- Email: [email protected]
Endoscopic ultrasound-guided fine needle aspiration and biopsy using a 22-gauge needle with side fenestration in pancreatic cystic lesions
Digestive and Liver Disease 46 (2014) 45–50
Contents lists available at ScienceDirect
Digestive and Liver Disease journal homepage: www.elsevier.com/locate/dld
Digestive Endoscopy
Endoscopic ultrasound-guided fine needle aspiration and biopsy using a 22-gauge needle with side fenestration in pancreatic cystic lesions Luca Barresi a,∗ , Ilaria Tarantino a , Mario Traina a , Antonino Granata a , Gabriele Curcio a , Neville Azzopardi a , Paola Baccarini b , Rosa Liotta c , Adele Fornelli b , Antonella Maimone d , Elio Jovine e , Vincenzo Cennamo d , Carlo Fabbri d a
Gastroenterology and Endoscopy Unit, ISMETT, Palermo, Italy Pathology Unit, AUSL Bologna Bellaria-Maggiore Hospital, Bologna, Italy c Pathology Unit, ISMETT, Palermo, Italy d Unit of Gastroenterology and Digestive Endoscopy, AUSL Bologna Bellaria-Maggiore Hospital, Bologna, Italy e Department of General Surgery, AUSL Bologna Bellaria-Maggiore Hospital, Bologna, Italy b
a r t i c l e
i n f o
Article history: Received 8 February 2013 Accepted 28 June 2013 Available online 31 July 2013 Keywords: Endosonography EUS-FNA Pancreatic cyst
a b s t r a c t Background: Cytologic diagnosis by endoscopic ultrasound-guided fine needle aspiration is associated with low sensitivity and adequacy. A newly designed endoscopic ultrasound-guided fine needle biopsy device, endowed with a side fenestration, is now available. Aims: We carried out a study with the aim of evaluating the feasibility, safety, and diagnostic yield of the 22-gauge needle with side fenestration for endoscopic ultrasound fine needle aspiration and biopsy of pancreatic cystic lesions. Methods: 58 patients with 60 pancreatic cystic lesions consecutively referred for endoscopic ultrasound guided-fine needle aspiration were enrolled in a prospective, dual centre study, and underwent fine needle aspiration and biopsy with the 22-gauge needle with side fenestration. Results: Fine needle aspiration and biopsy was technically feasible in all cases. In 39/60 (65%) pancreatic cystic lesions, the specimens were adequate for cyto-histologic assessment. In lesions with solid components, and in malignant lesions, adequacy was 94.4% (p = 0.0149) and 100% (p = 0.0069), respectively. Samples were adequate for histologic evaluation in 18/39 (46.1%) cases. There were only 2 (3.3%) mild complications. Conclusions: Fine needle aspiration and biopsy with the 22-gauge needle with side fenestration is feasible, and superior to conventional endoscopic ultrasound-guided fine needle aspiration cytology from cystic fluid, particularly in pancreatic cystic lesions with solid component or malignancy, with a higher diagnostic yield and with no increase in complication rate. © 2013 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Ltd. All rights reserved.
1. Introduction Pancreatic cystic lesions (PCLs) are increasingly found in clinical practice because of the widespread use of imaging modalities [1–3]. Pancreatic cysts are a complex group of benign, premalignant and malignant lesions with diverse clinical, radiologic and pathologic features [4]. However, diagnosis and classification remain a challenge despite advances in diagnostic tools. Radiologic and endosonographic morphologic features alone are not accurate enough for arriving at a final diagnosis in all cases [5–8]. However endoscopic ultrasound (EUS) has the advantage of
∗ Corresponding author at: Gastroenterology and Endoscopy Unit, ISMETT, Via Tricomi 1, Palermo 90127, Italy. Tel.: +39 331 1718159; fax: +39 091 2192400. E-mail address: [email protected] (L. Barresi).
allowing fine needle aspiration (FNA), with cystic fluid analysis and cytological characterization. In cystic fluid, the most widely used test to distinguish mucinous from non-mucinous lesions is carcinoembryonic antigen (CEA) and more recently K-ras mutation, but the accuracy of these tests is only moderate [8–10], and neither test is useful in identifying malignant lesions [8,10]. Pathologic analysis is the most reliable diagnostic test, but while cytology obtained by EUS-FNA from cystic fluid has a high specificity for identification of malignancy and PCL type, it has a low sensitivity [8,9], due to the small number of cells present in the aspirate, particularly in small PCLs [11]. Several studies have explored the possibility of obtaining a better cyto-histological sample with new devices such as the Echobrush (Cook Endoscopy, WinstonSalem, NC, USA) [12–14], or new techniques of FNA like cystic wall punctures (CWP) using a standard EUS-FNA needle [11,15] or
1590-8658/$36.00 © 2013 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.dld.2013.06.008
46
L. Barresi et al. / Digestive and Liver Disease 46 (2014) 45–50
EUS-trucut biopsy needle (Quick-Core; Cook Endoscopy, WinstonSalem, NC, USA) [16]. A newly designed EUS-guided fine needle biopsy was released on the market, featuring ProCore reverse bevel technology (EchoTip ProCore High Definition Ultrasound Biopsy Needle, Cook Endoscopy Inc., Limerick, Ireland), and equipped with a side fenestration that is 2 mm in length for the 22-gauge needle, and in which the tissue sample is embedded during puncture. With this new device, histological samples of the target organs can be obtained effectively and safely in different solid lesions [17,18]. The aim of this study was to evaluate the feasibility, safety, and diagnostic yield of this newly developed needle in pancreatic cystic lesions. 2. Patients and methods
Fig. 1. (a) The 22-gauge needle with the side fenestration, 2 mm in length; with stylet and (b) without stylet.
2.1. Patient inclusion and exclusion criteria All patients with PCLs observed at two referral centres (Gastroenterology and Endoscopy Unit, ISMETT, Palermo and Unit of Gastroenterology and Digestive Endoscopy, AUSL Bologna BellariaMaggiore Hospital, Bologna) were evaluated for inclusion in this prospective, observational study. All patients underwent abdominal MRI or if contraindicated, CT scan. Patients with a clear diagnosis of pancreatic pseudocysts (round, non-septated lesions, presence of internal debris, recent acute pancreatitis or radiologic imaging of chronic pancreatitis) were excluded from the study. Inclusion criteria were: (1) PCLs in patients fit for surgery or where cyto-histological diagnosis was requested for oncologic treatments, (2) PCLs ≥1.5 cm, and (3) PCLs of any size with worrisome features (thickened enhanced cyst walls, non-enhanced mural nodules, mean pancreatic duct (MPD) size of 5–9 mm, abrupt change in the MPD calibre with distal pancreatic atrophy, and lymphadenopathy). Exclusion criteria included: (1) patients <18-years old, (2) Severe haemorrhagic risk (platelet count <50,000 per L or prothrombin activity <50%), and (3) interposed vessels or EUSprocedure considered risky by endosonographer. A PCL diameter of 1.5 cm was considered the minimum size required to obtain useful information for diagnosis from FNA and cystic fluid analysis, as previously described [19]. Written informed consent was obtained from all patients and the study protocol was approved by the institutional review board. A dedicated database was created to record clinical characteristics of patients, endoscopy reports, cystic fluid analysis, pathologic results, surgery and surgical diagnosis and follow-up. 2.2. EUS-guided FNA/B techniques All patients underwent standard sedation with propofol under anesthesiologic monitoring. Prophylactic antibiotics were administered intravenously once before the procedure (piperacillin–tazobactam 4.5 mg or ciprofloxacin 400 mg) and orally (amoxicillin–clavulanic acid 1 g or ciprofloxacin 500 mg twice a day) for 5 days after the procedure. The exams were done by three experienced endosonographers (LB, IT, CF), all with at least 5 years of experience in pancreatic EUS (about 3000 EUS each). The instruments used were linear echoendoscopes (Olympus UCT-140; Olympus America, Inc., Melville, NY or Fujinon, Inc., Saitama, Japan). A cytopathologist experienced in pancreatic cytology was present in the EUS room for rapid on site evaluation during the exams. The morphologic evaluation of the PCLs and evaluation of findings suspicious for malignancy were done during the EUS procedure.
