Endobronchial Ultrasonography/Endoscopic Ultrasonography

CRITICAL ELEMENTS

Identification of Lymph Nodes Suspicious for Cancer Metastasis
Node Station Assessment Utilizing Endobronchial Ultrasonography/Endoscopic Ultrasonography

1. IDENTIFICATION OF LYMPH NODES SUSPICIOUS FOR CANCER METASTASIS

Recommendation: The standard endobronchial ultrasonography (EBUS) classification system of the sonographic features of lymph nodes is useful to determine whether lymph nodes are malignant or benign.

Type of Data: Retrospective.

Strength of Recommendation: Weak.

Rationale

Round lymph nodes whose short-axis diameter is larger than 1 cm, that have distinct margins and heterogeneous echogenicity, and that have coagulation necrosis sign but not central hilar structures are suspicious for malignancy and must be biopsied (Fig. 6-1).¹

During EBUS-guided transbronchial needle aspiration (EBUS-TBNA) or endoscopic ultrasonography (EUS)-guided fine needle aspiration (EUS-FNA), all mediastinal and hilar lymph nodes should be assessed, characterized, and documented systematically. Lymph nodes should be identified according to the International Association for the Study of Lung Cancer lymph node map (Fig. I-3).² EBUS-TBNA and EUS-FNA should proceed from N3 nodes to N2 nodes and then to N1 nodes to prevent needle contamination and avoid accidental disease overstaging.

FIGURE 6-1 Sonographic characteristics, by EBUS or EUS, of mediastinal lymph nodes that favor benignancy or malignancy in lung cancer.

Morphology ultrasonography should be used to assess mediastinal lymph node stations 2R, 4R, 2L, 4L, and 7 during EBUS and stations 8 and 9 during EUS. When rapid on-site cytologic evaluation (ROSE) is available, a sample from one lymph node station should be confirmed as adequate for evaluation before the next station is sampled. When ROSE is not available, at least three passes of one lymph node station should be obtained unless a tissue block that is adequate for evaluation is grossly obtained during the first or second pass.³

2. NODE STATION ASSESSMENT UTILIZING ENDOBRONCHIAL ULTRASONOGRAPHY/ENDOSCOPIC ULTRASONOGRAPHY

Recommendation: Morphology ultrasonography should be used to assess mediastinal lymph node stations 2R, 4R, 2L, 4L, and 7 during EBUS and lymph node stations 8 and 9 during EUS.

Type of Data: Retrospective.

Strength of Recommendation: Weak.

Rationale

The recommendations for the extent of lymph node analysis are based on published guidelines. 4 Convex probe EBUS can access the upper paratracheal (stations 2R and 2L), lower paratracheal (stations 4R and 4L), subcarinal (station 7) and retrotracheal (station 3p) lymph nodes, as well as the hilar (station 10), interlobar (station 11), and lobar (station 12) nodes in the lower lobes. However, convex probe EBUS cannot access the prevascular (station 3a), subaortic (station 5), para-aortic (station 6), paraesophageal (station 8), pulmonary ligament (station 9), or lobar lymph nodes in the upper and middle lobes (Table 6-1).

EUS can access the pulmonary ligament (station 9), paraesophageal (station 8), subcarinal (station 7), retrotracheal (station 3p), and paratracheal (stations 2 and 4) lymph nodes. EUS usually cannot access the prevascular (station 3a), subaortic (station 5), para-aortic (station 6), or N1 lymph nodes. Some authors have reported that EUS can access stations 5 and 6, but the assessment of these nodes with EUS is
more difficult than that of other stations with EUS and requires significant expertise and experience to accomplish successfully.⁵

TABLE 6-1 Lymph Nodes Accessible by EBUS/EUS

EBUS	EUS
2	2
3p	3p
4	4
7	7
10	8
11	9
12^*
EBUS, endobronchial ultrasonography; EUS, endoscopic ultrasonography.
^*Lower lobe lobar lymph nodes.

Technical Aspects

Although the instrumentation used for EUS and EBUS have some similarities, they also have distinct characteristics that limit the way in which each can be utilized. EBUS-TBNA is performed using a convex probe. A saline-filled balloon at the tip of the probe helps improve image quality. Samples are obtained using either a 21- or 22-gauge needle with real-time EBUS visualization. The needle has a number of safety features, including a mechanism that prevents the needle from protruding more than 40 mm and an internal sheath that prevents the needle from contaminating the working channel of the scope. In contrast, EUS-FNA is performed using a side-viewing videogastroscope with a dedicated curved linear array transducer attached to the tip of the probe. A number of different scopes from different companies are available for EUS-FNA, and all have minor differences in their overall diameter and size of the ultrasonography probe. A dedicated 22-gauge needle is usually used for EUS-FNA, but smaller (25-gauge) and larger (19-gauge) needles are also available. Similar to that performed using EBUS, tissue sampling using EUS is performed with the aid of real-time ultrasonography.

