- Email: [email protected]
EUS-guided FNA of solid pancreatic masses: a prospective, randomized trial comparing 22-gauge and 25-gauge needles
ORIGINAL ARTICLE: Clinical Endoscopy
EUS-guided FNA of solid pancreatic masses: a prospective, randomized trial comparing 22-gauge and 25-gauge needles Uzma D. Siddiqui, MD, Federico Rossi, MD, Lawrence S. Rosenthal, MD, Manmeet S. Padda, MD, Visvanathan Murali-Dharan, MD, Harry R. Aslanian, MD New Haven, Connecticut, Minneapolis, Minnesota, USA
Background: There is a lack of prospective, randomized studies comparing the diagnostic yield and complication rates of 22-gauge and 25-gauge needles during EUS-FNA of solid pancreatic masses. Objectives: Our primary aim was to compare the diagnostic yield of 22-gauge and 25-gauge needles. Secondary aims included determining the number of needle passes performed, ease of needle passage, and complications. Design: Prospective, randomized study. Setting: Tertiary referral centers at Yale University School of Medicine, New Haven, Connecticut, and Virginia Piper Cancer Institute, Minneapolis, Minnesota. Patients: Patients with a suspected solid pancreatic mass from February 2007 to June 2008 were enrolled. Interventions: Patients were randomized to EUS-FNA with a 22-gauge or 25-gauge needle. Main Outcome Measurements: A diagnostic result was defined as cytology findings positive for malignant cells. Results: A total of 131 patients were enrolled: EUS-FNA was performed with a 22-gauge needle in 64 patients and with a 25-gauge needle in 67 patients. Cytology was diagnostic in 120 (91.6%) of 131 patients overall: 56 (87.5%) of 64 with 22-gauge needles and 64 (95.5%) of 67 with 25-gauge needles (no statistically significant difference was found between the 2 groups; PZ.18). A similar number of passes was performed in both arms (mean [SD] 2.6 [1.2] each; PZ.96). There were no complications in either group. Limitation: A larger number of patients is needed to determine small differences in diagnostic yield. Conclusions: This is the first prospective, randomized trial comparing 22-gauge and 25-gauge needles in EUSFNA of solid pancreatic masses. We achieved equally high diagnostic yields by using a similar number of passes, showing that 25-gauge needles are an effective alternative to 22-gauge needles. (Gastrointest Endosc 2009;70:1093-7.)
EUS provides detailed imaging of the pancreas, and its role in the diagnosis of pancreatic malignancy is widely recognized. The addition of FNA under EUS guidance allows the performance of real-time tissue sampling of pancreatic masses with reported sensitivities between 71% and 90% for diagnosing pancreatic malignancy.1-4 The po-
Abbreviations: ASA, aspirin; NSAIDs, nonsteroidal anti-inflammatory drugs. DISCLOSURE: All authors disclosed no financial relationships relevant to this publication. Copyright ª 2009 by the American Society for Gastrointestinal Endoscopy 0016-5107/$36.00 doi:10.1016/j.gie.2009.05.037
www.giejournal.org
tential risks of performing EUS-guided FNA of pancreatic masses are low and include bleeding, infection, perforation, and pancreatitis at a rate of less than 1%.5-7 Previous studies used a variety of needle sizes, but there has not been a consensus as to which needle size provides the best diagnostic yield during EUS-FNA.4,8,9 EUS-FNA can be performed with 19-gauge, 22-gauge, or 25-gauge needles. The 19-gauge needles are rarely used because they are often cumbersome to pass through the echoendoscope, and there is concern for an increased risk of bleeding. The ability of 25-gauge needles to provide adequate cellularity has been questioned, and therefore 22-gauge needles have been most commonly used for the aspiration of pancreatic masses. This is the first prospective, randomized study to compare the cytologic yield Volume 70, No. 6 : 2009 GASTROINTESTINAL ENDOSCOPY 1093
EUS-guided FNA of solid pancreatic masses
and complication rates of 22-gauge and 25-gauge needles for EUS-FNA of solid pancreatic masses. Our primary study aim was to compare the diagnostic yield of 22-gauge and 25-gauge needles in obtaining a cytologic diagnosis of malignancy. Secondary aims included determining the number of passes needed to obtain a definitive diagnosis, ease of needle passage into the pancreatic mass, and the rate of needle malfunctions and complications.
Siddiqui et al
Capsule Summary What is already known on this topic d
What this study adds to our knowledge d
METHODS
EUS-FNA of pancreatic masses has been accomplished with a variety of needle sizes, but there is no consensus as to which size provides the best diagnostic yield.
In a prospective, randomized trial comparing 22-gauge and 25-gauge needles in EUS-FNA of solid pancreatic masses in 131 patients, no significant difference was found in diagnostic cytology or the number of passes.
Study design This was a prospective, randomized study that was approved by the Human Investigations Committee at both Yale University School of Medicine and the Virginia Piper Cancer Institute. Informed consent for voluntary participation was obtained from all participants.
