- Email: [email protected]
Advantage of EUS Trucut biopsy combined with fine-needle aspiration without immediate on-site cytopathologic examination
ORIGINAL ARTICLE: Clinical Endoscopy
Advantage of EUS Trucut biopsy combined with fine-needle aspiration without immediate on-site cytopathologic examination Ian Storch, DO, Merce Jorda, MD, PhD, Richard Thurer, MD, Luis Raez, MD, Caio Rocha-Lima, MD, Stephen Vernon, MD, Afonso Ribeiro, MD Miami, Florida, USA
Background: Endoscopic ultrasonographically guided fine-needle aspiration (EUS-FNA) is a safe and accurate method for obtaining diagnostic material from lesions within and immediately adjacent to the upper GI tract. Objective: To determine whether EUS Trucut biopsy (EUS-TCB) (Quickcore, Wilson-Cook, Winstom Salem, NC) can increase the accuracy of EUS-guided tissue sampling when combined with FNA when no cytopathologist is present. Design: Retrospective case review. Setting: University-based referral practice. Patients: All patients who had lesions that were accessible through the esophagus or stomach and that were greater than 20 mm and amenable to Trucut biopsy were included. Interventions: A total of 41 patients underwent both EUS-FNA and TCB with a separate pathologist evaluating each specimen. Main Outcome Measurements: The diagnostic performance of FNA, TCB, and its combination were compared. Results: The overall accuracy in our series was as follows: FNA, 76%; TCB, 76% (P not significant); and combination of FNA and TCB, 95% (P Z.007). In the 26 patients with malignant diagnoses, the accuracy of combination was 100% versus 77% for FNA (P Z .03). The median number of passes with the FNA and TCB was 4.4 (range 2-8) and 2.8 (range 2-5), respectively. One patient in the series had fever and chest pain after EUS biopsy. Limitations: Retrospective study. Conclusion: In our series EUS-TCB accuracy was equal to FNA when no on-site cytopathologist is present. TCB was helpful in the diagnosis of pancreatic masses, gastric submucosal lesions, lymphoma, and necrotic tumors. A 100% accuracy of FNA C TCB was seen in patients with malignant diseases and in patients who had failed or been refused biopsy by other modalities in the past. More data are needed before the exact role of TCB in the absence of on-site cytopathology can be accurately defined. (Gastrointest Endosc 2006;64:505-11.)
Copyright ª 2006 by the American Society for Gastrointestinal Endoscopy 0016-5107/$32.00 doi:10.1016/j.gie.2006.02.056
and cost.1-4 Although some institutions have an on-site cytopathologist readily available, this is not always costeffective for the institution or for the individual cytopathologist.4 Preliminary studies evaluating the ability of an endosonographer to determine the adequacy of an FNA specimen have shown that this method is not much better than chance.5 As rapid improvements occur in noninvasive diagnostic radiological modalities, it is vital for endosonographers to develop sampling techniques that can maximize tissue acquisition, sensitivity, and specificity while minimizing cost.
www.giejournal.org
Volume 64, No. 4 : 2006 GASTROINTESTINAL ENDOSCOPY 505
Endoscopic ultrasonographically guided fine-needle aspiration (EUS-FNA) has become a mainstream technique for tissue acquisition from submucosal lesions arising from the gut and lesions and lymph nodes in the structures directly adjacent to the GI tract. Support for the use of EUS over other methods of tissue acquisition has been increasing as a result of its low morbidity, mortality,
Advantages of TCB with FNA without immediate on-site cytopathologic examination
Recently, a 19-gauge Trucut biopsy (TCB) needle has been introduced (Quickcore, Wilson-Cook, Winston-Salem, NC) for endosonographic use. Although this technique has been shown to have good accuracy for transcutaneous biopsy,6 its use to increase accuracy with EUS has only been reported in small series of patients (Table 1).6-11 This review was undertaken to report our experience with EUS-TCB when performed routinely in addition to standard FNA techniques when no cytopathologist was present.
Storch et al
Capsule Summary What is already known on this topic d
d
EUS-FNA is the main tissue acquisition technique for submucosal gastrointestinal lesions, as well as in structures adjacent to the gastrointestinal tract. A 19-gauge TCB needle for endosonographic use has good accuracy for transcutaneous biopsy, but its use with EUS has been limited.
What this study adds to our knowledge
METHODS
d
In a retrospective case review of 41 patients who underwent both EUS-FNA and TCB, EUS-TCB accuracy was equal to FNA when no on-site cytopathologist was present. TCB plus FNA had a combined overall accuracy of 95%, a significant difference over either modality alone. In 26 patients with malignant disease, the accuracy of EUS-FNA was 77%, EUS-TCB 73%, and combination of EUS-FNA C TCB 100%.
This study was a retrospective case review of all patients who underwent EUS-TCB from October 2003 through July 2005. All procedures were performed in the endoscopy unit of the University of Miami Hospital and Clinics. All patients who were referred for EUS-guided biopsy and who had lesions 20 mm or greater accessible through the esophagus or stomach with no adjacent vascular structures that would limit full advancement of the Trucut tissue tray were included. Informed consent was obtained from all patients before the procedure. Collection of data for this study was approved by our Institution Review Board (IRB No. 200504290345). EUS was performed with use of the linear-array echoendoscope (GF-UC30P; Olympus America, Melville, NY) with the patient under titrated intravenous conscious sedation (meperidine and midazolam). Two patients required monitored anesthesia care because of a high cardiopulmonary risk (American Society of Anesthesiologists score IV). Under EUS guidance, each lesion was first sampled by FNA (mean 4.4 passes, range 2-8 passes) with a 22-gauge needle (Echotip, Wilson-Cook). At least 2 FNA passes were made initially in each lesion, with a goal of 5 passes for lymph nodes and 7 for pancreatic masses. Manual suction was applied to the FNA needle with a 10-mL syringe in all cases. The FNA needle was passed through the largest diameter possible in each lesion to maximize tissue sampling. The FNA cytologic specimen was placed onto a glass slide and fixed in absolute alcohol solution for staining. No on-site cytopathologist was available for slide review. Next, a core biopsy specimen was obtained from the same lesion with a 19-gauge TCB needle (QuickCore, Wilson-Cook). The Trucut needle was prepared by a nurse assistant: the spring-loaded handle was pulled back until a click was heard (confirming that the needle was in the firing positon) and then inserted through the echoendoscope working channel. Before the TCB needle was advanced into the lesion, the projected path was interrogated with power Doppler imaging to ensure absence of intervening blood vessels. For EUS-TCB, it was essential to straighten the tip of the echoendoscope to allow smooth advancement of the needle into the lesion. After the needle was inserted into the tumor, the instrument
The diagnostic performance of EUS-FNA and TCB were compared with the McNemar test, and the mean number of needle passes between the 2 techniques were compared with use of the paired t test. A P value less than .05 was considered statistically significant. Although there was multiple testing of outcome data arising from individual patients, it is noted that correction by the Bonferroni method would not have removed any instances of statistical significance with the following analytic approach. The primary results were taken as those involving comparisons of diagnostic results among patients (1) with malignant disease and (2) without malignant disease. Because these were separate groups, no correction for multiple testing
506 GASTROINTESTINAL ENDOSCOPY Volume 64, No. 4 : 2006
www.giejournal.org
d
d
tip was straightened a little and the tissue tray was advanced into the tumor completely (close to the outer edge of the lesion) so that the largest fragment possible could be obtained. Trucut biopsy passes were made until adequate core tissue was obtained, with a goal of 2 samples. The length of core samples was not measured. The TCB specimen was preserved in formalin for histopathologic analysis. An immediate ‘‘touch prep’’ was not performed. A mean of 2.8 Trucut passes were made in each lesion (range 2-5 passes). Histopathologic and cytologic samples were reviewed by 2 different pathologists for adequacy and diagnosis. A positive diagnosis of malignancy by an EUS biopsy specimen was accepted as a true positive. A benign diagnosis was confirmed by surgical tissue samples when available or clinical follow-up (medical records or phone call). Each patient was called 24 hours after EUS to monitor for complications.