Presence of solid components was defined at EUS as the presence of a solid cystic wall nodule, pericystic solid lesions, thickened wall or intracystic septa. All these aspects are associated with an increased risk of malignancy and so were considered together for the analysis. The PCL puncture was made with the 22-gauge, Echotip Ultra FNB needle, featuring ProCore reverse bevel technology, with side fenestration (Cook Medical, Ireland) (Fig. 1a and b). Because of the particular characteristics of this needle we called our technique fine needle aspiration and biopsy (FNA/B). The technique used was the so-called cystic wall puncture (CWP), previously described in the literature [15], and modified for our study based on the characteristics of the new needle. After the target lesion was endosonographically visualized, and under colour doppler control, FNA/B was performed either from the duodenum or from the stomach. The needle was advanced into the PCL under EUS guidance. Once the lesion was penetrated, the stylet was removed (Fig. 2a), and suction was applied using a 10-mL syringe to aspirate all the intra-cystic liquid. If at least 1 ml of intracystic fluid was obtained, a sample for CEA and amylase was first sent and if more fluid was available another 0.5 ml was sent for verification of K-ras mutation. The molecular analysis of K-ras oncogene (exon 2 and exon 3) was done by direct sequencing, using 3500 Genetic Analyzer (Applied Biosystems). A CEA value of ≥192 ng/ml was considered suggestive of mucinous PCLs [8]. Any additional fluid aspirated was sent for cytology. When the cystic walls collapsed on the needle (Fig. 2b), aspiration was continued for about 30 s to allow cystic wall tissue to enter the side fenestration of the needle. The needle was then moved 3–4 times about 5–10 mm back and forth within the lesion. If intracystic septae were present and the fluid was impossible to aspirate due to its density, we performed the FNA/B on the septae; otherwise we performed a smear with the few drops of cystic fluid aspirated for cytologic evaluation. If there was a solid component, FNA/B was performed directly on it by inserting the needle and leaving the aspiration open for 30 s and then moving 3–4 times about 5–10 mm back and forth within the lesion. Thereafter we waited for on-site evaluation by the pathologist to assess sample adequacy. If the sample was considered adequate the procedure was stopped; otherwise we performed further FNA/B on the solid component. 2.3. Cyto-histological analysis For each PCL, two or three smears were stained with Hemacolor or Papanicolau for cytological analysis. Small visible pieces were stored in formalin. Finally we pushed the stylet through the
L. Barresi et al. / Digestive and Liver Disease 46 (2014) 45–50
47
Fig. 2. (a) Needle inserted in pancreatic cystic lesion before aspiration of cystic fluid. Note the side fenestration. (b) Collapsed cystic walls on the needle after aspiration of cystic fluid.
needle and then flushed the needle with air or saline solution to expel the sample into formalin. Samples were embedded in paraffin, and serial sections were stained with haematoxylin–eosin for histological evaluation and various immunohistochemical analyses. The liquid cell suspension was centrifuged and embedded in agarose gel for cell-block, or processed for cytospin when the liquid cell suspension was very clear. The cyst was categorized by the cytopathologist as mucinous if there was identifiable epithelium with characteristics consistent with mucin-producing cyst epithelium rather than gastric or duodenal epithelium, as reported in other studies [20,21]. Material from FNA/B is rarely sufficient or characteristic enough to distinguish between mucinous cystic neoplasm (MCN) and intraductal papillary mucinous neoplasm (IPMN), so we decided to report the results of mucinous cysts generically. Serous cystoadenoma (SC) was characterized by bland cuboidal cells that stained positively for periodic acid-Schiff (PAS), diastase sensitive, confirming the presence of cytoplasm glycogen. The cells present in the fluid of pseudocysts are chronic inflammatory cells, histiocytes, and hemosiderin-laden foamy macrophages, and, by definition, should not contain any cyst-lining epithelial cells [21]. When possible, and clinically or histologically indicated, immunohistochemical analysis was done on sections from samples embedded in paraffin, to evaluate specific PCLs such as cystic neuroendocrine tumour (NET), solid pseudopapillary pancreatic tumour (SPPT) or acinar cell pancreatic carcinoma (ACC) with cystic aspect. NET and SPPT were classified as malignant tumours according to the 4th edition of “World Health Organization Classification of Tumours of the Digestive System” published in 2010. Due to the characteristics of the ProCore needle and the technique that we used in PCLs, we evaluated cytologic and histologic adequacy together because it was impossible to distinguish the specimens derived from the cystic fluid from those derived from the cystic wall. As a result, we decided to call this technique FNA/B. The pathologist attempted to identify tissue core sample, or microbiopsy [11,22] with specimens available for full histological evaluation when they contained a coherent tissue specimen from the target organ [22,23]. The histological samples contained enough tissue for identification of the type of PCLs and made it possible to identify the grade of dysplasia, allowed the possibility of performing multiple sections, which were then used for special stains and immunohistochemical evaluation, and, in some cases, made it possible to identify stromal infiltration (such as in invasive malignancy).
2.4. Definition of gold-standard In patients undergoing surgical resection, final diagnosis was based on surgical pathology. In patients not undergoing surgical resection, final diagnosis took into account all information available from demography, clinical evaluation, imaging techniques, intracystic fluid analysis and cyto-histology from FNA/B. All nonoperated patients were followed up as out-patients with 3-monthly visits and MRI or CT scan and/or EUS every 3–6 months (according to PCLs size) to monitor PCLs change in size or appearance of new symptoms according to the protocol of the international consensus guidelines 2012 [24]. However, PCLs without surgical or FNA/B cyto-histological diagnosis were considered indeterminate lesions. Malignancy was excluded if after 1 year follow-up there were no PCL changes or new symptoms.
2.5. Assessment of complications Patients had baseline haemoglobin and amylase/lipase levels measured. They were observed for 6 h after EUS-FNA/B and evaluated for pain, vomiting, fever, bleeding or any other new symptoms. At 6 h, a blood sample was sent to the laboratory for measurement of haemoglobin, amylase and lipase. A drop in haemoglobin of >2 g/dl was considered to be significant, and post-FNA/B pancreatitis was defined as abdominal pain and an increase in amylase/lipase >3 times the upper limit of normal. Fever was defined as a temperature >38 ◦ C. A telephone interview was done the day after the procedure and at 1, 2 and 4 weeks. Complications were defined as early if they occurred within 6 h and late if they occurred after 6 h. When the cyto-histological results were available, the patients were seen in the out-patient clinic to explain the results and plan further management.
3. Statistical analysis Continuous variables were summarized as mean and standard deviation, or range. Categorical variables were summarized as frequency and percentage. Differences in proportions were compared using the z-test. Assuming that the adequacy of FNA with cytology from cystic fluid is 31% [25] and assuming an improved adequacy of FNA/B with needle with side fenestration of 20%, a sample size of 58 patients is required to achieve 90% power. Statistical tests were
48
L. Barresi et al. / Digestive and Liver Disease 46 (2014) 45–50 Table 3 Fine needle aspiration and biopsy cytohistologic diagnosis.
