You are on page 1of 10

Article

Power Doppler Endoscopic Ultrasonography for the Differential Diagnosis Between Pancreatic Cancer and Pseudotumoral Chronic Pancreatitis
Adrian Sa ˘ftoiu, MD, PhD, Carmen Popescu, MD, Sergiu Cazacu, MD, Daniela Dumitrescu, MD, Claudia Valentina Georgescu, MD, PhD, Mihai Popescu, MD, Tudorel Ciurea, MD, PhD, Florin Gorunescu, MD, PhD
Objective. The accuracy of endoscopic ultrasonography (EUS) and EUS-guided fine-needle aspiration for the differential diagnosis of pancreatic masses is variable in the literature, being as low as 75% in some studies. The aim of the study was to assess the accuracy of power Doppler EUS for the differential diagnosis between pancreatic cancer and pseudotumoral chronic pancreatitis. Methods. We included 42 consecutive patients with pancreatic tumor masses (27 men and 15 women) examined by EUS between January 2002 and August 2004. Endoscopic ultrasonographic procedures included power Doppler EUS as well as EUS-guided fine-needle aspiration in all patients. Final diagnosis of pancreatic cancer was confirmed in 29 patients on the basis of a combination of information provided by imaging tests, follow-up of at least 6 months, and laparotomy in 18 patients for diagnostic or palliative reasons. Results. Sensitivity and specificity of the absence of power Doppler signals inside the suggestive pancreatic mass were 93% and 77%, respectively, with accuracy of 88%. Moreover, the addition of the information provided by the presence of peripancreatic collaterals improved the sensitivity and specificity to 97% and 92%, with accuracy of 95%. Conclusions. Power Doppler EUS provides useful information for the differential diagnosis of pancreatic masses. The results were in concordance with previous studies that showed a hypovascular pattern of pancreatic carcinoma, as well as the formation of collaterals in advanced cases due to the invasion of the splenic or portal veins. Further studies of dynamic EUS with contrast agents are necessary to better characterize pancreatic masses. Key words: chronic pancreatitis; endoscopic ultrasonography; pancreatic cancer; power Doppler ultrasonography.
Abbreviations CT, computed tomography; EUS, endoscopic ultrasonography; FNA, fine-needle aspiration; ROC, receiver operating characteristic; TUS, transabdominal ultrasonography Received August 16, 2005, from the Departments of Gastroenterology (A.S., S.C., T.C.), Radiology and Imaging (D.D., M.P.), Pathology (C.V.G.), and Biostatistics (F.G.) and Cytology Laboratory (C.P.), University of Medicine and Pharmacy Craiova, Craiova, Romania. Revision requested September 6, 2005. Revised manuscript accepted for publication November 3, 2005. Dr Sa ˘ftoiu thanks Prof Dr Peter Vilmann for critical review of the manuscript as well as for help and advice in mastering the endoscopic ultrasonographically guided fine-needle aspiration technique. Address correspondence to Adrian Sa ˘ftoiu, MD, PhD, Department of Gastroenterology, University of Medicine and Pharmacy Craiova, strada Horia 11, Craiova, Dolj 200490, Romania. E-mail: adry@umfcv.ro

P

ancreatic cancer is the 10th most common malignancy, being the fourth leading cause of cancer death. Most of the patients are diagnosed by imaging studies that visualize pancreatic masses.1 Because of the variable sensitivity and specificity, different methods are currently used: transabdominal ultrasonography (TUS), contrast-enhanced computed tomography (CT), magnetic resonance cholangiopancreatography, endoscopic ultrasonography (EUS), and endoscopic retrograde cholangiopancreatography.2 However, the differential diagnosis of pancreatic cancer and pseudotumoral chronic pancreatitis is particularly difficult despite the progress of imaging techniques. Tissue diagnosis (cytologic or histologic examination) is essential in these cases, with EUS-guided fine-needle

© 2006 by the American Institute of Ultrasound in Medicine • J Ultrasound Med 2006; 25:363–372 • 0278-4297/06/$3.50

Thus. and discordant cases were solved by consensus.C. The other 13 patients had a diagnosis of chronic pseudotumoral pancreatitis and were followed for at least 6 months. All EUS examinations with EUS-FNA were performed by 2 gastroenterologists (A. Ltd. where available.5 Hence. Aloka Co. To establish whether there were differences between the 2 groups of reviewers. clinical results.). Japan). or CT examinations. The gain was also set as a function of the flow in the portal vein. sex. and S. Pancreatic ductal adenocarcinoma has a poor prognosis. and its ability to differentiate between pancreatic cancer and pseudotumoral chronic pancreatitis is limited. other imaging procedures. or repeated EUS-FNA. The presence or absence of power Doppler signals inside the pancreatic mass was reviewed by both endoscopists (A.C. and final pathologic diagnosis. whereas surgical resection was still necessary to confirm the diagnosis.6. and sedation with midazolam or propofol was applied. being as low as 75% in some studies. complications.C. Patients