Version May 2024
INTRODUCTION — Endoscopic ultrasonography (EUS) is a combination of endoscopy and ultrasonography. EUS can be used to visualize and sample mass lesions of the pancreas, gastrointestinal tract, posterior mediastinum, and retroperitoneum.
The procedure and its indications, contraindications, and complications are discussed in this topic review. The evaluation of mediastinal masses and the modalities used to diagnose and stage lung cancer are presented separately. (See "Approach to the adult patient with a mediastinal mass" and "Overview of the initial evaluation, diagnosis, and staging of patients with suspected lung cancer" and "Procedures for tissue biopsy in patients with suspected non-small cell lung cancer" and "Selection of modality for diagnosis and staging of patients with suspected non-small cell lung cancer".)
TECHNIQUE AND PROCEDURE — EUS should be distinguished from endobronchial ultrasound (EBUS). Both visualize and guide sampling of mediastinal structures and lymph nodes, but EUS is performed during endoscopy of the gastrointestinal tract and EBUS is performed during bronchoscopy. They are complimentary; EBUS is used to examine the anterior mediastinum while EUS is used to examine the posterior mediastinum. Both procedures should only be performed by a skilled endoscopist (pulmonologist, gastroenterologist, surgeon), who has undergone appropriate training. (See "Endobronchial ultrasound: Technical aspects" and "Endobronchial ultrasound: Indications, contraindications, and complications".)
Preprocedure preparation — The preparation for patients is similar to that described for patients undergoing upper gastrointestinal endoscopy and EUS-guided sampling for non-mediastinal lesions. (See "Overview of upper gastrointestinal endoscopy (esophagogastroduodenoscopy)", section on 'Patient preparation'.)
The majority of patients do not need to discontinue aspirin or nonsteroidal anti-inflammatories when undergoing EUS-guided sampling. The management of antiplatelet and anticoagulant therapy in those undergoing endoscopy is typically individualized, managed in conjunction with the prescribing subspecialist, and is discussed separately. (See "Management of antiplatelet agents in patients undergoing endoscopic procedures" and "Management of anticoagulants in patients undergoing endoscopic procedures" and "Gastrointestinal endoscopy in patients with disorders of hemostasis".)
With the exception of those who are undergoing EUS-guided sampling for mediastinal cysts, most patients do not need prophylactic antibiotics. Based upon the higher than usual risk of infection in those with mediastinal cystic lesions, prophylactic antibiotics are typically administered in this population both before and for three to five days after the procedure is performed. Indications and suggested regimens for antibiotic prophylaxis are listed in the tables (table 1 and table 2). (See 'Procedure-related' below and "Overview of upper gastrointestinal endoscopy (esophagogastroduodenoscopy)", section on 'Patient preparation' and "Antibiotic prophylaxis for gastrointestinal endoscopic procedures".)
Equipment — EUS is performed with an echoendoscope (an endoscope with an ultrasound transducer engineered into its tip). The echoendoscope is flexible, approximately 11 to 13 mm in diameter depending on the type of scope (radial or linear), and can obtain reliable images at a depth ranging from 3 mm to 8 cm from the transducer. Two types of echoendoscopes exist, radial and curvilinear:
●The radial echoendoscope provides a 360-degree ultrasound image of the gastrointestinal tract and surrounding structures (image 1). It is purely a diagnostic tool. The radial EUS scope readily images lesions such as submucosal lesions of the esophagus, stomach, or duodenum as the layers of the esophagus, stomach, or duodenum are clearly seen. This scope allows the endoscopist to identify target lesions quickly, and also provides a familiar cross sectional view such as seen with computed tomography (CT).
●The curvilinear echoendoscope provides a 90 to 180-degree view, depending on the model of curvilinear echoendoscope, that is parallel to the shaft of the echoendoscope, thereby allowing real-time visualization of the fine needle during the tissue sampling. Color flow and Doppler features of the curvilinear echoendoscope also permit identification of vascular, ductular (such as biliary and pancreatic ducts), and cystic structures, which is important prior to fine needle aspiration (biopsy) since samples cannot be obtained if there are intervening blood vessels in the path of the fine needle aspiration (biopsy) needle. Landmarks in the mediastinum may be more difficult to see fully depending on the experience of the endoscopist. (See 'Sampling' below.)
Procedure — The procedure involves sonographic imaging through the esophagus to look for the typical anatomic landmarks and identify target abnormalities for sampling.
