Received Date: November 23, 2015; Accepted Date: March 23, 2016; Published Date: March 29, 2016
Citation: Klingenberg-Noftz RD, Sun L, Homann N, Ludwig D (2016) EUS Guided Fine-needle Aspiration (EUS Guided FNA) of Pancreatic Masses: Experiences from the Beginning of the Era and Implications to the Present Day. J Gastrointest Dig Syst 6:405. doi:10.4172/2161-069X.1000405
Copyright: © 2016 Klingenberg-Noftz RD, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Visit for more related articles at Journal of Gastrointestinal & Digestive System
Background: The work-up of pancreatic mass lesions requires an orchestrated employment of different diagnostic means. The “best”, meaning the most sufficient and goal achieving diagnostic pathway, remains to be established and has to be adopted according to each individual patient. The technique of endoscopic ultrasound (EUS) has been established with a high yield of diagnostic value regarding pancreatic lesions. In combination with fine needle aspirations of the suspected area guided by EUS (EUS guided FNA) the diagnostic value of pancreatic lesions is boosted with regard to the final diagnosis in a clinical setting.
Patients and Methods: During a 36-month period, 142 consecutive patients were referred to us for suspected pancreatic disease and evaluated for this study. Work-up for pancreatic lesions was performed including EUS and EUS-guided FNA. Definite diagnosis was established by explorative/curative laparoscopy/laparotomy or follow-up for up to 36 months in the aftermath of first admission. Results regarding diagnostic precision of EUS guided FNA cytology were evaluated retrospectively and correlated with other diagnostic means performed.
Results: 142 patients underwent work-up for pancreatic mass lesions of unknown genesis. EUS guided FNA was performed in all patients with a total of 13 (9.2%) minor complications (local control achieved), 2 (1.4%) major complications (1× bleeding, 1× perforation) and no fatal complication. Cytology obtained by EUS guided FNA found malignancy in 52(37%) and absence of malignant disease in 70 (49%) cases. Cytology has not rendered a definite result in 20 (14%) cases. Final diagnosis resulted in 42% (n = 59) malignant disease, 42% (n = 60) benign disease, 10 cases (7%) remained without definite diagnosis and 13 patients (9%) left hospitalization or were lost to follow-up before completing the diagnostic work-up. Sensitivity of EUS guided FNA regarding malignant pancreatic disease was 83.7%, the specificity was 95.1% (positive predictive value 93.2%, negative predictive 87.9%).
EUS guided FNA; Pancreatic mass lesions; Diagnostic work-up
Endoscopic ultrasound (EUS) guided fine needle aspiration (FNA) has initially been reported in 1992 for the diagnostic work-up of pancreatic mass lesions . The technique is at present widely accepted in the differential diagnostic of pancreatic, gastrointestinal and mediastinal diseases [2-4]. It has been shown to be of value for preoperative staging, for determining surgical resectability of pancreatic or other abdominal malignancies [5-7] and especially in obtaining material in cases prone to be managed conservatively. In a multicenter prospective study of 90 patients in whom previous diagnostic interventional radiological procedures failed to yield a definitive diagnosis, EUS-FNA has proved to be the modality of choice to establish a diagnosis . It is especially useful when computerized tomography (CT) scans do not show a focal mass lesion [6,7]. At present, data addressing sensitivity of EUS vs. CT/MRI remains undecided, but EUS, performed at expert level, appears to be the most sensitive method in diagnosing pancreatic mass lesions in sizes smaller than 20 mm, even though, data remains heterogenic in total [8-10].
Besides securing cytology for defining malignant disease, EUS-FNA is also applied to differentiate benign disease, most often from chronic pancreatitis. The diagnosis of severe chronic pancreatitis with extensive calcifications, ductal dilatation as well as intraductal concrements is fairly simple. However, diagnostic difficulties arise in patients with early, mild, minimal-change, autoimmune or focal pancreatitis that present similar to pancreatic neoplasms . Another diagnostic dilemma is evident with regard to following and diagnosing malignant transformation in patients with chronic pancreatitis which is considered a precancerous lesion. In this setting, percutaneous imaging procedures such as ultrasound (US) or CT frequently fail to visualize or differentiate discrete abnormalities of the parenchyma and ducts. Even though, preoperative sample aquisition may at times not be recommended in local or primarily curative setting of an undoubtedly malignant pancreatic tumor [12,13], only tissue sampling via EUS guided FNA and cytologic work-up could prevent operative intervention in a benign situation masquerading malignant disease [14,15]. But the risk of false negative FNA has to be taken in account.
