Usefulness of endoscopic ultrasound-guided fine needle aspiration in the diagnosis of hepatic, gallbladder and biliary tract Lesions

Abstract

Endoscopic ultrasound (EUS)-guided fine needle aspiration (FNA) of the liver is a safe procedure in the diagnosis and staging of hepatobiliary malignancies with a minimal major complication rate. EUS-FNA is useful for liver lesions poorly accessible to other imaging modalities of the liver. EUS-guided FNA of biliary neoplasia and malignant biliary stricture is superior to the conventional endoscopic brushing and biopsy.

Key Words: Endoscopic ultrasound; Fine needle aspiration; Hepatocellular carcinoma; Bile duct stricture; Gallbladder; Cholangiocarcinoma; Biliary drainage

Core tip: The present article reviews the usefulness of endoscopic ultrasound-guided fine needle aspiration (EUS-FNA) in patients with focal liver and biliary tract lesions. We conducted MEDLINE search using the terms “endoscopic ultrasound-guided fine needle aspiration”, “focal liver lesions” and “biliary tract lesions”, “EUS and biliary stricture”, EUS and focal liver mass”, “EUS and cholangiocarcinoma” and “EUS and gallbladder” to retrieve articles published between 1999 to 2014.

INTRODUCTION

Endoscopic ultrasonography (EUS) has become an indispensable diagnostic and therapeutic procedure in the field of gastroenterology coupling endoscopy with high frequency echo sonography. Endoscopic ultrasound-guided fine needle aspiration (EUS-FNA) is performed using the curved linear array echoendoscope (Figure 1) using various needles (Figure 2). The recently introduced forward viewing linear echoendoscope is gaining momentum in endoscopic ultrasound-guided interventions (Figure 1). EUS-FNA is minimally invasive that is utilized for procurement of tissue from unresectable tumors. EUS-guided fine needle aspiration is used increasingly for the diagnosis of mediastinal, pancreatic and gastric tumors, however, not much is known about EUS-FNA in hepatic lesions. EUS imaging of the liver is currently limited to the left lobe, the proximal right lobe, the hilum and part of the intrahepatic biliary tract. EUS-FNA may be considered as an alternative to liver percutaneous biopsy in patients at high risk of bleeding or with small lesions of the liver uncharacterized by cross-sectional abdominal imaging. EUS-guided biliary drainage (EUS-BD) was developed using a curved linear array echoendoscope for cases with failed endoscopic biliary drainage. Table 1 summarizes the use of endoscopic ultrasound-guided fine needle aspiration in the diagnosis and management of hepatic, gallbladder and biliary tract lesions.

Table 1 Summary of the use of endoscopic ultrasound-guided fine needle aspiration in the diagnosis of hepatic and biliary tract lesions.

Diagnosis of focal malignant and benign liver lesions

Diagnosis of malignant biliary stricture and neoplasia

Preoperative staging of hepatocellular carcinoma and lymph node metastasis

Ablation of focal malignant and benign liver lesions

Liver biopsy

Fluid acquisition and biopsy of peritoneal and omental deposits

Drainage of intrahepatic and extrahepatic biliary tree

Drainage of hepatic abscesses

Figure 1 The curved linear array videoechoendoscope (GF-UCT180) (Back); The new prototype forward viewing linear array videoechoendoscope (TGF-UC180J) (Front).

Figure 2 Various echoendoscopic needles used for fine needle aspiration.

