Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Jul 1.
Published in final edited form as: Gastrointest Endosc. 2020 Feb 25;92(1):108–119.e3. doi: 10.1016/j.gie.2020.02.021

EUS-guided fine-needle biopsy versus fine-needle aspiration in the diagnosis of subepithelial lesions: a large multicenter study

Diogo T H de Moura 1,2,3, Thomas R McCarty 1,2, Pichamol Jirapinyo 1,2, Igor B Ribeiro 3, Victor K Flumignan 4, Fedaa Najdawai 2,5, Marvin Ryou 1,2, Linda S Lee 1,2, Christopher C Thompson 1,2
PMCID: PMC7340004  NIHMSID: NIHMS1593895  PMID: 32105712

Abstract

Background and Aims

Although conventional endoscopic ultrasound-guided fine needle aspiration (EUS-FNA) has previously been considered first-line for sampling subepithelial lesions (SELs), variable accuracy has resulted in increased use of fine-needle biopsy (FNB) to improve diagnostic yield. The primary aim of this study was to compare FNA versus FNB for diagnosis of SEL.

Methods

This was a multicenter, retrospective study to evaluate the outcomes of EUS-FNA and EUS-FNB of SELs over a 3-year period. Demographics, lesion characteristics, sensitivity, specificity, accuracy, number of needle passes, diagnostic adequacy of rapid on-site evaluation (ROSE), cell-block accuracy, as well as adverse events were analyzed. Subgroup analyses were performed comparing FNA versus FNB by location as well as diagnostic yield with or without ROSE. Multivariable logistic regression was also performed.

Results

A total of 229 patients with SELs (n=115 FNA and n=114 FNB) underwent EUS-guided sampling. Mean age was 60.86±12.84 years. Most lesions were gastric in location (75.55%) and from the fourth layer (71.18%). Cell-block for FNB required fewer passes to achieve conclusive diagnosis (2.94±1.09 versus 3.55±1.55; P=0.003). Number of passes were not different for ROSE adequacy (P=0.167). Immunohistochemistry (IHC) was more able to be successfully performed in more FNB samples (69.30% versus 40.00%; P<0.001). Overall, sensitivity and accuracy were superior for FNB versus FNA ([79.41% versus 51.92%; P=0.001] and [88.03% versus 77.19%; P=0.030]). On the subgroup analysis, sensitivity and accuracy of FNB alone was superior to FNA+ROSE ([79.03% versus 46.67%; P=0.001] and [87.25% versus 68.00%; P=0.024]). There was no significant difference in diagnostic yield of FNB alone versus FNB+ROSE (P>0.05). Multivariate analysis showed no predictors associated with accuracy. One minor adverse event was reported in the FNA group.

Conclusions

EUS-FNB was superior to EUS-FNA in the diagnosis of SELs. EUS-FNB was also superior to EUS-FNA alone and EUS-FNA+ROSE. These results suggest EUS-FNB should be considered a first-line modality and may suggest a reduced role for ROSE in the diagnosis of SELs. However, a large randomized controlled trial is required to confirm our findings.

Keywords: EUS-guided tissue acquisition, subepithelial lesions, endoscopic ultrasound, endoscopy, FNA, FNB, mass, tumor, gastrointestinal

INTRODUCTION

Subepithelial lesions (SELs) are defined as being located under the mucosa and originating from the gastrointestinal wall or from extrinsic compression by an adjacent organ. SELs include a wide spectrum of lesions ranging from benign, such as lipoma and heterotopic pancreas, to malignant conditions, such as gastrointestinal stromal tumors (GISTs) and leiomyosarcomas (1–3). Frequently, these lesions are identified during esophagogastroduodenoscopy (EGD); however, precise diagnosis during EGD is challenging given inadequate tissue sampling due to forceps biopsy (4). Endoscopic ultrasound (EUS) is an alternative modality that provides information about the shape, echogenicity, layer of origin, and allows tissue sampling for definitive diagnosis (5–7). EUS-guided tissue sampling can generally be performed in 3 ways: fine-needle aspiration (FNA), trucut biopsy (TCB), and fine-needle biopsy (FNB) (6–8).

Endoscopic ultrasound-guided fine-needle aspiration (EUS-FNA) usually provides cytological specimens instead of a histologic sample. This occurs due to its insufficient tissue acquisition during the procedure that sometimes is not enough for immunohistochemical (IHC) and biomolecular testing, thus decreasing EUS-FNA’s diagnostic power and potentially limiting its use in SELs (7). Similar to the EUS-FNA, the EUS-TCB possesses limited diagnostic yield due to technical difficulties with tissue acquisition. The rate of SEL diagnosis from samples obtained using these modalities ranges from 52% to 82% (8–11).