Statistical analysis
Storch et al
Advantages of TCB with FNA without immediate on-site cytopathologic examination
TABLE 1. Studies of EUS-TCB
Location of biopsy accuracy Larghi et al11
No. of patients with attempted TCB tissue acquisition
Pancreas benign
22
Itoi et al
10
Levy et al
3
2
Pancreas malignant
14
9
Total
17
11
Pancreas malignant
16
2
Total
16
2
Malignant lymph node
6
0
Benign lymph node
4
0
Intramural tumor/submucosal mass
5
3
Pancreatic neoplasm
3
1
Liver
1
0
Benign extramural mass
1
1
20
5
Malignant lymph node
1
0
Benign lymph node
7
0
Intramural tumor/submucosal mass
2
0
Pancreatic neoplasm
2
0
Pancreatic benign diagnosis
1
0
Benign extramural mass
3
1
Gastric esophageal cyst
2
0
18
1
Total Varadarajulu et al
9
Total 8
Early et al
Malignant lymph node Intramural tumor/submucosal mass Pancreatic neoplasm
Current study
Trucut changed diagnosis from FNA
Total (%)
23
82
82
69
69
81
60
85
90
89
78
100
78
76
95
2
3 15
3
Malignant lymph node
19
4
Benign lymph node
10
0
Lung
3
1
Intramural tumor/submucosal mass
5
3
Pancreatic neoplasm
4
1
41
9
86
22
Total
Trucut accuracy (%)
10
Total
Total
22-gauge FNA accuracy (%)
was necessary. All other statistical tests were considered as secondary and were subject to correction with the Bonferroni method, recognizing all statistical tests performed. The uncorrected P values for a single test are presented.
RESULTS
www.giejournal.org
Volume 64, No. 4 : 2006 GASTROINTESTINAL ENDOSCOPY 507
A total of 44 procedures were performed in 41 patients. There were 19 men and 22 women with a mean age of
Advantages of TCB with FNA without immediate on-site cytopathologic examination
Storch et al
TABLE 2. Cases where other modalities failed or were refused
TABLE 3. Results of TCB and FNA by final diagnosis in all patients
Modality
TCB positive and FNA negative
No. of patients
Failed bronchoscopy
7
Gastric GIST
2
Refused bronchoscopy, high risk
3
Pancreatic adenocarcinoma
2
Refused CT-guided biopsy, high risk
3
Lung cancer
1
Failed medianoscopy
1
Metastatic mediastinal node (lung cancer)*
2
Retroperitoneal lymphoma
1
Mediastinal lymphoma*
1
CT, Computed tomography.
65 years (range 18-85 years). Three patients with mediastinal masses underwent 2 procedures. In all 3, the first EUS was diagnostic of malignancy; however, inadequate tissue was present for additional immunostaining. These patients were classified as having positive studies in the final accuracy calculations. One patient who had an EUS for mediastinal adenopathy had fever and chills 3 days after the procedure. Three passes of both FNA and TCB were performed into this lesion. The patient was treated with intravenous antibiotics for 3 days and then oral antibiotics for 11 days at an outside institution with resolution of symptoms. No surgery was required. Of the 41 patients who underwent EUS, 26 had a final diagnosis of malignancy and 15 patients had benign disorders. Sites of biopsy were 26 mediastinal nodes (21 subcarinal, 5 anteroposterior window) and 3 lung and 12 transgastric biopsies (6 submucosal lesions, 2 celiac nodes, and 4 pancreatic masses). The median follow-up in patients with benign histologic features was 6 months (range 2-11 months). The mean size of lesions that were biopsied was 33 mm (range 20-55 mm). The accuracy for FNA, TCB, and FNA C TCB was 76%, 76%, and 95%, respectively (P Z .007). A total of 14 (34%) patients were referred either because of failure by other modalities or refusal (Table 2). Eight patients had a prior biopsy procedure that had failed to provide the correct diagnosis. Three patients were not considered candidates for bronchoscopy or mediastinoscopy because of high cardiopulmonary risk (American Society of Anesthesiologists III and IV), 3 were turned down for CT-guided biopsy (2 pancreatic masses because of collateral blood vessels and lack of a safe window and 1 lung mass because of advanced chronic obstructive pulmonary disease with risk of life-threatening pneumothorax). The diagnostic accuracy for EUS-FNA was 78% and for EUS-TCB was 86% (not significant [NS]) in this group. The combined accuracy of FNA plus TCB was 100% (NS). In the subgroup of patients with malignant diseases (26 patients), the accuracy of EUS-FNA was 77%, EUSTCB 73%, and combination of EUS-FNA plus TCB 100% (P Z .03). In 16 patients with lung cancer, FNA accuracy was 87%, TCB 81%, and the combination 100% (NS). The absolute difference between the FNA C TCB versus
FNA occurred because TCB histologic study was diagnostic in 8 cases missed by FNA (Table 3). Three of these patients, 1 patient with lymphoma and 2 with lung cancer, had prior failed biopsy procedures. In 23 patients (56%), tissue was acquired on the first EUS-TCB pass. A total of 10 patients failed EUS-TCB; in 3 cases TCB failed to obtain any tissue for analysis (2 mediastinal and 1 distal antrum submucosal mass) and in 7 cases TCB provided nondiagnostic material. Overall, the average number of needle passes for EUSFNA was 4.3 (range 2-8) and for EUS-TCB 2.7 (range 1-5) (P ! .01). In patients with positive and negative FNA or TCB results, the average numbers of needle passes were as follows: positive FNA 4.4, negative FNA 4.1, positive
508 GASTROINTESTINAL ENDOSCOPY Volume 64, No. 4 : 2006
www.giejournal.org
TCB negative and FNA positive Pancreatic adenocarcinoma
1
Pancreatic mucinous neoplasm
1
Metastatic melanoma to mediastinal node
1
Metastatic mediastinal node (lung cancer)
4
Benign adenopathy
1
TCB positive and FNA positive Gastric GIST
2
Pancreatic adenocarcinoma
1
Lung metastasis of melanoma
1
Lung lymphoma
1
Sarcoidosis
2
Benign adenopathy
6
Metastatic mediastinal node (lung cancer)
8
Adenocarcinoma metastatic to celeac node
1
Lung cancer
1
TCB negative and FNA negative Gastric submucosa mass
1
Gastric GIST
1
GIST, Gastrointestinal stromal tumor. *Necrotic tumor.