Patients Patients presenting to Yale New Haven Hospital, New Haven, Connecticut, and the Virginia Piper Cancer Institute, Minneapolis, Minneapolis, from February 2007 to June 2008 with a suspected pancreatic mass were prospectively enrolled. Suspicion of a pancreatic mass was based on clinical symptoms (ie, jaundice, weight loss) and/or radiological findings (ie, abdominal imaging showing a mass, dilated common bile duct, and/or a dilated pancreatic duct). Patients in whom EUS identified a predominantly solid (O60% solid vs cystic) pancreatic mass with defined measurements were randomized to the performance of EUS-FNA with either a 22-gauge or 25-gauge needle. It is our standard clinical practice to perform FNA of all solid pancreatic masses. Patients were excluded if they had predominantly cystic pancreatic lesions, did not have a mass visualized on EUS, or had no malignancy identified on surgical pathology.
EUS technique
ing and cell blocks were prepared later in the cytology laboratory. EUS-FNA was performed by 3 experienced (O1500 EUS procedures) endosonographers.
Data collection Data collected included patient demographics, presenting symptoms, radiographic findings, and procedure details (ie, tumor characteristics, number of needle passes, ease of needle passage, and needle malfunction). Ease of needle passage was the endosonographer’s subjective observation of the needle passage from the echoendoscope and into the mass and described as either ‘‘easy’’ or ‘‘difficult.’’ Needle malfunctions were caused by a clogged needle (which could not be unclogged with saline solution flush or passage of the stylet) or a broken needle (included bending or loss of handle maneuverability), which required the use of a new needle. Information was also gathered regarding the use of nonsteroidal anti-inflammatory drugs (NSAIDs), aspirin (ASA), or anticoagulants within 7 days of the EUS-FNA. Postprocedure complications were defined as any symptoms requiring hospitalization or emergency department evaluation, new or worsening abdominal pain, or signs of internal hemorrhage. The attending cytologist (who was not present during the procedure) was blinded to the needle size used. Patients were contacted by telephone within 7 days to assess for any procedure-related complications and to relay FNA results. A diagnostic result was defined as cytologic findings positive for malignant cells of any type. Cytologic findings negative for malignancy, atypical, and suspicious were considered nondiagnostic.
Patients were randomized to undergo FNA with either a 22-gauge or 25-gauge needle (22G Endocoil or 25G Echotip, Wilson-Cook, Winston Salem, NC) by using a computer-generated random sequence. EUS was performed by using linear array Olympus GFUCT 140 series echoendoscopes (Center Valley, Pa). FNA was performed under EUS guidance with 10 mL of suction applied during aspiration. If the mass was located in the body or tail of the pancreas, the needle was passed through the gastric wall. If the mass was located in the uncinate or head of the pancreas, the needle was passed through the duodenal wall. Cytology technicians were present for all EUS-FNA procedures and reviewed each pass with the endosonographer. Needle passes were performed until sample adequacy was achieved. Diff-Quik histological staining of pancreatic aspirates was performed on-site, whereas Papanicolaou stain-
Based on previous published studies, we estimated that the expected rate of diagnostic yield from EUS-guided FNA by using a 22-gauge needle would be approximately 85%. By using a power of 80% and an a value of .05, it would be necessary to assess 57 tumors with each needle type by using c2 testing to detect a 25% difference in the rate of diagnostic yield between 22-gauge and 25-gauge needles. Direct comparisons of all the variables recorded
1094 GASTROINTESTINAL ENDOSCOPY Volume 70, No. 6 : 2009
www.giejournal.org
Statistical analysis
Siddiqui et al
EUS-guided FNA of solid pancreatic masses
TABLE 1. Baseline patient and tumor characteristics for all 131 patients and both the 22- and 25-gauge needle groups
Sex, female
Total no. (%) patients
22 gauge, no. (%) patients
25 gauge, no. (%) patients
P value*
49 (37.4)
29 (45.3)
20 (29.8)
.07
70.4
69.3
71.5
.29
83 (63.4)
44 (68.8)
39 (58.2)
.21
30.2
29.6
30.7
.59
33 (25.2)
17 (26.6)
16 (23.9)
.72
Mean age (y) Location of mass in pancreatic head Size of pancreatic mass (mm) Use of NSAIDs/ASA/ anticoagulants within 7 d
NSAIDs, Nonsteroidal anti-inflammatory drugs; ASA, aspirin. *P values are for comparison of 22-gauge and 25-gauge needles.
TABLE 2. Cytology diagnoses for 131 patients undergoing EUS-FNA of solid pancreatic masses Total no. patients
22 gauge, no. patients
25 gauge, no. patients
P value*
110
53
57
.72
Neuroendocrine tumor
4
1
3
.62
Pseudopapillary tumor
1
1
0
.49
Metastatic malignancy
5
1
4
.37
Negative
2
1
1
1.00
Suspicious and/or atypical
9
7
2
.09
Adenocarcinoma
*P values are for comparison of 22-gauge and 25-gauge needles.