FEASIBILITY OF ENDOSCOPIC ULTRASOUND-GUIDED FINE NEEDLE ASPIRATION OF FOCAL LIVER LESIONS

Focal liver lesions include simple liver cyst, focal nodular hyperplasia, hepatic adenoma, hepatic hemangioma, regenerative nodular hyperplasia, biliary cystadenoma, intrahepatic cholangiocarcinoma, hepatocellular carcinoma and metastatic liver lesions. The majority of these lesions can be diagnosed with certainty by cross-sectional abdominal imaging and by percutaneous liver biopsy. However, small lesions less than 1-cm in diameter may not be well characterized by abdominal ultrasound (US), computed tomography (CT) and/or magnetic resonance imaging (MRI). In general, the lowest ultrasound frequency available should be used to maximize penetration. EUS-guided liver biopsy using a 19-gauge FNA needle (non-Trucut) and EUS-guided Trucut needle appear to be feasible, safe and provide excellent diagnostic yield and specimen adequacy[1-3]. In a retrospective study by DeWitt et al[4], EUS-FNA of liver lesions that range from 3-40 mm in size was performed in 77 patients[4]. Of these lesions, 58% were diagnostic for malignancy, 33% were benign, and 9% were nondiagnostic. In a study by tenBerge et al[5], EUS-FNA was used to sample liver lesions in 167 patients. The indications were pancreatic mass in 37%, liver metastasis of unknown origin in 20%, esophageal, gastric and liver masses. EUS-FNA of the liver revealed malignancy in patients when abdominal ultrasonography-guided FNA and CT-guided FNA have failed. Crowe et al[6] compared 34 percutaneous computerized tomographic-guided fine needle aspiration liver biopsies and 16 EUS-FNA liver biopsies showed comparable results. These studies and others suggest that EUS-FNA is feasible and comparable to US/CT-guided biopsy in the diagnosis of patients with focal liver lesions.

Malignant focal/metastatic liver lesions

EUS can provide high resolution imaging of the left hepatic lobe to detect unsuspected metastatic disease during staging and may deter from unnecessary surgery[7,8]. EUS-FNA of liver lesions can provide useful information for future management. Hepatic metastasis is generally echo-poor without a distinct border such as the one seen in pancreatic and colon metastasis (Figure 3) or echo-rich such as seen in metastatic neuroendocrine tumors and renal cell carcinoma (Figure 4). EUS-FNA can detect tumors less than 3 mm in size[7]. Solid liver lesions accessible by EUS may be safely sampled by EUS-FNA. The use of stylet during FNA does not appear to confer any advantage with regards to the adequacy of specimen or diagnostic yield of malignancy[9]. In a prospective study of 132 subjects with newly diagnosed tumors, the diagnostic accuracy of EUS/EUS-FNA and CT scan in detecting hepatic metastasis was 98% and 92%, respectively (P = 0.0578)[10]. In a large single-center experience, the sensitivity of EUS-FNA for the diagnosis of liver cancer ranged from 82% to 94%[4]. In a prospective study of 41 patients, 33 of whom had clinical findings suggestive of liver malignancies, EUS-FNA provided biopsy specimens in 40/41 patients[11]. Combining histological and cytological features had a sensitivity of 94%, specificity of 100%, negative predictive value of 78%, and positive predictive value of 100%[11]. These data suggest that EUS-FNA is a sensitive diagnostic procedure in patients with focal malignant liver lesions especially to those confined to left hepatic lobe.

Figure 3 Curved linear echoendoscope showing a rounded hypoechoic left lobe liver lesion with no well-defined border representing liver metastasis in a

Figure 4 Hyperechoic rounded liver lesion (white arrow) representing a metastasis in patient with pancreatic neuroendocrine tumor with biliary obstruction and dilated intrahepatic duct (A); endoscopic ultrasound-guided fine needle aspiration of liver lesion using 22 gauge needle (B).