To overcome the deficiency of these techniques, a new type of needle, fine-needle biopsy (FNB), was designed. This type of needle allows for not simply cytologic examination, but histologic evaluation as well (6). Different from conventional FNA needles, this needle type has the capacity to preserve tissue architecture and potentially lead to improved diagnostic results. These characteristics are pivotal, especially with regard to accurate diagnosis of SELs, where highly detailed IHC evaluations are essential for the differentiation of GIST from other spindle cell tumors (11,12).

Even though EUS-FNB appears to be a better, or at least more-complete method for tissue sampling as compared with EUS-FNA, some studies have not confirmed this finding with variable data to date (6,13). The most recent guideline from the European Society of Gastrointestinal Endoscopy (ESGE) equally recommends FNA and FNB for tissue sampling of SELs with no comment regarding a preferred strategy of SEL sampling (14). This may be due to a paucity of large, high-quality studies related to SEL tissue sampling. Therefore, we aimed to perform the largest multicenter study to specifically compare EUS-FNA versus EUS-FNB for patients with SELs.

METHODS

Material and methods

This was a multicenter, retrospective study conducted at 5 hospitals in Massachusetts, United States (Brigham and Women’s Hospital, Massachusetts General Hospital, Brigham and Women’s Faulkner Hospital, Newton-Wellesley Hospital, and North Shore Medical Center) following the STAndards for the Reporting of Diagnostic accuracy studies (STARD) recommendations (appendix 1). All hospitals were affiliated with Partners Healthcare, although each hospital encompasses different physician groups with different EUS sampling practice protocols and varied levels of experience. The study was approved by the IRB from Partners Human Research (2003P001665). All consecutive patients, age ≥18 years, that had undergone EUS-guided tissue acquisition of SEL from January 2016 to January 2019 were identified from a shared prospective registry. Search terms included: endoscopic ultrasound AND subepithelial lesion OR solid OR mass OR lesion AND fine needle aspiration OR fine needle biopsy. From this initial search, the institutional database was queried for information on patient demographics (ie, sex, age, and comorbidities), lesion characteristics (ie, location, size, shape, layer of origin, heterogeneity, and echogenicity), and procedure details (ie, route of tissue sampling, size and type of the needle, number of passes, diagnostic adequacy of cytological specimen on rapid on-site evaluation [ROSE] when it was available, and diagnostic adequacy on cell-block and on slide examination). Additionally, the database contains information regarding patient follow-up, including adverse events, other diagnostic methods, oncological treatment, and surgery. Patients with incomplete reporting data or cases with more than one needle used were excluded from this analysis.

Procedural technique

All EUS-guided tissue sampling procedures were performed with a linear array echoendoscope (Olympus GF-UCT180, Olympus, Center Valley, Pa, USA) under deep sedation with monitored anesthesia care by experienced endosonographers or by gastroenterology fellows under direct, expert supervision. Several different needles were used during this period, including 19-gauge, 22-gauge, and 25-gauge FNA needles (Expect, Boston Scientific Corporation, Natick, Mass, USA or Echotip, Cook Medical, Winston-Salem, NC, USA or Beacon, Medtronic Corporation, Newton, Mass, USA) and 19-gauge, 20-gauge, 21-gauge, 22-gauge, and 25-gauge FNB needles (Acquire, Boston Scientific Corporation, Natick, Mass, USA, or SharkCore, Medtronic Corporation, Newton, Mass, USA, or ProCore, Cook Medical, Winston-Salem, NC, USA). Cases in which more than one needle was used were not included in this study to avoid bias. After the lesion was identified and punctured under EUS guidance, a general fanning technique was performed. Individual operator technique varied from each center, including stylet slow-pull technique and standard suction as well as number of to-and-fro movements with each needle pass.

Rapid on-site evaluation

ROSE, which may be used to check sample adequacy and establish a preliminary diagnosis using a rapid stain, was not available in all cases. In cases where ROSE was performed, FNA specimens were expressed onto slides and then smeared for onsite preparation. FNB specimens were prepared using the touch imprint technique. To accomplish this technique, the tissue surface was slighted pressed onto the slides before staining to reduce the creation of crushing artifacts. All slides were prepared using both wet-fixed (placed in 96% ethyl alcohol for Papanicolaou staining) and air-dried (in some cases stained with Diff-Quik) techniques. Per pass adequacy for ROSE was assessed in real-time when available with the minimum number of needle passes recorded to achieve adequate sampling.