Table 1 Patient (n = 58) and pancreatic cystic lesion (n = 60) characteristics. Age: years (mean ± SD) Gender: M Symptoms Major cyst diameter Cyst location Head Body Tail Uncinate EUS aspects Multilocular Unilocular Microcystic Multifocality Presence of solid component
64 ± 13.7 34.5% 8 (13.7%) Mean 30.5 mm (range 10–70) 29 (48.3%) 19 (31.6%) 7 (11.6%) 5 (8.3%) 35 (58.3%) 20 (33.3%) 5 (5.6%) 24 (40%) 18 (30%)
considered significant when the corresponding p value was <5%. SPSS 17 software was used to carry out the statistical calculations. 4. Results From May 2011 to October 2012, 58 consecutive patients with a total of 60 PCLs, underwent EUS-FNA/B and were prospectively enrolled in the study. Mean follow-up was 11 ± 4.2 months. The patients’ baseline characteristics and the characteristics of the cysts are shown in Table 1. Fifty (86.3%) patients were asymptomatic, 5 (8.6%) had a history of acute pancreatitis, and 3 (5.1%) a history of chronic abdominal pain. The mean number of needle passages was 1.25 (range 1–2), and EUS-FNA/B was technically feasible in all cases (we performed more than one needle passage only in PCLs with solid components in order to obtain an adequate sample). An adequate amount of fluid was aspirated for CEA and amylase in 32 (53.3%) patients, and for K-ras in 24 (40%). In mucinous PCLs mean CEA level was 2127 ng/ml and mean amylase 19650 U/L; in non-mucinous PCLs mean CEA was 0.8 ng/ml and mean amylase 125 U/L. There were 9/24 (37.5%) patients positive for mutation of K-ras. Cystic fluid evaluation for CEA and K-ras mutation in mucinous cystic lesions are shown in Table 2. The most frequent reason for low fluid aspiration (insufficient for evaluation of intracystic CEA and/or K-ras) was the presence of intracystic fluid that was too viscous, and impossible to aspirate in 12 (20%) patients. The other reasons were adequate material for cyto-histological evaluation at the first passage of the needle on solid lesions, obviating the need for a second needle-passage, or small quantity of intracystic fluid, fluid too viscous to be tested by the laboratory; and the absence in the fluid of cells for evaluation of K-ras mutation. Table 2 Cystic fluid evaluation for carcinoembryonic antigen and K-ras mutation in pancreatic mucinous cystic lesions. CEA ≥ 192 ng/ml
CEA ≤ 192 ng/ml
Unavailable
9 (32.1%)
7 (26.9%)
13 (44.8%)
4 (44.4%)
2 (22.2%)
3 (33.3%)
K-ras mutation
Mucinous PCLs (n = 29) Mucinous malignant cysts (n = 9)
21 (35%) 20 (33.3%)a 6 (10%) 2 (3.3%) 3 (5%) 1 (1.6%) 7 (11.6%)b
PCLs, pancreatic cystic lesions. a Three intraductal papillary mucinous neoplasms with mild dysplasia at FNA/B were carcinoma in situ (n = 1) and invasive carcinoma (n = 2). b Three serous cystoadenoma, three pseudocysts, one pancreatic tuberculosis.
F, female; M, male; EUS, endoscopic ultrasound.
Mucinous PCLs (n = 29) Mucinous malignant cysts (n = 9)
Inadequate specimens Mucinous cysts with no malignancy Mucinous carcinoma (in situ or invasive) Ductal adenocarcinoma with cystic degeneration Neuroendocrine cystic tumour Solid pseudopapillary tumour Benign PCLs
Present
Absent
Unavailable
5 (17.8%) 1 (11.1%)
7 (25%) 4 (44.4%)
17 (58.6%) 4 (44.4%)
PCLs, pancreatic cystic lesions; CEA, carcinoembryonic antigen; K-ras, V-Ki-ras2 Kirsten rat sarcoma viral oncogene.
The overall sample adequacy for cyto-histological diagnosis was 39/60 (65%) PCLs (Fig. 3). The cyto-histological results of FNA/B are shown in Table 3. In PCLs with solid components the cellular material was adequate for cyto-histological evaluation in 17/18 (94.4%) lesions. The difference between the adequacy for cyto-histological diagnosis of all samples and only those of lesions with solid component was statistically significant (p = 0.0149). There were 15 (25%) malignant PCLs: 9 degenerated IPMN, 2 adenocarcinomas, 3 NET, and 1 SPPT. Ten (66.6%) malignant cysts were observed in PCLs ≤30 mm, 6 (40%) in ≤20 mm and 2 (13.3%) in ≤15 mm. In the malignant group, adequacy for cyto-histological evaluation was 15/15 (100%), though in 3 lesions (20%) the FNA/B underrated the malignancy, revealing mucinous epithelium with mild atypia, instead of 1 carcinoma in situ and 2 invasive carcinomas in IPMN observed in the histological samples from surgery. Worthy of note is that 2 of these 3 lesions were without solid component, 2 with CEA <192 ng/ml and 1 in which CEA was not available. The K-ras mutation was unavailable in two lesions and negative in the other lesion. The difference between adequacy for cyto-histological diagnosis of all specimens and that of specimens from malignant lesion was statistically significant (p = 0.0069). In 18/39 (46.1%) adequate samples the material was judged adequate as a histological sample and in 4 malignant cases (three NET and one SPPT) the possibility of carrying out immunohistochemical testing was essential for the diagnosis. There were 17 (28.3%) patients who underwent surgical resection. The surgical pathology showed IPMN with mild/moderate dysplasia in 8 patients, IPMN with carcinoma (in situ or invasive) in 5, MCN with mild dysplasia in 1, NET in 2 and SPPT in 1. In 13/17 (76.4%) patients in this group, the FNA/B sample was adequate for diagnosis. The concordance between FNA/B diagnosis and surgical specimen regarding the type of lesions was 100%, but in 3 (17.6%) lesions FNA/B underrated the severity of dysplasia (the same 3 patients with malignant IPMN instead of IPMN with mild atypia). No false positive results of the FNA/B were observed in surgically treated patients. There were only two early complications (3.3%): an intracystic haemorrhage in a cystic NET that resolved spontaneously, and a fever that was treated with two days of hospitalization and intravenous antibiotics. No late complications were observed. 5. Discussion Diagnosis of PCLs remains a very difficult challenge for physicians and is usually achieved by combining demographic and clinical information, radiologic and EUS assessments, intracystic fluid analyses and cyto-histology. Our study showed that FNA/B with a 22-gauge ProCore needle was feasible in all cases, including in PCLs located in difficult positions like the pancreatic head and uncinate process, and had an overall high diagnostic yield for cyto-histological sampling (65%).
L. Barresi et al. / Digestive and Liver Disease 46 (2014) 45–50
49
Fig. 3. (a) Neuroendocrine tumour (haematoxylin–eosin stain). (b) Immunohistochemistry of neuroendocrine tumour positive for synaptophysin. The test needs an adequate amount of cells to be performed. (c) Strips of mucinous epithelium with mild dysplasia. The preserved tissue architecture of the targeted lesions allows a more precise grading of dysplasia. (d) Severe atypical cells infiltrating the stroma.