TABLE 3. Characteristics of needle passage during EUS-FNA
No. needle passes, mean [SD], (range) Easy needle passage, no. patients
25 gauge
2.6 [1.2] (1-7) 51
2.6 [1.2] (1-6) 57
P value .96 .42
Need to change scope position, no. patients
15
10
.22
Needle malfunction, no. patients
11
10
.72
were made overall and between the 22-gauge and 25gauge needle groups. The positive diagnostic rates were estimated from the total population. A c2 analysis was performed for the categorical covariates based on the binary response (diagnostic/nondiagnostic) and by using Yates’ correction for continuity where appropriate. Student’s t test was used for comparison of quantitative variables, including the number of passes performed during EUS-FNA in each needle group. The P values are presented uncorwww.giejournal.org
22 gauge
rected for multiple testing of outcome data arising from individual patients.
RESULTS A total of 133 patients were prospectively enrolled from 2 centers with 3 endosonographers performing the procedures. Two patients with nondiagnostic cytology findings Volume 70, No. 6 : 2009 GASTROINTESTINAL ENDOSCOPY 1095
EUS-guided FNA of solid pancreatic masses
were excluded because they were subsequently found to not have a pancreatic malignancy (1 patient had sclerosing pancreatitis on surgical pathology and 1 had a pancreatic abscess that resolved on radiographic imaging with 1 year of clinical follow-up). Therefore, 131 patients were included for analysis. There were 64 patients in the 22-gauge needle arm and 67 patients in the 25-gauge needle arm. Forty-nine (37.4%) patients were women; the mean (SD) age of the patients was 70.4 years (12.0), average tumor size was 30.2 mm (11.4), and 63.4% of masses were in the pancreatic head. There were no statistically significant differences in baseline patient and tumor characteristics between both arms (Table 1). Thirty-three (25.2%) patients used anticoagulants (warfarin, 6; intravenous heparin, 1; NSAIDs, 7; ASA, 19 with 1 patient also taking clopidogrel) within 7 days of the EUS-FNA. Patients taking warfarin stopped the medication for 3 days before the performance of EUS-FNA and typically restarted the medication on the day after EUS-FNA. The decision about bridging with heparin was made based on the indication of the anticoagulation and the primary physician’s recommendations. The overall diagnostic rate of neoplasm was 91.6% (120/131); the diagnoses are listed in Table 2. The diagnostic yield was 87.5% (56/64) in the 22-gauge needle arm and 95.5% (64/67) in the 25-gauge needle arm. There was no statistically significant difference between the diagnostic yield for neoplasm between the 22-gauge and 25-gauge groups (P Z .18; 95% CI, 0.0148 to 0.1752). Furthermore, there were no differences between the 2 study arms with respect to the number of passes made, ease of needle passage, or needle malfunctions (Table 3). Needle malfunction was caused by bending of the needle or loss of handle maneuverability in 8 patients in both arms, with the remainder of needle malfunctions caused by clogging. There were 11 patients with nondiagnostic cytology findings: atypical (5 patients), suspicious (4 patients), and negative cytology findings (2 patients). Of these 11 patients, 9 patients had malignancy confirmed by either surgical pathology or clinical follow-up. Two patients lacked follow-up data but were included in the overall analysis and were presumed to have malignancy. Seven of the 11 patients had adequate cellularity (obtained with 22-gauge needles in 5 patients and obtained with 25-gauge needles in 2 patients) on cytology specimens and 4 with paucicellular specimens (obtained with 22-gauge needles in 3 patients and obtained with a 25-gauge needle in 1 patient). There were no complications identified in any of the study patients.
DISCUSSION This is the first prospective, randomized study examining the diagnostic yield of 22-gauge and 25-gauge needles 1096 GASTROINTESTINAL ENDOSCOPY Volume 70, No. 6 : 2009
Siddiqui et al
in EUS-FNA of solid pancreatic masses. As is the case at many EUS centers, we had previously routinely used 22-gauge needles for FNA of solid masses because of skepticism regarding the ability of 25-gauge needles to obtain adequate cellularity and few published data supporting their use. The possibility of fewer bleeding complications and fewer bloody aspirates, which may facilitate on-site interpretation and improved ease of needle passage, initiated our interest in using 25-gauge needles. A retrospective analysis of 100 patients, reported by Nguyen et al9 in abstract form, found that 3 fewer needle passes were needed to obtain a diagnosis with a 25-gauge needle compared with a 22-gauge needle. Our data demonstrated an equal number of average needle passes in each group and did not suggest a difference in on-site interpretation with either needle size; however, this was not directly evaluated. We found no significant difference in the diagnostic yield of malignancy between the 2 needle sizes. Both needle sizes were capable of obtaining diagnostic cytology in malignancies other than adenocarcinoma, including 4 metastatic tumors and 3 neuroendocrine tumors diagnosed with 25-gauge needles. There was a trend toward the 25-gauge needle having a higher overall diagnostic yield. A similar finding was also seen in a smaller, nonrandomized study of 24 patients by Sakamoto et al,8 