Endoscopic retrograde cholangiopancreatography (ERCP) was traditionally considered to be the standard reference procedure for morphologic diagnosis in pancreatic cancer, but it carries a substantial risk of post-ERCP pancreatitis, only visualizes the intraductal system  only and does therefore not add any additional preoperative staging information. In the preoperative setting of potentially curative disease, routine ERCP was therefore mainly replaced by EUS. Data addressing preoperative stenting is suggestive for a variety of pre- and postoperative complications and therefore confined to cases with cholangitis, pruritus or in a palliative or neoadjuvant setting .
EUS pictures not only the intraductal system and vascularisation of the pancreas in great detail, but also enables visualization of all parts of the pancreas, including the head, the uncinate process, the neck, the body and the tail, as well as most of their adjacent organs or structures [18,19].
However, disadvantages of EUS appear with regard to abnormalities of the morphologic aspect of the pancreas that can present in asymptomatic patients . Aging, obesity or low body mass may alter the pancreatic parenchymatic aspect as well and may add to the necessity of an experienced and well trained EUS examiner.
Since a “gold standard” for diagnosing a chronic pancreatitis by noninvasive means does not exist, histologic tissue diagnosis must still be regarded as the best reference available.
To evaluate the diagnostic accuracy of EUS-FNA and to evaluate factors influencing diagnostic accuracy, we review our initial experience with EUS-FNA in a consecutive series of 142 patients that underwent EUS guided FNA in our institution and compare the results to present day literature.
One hundred forty two consecutive patients underwent 155 EUSFNA procedures between January 1999 and December 2001 in the Department of Gastroenterology of the Luebeck University Hospital. These patients were referred for EUS examination for suspicion of an upper gastrointestinal or pancreatic disease. EUS was performed after an overnight fasting period > 10 hours. All procedures were taken under conscious sedation utilizing Midazolam, Buscopan, Disoprivan and (or) Ketanest-S after informed consent was obtained. Patients were evaluated initially by transabdominal ultrasound and endoscopy of the upper gastrointestinal tract. EUS was performed by using PENTAX FG-36UX linear echoendoscope with EUS-525 Ultrasound Scanner (Hitachi Medical Systems, GmbH, Germany). Either the HITACHI (Hitachi Medical Systems, GmbH, Germany) 22G, 0.7 mm catheter or the Medi-Globe (Medi-Globe GmbH, Germany), 22G, 0.7 mm catheters were used. If tissue was to be obtained transgastrically, HITACHI (Hitachi Medical Systems GmbH, Germany) 19G, 1.1 mm catheter was employed. Color Doppler was used to exclude any vessel in the path of the needle before the lesions were punctured under constant ultrasound guidance.
During the FNA procedure, the catheter was inserted into the biopsy channel and the needle tip was then advanced incrementally under real-time EUS control until the tip was seen within the lesion.
For pancreatic head or neck lesions, the ultrasound transducer was placed in the duodenal bulb. For lesions of the pancreatic body or tail and celiac lymph nodes, the transducer was placed in the proximal stomach. The stylet was removed and suction applied through a 10 ml syringes as the needle was moved 2 to 5 within the lesion. The needle was then retracted into the catheter sheth and the entire catheter was removed. The aspirate was placed on glass slides for cytology smears and air dryed. Specimens obtained were transported to pathology for cytologic evaluation in the aftermath. An average of 2 needle passes was performed and tissue aspirated. Total time of EUS and EUS guided FNA varied from 20 to 40 min.
Up to 36 months of clinical follow-up was reviewed for this study in cases with negative or inconclusive FNA diagnosis. Final diagnosis of lesions was based on conclusive results of surgery, autopsy or clinical follow-up as adopted gold standard. EUS-FNA results and clinical data were analyzed in a retrospective setting.
A total of 142 patients underwent 155 EUS-FNA procedures. There were 50 women aged of 60.6 ± 11.5 yrs (mean ± SD) ant average and 92 men with mean age of 57.4 ± 12.0 yrs (mean ± SD). 133 patients (93.7%) underwent one EUS-FNA intervention, 9 patients (6.3%) underwent EUS guided FNA twice to improve specimen quality. Complications secondary to EUS guided FNA were recorded as follows: 13 patients (9.2%) suffered from mild complications, such as abdominal pain, fever, controlled by conservative measures. Two major complication occurred (1.4%), one upper GI bleeding and one gastrointestinal perforation followed by surgical intervention.