It is particularly interesting to note that in the subgroup analysis of malignant lesions and lesions with a solid component, which are the most likely to harbour malignancy, the adequacy for cytohistological diagnosis was 100% and 94.4%, respectively. However, in 3 patients with malignant lesions the FNA/B diagnosis correctly identified the lesions as mucinous, but underestimated the presence of malignancy. This is not surprising because it is known that PCL cyto-histology can underrate the final histological grade of neoplastic mucinous cyst, given that the cyst lining is quite heterogeneous, and the degree of epithelial atypia may not be representative of the highest degree of atypia of the cyst at histology [21]. However, a diagnosis of mucinous lesions was essential in referring these patients to surgery, especially because in two of these patients there was no solid component of the lesions, and the CEA and K-ras were either unavailable or negative for mucinous lesions. The needle with lateral fenestration produced a high percentage of samples adequate for histological evaluation (46.1% of adequate samples), with a more precise diagnosis and the possibility of employing special stains and immunohistochemistry. In particular, in four patients the possibility of having enough material to perform immunohistochemistry was essential to the diagnosis of three cystic pancreatic NETs and one SPPT. We found no increase in the rate of complications, with only two mild complications, which resolved rapidly without consequences. The recently published guidelines in Fukuoka (2012) [24] do not consider EUS-FNA as the first option for diagnosis and imaging techniques are still considered essential in the decision analysis. However, other studies have shown that both morphologic EUS [6–8] and radiologic imaging [5–26] have a probability of correct diagnosis in only about half of the cases. In PCLs, both CEA and K-ras are useful in discriminating mucinous from non-mucinous lesions, but are not useful in identifying malignant lesions [8,10]. In our study, both CEA and K-ras in the cystic fluid had marginal utility in
the final diagnosis because in both benign and malignant mucinous PCLs, CEA was often under the cut-off value for mucinous lesions and K-ras was negative or both tests were unavailable. The unavailability of CEA to define PCL has also been described in other studies [25]. Conversely several studies [21,27] have shown that EUS-FNA with evaluation of the presence of mucinous cells with high-grade atypia in cystic fluid adds value to risk assessment of malignancy in pancreatic mucinous cysts, including small (<3 cm) cysts. So cytology appears to be the most useful diagnostic test because it can establish the type of lesion under observation, and it also offers the possibility to diagnose underlying malignancy. However, cytology from cystic fluid aspirate has been shown to have low adequacy and sensitivity because of the relatively low cellularity of aspirated pancreatic cyst fluid. In the Cooperative Pancreatic Cystic Study, the sensitivity of cystic fluid cytology for diagnosing mucinous cysts was only 34% because of the low number of cells found in cystic fluid [8]. Problems with cytology in EUS-guided cyst aspiration were delineated in a prospective, dual centre study in which only 31% of the samples had adequate cellularity for analysis [25]. New tools and techniques for improving diagnostic yield, such as Echobrush and CWP with either the trucut needle or the standard needle, have been proposed. Echobrush has been tested in several studies with conflicting results [12–14] and a high rate of complications (8–10%), including one death [13,14]. A prospective study on a new technique for obtaining cytologic material from the cystic wall with a standard 22G needle was recently published by Hong [15]. The author reports an impressive 56/69 (81%) cellular adequacy for cytologic assessment from CWP in PCLs. The technique used for FNA/B in our study is another method for obtaining samples from cyst walls. Regardless of the technique or needle used, obtaining a cystic wall sample seems to be superior to conventional FNA cytological sampling from cystic fluid. We think it is advisable to carry out a comparative study
50
L. Barresi et al. / Digestive and Liver Disease 46 (2014) 45–50
between the needle with side fenestration using the technique described in our study and CWP with a standard 22G needle, as described by Hong in his study. Limitations of our study have to be considered. Since only 28.3% of our patients underwent surgery, malignant lesions may have been missed. Similar studies in the future will probably suffer from this limitation, due to a growing body of scientific literature that supports a strategy of surveillance for small asymptomatic cysts without risk factors. However, although, the relatively short follow up of this report cannot completely exclude malignancy, no patients have had modification of the PCLs or developed new symptoms. Another limitation is that we did not have a control group for comparison with traditional cytology from cystic fluid aspirate or with other techniques, such as CWP with a standard 22G needle. If we compare our results with the historical control of cytology from cyst fluid aspirate, we find a substantial (about twofold) improvement in adequacy with this new needle. In conclusion, this dual centre, prospective study on PCLs suggests that EUS-guided FNA/B with the 22-gauge needle with side fenestration is feasible and maybe superior to conventional FNA cytology from cystic fluid, with a higher diagnostic yield, especially in PCLs with solid component and malignant lesions, with no associated increase in the rate of complications. These results need to be confirmed in comparative studies. Conflict of interest statement None declared. References [1] Lee KS, Sekhar A, Rofsky NM, et al. Prevalence of incidental pancreatic cysts in the adult population on MR imaging. American Journal of Gastroenterology 2010;105:2079–84. [2] Laffan T, Horton K, Klein A, et al. Prevalence of unsuspected pancreatic cysts on MDCT. American Journal of Roentgenology 2008;191:802–7. [3] Spinelli KS, Fromwiller TE, Daniel RA, et al. Cystic pancreatic neoplasm: observe or operate. Annals of Surgery 2004;239:651–7. [4] Brugge WR, Lauwers GY, Sahani D, et al. Cystic neoplasm of the pancreas. New England Journal of Medicine 2004;351:1218–26. [5] Visser BC, Yeh BM, Qayyum A, et al. Characterization of cystic pancreatic masses: relative accuracy of CT and MRI. American Journal of Roentgenology 2007;189:648–56. [6] O’Toole D, Palazzo L, Hammel P, et al. Macrocystic pancreatic cystadenoma: the role of EUS and cyst fluid analysis in distinguishing mucinous and serous lesions. Gastrointestinal Endoscopy 2004 Jun;59: 823–9. [7] Kubo H, Nakamura K, Itaba S, et al. Differential diagnosis of cystic tumors of the pancreas by endoscopic ultrasonography. Endoscopy 2009;41:684–9.
[8] Brugge WR, Lewandrowski K, Lee-Lewandrowski E, et al. The diagnosis of pancreatic cystic neoplasms: a report of the Cooperative Pancreatic Cyst (CPC) study. Gastroenterology 2004;126:1330–6. [9] van der Waaij LA, van Dullemen HM, Porte RJ. Cyst fluid analysis in the differential diagnosis of cystic pancreatic lesions: a pooled analysis. Gastrointestinal Endoscopy 2005;62:383–9. [10] Khalid A, Zahid M, Finkelstein SD, et al. Pancreatic cyst fluid DNA analysis in evaluating pancreatic cysts: a report of the PANDA study. Gastrointestinal Endoscopy 2009;69:1095–102. [11] Frossard JL, Amouyal P, Amouyal G, et al. Performance of endosonographyguide fine needle aspiration and biopsy in the diagnosis of cystic pancreatic lesions. American Journal of Gastroenterology 2003;98:1516–24. [12] Al-Haddad M, Gill KRS, Raimondo M, et al. Safety and efficacy of cytology brushings versus standard fine-needle aspiration in evaluating cystic pancreatic lesions: a controlled study. Endoscopy 2010;42:127–32. [13] Thomas T, Bebb J, Mannath J, et al. EUS-guided pancreatic cyst brushing: a comparative study in a tertiary referral centre. JOP 2010;11:163–9. [14] Sendino O, Fernández-Esparrach G, Solé M, et al. Endoscopic ultrasonographyguided brushing increases cellular diagnosis of pancreatic cysts: a prospective study. Digestive and Liver Disease 2010;42:877–81. [15] Hong SK, Loren DE, Rogart JN, et al. Targeted cyst wall puncture and aspiration during EUS-FNA increases the diagnostic yield of premalignant and malignant pancreatic cysts. Gastrointestinal Endoscopy 2012;75:775–82. [16] Levy MJ, Smyrk TC, Reddy RP, et al. Endoscopic ultrasound-guided trucut biopsy of the cyst wall for diagnosing cystic pancreatic tumors. Clinical Gastroenterology and Hepatology 2005;3:974–9. [17] Iglesias-Garcia J, Poley JW, Larghi A, et al. Feasibility and yield of a new EUS histology needle: results from a multicenter, pooled, cohort study. Gastrointestinal Endoscopy 2011;73:1189–96. [18] Iwashita T, Nakai Y, Samarasena JB. High single-pass diagnostic yield of a new 25-gauge core biopsy needle for EUS-guided FNA biopsy in solid pancreatic lesions. Gastrointestinal Endoscopy 2013;77:909–15. [19] Walsh PM, Zuccaro G, Dumot JA, et al. Predicting success of endoscopic aspiration for suspected pancreatic cystic neoplasms. Journal of the Pancreas 2008;9:612–7. [20] Nagle JA, Wilbur DC, Pitman MB. Cytomorphology of gastric and duodenal epithelium and reactivity to B723. Diagnostic Cytopathology 2005;33:381–6. [21] Pitman MB, Lewandrowsky K, Shen J, et al. Pancreatic cysts: preoperative diagnosis and clinical management. Cancer Cytopathology 2010;118:1–13. [22] Fabre M. Diagnosis of cystic pancreatic lesions by endoscopic ultrasonographyguided fine-needle aspiration. Which sampling and why? Utility of cystic wall microbiopsy and fluid analysis by monolayer processing. Acta Endoscopy 2002;32:1–13. [23] Papanikolaou IS, Adler A, Wegener K. Prospective pilot evaluation of a new needle prototype for endoscopic ultrasonography-guided fine-needle aspiration: comparison of cytology and histology yield. European Journal of Gastroenterology and Hepatology 2008;20:342–8. [24] Tanaka M, Fernández-del Castillo C, Adsay V, Chari S. International consensus guidelines 2012 for management of IPMN and MCN of the pancreas. Pancreatology 2012;12:183–97. [25] de Jong K, Poley JW, van Hooft JE, et al. Endoscopic-ultrasound fine-needle aspiration of pancreatic cystic lesions provides inadequate material for cytology and laboratory analysis: initial results of a prospective study. Endoscopy 2011;43:585–90. [26] Correa-Gallego C, Ferrone C, Thayer SP, et al. Incidental pancreati cysts: do we really know what we are watching. Pancreatology 2010;10:144–50. [27] Genevay M, Mino-Kenudson M, Yaeger K, et al. Cytology adds value to imaging studies for risk assessment of malignancy in pancreatic mucinous cyst. Annals of Surgery 2011;254:977–83.