were placed in the left lateral decubitus position. and T. ultrasonographic. although the sensitivity of EUS-FNA for the differential diagnosis of pancreatic masses is variable in the literature. follow-up of at least 6 months by repeated clinical. and the gain was set to avoid the appearance of noise. Endoscopic ultrasonographically guided fine-needle aspiration is becoming the standard for obtaining cytologic diagnosis. followup (type and duration in months). for diagnostic or palliative reasons.3 Moreover. Furthermore. and S.D. An ideal scanning plane of the mass was chosen. The study did not use any procedures beyond our hospital protocol because all imaging tests (including EUS-FNA) are current proce364 . The procedures were recorded on a DVD recorder for latter review. A careful optimization of the power Doppler technique was performed. and efforts were made to interrogate the whole area and to assess the presence or absence of vascular signals. 42 consecutive patients with a suspicion of pancreatic tumor masses underwent power Doppler EUS. the examinations were further reviewed by 2 readers (D. by ultrasonography. J Ultrasound Med 2006. 25:363–372 Materials and Methods Study Design and Patients The study prospectively included all the patients with a previous suspicion of pancreatic masses by ultrasonography or CT examined in the Department of Gastroenterology of the University of Medicine and Pharmacy Craiova over a 30month period from January 2002 to August 2004. CT. The following information was gathered for each patient: age.7 The aim of our study was to prospectively assess the accuracy of linear EUS with power Doppler imaging to differentiate between pancreatic cancer and chronic pancreatitis in patients with pancreatic masses visualized by other imaging techniques. followed by EUS-FNA in all cases. CT. Japan) coupled with the corresponding ultrasonography system (ProSound 5000.S. TNM staging.4 Consequently. A final diagnosis of pancreatic cancer was confirmed in 29 patients on the basis of a combination of information provided by imaging tests (including EUS-FNA and cytologic confirmation of malignancy).) according to a common protocol. Olympus Optical Co. and cytologic examination of EUS-FNA).S. a receiver operating characteristic (ROC) analysis was done to establish the accuracy of power Doppler EUS for the differential diagnosis (see below). the sensitivity was reported to be unacceptably lower (54%) in the context of chronic pancreatitis.Power Doppler Endoscopic Ultrasonography of Pancreatic Masses aspiration (EUS-FNA) usually performed. Tokyo. with a 5-year survival rate for patients undergoing curative resection of less than 10% to 20%. Ltd. Tokyo. dures used in the daily clinical treatment of patients. the early diagnosis of pancreatic carcinoma in patients with chronic pancreatitis becomes crucial to establish the management and to improve prognosis. type of therapy (chemotherapy or curative or palliative surgery). The presence or absence of collateral circulation was also reviewed in all cases.) who were absent during the EUS examinations and were blinded to the results of other imaging procedures (TUS. EUS findings and diagnosis. Power Doppler EUS Power Doppler EUS procedures were performed with a linear echoendoscope (GF-UCT 140 AL5. All patients provided written informed consent to undergo the EUS examination and were followed for a minimum of 6 months or until death occurred. EUS-FNA in patients with pancreatic masses has a low negative predictive value. which was assumed to be laminar and fluent. as well as laparotomy with pathologic results in 18 patients.

and. that is. whereas the Giemsa stain was done after dry fixation and postfixation with methanol. atypical probably malignant.8 The needle (Olympus NA-10J-1) was directly visualized under real-time conditions during the procedure. atypical probably benign. not only blood or scant cellularity samples. specificity. and laparotomy in 18 patients for diagnostic or palliative reasons (Tables 1 and 2). the results of power Doppler EUS and EUS-FNA were classified by the examiners as definitely negative (1). nuclei. probably positive (4). The characteristics of the patient subgroups and EUS examinations are presented in Tables 1–3. with all suggestive cases regarded as malignant. and malignant. suggestive of malignancy.10 Consequently. Results The study included 42 consecutive patients with a clinical suspicion of pancreatic tumor masses (27 men and 15 women) examined by EUS. finally. local lymph nodes. thus ensuring the correct placement inside the suggestive pancreatic mass. using a technique previously published in detail.P was in attendance dur. by a comparison of the results of these techniques with the final diagnosis. Receiver operating characteristic analysis thus displays the range of tradeoffs between true-positive and false-positive rates possible with the test. as well as for EUS-FNA. possibly positive (3).9 Between 3 and 6 passes were necessary in each