. [40] reported that the lymph nodes detected by EUS-B and missed by EBUS were not located in the inferior mediastinum, but at stations 4L and 5. In their study EUS was performed using a convex EBUS scope. The increased sensitivity of the combined technique may not be due to the ability to evaluate inferior mediastinum, but rather due to better visualisation of stations 4L and 7 and the ability to access the aortopulmonary window lymph nodes [40].

Adding EUS to EBUS allows for complete staging of the mediastinum and greater evaluation of lymph nodes and structures below the diaphragm. It allows for greater evaluation of lymph nodes compared with either of the techniques alone. A recently reported meta-analysis of 13 studies showed that adding EUS/EUS-B to EBUS increased sensitivity by 12% and addition of EBUS to EUS/EUS-B increased sensitivity by 22%. The mean sensitivity of a combined approach was 86% with a negative predictive value of 92% [41]. There was no statistically significant difference in the yield when comparing the use of the EBUS versus EUS scope in the transoesophageal route for sampling of lymph nodes [42]. However, the EUS scope has a better range of imaging, and also allows for evaluating adrenal (both right and left), liver and coeliac axis metastases. Using a single EUS-B scope in patients with adrenal involvement allows both mediastinal nodal and adrenal evaluation with a single procedure. This staging strategy reduces patient discomfort, risk and costs. There was no difference in the yield between EUS-B versus conventional EUS in sampling left adrenal metastases [43]. A conventional EUS scope may be more helpful if the right adrenal gland is involved. Identifying the left adrenal gland is easier compared with the right adrenal gland; the right adrenal gland cannot be reached with the EUS-B technique. The two techniques (EBUS and EUS/EUS-B) are complementary and can potentially replace surgical staging in patients with NSCLC.

A prospective trial of 166 patients comparing combined EBUS/EUS and mediastinoscopy with final pathology results of lymph node sampling at pulmonary resection showed that EBUS/EUS was diagnostic for N2/N3/M1 disease in 14% of patients in whom standard mediastinoscopy findings were negative, thereby preventing futile thoracotomy and invasive video-assisted thoracoscopic procedures [44]. The sensitivity, specificity, negative predictive value and diagnostic accuracy of combined EBUS/EUS were 91%, 100%, 96% and 97%, respectively (tables 3 and 4) [44].

Stay updated, free articles. Join our Telegram channel

Join