Ultrasound imaging — The procedure begins once the patient is adequately sedated. When clinically indicated, EUS can be performed on patients who are on a mechanical ventilator or intubated specifically prior to the procedure. (See "Gastrointestinal endoscopy in adults: Procedural sedation administered by endoscopists".)
The radial echoendoscope is passed through the mouth, esophagus, and stomach until the tip reaches the duodenum. It is then withdrawn slowly in 1 to 2 cm increments while ultrasound imaging is performed. Anatomic landmarks and sonographic abnormalities are typically recorded.
●The upper retroperitoneum is visualized through the gastric wall, including the retroperitoneal lymph nodes, left adrenal gland, right adrenal gland [1], left lobe of the liver, and other retroperitoneal structures. (See "Endoscopic ultrasound-guided fine needle aspiration in the gastrointestinal tract".)
●The posterior mediastinum is visualized through the esophageal wall. The heart, pleura, spine, and vascular structures can be readily identified. In addition, mediastinal lymph nodes may be visualized in the subcarina (level 7), the paraesophageal area (level 8), adjacent to the inferior pulmonary ligament (level 9), and, possibly, the aortopulmonary window (level 5) (figure 1 and figure 2). Abnormalities may also be detected in the paratracheal areas (levels 2 and 4), adjacent to the ascending aorta (level 6), and hilar lymph nodes (level 10R and 10L), if they are large enough. In general, only abnormal sites are sampled.
●The superior mediastinal lymph nodes and masses in the paratracheal region (level 2R and 2L) can be imaged with EUS, if large enough.
If fine needle aspiration is needed, the radial echoendoscope is withdrawn and a curvilinear echoendoscope is advanced to the desired location for the purposes of fine needle aspiration or core needle biopsy (FNA/CNB). Color flow and Doppler imaging are used prior to sampling to ensure that there are no intervening blood vessels, which is a contraindication for EUS-guided sampling. (See 'Sampling' below.)
The duration of the procedure is typically 30 to 60 minutes. It is usually performed as an outpatient procedure.
Although EUS can readily image and identify mediastinal abnormalities, sonographic features alone are not reliable to distinguish between malignant or benign conditions. Thus, fine needle aspiration is performed if there is any uncertainty [2,3]. As examples:
●In one retrospective review of 162 patients with enlarged mediastinal lymph nodes including lesions that were benign (68), due to sarcoidosis (33), lymphoma (20), or metastases (41), the median long axis was shorter in the benign group compared with metastases and sarcoid groups (21 versus 28 mm) [3]; however, ultrasonography features of the lymph nodes were not reliable in differentiating between the groups. Thus, FNA is recommended when feasible.
●Another study of 113 patients compared EUS characteristics of tubercular and reactive lymphadenopathy. EUS findings that were more commonly seen in tuberculosis included hypoechoic echotexture, patchy anechoic/hypoechoic areas, calcification, sharply demarcated borders, exudative material on aspirate, conglomeration of lymph nodes, and larger size [2]. However, these findings were also seen in patients with reactive lymphadenopathy, albeit less frequently, and were therefore not specific. In addition, some of these features also overlap with features of malignancy (eg, well-defined borders, hypoechoic appearance). Again, FNA is recommended when feasible.
Sampling — Tissue samples can be obtained via FNA or CNB. (See 'Fine needle aspiration' below and 'EUS-guided core biopsy' below.)
Choosing among these options varies at the discretion of the operator and is influenced by endoscopist experience and institutional practice. The advantage of core biopsy sampling is that typically larger pieces of tissue may be obtained in one pass. This may increase diagnostic yield and reduce sampling time (eg, useful if cytogenetics is planned) depending on the targeted lesion. One study suggested similar sensitivity between a 22-gauge core biopsy needle (EUS-CNB) and standard EUS-FNA, but the core needle required fewer passes [4]. However, most needles have not been adequately compared, and we believe that the choice of needle should be individualized. In a retrospective study of 250 EUS-CNB lesions, some of which were mediastinal, a diagnosis was achieved in over 80 percent with a median number of three passes. This yield was 69 percent if rapid on-site cytologic assistance was not used [5].
EUS-FNA samples, have been successfully used to detect lung cancer-associated mutations such as epidermal growth factor receptor and k-ras. These are important for predicting therapeutic response to chemotherapeutic drugs (eg, the EGFR tyrosine kinase inhibitors) and tailoring management [6].