patient to obtain sufficient material. and the resulting monolayer smears were colored by both Giemsa and Papanicolaou stains in all cases. and pulmonary carcinomas) and the diagnosis of neuroendocrine differentiations or neuroendocrine tumors.Saftoiu et al ˘ EUS-FNA and Cytologic Analysis The EUS-FNA procedures were performed according to a common protocol. ascites. the primary tumor) to prevent a potential false-positive upstaging of the disease. The diagnosis of pancreatic cancer was confirmed in 29 patients on the basis of a combination of information provided by imaging tests (including EUS with FNA biopsy and cytologic confirmation of malignancy). or definitely positive (5). both in the center and periphery. A conclusion of negative or positive concerning the diagnosis of malignancy was recorded. and advised as to the need of additional passes. the Papanicolaou stain was done after wet fixation in ethanol for at least 5 minutes. The sensitivity. Endoscopic ultrasonographic procedures included color and power Doppler ultrasonography as well as EUS-FNA in all patients. Thus. If multiple lesions were sampled in the same patient. The patients were included prospectively based on a suspicion of a pancreatic mass by ultrasonography or CT. breast carcinomas. pancreatic lymphoma. Endoscopic ultrasonographically guided fine-needle aspiration was performed in various locations within the lesion. clinical follow-up of 6 months. positive predictive value. Thus. In ROC analysis. by subtle adjustments of the needle and scope positions (using both the up-down wheel and the elevator) to maximize the likelihood of obtaining diagnostic material. J Ultrasound Med 2006. Immunocytochemistry was used in selected cases for the confirmation of the pancreatic origin (acinar or ductal). Receiver operating characteristic curve analysis was done separately for power Doppler EUS results of the 2 groups of readers (blinded and unblinded) and also for the results of EUS-FNA. and implicit criteria for assessment may vary among examiners. probably negative (2). This method fits well to our task because image assessments are made subjectively. exclusion of metastases (eg. between January 2002 and August 2004. EUS-FNA was performed in a successive order (distant metastases. gastric carcinomas. the subgroup of patients with pancreatic cancer included 21 men and 8 women with a 365 . The aspirated material obtained by EUS-FNA was sprayed onto glass slides. The smears were individually characterized concerning the cells. nucleoli. and accuracy for the differential diagnosis between pancreatic cancer and chronic pseudotumoral pancreatitis were calculated for power Doppler EUS. prepared the smears. and chromatin and were further interpreted as benign. 25:363–372 Statistical Analysis All results are expressed as mean ± SD. the differential diagnosis implies a classification of the results as either positive (pancreatic cancer) or negative (chronic pancreatitis) based on a level of confidence of each case on an ordinal scale (1–5). A cytopathologist (C. which included a minimum of 3 passes with a minimum of 10 “to-and-fro” movements in each pass that were performed under continuous aspiration. negative predictive value.) ing all the procedures. verified the adequacy of specimens.

male. the sensitivity. 25:363–372 . male.9%]). the negative predictive value was only 81%. EUS-FNA Surgery CT. Although the positive predictive value was 100%. TNM Stage (CT + EUS) EUS-FNA Result Final Diagnosis (Follow-up) Follow-up. EUS-FNA Surgery Surgery CT. and M. and 93%. There were no false-positive results of EUS-FNA in the patients with chronic pancreatitis. EUS-FNA Surgery Surgery Surgery Surgery Surgery CT. EUS-FNA CT. EUS-FNA CT. 100%. y Sex Size. mean age ± SD of 62. body (5 patients [17. EUS-FNA CT. with the exception of power Doppler EUS. Four patients had clinical and EUS criteria of chronic pancreatitis (Table 2). EUS-FNA Surgery CT. EUS-FNA 8 2 14 9 F indicates female.2%]). Results of endoscopic ultrasonographically guided fineneedle aspiration (Figure 1A) with cytologic (Figure 1B) and immune cytochemical analysis were positive in 26 cases of pancreatic cancer and false-negative in 3 patients. respectively. mm Mass (Location) Power Doppler Collaterals (Location) The tumors appeared by EUS as hypoechoic masses. CT. Pancreatic Cancer Without Chronic Pancreatitis (n = 25) Case Age. There were no complications recorded after EUS-FNA. and M. 366 J Ultrasound Med 2006. mo 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 56 75 30 69 42 68 52 67 73 74 67 76 77 67 70 46 63 43 53 56 66 61 62 68 76 F M M F M M M M M F M M M F M F M F F M M F M F M 23 35 45 17 30 40 35 45 40 40 55 20 25 30 50 18 70 20 40 40 35 32 50 50 40 Head Head Head Head Head Head Head Head Head Body Body Uncinate Head Head Head Head Body Head Head Body Body Head Head Uncinate Head Absent Absent Absent Absent Absent Absent Absent Absent Absent Absent Absent Absent Absent Absent Absent Absent Absent Absent Absent Absent Absent