Storch et al
Advantages of TCB with FNA without immediate on-site cytopathologic examination
TCB 2.7, and negative TCB 2.8. There was no statistical difference between the number of passes in the positive and negative FNA-TCB groups.
EUS-guided tissue acquisition has emerged as an accurate, safe, and relatively inexpensive method of diagnosing lesions within the GI lumen or in organs or lymph nodes located adjacent to the esophagus, stomach, and duodenum. Although EUS-FNA has a high accuracy in many clinical settings, one of the current aims of EUS research is to define techniques that will maximize accuracy while minimizing cost. Recently there has been a strong interest in the 19-gauge Trucut needle, a method of tissue acquisition that allows endosonographers to obtain a standard core biopsy. One of the advantages of a Trucut biopsy is that specimen adequacy can potentially be determined by the endosonographer, whereas FNA samples require an on-site cytopathologist, a resource not readily available at many institutions. Another benefit of a core sample is that it may improve yield compared with standard FNA in necrotic tumors, sclerotic lesions, lymphoma, and GI stromal tumors.3 Although prior studies have demonstrated the accuracy of EUS-TCB to range from 69% to 85% (Table 1), no comparison between FNA and combination of FNA C TCB has ever been published. The rationale for adding TCB to FNA in our study was to maximize tissue acquisition/adequacy when no on-site cytopathologist was present because it has been reported that EUS-FNA in the absence of an on-site reviewer can result in 10% to 15% reduction in the rate of definitive cytologic diagnosis.12 Our study demonstrated TCB C FNA to have a combined overall accuracy of 95%, a statistically significant difference over either modality alone. In patients with malignant disease, TCB C FNA resulted in 100% accuracy compared with 77% with FNA alone (P Z .03). Our data provide further support for the current concept that in (1) pancreatic masses, (2) gastric submucosal masses, (3) lymphoma, and (4) necrotic tumors, TCB increases diagnostic accuracy over FNA alone (Table 3). Clinical judgment is still essential in these cases because FNA of the pancreatic body and tail, for example, carries a risk of possible gastric tumor seeding.13 The advantage of EUS biopsy in subepithelial lesions greater than 5 cm must also be carefully considered because surgery is usually indicated because of the malignancy risk at this size. It must be kept in mind that negative FNA results in presumed GI stromal tumors does not help guide management because sampling error has been shown to occur.14 Our mean number of FNA passes (4.3) was less than our initial goal of 5 passes for lymph nodes and 7 for pancreatic masses. In some high-risk patients, the number of FNA passes was intentionally decreased to shorten the
procedure time. The mean number of FNA C TCB passes was 7, which was within our goal range. The lower mean number of FNA passes probably did not affect the accuracy in our data because most cases missed by FNA were lesions in which cytology is problematic (lymph node with necrosis, lymphoma, and GI stromal tumors). Only one pancreatic cancer was missed by FNA; this patient underwent 7 FNA passes. Among our patients with positive or negative FNA or TCB, there was no significant difference in the number of needle passes performed. Our study included a large number of mediastinal lymph nodes. This is of interest because EUS-TCB through the esophagus can be easier than through the stomach or duodenum because of a straighter scope position. In 2 of our patients, however, TCB still failed to obtain adequate tissue. When EUS-TCB is performed on mediastinal nodes, there are several limitations. First, lymph nodes must have at least 15 to 20 mm length in the path of the Trucut needle to allow full extension of the tissue tray. Second, the proximity of the pulmonary arteries and left atrium in the subcarinal space can represent a challenge for EUSTCB in smaller lesions. It has been suggested that in nodes that have a high suspicion of being metastatic from lung cancer, an initial evaluation with EUS-FNA (in centers with on-site cytopathology) has an accuracy of 94%,15 theoretically obviating the need for EUS-TCB. In our series, in 3 of 16 cases of lung cancer (19%), TCB led to a change the diagnosis over FNA. Further evaluation of TCB in the mediastinum is needed to determine its true benefit when on-site cytopathology is unavailable. Our data suggest that, in patients with failed prior biopsy by bronchoscopy or mediastinoscopy, EUS-guided biopsy is helpful. Prior studies have demonstrated up to a 97% accuracy for EUS-FNA in patients with suspected centrally located lung tumors and 71% in patients with subcarinal lymph nodes in patients who have had prior transbronchial biopsy.16,17 Other reports have described the benefit of EUS in addition to mediastinoscopy for lung cancer staging, with an increase in accuracy of up to 16%.18,19 In our study, 2 cases of lymphoma and 3 metastatic lung cancers with significant necrosis diagnosed by FNA C TCB were previously missed by transbronchial needle biopsy or mediastinoscopy. Further study of EUS biopsy in patients who have previously failed other modalities will be needed to define the role of TCB in these cases. The role of EUS-guided FNA with flow cytometry in patients with suspected lymphoma has been increasingly recognized.20 A diagnosis of lymphoma, however, can be difficult to make on cytologic study alone using the accepted lymphoma classification system (working formulation). Even if a diagnosis of lymphoma is made, subclassification is often impossible21(ie, Hodgkin’s disease). Varadarajulu et al22 reported a case where lymphoma was diagnosed by EUS-FNA, flow cytometry, and TCB. In our series, EUS-TCB changed management in two patients with lymphoma who had flow cytometry
www.giejournal.org
Volume 64, No. 4 : 2006 GASTROINTESTINAL ENDOSCOPY 509
DISCUSSION
Advantages of TCB with FNA without immediate on-site cytopathologic examination
Figure 1. A, EUS demonstrating a fistula within a mass in a patient with dysphagia. B, Trucut specimen defining the tumor as malignant lymphoma, large B-cell type. (H&E, orig. mag. 20.)