Fine needle aspiration — The technical procedure of EUS-guided fine needle aspiration (EUS-FNA) is similar to that used for EUS-guided aspiration of gastrointestinal lesions, which is described in detail separately. (See "Endoscopic ultrasound-guided fine needle aspiration in the gastrointestinal tract", section on 'Technique'.)
In brief, once the target lesion has been identified and no intervening blood vessels are seen, a 22- or 25-gauge fine aspiration needle is passed through the working channel of the curvilinear echoendoscope. Under direct ultrasound guidance, the needle is directed through the wall of the esophagus and into the target. Aspiration is then performed by simultaneously moving the needle in a to-and-fro motion inside the target lesion while applying negative pressure using a 10 or 20 mL syringe. The necessity of using the stylet after the initial pass has been debated (the needle comes preloaded with the stylet), and some have found that using the needle without the stylet does not impact the quality of the cytologic sample in terms of contamination, blood contamination, and cellularity [7]. Following removal of the aspiration needle, air is forced through the needle using the same syringe, which expels the aspirated sample onto a glass slide. Two direct smears are usually made from a single pass; one slide is dried with 95 percent alcohol for cytologic analysis in the laboratory while the other slide is stained for rapid on-site evaluation (ROSE) (image 2). Multiple passes can be performed to enhance the diagnostic yield as well as for special staining, cell block preparation for additional immunohistochemical staining, or for microbiology purposes. The same procedure can be used to aspirate tissue for microbiologic analysis, except the aspirate is placed in a small amount of sterile saline. In general, the same needle is used for each pass unless the needle comes in contact with formalin. (See "Bronchoscopy: Transbronchial needle aspiration", section on 'Technical considerations' and "Selection of modality for diagnosis and staging of patients with suspected non-small cell lung cancer", section on 'EBUS/EUS needle aspiration' and "Procedures for tissue biopsy in patients with suspected non-small cell lung cancer", section on 'Limitations'.)
EUS-guided core biopsy — EUS-CNB is done in a similar fashion. The needles range in size from 25- to 19-gauge. Compared with EUS-FNA, the goal of EUS-CNB is to obtain a core of tissue with preserved architecture. The technique of obtaining a core biopsy sample is similar to that of EUS-FNA. However, a vacuum suction is applied for 60 seconds prior to one or two to-and-fro motions. The core can be smeared on a glass slide for ROSE. Multiple passes can also be done to enhance the diagnostic yield. Core samples can also be submitted in formalin for formal pathologic interpretation or in sterile saline for culture. In our experience, the risks of bleeding are similar to EUS-FNA.
Postprocedure care — After the procedure, patients are recovered from sedation or anesthesia. A chest radiograph or CT of the chest is not routinely performed unless pneumomediastinum from esophageal perforation is suspected or a patient develops symptoms of a complication (eg, chest pain, dyspnea, fever). Air is not routinely seen after EUS-FNA or EUS-CNB. A fever after the procedure is not typical and warrants further investigation. The equipment is cleaned per procedural protocol. Recovery room time is typically 60 minutes for both EUS-FNA and EUS-guided core biopsy. (See 'Complications' below and "Flexible bronchoscopy in adults: Overview", section on 'Infection control' and "Procedural sedation in adults in the emergency department: Medication selection, dosing, and discharge criteria", section on 'Discharge criteria'.)
INDICATIONS — The most frequent indications for EUS of the mediastinum are:
●To further characterize mediastinal abnormalities or adenopathy
●Staging or diagnosis of lung cancer (non-small cell and small cell)
●Local staging of esophageal cancer (squamous and adenocarcinoma)
However, EUS-guided sampling can also be used to diagnose a wide range of other benign and malignant lesions of the mediastinum, including sarcoidosis, tuberculosis, metastatic disease involving the mediastinum, or lesions within the esophageal wall (stromal tumors, metastatic disease). EUS offers a distinct advantage as it is able to access lymph nodes, masses, and lesions in the posterior mediastinum, in areas that are not accessible to CT. EUS-guided sampling (EUS-FNA and EUS-CNB) is best suited to target lesions that are within the esophageal wall, adjacent to the esophageal wall, in the posterior mediastinum, or in the inferior mediastinum (figure 1 and figure 2) [8].