Present Present Absent Absent Absent Head Head. The pancreatic tumors were located in the head (20 patients [69%]). EUS-FNA Surgery Surgery CT.2 ± 12. or tail (2 patients [6. Consequently. uncinate process (2 patients [6.2 years. Pancreatic Cancer With Chronic Pancreatitis (n = 4) Case Age. specificity. and accuracy of EUS-FNA were 90%. tail Absent Head Absent Absent T1 N0 M0 T4 N1 M0 T4 N1 M1 T1 N0 M0 T2 N0 M0 T4 N1 M1 T4 N1 M1 T4 N1 M1 T4 N1 M0 T4 N1 M0 T4 N1 M0 T2 N0 M0 T4 N0 M0 T4 N0 M0 T4 N1 M0 T2 N1 M0 T4 N1 M0 T2 N0 M0 T4 N0 M0 T3 N1 M0 T4 N1 M1 T4 N1 M0 T4 N1 M0 T4 N1 M1 T4 N1 M0 Positive Positive Positive Negative Positive Positive Positive Positive Positive Positive Positive Negative Positive Positive Positive Positive Positive Positive Positive Positive Positive Positive Positive Positive Positive Surgery Surgery CT. mm Mass (Location) Power Doppler Collaterals (Location) TNM Stage (CT + EUS) EUS-FNA Final Diagnosis Result (Follow-up) Follow-up. inhomogeneous. The diagnosis of pancreatic cancer in patients with chronic pancreatitis was initially suspected on imaging (presence of a mass by ultrasonography. of variable dimensions between 17 and 50 mm in diameter. y Sex Size. EUS-FNA Surgery 27 20 6 27 17 2 3 13 12 5 6 10 10 8 3 3 9 4 2 8 8 3 7 6 6 F indicates female. The imaging findings in this subgroup of patients did not differ from those in the patients with pseudotumoral chronic pancreatitis. and EUS) and confirmed by EUS-FNA in 3 patients and surgery with pathologic results in another patient. body Absent Absent Absent Head Absent Head Body Absent Absent Absent Absent Head Absent Absent Absent Absent Absent Body. EUS-FNA Surgery Surgery Surgery CT.Power Doppler Endoscopic Ultrasonography of Pancreatic Masses Table 1.9%]) of the pancreas. mo 1 2 3 4 67 54 78 48 M M F M 35 50 20 40 Tail Head Head Tail Absent Absent Absent Absent Body. Table 2. tail Body Absent Absent T4 T4 T2 T4 N0 N1 N0 N1 M0 M1 M0 M0 Negative Positive Positive Positive Surgery CT.

Sensitivity and specificity of the absence of power Doppler signals inside the suggestive pancreatic mass were 93% and 77%. hyperechoic margins of the pancreatic duct. and pseudocysts. mm Mass (Location) Power Doppler Collaterals (Location) EUS Criteria* EUS-FNA Result Final Diagnosis (Follow-up) Follow-up. with indefinite margins. CT Ultrasonography. B. visualized as a hypoechoic. Endoscopic ultrasonographically guided FNA with real-time visualization of the needle inside the tumor was subsequently performed. and tail. mo 1 2 3 4 5 6 7 8 9 10 11 12 13 37 55 36 55 50 54 64 51 35 65 61 60 40 F M F F M M M F M M M M F 40 30 25 40 45 15 30 28 33 50 30 25 22 Tail Head Head Head Head Head Head Head Head Head Head Head Body Present Present Absent Present Present Absent Absent Present Present Present Present Present Present Absent Absent Body Absent Absent Absent Absent Absent Absent Absent Absent Absent Head 5 4 6 6 4 5 6 4 5 7 4 4 6 Negative Negative Negative Negative Negative Negative Negative Negative Negative Negative Negative Negative Negative Surgery Ultrasonography.7 months and consisted of repeated clinical. respectively. Moreover. 13 patients were dead or had a diagnosis of liver metastases during the follow-up. and 1 mm at the level of head. Portal vein invasion was evident posterior to the pancreatic head (arrowheads). whereas 2 patients had hypervascular tumors (Figure 3). *EUS criteria of chronic pancreatitis included inhomogeneity of parenchyma. inhomogeneous mass. pancreatic duct dilatation (>3. 2. Chronic Pancreatitis (n = 13) Case Age. hyperechoic foci or bands. The positive predictive value and negative predictive value were 90% and 83%.Saftoiu et al ˘ Table 3. male. with accuracy of 88%. ultrasonographic. The mean follow-up of the patients with pancreatic cancer was 8. CT EUS-FNA EUS-FNA Surgery EUS-FNA Ultrasonography. Moreover. “honeycomb” appearance of the parenchyma. Pancreatic head adenocarcinoma of approximately 30 × 26 mm in diameter (arrows). The positive predictive value and negative predictive value were 97% and 92%. CT EUS-FNA 25 24 25 11 17 11 23 15 6 24 28 11 28 F indicates female. CT Surgery Ultrasonography. A B J Ultrasound Med 2006. calcifications. respectively). Chronic pseudotumoral pancreatitis was diagnosed in 13 patients. who were subsequently followed for a minimum of 6 months (Table 3). Although the patients had suggestive masses on Figure 1. irregular pancreatic duct. and M. body. The patients with pancreatic tumors had absent power Doppler signals inside the tumor mass in 27 cases (Figure 2). The cytologic smears showed clumps of atypical cells (arrows) with a small island of normal acinar cells (arrowheads) in a background of erythrocytes (Papanicolaou stain). 25:363–372 367 . respectively.9 ± 6. the addition of the information provided by the presence of peripancreatic or periduodenal collaterals (Figure 4) improved the sensitivity and specificity to 97% and 92%. A. or CT examinations. with accuracy of 95%. y Sex Size. CT EUS-FNA Ultrasonography.

