Gastrointestinal Imaging: Endoscopic Ultrasound - Gastroenterology

Gastrointestinal Imaging: Endoscopic Ultrasound

MICHAEL F. BYRNE and PAUL S. JOWELLDivision of Gastroenterology, Department of Medicine, Duke University Medical Center, Durham, North Carolina

The development of endoscopic ultrasound (EUS)since its introduction in the early 1980s has added

greatly to the quality of imaging of the gastrointestinaltract. EUS is probably the investigation of choice forlocal staging of several gastrointestinal tumors and eval-uation of submucosal masses. In addition to well-estab-lished indications, newer applications of EUS are emerg-ing. For example, EUS is proving useful in evaluation ofother pathologies such as chronic pancreatitis and cho-ledocholithiasis. However, the applications of EUS are nolonger limited to the gastrointestinal system, and recentstudies have suggested that it has a significant role toplay in the staging of nongastrointestinal tumors, such asnon–small cell lung cancer.

EUS has progressed from being a purely imagingmodality to one that can provide a tissue diagnosis byfine needle aspiration (FNA) and deliver therapy (inter-ventional EUS). Indeed, EUS-guided FNA should nowbe regarded as a routine extension of EUS. The ability toobtain tissue under EUS has made its acceptance morewidespread. The field of interventional EUS is one that isin its relative infancy, but many potential applicationsare under investigation, and there is real promise withseveral of these applications. It is likely, as with thedevelopment of interventional endoscopic techniques,that many of these procedures will be accepted as stan-dard in the future. There continue to be technologicaladvances in the instrumentation used, such as mini-probes and thinner video echoendoscopes. Because theimpact of EUS on patient care is increasingly docu-mented in terms of both improved diagnostic accuracyand reduction in costs, the clinical applications of thistool will be difficult to ignore. After years of beinglimited mainly to academic centers, EUS is finally beingperformed more widely, but there are training issues thatremain to be addressed. This review describes the estab-lished and more tentative indications for EUS (Table 1)and covers the applications of EUS in FNA and as aninterventional tool. Potential future applications are alsodiscussed.

Instrumentation

EUS is performed using either dedicated echoen-doscopes or standard endoscopes through which mini-probes are passed. Two main types of echoendoscopes areused in clinical practice—radial and curvilinear array.1–3

Radial imaging uses a rotating 360° transducer andprovides an image in a plane perpendicular to the direc-tion of insertion of the echoendoscope. The image ob-tained is thus analogous to that of cross-sectional com-puted tomography (CT) and is more easily interpretableas a rule than the images produced by curvilinear arrays.A new echoendoscope using an electronic transverse arraytransducer is due to be released shortly. This gives animage in the same axis as the radial echoendoscope exceptthat the image is about 270°. Linear devices incorporatea transducer placed at the tip of an oblique viewingendoscope and provide sector images in a plane parallelto the direction of insertion of the echoendoscope. To thenovice, this latter type of imaging is less intuitive, butuse of linear devices offers the significant advantage offacilitating intervention. For example, FNA and biopsyis usually performed using a linear device because one isable to view the entire length of insertion of the needle.Many linear echoendoscopes and the new electronictransverse array echoendoscopes also offer features such aspulse Doppler and color Doppler.

Standard echoendoscopes use ultrasound frequenciesbetween 5 and 12 MHz.1–3 This allows for imaging oftissue up to 5–6 cm from the transducer but givesrelatively low resolution. A significant advance in EUScame with the development of catheter ultrasound probesor miniprobes, which are passed through the operating

Abbreviations used in this paper: CBD, common bile duct; CT,computed tomography; ER, endoscopic resection; ERCP, endoscopicretrograde cholangiopancreatography; EUS, endoscopic ultrasound;FNA, fine needle aspiration; IDUS, intraductal ultrasound; MRCP, mag-netic resonance cholangiopancreatography; MRI, magnetic resonanceimaging.

© 2002 by the American Gastroenterological Association0016-5085/02/$35.00

doi:10.1053/gast.2002.33576

GASTROENTEROLOGY 2002;122:1631–1648

channel of standard endoscopes. Frequencies between 7.5and 20 MHz are used with these miniprobes, allowingfor high resolution of structures within 1–2 cm of thetransducer (Figure 1A and B).4 Miniprobes are particu-larly useful for examining the layers of the gastrointes-tinal wall. In addition, use of miniprobes allows theoperator to use a standard endoscope, which is easier touse than the stiffer echoendoscopes. Intraductal ultra-sound (IDUS) with high-frequency catheter probes isincreasingly being used in imaging of the biliary tree andthe pancreas. The diagnostic abilities of standard EUS,with or without FNA, is described first for all majorsystems. EUS as a current or potential therapeutic tool isthen addressed.

General PrinciplesThe Gut Wall

Standard EUS is able to show 5 layers of thegastrointestinal wall of alternating hyperechoic and hy-poechoic layers as described in Table 2. With higher-frequency probes, 7 or more layers can be seen.5,6

Cancer Staging

Cancer staging is probably the most commonindication for EUS. Because it can delineate the compo-nent layers of the gut wall, EUS is very well suited toclassifying gastrointestinal cancers arising from the mu-cosa using the widely accepted TNM classification.3 It isalso useful for some extraluminal malignancies such aspancreatic cancer, but, in general, EUS is not useful fordetecting distant metastases, other than for esophagealcancer (in which celiac nodes can be seen)3 and for somecancers with liver metastases. A recent abstract reporteda retrospective analysis of 167 cases of EUS FNA of livermasses and found that adequate specimens were obtainedin 96% of cases and that they were positive for malig-nancy in 86% of cases.7 There was a high success rate forEUS FNA in cases in which previous EUS FNA or CTFNA had failed and the overall complication rate was low

at around 4%. Hence, EUS FNA may actually giveimportant information about the M stage of a tumor.

Lymph Nodes

Assessment of nodes plays a critical part in thestaging of gastrointestinal cancers. EUS offers a greatadvance in lymph node imaging because one can visual-ize nodes as small as 2–3 mm. The sonographic appear-

Table 1. Indications for EUS

Established indications for EUSDebatable

indications for EUS

Staging of gastroesophageal cancers Acute pancreatitisStaging of rectal cancerStaging of pancreatic, ampullary, and

biliary cancersEvaluation of submucosal massesEvaluation of pancreatic cysts

Detection of pancreasdivisum

Assessment of portalhypertension

Assessment of IBDDiagnosis of chronic pancreatitisDetection of choledocholithiasisStaging of lung cancer

Figure 1. (A) Endoscopic image showing submucosal mass in loweresophagus. (B) EUS with a 20-MHz miniprobe shows an anechoicstructure (single arrowhead) in the submucosa. This likely representsa duplication cyst. The muscularis propria is also labeled (doublearrowhead).

1632 BYRNE AND JOWELL GASTROENTEROLOGY Vol. 122, No. 6

ance of nodes can suggest the likelihood of a benign ormalignant nature, although there are no entirely reliablecriteria. Features suggestive of malignant lymph nodesinclude large size (�1 cm), hypoechoic pattern, sharpborders, and rounded shape,8,9 but the reliability of EUSfor determining the nature of lymph nodes has improvedgreatly with the use of EUS-guided FNA.10

FNA

Although EUS has undoubtedly improved theaccuracy of local staging of tumors, reliable differentia-tion of benign from malignant lesions remained difficult,a problem that has been overcome somewhat with the useof EUS-guided FNA. This practice has added greatly tothe diagnostic power of EUS and has led to a greatincrease in the number of indications for EUS. The use ofFNA is highlighted in the various systems described inthis review.

EUS-guided FNA is usually performed using linear-array echoendoscopes because the whole length of theneedle can be seen in real time (Figure 2). In addition,linear-array systems have pulse and color Doppler, theuse of which improves the safety of FNA because one canconfirm blood flow and avoid vessels.11 FNA is possibleusing radial array systems but is technically more diffi-cult, carries a higher complication rate, and is not rec-ommended.11,12 Needles ranging in size from 22–25gauge are routinely used for FNA, with a 19-gaugeneedle available for select indications.

EUS-guided FNA using linear array has a complica-tion rate less than 1%.3,12,13 A recent study described alow complication rate (0.5%) for FNA of solid lesionsbut a higher rate of 14% for FNA of cystic lesions,14

predominantly because of infectious or hemorrhagicevents after puncturing cystic lesions. The theoretic riskof tumor seeding has not yet been reported.15 As withevery technique, the success of EUS FNA is somewhatoperator dependent, but the overall diagnostic rate for

EUS-guided FNA is probably over 80%. In a series of327 lesions, EUS FNA had an overall accuracy for thediagnosis of malignancy of 86%.16 It appears to be mostsensitive for lymph nodes and mediastinal masses (up to95% sensitivity) and less so for pancreatic cancers. In aseparate study of 457 patients, the investigators con-cluded that EUS FNA accurately evaluates peri-intestinallesions and improves lymph node staging accuracy.14

One key factor in diagnostic yield is the presence of anexperienced cytopathologist at the time of the procedure.This is our practice and undoubtedly adds greatly to thesuccess rate of FNA.

Despite the invaluable contribution of FNA to diag-nosis of lesions, this modality still has limitations. Forpancreatic cancer, for example, there continue to be falsenegatives, but development of core biopsy needles mayimprove this situation. However, it is interesting to notethat getting larger samples using suction during FNAmay not necessarily help in diagnosis because samplesmay be quite bloody (something which hinders detectionof malignant cells).17 In fact, lymph node sampling mayactually be more diagnostic if done without suctioning.

Gastroesophageal MassesEsophageal Cancer

There are a variety of therapeutic options foresophageal cancer including surgery, radiotherapy, che-motherapy, photodynamic therapy, palliation, and mul-timodal combination therapy.18 The appropriate treat-ment for esophageal carcinoma is particularly dependent

Table 2. Layers of Gastrointestinal Wall Seen at EUS

EUSlayers

Standard EUS(5–7.5 MHz)

High frequencyEUS (12–20 MHz)

1 Superficial mucosa Superficial mucosa2 Deep mucosa Remainder of lamina propria3 Submucosa Interface of layer 2 and

muscularis mucosae4 Muscularis propria Remainder of muscularis

mucosae5 Serosa or adventitia Submucosa6 Circular muscularis propria7 Septum between layers 6 and 88 Longitudinal muscularis propria9 Serosa or adventitia

Figure 2. EUS image of a mass (single arrowhead) in the head ofpancreas. EUS FNA was performed and confirmed pancreatic adeno-carcinoma. The tip of the needle (double arrowhead) is seen in themass. A bile duct stent previously placed is also visualized (triplearrowhead).