in which there was a trend toward a superior diagnostic yield of 25-gauge needles compared with 22-gauge and 19gauge core needles (Trucut Needle; Wilson-Cook, Winston-Salem, NC). Our overall diagnostic yield of 91.6% is in the upper range of results reported in the multicenter evaluation by Savides et al4 of EUS-FNA yield in pancreatic masses, which identified an average diagnostic accuracy of 71% (range 39%-93%). Data on needle sizes used were not reported in this multicenter study. Factors contributing to our results may include less variation in technique between a smaller number of endosonographers, the high number of EUS-FNA procedures performed annually, and the presence of experienced cytotechnicians during the procedure. A study limitation is sample size calculations based on an assumed difference (because of the lack of previously published comparative data) of 25% between needle sizes, which was greater than the 8% actual difference seen in our study results. Because the observed differences in yield rates were quite different from those assumed in the power calculations, it is noted that the actual power of the sample sizes in this study to detect the observed difference was approximately 50% if tested against a 1-sided alternative and less than 40% with 2-sided testing. A larger sample size would have been needed to detect a smaller difference between the needles’ diagnostic yields. Approximately 20% of our study subjects used NSAIDs or ASA within the week before their EUS-FNA. Despite this, there were no complications, including bleeding, seen in any patients. We advise patients to stop NSAIDs or ASA for 7 days before elective EUS procedures when www.giejournal.org
Siddiqui et al
taken for low-risk conditions. In our current practice, in patients requiring urgent EUS-FNA procedures because of the new finding of a mass lesion or with high-risk cardiovascular conditions, such as recent cardiac stent placement or a history of nonhemorrhagic stroke, EUS-FNA is performed without discontinuing these medications. Although our study was not designed to assess the safety of NSAID/ASA use during EUS-FNA, our data support our previous clinical experience that EUS-FNA can be performed safely in the setting of NSAID and ASA use. Further studies specifically designed to address this issue are warranted. In summary, we found that 25-gauge needles are an effective alternative to 22-gauge needles in obtaining diagnostic EUS-guided aspirates of solid pancreatic tumors. No complications were identified in either group, including patients with recent NSAID or ASA use. Our study results correlate with those of previous studies also showing a trend toward a higher diagnostic yield for 25-gauge needles. Additional study in a larger group of patients will be required to detect small differences in initial (on-site) or final diagnostic yield of malignancy.
EUS-guided FNA of solid pancreatic masses
3. Harewood GC, Wiersema MJ. Endosonography-guided fine needle aspiration biopsy in the evaluation of pancreatic masses. Am J Gastroenterol 2002;97:1386-91. 4. Savides TJ, Donohue M, Hunt G, et al. EUS-guided FNA diagnostic yield of malignancy in solid pancreatic masses: a benchmark for quality performance measurement. Gastrointest Endosc 2007;66:277-82. 5. Eloubeidi MA, Tamhane A, Varadarajulu S, et al. Frequency of major complications after EUS-guided FNA of solid pancreatic masses: a prospective evaluation. Gastrointest Endosc 2006;63:622-9. 6. Affi A, Vazquez-Sequieros E, Norton ID, et al. Acute extraluminal hemorrhage associated with EUS-guided fine needle aspiration: frequency and clinical significance. Gastrointest Endosc 2001;53:221-5. 7. Varadarajulu S, Eloubeidi MA. Frequency and significance of acute intracystic hemorrhage during EUS-FNA of cystic lesions of the pancreas. Gastrointest Endosc 2004;60:631-5. 8. Sakamoto H, Kitano M, Komaki T, et al. Prospective comparative study of the EUS guided 25-gauge FNA needle with the 19-gauge Trucut needle and 22-gauge FNA needle in patients with solid pancreatic masses. J Gastroenterol Hepatol 2009;24:384-90. 9. Nguyen TTH, Lee CE, Whang CS, et al. A comparison of the diagnostic yield and specimen adequacy between 22 and 25 gauge needles for endoscopic ultrasound guided fine needle aspiration (EUS-FNA) of solid pancreatic lesions (SPL): is bigger better? [abstract]. Gastrointest Endosc 2008;67:AB100.
Received March 25, 2009. Accepted May 29, 2009.
REFERENCES 1. Eloubeidi MA, Chen VK, Eltoum IA, et al. Endoscopic ultrasound-guided fine needle aspiration biopsy of patients with suspected pancreatic cancer: diagnostic accuracy and acute and 30-day complications. Am J Gastroenterol 2003;98:2663-8. 2. Gress F, Gottlieb K, Sherman S, et al. Endoscopic ultrasonographyguided fine-needle aspiration biopsy of suspected pancreatic cancer. Ann Intern Med 2001;134:459-64.
www.giejournal.org
Current affiliations: Yale University School of Medicine (U.D.S., L.S.R. M.S.P., M.S.P., V.M.-D., H.R.A.), New Haven, Connecticut, Minnesota Gastroenterology (F.R.), Virginia Piper Cancer Institute, Minneapolis, Minnesota, USA. Reprint requests: Uzma D. Siddiqui, MD, Department of Internal Medicine, Section of Digestive Diseases, Yale University School of Medicine, LMP 1080, 333 Cedar Street, New Haven, CT 06520. If you would like to chat with an author of this article, you may contact him at [email protected].