FNA evaluation

The samples obtained through EUS-FNA were transferred to 3 to12 slides. Each smear was made with slight pressure to avoid crushing artifacts with half of the slides being placed immediately in the 96% ethyl alcohol solution and the others fixed in the air. When possible, part of the material was placed in formalin solution for the preparation of the cell-block. The specimen was subsequently sent to the pathology division, for processing and staining by Papanicolaou method (ie, slides in alcohol solution), Diff-Quick Staining Protocol (ie, air-dried slides), and hematoxylin and eosin stain (ie, cell-block). After these processes, the cytological material was examined by experienced cytopathologists.

FNB evaluation

FNB samples were fixed in buffered formalin (10% formalin) and dehydrated before being embedded in paraffin. The tissue was then sliced into a 4- to 6-μm sections and stained with hematoxylin-eosin. Additionally, in some cases, FNB specimens were prepared in slides using the touch imprint technique. All analyses were performed by experienced pathologists.

Immunohistochemical evaluation

IHC staining was performed for differential diagnosis of SELs. For spindle cell lesions, IHC staining for c-kit, CD34, CD117, DOG-1, a-SMA, Desmin, and S-100 protein was performed.

Outcome measures

The primary outcome was the diagnostic yield (sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), positive likelihood ratio (+LR), negative likelihood ratio (−LR), and accuracy) of EUS-FNA and EUS-FNB from cytologic or histologic analysis with and without IHC staining. Inconclusive specimen results were considered as non-neoplastic lesions as to not overestimate diagnostic yield. Secondary outcomes included the proportion of adequate cellularity for ROSE evaluation, median number of needle passes, diagnostic result from histologic (cell-block) and cytologic (slides) analysis, as well as adverse events related to the procedure. Surgical resection pathology was considered the reference standard for comparison to determine diagnostic accuracy. However, because not all patients underwent surgery, patient follow-up and IHC results were also considered as a reference standard.

Statistical analyses

Baseline patient characteristics and procedure characteristics were summarized as means ± standard deviation for continuous data and frequencies and proportions for categorical data. As diagnostic tests were performed on 2 independent groups of patients, a bivariate model was used to compute the pooled sensitivity and specificity, and diagnostic accuracy. Two-sample t-tests for binomial proportions were used (15). Continuous data were compared using the 2-sample t-test or Wilcoxon rank-sum test and categorical data were compared using the Chi-square or the Fisher exact test as appropriate (16). Statistical significance was defined as a P<0.05.

Subgroup analyses were then performed to evaluate diagnostic yield of FNA and FNB for each location (esophageal, gastric, and duodenal) of SEL. Additional analyses were also performed to identify the diagnostic yield of FNA alone, FNA with ROSE, FNB alone, and FNB with ROSE. From this data, sensitivity, specificity, PPV, NPV, +LR, −LR, and accuracy were compared with determine if ROSE was beneficial. In an effort to identify factors associated with diagnostic performance between FNA and FNB needle types, a multivariable logistic regression was performed with adjustment for clinically significant univariate findings as well as age, gender, number of passes, needle size, needle type, and application of ROSE, cell-block, and IHC. Results of the regression analysis were expressed as beta-coefficient (β) and odds ratio (OR). Statistical analyses were performed using the Stata 15.0 software package (Stata Corp LP, College Station, Tex, USA).

RESULTS

Baseline Patient and Lesion Characteristics

A total of 247 patients were enrolled in this study between January 2016 to January 2019. Of these, 18 were excluded because more than one needle was used. Therefore, 229 patients (116 female and 113 male) were included in the analyses. Mean age of patients was 60.86 ±12.84 years old. There was no significant age or gender difference noted between FNA and FNB cohorts (P=0.152 and P=0.844, respectively). Of the 229 patients, 115 underwent FNA and 114 FNB. Technical success occurred in all cases. A majority of SELs were located in the stomach (75.55%), followed by esophagus (11.79%), then duodenum (7.86%) and rectum (1.74%). Among lesions located within the stomach, gastric body was the most common region (46.34%). Mean size of all lesions was 21.22 ±13.27 mm with similar sizes between FNA and FNB patients. A total of 15 lesions were sub-centimeter (ie, <10 mm) in size [5 lesions with FNA sampling and 10 lesions with FNB sampling; P=0.178]. SELs were also most commonly located in the 4th layer of tissue (71.18%). Additional baseline characteristics for all included patients as well as stratification by FNA or FNB cohort are demonstrated in Table 1.