May 2002 ENDOSCOPIC ULTRASOUND 1633

on accurate staging, and EUS plays an increasingly pivo-tal role in this staging, which uses the TNM classifica-tion. Superiority of EUS over CT for local staging ofesophageal cancer has been shown in several studies,19,20

and the usefulness of EUS is emphasized by the closecorrelation between survival rates and EUS staging.21 Ina recent study, the authors were able to predict intra-esophageal or extraesophageal extension of tumor with ahigh degree of accuracy based on EUS measurements ofthe maximal thickness of the esophageal mass.22 Thesensitivity of T staging for esophageal cancer by EUS isbetween 85% and 95%.19,20 Using EUS, it is possible toassess whether esophageal tumors are limited to themucosa or submucosa (stage T1 lesions) (Figure 3A andB) or have invaded into (T2) or through (T3) the mus-cularis propria. Extraesophageal invasion (stage T4) ofadjacent organs can also be assessed by EUS.3 However,the sensitivity for N staging is lower but still high at70%–80%.19,20 This compares with sensitivities ofaround 50% for both T and N staging by standard CT.If the nodal appearance is equivocal, EUS FNA can beused, increasing the accuracy of N staging to around90%.23 However, if the needle passes through the pri-mary tumor en route to the lymph node, sampling errormay occur with a false-positive lymph node.

EUS FNA may also be able to offer information inrelation to the M stage of an esophageal cancer. In astudy of 198 patients with esophageal cancer, enlargeddistant lymph nodes (mediastinal or celiac) were detectedand biopsies were performed by EUS FNA in 40 cases24

with 97% sensitivity and 100% specificity. EUS FNA ofliver masses may also give information in relation to theM stage of an esophageal tumor.7 The usefulness of EUSFNA in determining the M stage of esophageal cancerremains to be proven in further studies, but early resultsare encouraging. However, EUS will not usually detectpretracheal adenopathy, and thus the current recommen-dation is that combination with CT allows completion ofthe TNM classification of an esophageal tumor.25

Although EUS is undoubtedly useful in pretreatmentstaging of esophageal cancer, it is not clear if EUSstaging before treatment predicts response to neoadju-vant therapy, and it would appear that it is much lessreliable for staging after neoadjuvant therapy. One groupfound that neither the pretreatment T or N stage corre-lated with a complete pathologic response.26 In anotherstudy in which 59 patients with esophageal cancer hadEUS before and after neoadjuvant therapy, EUS was only37% and 38% accurate for T and N staging, respectively,after therapy.27 Reduction in tumor size may be of somehelp prognostically,28 but a difficult problem with EUS

is that it cannot reliably distinguish inflammatory tissuefrom cancer.

Another problem with EUS and staging of esophagealcancers arises with advanced and stenotic lesions. It mayprove impossible to advance the EUS probe through thestenosis,29 and in this situation some authors have sug-gested dilatation to allow for EUS evaluation. Althoughan increased incidence of perforation has been reported,study results vary, and it now seems to be safe if generalprinciples of esophageal dilatation are applied.30,31 Be-cause resectability rates are very low for these advancedtumors, it is debatable whether EUS staging is beneficialor not.29 Use of blind probes, which can often be passed

Figure 3. (A) Endoscopic image of a TI lesion (adenocarcinoma) atthe gastroesophageal junction. (B) EUS shows this as a hypoechoicmass (single arrowhead) superficial to the muscularis propria (doublearrowhead).


beyond the lesion without the need for dilatation, may bea suitable compromise. Finally, miniprobes are useful forsmall esophageal lesions, which can be very difficult tovisualize with the standard echoendoscope.32

Gastric Cancer

EUS has also added greatly to the staging ofgastric cancers, but there are some limitations not en-countered with esophageal cancer. As with esophagealcancer, superiority of EUS over standard CT in assessingboth T and N stage has been shown,33–35 and EUS alsocompares favorably with intraoperative assessment.33

However, a relative inability to distinguish the muscu-laris propria from the serosa using EUS presents a prob-lem when trying to define a gastric lesion as T2 or T3.Inflammation and/or fibrosis associated with a cancer canlead to overstaging of a gastric tumor. In addition, aninability to distinguish malignant infiltration fromedema surrounding benign tumors or ulcers is also aproblem with EUS in the stomach and can make it verydifficult to distinguish a benign from a malignant gastriculcer.3

However, for cancers limited to the mucosa, EUSconfers significant advantages over CT and, with theadvent of mucosal resection for early gastric cancers,appropriate staging of a gastric cancer limited to themucosa or invading the muscularis mucosa is very im-portant.3 This detail is best provided with the use ofhigh-frequency miniprobes, which can delineate the gas-tric wall as a 9- or 11-layer structure as opposed to the 5layers seen with lower frequency EUS (Table 2). A veryclose correlation between EUS for early gastric cancersand the histology of resected specimens has been de-scribed.36 Combination of EUS and endoscopic mucosalresection for early gastric cancers looks like a very prom-ising strategy.

For infiltrative gastric malignancies, EUS is also help-ful in assigning a diagnosis and assessing extent of dis-ease. Linitis plastica and lymphoma infiltrate and destroythe gastric wall, and this can be assessed by EUS, thusguiding therapy.37–39 Response to treatment can also beassessed by EUS. Special mention of mucosa-associatedlymphoid tissue lymphoma is warranted because it hasbeen shown that the response to Helicobacter pylori erad-ication can be predicted by EUS determination of infil-tration beyond the submucosa.37 In a small but revealingstudy, 12 of 14 patients with tumor limited to themucosa or submucosa were in complete remission afterH. pylori eradication treatment compared with none of 8patients with invasion of the muscularis propria whoreceived the same treatment.40

EUS has been studied for the evaluation of the gastricwall after treatment for gastric cancer. It is suggestedthat the presence of a thickened gastric wall on follow-upprobably equates to presence of persistent tumor, even inthe setting of negative endoscopic biopsy specimens.41

Using large or jumbo biopsy forceps may help but if EUSshows abnormalities in deeper layers beyond the secondlayer, FNA should be considered.42 However, the assess-ment of tumor recurrence after gastric surgery or che-motherapy remains a very difficult problem.

Occasionally, large gastric folds are noted at endos-copy, but EUS can aid diagnosis if there is concern aboutan infiltrative neoplasm causing this appearance.5 In-volvement of the third or fourth layers of the gastric wallis much more suggestive of malignancy, but FNA underEUS increases the specificity of EUS in this situation13;the sensitivity of EUS FNA for tumors of the gastric wallis about 60%.13,43

Submucosal Tumors

Until the advent of EUS, it was difficult to knowhow best to image submucosal tumors found at endos-copy. EUS is able to distinguish intramural lesions fromextraluminal compression and can also determine whethera lesion is solid, cystic, or vascular. It has also aided thedecision regarding the suitability of a submucosal tumorfor endoscopic mucosal resection. By determining whichof the EUS-visible gastrointestinal tract wall layers areinvolved by the lesion and knowing if a lesion is hypo- orhyperechoic, the nature of a lesion can be more confi-dently determined (Figure 4). For example, in the uppergastrointestinal tract, stromal tumors such as leiomyo-mas are usually homogeneously echopoor and originatein the fourth (or occasionally second) wall layer, whereasfibromas and lipomas are more echorich and originate inthe submucosa.44 However, differentiation of benignfrom malignant submucosal tumors is more difficultusing EUS alone. Certain features are suggestive, but notconfirmatory, of a benign nature. For example, in a seriesof 56 histologically confirmed stromal tumors, it wasfound that small tumors (�3 cm) with regular marginsand homogeneous echo texture were predictive of benignlesions, whereas inhomogeneous lesions with irregularmargins and lymph nodes with malignant patterns sug-gested malignancy.45 In fact, it was suggested that thepresence of 2 of these 3 features gave a positive predictivevalue of 100%.45 The problem occurs when these tumorsdo not have most or all of these features suggestive of abenign or malignant lesion, and this would account forthe great variability in the literature in the accuracy ofpredicted malignancy.38,46,47 However, it is fair to saythat when faced with a submucosal tumor, EUS is help-


ful in distinguishing true submucosal tumors from casesof extraluminal compression. In some cases in which theEUS features suggest a benign nature and surgery iscontraindicated or the patient and physician are reluctantto pursue surgical resection, monitoring any change insize of the lesion by EUS may be the most appropriatestrategy.

EUS FNA has helped somewhat in the differentiationof benign from malignant submucosal lesions but hassignificant limitations. FNA may prove most helpful inidentifying the origin (cell type) of the lesion and maydiagnose malignancy in patients with metastases to thegut wall, but cytopathology cannot differentiate benignfrom malignant stromal tumors. The sensitivity of EUS-guided FNA cytology for submucosal tumors is only60%–64%,13,43 but development of EUS core-biopsyneedles may improve this reliability. If the lesion is smalland confined to the first 3 EUS layers, snare excision aftersaline injection may be indicated.5

Pancreaticobiliary SystemPancreatic Carcinoma

The advent of newer imaging modalities over thelast decade has improved the detection rate for pancreaticcancers, but the ideal diagnostic tool should allow de-tection of small, potentially curable lesions. Some con-sider EUS to be the gold standard in imaging of tumorsof the pancreas, and it appears to be superior to several

other imaging modalities including endoscopic retro-grade cholangiopancreatography (ERCP), angiography,and spiral CT, particularly for small masses less than 2–3cm in diameter (Figure 5).48–50 The prognosis of pancre-atic cancer remains poor, and hence early detection ofsmall resectable masses is probably the most significantrole for EUS. It has been reported in several studies thatEUS has a sensitivity over 95% for imaging pancreatictumors 2 cm or less in diameter.49,51–53 One study ofsmall tumors of the pancreas (�2 cm) revealed that EUSdetected all 25 tumors, CT 19 of 25, ERCP 19 of 25,magnetic resonance imaging (MRI) 10 of 25, and trans-abdominal ultrasound 10 of 25. However, classifying apancreatic lesion as benign or malignant using EUS ismore difficult.52

Once a pancreatic tumor is identified, EUS is veryaccurate for local staging, and most studies confirmsuperiority of EUS over standard CT in this regard.53

EUS predicts T stage and N stage with accuracies ofaround 80% and 70%, respectively.54 Use of dual phasehelical CT has improved the accuracy of CT, and somesuggest it is comparable to EUS.50 When determiningresectability of a pancreatic tumor, it is critical to knowwhether there is invasion of the portal and splenic veinsand invasion or encasement of the celiac axis and superiormesenteric artery. Involvement of the portal system canbe determined on EUS by loss of the sonographic inter-face between vessel and tumor and by irregularity in thelumen of venous structures and is superior comparedwith standard CT, angiography, and transabdominal ul-trasound.54,55 However, EUS is limited in determining

Figure 4. Stromal tumor in the rectum measuring 9 mm � 14 mm(single arrowhead), contiguous with the muscularis propria (doublearrowhead).