visualized at the level of the pancreatic head (arrows).930–1. they also had increased vascularization depicted by power Doppler imaging inside the tumor mass in 10 of 13 patients (Figure 5). visualized at the level of the pancreatic body (arrows). especially for the detection of small pancreatic tumors of less than 2 to 3 cm. The areas under the ROC curve were 0. hypoechoic masses on the EUS examination. the absence of power Doppler signals inside the suggestive pancreatic mass and the presence of scattered signals inside the pancreatic parenchyma (Figure 6A) were helpful for the correct diagnosis. and staging and resectability. with absent power Doppler signal inside. Because of the suspicion of pancreatic cancer.5 months.000) for EUS-FNA and 0. with the same results in both blinded and unblinded readers.2. 368 J Ultrasound Med 2006. being either equivalent or superior for the diagnosis of pancreatic cancer. CT.857 (95% confidence interval.1342). with absent power Doppler signals and retrograde dilatation of the main pancreatic duct (arrowheads). The difference was not statistically significant (P = . Hypoechoic tumor mass due to pancreatic adenocarcinoma of approximately 20 × 16 mm. with indefinite margins and power Doppler signals inside. Separate ROC analyses were conducted for the results of power Doppler EUS.000) for power Doppler EUS. as well as for the interpretation of EUS-FNA (Figure 7). again with negative results. Discussion Endoscopic ultrasonography coupled with EUSFNA is very useful for the detection of pancreatic tumors. 0. of approximately 35 × 30 mm.Power Doppler Endoscopic Ultrasonography of Pancreatic Masses Figure 2. inhomogeneous mass due to pancreatic adenocarcinoma. Collateral circulation was visualized in power Doppler mode in front of the pancreatic body. visualized at the level of the pancreatic head (arrows). The patients with concomitant chronic pancreatitis and pancreatic cancer were the most difficult to examine because of the inhomogeneous pancreatic parenchyma. of 40 × 35 mm.1 ± 7.1 Endoscopic ultrasonography was previously reported to be superior to TUS and conventional CT. confirmed by EUS-FNA (Figure 6B). 0. 25:363–372 . Hypoechoic tumor due to pancreatic adenocarcinoma.13 A recent prospective study assessed multidetector CT and EUS for the detection of Figure 4. Hypoechoic. with hyperechoic foci and bands. The differential diagnosis consisted of a classification of the results as either positive (pancreatic cancer) or negative (chronic pancreatitis) based on an ordinal scale of 1 to 5 in each case.975 (95% confidence interval. In these cases. differential diagnosis between benign and malignant tumors. EUS-FNA was repeated in 5 of 13 patients.709–1. ultrasonography or CT and even inhomogeneous. tion of imaging tests (ultrasonography. The presence of pancreatic cancer was excluded by a combinaFigure 3.11 Endoscopic ultrasonography was further compared with helical CT. however. and EUS-FNA) repeated during the mean follow-up of 19.12.