Table 1.

Baseline Patient and Lesion Characteristics

Results Total FNA FNB P value
Patient Characteristics
 No. of Patients 229 115 114
 Age (years) 60.86 (12.84) 59.64 (12.10) 62.07 (13.37) 0.152
 Gender 0.844
 No. of Males (%) 113 (49.34%) 56 (49.55%) 57 (50.44%)
 No. of Females (%) 116 (50.65%) 59 (50.86%) 57 (50.44%)
Lesion site 0.141
 Esophagus 27 (11.79%) 16 (13.91%) 11 (9.64%)
 GEJ 7 (3.05) 5 (4.35%) 2 (1.76%)
 Stomach 173 (75.55%) 81 (70.43%) 92 (80.70%)
 Duodenum 18 (7.86%) 9 (7.83%) 9 (7.90%)
 Rectum 4 (1.74%) 4 (3.48%) 0 (0.00%)
Gastric lesion site 0.756
 Cardia 29 (13.29%) 14 (16.05%) 15 (16.30%)
 Fundus 28 (16.76%) 11 (13.58%) 17 (19.57%)
 Body 80 (46.34%) 39 (48.15%) 41 (44.56%)
 Antrum 36 (30.81%) 18 (22.22%) 18 (19.57%)
Origin Layer 0.844
 2nd 30 (13.10%) 14 (12.18%) 16(14.04%)
 3rd 36 (15.72%) 19 (16.52%) 17 (14.91%)
 4th 163 (71.18%) 82 (71.30%) 81 (79.05%)
Lesion size 21.22 (13.27) 21.29 (14.87) 21.15 (11.52) 0.939
Open in a new tab

Needle and Sampling Characteristics

Multiple needle sizes were used in this study, including 19-gauge, 20-gauge, 21-gauge, 22-gauge, and 25-gauge. Of these, 22-gauge and 25-gauge were more commonly used (67.84% and 27.75%, respectively). The needle size was similar regardless of needle type (P=0.225). Despite similar size and location of SELs, as well as needle gauges, FNB resulted in fewer number of passes (2.89 ±1.12 versus 3.44 ±1.53; P=0.002). ROSE was performed on 14.47% of total SEL samples with a comparable percent of samples being adequate for evaluation (FNA: 83.33% versus FNB: 80.95%; P=0.507). Furthermore, the number of passes for ROSE adequacy was similar between both FNA (4.53 ±1.28) and FNB (3.85 ±1.34) (P=0.167). Cell-block was more common among FNB samples (96.49% versus 80.87%; P<0.001) with fewer number of passes required to achieve a conclusive diagnosis (2.94 ±1.09 versus 3.5 ±1.55; P=0.003). IHC staining was also more common among FNB samples (69.30% versus 40.00%; P<0.001) – Figure 1 and Figure 2. A further breakdown of needle type and sampling characteristics is illustrated in Table 2.

Figure 1.

Figure 1.

Figure 1.

Open in a new tab

Hemotoxylin and eosin staining as well as immunohistochemical staining of FNA-obtained gastric lesion demonstrating gastrointestinal stromal tumor (GIST). A, H&E, low power (orig. mag. x4). B, H&E, high power (orig. mag. x40). C, Negative staining for desmin. D, Positive staining for discovered on GIST (DOG)-1. E, Negative staining for tyrosine-protein kinase KIT. F, Negative staining for smooth muscle actin (SMA).

Figure 2.

Figure 2.

Figure 2.

Open in a new tab

Hemotoxylin and eosin staining as well as immunohistochemical staining of FNB-obtained esophageal lesion demonstrating leiomyoma. A, H&E, low power (orig. mag. x4). B, H&E, high power (orig. mag. x40). C, Positive staining for desmin. D, Negative staining for discovered on GIST (DOG)-1. E, Negative staining for tyrosine-protein kinase KIT. F, Positive staining for smooth muscle actin (SMA).

Table 2.