EUS is not suitable for lesions in the anterior or superior mediastinum. Since it is minimally invasive, EUS is performed on outpatient basis with moderate or deep sedation. EUS-guided sampling may be preferred over surgical approaches in patients who are not surgical candidates. In addition, EUS-FNA may be an alternative modality for obtaining pleural fluid or tissue in extrathoracic sites including the celiac lymph nodes, left adrenal gland, upper retroperitoneum, and left lobe of the liver [9-16]. Therapeutic applications are uncommon but include cyst or abscess drainage.
Limitations of EUS-FNA include availability of equipment and expertise, and the rare occurrence of false-negative or false-positive results when an aspirated lymph node is adjacent to a mediastinal mass [17].
Diagnostic — EUS-guided sampling can be used to diagnose benign and malignant disease involving the mediastinum.
Malignant diseases — Lung cancer (non-small cell and small cell) is the most common type of malignant mediastinal disease detected by EUS-FNA, but other types of malignancies can be diagnosed using this modality.
Lung cancer — EUS-FNA can access and therefore, diagnose tumors that invade the posterior and inferior mediastinum that are close to the esophagus (typically identified on CT or positron emission tomography). Rarely, if lesions are large and close enough to the esophagus, masses in the superior mediastinum may be sampled using EUS. Data to support the diagnostic sensitivity of EUS-guided sampling for patients with lung cancer are provided separately. (See "Selection of modality for diagnosis and staging of patients with suspected non-small cell lung cancer", section on 'Modality choice - mediastinal sampling' and "Procedures for tissue biopsy in patients with suspected non-small cell lung cancer", section on 'Transesophageal endoscopic ultrasound'.)
Other malignancies — EUS-guided sampling is also a useful technique for detecting malignant mediastinal diseases other than lung or esophageal cancer [18-24]. In a retrospective study of 49 patients who were not suspected of having lung cancer EUS-FNA of mediastinal lymphadenopathy or a mediastinal mass [18]. EUS-FNA was used to identify malignant mediastinal disease in 45 percent and the majority of the remaining patients had benign disease. Among the 22 patients with malignancy, diagnoses included breast cancer (six), colon cancer (two), renal cell cancer (two), testicular cancer (two), esophageal cancer (one), laryngeal cancer (one), metastatic disease from an unknown site (four), and small cell lung cancer (one) (image 3). EUS-FNA made the correct diagnosis in 94 percent of cases.
Other rare diagnoses that can be made with EUS-FNA include lymphoma, nerve sheath tumors [19-21], and thymic carcinoma [22].
EUS-FNA has also been useful in detecting recurrent malignancy in the posterior mediastinum among patients who underwent surgery for their primary malignancy and were in clinical and/or radiographic remission before the initial suspicion of tumor recurrence [25].
Although one case series reported that peri mediastinal parenchymal pulmonary nodules could be successfully located and biopsied using EUS, this is not standard practice [26,27]. Other, more commonly used image-guided approaches used to biopsy peripheral pulmonary nodules are discussed separately. (See "Image-guided bronchoscopy for biopsy of peripheral pulmonary lesions".)
Malignant pleural effusions can also be readily and safely sampled with EUS [14]. EUS-FNA of pleural fluid was used to identify metastases from endometrial cancer in a patient who presented with right pleural effusion and diffuse pleural thickening [15]. However, the typical approach to sampling a pleural effusion suspected to be malignant is thoracentesis, the details of which are discussed separately. (See "Selection of modality for diagnosis and staging of patients with suspected non-small cell lung cancer", section on 'Suspected advanced disease' and "Procedures for tissue biopsy in patients with suspected non-small cell lung cancer", section on 'Intrathoracic' and "Overview of the initial evaluation, diagnosis, and staging of patients with suspected lung cancer", section on 'Clinical-directed imaging'.)
Benign diseases — Reactive lymphadenopathy, infectious, and granulomatous diseases (eg, sarcoidosis, tuberculosis, histoplasmosis) are the most common benign conditions that can manifest with mediastinal lymphadenopathy. EUS-FNA is a safe alternative to other diagnostic methods (eg, mediastinoscopy) and has a high diagnostic yield in patients with benign mediastinal disease if the pretest probability is high [28-33]. As examples:
●In a retrospective study, 49 patients with a mediastinal mass and no history of pulmonary malignancy underwent EUS-FNA. Benign conditions were detected in 24 patients (49 percent) and included [18] histoplasmosis (eight), sarcoidosis (one), leiomyoma (two), duplication cyst (two), teratoma (one), and benign lymph nodes (10).