Volume 70, No. 6 : 2009 GASTROINTESTINAL ENDOSCOPY 1097
EUS-guided FNA of solid pancreatic masses: a prospective, randomized trial comparing 22-gauge and 25-gauge needles Uzma D. Siddiqui, MD, Federico Rossi, MD, Lawrence S. Rosenthal, MD, Manmeet S. Padda, MD, Visvanathan Murali-Dharan, MD, Harry R. Aslanian, MD New Haven, Connecticut, Minneapolis, Minnesota, USA
Background: There is a lack of prospective, randomized studies comparing the diagnostic yield and complication rates of 22-gauge and 25-gauge needles during EUS-FNA of solid pancreatic masses. Objectives: Our primary aim was to compare the diagnostic yield of 22-gauge and 25-gauge needles. Secondary aims included determining the number of needle passes performed, ease of needle passage, and complications. Design: Prospective, randomized study. Setting: Tertiary referral centers at Yale University School of Medicine, New Haven, Connecticut, and Virginia Piper Cancer Institute, Minneapolis, Minnesota. Patients: Patients with a suspected solid pancreatic mass from February 2007 to June 2008 were enrolled. Interventions: Patients were randomized to EUS-FNA with a 22-gauge or 25-gauge needle. Main Outcome Measurements: A diagnostic result was defined as cytology findings positive for malignant cells. Results: A total of 131 patients were enrolled: EUS-FNA was performed with a 22-gauge needle in 64 patients and with a 25-gauge needle in 67 patients. Cytology was diagnostic in 120 (91.6%) of 131 patients overall: 56 (87.5%) of 64 with 22-gauge needles and 64 (95.5%) of 67 with 25-gauge needles (no statistically significant difference was found between the 2 groups; PZ.18). A similar number of passes was performed in both arms (mean [SD] 2.6 [1.2] each; PZ.96). There were no complications in either group. Limitation: A larger number of patients is needed to determine small differences in diagnostic yield. Conclusions: This is the first prospective, randomized trial comparing 22-gauge and 25-gauge needles in EUSFNA of solid pancreatic masses. We achieved equally high diagnostic yields by using a similar number of passes, showing that 25-gauge needles are an effective alternative to 22-gauge needles. (Gastrointest Endosc 2009;70:1093-7.)
EUS provides detailed imaging of the pancreas, and its role in the diagnosis of pancreatic malignancy is widely recognized. The addition of FNA under EUS guidance allows the performance of real-time tissue sampling of pancreatic masses with reported sensitivities between 71% and 90% for diagnosing pancreatic malignancy.1-4 The po-
Abbreviations: ASA, aspirin; NSAIDs, nonsteroidal anti-inflammatory drugs. DISCLOSURE: All authors disclosed no financial relationships relevant to this publication. Copyright ª 2009 by the American Society for Gastrointestinal Endoscopy 0016-5107/$36.00 doi:10.1016/j.gie.2009.05.037
www.giejournal.org
tential risks of performing EUS-guided FNA of pancreatic masses are low and include bleeding, infection, perforation, and pancreatitis at a rate of less than 1%.5-7 Previous studies used a variety of needle sizes, but there has not been a consensus as to which needle size provides the best diagnostic yield during EUS-FNA.4,8,9 EUS-FNA can be performed with 19-gauge, 22-gauge, or 25-gauge needles. The 19-gauge needles are rarely used because they are often cumbersome to pass through the echoendoscope, and there is concern for an increased risk of bleeding. The ability of 25-gauge needles to provide adequate cellularity has been questioned, and therefore 22-gauge needles have been most commonly used for the aspiration of pancreatic masses. This is the first prospective, randomized study to compare the cytologic yield Volume 70, No. 6 : 2009 GASTROINTESTINAL ENDOSCOPY 1093
EUS-guided FNA of solid pancreatic masses
and complication rates of 22-gauge and 25-gauge needles for EUS-FNA of solid pancreatic masses. Our primary study aim was to compare the diagnostic yield of 22-gauge and 25-gauge needles in obtaining a cytologic diagnosis of malignancy. Secondary aims included determining the number of passes needed to obtain a definitive diagnosis, ease of needle passage into the pancreatic mass, and the rate of needle malfunctions and complications.
Siddiqui et al
Capsule Summary What is already known on this topic d
What this study adds to our knowledge d
METHODS
EUS-FNA of pancreatic masses has been accomplished with a variety of needle sizes, but there is no consensus as to which size provides the best diagnostic yield.
In a prospective, randomized trial comparing 22-gauge and 25-gauge needles in EUS-FNA of solid pancreatic masses in 131 patients, no significant difference was found in diagnostic cytology or the number of passes.
Study design This was a prospective, randomized study that was approved by the Human Investigations Committee at both Yale University School of Medicine and the Virginia Piper Cancer Institute. Informed consent for voluntary participation was obtained from all participants.