Contents lists available at ScienceDirect
Digestive and Liver Disease journal homepage: www.elsevier.com/locate/dld
Digestive Endoscopy
Endoscopic ultrasound-guided fine needle aspiration and biopsy using a 22-gauge needle with side fenestration in pancreatic cystic lesions Luca Barresi a,∗ , Ilaria Tarantino a , Mario Traina a , Antonino Granata a , Gabriele Curcio a , Neville Azzopardi a , Paola Baccarini b , Rosa Liotta c , Adele Fornelli b , Antonella Maimone d , Elio Jovine e , Vincenzo Cennamo d , Carlo Fabbri d a
Gastroenterology and Endoscopy Unit, ISMETT, Palermo, Italy Pathology Unit, AUSL Bologna Bellaria-Maggiore Hospital, Bologna, Italy c Pathology Unit, ISMETT, Palermo, Italy d Unit of Gastroenterology and Digestive Endoscopy, AUSL Bologna Bellaria-Maggiore Hospital, Bologna, Italy e Department of General Surgery, AUSL Bologna Bellaria-Maggiore Hospital, Bologna, Italy b
a r t i c l e
i n f o
Article history: Received 8 February 2013 Accepted 28 June 2013 Available online 31 July 2013 Keywords: Endosonography EUS-FNA Pancreatic cyst
a b s t r a c t Background: Cytologic diagnosis by endoscopic ultrasound-guided fine needle aspiration is associated with low sensitivity and adequacy. A newly designed endoscopic ultrasound-guided fine needle biopsy device, endowed with a side fenestration, is now available. Aims: We carried out a study with the aim of evaluating the feasibility, safety, and diagnostic yield of the 22-gauge needle with side fenestration for endoscopic ultrasound fine needle aspiration and biopsy of pancreatic cystic lesions. Methods: 58 patients with 60 pancreatic cystic lesions consecutively referred for endoscopic ultrasound guided-fine needle aspiration were enrolled in a prospective, dual centre study, and underwent fine needle aspiration and biopsy with the 22-gauge needle with side fenestration. Results: Fine needle aspiration and biopsy was technically feasible in all cases. In 39/60 (65%) pancreatic cystic lesions, the specimens were adequate for cyto-histologic assessment. In lesions with solid components, and in malignant lesions, adequacy was 94.4% (p = 0.0149) and 100% (p = 0.0069), respectively. Samples were adequate for histologic evaluation in 18/39 (46.1%) cases. There were only 2 (3.3%) mild complications. Conclusions: Fine needle aspiration and biopsy with the 22-gauge needle with side fenestration is feasible, and superior to conventional endoscopic ultrasound-guided fine needle aspiration cytology from cystic fluid, particularly in pancreatic cystic lesions with solid component or malignancy, with a higher diagnostic yield and with no increase in complication rate. © 2013 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Ltd. All rights reserved.
1. Introduction Pancreatic cystic lesions (PCLs) are increasingly found in clinical practice because of the widespread use of imaging modalities [1–3]. Pancreatic cysts are a complex group of benign, premalignant and malignant lesions with diverse clinical, radiologic and pathologic features [4]. However, diagnosis and classification remain a challenge despite advances in diagnostic tools. Radiologic and endosonographic morphologic features alone are not accurate enough for arriving at a final diagnosis in all cases [5–8]. However endoscopic ultrasound (EUS) has the advantage of
∗ Corresponding author at: Gastroenterology and Endoscopy Unit, ISMETT, Via Tricomi 1, Palermo 90127, Italy. Tel.: +39 331 1718159; fax: +39 091 2192400. E-mail address: [email protected] (L. Barresi).
allowing fine needle aspiration (FNA), with cystic fluid analysis and cytological characterization. In cystic fluid, the most widely used test to distinguish mucinous from non-mucinous lesions is carcinoembryonic antigen (CEA) and more recently K-ras mutation, but the accuracy of these tests is only moderate [8–10], and neither test is useful in identifying malignant lesions [8,10]. Pathologic analysis is the most reliable diagnostic test, but while cytology obtained by EUS-FNA from cystic fluid has a high specificity for identification of malignancy and PCL type, it has a low sensitivity [8,9], due to the small number of cells present in the aspirate, particularly in small PCLs [11]. Several studies have explored the possibility of obtaining a better cyto-histological sample with new devices such as the Echobrush (Cook Endoscopy, WinstonSalem, NC, USA) [12–14], or new techniques of FNA like cystic wall punctures (CWP) using a standard EUS-FNA needle [11,15] or
1590-8658/$36.00 © 2013 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.dld.2013.06.008
46
L. Barresi et al. / Digestive and Liver Disease 46 (2014) 45–50
EUS-trucut biopsy needle (Quick-Core; Cook Endoscopy, WinstonSalem, NC, USA) [16]. A newly designed EUS-guided fine needle biopsy was released on the market, featuring ProCore reverse bevel technology (EchoTip ProCore High Definition Ultrasound Biopsy Needle, Cook Endoscopy Inc., Limerick, Ireland), and equipped with a side fenestration that is 2 mm in length for the 22-gauge needle, and in which the tissue sample is embedded during puncture. With this new device, histological samples of the target organs can be obtained effectively and safely in different solid lesions [17,18]. The aim of this study was to evaluate the feasibility, safety, and diagnostic yield of this newly developed needle in pancreatic cystic lesions. 2. Patients and methods
Fig. 1. (a) The 22-gauge needle with the side fenestration, 2 mm in length; with stylet and (b) without stylet.
2.1. Patient inclusion and exclusion criteria All patients with PCLs observed at two referral centres (Gastroenterology and Endoscopy Unit, ISMETT, Palermo and Unit of Gastroenterology and Digestive Endoscopy, AUSL Bologna BellariaMaggiore Hospital, Bologna) were evaluated for inclusion in this prospective, observational study. All patients underwent abdominal MRI or if contraindicated, CT scan. Patients with a clear diagnosis of pancreatic pseudocysts (round, non-septated lesions, presence of internal debris, recent acute pancreatitis or radiologic imaging of chronic pancreatitis) were excluded from the study. Inclusion criteria were: (1) PCLs in patients fit for surgery or where cyto-histological diagnosis was requested for oncologic treatments, (2) PCLs ≥1.5 cm, and (3) PCLs of any size with worrisome features (thickened enhanced cyst walls, non-enhanced mural nodules, mean pancreatic duct (MPD) size of 5–9 mm, abrupt change in the MPD calibre with distal pancreatic atrophy, and lymphadenopathy). Exclusion criteria included: (1) patients <18-years old, (2) Severe haemorrhagic risk (platelet count <50,000 per L or prothrombin activity <50%), and (3) interposed vessels or EUSprocedure considered risky by endosonographer. A PCL diameter of 1.5 cm was considered the minimum size required to obtain useful information for diagnosis from FNA and cystic fluid analysis, as previously described [19]. Written informed consent was obtained from all patients and the study protocol was approved by the institutional review board. A dedicated database was created to record clinical characteristics of patients, endoscopy reports, cystic fluid analysis, pathologic results, surgery and surgical diagnosis and follow-up. 2.2. EUS-guided FNA/B techniques All patients underwent standard sedation with propofol under anesthesiologic monitoring. Prophylactic antibiotics were administered intravenously once before the procedure (piperacillin–tazobactam 4.5 mg or ciprofloxacin 400 mg) and orally (amoxicillin–clavulanic acid 1 g or ciprofloxacin 500 mg twice a day) for 5 days after the procedure. The exams were done by three experienced endosonographers (LB, IT, CF), all with at least 5 years of experience in pancreatic EUS (about 3000 EUS each). The instruments used were linear echoendoscopes (Olympus UCT-140; Olympus America, Inc., Melville, NY or Fujinon, Inc., Saitama, Japan). A cytopathologist experienced in pancreatic cytology was present in the EUS room for rapid on site evaluation during the exams. The morphologic evaluation of the PCLs and evaluation of findings suspicious for malignancy were done during the EUS procedure.