Needle Type and Sample Characteristics

Results Total FNA FNB P value
Needle Size 0.225
 19-gauge 7 (3.09%) 2 (1.73%) 5 (4.39%)
 20-gauge 2 (0.88%) 0 (0.00%) 2 (1.75%)
 21-gauge 1 (0.44%) 0 (0.00%) 1 (0.88%)
 22-gauge 154 (67.84%) 77 (66.95%) 79 (69.30%)
 25-gauge 63 (27.75%) 36 (31.30%) 27 (23.68%)
No of Passes 3.17 (1.37) 3.44 (1.53) 2.89 (1.12) 0.002
No. of Samples with ROSE 0.087
 Yes 40 (14.47%) 25 (21.74%) 15 (13.16%)
 No 189 (85.53%) 90 (78.26%) 99 (86.84%)
Adequate Sample for ROSE 0.507
 Yes 30 (83.33%) 17 (80.95%) 13 (86.67%)
 No 6 (16.67%) 4 (19.05%) 2 (13.33%)
No. of Passes for ROSE Adequacy 4.23 (1.33) 4.53 (1.28) 3.85 (1.34) 0.167
No. of Samples with Cell-block <0.001
 Yes 203 (88.65%) 93 (80.87%) 110 (96.49%)
 No 26 (11.35%) 22 (19.13%) 4 (3.51%)
No. of Passes for Cell-block Diagnosis 3.20 (1.35) 3.55 (1.55) 2.94 (1.09) 0.003
No. of Samples with IHC <0.001
 Yes 125 (54.59%) 46 (40.00%) 79 (69.30%)
 No 104 (45.41%) 69 (60.00%) 35 (30.70%)
Diagnostic Findings with IHC <0.001
 GIST 49 (39.20%) 15 (32.62%) 34 (43.05%)
 Leiomyoma 34 (27.20%) 13 (28.26%) 21 (26.58%)
 Adenocarcinoma* 6 (4.80%) 1 (2.17%) 5 (6.33%)
 Lymphoma 2 (1.60%) 0 (0.00%) 2 (2.53%)
 Other 20 (16.00%) 11 (23.91%) 9 (11.39%)
 Inconclusive 14 (11.20%) 6 (13.04%) 8 (10.12%)
Open in a new tab
*

All adenocarcinomas were metastatic lesions arising from the second layer.

Comparison of Diagnostic Accuracy

Overall sensitivity, specificity, and accuracy of FNA and FNB for SELs was 67.50% (95% CI, 58.35 to 75.77), 99.10% (95% CI, 95.08 to 99.98), and 82.68% (95% CI, 77.18 to 87.33). Compared with FNA, FNB resulted in significantly better sensitivity [79.41% (95% CI, 67.8 to 88.26) versus 51.92% (95% CI, 37.63 to 65.99); P=0.001] and accuracy [88.03% (95% CI, 80.74 to 93.30) versus 77.19% (95% CI, 68.40 to 84.53); P=0.0300]. Both methods presented similar high specificity [100.00% (95% CI, 92.75 to 100.00) versus 98.39% (95% CI, 91.34 to 99.96); P=0.279]. Adverse events were minimal (one minor hemorrhage after puncture treated with epinephrine injection and hemostatic clip placement) and not statistically different between FNA and FNB (0.87% versus 0.00%; P=0.321). Complete diagnostic test characteristics are shown in Table 3.

Table 3.

Comparison of Diagnostic Accuracy

Diagnostic Test Characteristics Total FNA FNB P value
Sensitivity 67.50% (95% CI, 58.35 to 75.77) 51.92% (95% CI, 37.63 to 65.99) 79.41% (95% CI, 67.8 to 88.26) 0.001
Specificity 99.10% (95% CI, 95.08 to 99.98) 98.39% (95% CI, 91.34 to 99.96) 100.00% (95% CI, 92.75 to 100.00) 0.279
Positive Likelihood Ratio 74.92 (95% CI, 10.61 to 529.32) 32.19 (95% CI, 4.53 to 228.92) NA NA
Negative Likelihood Ratio 0.33 (95% CI, 0.25 to 0.42) 0.49 (95% CI, 0.37 to 0.65) 0.21 (95% CI, 0.13 to 0.33) 0.002
Positive Predictive Value 98.78% (95% CI, 91.98 to 99.83) 96.43% (95% CI, 79.15 to 99.48) 100.00% 0.039
Negative Predictive Value 73.83% (95% CI, 68.53 to 78.51) 70.93% (95% CI, 64.74 to 76.43) 77.78% (95% CI, 68.70 to 84.81) 0.234
Accuracy 82.68% (95% CI, 77.18 to 87.33) 77.19% (95% CI, 68.40 to 84.53) 88.03% (95% CI, 80.74 to 93.30) 0.030
Adverse Events 1 (0.44) 1 (0.87) 0 (0.00) 0.321
Open in a new tab