performed on an FNA sample (Fig. 1A and B). In one patient with large cell lymphoma, histologic analysis and immunostaining was needed for final diagnostic confirmation of subtype. In the second patient, the presence of intense necrosis led to an initial benign diagnosis by FNA, and a final diagnosis of large cell lymphoma was made by TCB. Our series adds further support to the utility of EUS-TCB for final subclassification in patients with suspected lymphoma. Although TCB C FNA may increase accuracy of EUS tissue sampling when no cytopathologist is present, there are potential drawbacks to combination biopsy. The use of FNA C TCB results in more needle passes per case and higher cost yet still appears safe in our initial review. 510 GASTROINTESTINAL ENDOSCOPY Volume 64, No. 4 : 2006
Storch et al
Only one minor complication (fever and chest pain that resolved with antibiotics) out of 44 procedures was observed (a rate of 2.2%). The complication rate of EUSFNA alone has been stated to be about 0.5%.2 Complications reported in other series of TCB alone include 2 cases of mediastinitis and 1 case of postbiopsy bleeding.8,9 More data on the safety of TCB is still needed. Other problems with the Trucut needle include its stiffness, which makes advancement in certain scope positions challenging, particularly in the duodenum. Although our series does not include cases sampled through the duodenum, the published experience thus far is disappointing.6-11 We have also encountered occasional difficulties in the distal antrum, including misfire of the needle once it is positioned inside the lesion. At times, the long throw of the tissue tray makes obtaining an adequate specimen in the mediastinum difficult. Refinements in the needle, such as a smaller gauge and sturdier design, will be necessary to improve its performance. It was not the intent of our study to determine whether Trucut biopsy can replace FNA C on-site cytopathologic study. It is important to realize that the addition of TCB to standard FNA without an on-site cytopathologist on average costs $200 (based on the Medicare cost) more than EUS-FNA with on-site cytopathologic examination. Because no direct comparison between the accuracy of FNA C on-site cytopathologic study versus FNA C TCB without an on-site cytopathologist has been performed, it is impossible to determine which is more cost-effective. In centers where on-site cytopathologic examination is not available, the question is whether TCB should be done in all cases versus repeat EUS with TCB when the first FNA has failed. Our study demonstrates that the increased accuracy of FNA C TCB over FNA alone was primarily in patients with pancreatic masses, lymphoma, GI stromal tumors, and necrotic lesions. We thus suggest that in centers without on-site cytopathologic capabilities, TCB during the initial EUS be reserved for these lesions and for cases of previously failed EUS-FNA until more data are available. In summary, we have shown overall EUS-TCB accuracy to be, at best, equal to FNA when no on-site cytopathologist is present. The 100% accuracy in this series among patients with malignant diseases, patients who failed prior biopsies by other modalities, or who were refused other procedures because of medical risk supports the use of a combined FNA-TCB in these situations. TCB seems to hold promise in specific clinical scenarios; however, more data will be needed before FNA C TCB can be recommended in all patients when on-site cytopathologic study is unavailable.
DISCLOSURE None of the authors has received any study grants or financial incentive from manufacturers of endoscopic www.giejournal.org
Storch et al
Advantages of TCB with FNA without immediate on-site cytopathologic examination
equipment or other devices mentioned in this study. The authors do not hold any shares of the companies marketing devices used in this study.
REFERENCES 1. Erickson R. EUS-guided FNA. Gastrointest Endosc 2004;60:267-4. 2. Lee LS, Saltzman JR, Bounds BC, et al. EUS-guided fine needle aspiration of pancreatic cysts: a retrospective analysis of complications and their predictors. Clin Gastroenterol Hepatol 2005;3:231-6. 3. Wiersma MJ, Vilmann P, Giovannini M, et al. Endosonography-guided fine-needle aspiration biopsy: diagnostic accuracy and complication assessment. Gastroenterology 1997;112:1087-95. 4. Jhala NC, Jhala DN, Chhiend DC, et al. Endoscopic ultrasound-guided fine-needle aspiration: a cytopathologist’s perspective. Am J Clin Pathol 2003;120:351-67. 5. Hikichi T, Irisawa A, Buthani MS, et al. Endoscopic ultrasound-guided fine-needle aspiration with immediate cytology of solid pancreatic masses by endoscopists [abstract]. Gastrointest Endosc 2005;61: AB281. 6. Levy M, Reddy R, Wiersema M, et al. EUS-guided trucut biopsy in establishing autoimmune pancreatitis as the cause of obstructive jaundice. Gastrointest Endosc 2005;61:467-72. 7. Tam W, Aithal G, Dean J, et al. A prospective multicenter study of the safety and accuracy of combined endoscopic ultrasound (EUS)– guided fine needle aspiration (FNA) and Trucut needle biopsy (TNB) [abstract]. Gastrointest Endosc 2004;59:AB212:T1552. 8. Early D, Edmundowicz S, Azar R, et al. EUS-guided trucut biopsy. Gastroenterology 2003;57:W1233. 9. Varadarajulu S, Fraig M, Schmulewitz N, et al. Comparison of EUSguided 19-gauge Trucut needle biopsy with EUS-guided fine-needle aspiration. Endoscopy 2004;36:397-401. 10. Levy M, Jondal M, Clain J, et al. Preliminary experience with an EUS-guided trucut biopsy needle compared with EUS-guided FNA. Gastrointest Endosc 2003;57:101-6. 11. Larghi A, Verna E, Stavropoulos S, et al. EUS-guided trucut needle biopsies in patients with solid pancreatic masses: a prospective study. Gastrointest Endosc 2004;59:185-90. 12. Eloubeidi MA, Cohn M, Cerfolio RJ, et al. Endoscopic ultrasoundguided fine-needle aspiration in the diagnosis of foregut duplication cysts: the value of demonstrating detached ciliary tufts in cyst fluid. Cancer 2004;102:253-8.
www.giejournal.org
13. Paquin SC, Gariepy G, Lepanto L, et al. A first report of tumor seeding because of EUS-guided FNA of a pancreatic adenocarcinoma. Gastrointest Endosc 2005;61:610-1. 14. Fu K, Eloubeidi MA, Jhala NC, et al. Diagnosis of gastrointestinal stromal tumor by endoscopic ultrasound-guided fine needle aspiration biopsyda potential pitfall. Ann Diagn Pathol 2002;6:294-301. 15. Caddy G, Conron M, Wright G, et al. The accuracy of EUS-FNA in assessing mediastinal lymphadenopathy and staging patients with NSCLC. Eur Respir J 2005;25:410-5. 16. Annema JT, Veselic M, Rabe KF. EUS-guided FNA of centrally located lung tumours following a non-diagnostic bronchoscopy. Lung Cancer 2005;48:357-61. 17. Annema JT, Veselic M, Rabe KF. Analysis of subcarinal lymph nodes in (suspected) non-small-cell lung cancer after a negative transbronchial needle aspirationdwhat’s next? A preliminary report. Respiration 2004;71:630-4. 18. Annema JT, Versteegh MI, Veselic M, et al. Endoscopic ultrasound added to mediastinoscopy for preoperative staging of patients with lung cancer. JAMA 2005;294:931-6. 19. Van Schil P. Accuracy of mediastinoscopy compared to endoscopic ultrasound with fine needle aspiration (EUS-FNA). Lung Cancer 2005;50:273-4. 20. Ribeiro A, Vazquez-Sequeiros E, Wiersma LM, et al. EUS-guided fineneedle aspiration combined with flow cytometry and immunocytochemistry in the diagnosis of lymphoma. Gastrointest Endosc 2001;53:485-91. 21. Hehn S, Grogan T, Miller T. Utility of fine-needle aspiration as a diagnostic technique in lymphoma. J Clin Oncol 2004;22:3046–5. 22. Itoi T, Itokawa F, Sofuni A, et al. Puncture of solid pancreatic tumors guided by endoscopic ultrasonography: a pilot study series comparing Trucut and 19-gauge and 22-gauge aspiration needles. Endoscopy 2005;37:362-6.
Received September 21, 2005. Accepted February 20, 2006. Current affiliations: Division of Gastroenterology (I. S., A. R.), Division of Hematology/Oncology (L. R., C. R.), Department of Surgical Pathology (M. J., S. V.), and Division of Thoracic Surgery (R. T.), Sylvester Comprehensive Cancer Center. Miller School of Medicine, University of Miami, Miami, Florida. Reprint requests: Afonso Ribeiro, MD, Division of Gastroenterology, University of Miami, PO Box 016960, Miami, FL 33101.