Hepatocellular carcinoma

EUS-FNA may be useful in the diagnosis of focal liver lesions, early hepatocellular carcinoma, and evaluation of perihepatic adenopathy[12-15]. Hepatocellular carcinoma (HCC) may appear on EUS images either as hypoechoic or hyperechoic[16]. Burrel et al[17] showed that lesions smaller than 1cm in diameter are missed in a significant percentage (70%) of the patients by modalities such as CT imaging[14,18] and magnetic resonance imaging[18]. EUS and EUS-FNA are particularly valuable for the preoperative staging of hepatocellular and metastatic liver carcinoma. In a study by Awad et al[18], EUS identified liver lesions 0.3-14 cm in size in all 14 study patients with hepatocellular cancer and metastatic lesions who underwent both dynamic CT scans and EUS[18]. Moreover, in 28% of the patients, EUS identified new lesions less than 0.5 cm in size. In a prospective single-center study evaluating 17 patients who underwent cross-sectional imaging and EUS, 9 had liver tumors[16]. EUS-FNA established a tissue diagnosis in 8 of the 9 cases. The diagnostic accuracy of transabdominal ultrasonography, abdominal CT, MRI, and EUS/EUS-FNA were 38%, 69%, 92%, and 94%, respectively[16]. Another retrospective study evaluated the sensitivity and complications of EUS-FNA of liver nodules in 14 patients, performed by single endoscopist[19]. Twenty-one percent of the cases were hepatocellular carcinoma. The sensitivity of diagnosis of malignant liver lesions utilizing cytology was 78.5%. However, combining clinical course and pathology increased the sensitivity to 100%. These data suggest that EUS has an excellent diagnostic accuracy in patients with HCC.

Moreover, EUS-guided fine needle aspiration of portal vein thrombus to detect malignancy has been described in literature[20,21]. More recently, a newly developed promising technique utilizing real time-sonoelastography (RTE) by EUS might improve the characterization and differentiation between benign and malignant focal liver lesions[22].

Screening and treatment of HCC

The use of EUS-FNA in screening for HCC is limited by the semi-invasive nature of the procedure as well as its inability to evaluate all liver segments at this time[13]. Nevertheless, EUS can provide an additional option for treatment in patients with hepatocellular carcinoma who are difficult to treat utilizing percutaneous ablative therapy such as endoscopic ultrasound-guided ethanol injection[23,24] and EUS-guided Nd:YAG laser ablation of a caudate lobe hepatocellular carcinoma[25].

Benign focal liver lesions

Large hepatic cysts are amenable to percutaneous drainage or surgical resection. EUS-guided ethanol injection has been shown to be effective in treating patients with large hepatic cysts especially in the left hepatic lobe. In a retrospective study evaluating 17 patients with 19 hepatic cysts (median cyst volume before therapy was 368.9 mL)[26], ten cysts were drained by the percutaneous approach and 8 cysts underwent EUS-guided aspiration and lavage treatment. During 15-mo follow-up, the cysts showed nearly 100% reduction in the EUS-guided group compared to 97% reduction in the percutaneous group. Furthermore, EUS-FNA has also shown excellent success rates in selected patients with hepatic abscesses. In a recent review of the literature by Singhal et al[27], seven studies have reported 100% technical and clinical success rates of EUS-guided drainage of hepatic abscesses in patients refractory or not amenable to percutaneous drainage.

Ascites and peritoneal metastasis

EUS-guided paracentesis is valuable in the cytologic diagnosis and staging of malignant ascites[28,29]. EUS frequently identifies ascites missed by other imaging modalities and may identify malignancy[30]. It is particularly useful when CT imaging does not identify abnormalities[31]. EUS-FNA can be performed safely for therapeutic paracentesis[32]. In a retrospective single center study that evaluated 101 patients who underwent EUS-guided paracentesis, the specificity, sensitivity, positive and negative predictive values, and diagnostic accuracy were 100%, 80%, 100%, 95% and 96%, respectively[29]. Furthermore, EUS-FNA can be used effectively and safely to obtain tissue from the peritoneum for diagnosis of tuberculous peritonitis[33]. EUS-FNA allows the sampling of peritoneal metastatic lesions, which appear on EUS images as hyperechoic compared to surrounding anechoic ascitic fluid (Figure 5). In a small study involving 12 patients with undiagnosed ascites, peritoneal deposits noted in 10 (83.3%) patients[34]. The cytological results were positive for malignancy in 6 of those patients, while the remaining four patients had inflammatory cells.