Figure 5. A hypoechoic lesion measuring 1.5 cm in the head of thepancreas (single arrowhead). It is seen to invade the CBD (doublearrowhead) and was confirmed at EUS FNA as a pancreatic adenocar-cinoma.


arterial encasement and involvement of the superior mes-enteric vein.54 While performing EUS for evaluation ofpancreatic masses, one can obtain good views of the leftlobe of the liver to look for metastatic spread (Figure 6Aand B), but CT remains superior for imaging of the righthepatic lobe.56

There has been recent suggestion that the staginginformation derived from EUS for pancreatic tumors maybe biased by information available from other modalitiessuch as CT or MRI and other clinical and laboratory data.For example, one group showed sensitivity for stagingtumors (stages T1–3) of the pancreas of 72.2% forblinded EUS and 100% for unblinded EUS when thephysician was aware of the patients’ CT reports.57 Thisand other studies suggest that for imaging of pancreatic

tumors, a combined imaging approach is best using EUS,CT, and MRI findings together.58

Specificity of EUS for pancreatic cancers has beentraditionally lower than the sensitivity. However, EUS-guided FNA is being used increasingly to sample pan-creatic masses, and reported accuracy rates range between85% and 95%.12,14,59 No cases of tumor seeding havebeen reported.5 Having a pathologist present at the timeof FNA should reduce the number of needle passesrequired to make a diagnosis. FNA under EUS of pan-creatic masses is technically more demanding than FNAof lymph nodes because pancreatic masses can often bequite fibrotic, and it may be more difficult to advance theneedle to the desired biopsy site especially if the endo-scope is torqued in the duodenum.5 Hence, operatorexperience is undoubtedly a factor in success.

Neuroendocrine Pancreatic Tumors

EUS has significantly improved the ability tolocalize neuroendocrine tumors (Figure 7). Other imag-ing modalities used to date include CT, MRI, transab-dominal ultrasound, angiography, and somatostatin scin-tigraphy, but irrespective of imaging technique used,many neuroendocrine tumors are missed. In addition, upto 30% of patients with insulinomas or gastrinomas whoundergo surgery fail to have localization of the tumorduring surgery.60

EUS is very well suited for detection of neuroendo-crine tumors, particularly insulinomas, which are typi-cally small, hypoechoic, often solitary lesions, with ahyperechoic rim within the pancreatic parenchyma.61,62

A series of 37 patients who had had nondiagnostic CTFigure 6. EUS appearances of liver metastases. (A) A 5-mm hypo-echoic lesion and (B) a 35-mm � 30-mm hyperechoic lesion. Thesemasses in 2 different patients were both in the left lobe of the liverand EUS FNA confirmed metastatic pancreatic adenocarcinomas.

Figure 7. Pancreatic neuroendocrine tumors. A 6-mm hypoechoicmass (single arrowhead) is seen immediately adjacent to the pancre-atic head (double arrowhead). This was confirmed as a gastrinoma ina patient with known Zollinger-Ellison syndrome.


and transabdominal imaging proceeded to have preoper-ative EUS.63 Twenty-two of these patients also under-went selective angiography. EUS had a sensitivity fortumor detection of 82% compared with 27% for angiog-raphy, and the specificity of EUS was estimated at 95%.This superiority of EUS, particularly for insulinomas, indetection of neuroendocrine tumors has been confirmedin more recent studies.64 Taken alone, the sensitivity ofEUS for gastrinomas is lower (around 60%) because moreof these lesions lie outside of the pancreas than insulino-mas.65 It is likely that management of neuroendocrinetumors will benefit from pretreatment EUS FNA.5

Pancreatic Cysts

Pancreatic cysts can be broadly divided into non-neoplastic cysts and primary neoplastic cysts. The non-neoplastic cysts include pseudocysts (postinflammatory),simple cysts, and duplication cysts.66,67 Neoplastic cysts(approximately 10% of pancreatic cysts) include tumorswith low malignant potential (serous cystadenoma) (Fig-ure 8A and B) and those with high malignant potential(mucinous cystadenoma [Figure 9], mucinous cysta-denocarcinoma [Figure 10], adenocarcinoma with cysticdegeneration, and intraductal papillary mucinous tu-mor).68,69

EUS is useful in differentiation of these lesions, butthere are limitations. Several authors have described EUSfeatures in keeping with benign or malignant processes.For example, it has been suggested that well-defined,simple, uniloculated cysts are probably benign and thatcomplex cystic lesions with thick walls and septations orwith solid lesion protrusion into the cyst lumen are likelymalignant.70 Some series have suggested by using EUSfindings alone that one can predict the nature of apancreatic cyst with greater than 90% accuracy.70,71

However, as highlighted in a recent study, EUS alonecannot be considered the gold standard. In a series of 48patients, EUS could not reliably differentiate benignfrom malignant cystic lesions of the pancreas.67 Al-though some limitations of this study were brought forthin an accompanying editorial72 (such as use of still EUSimages alone and not using pancreatic duct dilation as adistinguishing feature), it nonetheless shows that furtheradvances are required in the diagnosis and managementof pancreatic cystic lesions. One advance in this regard isthe use of EUS-guided FNA. In one study, pancreaticcystic lesions were aspirated under EUS and the aspiratesent for cytologic analysis and quantification of severaltumor markers such as carcinoembryonic antigen andcarbohydrate antigen 19-9.73 Positive cytology or ele-vated tumor markers were 86% accurate for diagnosinga cystic neoplasm. A recent multicenter study showed

that combination of fluid cytology, carcinoembryoniclevels, and EUS features increased the sensitivity of EUSto diagnose malignant cysts to 89%,74 but it should stillbe remembered that a negative FNA does not excludemalignancy. Prophylactic antibiotics should be givenbecause there has been some concern that FNA of cysticlesions may lead to a higher rate of infection than FNAof solid lesions.75,76 Thus, although undoubtedly im-proving the imaging of pancreatic cysts, it would appearthat a combined approach using history, FNA-aspirateanalysis, EUS morphology, and any other imaging infor-mation will give the most reliable results.

Ampullary Tumors

The prognosis of ampullary cancers is better thanfor pancreatic cancers, but there have been fewer studies

Figure 8. EUS appearances of pancreatic cystic lesions. (A and B)Serous cystadenomas in 2 different patients but (A) 1 appears as amultilocular cystic lesion and (B) the other as a unilocular large cyst.Abbreviations: C, cyst; PD, pancreatic duct; HOP, head of pancreas.


looking at EUS imaging of these lesions. One groupfound that EUS was very accurate for T and reasonablyaccurate for N staging of ampullary cancers (over 80%for T staging, 66% for N0, and 75% for N1 staging)77

but commented that EUS is unable to determine whichof the TI tumors are limited to the sphincter of Oddi.This is an obvious disappointment because surgical am-pullectomy for these lesions constitutes a cure. High-frequency (20 MHz) IDUS may allow this distinctionand may lead to a reduction in the need for Whippleresection.78 In a prospective study comparing EUS, CT,and IDUS in the imaging of polypoid major ampullarylesions, IDUS was more sensitive (100% vs. 62.5%) andspecific (75% vs. 50%) than EUS for tumor staging.79

Combining ERCP with catheter probe sonography offersa new diagnostic modality that has some potential ad-vantages for local staging of small tumors of the mainduodenal papilla and also conveniently can be done asone endoscopic procedure.18 The authors of the previ-ously mentioned study79 suggest that IDUS findingsshould serve as the basis for minimally invasive tech-niques (such as endoscopic ampullectomy) for resectionof seemingly benign tumors of the papilla or smallcarcinomas, but although this approach shows promise,it requires validation.

Biliary Tumors

EUS also has a useful role to play in the staging ofextrahepatic bile duct carcinoma.80 It can determinewhether there is invasion of the portal system and/or

pancreas, an important determinant of resectability.81

IDUS may facilitate in the diagnosis and staging of thisand other bile duct pathologies. A comparison of IDUSand EUS to predict resectability in patients with biliaryobstruction showed that T staging was more accuratewith IDUS but N staging inferior.82 Perhaps this isanother example in which a combination approach is bestusing information both from IDUS and EUS. One groupmade recommendations that if IDUS reveals no localizedintraductal lesion, no further investigation is required,but that if there is a lesion with preserved bile ductstructure, tissue should be obtained by ERCP or percu-taneous transhepatic cholangiography.83 If, however, thebile duct wall is interrupted by a protruding tumor, theysuggest that the patient should undergo surgical explo-ration. This approach to biliary lesions is not widelyaccepted, and additional experience is needed with IDUSbefore general recommendations can be made.

Chronic Pancreatitis

Until recently, ERCP has been considered thegold standard investigation in the diagnosis of chronicpancreatitis, but the use of EUS in this setting is grow-ing rapidly and may be overtaking ERCP. However,there are limitations with EUS that need to be taken intoaccount. There is again, as with many EUS studies todate, variability in results between studies in the EUSevaluation of chronic pancreatitis. One group comparedEUS, ERCP, and secretin tests in 80 consecutive patientswith recurrent pancreatitis84 and described good corre-lation between ERCP and EUS for severe and moderatedisease but poor correlation for mild disease. Chronicpancreatitis was diagnosed in 63 patients using EUS and

Figure 9. A mucinous cystadenoma appearing as a small unilocularcyst in the pancreatic body measuring 10 � 14 mm.