Volume 64, No. 4 : 2006 GASTROINTESTINAL ENDOSCOPY 511
Advantage of EUS Trucut biopsy combined with fine-needle aspiration without immediate on-site cytopathologic examination Ian Storch, DO, Merce Jorda, MD, PhD, Richard Thurer, MD, Luis Raez, MD, Caio Rocha-Lima, MD, Stephen Vernon, MD, Afonso Ribeiro, MD Miami, Florida, USA
Background: Endoscopic ultrasonographically guided fine-needle aspiration (EUS-FNA) is a safe and accurate method for obtaining diagnostic material from lesions within and immediately adjacent to the upper GI tract. Objective: To determine whether EUS Trucut biopsy (EUS-TCB) (Quickcore, Wilson-Cook, Winstom Salem, NC) can increase the accuracy of EUS-guided tissue sampling when combined with FNA when no cytopathologist is present. Design: Retrospective case review. Setting: University-based referral practice. Patients: All patients who had lesions that were accessible through the esophagus or stomach and that were greater than 20 mm and amenable to Trucut biopsy were included. Interventions: A total of 41 patients underwent both EUS-FNA and TCB with a separate pathologist evaluating each specimen. Main Outcome Measurements: The diagnostic performance of FNA, TCB, and its combination were compared. Results: The overall accuracy in our series was as follows: FNA, 76%; TCB, 76% (P not significant); and combination of FNA and TCB, 95% (P Z.007). In the 26 patients with malignant diagnoses, the accuracy of combination was 100% versus 77% for FNA (P Z .03). The median number of passes with the FNA and TCB was 4.4 (range 2-8) and 2.8 (range 2-5), respectively. One patient in the series had fever and chest pain after EUS biopsy. Limitations: Retrospective study. Conclusion: In our series EUS-TCB accuracy was equal to FNA when no on-site cytopathologist is present. TCB was helpful in the diagnosis of pancreatic masses, gastric submucosal lesions, lymphoma, and necrotic tumors. A 100% accuracy of FNA C TCB was seen in patients with malignant diseases and in patients who had failed or been refused biopsy by other modalities in the past. More data are needed before the exact role of TCB in the absence of on-site cytopathology can be accurately defined. (Gastrointest Endosc 2006;64:505-11.)
Copyright ª 2006 by the American Society for Gastrointestinal Endoscopy 0016-5107/$32.00 doi:10.1016/j.gie.2006.02.056
and cost.1-4 Although some institutions have an on-site cytopathologist readily available, this is not always costeffective for the institution or for the individual cytopathologist.4 Preliminary studies evaluating the ability of an endosonographer to determine the adequacy of an FNA specimen have shown that this method is not much better than chance.5 As rapid improvements occur in noninvasive diagnostic radiological modalities, it is vital for endosonographers to develop sampling techniques that can maximize tissue acquisition, sensitivity, and specificity while minimizing cost.
www.giejournal.org
Volume 64, No. 4 : 2006 GASTROINTESTINAL ENDOSCOPY 505
Endoscopic ultrasonographically guided fine-needle aspiration (EUS-FNA) has become a mainstream technique for tissue acquisition from submucosal lesions arising from the gut and lesions and lymph nodes in the structures directly adjacent to the GI tract. Support for the use of EUS over other methods of tissue acquisition has been increasing as a result of its low morbidity, mortality,
Advantages of TCB with FNA without immediate on-site cytopathologic examination
Recently, a 19-gauge Trucut biopsy (TCB) needle has been introduced (Quickcore, Wilson-Cook, Winston-Salem, NC) for endosonographic use. Although this technique has been shown to have good accuracy for transcutaneous biopsy,6 its use to increase accuracy with EUS has only been reported in small series of patients (Table 1).6-11 This review was undertaken to report our experience with EUS-TCB when performed routinely in addition to standard FNA techniques when no cytopathologist was present.
Storch et al
Capsule Summary What is already known on this topic d
d
EUS-FNA is the main tissue acquisition technique for submucosal gastrointestinal lesions, as well as in structures adjacent to the gastrointestinal tract. A 19-gauge TCB needle for endosonographic use has good accuracy for transcutaneous biopsy, but its use with EUS has been limited.
What this study adds to our knowledge
METHODS
d
In a retrospective case review of 41 patients who underwent both EUS-FNA and TCB, EUS-TCB accuracy was equal to FNA when no on-site cytopathologist was present. TCB plus FNA had a combined overall accuracy of 95%, a significant difference over either modality alone. In 26 patients with malignant disease, the accuracy of EUS-FNA was 77%, EUS-TCB 73%, and combination of EUS-FNA C TCB 100%.
This study was a retrospective case review of all patients who underwent EUS-TCB from October 2003 through July 2005. All procedures were performed in the endoscopy unit of the University of Miami Hospital and Clinics. All patients who were referred for EUS-guided biopsy and who had lesions 20 mm or greater accessible through the esophagus or stomach with no adjacent vascular structures that would limit full advancement of the Trucut tissue tray were included. Informed consent was obtained from all patients before the procedure. Collection of data for this study was approved by our Institution Review Board (IRB No. 200504290345). EUS was performed with use of the linear-array echoendoscope (GF-UC30P; Olympus America, Melville, NY) with the patient under titrated intravenous conscious sedation (meperidine and midazolam). Two patients required monitored anesthesia care because of a high cardiopulmonary risk (American Society of Anesthesiologists score IV). Under EUS guidance, each lesion was first sampled by FNA (mean 4.4 passes, range 2-8 passes) with a 22-gauge needle (Echotip, Wilson-Cook). At least 2 FNA passes were made initially in each lesion, with a goal of 5 passes for lymph nodes and 7 for pancreatic masses. Manual suction was applied to the FNA needle with a 10-mL syringe in all cases. The FNA needle was passed through the largest diameter possible in each lesion to maximize tissue sampling. The FNA cytologic specimen was placed onto a glass slide and fixed in absolute alcohol solution for staining. No on-site cytopathologist was available for slide review. Next, a core biopsy specimen was obtained from the same lesion with a 19-gauge TCB needle (QuickCore, Wilson-Cook). The Trucut needle was prepared by a nurse assistant: the spring-loaded handle was pulled back until a click was heard (confirming that the needle was in the firing positon) and then inserted through the echoendoscope working channel. Before the TCB needle was advanced into the lesion, the projected path was interrogated with power Doppler imaging to ensure absence of intervening blood vessels. For EUS-TCB, it was essential to straighten the tip of the echoendoscope to allow smooth advancement of the needle into the lesion. After the needle was inserted into the tumor, the instrument
The diagnostic performance of EUS-FNA and TCB were compared with the McNemar test, and the mean number of needle passes between the 2 techniques were compared with use of the paired t test. A P value less than .05 was considered statistically significant. Although there was multiple testing of outcome data arising from individual patients, it is noted that correction by the Bonferroni method would not have removed any instances of statistical significance with the following analytic approach. The primary results were taken as those involving comparisons of diagnostic results among patients (1) with malignant disease and (2) without malignant disease. Because these were separate groups, no correction for multiple testing
506 GASTROINTESTINAL ENDOSCOPY Volume 64, No. 4 : 2006
www.giejournal.org
d
d
tip was straightened a little and the tissue tray was advanced into the tumor completely (close to the outer edge of the lesion) so that the largest fragment possible could be obtained. Trucut biopsy passes were made until adequate core tissue was obtained, with a goal of 2 samples. The length of core samples was not measured. The TCB specimen was preserved in formalin for histopathologic analysis. An immediate ‘‘touch prep’’ was not performed. A mean of 2.8 Trucut passes were made in each lesion (range 2-5 passes). Histopathologic and cytologic samples were reviewed by 2 different pathologists for adequacy and diagnosis. A positive diagnosis of malignancy by an EUS biopsy specimen was accepted as a true positive. A benign diagnosis was confirmed by surgical tissue samples when available or clinical follow-up (medical records or phone call). Each patient was called 24 hours after EUS to monitor for complications.