EUS guided FNA was indicated because of search for cancer with unknown primary tumor (CUP syndrome) in 78 cases (55%), staging of a cancer already diagnosed (26 cases, 19%), differential diagnostics in inflammatory or benign pancreatic disease for 19 cases (13%), unclear or other in 13% of the cases (19 patients).
EUS imaging found tumors ranging in size from <10 mm, 2 cases (1.4%), <20 mm, 22 cases (15.5%), <30 mm, 9 cases (6%), <40 mm, 30 cases (21.1%), <50 mm, 15 cases (11%), >50 mm, 7 cases (5%). In 31 cases no data about the tumor size was recorded.
According to the cytological report of obtained FNA specimens, 52 tumors were found to be of malignant origin (37%), 70 cases (49%) identified benign tumor or focal inflammation of pancreas. In 20 cases (14%) a definite diagnosis was not established (Table 1).
Table 1: Cytology of FNA.
After follow-up (36 months) in a consumptive view of all existing clinical data, final diagnosis established malignancy in 59 cases (42%), benign disease in 60 cases (42%), while 10 cases (7%) remained without definite diagnosis after 36 months follow-up and 13 patients (9%) were lost to follow-up and could not be settled in later work up.
Table 2 shows the high congruence of diagnosis established by EUS guided FNA with diagnoses established by consumptive view taking all clinical information and follow-up in account.
Table 2: Comparing EUS-FNA with final diagnosis. *Chi-Square test cannot analyze when a column has no value.
Numerous studies have shown that EUS is the single most accurate test for imaging and staging of pancreatic tumors [21-23]. In the current study, we report our experience of 142 consecutive patients that underwent EUS-FNA in a total of 155 times from January 1999 to December 2001 and compare our results and implications to present day data.
Exact preoperative staging of patients with pancreatic neoplasms is decisive with regard to patient survival [24-26]. Advanced tumor staging and declined physical status inversely correlate with patient survival, which has basically remained unchanged during the past decade despite medical progress.
In the differential diagnosis of pancreatic mass lesions EUS guided FNA offers a high yield regarding malignancy with regard to sensitivity, specificity as well as positive and negative predictive values (Table 3).
|Detecting pancreatic cancer|
Table 3: Sensitivities, specificities, NPV and PPV of EUS-FNA. NPV: Negative Predictive Value; PPV: Positive Predictive Value.
In our study, EUS-FNA was indicated for the overwhelming part, because malignant disease was suspected clinically or by prior diagnostics. Sensitivity and specificity of EUS-FNA for detecting pancreatic cancer was 87.3% and 95.1% respectively. This accuracy coincides with results of other study groups [6,27]. Most of the pancreatic mass lesions detected by EUS in our study were less than 50 mm (>75%) in diameter and were frequently undetected or postinterventionally confirmed by other image donating diagnostics such as CT scan or MRT. Current sensitivities (up to 95%) and specificities (100%) of EUS guided FNA are described in various studies and support EUS guided FNA as the method of choice for defining the histologic character of pancreatic mass lesions [28-30].
In a clinical setting pancreatic mass lesions could be evaluated evaluated by a multitude of invasive and noninvasive diagnostic measures. The optimal, meaning most successful, with regard to the definite final diagnosis, precise, considering preoperative planning, and least harmful, with regard to periinterventional complications, diagnostic pathway is yet to be established and will, however, have be adopted to each individual patient.
With regard to diagnostic accuracy some studies have shown that EUS alone is of comparable [31,32] or even greater [23,33,34] accuracy to CT scan plus ERCP with regard to the staging of pancreatic lesions. Decision making with regard to the diagnostic approach to pancreatic mass lesions, has been greatly facilitated and a straightforward diagnostic approach was recommended by recent guidelines . CT scan is here described as second diagnostic step in the work-up following transabdominal ultrasonography. EUS and EUS guided FNA was recommended to be employed in succession, following picture donating radiology. However, at times, we favor EUS and obtaining material via FNA in succession of transabdominal ultrasonography, prior to CT or MRI scans, if necessary at all, when a primarily palliative setting appears. Therefore we advocate performing EUS guided FNA as early as possible in the diagnostic and staging work-up of solid pancreatic masses. This aspect is in line with growing evidence with regard to establishing protocols in neoadjuvant chemotherapy settings especially in locally advanced pancreatic cancer, for which a definite histological diagnosis must be achieved.