Patients Patients presenting to Yale New Haven Hospital, New Haven, Connecticut, and the Virginia Piper Cancer Institute, Minneapolis, Minneapolis, from February 2007 to June 2008 with a suspected pancreatic mass were prospectively enrolled. Suspicion of a pancreatic mass was based on clinical symptoms (ie, jaundice, weight loss) and/or radiological findings (ie, abdominal imaging showing a mass, dilated common bile duct, and/or a dilated pancreatic duct). Patients in whom EUS identified a predominantly solid (O60% solid vs cystic) pancreatic mass with defined measurements were randomized to the performance of EUS-FNA with either a 22-gauge or 25-gauge needle. It is our standard clinical practice to perform FNA of all solid pancreatic masses. Patients were excluded if they had predominantly cystic pancreatic lesions, did not have a mass visualized on EUS, or had no malignancy identified on surgical pathology.
EUS technique
ing and cell blocks were prepared later in the cytology laboratory. EUS-FNA was performed by 3 experienced (O1500 EUS procedures) endosonographers.
Data collection Data collected included patient demographics, presenting symptoms, radiographic findings, and procedure details (ie, tumor characteristics, number of needle passes, ease of needle passage, and needle malfunction). Ease of needle passage was the endosonographer’s subjective observation of the needle passage from the echoendoscope and into the mass and described as either ‘‘easy’’ or ‘‘difficult.’’ Needle malfunctions were caused by a clogged needle (which could not be unclogged with saline solution flush or passage of the stylet) or a broken needle (included bending or loss of handle maneuverability), which required the use of a new needle. Information was also gathered regarding the use of nonsteroidal anti-inflammatory drugs (NSAIDs), aspirin (ASA), or anticoagulants within 7 days of the EUS-FNA. Postprocedure complications were defined as any symptoms requiring hospitalization or emergency department evaluation, new or worsening abdominal pain, or signs of internal hemorrhage. The attending cytologist (who was not present during the procedure) was blinded to the needle size used. Patients were contacted by telephone within 7 days to assess for any procedure-related complications and to relay FNA results. A diagnostic result was defined as cytologic findings positive for malignant cells of any type. Cytologic findings negative for malignancy, atypical, and suspicious were considered nondiagnostic.
Patients were randomized to undergo FNA with either a 22-gauge or 25-gauge needle (22G Endocoil or 25G Echotip, Wilson-Cook, Winston Salem, NC) by using a computer-generated random sequence. EUS was performed by using linear array Olympus GFUCT 140 series echoendoscopes (Center Valley, Pa). FNA was performed under EUS guidance with 10 mL of suction applied during aspiration. If the mass was located in the body or tail of the pancreas, the needle was passed through the gastric wall. If the mass was located in the uncinate or head of the pancreas, the needle was passed through the duodenal wall. Cytology technicians were present for all EUS-FNA procedures and reviewed each pass with the endosonographer. Needle passes were performed until sample adequacy was achieved. Diff-Quik histological staining of pancreatic aspirates was performed on-site, whereas Papanicolaou stain-
Based on previous published studies, we estimated that the expected rate of diagnostic yield from EUS-guided FNA by using a 22-gauge needle would be approximately 85%. By using a power of 80% and an a value of .05, it would be necessary to assess 57 tumors with each needle type by using c2 testing to detect a 25% difference in the rate of diagnostic yield between 22-gauge and 25-gauge needles. Direct comparisons of all the variables recorded
1094 GASTROINTESTINAL ENDOSCOPY Volume 70, No. 6 : 2009
www.giejournal.org
Statistical analysis
Siddiqui et al
EUS-guided FNA of solid pancreatic masses
TABLE 1. Baseline patient and tumor characteristics for all 131 patients and both the 22- and 25-gauge needle groups
Sex, female
Total no. (%) patients
22 gauge, no. (%) patients
25 gauge, no. (%) patients
P value*
49 (37.4)
29 (45.3)
20 (29.8)
.07
70.4
69.3
71.5
.29
83 (63.4)
44 (68.8)
39 (58.2)
.21
30.2
29.6
30.7
.59
33 (25.2)
17 (26.6)
16 (23.9)
.72
Mean age (y) Location of mass in pancreatic head Size of pancreatic mass (mm) Use of NSAIDs/ASA/ anticoagulants within 7 d
NSAIDs, Nonsteroidal anti-inflammatory drugs; ASA, aspirin. *P values are for comparison of 22-gauge and 25-gauge needles.
TABLE 2. Cytology diagnoses for 131 patients undergoing EUS-FNA of solid pancreatic masses Total no. patients
22 gauge, no. patients
25 gauge, no. patients
P value*
110
53
57
.72
Neuroendocrine tumor
4
1
3
.62
Pseudopapillary tumor
1
1
0
.49
Metastatic malignancy
5
1
4
.37
Negative
2
1
1
1.00
Suspicious and/or atypical
9
7
2
.09
Adenocarcinoma
*P values are for comparison of 22-gauge and 25-gauge needles.