Both EBUS-TBNA and EUS-FNA can be performed on an outpatient basis utilizing either conscious sedation or general anesthesia. (When general anesthesia is used, the cough reflex is minimal, which may be an advantage during the procedure.) An endotracheal tube (size 8 minimum) or laryngeal mask airway is used to accommodate the scope. The size of the scope limits nasal insertion.

EBUS-TBNA should include an examination of the airway using regular flexible bronchoscopy. Ultrasonically visible vascular landmarks should be used to identify the specific lymph node stations according to the International Association for the Study of Lung Cancer lymph node map.² Doppler ultrasonography is used to identify surrounding vessels as well as the blood flow within lymph nodes. The needle is introduced into the lymph node with a sharp stab. The node is aspirated by removing the internal stylet and using a Vac-Lok syringe to apply negative pressure. (Because negative pressure can cause bloody samples of hypervascular lymph nodes, EBUSTBNA of such nodes can be done without suction.) The needle is moved back and forth inside the lymph node to obtain samples. Finally, the needle is retracted into its sheath and removed from the bronchoscope.

The technique for EUS-FNA is similar to that for EBUS-TBNA. Anatomical landmarks such as the inferior vena cava, right and left atrium, azygos vein, main pulmonary artery, and aorta are identified. Any lymph nodes encountered are described and numbered according to the International Association for the Study of Lung Cancer lymph node map (see Fig. I-3 in the Introduction to Section II). Lymph nodes are then biopsied, usually with a 22-gauge needle, using real-time ultrasonography guidance. Finally, one cannot stress enough the importance of proper specimen handling during EBUS-TBNA and/or EUS-FNA. Institutional standards differ, and close collaboration between the surgeon and cytopathologist is essential. However, when ROSE is available, both EBUSTBNA and EUS-FNA can be performed without the aid of a cytopathologist.

REFERENCES

1. Fujiwara T, Yasufuku K, Nakajima T, et al. The utility of sonographic features during endobronchial ultrasound-guided transbronchial needle aspiration for lymph node staging in patients with lung cancer: a standard endobronchial ultrasound image classification system. Chest 2010;138:641-647.

2. Rusch VW, Asamura H, Watanabe H, et al. The IASLC Lung Cancer Staging Project. A proposal for a new international lymph node map in the forthcoming seventh edition of the TNM classification for lung cancer. J Thorac Oncol 2009;4:568-577.

3. Lee HS, Lee GK, Lee HS, et al. Real-time endobronchial ultrasound-guided transbronchi al needle aspiration in mediastinal staging of non-small cell lung cancer: how many aspirations per target lymph node station? Chest 2008;134(2):368-374.

4. Howington JA, Blum MG, Chang AC, et al. Treatment of stage I and II non-small cell lung cancer: diagnosis and management of lung cancer, 3rd ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2013;143(5)(suppl):e278S-e313S.

5. Liberman M, Durnceau A, Grunenwald E, et al. Initial experience with a new technique of endoscopic and ultrasonographic access for biopsy of para-aortic (station 6) mediastinal lymph nodes without traversing the aorta. J Thorac Cardiovasc Surg 2012;144(4):787-792; discussion 792-793.

Endobronchial Ultrasonography/Endoscopic Ultrasonography: Key Question

When should negative endobronchial ultrasonography findings be confirmed by a more invasive procedure?

INTRODUCTION

Accurate staging is of paramount importance in the management of non-small cell lung cancer (NSCLC). The treatment of NSCLC, whether surgery, chemotherapy, radiotherapy, or a multimodality approach, depends on the stage of the disease. In NSCLC patients without metastatic disease, the status of the mediastinum is the most important determinant of candidacy for curative-intent treatment.

The first-line tests in the diagnosis and staging of NSCLC are imaging studies, including computed tomography (CT) and positron emission tomography (PET). Based on the most recent American College of Chest Physicians (ACCP) guidelines, the sensitivity and specificity of CT in identifying mediastinal lymph node metastasis are 55% and 81%, respectively, and those of PET are 77% and 86%, respectively.¹ These imaging studies have substantial false positive and false negative rates, and PET findings in particular must be confirmed with lymph node biopsy.