Figure 10. An EUS image of a mucinous cystadenocarcinoma withintraluminal ingrowth (arrows).


only 38 using ERCP, and it was reported that EUS hada 100% negative predictive value but a relatively low(60%) positive predictive value. However, there was nohistologic correlation in this study. This and other stud-ies suggested that EUS is more sensitive for detecting theparenchymal changes of chronic pancreatitis before thedevelopment of ductal lesions visible at ERCP and thusmay be better at diagnosing early pancreatitis.85,86 EUSfeatures compatible with chronic pancreatitis are de-scribed in Table 3.84,87

It can be difficult to distinguish chronic pancreatitisfrom pancreatic cancer using EUS, particularly in cases ofsevere pancreatitis, and it is here that caution is needed.62

With advanced inflammatory disease, widespread heter-ogeneous or hypoechoic areas may be mislabeled as can-cers, and studies suggest that the specificity of EUS formaking this distinction is no better than 75%.60,88 Theincreasing use of EUS FNA may improve this situation,89

but even FNA seems to have a lower sensitivity fordiagnosing cancer in patients with underlying calcificpancreatitis compared with patients without underlyingchronic pancreatitis.

Overall, it seems that EUS is good at detecting pa-renchymal changes in chronic pancreatitis and may beparticularly useful in early cases. However, further stud-ies in which histologic confirmation is obtained areneeded to validate these suggestions.

Acute Pancreatitis

As described elsewhere in this review, EUS is veryaccurate for detection of common bile duct (CBD) stonesand is comparable with ERCP for diagnosis of chole-docholithiasis. Recent studies have examined the role ofEUS in acute biliary pancreatitis,90–92 and the resultssuggest that EUS may play a role in determining whichof these patients has choledocholithiasis and would thusbenefit from early ERCP and stone extraction. Whetherall patients with acute pancreatitis of presumed biliaryorigin should undergo early EUS to select which patientsshould have early therapeutic ERCP is a question stillunanswered. Magnetic resonance cholangiopancreatogra-phy (MRCP) offers the advantage of being noninvasive

and gives results comparable with EUS in this settingand may be the most appropriate investigation acutely.Whether EUS will prove useful in the investigation ofacute pancreatitis of nonbiliary cause remains to be de-termined.

EUS is also able to provide information in relation tothe morphology of the pancreas such as echogenicity anddegree of peripancreatic fluid in the setting of acutepancreatitis. This may help in predicting prognosis; onestudy showed that a score based on EUS appearance ofthe pancreas in acute pancreatitis correlated well withnumber of days in hospital and number of days inintensive care.91

Choledocholithiasis

Transabdominal ultrasound is not very sensitiveat detection of biliary tract stones because of interferencefrom bowel gas in the duodenum,18 but EUS overcomesthis problem because the transducer is placed directly inthe duodenal bulb. Several studies have confirmed thatEUS and ERCP have very similar accuracy rates fordetecting CBD stones, most describing rates for bothmodalities over 90%.93,94 Based on these findings and thefact that EUS is less costly than ERCP and has a lesserrisk of pancreatitis, EUS may be preferable to ERCP forpatients with a low and intermediate risk for choledocho-lithiasis, whereas ERCP is the preferred investigation forpatients with a high risk.93,95 EUS has also been com-pared directly with MRCP in the detection of chole-docholithiasis, and one study reported a sensitivity ofboth modalities of 100% but a specificity of 95% forEUS and only 73% for MRCP.94 However, MRCP doesoffer the advantage of being noninvasive. IDUS is alsounder investigation in the diagnosis of choledocholithi-asis. Interestingly, in a study of patients who had had anendoscopic papillary dilation and stone extraction, IDUSdetected small residual stones in 27 of the total 81patients with normal cholangiography.96 The decision asto which imaging modality should be used in patientswith suspected choledocholithiasis still needs to be re-solved.

Gallbladder Imaging

EUS can be used in the staging of carcinoma ofthe gallbladder97 and is very sensitive for pedunculatedcancers in the gallbladder and less so for flat or broad-based cancers.98 However, the clinical role of EUS for thisindication is currently unclear. At the present time, theprimary indication for EUS of the gallbladder is forevaluating the cause of biliary pancreatitis because smallstones or sludge not seen on transabdominal ultrasoundcan often be seen using EUS.

Table 3. EUS Features Compatible WithChronic Pancreatitis

Parenchymal Ductal

Heterogeneous parenchyma Hyperechoic ductal wallsHyperechoic foci Ductal dilationSmall cystic cavities Ductal irregularityProminent interlobular septae Side-branch ectasiaShadowing fociEchogenic strands


Pancreas Divisum and Other AnatomicAnomalies

Although ERCP remains the gold standard fordiagnosing pancreas divisum, EUS can detect pancreasdivisum with a sensitivity of 66% and a specificity of83% using the stack sign (when the main pancreatic ductlies parallel to the distal bile duct in the long endoscopeposition) as indicative of normal pancreatic anatomy.99

EUS can also detect anomalous pancreaticobiliary unionin which the common channel is longer than normal andthe junction between CBD and pancreatic duct outsidethe duodenal wall.100 This condition is associated with asignificantly increased risk of malignancy of the biliarytree.101

Portal Hypertension

Several groups have shown that EUS can detectperiesophageal collateral veins.58 It is reported that de-tection of these veins predicts greater recurrence ofesophageal varices after obliterative therapy102 and alsothat EUS is better at detection of gastric varices thanstandard endoscopy.18 As of yet, the clinical role of EUSremains to be determined, but improved detection ofgastric varices may be helpful, for example, in cirrhoticpatients who have an upper gastrointestinal bleed withno evidence of esophageal varices.

Large BowelRectal Carcinoma

There is a growing body of literature describingthe use of EUS in staging of rectal carcinoma, but thereare many conflicting results and no apparent uniformconsensus. Accurate staging of rectal cancer is importantto guide appropriate resection (Figure 11) because lesionsthat are confined to the mucosa can be resected transa-nally.17 Distal, invasive tumors require an abdominoper-ineal resection, and proximal, invasive tumors can beremoved with a low anterior resection.56 Patients withrectal cancer and lymph node metastasis or invasionthrough the muscularis propria generally get adjuvant orneoadjuvant chemotherapy and radiotherapy.

One study suggests that EUS staging of rectal canceris similar in sensitivity to that of other luminal can-cers.103 However, this view is not held by all, and cer-tainly there are problems in T staging such as tumor-associated inflammation and microscopic spread and in Nstaging such as small malignant nodes or large benignnodes.38,104 Obviously, these problems lead to under- oroverstaging of cancers and affect the accuracy of staging,but, overall, it seems that EUS is reliable preoperativelyfor T staging (accuracy rates between 73% and 94%) but

less so for N staging (around 70% at best).105 Miniprobesmay have better accuracy rates for T and N staging ofrectal cancer.106

Restaging of rectal cancer by EUS after therapy hasbeen reported to be inaccurate in one study.107 However,others report that EUS is a valuable tool for early detec-tion of recurrence of rectal cancer.108 The role of EUS inthis situation may be most useful in performing FNA oflymph nodes or soft-tissue masses in patients suspectedof having recurrence.

Inflammatory Bowel Disease

Bowel wall thickening, especially small bowel,can be detected by standard ultrasound, and superiormesenteric artery flow, increased during active disease,may be measurable by Doppler studies.109,110 However,the reproducibility and clinical applicability of theseparameters are not certain.110 Some studies have evalu-ated the use of EUS in assessment of inflammatory boweldisease111,112 and describe certain features such as in-creased bowel wall thickness, lymphadenopathy, and en-larged perirectal vessels. EUS may also be useful inevaluating possible abscesses and fistulae in the perirectalregion. In addition, EUS may be able to distinguishulcerative colitis from Crohn’s disease based on whichbowel wall layers are involved.113

Mediastinal MassesThere is growing interest and experience in the

use of EUS FNA for the evaluation of mediastinal masses

Figure 11. EUS image of a TI rectal tumor (single arrowhead), whichis superficial to the muscularis propria (double arrowhead).


(Figure 12A and B). The middle and posterior medias-tinum are inaccessible to percutaneous ultrasound,114

and, traditionally, lesions in these locations have beenimaged and biopsies performed using CT, MRI, bron-choscopy and transbronchial biopsy, and mediastinos-copy/mediastinotomy.115,116 EUS staging of lung cancerhas been shown to be safe and accurate in comparison toother modalities as well as being cost-effective.117–119 Forpatients with lung cancer, the cost per year of expectedsurvival was $1729 with an EUS strategy and $2411with a mediastinoscopy/mediastinotomy strategy.119 Onestudy describes detection of malignant adenopathy byEUS in 84% of cases (96% with EUS FNA) comparedwith 49% for CT.118 Because the detection of contralat-eral or subcarinal adenopathy in cases of non–small cellcarcinoma of the lung essentially precludes resectability,EUS FNA should obviate the need for unnecessary sur-

gery.120 In addition, EUS FNA of nodes in the medias-tinum has a role to play in the diagnosis of other con-ditions such as sarcoidosis121 or metastatic tumors. Itshould be noted that mediastinal EUS has not yet be-come widely accepted as a staging modality for lungcancer among pulmonologists and thoracic surgeons. Ifand when it receives this recognition, it remains to beseen who will adopt this tool as part of their service.

Interventional EUSThe number of therapeutic procedures being at-

tempted under EUS is rapidly growing, and the intro-duction of EUS-guided FNA systems has led to manyadvances in the field of intervention. Following is areview of some of the procedures performed under EUS,some of which are still essentially under development(Table 4).

Celiac Plexus Blocks

Patients with inoperable pancreatic cancer oftenhave significant abdominal pain that can be difficult tocontrol.122,123 Injection of bupivicaine and alcohol intothe celiac ganglia has led to a marked reduction in painand decreased the need for narcotics in up to 88% ofpatients.122 There is also a suggestion that this techniquemay be safer than CT-guided celiac neurolysis, whichuses a posterior approach and has very rarely been com-plicated by paraplegia.122,124 The role of EUS-guidedceliac plexus blocks in chronic pancreatitis warrants fur-ther study, although early data are not encouraging.125

Pseudocyst Drainage

EUS has augmented the endoscopic managementof pancreatic pseudocysts.126–128 Endoscopic cystogastro-stomy was often done in a blind fashion, and this mayhave increased the risk of complications, especially ifthere was not a significant intragastric bulge.129 EUS canlocate a puncture site devoid of vessels and less than 1 cm

Figure 12. Lung cancer staging. A mediastinal lymph node is seen on(A) MRI (arrow) and also on (B) EUS (arrow). This was a patient withnon–small cell cancer and recurrent nodal disease.