Figure 5 Peritoneal deposits in a patient with malignant ascites. Peritoneal implants appear as hypoechoic in comparison to the surrounding tissue but hyperechoic in comparison to the anechoic ascitic fluid (A); endoscopic ultrasound-guided fine needle aspiration of a large peritoneal deposit (B).

ENDOSCOPIC ULTRASOUND-GUIDED FINE NEEDLE ASPIRATION OF BILE DUCT, GALLBLADDER AND AMPULLARY LESIONS

Cholangiocarcinoma and proximal biliary strictures

Preoperative tissue diagnosis is required for hilar neoplasia [cholangiocarcinoma (CCA)] to avoid risk of unnecessary extensive surgery. Endoscopic transpapillary brush cytology and forceps biopsy are used for the pathological diagnosis of malignant biliary strictures. Endoscopic retrograde cholangiography (ERC) is currently the main diagnostic procedure performed to obtain sampling of the biliary tree. However, the sensitivity and specificity of obtaining a sample in biliary neoplasia is variable. EUS is capable of visualizing the hilum at the duodenal bulb by tracing the common bile duct (CBD) towards the liver hilum. In a meta-analysis of 36 studies by Garrow et al[35] EUS has a sensitivity of 78% and specificity of 84% in detecting malignant biliary strictures. Nayar et al[36] reported on 32 patients who underwent 36 procedures for hilar lesions. The overall sensitivity, accuracy, specificity, positive predictive value and negative predictive value of EUS-FNA were 52%, 68%, 100%, 100% and 54%, respectively. Fritscher-Ravens et al[37] prospectively evaluated 44 patients with hilar strictures diagnosed by CT and/or Endoscopic retrograde cholangiopancreaticography (ERCP) that were suspicious for hilar cholangiocarcinoma but had inconclusive tissue diagnosis. The sensitivity, accuracy, and specificity of EUS-FNA in this study were 89%, 91%, and 100%, respectively. Moreover, EUS and EUS-FNA changed preplanned surgical approach in about half of these patients[37]. The above studies suggest that hilar neoplasia can be sampled by EUS-FNA although the accuracy and sensitivity were not robust. Moreover, EUS-FNA may be considered in evaluating regional lymph nodes to evaluate for metastasis in patients with unresectable hilar cholangiocarcinoma[38,39]. EUS-FNA in patients with cholangiocarcinoma did not appear to adversely affect the overall survival[40].

Distal malignant biliary stricture

The sensitivity of EUS-FNA is much higher in distal malignant biliary strictures than proximal strictures. Malignant distal biliary strictures are most commonly secondary to pancreatic malignancy and/or distal bile duct cholangiocarcinoma (Figure 6). In a recent prospective comparative one-year study of 51 patients who underwent EUS and ERCP in the same session for evaluation of malignant biliary obstruction[41], EUS-FNA was superior to ERCP in tissue sampling for evaluating suspected malignant biliary obstruction, especially for pancreatic masses with an overall accuracy and sensitivity of 94% and 94% for EUS-FNA, and 53% and 50% for ERCP sampling, respectively. In an observation study of prospectively collected data of 228 patients with biliary strictures who underwent EUS[42]. Cholangiocarcinoma was detected in eighty-one, Fifty-one of the patients (63%) had distal and 30 (37%) had proximal CCA. The overall sensitivity of EUS-FNA for the diagnosis of CCA was 73% and was significantly higher in distal compared to proximal CCA (81% vs 59%, respectively; P = 0.04). Furthermore, a retrospective analysis of 342 patients who underwent EUS-FNA after presenting with biliary stricture and obstructive jaundice[43] showed an overall 92.4% accuracy of EUS-FNA for diagnosing malignancy with 91.5% sensitivity and 80.9% negative predictive value. These studies and others demonstrate the higher sensitivity of EUS-FNA in distal biliary stricture. Moreover, EUS-FNA appears equivalent to ERCP sampling for biliary tumors and indeterminate strictures[41] and may provide a diagnosis of malignancy when ERCP sampling is indeterminate[44]. Moreover, EUS-FNA can have a role in diagnosing other lesions that may mimic cholangiocarcinoma and present either as a mass or with obstructive jaundice. Such lesions as epithelial vs nonepithelial tumors, neuroendocrine tumors, lymphoma, and metastasis from other primaries[45,46].