Statistical analysis
Storch et al
Advantages of TCB with FNA without immediate on-site cytopathologic examination
TABLE 1. Studies of EUS-TCB
Location of biopsy accuracy Larghi et al11
No. of patients with attempted TCB tissue acquisition
Pancreas benign
22
Itoi et al
10
Levy et al
3
2
Pancreas malignant
14
9
Total
17
11
Pancreas malignant
16
2
Total
16
2
Malignant lymph node
6
0
Benign lymph node
4
0
Intramural tumor/submucosal mass
5
3
Pancreatic neoplasm
3
1
Liver
1
0
Benign extramural mass
1
1
20
5
Malignant lymph node
1
0
Benign lymph node
7
0
Intramural tumor/submucosal mass
2
0
Pancreatic neoplasm
2
0
Pancreatic benign diagnosis
1
0
Benign extramural mass
3
1
Gastric esophageal cyst
2
0
18
1
Total Varadarajulu et al
9
Total 8
Early et al
Malignant lymph node Intramural tumor/submucosal mass Pancreatic neoplasm
Current study
Trucut changed diagnosis from FNA
Total (%)
23
82
82
69
69
81
60
85
90
89
78
100
78
76
95
2
3 15
3
Malignant lymph node
19
4
Benign lymph node
10
0
Lung
3
1
Intramural tumor/submucosal mass
5
3
Pancreatic neoplasm
4
1
41
9
86
22
Total
Trucut accuracy (%)
10
Total
Total
22-gauge FNA accuracy (%)
was necessary. All other statistical tests were considered as secondary and were subject to correction with the Bonferroni method, recognizing all statistical tests performed. The uncorrected P values for a single test are presented.
RESULTS
www.giejournal.org
Volume 64, No. 4 : 2006 GASTROINTESTINAL ENDOSCOPY 507
A total of 44 procedures were performed in 41 patients. There were 19 men and 22 women with a mean age of
Advantages of TCB with FNA without immediate on-site cytopathologic examination
Storch et al
TABLE 2. Cases where other modalities failed or were refused
TABLE 3. Results of TCB and FNA by final diagnosis in all patients
Modality
TCB positive and FNA negative
No. of patients
Failed bronchoscopy
7
Gastric GIST
2
Refused bronchoscopy, high risk
3
Pancreatic adenocarcinoma
2
Refused CT-guided biopsy, high risk
3
Lung cancer
1
Failed medianoscopy
1
Metastatic mediastinal node (lung cancer)*
2
Retroperitoneal lymphoma
1
Mediastinal lymphoma*
1
CT, Computed tomography.
65 years (range 18-85 years). Three patients with mediastinal masses underwent 2 procedures. In all 3, the first EUS was diagnostic of malignancy; however, inadequate tissue was present for additional immunostaining. These patients were classified as having positive studies in the final accuracy calculations. One patient who had an EUS for mediastinal adenopathy had fever and chills 3 days after the procedure. Three passes of both FNA and TCB were performed into this lesion. The patient was treated with intravenous antibiotics for 3 days and then oral antibiotics for 11 days at an outside institution with resolution of symptoms. No surgery was required. Of the 41 patients who underwent EUS, 26 had a final diagnosis of malignancy and 15 patients had benign disorders. Sites of biopsy were 26 mediastinal nodes (21 subcarinal, 5 anteroposterior window) and 3 lung and 12 transgastric biopsies (6 submucosal lesions, 2 celiac nodes, and 4 pancreatic masses). The median follow-up in patients with benign histologic features was 6 months (range 2-11 months). The mean size of lesions that were biopsied was 33 mm (range 20-55 mm). The accuracy for FNA, TCB, and FNA C TCB was 76%, 76%, and 95%, respectively (P Z .007). A total of 14 (34%) patients were referred either because of failure by other modalities or refusal (Table 2). Eight patients had a prior biopsy procedure that had failed to provide the correct diagnosis. Three patients were not considered candidates for bronchoscopy or mediastinoscopy because of high cardiopulmonary risk (American Society of Anesthesiologists III and IV), 3 were turned down for CT-guided biopsy (2 pancreatic masses because of collateral blood vessels and lack of a safe window and 1 lung mass because of advanced chronic obstructive pulmonary disease with risk of life-threatening pneumothorax). The diagnostic accuracy for EUS-FNA was 78% and for EUS-TCB was 86% (not significant [NS]) in this group. The combined accuracy of FNA plus TCB was 100% (NS). In the subgroup of patients with malignant diseases (26 patients), the accuracy of EUS-FNA was 77%, EUSTCB 73%, and combination of EUS-FNA plus TCB 100% (P Z .03). In 16 patients with lung cancer, FNA accuracy was 87%, TCB 81%, and the combination 100% (NS). The absolute difference between the FNA C TCB versus
FNA occurred because TCB histologic study was diagnostic in 8 cases missed by FNA (Table 3). Three of these patients, 1 patient with lymphoma and 2 with lung cancer, had prior failed biopsy procedures. In 23 patients (56%), tissue was acquired on the first EUS-TCB pass. A total of 10 patients failed EUS-TCB; in 3 cases TCB failed to obtain any tissue for analysis (2 mediastinal and 1 distal antrum submucosal mass) and in 7 cases TCB provided nondiagnostic material. Overall, the average number of needle passes for EUSFNA was 4.3 (range 2-8) and for EUS-TCB 2.7 (range 1-5) (P ! .01). In patients with positive and negative FNA or TCB results, the average numbers of needle passes were as follows: positive FNA 4.4, negative FNA 4.1, positive
508 GASTROINTESTINAL ENDOSCOPY Volume 64, No. 4 : 2006
www.giejournal.org
TCB negative and FNA positive Pancreatic adenocarcinoma
1
Pancreatic mucinous neoplasm
1
Metastatic melanoma to mediastinal node
1
Metastatic mediastinal node (lung cancer)
4
Benign adenopathy
1
TCB positive and FNA positive Gastric GIST
2
Pancreatic adenocarcinoma
1
Lung metastasis of melanoma
1
Lung lymphoma
1
Sarcoidosis
2
Benign adenopathy
6
Metastatic mediastinal node (lung cancer)
8
Adenocarcinoma metastatic to celeac node
1
Lung cancer
1
TCB negative and FNA negative Gastric submucosa mass
1
Gastric GIST
1
GIST, Gastrointestinal stromal tumor. *Necrotic tumor.