Presence of solid components was defined at EUS as the presence of a solid cystic wall nodule, pericystic solid lesions, thickened wall or intracystic septa. All these aspects are associated with an increased risk of malignancy and so were considered together for the analysis. The PCL puncture was made with the 22-gauge, Echotip Ultra FNB needle, featuring ProCore reverse bevel technology, with side fenestration (Cook Medical, Ireland) (Fig. 1a and b). Because of the particular characteristics of this needle we called our technique fine needle aspiration and biopsy (FNA/B). The technique used was the so-called cystic wall puncture (CWP), previously described in the literature [15], and modified for our study based on the characteristics of the new needle. After the target lesion was endosonographically visualized, and under colour doppler control, FNA/B was performed either from the duodenum or from the stomach. The needle was advanced into the PCL under EUS guidance. Once the lesion was penetrated, the stylet was removed (Fig. 2a), and suction was applied using a 10-mL syringe to aspirate all the intra-cystic liquid. If at least 1 ml of intracystic fluid was obtained, a sample for CEA and amylase was first sent and if more fluid was available another 0.5 ml was sent for verification of K-ras mutation. The molecular analysis of K-ras oncogene (exon 2 and exon 3) was done by direct sequencing, using 3500 Genetic Analyzer (Applied Biosystems). A CEA value of ≥192 ng/ml was considered suggestive of mucinous PCLs [8]. Any additional fluid aspirated was sent for cytology. When the cystic walls collapsed on the needle (Fig. 2b), aspiration was continued for about 30 s to allow cystic wall tissue to enter the side fenestration of the needle. The needle was then moved 3–4 times about 5–10 mm back and forth within the lesion. If intracystic septae were present and the fluid was impossible to aspirate due to its density, we performed the FNA/B on the septae; otherwise we performed a smear with the few drops of cystic fluid aspirated for cytologic evaluation. If there was a solid component, FNA/B was performed directly on it by inserting the needle and leaving the aspiration open for 30 s and then moving 3–4 times about 5–10 mm back and forth within the lesion. Thereafter we waited for on-site evaluation by the pathologist to assess sample adequacy. If the sample was considered adequate the procedure was stopped; otherwise we performed further FNA/B on the solid component. 2.3. Cyto-histological analysis For each PCL, two or three smears were stained with Hemacolor or Papanicolau for cytological analysis. Small visible pieces were stored in formalin. Finally we pushed the stylet through the
L. Barresi et al. / Digestive and Liver Disease 46 (2014) 45–50
47
Fig. 2. (a) Needle inserted in pancreatic cystic lesion before aspiration of cystic fluid. Note the side fenestration. (b) Collapsed cystic walls on the needle after aspiration of cystic fluid.
needle and then flushed the needle with air or saline solution to expel the sample into formalin. Samples were embedded in paraffin, and serial sections were stained with haematoxylin–eosin for histological evaluation and various immunohistochemical analyses. The liquid cell suspension was centrifuged and embedded in agarose gel for cell-block, or processed for cytospin when the liquid cell suspension was very clear. The cyst was categorized by the cytopathologist as mucinous if there was identifiable epithelium with characteristics consistent with mucin-producing cyst epithelium rather than gastric or duodenal epithelium, as reported in other studies [20,21]. Material from FNA/B is rarely sufficient or characteristic enough to distinguish between mucinous cystic neoplasm (MCN) and intraductal papillary mucinous neoplasm (IPMN), so we decided to report the results of mucinous cysts generically. Serous cystoadenoma (SC) was characterized by bland cuboidal cells that stained positively for periodic acid-Schiff (PAS), diastase sensitive, confirming the presence of cytoplasm glycogen. The cells present in the fluid of pseudocysts are chronic inflammatory cells, histiocytes, and hemosiderin-laden foamy macrophages, and, by definition, should not contain any cyst-lining epithelial cells [21]. When possible, and clinically or histologically indicated, immunohistochemical analysis was done on sections from samples embedded in paraffin, to evaluate specific PCLs such as cystic neuroendocrine tumour (NET), solid pseudopapillary pancreatic tumour (SPPT) or acinar cell pancreatic carcinoma (ACC) with cystic aspect. NET and SPPT were classified as malignant tumours according to the 4th edition of “World Health Organization Classification of Tumours of the Digestive System” published in 2010. Due to the characteristics of the ProCore needle and the technique that we used in PCLs, we evaluated cytologic and histologic adequacy together because it was impossible to distinguish the specimens derived from the cystic fluid from those derived from the cystic wall. As a result, we decided to call this technique FNA/B. The pathologist attempted to identify tissue core sample, or microbiopsy [11,22] with specimens available for full histological evaluation when they contained a coherent tissue specimen from the target organ [22,23]. The histological samples contained enough tissue for identification of the type of PCLs and made it possible to identify the grade of dysplasia, allowed the possibility of performing multiple sections, which were then used for special stains and immunohistochemical evaluation, and, in some cases, made it possible to identify stromal infiltration (such as in invasive malignancy).
2.4. Definition of gold-standard In patients undergoing surgical resection, final diagnosis was based on surgical pathology. In patients not undergoing surgical resection, final diagnosis took into account all information available from demography, clinical evaluation, imaging techniques, intracystic fluid analysis and cyto-histology from FNA/B. All nonoperated patients were followed up as out-patients with 3-monthly visits and MRI or CT scan and/or EUS every 3–6 months (according to PCLs size) to monitor PCLs change in size or appearance of new symptoms according to the protocol of the international consensus guidelines 2012 [24]. However, PCLs without surgical or FNA/B cyto-histological diagnosis were considered indeterminate lesions. Malignancy was excluded if after 1 year follow-up there were no PCL changes or new symptoms.
2.5. Assessment of complications Patients had baseline haemoglobin and amylase/lipase levels measured. They were observed for 6 h after EUS-FNA/B and evaluated for pain, vomiting, fever, bleeding or any other new symptoms. At 6 h, a blood sample was sent to the laboratory for measurement of haemoglobin, amylase and lipase. A drop in haemoglobin of >2 g/dl was considered to be significant, and post-FNA/B pancreatitis was defined as abdominal pain and an increase in amylase/lipase >3 times the upper limit of normal. Fever was defined as a temperature >38 ◦ C. A telephone interview was done the day after the procedure and at 1, 2 and 4 weeks. Complications were defined as early if they occurred within 6 h and late if they occurred after 6 h. When the cyto-histological results were available, the patients were seen in the out-patient clinic to explain the results and plan further management.
3. Statistical analysis Continuous variables were summarized as mean and standard deviation, or range. Categorical variables were summarized as frequency and percentage. Differences in proportions were compared using the z-test. Assuming that the adequacy of FNA with cytology from cystic fluid is 31% [25] and assuming an improved adequacy of FNA/B with needle with side fenestration of 20%, a sample size of 58 patients is required to achieve 90% power. Statistical tests were
48
L. Barresi et al. / Digestive and Liver Disease 46 (2014) 45–50 Table 3 Fine needle aspiration and biopsy cytohistologic diagnosis.