Figure 6 Malignant distal biliary strictures are most commonly secondary to pancreatic malignancy and/or distal bile duct cholangiocarcinoma. A: Distal common bile duct stricture secondary to a large heterogenous hypoechoic pancreas head mass with irregular border; B: Endoscopic ultrasound-guided fine needle aspiration of pancreas head mass/stricture; C: Distal irregular common bile duct stricture seen on cholangiogram.

Endoscopic ultrasound-guided biliary access/drainage

ERCP is currently the standard of care for biliary drainage, however the failed cannulation rates ranges 3% to 5% in experienced hands. EUS-guided biliary drainage includes EUS-guided choledochoduodenostomy[47], hepaticogastrostomy[48], and EUS-guided transpapillary rendezvous biliary drainage[49]. The procedure technique has been described as follows[50]: the linear-array EUS scope is placed against the cardia or lesser curve of the stomach in a patient with dilated left intrahepatic biliary tree for hepaticogastrostomy or against the bulb of the duodenum for choledochoduodenostomy. The dilated bile duct or left intrahepatic duct which appears as hyperechoic structure running alongside the portal venous system without Doppler flow signals is then identified and punctured using a 19-guage or 22-guage needle. The stylet is then removed followed by contrast injection to visualize the biliary tree under fluoroscopy. A 0.035’’ or 0.021’’ guidewire is subsequently passed via the FNA needle into the bile duct or dilated intrahepatic duct. The needle knife is then used to make an incision of the gastric or duodenal wall under EUS guidance for preparation of dilation of the transmural tract. Dilation can be performed using 4.5F to 5F ERCP cannula, 4-mm or 6-mm dilating biliary balloon. A plastic biliary stent or self-expandable fully covered metal stent can then be placed[51,52]. In a large multicenter, nonrandomized retrospective study of 240 patients who underwent EUS-guided bile duct access and drainage[53], success was achieved in 87% of the cases. Similarly, in extrahepatic and intrahepatic approaches, the success rate was 84.3% vs 90.4%; respectively.

Gallbladder lesions

EUS-FNA has gained momentum in sampling gallbladder masses for diagnostic and staging purposes with accuracy reaching 100% in early stages. Sadamoto et al[54] reported EUS accuracy of 100% for in situ tumors (Tis), 76% for T1, 85% for T2, and 93% for T3 and T4 lesions. In one series, EUS-FNA provided accurate diagnosis of six patients with obstructive jaundice (five with gallbladder adenocarcinomas) where CT scans mostly failed to detect the causing lesions[55]. Jacobson et al[56] described similar findings in four out of five patients diagnosed with adenocarcinoma of the gallbladder. Meara et al[57] reported sensitivity of 80% and specificity of 100% in diagnosing gallbladder wall lesions.

EUS and transabdominal US are usually viewed as good tools to evaluate gallbladder polyps with superior sensitivities for EUS 97% vs transabdominal US 71% in one study[58]. Diagnostic distinction between malignant and non-malignant polyps for the purpose of staging and determining next steps management, remains mostly dependent on the ultrasonographic features of the polyps rather than tissue sampling[54]. No reports of the use of EUS-FNA in approaching gallbladder polyps were found. Endoscopic ultrasound-guided transmural gallbladder drainage with placement of self-expandable stent has been reported and is technically successful for the management of acute cholecystitis in high risk patients[59-61]. Table 2 summarizes the sensitivity, specificity and diagnostic accuracy in the reported studies.