Storch et al
Advantages of TCB with FNA without immediate on-site cytopathologic examination
TCB 2.7, and negative TCB 2.8. There was no statistical difference between the number of passes in the positive and negative FNA-TCB groups.
EUS-guided tissue acquisition has emerged as an accurate, safe, and relatively inexpensive method of diagnosing lesions within the GI lumen or in organs or lymph nodes located adjacent to the esophagus, stomach, and duodenum. Although EUS-FNA has a high accuracy in many clinical settings, one of the current aims of EUS research is to define techniques that will maximize accuracy while minimizing cost. Recently there has been a strong interest in the 19-gauge Trucut needle, a method of tissue acquisition that allows endosonographers to obtain a standard core biopsy. One of the advantages of a Trucut biopsy is that specimen adequacy can potentially be determined by the endosonographer, whereas FNA samples require an on-site cytopathologist, a resource not readily available at many institutions. Another benefit of a core sample is that it may improve yield compared with standard FNA in necrotic tumors, sclerotic lesions, lymphoma, and GI stromal tumors.3 Although prior studies have demonstrated the accuracy of EUS-TCB to range from 69% to 85% (Table 1), no comparison between FNA and combination of FNA C TCB has ever been published. The rationale for adding TCB to FNA in our study was to maximize tissue acquisition/adequacy when no on-site cytopathologist was present because it has been reported that EUS-FNA in the absence of an on-site reviewer can result in 10% to 15% reduction in the rate of definitive cytologic diagnosis.12 Our study demonstrated TCB C FNA to have a combined overall accuracy of 95%, a statistically significant difference over either modality alone. In patients with malignant disease, TCB C FNA resulted in 100% accuracy compared with 77% with FNA alone (P Z .03). Our data provide further support for the current concept that in (1) pancreatic masses, (2) gastric submucosal masses, (3) lymphoma, and (4) necrotic tumors, TCB increases diagnostic accuracy over FNA alone (Table 3). Clinical judgment is still essential in these cases because FNA of the pancreatic body and tail, for example, carries a risk of possible gastric tumor seeding.13 The advantage of EUS biopsy in subepithelial lesions greater than 5 cm must also be carefully considered because surgery is usually indicated because of the malignancy risk at this size. It must be kept in mind that negative FNA results in presumed GI stromal tumors does not help guide management because sampling error has been shown to occur.14 Our mean number of FNA passes (4.3) was less than our initial goal of 5 passes for lymph nodes and 7 for pancreatic masses. In some high-risk patients, the number of FNA passes was intentionally decreased to shorten the
procedure time. The mean number of FNA C TCB passes was 7, which was within our goal range. The lower mean number of FNA passes probably did not affect the accuracy in our data because most cases missed by FNA were lesions in which cytology is problematic (lymph node with necrosis, lymphoma, and GI stromal tumors). Only one pancreatic cancer was missed by FNA; this patient underwent 7 FNA passes. Among our patients with positive or negative FNA or TCB, there was no significant difference in the number of needle passes performed. Our study included a large number of mediastinal lymph nodes. This is of interest because EUS-TCB through the esophagus can be easier than through the stomach or duodenum because of a straighter scope position. In 2 of our patients, however, TCB still failed to obtain adequate tissue. When EUS-TCB is performed on mediastinal nodes, there are several limitations. First, lymph nodes must have at least 15 to 20 mm length in the path of the Trucut needle to allow full extension of the tissue tray. Second, the proximity of the pulmonary arteries and left atrium in the subcarinal space can represent a challenge for EUSTCB in smaller lesions. It has been suggested that in nodes that have a high suspicion of being metastatic from lung cancer, an initial evaluation with EUS-FNA (in centers with on-site cytopathology) has an accuracy of 94%,15 theoretically obviating the need for EUS-TCB. In our series, in 3 of 16 cases of lung cancer (19%), TCB led to a change the diagnosis over FNA. Further evaluation of TCB in the mediastinum is needed to determine its true benefit when on-site cytopathology is unavailable. Our data suggest that, in patients with failed prior biopsy by bronchoscopy or mediastinoscopy, EUS-guided biopsy is helpful. Prior studies have demonstrated up to a 97% accuracy for EUS-FNA in patients with suspected centrally located lung tumors and 71% in patients with subcarinal lymph nodes in patients who have had prior transbronchial biopsy.16,17 Other reports have described the benefit of EUS in addition to mediastinoscopy for lung cancer staging, with an increase in accuracy of up to 16%.18,19 In our study, 2 cases of lymphoma and 3 metastatic lung cancers with significant necrosis diagnosed by FNA C TCB were previously missed by transbronchial needle biopsy or mediastinoscopy. Further study of EUS biopsy in patients who have previously failed other modalities will be needed to define the role of TCB in these cases. The role of EUS-guided FNA with flow cytometry in patients with suspected lymphoma has been increasingly recognized.20 A diagnosis of lymphoma, however, can be difficult to make on cytologic study alone using the accepted lymphoma classification system (working formulation). Even if a diagnosis of lymphoma is made, subclassification is often impossible21(ie, Hodgkin’s disease). Varadarajulu et al22 reported a case where lymphoma was diagnosed by EUS-FNA, flow cytometry, and TCB. In our series, EUS-TCB changed management in two patients with lymphoma who had flow cytometry
www.giejournal.org
Volume 64, No. 4 : 2006 GASTROINTESTINAL ENDOSCOPY 509
DISCUSSION
Advantages of TCB with FNA without immediate on-site cytopathologic examination
Figure 1. A, EUS demonstrating a fistula within a mass in a patient with dysphagia. B, Trucut specimen defining the tumor as malignant lymphoma, large B-cell type. (H&E, orig. mag. 20.)