Subgroup Analyses

Cumulative data was then stratified by location of SEL (ie, esophagus, gastric, and duodenum). For esophageal lesions, FNB had a significantly higher sensitivity (FNB 66.67% [95% CI, 22.28 to 95.67], versus FNA 00.00% [95% CI, 0.00 to 97.50], P<0.001). There was no difference between FNA and FNB with regard to specificity or accuracy (P>0.05). For gastric lesions, FNB demonstrated a significant better sensitivity and accuracy as compared with FNA (81.03% [95% CI, 68.59 to 90.13] versus 56.10% [95% CI, 39.75 to 71.53]; P<0.001) and (89.62% [95% CI, 82.19 to 94.70] versus 77.50% [95% CI, 66.79 to 86.09]; P=0.025), respectively]. There was no difference in diagnostic yield among SELs located in the duodenum (P>0.05) (Table 4). Comparison of FNA versus FNB for sub-centimeter lesions revealed FNB to be associated with better diagnostic accuracy; however, this difference was not significant (66.67% [95% CI, 9.43 to 99.16%] versus 81.82% [95% CI, 48.22 to 97.72%]; P=0.527).

Table 4.

Subgroup Analyses by Location

Esophageal Total FNA FNB P value
Sensitivity 57.14% (95% CI, 18.41 to 90.10) 0.00% (95% CI, 0.00 to 97.50) 66.67% (95% CI, 22.28 to 95.67) <0.001
Specificity 95.00% (95% CI, 75.13 to 99.87) 93.33% (95% CI, 68.05 to 100.00) 100.00% (95% CI, 47.82 to 100.00) 0.391
Positive Likelihood Ratio 11.43 (95% CI, 1.52 to 85.74) 0.00% NA NA
Negative Likelihood Ratio 0.45 (95% CI, 0.19 to 1.07) 1.07 (95% CI, 0.94 to 1.23) 0.33 (95% CI, 0.31 to 1.03) 0.832
Positive Predictive Value 80.00% (95% CI, 34.78 to 96.78) 0.00% 100.00% <0.001
Negative Predictive Value 86.36% (95% CI, 72.80 to 93.74) 93.33% (95% CI, 92.44 to 94.13) 71.43% (95% CI, 44.64 to 88.57) 0.13
Accuracy 85.19% (95% CI, 66.27 to 95.81) 87.50% (95% CI, 61.65 to 98.45) 81.82% (95% CI, 48.22 to 97.72) 0.689
Gastric
Sensitivity 70.71% (95% CI, 60.71 to 79.43) 56.10% (95% CI, 39.75 to 71.53) 81.03% (95% CI, 68.59 to 90.13) <0.001
Specificity 100.00% (95% CI, 95.85 to 100.00) 100.00% (95% CI, 90.97 to 100.00) 100.00% (95% CI, 92.60 to 100.00) NA
Positive Likelihood Ratio NA NA NA NA
Negative Likelihood Ratio 0.29 (95% CI, 0.22 to 0.40) 0.44 (95% CI, 0.31 to 0.62) 0.19 (95% CI, 0.11 to 0.32) 0.914
Positive Predictive Value 100.00% 100.00% 100.00% NA
Negative Predictive Value 75.00% (95% CI, 68.84 to 80.29) 68.42% (95% CI, 60.52 to 75.38) 81.36% (95% CI, 71.94 to 88.14) 0.042
Accuracy 84.41% (95% CI, 78.38 to 89.30) 77.50% (95% CI, 66.79 to 86.09) 89.62% (95% CI, 82.19 to 94.70) 0.025
Duodenum
Sensitivity 50.00% (95% CI, 18.71 to 81.29) 33.33% (95% CI, 4.33 to 77.72) 75.00% (95% CI, 19.41 to 99.37) 0.076
Specificity 100.00% (95% CI, 66.37 to 100.00) 100.00% (95% CI, 29.24 to 100.00) 100.00% (95% CI, 54.07 to 100.00) NA
Positive Likelihood Ratio NA NA NA NA
Negative Likelihood Ratio 0.50 (95% CI, 0.27 to 0.93) 0.67 (95% CI, 0.37 to 1.17) 0.25 (95% CI, 0.05 to 1.36) 0.894
Positive Predictive Value 100.00% 100.00% 100.00% NA
Negative Predictive Value 64.29% (95% CI, 49.20 to 76.99) 42.86% (95% CI, 29.87 to 56.91) 85.71% (95% CI, 52.36 to 97.04) 0.056
Accuracy 73.68% (95% CI, 48.80 to 90.85) 55.56% (95% CI, 21.20 to 86.30) 90.00% (95% CI, 55.50 to 99.75) 0.098
Open in a new tab