●In another retrospective study of 34 patients with undiagnosed mediastinal lymphadenopathy or mediastinal mass, EUS-FNA was used to diagnose tuberculosis (five), duplication cysts (two), cryptococcosis, and reactive lymphadenitis (one each) [24].
Sarcoidosis — EUS-FNA has been used to make a diagnosis of sarcoid (picture 1) and may be useful in patients with posterior mediastinal adenopathy and a nondiagnostic bronchoscopy.
●In a randomized multicenter study of 304 patients with suspected stage I or II sarcoidosis with mediastinal adenopathy, transbronchial needle aspiration (TBNA) was compared with EBUS-FNA and EUS-FNA. Endosonography resulted in a higher diagnostic yield (80 versus 53 percent) [34].
●In a retrospective study of 127 patients who had mediastinal lymphadenopathy of unclear etiology, EUS-FNA was used to identify sarcoidosis with a sensitivity and specificity of 89 and 96 percent, respectively [33].
●In a prospective study, 50 patients with suspected sarcoidosis (71 percent of whom had a prior nondiagnostic bronchoscopy) underwent EUS-FNA [29]. Sarcoidosis was confirmed in 41 patients (82 percent). The location of aspirated lymph nodes included the following: paratracheal, aortopulmonary window, and paraesophageal regions.
●In a retrospective study of 101 patients (55 percent of whom had a prior nondiagnostic bronchoscopy) with suspected pulmonary sarcoidosis (stage I and II), the sensitivity of EUS-FNA was 87 percent [35]. Sensitivity was highest in patients with stage I disease (92 percent) compared with stage II disease (77 percent). The diagnostic yield was improved if a cell block (additional passes) was submitted.
Some studies have reported an increased risk of infection in association with EUS-FNA in patients with sarcoidosis [36,37]. However, whether case reports such as these should prompt prophylactic antibiotics prior to the procedure in patients suspected as having sarcoidosis is unclear. In general, it is not our practice to administer prophylactic antibiotics unless the patient is immunosuppressed (eg, on high dose steroids or other immune suppressant agent). (See 'Preprocedure preparation' above.)
The value of EBUS-TBNA in the diagnosis of lymph nodal sarcoidosis and the general approach to the diagnosis of sarcoidosis are discussed separately. (See "Bronchoscopy: Transbronchial needle aspiration", section on 'Sarcoidosis' and "Clinical manifestations and diagnosis of sarcoidosis", section on 'Endoscopic ultrasound-guided needle aspiration or cryobiopsy'.)
Tuberculosis — Case reports describe the successful diagnosis of tuberculosis (TB) using EUS-FNA [2,18,24,38]. In a retrospective study of 60 patients with mediastinal lymphadenopathy and clinical suspicion of TB, EUS-FNA had a sensitivity and specificity of 71 and 100 percent [38]. The endoscopic appearance of tubercular lymph nodes is described separately. (See 'Ultrasound imaging' above.)
Other — Among case series of patients with lymphadenopathy of unknown etiology, other benign pathologic abnormalities have been described including cryptococcosis, histoplasmosis, esophageal bronchogenic cyst, bacterial abscess, leiomyoma, duplication cyst, teratoma, and benign reactive lymphadenopathy [18,24,39].
Staging — EUS-FNA has been used for local staging of lung cancer. It may also be used to identify metastases from extrathoracic tumors.
Staging lung cancer — Non-small cell lung cancer (NSCLC), the most common form of lung cancer, is usually staged according to the tumor-node-metastasis (TNM) system (table 3 and table 4) [40,41]. EUS-guided sampling can help characterize the primary tumor (T), assess the mediastinal lymph nodes (N) for evidence of metastatic disease, and evaluate some sites of distant metastasis (M) such as the left lobe of the liver and adrenal glands [16,42]. Among these contributions, however, mediastinal lymph node evaluation is its primary role. One report suggests a possible role in restaging after neoadjuvant therapy [43]. EUS-guided sampling, particularly with EUS-FNA and endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) are often combined, as these two methods are complementary in evaluating the mediastinum. In theory, the entire mediastinum is accessible to the combination of EUS-FNA plus EBUS-guided TBNA. Details regarding the sensitivity and specificity of these techniques in lung cancer diagnosis and staging are provided separately. Perhaps the greatest advantage of EUS-FNA over EBUS-TBNA is the ability to reach the left adrenal gland, upper retroperitoneal lymph nodes, left lobe of the liver, and abdominal ascites, therefore confirming stage IV disease. (See "Tumor, node, metastasis (TNM) staging system for lung cancer" and "Selection of modality for diagnosis and staging of patients with suspected non-small cell lung cancer", section on 'Modality choice - mediastinal sampling' and "Procedures for tissue biopsy in patients with suspected non-small cell lung cancer", section on 'Transesophageal endoscopic ultrasound'.)