TABLE 3. Characteristics of needle passage during EUS-FNA
No. needle passes, mean [SD], (range) Easy needle passage, no. patients
25 gauge
2.6 [1.2] (1-7) 51
2.6 [1.2] (1-6) 57
P value .96 .42
Need to change scope position, no. patients
15
10
.22
Needle malfunction, no. patients
11
10
.72
were made overall and between the 22-gauge and 25gauge needle groups. The positive diagnostic rates were estimated from the total population. A c2 analysis was performed for the categorical covariates based on the binary response (diagnostic/nondiagnostic) and by using Yates’ correction for continuity where appropriate. Student’s t test was used for comparison of quantitative variables, including the number of passes performed during EUS-FNA in each needle group. The P values are presented uncorwww.giejournal.org
22 gauge
rected for multiple testing of outcome data arising from individual patients.
RESULTS A total of 133 patients were prospectively enrolled from 2 centers with 3 endosonographers performing the procedures. Two patients with nondiagnostic cytology findings Volume 70, No. 6 : 2009 GASTROINTESTINAL ENDOSCOPY 1095
EUS-guided FNA of solid pancreatic masses
were excluded because they were subsequently found to not have a pancreatic malignancy (1 patient had sclerosing pancreatitis on surgical pathology and 1 had a pancreatic abscess that resolved on radiographic imaging with 1 year of clinical follow-up). Therefore, 131 patients were included for analysis. There were 64 patients in the 22-gauge needle arm and 67 patients in the 25-gauge needle arm. Forty-nine (37.4%) patients were women; the mean (SD) age of the patients was 70.4 years (12.0), average tumor size was 30.2 mm (11.4), and 63.4% of masses were in the pancreatic head. There were no statistically significant differences in baseline patient and tumor characteristics between both arms (Table 1). Thirty-three (25.2%) patients used anticoagulants (warfarin, 6; intravenous heparin, 1; NSAIDs, 7; ASA, 19 with 1 patient also taking clopidogrel) within 7 days of the EUS-FNA. Patients taking warfarin stopped the medication for 3 days before the performance of EUS-FNA and typically restarted the medication on the day after EUS-FNA. The decision about bridging with heparin was made based on the indication of the anticoagulation and the primary physician’s recommendations. The overall diagnostic rate of neoplasm was 91.6% (120/131); the diagnoses are listed in Table 2. The diagnostic yield was 87.5% (56/64) in the 22-gauge needle arm and 95.5% (64/67) in the 25-gauge needle arm. There was no statistically significant difference between the diagnostic yield for neoplasm between the 22-gauge and 25-gauge groups (P Z .18; 95% CI, 0.0148 to 0.1752). Furthermore, there were no differences between the 2 study arms with respect to the number of passes made, ease of needle passage, or needle malfunctions (Table 3). Needle malfunction was caused by bending of the needle or loss of handle maneuverability in 8 patients in both arms, with the remainder of needle malfunctions caused by clogging. There were 11 patients with nondiagnostic cytology findings: atypical (5 patients), suspicious (4 patients), and negative cytology findings (2 patients). Of these 11 patients, 9 patients had malignancy confirmed by either surgical pathology or clinical follow-up. Two patients lacked follow-up data but were included in the overall analysis and were presumed to have malignancy. Seven of the 11 patients had adequate cellularity (obtained with 22-gauge needles in 5 patients and obtained with 25-gauge needles in 2 patients) on cytology specimens and 4 with paucicellular specimens (obtained with 22-gauge needles in 3 patients and obtained with a 25-gauge needle in 1 patient). There were no complications identified in any of the study patients.
DISCUSSION This is the first prospective, randomized study examining the diagnostic yield of 22-gauge and 25-gauge needles 1096 GASTROINTESTINAL ENDOSCOPY Volume 70, No. 6 : 2009
Siddiqui et al
in EUS-FNA of solid pancreatic masses. As is the case at many EUS centers, we had previously routinely used 22-gauge needles for FNA of solid masses because of skepticism regarding the ability of 25-gauge needles to obtain adequate cellularity and few published data supporting their use. The possibility of fewer bleeding complications and fewer bloody aspirates, which may facilitate on-site interpretation and improved ease of needle passage, initiated our interest in using 25-gauge needles. A retrospective analysis of 100 patients, reported by Nguyen et al9 in abstract form, found that 3 fewer needle passes were needed to obtain a diagnosis with a 25-gauge needle compared with a 22-gauge needle. Our data demonstrated an equal number of average needle passes in each group and did not suggest a difference in on-site interpretation with either needle size; however, this was not directly evaluated. We found no significant difference in the diagnostic yield of malignancy between the 2 needle sizes. Both needle sizes were capable of obtaining diagnostic cytology in malignancies other than adenocarcinoma, including 4 metastatic tumors and 3 neuroendocrine tumors diagnosed with 25-gauge needles. There was a trend toward the 25-gauge needle having a higher overall diagnostic yield. A similar finding was also seen in a smaller, nonrandomized study of 24 patients by Sakamoto et al,8 in which there was a trend