Diagnostic Yield With and Without ROSE

A comparison between methods with and without ROSE was also performed. Overall, FNB alone and FNB with ROSE presented higher sensitivity, specificity, and accuracy compared with FNA with and without ROSE (P<0.05). EUS-FNA alone and with ROSE demonstrated similar accuracy (79.78% [95% CI, 69.93 to 87.55] versus 68.00% [95% CI, 46.50 to 85.05]; P=0.217). Accuracy of EUS-FNB alone was superior to EUS-FNA with ROSE (87.25% [95% CI, 79.19 to 93.04] versus 68.00% [95% CI, 46.50 to 85.05]; P=0.021]. Sensitivity, specificity, and accuracy of EUS-FNB alone was not significantly different when compared with EUS-FNB with ROSE (83.33% [95% CI, 35.88 to 99.58] versus 79.03% [95% CI, 66.82 to 88.34]; P=0.701), (100.00% [95% CI, 66.37 to 100.00] versus 100.00% [95% CI, 91.19 to 100.00]; P=NA), and (87.25 [95% CI, 79.19 to 93.04] versus 93.33% [95% CI, 68.05 to 99.83], P =0.500) (Table 5 and Supplementary Table 1).

Table 5.

Comparison Between Methods With and Without ROSE

FNA Alone FNA with ROSE FNB Alone FNB with ROSE
Sensitivity 54.05% (95% CI, 36.92 to 70.51) 46.67% (95% CI, 21.27 to 73.41) 79.03% (95% CI, 66.82 to 88.34) 83.33% (95% CI, 35.88 to 99.58)
Specificity 98.08% (95% CI, 89.74 to 99.95) 100.00% (95% CI, 69.15 to 100.00) 100.00% (95% CI, 91.19 to 100.00) 100.00% (95% CI, 66.37 to 100.00)
Positive Likelihood Ratio 28.11 (95% CI, 3.94 to 200.28) NA NA NA
Negative Likelihood Ratio 0.47 (95% CI, 0.33 to 0.67) 0.53 (95% CI, 0.33 to 0.86) 0.21 (95% CI, 0.13 to 0.34) 0.17 (95% CI, 0.03 to 1.00)
Positive Predictive Value 95.24% (95% CI, 73.73 to 99.30) 100.00% 100.00%% 100.00%%
Negative Predictive Value 75.00% (95% CI, 67.85 to 81.00) 55.56% (95% CI, 43.78 to 66.74) 75.47% (95% CI, 65.49 to 83.30) 90.00% (95% CI, 60.06 to 98.18)
Accuracy 79.78% (95% CI, 69.93 to 87.55) 68.00% (95% CI, 46.50 to 85.05) 87.25% (95% CI, 79.19 to 93.04) 93.33% (95% CI, 68.05 to 99.83)
Open in a new tab

Multivariate Logistic Regression

Multivariate analysis was then performed controlling for age, gender, number of passes, needle type, needle size, application of ROSE, application of cell-block, and application of IHC on accuracy. Based upon the results of this multivariate logistic regression, and controlled for the variables above, there was no significant predictor for better accuracy.

DISCUSSION

Management and diagnosis of SELs remains challenging with histological diagnosis being essential for the differentiation between neoplastic and non-neoplastic lesions. The results of our analysis demonstrate that EUS-FNB had a higher sensitivity, PPV, and accuracy as compared with EUS-FNA in the diagnosis of SELs. Additionally, the diagnostic adequacy on cell-block and IHC staining was higher, requiring fewer number of passes. These findings are similar to previous literature from a large series of patients with sampling from multiple types of solid lesions (17).

To date, this multicenter study remains the largest study of its kind to evaluate the results of EUS-guided tissue sampling in gastrointestinal SELs. Similar to other studies, the most common location of these lesions was the stomach, followed by esophagus, duodenum, and finally the rectum. Additionally, gastric SELs originating from the fourth layer was the most-common lesion layer type represented in the study, likely reflecting real-world clinical practice (5, 18, 19). Among gastric lesions, location within the body of the stomach was the most common with gastric cardia lesions being the least. Although lesions of the cardia were limited in this study, it is crucial to recognize that EUS with pre-operative histologic diagnosis may be helpful in determining the requirement of surgery. In a previous study of gastric cardia lesions, 20% of cases were diagnosed with GIST associated with a high malignant potential (20). Based upon the results of our study, there were no significant difference in baseline characteristics between FNA and FNB cohorts, thus making our comparison more reliable. Among the 229 patients including in this study, a total of 60 patients with EUS-tissue sampling ultimately underwent surgical resection based upon EUS findings. An additional 28 patients with subepithelial lesions >20 mm were found to not have evidence of malignancy and thus did not require surgery post-EUS tissue sampling, underscoring the importance of EUS as useful tool for medical decision-making.