The greatest portion of patients in our study was found to be diagnosed with a chronic pancreatitis (40 cases, 28.2%) by EUS and EUS-FNA. Several studies have shown that EUS is the most sensitive method in the diagnostic work-up of chronic pancreatitis . Lees described endosonographic criteria to differentiate chronic pancreatitis with regard to parenchymatic (six criteria) and ductal signs (eight criteria) . What however, should be considered the “gold standard” to which EUS is compared? Possible gold candidates include histology cytology, pancreatography and pancreatic function tests. Traditionally, pancreatography, via endoscopic retrograde access (ERP), was accepted as the most accurate, nonpathologic imaging method with regard to chronic pancreatitis, though yielding a substantial periinterventional risk for the patient. It was therefore replaced by MRI and MRCP which renders an even higher diagnostic accuracy compared to ERP . EUS is known to produce high diagnostic values with regard to chronic pancreatitis [39,40] and is recommanded be used complementary with MRI techniques in this matter .
A diagnostic challenge represents the differentiation of focal chronic pancreatitis from neoplastic processes of the pancreas. With regard to this, pathologic work-up is necessary. Brand et al. prospectively evaluated focal pancreatic lesions and found that 13 of 34 patients having focal, chronic pancreatitis simulating tumor had no other associated diffuse parenchymal alteration usually present in chronic pancreatitis . On the other hand, focal pancreatitis is frequently found in patients undergoing pancreatic resection for presumed cancer. However, chronic pancreatitis is considered a precancerous condition as a significantly increased risk of pancreatic cancer in all patients with chronic pancreatitis of any cause  is known. Malignant transformation in the condition of a chronic pancreatitis cannot be ruled out by EUS or EUS guided FNA considering false negative aspiration cytologies. This rate could be reduced by 10-15% via on-site cytopathologic examination of the specimens [44,45], but availability of a trained cytopathologist for on-site examination remains scarce. The dilemma appears unchanged and harbors a relevant and at times detrimental risk for the patient. No known diagnostic measure, including CT/MRI+MRCP, yields sufficient diagnostic safety in this matter. All necessary diagnostic means, including exploratory laparoscopy/laparotomy should, if in doubt, be employed and/or a definite follow-up, for instance (3) 6-12 months intervalls, be established  if reasonable.
EUS guided FNA is a safe diagnostic tool with a low complication rate conflicted to it. In our series of 142 patients, 13 (9.2%) patients experienced mild complications such as post-FNA fever or abdominal pain which ceased within two days, treated conservatively. One (0.7%) patient suffered from an upper GI bleeding after FNA, one other (0.7%) patient experienced a perforation and had to undergo surgery. Perforation is an inert risk conflicted to the employment of endoscopic techniques. Traditionally, surgical intervention, by laparoscopic or laparotomic approach, was the only choice with regard to damage control and reconstitution of gastrointestinal integrity. Recently, the development and widespread use of over the scope clipping systems, may constitute a valuable and effective alternative in a situation such as we have experienced. This technique may help to avoid a surgical approach with regard to endoscopic complication management .
Currently, rates of adverse events conflicted be EUS guided FNA of pancreatic masses are described to range from 0.5-2% including risk of pancreatitis and bleeding [48-50]. Discomfort, as described after FNA in our study by approximately 10% of the cases, was rarely registered in current studies. We have noted no pancreatitis in the aftermath of an EUS guided FNA. Rates for relevant complications range within currently described limits, but have to be considered to be a constant reminder for the need for constant endoscopic training and keen awareness the regarding potential detrimental risks immanent to EUS and EUS guided FNA [51,52].
We found no evidence of needle-tract or cutaneous malignant seeding having taken place by EUS guided FNA in the histopathologic specimens. The puncture sites located in the duodenum were resected during the following surgery. Which appears to be reflected in current findings, as being restricted to single cases [48,49]. No fatal complication was recorded following interventions described in this series.
In conclusion, we have demonstrated the diagnostic value of EUS guided FNA in our single center experience originating from the beginning of the EUS era. We have found diagnostic specificities, sensitivities and rates of adverse events that are compatible to current day data and therefore support the value of EUS and EUS guided FNA in the work up of pancreatic mass lesions at present.
The method itself has proved to have stepped beyond experimental stage in the past decade and is of highest value in the hands of an experienced examiner.
However, as controversies about the optimal diagnostic pathway regarding the work up of pancreatic mass lesions appear to have settled recently, we advocate the use of EUS and EUS guided FNA at times even sooner as recommended in the diagnostic work- up.
Securely detecting or ruling out malignant transformation in patients presenting with chronic pancreatitis and its complications remains a be to be constant challenge to the treating gastroenterologist. This dilemma has not changed since the acquisition of our data to the present day and may only be solved by careful, precise diagnostics and determined medical follow-up.
Make the best use of Scientific Research and information from our 700 + peer reviewed, Open Access Journals