Table 2 Summary of the sensitivity, specificity and diagnostic accuracy of endoscopic ultrasound-guided fine needle aspiration in the diagnosis of focal hepatic, gallbladder and biliary tract lesions.

Study, year, number	Sensitivity (%)	Specificity (%)	Diagnosticaccuracy (%)
Focal malignant liver lesions
DeWitt et al[4], 2003, n = 77	82-94	-	-
Hollerbach et al[11], 2003, n = 44	94	100	-
Singh et al[16], 2007, n = 17	89	100	94
	CT (71)	67	69
	MR (86)	100	92
Prachayakul et al[19], 2012, n = 14	78.5	-	-
Malignant biliary tract and gallbladder lesions
Garrow et al[35], 2007, 36 studies, n = 3532	78	84	90
Nayar et al[36], 2011, n = 32	52	100	68
Fritscher-Ravens et al[37], 2004, n = 44	89	100	91
Weilert et al[41], 2014, n = 51	94	100	94
	ERCP brushing (50)	100	53
Mohamadnejad et al[42], 2011, n = 228	73	-	-
	ERCP brushing (27)		-
Tummala et al[43], 2013, n = 342	91.5	-	92.4
Meara et al[57], 2006, n = 53	80	100	-
	ERCP brushing (13)	75	-

CT: Computed tomography; MR: Magnetic resonance; ERCP: Endoscopic retrograde cholangiopancreaticography.

Ampullary tumors

EUS-FNA can provide an excellent diagnostic accuracy in distinguishing between benign and malignant ampullary tumors in comparison to surface biopsy with duodenoscopy, and/or intra-ampullary biopsy, and/or brush cytology with ERCP, and/or intra-ampullary biopsy after endoscopic sphincterotomy (100% vs 70%)[62]. Furthermore, the diagnostic accuracy of EUS-FNA for ampullary tumors supersedes that without EUS-FNA. In a retrospective study by Roberts et al[63], rates of diagnostic accuracy in high-grade dysplasia, low-grade dysplasia, and adenocarcinoma were 20%, 72%, and 96%, respectively, in the non-EUS group, and 50%, 93%, and 100%, respectively, in the EUS group.

ENDOSCOPIC ULTRASOUND AND RAPID ON-SITE CYTOLOGY EVALUATION

The diagnostic accuracy of EUS-FNA is dependent on how the sample is processed after acquisition. The presence of a rapid on-site cytology evaluation (ROSE) by a cytopathologist in the vicinity where the sample is obtained has been shown to improve the diagnostic yield of the procedure[64]. ROSE may allow a less number of needle passes and ensure adequacy of the sample obtained by onsite staining prior to completion of procedure. In general, the diagnostic yield of EUS-FNA with ROSE in most studies exceeds 90%. Meara et al[57] reported on 53 cases undergone EUS-FNA from 46 bile duct and seven gallbladder lesions where ROSE was available. All cases initially diagnosed as suspicious/malignant were confirmed on the final cytological interpretation. The specificity for EUS-FNA was 100% with sensitivity rates of 80% and 87% from clinically suspected malignancies of gallbladder and biliary tract, respectively. A retrospective study by Jhala et al[65] provided on-site diagnosis of malignancy on 485 EUS-FNA of the pancreas (n = 305), lymph nodes (n = 91), biliary tree (n = 47), liver (n = 15), gastrointestinal tract (n = 19), and adrenal gland (n = 8). A significantly higher degree of concordance was noted for unequivocal diagnosis of malignancy vs no malignancy (98.9% vs 67.2%) between on-site and final cytologic diagnosis. These studies have demonstrated ROSE by cytopathologist and interpretation significantly improves the diagnostic yield of EUS-FNA.