Table 4. Present and Future Therapeutic Applicationsof EUS

Celiac plexus blocksFNI for achalasiaEUS directed endoscopic resectionFNI of tumors with immunotherapeutic, chemotherapeutic, gene

therapy, or radionuclide agentsEUS-directed cholangiographyPancreatic pseudocyst drainageRadiofrequency ablation of tumorsEstablishment of hepatico-luminal anastomoses for inoperable

biliary obstructionTransrectal abscess drainage

FNI, fine needle injection.


between the gastric and cyst lumens, and one can markthis site by a double biopsy.130 In a series of 32 pa-tients,127 preinterventional EUS provided essential infor-mation that resulted in a major change in therapeuticmanagement in one third of patients. Some argue thatendoscopic drainage should not be done without priorendosonographic examination. With the advent of largerechoendoscopes, the entire procedure, including stentplacement, can be performed under EUS guidance.131

Fine Needle Injection for Achalasia

One of the treatments used for symptom relief inachalasia is the injection of botulinum toxin into thelower esophageal sphincter. However, although initialresults are good, there is a high relapse rate necessitatingfrequent repeat injections.132 It has been suggested thatthe often short-lived effect of toxin injection is because ofincomplete delivery of the toxin to the muscularis pro-pria layer of the lower esophageal sphincter. Accordingly,EUS-guided injection has been attempted to try to im-prove this situation. Toxin is delivered to regions of focalthickening in the muscularis propria at the gastroesoph-ageal junction. Initial data are encouraging and suggestthat the relapse rate may be lower using an EUS ap-proach, but results of larger trials are awaited.133

EUS-Guided Sclerotherapy

There are limited data suggesting that the re-bleeding rate with EUS-guided sclerotherapy may belower than standard band ligation, but no recommenda-tions can be made at this time.134 It seems unlikely thatEUS will oust standard endoscopic therapy for varices.

EUS Fine Needle Injection of Tumors WithTherapeutic Agents

A particularly exciting area of ongoing researchis the EUS-guided delivery of a variety of therapeuticagents to tumors. The development of EUS FNA allowedthis possibility, and it is still a very young discipline butpreliminary results are encouraging. In 1 study of pan-creatic cancer, a local cytoimplant (containing activatedT lymphocytes) was placed by using EUS guidance.135

Longer-term results are awaited with interest.

EUS-Directed Cholangiopancreatography

ERCP is not always successful in accessing thepancreaticobiliary tree for a variety of reasons includinganatomic problems such as with Roux-en-Y reconstruc-tions or luminal obstruction by tumor. EUS is able tovisualize the pancreatic and biliary ducts in almost allsituations. In some such circumstances, EUS-guidedtransgastrointestinal injection of contrast material into

the biliary tree or pancreatic duct can be performed.136,137

Again, there are limited data in this regard, but thistechnique is unlikely to be used often because it does notallow for other therapeutic measures such as stone ex-traction and stent placement, and MRCP is improvingall the time as a noninvasive modality for imaging of thebiliary system.5

EUS-Directed or Assisted Resection

Endoscopic resection (ER) of mucosal and submu-cosal lesions has gained a lot of publicity in recent years.However, some safety fears have been expressed, and theissue of completeness of resection of a lesion has also beencalled into question. Guiding the submucosal injectionof saline using EUS may help but generally will probablybe unnecessary.15 It would appear from data so far thatthe greatest help offered by EUS for mucosal resection isin the selection of appropriate patients for ER. Oneretrospective study on early gastric cancers showed asensitivity of 93% and specificity of 86% regarding theassessment of patients for ER.138 The patients were clas-sified as unsuitable for endoscopic mucosal resection ofearly gastric cancer if EUS showed involvement of thethird layer and as suitable if such involvement was notseen. If a submucosal lesion is present, ER may bepossible if EUS confirms that the lesion is superficial tothe muscularis propria. EUS may also be useful after ERto assess lesion removal.

EUS-Guided Paracentesis

Small amounts of ascites may be missed on stan-dard ultrasound or CT. If there is a question of under-lying malignancy, EUS-guided FNA may aid in diagno-sis.5,139

Future DevelopmentsThere are many exciting developments in the field

of interventional EUS. EUS-guided fine needle injectionof tumors with immune modulators is under study asdescribed earlier.135 It is likely with advances in tumortherapy and echoendoscope technology that EUS fineneedle injection may also deliver chemotherapeuticagents, gene therapy, and radionuclide preparations di-rectly into tumors.5 Delivery of radionuclide agents byEUS would require a shielded delivery system.15

Another exciting development is that of EUS-guidedradiofrequency ablation of tumors using high-intensityultrasound probes. This has been tried in a porcinepancreatic model and produced discrete areas of coagu-lation necrosis in the pancreas (Figure 13),140 but furtheranimal and human data are needed. This may prove


especially useful in treating cystic lesions of the pancreas.Another group has investigated the use of microwavecoagulation therapy during laparoscopic ultrasound fortreatment of small liver tumors.141 This could potentiallybe applied to EUS systems if results proved success of thetechnique.

Work is ongoing looking at the use of contrast agentsthat may aid echoendosonography. Injection of suspen-sions of galactose microparticles improved color Dopplersignals in assessment of malignant vascular invasion inpancreatic cancer.142 Others have described favorablefindings with the use of IV sonicates of albumin to betterdemarcate gut wall tumors.143

There has been recent excitement at the use of ultra-sound virtual endoscopic imaging in which 3-dimen-sional images can be displayed in a short time duringultrasound examinations.144 Similar computer processingmethods are used for CT and MRI multiplanar imagingand 3-dimensional reformatting. In a series of patientswith pancreaticobiliary disease, use of 3-dimensional

IDUS allowed accurate definition of the surroundingvasculature and assessment of the relationship of tumorto other organs.145 Other potential invasive applica-tions of EUS include transrectal drainage of perirectalabscesses5 and the establishment of anastomoses betweenthe liver and stomach in cases of inoperable malignantobstruction of the biliary tract.146

SummaryEUS is now firmly established as the investigation

of choice in the locoregional staging of several gastroin-testinal tumors and submucosal masses. In some situa-tions, it is recommended that combination of informa-tion from other imaging modalities with EUS findings isneeded to maximize accuracy. However, the increasinguse of EUS-guided FNA should improve the accuracy ofEUS as a stand-alone investigation.

Other advances have been made in the diagnosis ofchronic pancreatitis with EUS, especially patients withearly changes. EUS may also emerge as the investigationof choice in patients with biliary pancreatitis of low tomoderate risk of persistent choledocholithiasis. EUSstaging of lung cancer is very accurate, and undoubtedlyEUS will continue to be used more and more for thisindication.

With the increasing use of EUS FNA, other potentialtherapeutic applications of EUS were developed, andsome of these show great promise. As technical advancesare made with scope design, accessory devices, andprobes, it is likely that many of these potential thera-peutic applications will become routine procedures. Thediscipline of EUS intervention is a young one, and othernew therapeutic advances are a certainty.

EUS has matured over the last few years as a pivotalinvestigation in many disorders of the gastrointestinaltract and mediastinum. Recognition of this has finallyled to a wider availability of EUS in clinical practice, andundoubtedly EUS will be regarded over the next fewyears as a standard investigation rather than one ofintellectual curiosity.

References1. Role of endoscopic ultrasonography. American Society for Gas-

trointestinal Endoscopy. Gastrointest Endosc 2000;52:852–859.

2. Mallery S, Van Dam J. Interventional endoscopic ultrasonogra-phy: current status and future direction. J Clin Gastroenterol1999;29:297–305.

3. Brugge WR. Endoscopic ultrasonography: the current status.Gastroenterology 1998;115:1577–1583.

4. Bhutani MS. “Probing” the endoscopic ultrasound (EUS) cathe-ter probe: a small step for EUS or a giant leap? GastrointestEndosc 1998;48:542–545.

Figure 13. Radiofrequency ablation. This image shows a gross sec-tion of the pancreas after RF ablation. Note the sharp border. (Cour-tesy of W. Brugge.)


5. Bhutani MS. Interventional endoscopic ultrasonography: stateof the art at the new millenium. Endoscopy 2000;32:62–71.

6. Kimmey MB, Martin RW, Haggitt RC, Wang KY, Franklin DW,Silverstein FE. Histologic correlates of gastrointestinal ultra-sound images. Gastroenterology 1989;96:433–441.

7. Berge J, Hawes R, Hoffman B, Van Enckevort C, Giovannini M,Erickson R, Catalano M, Fogel R, Mallery S, Faigel D, Wallace M.EUS guided fine needle aspiration of the liver. Indications, yield,and safety from an International survey of 167 cases (abstr).Gastrointest Endosc 2001;53:AB176.

8. Catalano MF, Sivak MV, Jr., Rice T, Gragg LA, Van Dam J.Endosonographic features predictive of lymph node metastasis.Gastrointest Endosc 1994;40:442–446.

9. Akahoshi K, Misawa T, Fujishima H, Chijiiwa Y, Nawata H. Re-gional lymph node metastasis in gastric cancer: evaluation withendoscopic US. Radiology 1992;182:559–564.

10. Bhutani MS, Hawes RH, Hoffman BJ. A comparison of theaccuracy of echo features during endoscopic ultrasound (EUS)and EUS-guided fine-needle aspiration for diagnosis of malig-nant lymph node invasion. Gastrointest Endosc 1997;45:474–479.

11. Antillon MR, Chang KJ. Endoscopic and endosonography guidedfine-needle aspiration. Gastrointest Endosc Clin N Am 2000;10:619–636.

12. Gress FG, Hawes RH, Savides TJ, Ikenberry SO, Lehman GA.Endoscopic ultrasound-guided fine-needle aspiration biopsy us-ing linear array and radial scanning endosonography. Gastroin-test Endosc 1997;45:243–250.

13. Giovannini M, Seitz JF, Monges G, Perrier H, Rabbia I. Fine-needle aspiration cytology guided by endoscopic ultrasonogra-phy: results in 141 patients. Endoscopy 1995;27:171–177.

14. Wiersema MJ, Vilmann P, Giovannini M, Chang KJ, WiersemaLM. Endosonography-guided fine-needle aspiration biopsy: diag-nostic accuracy and complication assessment. Gastroenterol-ogy 1997;112:1087–1095.

15. Mortensen MB. The role of gastrointestinal endosonography indiagnostic and therapeutic interventional procedures. Eur J Ul-trasound 1999;10:93–104.

16. Williams DB, Sahai AV, Aabakken L, Penman ID, van Velse A,Webb J, Wilson M, Hoffman BJ, Hawes RH. Endoscopic ultra-sound guided fine needle aspiration biopsy: a large single cen-tre experience. Gut 1999;44:720–726.