Traditionally, the gold standard for sampling lymph nodes in the mediastinum to confirm imaging studies’ findings has been mediastinoscopy. Recently, however, endobronchial ultrasonography (EBUS)- or endoscopic ultrasonography (EUS)-guided needle aspiration techniques have also emerged as options for staging the mediastinum. In many institutions, EBUS-guided transbronchial needle aspiration (EBUSTBNA) has replaced mediastinoscopy as the first-line means of sampling mediastinal lymph nodes, and mediastinoscopy is reserved to confirm the absence of N2 disease.¹,²

As surgeons gain experience using EBUS-TBNA for lung cancer staging, the role of confirmatory mediastinoscopy and other invasive approaches to confirm EBUS-TNBA findings continues to diminish. Despite the impressive results obtained with EBUSTBNA overall, confirmatory mediastinoscopy is still useful in certain situations. These situations are discussed in the following texts.

LITERATURE REVIEW

We conducted a Medline search of English language studies published from 2003 to 2013. We used the Medical Subject Heading term “endobronchial ultrasound”; the terms “negative predictive value,” “false negative,” “accuracy,” and “sensitivity” were introduced to limit the number of studies identified. A total of 80 articles were found. Of the 80 abstracts we reviewed, 24 were excluded because the studies included patients who did not have lung cancer or examined EBUS in a role other than its use in invasive mediastinal staging. A careful review of the remaining 56 articles
identified 35 articles that included data about the false negative rate, sensitivity, or negative predictive value of EBUS in the assessment of N2 disease in patients with NSCLC. This is summarized in Figure 6-2.

FIGURE 6-2 CONSORT diagram summarizing the literature search used for review of endobronchial ultrasound performance.

FINDINGS

Factors that affect the negative predictive value of EBUS

Multiple studies have demonstrated the utility of EBUS-TBNA as a staging modality in patients with NSCLC (Table 6-2).³,⁴,⁵,⁶,⁷,⁸,⁹,¹⁰,¹¹,¹²,¹³,¹⁴,¹⁵,¹⁶,¹⁷,¹⁸,¹⁹,²⁰,²¹,²²,²³,²⁴,²⁵,²⁶,²⁷,²⁸,²⁹,³⁰ The most important parameter defining the role of a staging modality in the preoperative staging of NSCLC is its ability to rule out mediastinal lymph node disease. In other words, a test for mediastinal disease must have a high sensitivity and high negative predictive value to prevent surgeons from performing a futile thoracotomy or thoracoscopic resection. However, both sensitivity and negative predictive value can be influenced by a number of factors.

TABLE 6-2 Studies Reporting Sensitivity and/or Negative Predictive Value of Real-Time Endobronchial Ultrasonography in the Mediastinal Staging of Confirmed or Highly Suspected Lung Cancer