with a suggestion of a pancreatic head tumor mass (arrows). which may influence treatment of the patients. Hence. usually 90% to 95% but as low as 75% in some studies. 25:363–372 369 . with a negative predictive value of 83%.3 The same results were obtained in our study. even EUS can miss diffusely infiltrative pancreatic tumors or neoplasms in the setting of chronic pancreatitis.22 The diagnostic yield of EUS-FNA in pancreatic cancer is also dependent on the technique and the experience of the team (endoscopist and cytopathologist). Three patients had negative EUS-FNA findings despite multiple passes. In our study. However. inhomogeneous mass at the level of the pancreatic head (arrows). concluding that EUS is superior because it identifies tumors that are undetected by CT. Accuracy of EUS with EUS-FNA for the differential diagnosis of pancreatic masses is variable. The mass had no power Doppler signals inside. Chronic pancreatitis with an inhomogeneous pancreatic head containing calcifications. the accuracy of the absence of power Doppler signals inside the mass in diagnosing pancreatic cancer was 88%. Other tumor-related factors may also decrease the cellularity of samples.14 However. comparable with the accuracy and negative predictive value of EUS-FNA of 93% and 81%. follow-up EUS after 2 to 3 months might be necessary. respectively. Chronic pancreatitis with a hypoechoic.Saftoiu et al ˘ Figure 5. A B J Ultrasound Med 2006. with excellent accuracy of 93% but a lower negative predictive value of only 81%. whereas EUS-FNA results were negative for atypical cells. the negative predictive value was reported to be low because of falsenegative results. B. it has a limited ability to differentiate between inflammatory masses and cancer.17–21 Most studies report high specificity (close to 100%) and positive predictive value. with disastrous results.16 The next step was to use EUS-FNA for a correct discrimination between pancreatic tumor masses.21. of approximately 50 × 45 mm. whereas the diagnosis of adenocarcinoma was confirmed by EUS-guided FNA. in a background of erythrocytes (Giemsa stain). Figure 6.8. A. Power Doppler EUS showed intense vascularization of the inflammatory mass (arrows).3. Thus. the ability of imaging methods (EUS with power Doppler imaging) was similar to that of EUS-FNA for the differential diagnosis of pancreatic cancer and pseudotumoral inflammatory masses.15 Although EUS has high sensitivity for the detection of pancreatic masses. pancreatic cancer. including extensive fibrosis (desmoplastic reaction) and necrosis. as well as the degree of differentiation (welldifferentiated tumors require a larger number of needle passes than moderately or poorly differentiated tumors). indicating the presence of adenocarcinoma (arrows). probably because of sampling errors caused by the scant cellularity of specimens due to the small size of the lesions (2 patients) or the presence of chronic pancreatitis in another patient. in case of a high suspicion of pancreatic neoplasm. The cytologic smears showed tridimensional clumps of atypical cells.

all these methods based on TUS are limited only for patients in whom the pancreatic tumor and parenchyma can be clearly observed in a single view. repeated EUS-FNA.26 Contrast-enhanced EUS was also used for the differential diagnosis of pancreatic tumor masses for a better assessment of perfusion in the pancreatic tissue and inside the mass. even in the absence of tissue confirmation that may occur in patients with chronic pancreatitis and pseudotumoral inflammatory masses. in which a diagnosis of pancreatic cancer cannot be excluded. did not significantly increase the sensitivity for the detection of pancreatic lesions compared with conventional B-mode ultrasonography (70% versus 60%.2 The introduction of new techniques.Power Doppler Endoscopic Ultrasonography of Pancreatic Masses Figure 7. thus reducing the associated morbidity and mortality. P = . EUS with power Doppler imaging provides information about the etiology of the tumor mass.857. whereas markedly hypervascular lesions were inflammatory masses. Nevertheless. such as phase inversion tissue harmonic imaging. The ability of power Doppler EUS to differentiate pancreatic masses has to be viewed as an adjunct to the other imaging techniques rather than a replacement of tissue confirmation. However. Dynamic imaging is increasingly used for the differential diagnosis of pancreatic masses. the results were comparable with cytopathologic results (percutaneously or EUS guided) with high sensitivity (94%) and specificity (100%). Consequently. with better accuracy of 95% and a higher negative predictive value of 92%.2 370 Thus. especially in the setting of chronic pancreatitis. 25:363–372 . Comparative ROC analysis of power Doppler EUS and EUS-FNA based on a classification of the results as either positive (pancreatic cancer) or negative (chronic pancreatitis). New progress in ultrasonography systems and transducer technology has facilitated the application of contrast-enhanced coded phase (pulse) inversion harmonic or wideband ultrasonographic techniques for a better characterization of pancreatic masses.975 and 0. even without the enhancement produced by contrast agents. poor visualization of the tumor mass and operator dependence limit the transabdominal visualization of the pancreas. Moreover.46). determined according to an ordinal scale of 1 to 5 in each case.23–25 Pancreatic carcinoma was described as usually hypovascular compared with the rest of the parenchyma. P = . overlying bowel gas.31 The results of our study showed that power Doppler EUS was an accurate discriminative test. contrast-enhanced power Doppler ultrasonography was used for the transabdominal characterization of pancreatic masses. or presence of ascites. the addition of the information provided by the appearance of collaterals enhanced the diagnostic value of power Doppler ultrasonography.26–28 These techniques showed the presence of small tumor vessels in approximately 65% of pancreatic ductal carcinomas. Currently. Moreover. there is no imaging method that can reliably provide this capability in patients with pancreatic masses.30. the relative frequency is inferior to the frequency of collateral appearance in patients with pancreatic cancer. The pancreatic carcinoma masses were relatively hypovascular compared with surrounding parenchyma. with a difference between areas under the ROC curves that was not statistically significant (0. disturbing the examinations in up to one third of patients because of large body habitus.31 Pancreatic carcinoma was again shown to be relatively hypovascular compared with surrounding pancreatic tissue.29 In fact. or surgery).1342). especially in the cases with negative EUS-FNA findings. The high negative predictive value might also allow a decrease of the number of unnecessary surgical interventions.32 J Ultrasound Med 2006. especially when the information provided by the presence of collaterals was added. categorizing the risk of malignancy is very important for the clinical decision-making process and subsequent treatment (follow-up CT and EUS. Receiver operating characteristic analyses also showed excellent accuracy for power Doppler EUS and EUS-FNA. whereas inflammatory masses are isovascular or hypervascular. Although collaterals might also appear in chronic pancreatitis because of segmentary portal hypertension.