performed on an FNA sample (Fig. 1A and B). In one patient with large cell lymphoma, histologic analysis and immunostaining was needed for final diagnostic confirmation of subtype. In the second patient, the presence of intense necrosis led to an initial benign diagnosis by FNA, and a final diagnosis of large cell lymphoma was made by TCB. Our series adds further support to the utility of EUS-TCB for final subclassification in patients with suspected lymphoma. Although TCB C FNA may increase accuracy of EUS tissue sampling when no cytopathologist is present, there are potential drawbacks to combination biopsy. The use of FNA C TCB results in more needle passes per case and higher cost yet still appears safe in our initial review. 510 GASTROINTESTINAL ENDOSCOPY Volume 64, No. 4 : 2006
Storch et al
Only one minor complication (fever and chest pain that resolved with antibiotics) out of 44 procedures was observed (a rate of 2.2%). The complication rate of EUSFNA alone has been stated to be about 0.5%.2 Complications reported in other series of TCB alone include 2 cases of mediastinitis and 1 case of postbiopsy bleeding.8,9 More data on the safety of TCB is still needed. Other problems with the Trucut needle include its stiffness, which makes advancement in certain scope positions challenging, particularly in the duodenum. Although our series does not include cases sampled through the duodenum, the published experience thus far is disappointing.6-11 We have also encountered occasional difficulties in the distal antrum, including misfire of the needle once it is positioned inside the lesion. At times, the long throw of the tissue tray makes obtaining an adequate specimen in the mediastinum difficult. Refinements in the needle, such as a smaller gauge and sturdier design, will be necessary to improve its performance. It was not the intent of our study to determine whether Trucut biopsy can replace FNA C on-site cytopathologic study. It is important to realize that the addition of TCB to standard FNA without an on-site cytopathologist on average costs $200 (based on the Medicare cost) more than EUS-FNA with on-site cytopathologic examination. Because no direct comparison between the accuracy of FNA C on-site cytopathologic study versus FNA C TCB without an on-site cytopathologist has been performed, it is impossible to determine which is more cost-effective. In centers where on-site cytopathologic examination is not available, the question is whether TCB should be done in all cases versus repeat EUS with TCB when the first FNA has failed. Our study demonstrates that the increased accuracy of FNA C TCB over FNA alone was primarily in patients with pancreatic masses, lymphoma, GI stromal tumors, and necrotic lesions. We thus suggest that in centers without on-site cytopathologic capabilities, TCB during the initial EUS be reserved for these lesions and for cases of previously failed EUS-FNA until more data are available. In summary, we have shown overall EUS-TCB accuracy to be, at best, equal to FNA when no on-site cytopathologist is present. The 100% accuracy in this series among patients with malignant diseases, patients who failed prior biopsies by other modalities, or who were refused other procedures because of medical risk supports the use of a combined FNA-TCB in these situations. TCB seems to hold promise in specific clinical scenarios; however, more data will be needed before FNA C TCB can be recommended in all patients when on-site cytopathologic study is unavailable.
DISCLOSURE None of the authors has received any study grants or financial incentive from manufacturers of endoscopic www.giejournal.org
Storch et al
Advantages of TCB with FNA without immediate on-site cytopathologic examination
equipment or other devices mentioned in this study. The authors do not hold any shares of the companies marketing devices used in this study.
REFERENCES 1. Erickson R. EUS-guided FNA. Gastrointest Endosc 2004;60:267-4. 2. Lee LS, Saltzman JR, Bounds BC, et al. EUS-guided fine needle aspiration of pancreatic cysts: a retrospective analysis of complications and their predictors. Clin Gastroenterol Hepatol 2005;3:231-6. 3. Wiersma MJ, Vilmann P, Giovannini M, et al. Endosonography-guided fine-needle aspiration biopsy: diagnostic accuracy and complication assessment. Gastroenterology 1997;112:1087-95. 4. Jhala NC, Jhala DN, Chhiend DC, et al. Endoscopic ultrasound-guided fine-needle aspiration: a cytopathologist’s perspective. Am J Clin Pathol 2003;120:351-67. 5. Hikichi T, Irisawa A, Buthani MS, et al. Endoscopic ultrasound-guided fine-needle aspiration with immediate cytology of solid pancreatic masses by endoscopists [abstract]. Gastrointest Endosc 2005;61: AB281. 6. Levy M, Reddy R, Wiersema M, et al. EUS-guided trucut biopsy in establishing autoimmune pancreatitis as the cause of obstructive jaundice. Gastrointest Endosc 2005;61:467-72. 7. Tam W, Aithal G, Dean J, et al. A prospective multicenter study of the safety and accuracy of combined endoscopic ultrasound (EUS)– guided fine needle aspiration (FNA) and Trucut needle biopsy (TNB) [abstract]. Gastrointest Endosc 2004;59:AB212:T1552. 8. Early D, Edmundowicz S, Azar R, et al. EUS-guided trucut biopsy. Gastroenterology 2003;57:W1233. 9. Varadarajulu S, Fraig M, Schmulewitz N, et al. Comparison of EUSguided 19-gauge Trucut needle biopsy with EUS-guided fine-needle aspiration. Endoscopy 2004;36:397-401. 10. Levy M, Jondal M, Clain J, et al. Preliminary experience with an EUS-guided trucut biopsy needle compared with EUS-guided FNA. Gastrointest Endosc 2003;57:101-6. 11. Larghi A, Verna E, Stavropoulos S, et al. EUS-guided trucut needle biopsies in patients with solid pancreatic masses: a prospective study. Gastrointest Endosc 2004;59:185-90. 12. Eloubeidi MA, Cohn M, Cerfolio RJ, et al. Endoscopic ultrasoundguided fine-needle aspiration in the diagnosis of foregut duplication cysts: the value of demonstrating detached ciliary tufts in cyst fluid. Cancer 2004;102:253-8.
www.giejournal.org
13. Paquin SC, Gariepy G, Lepanto L, et al. A first report of tumor seeding because of EUS-guided FNA of a pancreatic adenocarcinoma. Gastrointest Endosc 2005;61:610-1. 14. Fu K, Eloubeidi MA, Jhala NC, et al. Diagnosis of gastrointestinal stromal tumor by endoscopic ultrasound-guided fine needle aspiration biopsyda potential pitfall. Ann Diagn Pathol 2002;6:294-301. 15. Caddy G, Conron M, Wright G, et al. The accuracy of EUS-FNA in assessing mediastinal lymphadenopathy and staging patients with NSCLC. Eur Respir J 2005;25:410-5. 16. Annema JT, Veselic M, Rabe KF. EUS-guided FNA of centrally located lung tumours following a non-diagnostic bronchoscopy. Lung Cancer 2005;48:357-61. 17. Annema JT, Veselic M, Rabe KF. Analysis of subcarinal lymph nodes in (suspected) non-small-cell lung cancer after a negative transbronchial needle aspirationdwhat’s next? A preliminary report. Respiration 2004;71:630-4. 18. Annema JT, Versteegh MI, Veselic M, et al. Endoscopic ultrasound added to mediastinoscopy for preoperative staging of patients with lung cancer. JAMA 2005;294:931-6. 19. Van Schil P. Accuracy of mediastinoscopy compared to endoscopic ultrasound with fine needle aspiration (EUS-FNA). Lung Cancer 2005;50:273-4. 20. Ribeiro A, Vazquez-Sequeiros E, Wiersma LM, et al. EUS-guided fineneedle aspiration combined with flow cytometry and immunocytochemistry in the diagnosis of lymphoma. Gastrointest Endosc 2001;53:485-91. 21. Hehn S, Grogan T, Miller T. Utility of fine-needle aspiration as a diagnostic technique in lymphoma. J Clin Oncol 2004;22:3046–5. 22. Itoi T, Itokawa F, Sofuni A, et al. Puncture of solid pancreatic tumors guided by endoscopic ultrasonography: a pilot study series comparing Trucut and 19-gauge and 22-gauge aspiration needles. Endoscopy 2005;37:362-6.
Received September 21, 2005. Accepted February 20, 2006. Current affiliations: Division of Gastroenterology (I. S., A. R.), Division of Hematology/Oncology (L. R., C. R.), Department of Surgical Pathology (M. J., S. V.), and Division of Thoracic Surgery (R. T.), Sylvester Comprehensive Cancer Center. Miller School of Medicine, University of Miami, Miami, Florida. Reprint requests: Afonso Ribeiro, MD, Division of Gastroenterology, University of Miami, PO Box 016960, Miami, FL 33101.
Volume 64, No. 4 : 2006 GASTROINTESTINAL ENDOSCOPY 511