Author, Year	Study Design	No. of Patients	Prevalence of N2 Disease (%)	NPV	Sensitivity	Accuracy	Key Findings	Potential for Significant Bias
Cornwell et al, 2013³	Retrospective	62	5	93	67	94	In patients with clinical stage I NSCLC, EBUS results in a lower incidence of nontherapeutic thoracotomy than noninvasive staging does, but this difference is not significant.	Yes
Herth et al, 2008⁴	Prospective observational	97	10	98.9	89	NA	EBUS is accurate in NSCLC staging for patients with clinical stage I NSCLC determined by CT and PET findings.	Yes
Herth et al, 2006⁵	Prospective observational	100	121	96.3	92.3	NA	EBUS is beneficial in patients with clinical stage I NSCLC. It prevents nontherapeutic thoracotomy in 1 of 6 patients. PET was not used routinely.	Yes
Hwangbo et al, 2009⁶	Prospective observational	126	26	96.7	90	97.4	EBUS is useful for confirming N2 disease detected by PET. It is also useful for detecting N2 disease in patients with radiographic N0 disease.	Yes
Lee et al, 2008⁷	Retrospective	102	30	96.9	93.8	97.9	Optimal results with EBUS are obtained when at least 3 aspirations of each lymph node are performed.	Yes
Yasufuku et al, 2011⁸^*	Prospective controlled trial	153	35	91	81	93	EBUS is equivalent to mediastinoscopy in the mediastinal staging of NSCLC.	No
Feller-Kopman et al, 2009⁹^*	Retrospective	131	35	89.7	85	NA	EBUS is an accurate and sensitive method for diagnosing and staging NSCLC.	Yes
Petersen et al, 2009¹⁰	Retrospective	157	43	90	85	NA	EBUS is accurate in staging the mediastinum in NSCLC patients. The routine confirmation of negative EBUS findings with mediastinoscopy has a minor role in NSCLC staging.	Yes
Sanz-Santos et al, 2012¹¹	Retrospective	296	51	93.6	NA	NA	EBUS can be used to sample lymph node regions 4R, 4L, and 7 in more than 80% of patients. In such patients, EBUS has an NPV of >90% for mediastinal malignancy.	Yes
Nakajima et al, 2013¹²^*	Retrospective	438	52	90	97	98	ROSE during EBUS results in a low incidence of nondiagnostic samples.	Yes
Jhun et al, 2012¹³	Retrospective	151	55	84.3	91.6	93.8	The diagnostic yield of EBUS is lower for left paratracheal lymph nodes. The diagnostic yield is not related to lymph node size.	Yes
Szlubowski et al, 2009¹⁴	Retrospective	226	57	89	83.5	92.9	EBUS is an effective and safe technique for mediastinal staging in NSCLC patients. In patients with negative EBUS results, surgical exploration of the mediastinum should be performed.	Yes
Bauwens et al, 2008¹⁵	Retrospective	106	58	91	95	97	EBUS is a reasonable first step in the confirmation of N2 disease in NSCLC patients. Surgical mediastinal staging should be used to confirm negative EBUS findings.	Yes
Joesph et al, 2013¹⁶^*	Retrospective	131	58	90	92	NA	ROSE does not affect clinical decisions made during staging EBUS.	Yes
Lee et al, 2012¹⁷	Retrospective	73	62	94	95	97	EBUS can be used to accurately assess the mediastinum in patients with NSCLC and radiographic N2 disease.	Yes
Cerfolio et al, 2010¹⁸	Retrospective	72	63	79	57	83	EBUS and EUS have high false negative rates, and negative results should be confirmed prior to thoracotomy.	Yes
Navani et al, 2012¹⁹	Retrospective	774	65	88	72	NA	EBUS samples are suitable for use in NSCLC subtyping and EGFR mutation analysis.	Yes
Kuo et al, 2011²⁰	Retrospective	43	65	85.7	80.6	91	The diagnostic accuracy of EBUS is higher than that of PET in a tuberculosis-endemic population.	Yes
Hu et al, 2013²¹	Retrospective	231	67^‡	92	88	87	Proficiency using EBUS requires 22 cases. Lymph node size is a predictor of success.	Yes
Yasufuku et al, 2005²²^*	Prospective observational	105	67	89.5	94.6	96.3	EBUS is an accurate staging procedure in patients with NSCLC.	Yes
Rintoul et al, 2009²³	Retrospective	109	71	60	91	92	EBUS can be used to accurately evaluate PET-positive hilar and mediastinal lymph nodes. Negative findings should be confirmed by surgical means.	Yes
Cetinkaya et al, 2011²⁴	Retrospective	52	80	83	95	96	EBUS is safe and accurate in NSCLC staging.	Yes
Ernst et al, 2008²⁵	Prospective cross-over	60	89	78	87	NA	The difference between EBUS and mediastinoscopy in determining the N status of patients with NSCLC is not statistically significant.	Yes
Gu et al, 2009²⁶	Meta-analysis	1,299	NA	93	NA	NA	EBUS has a high NPV and is costeffective in the mediastinal staging of NSCLC.
Adams et al, 2009²⁷	Meta-analysis	782	NA	NA	88	NA	EBUS has high sensitivity in the mediastinal staging of NSCLC.
Abu-Hijleh et al, 2013²⁸^*	Retrospective	200	NA	75	87	91	EBUS is similar to surgical staging in patients with NSCLC. The NPV of EBUS is highest after the initial 25-50 cases. The accuracy of EBUS is independent of lymph node size or location and number of passes.
Dong et al, 2013²⁹^*	Meta-analysis	1,066	NA	93	90	96	EBUS is accurate and safe in staging NSCLC.
Whitson et al, 2013³⁰^*	Retrospective	120	NA	66^† 85	83^† 93	87^† 95	The inclusion of nondiagnostic results yields a lower NPV, sensitivity, and accuracy.
^*ROSE was used. ^†For the Whitson study, the first set of numbers includes nondiagnostic specimens. The values for when nondiagnostic studies are included are shown below in the same field. ^‡Incidence of N1 and N2 disease.
CT, computed tomography; EBUS, endobronchial ultrasonography; EGFR, epidermal growth factor receptor; NA, not available; NPV, negative predictive value; NSCLC, non-small cell lung cancer; PET, positron emission tomography; ROSE, rapid on-site evaluation.