org. et al. as well as the overall accuracy of the imaging methods. Gastrointest Endosc 1996. 43:238–242. However. et al. 3. Gut 2005. Falconi M. et al. and mean follow-up of 8. 52:367–371. Pancreatic tumors: comparison of dual-phase helical CT and endoscopic sonography. Takhar AS. Chang KJ. Does anyone survive pancreatic ductal adenocarcinoma? A nationwide study re-evaluating the data of the Finnish Cancer Registry. The other 13 patients with a diagnosis of chronic pseudotumoral pancreatitis based on imaging studies and negative EUS-FNA findings were followed for a mean of 19. 2:606–621. Yeo CJ. Dhinga R. ROC Analysis: Web-Based Calculator for ROC Curves. J Ultrasound Med 2006. 5:289–303. Gastrointest Endosc 1999.jrocfit. et al. Role of EUS in the preoperative staging of pancreatic cancer: a large singlecenter experience. and CT scanning for evaluation of pancreatic adenocarcinoma. Am J Gastroenterol 2003. 6. EUS. Gastrointest Endosc 2002. Gress FG. Horwhat JD. Eloubeidi MA.1 months by ultrasonography. The presence of power Doppler signals inside the inflammatory masses is variable as a function of inflammation and necrosis. Defining the diagnostic algorithm in pancreatic cancer. Bhutani MS. Bassi C. the diagnosis of pancreatic cancer was confirmed in 29 patients on the basis of a combination of EUS-FNA (positive cytologic results in 26 patients). Consequently. Another limitation was that the endoscopists were not blinded to the results of ultrasonographic or CT scanning obtained before the EUS-FNA examination combined with power Doppler imaging. a considerable selection bias might have influenced the results. Am J Gastroenterol 2002. whereas the results of EUS with power Doppler imaging were reviewed by other authors blinded by the results of previous imaging studies. 98:2663–2668. as well as the formation of collaterals in advanced cases due to the invasion of the splenic or portal veins. Comparison of endoscopic ultrasonography and multidetector computed tomography for detecting and staging pancreatic cancer. BMJ 2004. 13. Legmann P. Devereaux B. et al. Canto MI. 2005. Early detection of pancreatic cancer following the diagnosis of chronic pancreatitis. 9. Mertz HR. PET. AJR Am J Roentgenol 1998. Knofel WT. Further prospective studies of dynamic EUS with contrast agents are necessary to better characterize pancreatic masses. Carpelan-Holmström M. Comparison of endoscopic ultrasound-guided fine needle aspiration for focal pancreatic lesions in patients with normal parenchyma and chronic pancreatitis. 371 . The third limitation was the small number of patients with chronic pseudotumoral pancreatitis. However. 4. 10. Endoscopic ultrasound-guided fine-needle aspiration biopsy of patients with suspected pancreatic cancer: diagnostic accuracy and acute and 30-day complications. 2. Baltimore. 170:1315–1322. Clin Gastroenterol Hepatol 2004. Talamini G. 12. et al. Ann Intern Med 2004. 11. Eng J. 15. 50:786–791. Chriswell M. Eltoum IA.9 months (including 13 patients who were dead or had a diagnosis of liver metastases). Lobo DN. CT. Hancke S. Brand L. Recent developments in diagnosis of pancreatic cancer. Hawes RH. Maximizing the yield of EUS-guided fine-needle aspiration. Goggins M. Delbeke D. 5. DeWitt J. laparotomy with biopsies (positive pathologic results in 16 patients). Digestion 1999. The pathologists were also blinded to the results of imaging tests. Vignaux O. Fritscher-Ravens A. 329:668–673. 54:385–387. 8. 97:2768–2775.Saftoiu et al ˘ Consequently. Available at: http://www. Accessed September 15. One limitation of this study was the absence of histopathologic confirmation in 11 patients with pancreatic cancer who did not undergo surgery. Endoscopy 2004. or repeated EUS with FNA. Dousset B. Savides TJ. Palaniappan P. 25:363–372 References 1. 7. Pukkala E. which might have influenced the statistical significance of the positive and negative predictive values. The results of our study are in concordance with those of previous studies that showed a hypovascular pattern of pancreatic carcinoma. JOP J Pancreas [serial online] 2004. we included consecutive patients who had a suggestion of a pancreatic mass. 60:554–561. 14. Gress PG. In conclusion. 141:753–763. The NO Endosonographic Detection of Tumor (NEST) study: a case series of pancreatic cancers missed on endoscopic ultrasonography. Gress FG. Nordling S. power Doppler EUS provides useful information for the differential diagnosis of pancreatic masses. with a considerable bias that might have been determined by the small number of patients included. Gastrointest Endosc 2000. et al. Chen VK. combining the information provided by the absence of power Doppler signals and the presence of pancreatic collaterals yielded the best accuracy for the differential diagnosis between pseudotumoral chronic pancreatitis and pancreatic cancer. 56:S28–S34. Vilmann P. et al. Screening for pancreatic neoplasia in high-risk individuals: an EUS-based approach. Giovannini M. Sechopoulos P. MD: Johns Hopkins University. et al. et al. A new biopsy handle instrument for endoscopic ultrasound guided biopsy. 36:385–389.