Table 1 Patient (n = 58) and pancreatic cystic lesion (n = 60) characteristics. Age: years (mean ± SD) Gender: M Symptoms Major cyst diameter Cyst location Head Body Tail Uncinate EUS aspects Multilocular Unilocular Microcystic Multifocality Presence of solid component
64 ± 13.7 34.5% 8 (13.7%) Mean 30.5 mm (range 10–70) 29 (48.3%) 19 (31.6%) 7 (11.6%) 5 (8.3%) 35 (58.3%) 20 (33.3%) 5 (5.6%) 24 (40%) 18 (30%)
considered significant when the corresponding p value was <5%. SPSS 17 software was used to carry out the statistical calculations. 4. Results From May 2011 to October 2012, 58 consecutive patients with a total of 60 PCLs, underwent EUS-FNA/B and were prospectively enrolled in the study. Mean follow-up was 11 ± 4.2 months. The patients’ baseline characteristics and the characteristics of the cysts are shown in Table 1. Fifty (86.3%) patients were asymptomatic, 5 (8.6%) had a history of acute pancreatitis, and 3 (5.1%) a history of chronic abdominal pain. The mean number of needle passages was 1.25 (range 1–2), and EUS-FNA/B was technically feasible in all cases (we performed more than one needle passage only in PCLs with solid components in order to obtain an adequate sample). An adequate amount of fluid was aspirated for CEA and amylase in 32 (53.3%) patients, and for K-ras in 24 (40%). In mucinous PCLs mean CEA level was 2127 ng/ml and mean amylase 19650 U/L; in non-mucinous PCLs mean CEA was 0.8 ng/ml and mean amylase 125 U/L. There were 9/24 (37.5%) patients positive for mutation of K-ras. Cystic fluid evaluation for CEA and K-ras mutation in mucinous cystic lesions are shown in Table 2. The most frequent reason for low fluid aspiration (insufficient for evaluation of intracystic CEA and/or K-ras) was the presence of intracystic fluid that was too viscous, and impossible to aspirate in 12 (20%) patients. The other reasons were adequate material for cyto-histological evaluation at the first passage of the needle on solid lesions, obviating the need for a second needle-passage, or small quantity of intracystic fluid, fluid too viscous to be tested by the laboratory; and the absence in the fluid of cells for evaluation of K-ras mutation. Table 2 Cystic fluid evaluation for carcinoembryonic antigen and K-ras mutation in pancreatic mucinous cystic lesions. CEA ≥ 192 ng/ml
CEA ≤ 192 ng/ml
Unavailable
9 (32.1%)
7 (26.9%)
13 (44.8%)
4 (44.4%)
2 (22.2%)
3 (33.3%)
K-ras mutation
Mucinous PCLs (n = 29) Mucinous malignant cysts (n = 9)
21 (35%) 20 (33.3%)a 6 (10%) 2 (3.3%) 3 (5%) 1 (1.6%) 7 (11.6%)b
PCLs, pancreatic cystic lesions. a Three intraductal papillary mucinous neoplasms with mild dysplasia at FNA/B were carcinoma in situ (n = 1) and invasive carcinoma (n = 2). b Three serous cystoadenoma, three pseudocysts, one pancreatic tuberculosis.
F, female; M, male; EUS, endoscopic ultrasound.
Mucinous PCLs (n = 29) Mucinous malignant cysts (n = 9)
Inadequate specimens Mucinous cysts with no malignancy Mucinous carcinoma (in situ or invasive) Ductal adenocarcinoma with cystic degeneration Neuroendocrine cystic tumour Solid pseudopapillary tumour Benign PCLs
Present
Absent
Unavailable
5 (17.8%) 1 (11.1%)
7 (25%) 4 (44.4%)
17 (58.6%) 4 (44.4%)
PCLs, pancreatic cystic lesions; CEA, carcinoembryonic antigen; K-ras, V-Ki-ras2 Kirsten rat sarcoma viral oncogene.
The overall sample adequacy for cyto-histological diagnosis was 39/60 (65%) PCLs (Fig. 3). The cyto-histological results of FNA/B are shown in Table 3. In PCLs with solid components the cellular material was adequate for cyto-histological evaluation in 17/18 (94.4%) lesions. The difference between the adequacy for cyto-histological diagnosis of all samples and only those of lesions with solid component was statistically significant (p = 0.0149). There were 15 (25%) malignant PCLs: 9 degenerated IPMN, 2 adenocarcinomas, 3 NET, and 1 SPPT. Ten (66.6%) malignant cysts were observed in PCLs ≤30 mm, 6 (40%) in ≤20 mm and 2 (13.3%) in ≤15 mm. In the malignant group, adequacy for cyto-histological evaluation was 15/15 (100%), though in 3 lesions (20%) the FNA/B underrated the malignancy, revealing mucinous epithelium with mild atypia, instead of 1 carcinoma in situ and 2 invasive carcinomas in IPMN observed in the histological samples from surgery. Worthy of note is that 2 of these 3 lesions were without solid component, 2 with CEA <192 ng/ml and 1 in which CEA was not available. The K-ras mutation was unavailable in two lesions and negative in the other lesion. The difference between adequacy for cyto-histological diagnosis of all specimens and that of specimens from malignant lesion was statistically significant (p = 0.0069). In 18/39 (46.1%) adequate samples the material was judged adequate as a histological sample and in 4 malignant cases (three NET and one SPPT) the possibility of carrying out immunohistochemical testing was essential for the diagnosis. There were 17 (28.3%) patients who underwent surgical resection. The surgical pathology showed IPMN with mild/moderate dysplasia in 8 patients, IPMN with carcinoma (in situ or invasive) in 5, MCN with mild dysplasia in 1, NET in 2 and SPPT in 1. In 13/17 (76.4%) patients in this group, the FNA/B sample was adequate for diagnosis. The concordance between FNA/B diagnosis and surgical specimen regarding the type of lesions was 100%, but in 3 (17.6%) lesions FNA/B underrated the severity of dysplasia (the same 3 patients with malignant IPMN instead of IPMN with mild atypia). No false positive results of the FNA/B were observed in surgically treated patients. There were only two early complications (3.3%): an intracystic haemorrhage in a cystic NET that resolved spontaneously, and a fever that was treated with two days of hospitalization and intravenous antibiotics. No late complications were observed. 5. Discussion Diagnosis of PCLs remains a very difficult challenge for physicians and is usually achieved by combining demographic and clinical information, radiologic and EUS assessments, intracystic fluid analyses and cyto-histology. Our study showed that FNA/B with a 22-gauge ProCore needle was feasible in all cases, including in PCLs located in difficult positions like the pancreatic head and uncinate process, and had an overall high diagnostic yield for cyto-histological sampling (65%).
L. Barresi et al. / Digestive and Liver Disease 46 (2014) 45–50
49
Fig. 3. (a) Neuroendocrine tumour (haematoxylin–eosin stain). (b) Immunohistochemistry of neuroendocrine tumour positive for synaptophysin. The test needs an adequate amount of cells to be performed. (c) Strips of mucinous epithelium with mild dysplasia. The preserved tissue architecture of the targeted lesions allows a more precise grading of dysplasia. (d) Severe atypical cells infiltrating the stroma.