Staging esophageal cancer — EUS is used in local staging of esophageal cancer (squamous cell carcinoma and adenocarcinoma). It is also useful in cases of biopsy-negative esophageal strictures when malignancy is suspected [44]. The information obtained is used to determine the next appropriate treatment. Details are discussed separately. (See "Endoscopic ultrasound for evaluating patients with esophageal cancer".)
Staging extrathoracic tumors — EUS-FNA has been used to successfully diagnose mediastinal metastases from extrathoracic tumors including pancreatic, breast, upper gastrointestinal, head and neck, salivary, and colorectal cancers, although many of these studies also included EBUS-TBNA as a staging modality in their analysis [23,45,46].
Therapeutic — Rare case reports suggest that EUS may be used for abscess or cyst drainage [47,48]. Preclinical studies have reported successful radiofrequency ablation (RFA) of mediastinal lesions in pigs using EUS, but this has not been studied in humans and remains investigational [49].
CONTRAINDICATIONS — There are few contraindications to EUS-guided sampling. The most common contraindications are those related to routine upper endoscopy and include:
●Patients who cannot tolerate routine moderate sedation (eg, poor performance status), monitored anesthesia care (MAC), or general anesthesia
●Patients with some bleeding disorders may still undergo EUS, but FNA (biopsy) would be determined on a case by case basis
●Patients who are hemodynamically unstable
●Patients with esophageal obstruction can still undergo EUS but the imaging is limited to an area proximal to the level of obstruction
COMPLICATIONS — EUS and EUS-guided sampling are generally safe procedures [50-52]. Common complications are often due to the effect of procedural sedation (eg, hypotension, cardio-respiratory arrest) and less commonly due to the endoscopy itself or sampling (eg, hemorrhage, viscus rupture, infection). Complications, most often infection, appear to be highest in those who undergo EUS-guided sampling for cystic rather than solid lesions of the mediastinum, and possibly in those with sarcoidosis [36,50,51,53]. Injury to the thoracic duct resulting in chylothorax following EUS-guided fine needle aspiration of a large mediastinal fibrous tumor has been reported [54]. Death is rare and likely related to the underlying performance status of the patient.
One systematic review of 190 studies that included EUS-guided fine needle aspiration (FNA) and endobronchial ultrasound transbronchial needle aspiration (EBUS-TBNA) reported severe adverse events in 0.14 percent and minor adverse events in 0.22 percent [55]. Another study reported a similar low serious adverse event rate in patients undergoing EUS-FNA (0.3 percent) [51]. In another study death was reported rare and more likely to occur in those with poor performance status (<0.05 percent) [56].
Endoscopy-related — Many of the complications associated with EUS are due to upper gastrointestinal endoscopy (eg, esophageal or pharyngeal wall perforation, infection) and/or the associated sedation and anesthesia (eg, hypotension, respiratory arrest). Esophageal rupture during EUS staging of the mediastinum is rare but is serious, requiring surgical management [57]. These complications are discussed separately. (See "Anesthesia for gastrointestinal endoscopy in adults" and "Gastrointestinal endoscopy in adults: Procedural sedation administered by endoscopists", section on 'Adverse events' and "Overview of upper gastrointestinal endoscopy (esophagogastroduodenoscopy)", section on 'Complications'.)
Procedure-related — Reported complications of EUS-guided sampling are very rare and include [53,58-61]:
●Infection (mediastinitis, pneumonia, candida)
●Hemorrhage (usually minor; hemorrhage requiring transfusion is rare)
●Esophageal wall or posterior pharyngeal perforation (very rare)
●Pneumothorax or pneumomediastinum (rare but can occur if the pleural membrane is breached when sampling pleural fluid)
●Fistula formation
●Needle tracking with malignant or infectious cells (rare)
Among studies, the rate of EUS-FNA-related complications is higher in cystic than solid lesions, particularly infectious complications [53,58-60]. As an example, a prospective cohort study of 457 patients (554 lesions), complications were less frequent with solid lesions (eg, mediastinal masses, lymphadenopathy) than cystic lesions (0.5 versus 14 percent) [53]. Infectious complications may also be higher in patients with sarcoidosis. (See 'Sarcoidosis' above.)