17. Hawes RH. Endoscopic ultrasound. Gastrointest Endosc Clin NAm 2000;10:161–174.

18. Chak A. Endoscopic ultrasonography. Endoscopy 2000;32:146–152.

19. Botet JF, Lightdale CJ, Zauber AG, Gerdes H, Urmacher C,Brennan MF. Preoperative staging of esophageal cancer: com-parison of endoscopic US and dynamic CT. Radiology 1991;181:419–425.

20. Ziegler K, Sanft C, Zeitz M, Friedrich M, Stein H, Haring R,Riecken EO. Evaluation of endosonography in TN staging ofoesophageal cancer. Gut 1991;32:16–320.

21. Chak A, Canto M, Gerdes H, Lightdale CJ, Hawes RH, WiersemaMJ, Kallimanis G, Tio TL, Rice TW, Boyce HW Jr., Sivak MV Jr.Prognosis of esophageal cancers preoperatively staged to belocally invasive (T4) by endoscopic ultrasound (EUS): a multi-center retrospective cohort study. Gastrointest Endosc 1995;42:501–506.

22. Brugge WR, Lee MJ, Carey RW, Mathisen DJ. Endoscopic ultra-sound staging criteria for esophageal cancer. Gastrointest En-dosc 1997;45:147–152.

23. Hoffman BJ, Hawes RH. Endoscopic ultrasonography-guidedpuncture of the lymph nodes: first experience and clinical con-sequences. Gastrointest Endosc Clin N Am 1995;5:587–593.

24. Giovannini M, Monges G, Seitz JF, Moutardier V, Bernardini D,Thomas P, Houvenaeghel G, Delpero JR, Giudicelli R, Fuentes P.

Distant lymph node metastases in esophageal cancer: impactof endoscopic ultrasound-guided biopsy. Endoscopy 1999;31:536–540.

25. Van Dam J. Endosonography of the esophagus. GastrointestEndosc Clin N Am 1994;4:803–826.

26. Mallery S, DeCamp M, Bueno R, Mentzer SJ, Sugarbaker DJ,Swanson SJ, Van Dam J. Pretreatment staging by endoscopicultrasonography does not predict complete response to neoad-juvant chemoradiation in patients with esophageal carcinoma.Cancer 1999;86:764–769.

27. Zuccaro G, Jr., Rice TW, Goldblum J, Medendorp SV, Becker M,Pimentel R, Gitlin L, Adelstein DJ. Endoscopic ultrasound cannotdetermine suitability for esophagectomy after aggressive che-moradiotherapy for esophageal cancer. Am J Gastroenterol1999;94:906–912.

28. Isenberg G, Chak A, Canto MI, Levitan N, Clayman J, Pollack BJ,Sivak MV, Jr. Endoscopic ultrasound in restaging of esophagealcancer after neoadjuvant chemoradiation. Gastrointest Endosc1998;48:158–163.

29. Catalano MF, Van Dam J, Sivak MV, Jr. Malignant esophagealstrictures: staging accuracy of endoscopic ultrasonography.Gastrointest Endosc 1995;41:535–539.

30. Van Dam J, Rice TW, Catalano MF, Kirby T, Sivak MV Jr. High-grade malignant stricture is predictive of esophageal tumorstage. Risks of endosonographic evaluation. Cancer 1993;71:2910–2917.

31. Kallimanis GE, Gupta PK, al-Kawas FH, Tio LT, Benjamin SB,Bertagnolli ME, Nguyen CC, Gomes MN, Fleischer DE. Endo-scopic ultrasound for staging esophageal cancer, with or with-out dilation, is clinically important and safe. Gastrointest En-dosc 1995;41:540–546.

32. Wallace MB, Hoffman BJ, Sahai AS, Inoue H, Van Velse A,Hawes RH. Imaging of esophageal tumors with a water-filledcondom and a catheter US probe. Gastrointest Endosc 2000;51:597–600.

33. Ziegler K, Sanft C, Zimmer T, Zeitz M, Felsenberg D, Stein H,Germer C, Deutschmann C, Riecken EO. Comparison of com-puted tomography, endosonography, and intraoperative assess-ment in TN staging of gastric carcinoma. Gut 1993;34:604–610.

34. Caletti G, Ferrari A, Brocchi E, Barbara L. Accuracy of endo-scopic ultrasonography in the diagnosis and staging of gastriccancer and lymphoma. Surgery 1993;113:14–27.

35. Grimm H, Binmoeller KF, Hamper K, Koch J, Henne-Bruns D,Soehendra N. Endosonography for preoperative locoregionalstaging of esophageal and gastric cancer. Endoscopy 1993;25:224–230.

36. Ohashi S, Segawa K, Okamura S, Mitake M, Urano H, Shimo-daira M, Takeda T, Kanamori S, Naito T, Takeda K, Itoh B, GotoH, Niwa Y, Hayakawa T. The utility of endoscopic ultrasonogra-phy and endoscopy in the endoscopic mucosal resection ofearly gastric cancer. Gut 1999;45:599–604.

37. Palazzo L, Roseau G, Ruskone-Fourmestraux A, Rougier P,Chaussade S, Rambaud JC, Couturier D, Paolaggi JA. Endo-scopic ultrasonography in the local staging of primary gastriclymphoma. Endoscopy 1993;25:502–508.

38. Caletti GC, Lorena Z, Bolondi L, Guizzardi G, Brocchi E, BarbaraL. Impact of endoscopic ultrasonography on diagnosis and treat-ment of primary gastric lymphoma. Surgery 1988;103:315–320.

39. Tio TL, den Hartog Jager FC, Tijtgat GN. Endoscopic ultrasonog-raphy of non-Hodgkin lymphoma of the stomach. Gastroenterol-ogy 1986;91:401–408.

40. Sackmann M, Morgner A, Rudolph B, Neubauer A, Thiede C,Schulz H, Kraemer W, Boersch G, Rohde P, Seifert E, Stolte M,Bayerdoerffer E. Regression of gastric MALT lymphoma aftereradication of Helicobacter pylori is predicted by endosono-


graphic staging. MALT Lymphoma Study Group. Gastroenterol-ogy 1997;113:1087–1090.

41. Levy M, Hammel P, Lamarque D, Marty O, Chaumette MT,Haioun C, Blazquez M, Delchier JC. Endoscopic ultrasonographyfor the initial staging and follow-up in patients with low-gradegastric lymphoma of mucosa-associated lymphoid tissuetreated medically. Gastrointest Endosc 1997;46:328–333.

42. Caletti G, Fusaroli P, Bocus P. Endoscopic ultrasonography inlarge gastric folds. Endoscopy 1998;30(suppl 1):A72–A75.

43. Chang KJ. Endoscopic ultrasound (EUS)-guided fine needle as-piration (FNA) in the USA. Endoscopy 1998;30(suppl 1):A159–A160.

44. Rosch T. Endoscopic ultrasonography. Endoscopy 1992;24:144–153.

45. Palazzo L, Landi B, Cellier C, Cuillerier E, Roseau G, Barbier JP.Endosonographic features predictive of benign and malignantgastrointestinal stromal cell tumours. Gut 2000;46:88–92.

46. Chak A, Canto MI, Rosch T, Dittler HJ, Hawes RH, Tio TL,Lightdale CJ, Boyce HW, Scheiman J, Carpenter SL, Van Dam J,Kochman ML, Sivak MV, Jr. Endosonographic differentiation ofbenign and malignant stromal cell tumors. Gastrointest Endosc1997;45:468–473.

47. Rosch T. Endoscopic ultrasonography in upper gastrointestinalsubmucosal tumors: a literature review. Gastrointest EndoscClin N Am 1995;5:609–614.

48. Muller MF, Meyenberger C, Bertschinger P, Schaer R, MarincekB. Pancreatic tumors: evaluation with endoscopic US, CT, andMR imaging. Radiology 1994;190:745–751.

49. Palazzo L, Roseau G, Gayet B, Vilgrain V, Belghiti J, Fekete F,Paolaggi JA. Endoscopic ultrasonography in the diagnosis andstaging of pancreatic adenocarcinoma. Results of a prospectivestudy with comparison to ultrasonography and CT scan. Endos-copy 1993;25:143–150.

50. Legmann P, Vignaux O, Dousset B, Baraza AJ, Palazzo L, Du-montier I, Coste J, Louvel A, Roseau G, Couturier D, Bonnin A.Pancreatic tumors: comparison of dual-phase helical CT andendoscopic sonography. AJR Am J Roentgenol 1998;170:1315–1322.

51. Ariyama J, Suyama M, Satoh K, Wakabayashi K. Endoscopicultrasound and intraductal ultrasound in the diagnosis of smallpancreatic tumors. Abdom Imaging 1998;23:380–386.

52. Rosch T, Lorenz R, Braig C, Feuerbach S, Siewert JR, Schusd-ziarra V, Classen M. Endoscopic ultrasound in pancreatic tumordiagnosis. Gastrointest Endosc 1991;37:347–352.

53. Yasuda K, Mukai H, Nakajima M. Endoscopic ultrasonographydiagnosis of pancreatic cancer. Gastrointest Endosc Clin N Am1995;5:699–712.

54. Rosch T. Staging of pancreatic cancer. Analysis of literatureresults. Gastrointest Endosc Clin N Am 1995;5:735–739.

55. Brugge WR, Lee MJ, Kelsey PB, Schapiro RH, Warshaw AL. Theuse of EUS to diagnose malignant portal venous system inva-sion by pancreatic cancer. Gastrointest Endosc 1996;43:561–567.

56. Mallery S, Van Dam J. Current status of diagnostic and thera-peutic endoscopic ultrasonography. Radiol Clin North Am 2001;39:449–463.

57. Rosch T, Dittler HJ, Strobel K, Meining A, Schusdziarra V, LorenzR, Allescher HD, Kassem AM, Gerhardt P, Siewert JR, Hofler H,Classen M. Endoscopic ultrasound criteria for vascular invasionin the staging of cancer of the head of the pancreas: a blindreevaluation of videotapes. Gastrointest Endosc 2000;52:469–477.

58. Bhutani MS. Endoscopic ultrasonography. Endoscopy 2000;32:853–862.

59. Chang KJ, Nguyen P, Erickson RA, Durbin TE, Katz KD. Theclinical utility of endoscopic ultrasound-guided fine-needle aspi-

ration in the diagnosis and staging of pancreatic carcinoma.Gastrointest Endosc 1997;45:387–393.