COMPLICATIONS OF ENDOSCOPIC ULTRASOUND-GUIDED FINE NEEDLE ASPIRATION OF HEPATIC AND BILE DUCT LESIONS

In a retrospective questionnaire sent to 130 EUS-FNA centers across the world[5]. 167 cases of EUS-FNA of the liver were reported by 21 centers. A complication was reported in 6 (4%) of the 167 cases including the following: death in 1 patient, bleeding (1), fever (2), and pain (2)[5]. EUS-guided liver biopsy appears to be safe and associated with no significant complications[2-4,66]. Several studies have reported no adverse events related to EUS-FNA of bile duct strictures, gallbladder and masses[41,42,56,57,67]. However, EUS-FNA of malignant biliary lesions was reported to have a risk of bleeding, infection, or pancreatitis in less than 2% of the cases[68]. Hemobilia was reported in 1.3% of patients who underwent EUS-FNA of malignant biliary stricture[42]. Bacteremia after EUS-FNA is rare. However, prophylactic antibiotics should be given prior and after EUS-FNA of biliary tract in patients with biliary obstruction. EUS-guided diagnostic abdominal paracentesis was not associated with any complication in one study[28]. Bile peritonitis has been reported after inadvertent biliary puncture during EUS-FNA[69]. Complications of EUS-guided biliary drainage included pneumoperitoneum 5%, bleeding 11%, bile leak/peritonitis 10%, and cholangitis 5%[53]. Needle track tumor seeding has been reported and is a risk after EUS-FNA of malignant biliary neoplasia[70,71]. EUS-FNA of malignant biliary stricture is considered a contraindication in patients eligible for liver transplantation. Cholecystitis and bile peritonitis have been reported after EUS-FNA of gallbladder lesions[72]. Bleeding after EUS-FNA of solid tumor is rare and appears as an expanding extraluminal echopoor region adjacent to the sampled lesion[73].

LIMITATIONS OF ENDOSCOPIC ULTRASOUND-GUIDED FINE NEEDLE ASPIRATION OF HEPATIC AND BILE DUCT LESIONS

The head of pancreas and CBD are not visualized after Roux-en-Y surgery and Billroth II surgery if the afferent limb is not intubated. Presence of vascular structures or collaterals in needle path may limit EUS-FNA of focal lesions. Because the right liver lobe is farther away from the probe, it is generally not seen except in small parts. The presence of pneumobilia, fatty infiltration, calcifications and extensive fibrosis may interfere with ultrasound beam and images. Endosonographer’s experience, time consumed to image the liver and patient’s body habitus are of critical importance to clearly identify and diagnose focal liver lesions. The miss rate for resectable pancreaticobiliary malignancy by EUS-FNA is rather small. Moreover, EUS and EUS-FNA may not be widely available and require an expertise with dedicated echosonographer in the field. With improving resolution and widespread use of EUS with dedicated formal training, small liver metastasis and other focal liver lesions are being increasingly detected. EUS does not use intravenous contrast to evaluate the nature of focal liver lesions and thus correlation with other cross-sectional imaging such as CT and/or MR is needed. However, the technology has dramatically improved. The use of color and power Doppler imaging, three-dimensional imaging, electronic scanning, tissue harmonic imaging, elastography, and recently contrast-enhanced images have improved the diagnostic capability. The depth of tumor infiltration and differentiation between infiltrating or exophytic lesions can now be assessed with greater accuracy[74-76].

CONCLUSION

Endoscopic ultrasound-guided fine needle aspiration of the liver, gallbladder and biliary tract is feasible and provides an excellent diagnostic accuracy. The presence of ROSE has increased the diagnostic yield. EUS-FNA is capable to differentiate between focal benign or malignant liver lesions. The widespread of EUS and increase formal training have enhanced the diagnostic and therapeutic armamentarium of EUS in hepatobiliary disorders. EUS-FNA should be considered as an adjunct to other cross-sectional imaging in the differentiation between benign and focal hepatobiliary disorders. EUS-guided interventions such as fine-needle injections, tumor ablative therapies and biliary drainage have increased the application of EUS and is considered as an adjunct to other modalities.