Patients with infectious (eg, chest pain, fever) or hemorrhagic complications (hematemesis) or patients with pneumomediastinum or pneumothorax (chest pain dyspnea) should be treated supportively. (See "Treatment of secondary spontaneous pneumothorax in adults" and "Diagnosis, management, and prevention of pulmonary barotrauma during invasive mechanical ventilation in adults".)
Patients rarely complain of odynophagia or chest pain after EUS-FNA, EUS-CNB, or EBUS-TBNA. If these symptoms are present, further investigation is warranted to rule out complications.
Mediastinal-esophageal fistula formation has been reported after EUS-FNA of mediastinal adenopathy that was due to tuberculosis (TB) [62,63]. During treatment, patients need to avoid oral feeding and be fed parenterally or enterally using a post-pyloric feeding tube. Symptoms generally resolve a few months after instituting TB treatment.
Contamination of the aspiration or biopsy needle with cells that can be transferred to the esophagus is rare but has been reported [60,61]. In particular, EUS-FNA of potentially resectable primary liver cancer or primary renal cancer is not advised.
Elastography — A newer technique used to differentiate between benign and malignant lymph nodes is elastography. This feature is a real-time noninvasive strain imaging applied during EUS that measures the relative tissue stiffness. Malignant lymph nodes have increased tissue stiffness. In a meta-analysis, the sensitivity and specificity in detecting malignant lymph nodes during EUS was 88 and 85 percent, respectively [64]. This feature is useful in improving diagnostic yield, especially in uncertain cases. This feature is not currently widely available.
SUMMARY AND RECOMMENDATIONS
●Technique and procedure – Endoscopic ultrasound (EUS) is performed with an echoendoscope (an endoscope with an ultrasound transducer in its tip). While radial EUS can image the mediastinum, curvilinear EUS provides real-time ultrasonographic guidance for fine needle aspiration (EUS-FNA) and EUS-guided core needle biopsy (EUS-CNB). EUS should only be performed by a skilled endoscopist. Pre- and post-procedural care is similar to that for patients who undergo upper gastrointestinal endoscopy. (See 'Technique and procedure' above and "Endoscopic ultrasound-guided fine needle aspiration in the gastrointestinal tract" and "Overview of upper gastrointestinal endoscopy (esophagogastroduodenoscopy)".)
●Indications – The primary indication for mediastinal-guided EUS-guided sampling is for the diagnosis and staging of lung cancer and esophageal cancer but can also be used to diagnose a wide range of other benign (eg, sarcoidosis, tuberculosis) and malignant lesions of the mediastinum (eg, thymic carcinoma, metastases from extrathoracic sites).(See 'Indications' above.)
•EUS-guided sampling is best suited to target lesions that are paraesophageal, in the posterior mediastinum, or in the inferior mediastinum (figure 1 and figure 2) and is not suitable for lesions in the anterior or superior mediastinum.
•EUS-guided sampling may also be an alternative modality in patients who are not surgical candidates and may occasionally be used to obtain pleural fluid or tissue in extrathoracic sites including the celiac lymph nodes, left adrenal gland, and left lobe of the liver.
•Therapeutic applications are rare (cyst or abscess drainage).
●Contraindications – There are few contraindications to EUS-guided sampling which are most commonly related to upper endoscopy itself including poor performance status, uncorrectable coagulopathy, and hemodynamic instability. (See 'Contraindications' above.)
●Complications – EUS and EUS-guided sampling are generally safe procedures and complications are infrequent (<1 percent). (See 'Complications' above.)
•Common complications are often due to the effect of procedural sedation (eg, hypotension, respiratory arrest) and less commonly due to the procedure itself or sampling (eg, hemorrhage, viscus rupture, infection).
•Complications appear to be highest in those who undergo EUS-guided sampling for cystic rather than solid lesions of the mediastinum and possibly in those with sarcoidosis.
•Patients who undergo EUS-mediated biopsy of cystic lesions are prone to infection such that prophylactic antibiotics should be administered (table 1).
•Death is rare and likely related to the underlying performance status of the patient.
آیا می خواهید مدیلیب را به صفحه اصلی خود اضافه کنید؟