60. Rosch T, Braig C, Gain T, Feuerbach S, Siewert JR, SchusdziarraV, Classen M. Staging of pancreatic and ampullary carcinoma byendoscopic ultrasonography. Comparison with conventionalsonography, computed tomography, and angiography. Gastroen-terology 1992;102:188–199.

61. Glover JR, Shorvon PJ, Lees WR. Endoscopic ultrasound forlocalisation of islet cell tumours. Gut 1992;33:108–110.

62. Tenner SM, Banks PA, Wiersema MJ, Van Dam J. Evaluation ofpancreatic disease by endoscopic ultrasonography. Am J Gas-troenterol 1997;92:18–26.

63. Rosch T, Lightdale CJ, Botet JF, Boyce GA, Sivak MV Jr., YasudaK, Heyder N, Palazzo L, Dancygier H, Schusdziarra V, Classen M.Localization of pancreatic endocrine tumors by endoscopic ul-trasonography. N Engl J Med 1992;326:1721–1726.

64. Zimmer T, Ziegler K, Bader M, Fett U, Hamm B, Riecken EO,Wiedenmann B. Localisation of neuroendocrine tumours of theupper gastrointestinal tract. Gut 1994;35:471–475.

65. Ruszniewski P, Amouyal P, Amouyal G, Grange JD, Mignon M,Bouche O, Bernades P. Localization of gastrinomas by endo-scopic ultrasonography in patients with Zollinger-Ellison syn-drome. Surgery 1995;117:629–635.

66. Fernandez-del Castillo C, Warshaw AL. Cystic tumors of thepancreas. Surg Clin North Am 1995;75:1001–1016.

67. Ahmad NA, Kochman ML, Lewis JD, Ginsberg GG. Can EUSalone differentiate between malignant and benign cystic lesionsof the pancreas? Am J Gastroenterol 2001;96:3295–300.

68. Meyer W, Kohler J, Gebhardt C. Cystic neoplasms of the pan-creas—cystadenomas and cystadenocarcinomas. Langen-becks Arch Surg 1999;384:44–49.

69. Cubilla AL, Fitzgerald PJ. Cancer of the exocrine pancreas: thepathologic aspects. CA Cancer J Clin 1985;35:2–18.

70. Koito K, Namieno T, Nagakawa T, Shyonai T, Hirokawa N, MoritaK. Solitary cystic tumor of the pancreas: EUS-pathologic corre-lation. Gastrointest Endosc 1997;45:268–276.

71. Maguchi H, Osanai M, Yanagawa N, Takahashi K, Itoh H, Ka-tanuma A, Obara T, Kohgo Y. Endoscopic ultrasonography diag-nosis of pancreatic cystic disease. Endoscopy 1998;30(suppl1):A108–A110.

72. Hernandez LC, Bhutani MS. Endoscopic ultrasound and pancre-atic cysts: a sticky situation! Am J Gastroenterol 2001;96:3229–230.

73. Hammel P, Levy P, Voitot H, Levy M, Vilgrain V, Zins M, Flejou JF,Molas G, Ruszniewski P, Bernades P. Preoperative cyst fluidanalysis is useful for the differential diagnosis of cystic lesionsof the pancreas. Gastroenterology 1995;108:1230–1235.

74. Brugge WR, Saltzman JR, Scheiman JM, Wallace MB, Jowell PS,Pochapin MB, Amitabh C, Stevens PD, Gress FG, LewandrowskiK. Diagnosis of cystic neoplasms of the pancreas by EUS: Thereport of the cooperative pancreatic cyst study. GastrointestEndosc 2001;53:AB71.

75. Kochman ML, Wiersema MJ, Hawes RH, Canal D, Wiersema L.Preoperative diagnosis of cystic lymphangioma of the colonby endoscopic ultrasound. Gastrointest Endosc 1997;45:204–206.

76. Bhutani MS. Endoscopic ultrasound in pancreatic diseases.Indications, limitations, and the future. Gastroenterol Clin NorthAm 1999;28:747–770.

77. Palazzo L. Staging of ampullary carcinoma by endoscopic ultra-sonography. Endoscopy 1998;30(suppl 1):A128–A131.

78. Itoh A, Goto H, Naitoh Y, Hirooka Y, Furukawa T, Hayakawa T.Intraductal ultrasonography in diagnosing tumor extension ofcancer of the papilla of Vater. Gastrointest Endosc 1997;45:251–260.

79. Menzel J, Hoepffner N, Sulkowski U, Reimer P, Heinecke A,Poremba C, Domschke W. Polypoid tumors of the major duode-


nal papilla: preoperative staging with intraductal US, EUS, andCT—a prospective, histopathologically controlled study. Gastroi-ntest Endosc 1999;49:349–357.

80. Mukai H, Yasuda K, Nakajima M. Tumors of the papilla anddistal common bile duct. Diagnosis and staging by endoscopicultrasonography. Gastrointest Endosc Clin N Am 1995;5:763–772.

81. Nakazawa S. Recent advances in endoscopic ultrasonography. JGastroenterol 2000;35:257–260.

82. Menzel J, Poremba C, Dietl KH, Domschke W. Preoperativediagnosis of bile duct strictures—comparison of intraductalultrasonography with conventional endosonography. Scand JGastroenterol 2000;35:77–82.

83. Tamada K, Ueno N, Tomiyama T, Oohashi A, Wada S, NishizonoT, Tano S, Aizawa T, Ido K, Kimura K. Characterization of biliarystrictures using intraductal ultrasonography: comparison withpercutaneous cholangioscopic biopsy. Gastrointest Endosc1998;47:341–349.

84. Catalano MF, Lahoti S, Geenen JE, Hogan WJ. Prospectiveevaluation of endoscopic ultrasonography, endoscopic retro-grade pancreatography, and secretin test in the diagnosis ofchronic pancreatitis. Gastrointest Endosc 1998;48:11–17.

85. Deviere J, Finet L, Dunham F, Cremer M. Endoscopic ultrasonog-raphy in chronic pancreatitis. Endoscopy 1994;26:808–809.

86. Zuccaro G Jr, Sivak MV Jr. Endoscopic ultrasonography in thediagnosis of chronic pancreatitis. Endoscopy 1992;24(suppl1):347–349.

87. Sahai AV, Zimmerman M, Aabakken L, Tarnasky PR, Cunning-ham JT, van Velse A, Hawes RH, Hoffman BJ. Prospectiveassessment of the ability of endoscopic ultrasound to diagnose,exclude, or establish the severity of chronic pancreatitis foundby endoscopic retrograde cholangiopancreatography. Gastroin-test Endosc 1998;48:18–25.

88. Kaufman AR, Sivak MV Jr. Endoscopic ultrasonography in thedifferential diagnosis of pancreatic disease. Gastrointest En-dosc 1989;35:214–219.

89. Vilmann P, Jacobsen GK, Henriksen FW, Hancke S. Endoscopicultrasonography with guided fine needle aspiration biopsy inpancreatic disease. Gastrointest Endosc 1992;38:172–173.

90. Sugiyama M, Atomi Y. Acute biliary pancreatitis: the roles ofendoscopic ultrasonography and endoscopic retrograde cholan-giopancreatography. Surgery 1998;124:14–21.

91. Chak A, Hawes RH, Cooper GS, Hoffman B, Catalano MF, WongRC, Herbener TE, Sivak MV Jr. Prospective assessment of theutility of EUS in the evaluation of gallstone pancreatitis. Gas-trointest Endosc 1999;49:599–604.

92. Tandon M, Topazian M. Endoscopic ultrasound in idiopathicacute pancreatitis. Am J Gastroenterol 2001;96:705–709.

93. Canto MI, Chak A, Stellato T, Sivak MV Jr. Endoscopic ultra-sonography versus cholangiography for the diagnosis of chole-docholithiasis. Gastrointest Endosc 1998;47:439–448.

94. de Ledinghen V, Lecesne R, Raymond JM, Gense V, AmourettiM, Drouillard J, Couzigou P, Silvain C. Diagnosis of choledocho-lithiasis: EUS or magnetic resonance cholangiography? A pro-spective controlled study. Gastrointest Endosc 1999;49:26–31.

95. Sahai AV, Mauldin PD, Marsi V, Hawes RH, Hoffman BJ. Bileduct stones and laparoscopic cholecystectomy: a decision anal-ysis to assess the roles of intraoperative cholangiography, EUS,and ERCP. Gastrointest Endosc 1999;49:334–343.

96. Ohashi A, Ueno N, Tamada K, Tomiyama T, Wada S, Miyata T,Nishizono T, Tano S, Aizawa T, Ido K, Kimura K. Assessment ofresidual bile duct stones with use of intraductal US duringendoscopic balloon sphincteroplasty: comparison with ballooncholangiography. Gastrointest Endosc 1999;49:328–333.

97. Mitake M, Nakazawa S, Naitoh Y, Kimoto E, Tsukamoto Y, AsaiT, Yamao K, Inui K, Morita K, Hayashi Y. Endoscopic ultrasonog-

raphy in diagnosis of the extent of gallbladder carcinoma. Gas-trointest Endosc 1990;36:562–566.

98. Inui K, Nakazawa S. Diagnosis of depth of invasion of gallblad-der carcinoma with endosonography. Nippon Geka GakkaiZasshi 1998;99:696–699.

99. Bhutani MS, Hoffman BJ, Hawes RH. Diagnosis of pancreasdivisum by endoscopic ultrasonography. Endoscopy 1999;31:167–169.

100. Mitake M, Nakazawa S, Naitoh Y, Kimoto E, Tsukamoto Y,Yamao K, Inui K. Value of endoscopic ultrasonography in thedetection of anomalous connections of the pancreatobiliaryduct. Endoscopy 1991;23:117–120.

101. Kuga H, Yamaguchi K, Shimizu S, Yokohata K, Chijiiwa K,Tanaka M. Carcinoma of the pancreas associated with anoma-lous junction of pancreaticobiliary tracts: report of two casesand review of the literature. J Hepatobiliary Pancreat Surg 1998;5:113–116.

102. De Angelis C, Carucci P, Maglione V. Clinical application ofendoscopic ultrasound in portal hypertension: EUS can predictvariceal recurrence in endoscopically treated patients. Gastroi-ntest Endosc 2000;51:AB160.

103. Glaser F, Schlag P, Herfarth C. Endorectal ultrasonography forthe assessment of invasion of rectal tumours and lymph nodeinvolvement. Br J Surg 1990;77:883–887.