It is particularly interesting to note that in the subgroup analysis of malignant lesions and lesions with a solid component, which are the most likely to harbour malignancy, the adequacy for cytohistological diagnosis was 100% and 94.4%, respectively. However, in 3 patients with malignant lesions the FNA/B diagnosis correctly identified the lesions as mucinous, but underestimated the presence of malignancy. This is not surprising because it is known that PCL cyto-histology can underrate the final histological grade of neoplastic mucinous cyst, given that the cyst lining is quite heterogeneous, and the degree of epithelial atypia may not be representative of the highest degree of atypia of the cyst at histology [21]. However, a diagnosis of mucinous lesions was essential in referring these patients to surgery, especially because in two of these patients there was no solid component of the lesions, and the CEA and K-ras were either unavailable or negative for mucinous lesions. The needle with lateral fenestration produced a high percentage of samples adequate for histological evaluation (46.1% of adequate samples), with a more precise diagnosis and the possibility of employing special stains and immunohistochemistry. In particular, in four patients the possibility of having enough material to perform immunohistochemistry was essential to the diagnosis of three cystic pancreatic NETs and one SPPT. We found no increase in the rate of complications, with only two mild complications, which resolved rapidly without consequences. The recently published guidelines in Fukuoka (2012) [24] do not consider EUS-FNA as the first option for diagnosis and imaging techniques are still considered essential in the decision analysis. However, other studies have shown that both morphologic EUS [6–8] and radiologic imaging [5–26] have a probability of correct diagnosis in only about half of the cases. In PCLs, both CEA and K-ras are useful in discriminating mucinous from non-mucinous lesions, but are not useful in identifying malignant lesions [8,10]. In our study, both CEA and K-ras in the cystic fluid had marginal utility in
the final diagnosis because in both benign and malignant mucinous PCLs, CEA was often under the cut-off value for mucinous lesions and K-ras was negative or both tests were unavailable. The unavailability of CEA to define PCL has also been described in other studies [25]. Conversely several studies [21,27] have shown that EUS-FNA with evaluation of the presence of mucinous cells with high-grade atypia in cystic fluid adds value to risk assessment of malignancy in pancreatic mucinous cysts, including small (<3 cm) cysts. So cytology appears to be the most useful diagnostic test because it can establish the type of lesion under observation, and it also offers the possibility to diagnose underlying malignancy. However, cytology from cystic fluid aspirate has been shown to have low adequacy and sensitivity because of the relatively low cellularity of aspirated pancreatic cyst fluid. In the Cooperative Pancreatic Cystic Study, the sensitivity of cystic fluid cytology for diagnosing mucinous cysts was only 34% because of the low number of cells found in cystic fluid [8]. Problems with cytology in EUS-guided cyst aspiration were delineated in a prospective, dual centre study in which only 31% of the samples had adequate cellularity for analysis [25]. New tools and techniques for improving diagnostic yield, such as Echobrush and CWP with either the trucut needle or the standard needle, have been proposed. Echobrush has been tested in several studies with conflicting results [12–14] and a high rate of complications (8–10%), including one death [13,14]. A prospective study on a new technique for obtaining cytologic material from the cystic wall with a standard 22G needle was recently published by Hong [15]. The author reports an impressive 56/69 (81%) cellular adequacy for cytologic assessment from CWP in PCLs. The technique used for FNA/B in our study is another method for obtaining samples from cyst walls. Regardless of the technique or needle used, obtaining a cystic wall sample seems to be superior to conventional FNA cytological sampling from cystic fluid. We think it is advisable to carry out a comparative study
50
L. Barresi et al. / Digestive and Liver Disease 46 (2014) 45–50
between the needle with side fenestration using the technique described in our study and CWP with a standard 22G needle, as described by Hong in his study. Limitations of our study have to be considered. Since only 28.3% of our patients underwent surgery, malignant lesions may have been missed. Similar studies in the future will probably suffer from this limitation, due to a growing body of scientific literature that supports a strategy of surveillance for small asymptomatic cysts without risk factors. However, although, the relatively short follow up of this report cannot completely exclude malignancy, no patients have had modification of the PCLs or developed new symptoms. Another limitation is that we did not have a control group for comparison with traditional cytology from cystic fluid aspirate or with other techniques, such as CWP with a standard 22G needle. If we compare our results with the historical control of cytology from cyst fluid aspirate, we find a substantial (about twofold) improvement in adequacy with this new needle. In conclusion, this dual centre, prospective study on PCLs suggests that EUS-guided FNA/B with the 22-gauge needle with side fenestration is feasible and maybe superior to conventional FNA cytology from cystic fluid, with a higher diagnostic yield, especially in PCLs with solid component and malignant lesions, with no associated increase in the rate of complications. These results need to be confirmed in comparative studies. Conflict of interest statement None declared. References [1] Lee KS, Sekhar A, Rofsky NM, et al. Prevalence of incidental pancreatic cysts in the adult population on MR imaging. American Journal of Gastroenterology 2010;105:2079–84. [2] Laffan T, Horton K, Klein A, et al. Prevalence of unsuspected pancreatic cysts on MDCT. American Journal of Roentgenology 2008;191:802–7. [3] Spinelli KS, Fromwiller TE, Daniel RA, et al. Cystic pancreatic neoplasm: observe or operate. Annals of Surgery 2004;239:651–7. [4] Brugge WR, Lauwers GY, Sahani D, et al. Cystic neoplasm of the pancreas. New England Journal of Medicine 2004;351:1218–26. [5] Visser BC, Yeh BM, Qayyum A, et al. Characterization of cystic pancreatic masses: relative accuracy of CT and MRI. American Journal of Roentgenology 2007;189:648–56. [6] O’Toole D, Palazzo L, Hammel P, et al. Macrocystic pancreatic cystadenoma: the role of EUS and cyst fluid analysis in distinguishing mucinous and serous lesions. Gastrointestinal Endoscopy 2004 Jun;59: 823–9. [7] Kubo H, Nakamura K, Itaba S, et al. Differential diagnosis of cystic tumors of the pancreas by endoscopic ultrasonography. Endoscopy 2009;41:684–9.
[8] Brugge WR, Lewandrowski K, Lee-Lewandrowski E, et al. The diagnosis of pancreatic cystic neoplasms: a report of the Cooperative Pancreatic Cyst (CPC) study. Gastroenterology 2004;126:1330–6. [9] van der Waaij LA, van Dullemen HM, Porte RJ. Cyst fluid analysis in the differential diagnosis of cystic pancreatic lesions: a pooled analysis. Gastrointestinal Endoscopy 2005;62:383–9. [10] Khalid A, Zahid M, Finkelstein SD, et al. Pancreatic cyst fluid DNA analysis in evaluating pancreatic cysts: a report of the PANDA study. Gastrointestinal Endoscopy 2009;69:1095–102. [11] Frossard JL, Amouyal P, Amouyal G, et al. Performance of endosonographyguide fine needle aspiration and biopsy in the diagnosis of cystic pancreatic lesions. American Journal of Gastroenterology 2003;98:1516–24. [12] Al-Haddad M, Gill KRS, Raimondo M, et al. Safety and efficacy of cytology brushings versus standard fine-needle aspiration in evaluating cystic pancreatic lesions: a controlled study. Endoscopy 2010;42:127–32. [13] Thomas T, Bebb J, Mannath J, et al. EUS-guided pancreatic cyst brushing: a comparative study in a tertiary referral centre. JOP 2010;11:163–9. [14] Sendino O, Fernández-Esparrach G, Solé M, et al. Endoscopic ultrasonographyguided brushing increases cellular diagnosis of pancreatic cysts: a prospective study. Digestive and Liver Disease 2010;42:877–81. [15] Hong SK, Loren DE, Rogart JN, et al. Targeted cyst wall puncture and aspiration during EUS-FNA increases the diagnostic yield of premalignant and malignant pancreatic cysts. Gastrointestinal Endoscopy 2012;75:775–82. [16] Levy MJ, Smyrk TC, Reddy RP, et al. Endoscopic ultrasound-guided trucut biopsy of the cyst wall for diagnosing cystic pancreatic tumors. Clinical Gastroenterology and Hepatology 2005;3:974–9. [17] Iglesias-Garcia J, Poley JW, Larghi A, et al. Feasibility and yield of a new EUS histology needle: results from a multicenter, pooled, cohort study. Gastrointestinal Endoscopy 2011;73:1189–96. [18] Iwashita T, Nakai Y, Samarasena JB. High single-pass diagnostic yield of a new 25-gauge core biopsy needle for EUS-guided FNA biopsy in solid pancreatic lesions. Gastrointestinal Endoscopy 2013;77:909–15. [19] Walsh PM, Zuccaro G, Dumot JA, et al. Predicting success of endoscopic aspiration for suspected pancreatic cystic neoplasms. Journal of the Pancreas 2008;9:612–7. [20] Nagle JA, Wilbur DC, Pitman MB. Cytomorphology of gastric and duodenal epithelium and reactivity to B723. Diagnostic Cytopathology 2005;33:381–6. [21] Pitman MB, Lewandrowsky K, Shen J, et al. Pancreatic cysts: preoperative diagnosis and clinical management. Cancer Cytopathology 2010;118:1–13. [22] Fabre M. Diagnosis of cystic pancreatic lesions by endoscopic ultrasonographyguided fine-needle aspiration. Which sampling and why? Utility of cystic wall microbiopsy and fluid analysis by monolayer processing. Acta Endoscopy 2002;32:1–13. [23] Papanikolaou IS, Adler A, Wegener K. Prospective pilot evaluation of a new needle prototype for endoscopic ultrasonography-guided fine-needle aspiration: comparison of cytology and histology yield. European Journal of Gastroenterology and Hepatology 2008;20:342–8. [24] Tanaka M, Fernández-del Castillo C, Adsay V, Chari S. International consensus guidelines 2012 for management of IPMN and MCN of the pancreas. Pancreatology 2012;12:183–97. [25] de Jong K, Poley JW, van Hooft JE, et al. Endoscopic-ultrasound fine-needle aspiration of pancreatic cystic lesions provides inadequate material for cytology and laboratory analysis: initial results of a prospective study. Endoscopy 2011;43:585–90. [26] Correa-Gallego C, Ferrone C, Thayer SP, et al. Incidental pancreati cysts: do we really know what we are watching. Pancreatology 2010;10:144–50. [27] Genevay M, Mino-Kenudson M, Yaeger K, et al. Cytology adds value to imaging studies for risk assessment of malignancy in pancreatic mucinous cyst. Annals of Surgery 2011;254:977–83.