Kitano M. Endoscopy 1997. Tanaka S. Brand B. Yield of endoscopic ultrasound-guided fine-needle aspiration biopsy in patients with suspected pancreatic carcinoma. Scand J Gastroenterol 2002. Phase-inversion tissue harmonic imaging compared with conventional B-mode ultrasound in the evaluation of pancreatic lesions. Am J Gastroenterol 2004. Becker D. Strobel D. Chen VK. et al. 53:854–859. 15:527–530. wide-band Doppler ultrasonography: correlation with tumor characteristics and vascular endothelial growth factor. Iseman DT. Contrastenhanced endoscopic ultrasonography with galactose microparticles: SHU508A (Levovist). Ueno N. Endoscopy 2003. Erickson RA. et al. AJR Am J Roentgenol 1997. 112:1087–1095. Molke KL. Ohshima T. Pancreas 2004. 29. 29:635–639. Bernatik T. Morita K. Endoscopic ultrasound for differential diagnosis of focal pancreatic lesions. 25. Yamaguchi T. 30. 23. 18. Cancer 2003. Gastrointest Endosc 2001. Contrast-enhanced color Doppler ultrasonography in diagnosis of pancreatic tumor: two case reports. et al. 53:784–789. et al. Gastrointest Endosc 1997. Eloubeidi MA. 169:1263– 1267. Erickson RA. Vilmann P. Chang KJ. Differentiation of pancreatic tumors by conventional ultrasound. J Ultrasound Med 1996. Kudo M. Unkrodt K. 21. et al. Schmidt T. Maekawa K. 37:1313–1320. 15:1439–1443. 100:844–850. Jhala D. 26. 35:1221– 1228. et al. Giovannini M. 20. van Velse A. Nagakawa T. Gut 2004. Scand J Gastroenterol 2000. 17. et al. Dynamic imaging of pancreatic diseases by contrast enhanced coded phase inversion harmonic ultrasonography. Evaluation of bloody flow in pancreatic ducal carcinoma using contrastenhanced. Eur Radiol 2004. Ocran KW. Gastrointest Endosc 2004. Koito K. 28. et al.Power Doppler Endoscopic Ultrasonography of Pancreatic Masses 16. Faigel DO. confirmed by surgery. AJR Am J Roentgenol 2002. 45:387–393. et al. Wermke W. Haage P. 25:363–372 . Neye H. Ginsberg GG. Hahn EG. Echo-enhanced and power Doppler EUS for the discrimination between focal pancreatitis and pancreatic carcinoma. Eloubeidi MA. Hohl C. 24. et al. Endoscopic ultrasound-guided fine needle aspiration and multidetector spiral CT in the diagnosis of pancreatic cancer. 372 J Ultrasound Med 2006. 27. 178:1133–1137. Rickes S. Chhieng DC. 19. unenhanced and echo-enhanced power Doppler sonography. Hawes RH. Binmoeller KF. 99:285–292. The clinical utility of endoscopic ultrasound-guided real-time fine-needle aspiration in the diagnosis and staging of pancreatic carcinoma. Hoffman BJ. 22. et al. Pfaff T. Bhutani MS. Bentz JS. Ioka T. Endoscopic ultrasound-guided real-time fine-needle aspiration biopsy of the pancreas in cancer patients with pancreatic lesions. Nguyen P. EUS-guided FNA. 31. 35:1015–1019. et al. Gastroenterology 1997. Agarwal B. Tano S. Prevalence and significance of periduodenal venous collaterals in patients evaluated for pancreaticobiliary disorders by endosonography. 28:335–343. Tomiyama T. Takeshi I. J Clin Oncol 1997. Endosonography-guided fine-needle aspiration biopsy: diagnostic accuracy and complication assessment. Namieno T. Abu-Hamda E. 32. 14:1109–1117. Wiersema MJ. Inflammatory pancreatic masses: differential from ductal carcinomas with contrast-enhanced sonography using carbon dioxide microbubbles. Dynamic sonography of pancreatic tumors: comparison with dynamic CT. Oshikawa O. 60:267–279.