104. McClave SA, Jones WF, Woolfolk GM, Schrodt GR, WiersemaMJ. Mistakes on EUS staging of colorectal carcinoma: error ininterpretation or deception from innate pathologic features?Gastrointest Endosc 2000;51:682–689.

105. Dershaw DD, Enker WE, Cohen AM, Sigurdson ER. Transrectalultrasonography of rectal carcinoma. Cancer 1990;66:2336–2340.

106. Hamada S, Akahoshi K, Chijiiwa Y, Sasaki I, Nawata H. Preop-erative staging of colorectal cancer by a 15 MHz ultrasoundminiprobe. Surgery 1998;123:264–269.

107. Lin D, Vanagunas A, Stryker S. Endoscopic ultrasound restagingof rectal cancer is inaccurate following neoadjuvant chemora-diation therapy. Gastrointest Endosc 2000;51:AB172.

108. Rotondano G, Esposito P, Pellecchia L, Novi A, Romano G. Earlydetection of locally recurrent rectal cancer by endosonography.Br J Radiol 1997;70:567–571.

109. van Oostayen JA, Wasser MN, van Hogezand RA, Griffioen G,Biemond I, Lamers CB, de Roos A. Doppler sonography evalu-ation of superior mesenteric artery flow to assess Crohn’sdisease activity: correlation with clinical evaluation, Crohn’sdisease activity index, and alpha 1-antitrypsin clearance in fe-ces. AJR Am J Roentgenol 1997;168:429–433.

110. Byrne M, Farrell M, Abass S, Varghese J, Fitzgerald A, ThorntonF, Murray F, Lee M. Assessment of Crohn’s disease activity byDoppler sonography of the superior mesenteric artery, clinicalevaluation and the Crohn’s disease activity index-a prospectivestudy. Clinical Radiology 2001;56:973–978.

111. Shimizu S, Tada M, Kawai K. Endoscopic ultrasonography ininflammatory bowel diseases. Gastrointest Endosc Clin N Am1995;5:851–859.

112. Gast P, Belaiche J. Rectal endosonography in inflammatorybowel disease: differential diagnosis and prediction of remis-sion. Endoscopy 1999;31:158–166.

113. Soweid AM, Chak A, Katz JA, Sivak MV Jr. Catheter probeassisted endoluminal US in inflammatory bowel disease. Gas-trointest Endosc 1999;50:41–46.

114. Panelli F, Erickson RA, Prasad VM. Evaluation of mediastinalmasses by endoscopic ultrasound and endoscopic ultrasound-guided fine needle aspiration. Am J Gastroenterol 2001;96:401–408.

115. Ronson RS, Duarte I, Miller JI. Embryology and surgical anatomyof the mediastinum with clinical implications. Surg Clin NorthAm 2000;80:157–169.


116. Laurent F, Latrabe V, Lecesne R, Zennaro H, Airaud JY, Rautu-rier JF, Drouillard J. Mediastinal masses: diagnostic approach.Eur Radiol 1998;8:1148–1159.

117. Serna DL, Aryan HE, Chang KJ, Brenner M, Tran LM, Chen JC. Anearly comparison between endoscopic ultrasound-guided fine-needle aspiration and mediastinoscopy for diagnosis of medi-astinal malignancy. Am Surg 1998;64:1014–1018.

118. Gress FG, Savides TJ, Sandler A, Kesler K, Conces D, Cum-mings O, Mathur P, Ikenberry S, Bilderback S, Hawes R. Endo-scopic ultrasonography, fine-needle aspiration biopsy guided byendoscopic ultrasonography, and computed tomography in thepreoperative staging of non-small-cell lung cancer: a compari-son study. Ann Intern Med 1997;127:604–612.

119. Aabakken L, Silvestri GA, Hawes R, Reed CE, Marsi V, HoffmanB. Cost-efficacy of endoscopic ultrasonography with fine-needleaspiration vs. mediastinotomy in patients with lung cancer andsuspected mediastinal adenopathy. Endoscopy 1999;31:707–711.

120. Martini N. Operable lung cancer. CA Cancer J Clin 1993;43:201–214.

121. Mishra G, Sahai AV, Penman ID, Williams DB, Judson MA, LewinDN, Hawes RH, Hoffman BJ. Endoscopic ultrasonography withfine-needle aspiration: an accurate and simple diagnostic mo-dality for sarcoidosis. Endoscopy 1999;31:377–382.

122. Wiersema MJ, Wiersema LM. Endosonography-guided celiacplexus neurolysis. Gastrointest Endosc 1996;44:656–662.

123. Harada N, Wiersema MJ, Wiersema LM. Endosonography-guided celiac plexus neurolysis. Gastrointest Endosc Clin N Am1997;7:237–245.

124. van Dongen RT, Crul BJ. Paraplegia following coeliac plexusblock. Anaesthesia 1991;46:862–863.

125. Gress F, Schmitt C, Sherman S, Ciaccia D, Ikenberry S, LehmanG. Endoscopic ultrasound-guided celiac plexus block for man-aging abdominal pain associated with chronic pancreatitis: aprospective single center experience. Am J Gastroenterol 2001;96:409–416.

126. Kozarek RA, Brayko CM, Harlan J, Sanowski RA, Cintora I, KovacA. Endoscopic drainage of pancreatic pseudocysts. Gastroin-test Endosc 1985;31:322–327.

127. Fockens P, Johnson TG, van Dullemen HM, Huibregtse K, TytgatGN. Endosonographic imaging of pancreatic pseudocysts be-fore endoscopic transmural drainage. Gastrointest Endosc1997;46:412–416.

128. Norton ID, Clain JE, Wiersema MJ, DiMagno EP, Petersen BT,Gostout CJ. Utility of endoscopic ultrasonography in endoscopicdrainage of pancreatic pseudocysts in selected patients. MayoClin Proc 2001;76:794–798.

129. Sahel J, Bastid C, Pellat B, Schurgers P, Sarles H. Endoscopiccystoduodenostomy of cysts of chronic calcifying pancreatitis: areport of 20 cases. Pancreas 1987;2:447–453.

130. Gerolami R, Giovannini M, Laugier R. Endoscopic drainage ofpancreatic pseudocysts guided by endosonography. Endoscopy1997;29:106–108.

131. Seifert H, Dietrich C, Schmitt T, Caspary W, Wehrmann T. En-doscopic ultrasound-guided one-step transmural drainage ofcystic abdominal lesions with a large-channel echo endoscope.Endoscopy 2000;32:255–259.

132. Pasricha PJ, Ravich WJ, Hendrix TR, Sostre S, Jones B, Kalloo

AN. Intrasphincteric botulinum toxin for the treatment of acha-lasia. N Engl J Med 1995;332:774–778.

133. Hoffman BJ, Knapple WL, Bhutani MS, Verne GN, Hawes RH.Treatment of achalasia by injection of botulinum toxin underendoscopic ultrasound guidance. Gastrointest Endosc 1997;45:77–79.

134. Lahoti S, Catalano MF, Alcocer E, Hogan WJ, Geenen JE. Oblit-eration of esophageal varices using EUS-guided sclerotherapywith color Doppler. Gastrointest Endosc 2000;51:331–333.

135. Chang KJ, Nguyen PT, Thompson JA, Kurosaki TT, Casey LR,Leung EC, Granger GA. Phase I clinical trial of allogeneic mixedlymphocyte culture (cytoimplant) delivered by endoscopic ultra-sound-guided fine-needle injection in patients with advancedpancreatic carcinoma. Cancer 2000;88:1325–1335.

136. Wiersema MJ, Sandusky D, Carr R, Wiersema LM, Erdel WC,Frederick PK. Endosonography-guided cholangiopancreatogra-phy. Gastrointest Endosc 1996;43:102–106.

137. Gress F, Ikenberry S, Sherman S, Lehman G. Endoscopic ultra-sound-directed pancreatography. Gastrointest Endosc 1996;44:736–739.

138. Akahoshi K, Chijiiwa Y, Hamada S, Sasaki I, Maruoka A, Kabe-mura T, Nawata H. Endoscopic ultrasonography: a promisingmethod for assessing the prospects of endoscopic mucosalresection in early gastric cancer. Endoscopy 1997;29:614–619.

139. Chang KJ, Albers CG, Nguyen P. Endoscopic ultrasound-guidedfine needle aspiration of pleural and ascitic fluid. Am J Gastro-enterol 1995;90:148–150.

140. Goldberg SN, Mallery S, Gazelle GS, Brugge WR. EUS-guidedradiofrequency ablation in the pancreas: results in a porcinemodel. Gastrointest Endosc 1999;50:392–401.

141. Ido K, Isoda N, Sugano K. Microwave coagulation therapy forliver cancer: laparoscopic microwave coagulation. J Gastroen-terol 2001;36:145–152.

142. Bhutani MS, Hoffman BJ, van Velse A, Hawes RH. Contrast-enhanced endoscopic ultrasonography with galactose micropar-ticles: SHU508 A (Levovist). Endoscopy 1997;29:635–639.

143. Nomura N, Goto H, Niwa Y, Arisawa T, Hirooka Y, Hayakawa T.Usefulness of contrast-enhanced EUS in the diagnosis of upperGI tract diseases. Gastrointest Endosc 1999;50:555–560.

144. Nakata N, Miyamoto Y, Tsujimoto F, Harada J, Tada S, FukudaK. Ultrasound virtual endoscopic imaging. Semin Ultrasound CTMR 2001;22:78–84.

145. Kanemaki N, Nakazawa S, Inui K, Yoshino J, Yamao J,Okushima K. Three-dimensional intraductal ultrasonography:preliminary results of a new technique for the diagnosis ofdiseases of the pancreatobiliary system. Endoscopy 1997;29:726–731.

146. Sahai AV, Hoffman BJ, Hawes RH. Endoscopic ultrasound-guided hepaticogastrostomy to palliate obstructive jaundice:preliminary results in pigs. Gastrointest Endosc 1998;47:AB37.

Received February 21, 2002. Accepted March 15, 2002.Address requests for reprints to: Michael F. Byrne, M.D., MA

(Cantab), MRCP, Box 3189, Duke University Medical Center, Durham,North Carolina 27710. e-mail: [email protected]; fax: (919)684-4695.


Gastrointestinal Imaging: Endoscopic Ultrasound - Gastroenterology

Documents

Transcript of Gastrointestinal Imaging: Endoscopic Ultrasound - Gastroenterology