toward a superior diagnostic yield of 25-gauge needles compared with 22-gauge and 19gauge core needles (Trucut Needle; Wilson-Cook, Winston-Salem, NC). Our overall diagnostic yield of 91.6% is in the upper range of results reported in the multicenter evaluation by Savides et al4 of EUS-FNA yield in pancreatic masses, which identified an average diagnostic accuracy of 71% (range 39%-93%). Data on needle sizes used were not reported in this multicenter study. Factors contributing to our results may include less variation in technique between a smaller number of endosonographers, the high number of EUS-FNA procedures performed annually, and the presence of experienced cytotechnicians during the procedure. A study limitation is sample size calculations based on an assumed difference (because of the lack of previously published comparative data) of 25% between needle sizes, which was greater than the 8% actual difference seen in our study results. Because the observed differences in yield rates were quite different from those assumed in the power calculations, it is noted that the actual power of the sample sizes in this study to detect the observed difference was approximately 50% if tested against a 1-sided alternative and less than 40% with 2-sided testing. A larger sample size would have been needed to detect a smaller difference between the needles’ diagnostic yields. Approximately 20% of our study subjects used NSAIDs or ASA within the week before their EUS-FNA. Despite this, there were no complications, including bleeding, seen in any patients. We advise patients to stop NSAIDs or ASA for 7 days before elective EUS procedures when www.giejournal.org
Siddiqui et al
taken for low-risk conditions. In our current practice, in patients requiring urgent EUS-FNA procedures because of the new finding of a mass lesion or with high-risk cardiovascular conditions, such as recent cardiac stent placement or a history of nonhemorrhagic stroke, EUS-FNA is performed without discontinuing these medications. Although our study was not designed to assess the safety of NSAID/ASA use during EUS-FNA, our data support our previous clinical experience that EUS-FNA can be performed safely in the setting of NSAID and ASA use. Further studies specifically designed to address this issue are warranted. In summary, we found that 25-gauge needles are an effective alternative to 22-gauge needles in obtaining diagnostic EUS-guided aspirates of solid pancreatic tumors. No complications were identified in either group, including patients with recent NSAID or ASA use. Our study results correlate with those of previous studies also showing a trend toward a higher diagnostic yield for 25-gauge needles. Additional study in a larger group of patients will be required to detect small differences in initial (on-site) or final diagnostic yield of malignancy.
EUS-guided FNA of solid pancreatic masses
3. Harewood GC, Wiersema MJ. Endosonography-guided fine needle aspiration biopsy in the evaluation of pancreatic masses. Am J Gastroenterol 2002;97:1386-91. 4. Savides TJ, Donohue M, Hunt G, et al. EUS-guided FNA diagnostic yield of malignancy in solid pancreatic masses: a benchmark for quality performance measurement. Gastrointest Endosc 2007;66:277-82. 5. Eloubeidi MA, Tamhane A, Varadarajulu S, et al. Frequency of major complications after EUS-guided FNA of solid pancreatic masses: a prospective evaluation. Gastrointest Endosc 2006;63:622-9. 6. Affi A, Vazquez-Sequieros E, Norton ID, et al. Acute extraluminal hemorrhage associated with EUS-guided fine needle aspiration: frequency and clinical significance. Gastrointest Endosc 2001;53:221-5. 7. Varadarajulu S, Eloubeidi MA. Frequency and significance of acute intracystic hemorrhage during EUS-FNA of cystic lesions of the pancreas. Gastrointest Endosc 2004;60:631-5. 8. Sakamoto H, Kitano M, Komaki T, et al. Prospective comparative study of the EUS guided 25-gauge FNA needle with the 19-gauge Trucut needle and 22-gauge FNA needle in patients with solid pancreatic masses. J Gastroenterol Hepatol 2009;24:384-90. 9. Nguyen TTH, Lee CE, Whang CS, et al. A comparison of the diagnostic yield and specimen adequacy between 22 and 25 gauge needles for endoscopic ultrasound guided fine needle aspiration (EUS-FNA) of solid pancreatic lesions (SPL): is bigger better? [abstract]. Gastrointest Endosc 2008;67:AB100.
Received March 25, 2009. Accepted May 29, 2009.
REFERENCES 1. Eloubeidi MA, Chen VK, Eltoum IA, et al. Endoscopic ultrasound-guided fine needle aspiration biopsy of patients with suspected pancreatic cancer: diagnostic accuracy and acute and 30-day complications. Am J Gastroenterol 2003;98:2663-8. 2. Gress F, Gottlieb K, Sherman S, et al. Endoscopic ultrasonographyguided fine-needle aspiration biopsy of suspected pancreatic cancer. Ann Intern Med 2001;134:459-64.
www.giejournal.org
Current affiliations: Yale University School of Medicine (U.D.S., L.S.R. M.S.P., M.S.P., V.M.-D., H.R.A.), New Haven, Connecticut, Minnesota Gastroenterology (F.R.), Virginia Piper Cancer Institute, Minneapolis, Minnesota, USA. Reprint requests: Uzma D. Siddiqui, MD, Department of Internal Medicine, Section of Digestive Diseases, Yale University School of Medicine, LMP 1080, 333 Cedar Street, New Haven, CT 06520. If you would like to chat with an author of this article, you may contact him at [email protected].
Volume 70, No. 6 : 2009 GASTROINTESTINAL ENDOSCOPY 1097