Procurement of histological samples that yield an adequate amount of tissue suitable for IHC staining is pivotal for personalized management of SELs, especially in spindle cell lesions, where IHC staining is imperative. Although EUS-FNA may be accurate to identify spindle cells and mitotic activity, it has been limited in providing tissue for IHC (21). An alternative technique is to collect FNA samples in a formalin solution for preparation of a cell-block with subsequent sectioning and histological interpretation (22–25). However, this technique has demonstrated lower diagnostic rates (81% of patients), independent of the number of FNA passes (26). In our study, cell-block analysis was possible in 80.87% of patients after FNA and 96.49% after FNB (P<0.001). The superiority of EUS-FNB in our study is similar to other studies comparing these 2 techniques (18,19). Additionally, our study showed that the number of passes required to achieve cell-block diagnosis after FNA was higher than FNB (3.55 versus 2.94, P=0.003). Furthermore, similar to other studies, we demonstrated higher rates of IHC staining diagnoses after FNB when compared with FNA (69.30% versus 40.00%, P<0.001) (12,19). A limiting factor in EUS-FNA achieving accurate diagnoses is the pauci-cellular nature of the aspirate with a significant proportion of the collected tissue being distorted or consumed during automated processing and sectioning (17, 21). The low power (4x) images in for FNA and FNB highlight the difference in sample tissue characteristics for both modalities with distorted tissue in Figure 1A and preserved architecture in Figure 2A.

Despite the superiority of FNB regarding IHC staining, our rate of 69.30% was inferior to the 100% rate of IHC staining described by El Chafic, et al (19). This difference may be related to the small number of cases (15 FNBs) included in this previous study and due to the larger mean size of the lesion (27.68 mm ±15.70 versus 21.22 mm ±13.27) of their study. A recent study (26) also evaluated FNB in 20 small (<3 cm) SELs (mean lesion size of 16 mm), and found that despite a diagnostic yield of 75%, core biopsy specimens were obtained in only 25% of cases, again suggesting small lesions are more difficult to sample regardless of technique. Usually, FNA or FNB are not performed in sub-centimeter lesions. In our series, a total of 15 lesions <10 mm were included in our analysis, finding no difference in sampling between 2 strategies. Notably, accuracy of FNA versus FNB was not statistically significant among these sub-centimeter lesions though our study may be underpowered to detect a significant difference.

Unlike several previous studies in the literature, we analyzed the sensitivity, specificity, +LR, −LR, PPV, NPV, and accuracy of EUS-FNA as compared with EUS-FNB (13,18,19,28). In our study, EUS-FNB possessed a significantly higher sensitivity (79.41% versus 51.92%, P=0.001) and accuracy (88.03% versus 77.19%, P=0.030) when compared with EUS-FNA. These results are similar to a study by Hedenstrom et al (1), which showed higher sensitivity and accuracy for FNB when compared with EUS-FNA (90% versus 52% and 83% versus 49%, respectively). The accuracy reported in our large series is also similar to that of multiple smaller studies ranging from 52% to 82% for FNA and 66% to 100% for FNB (9,18,19,29–31).

To determine whether lesion location influenced FNA or FNB sampling, we performed subgroup analyses comparing FNA and FNB for esophageal, gastric, and duodenal SELs. EUS-FNB demonstrated a significantly higher sensitivity than EUS-FNA in the esophageal subgroup; however, no difference in accuracy was found. Among patients with gastric SELs, and contradicting a previous study of only 23 patients demonstrating equivalent results, EUS-FNB demonstrated better sensitivity and higher accuracy as compared with EUS-FNA (13). The superiority of FNB for gastric lesions has been illustrated in prior studies using an old generation biopsy needle (EUS-TCB). However, when EUS-TCB was used, several technical failures were reported, different from our study where no technical failures were reported (8,32). This is likely due to the improved tissue cutting design and flexibility of the newer needles. In the duodenal subgroup, FNB was found to have superior sensitivity (75% versus 33%) and accuracy (90% versus 55.56%) as compared with FNA; however, no statistical difference was noted, likely secondary to the limited number of duodenal SELs included.

Current literature remains mixed as to whether to consider an inconclusive (nondiagnostic) result as benign or to exclude the result entirely from the analysis. This fact is related to the heterogeneity of the results published in the literature (22). When excluding inconclusive results, an increase in sensitivity is observed. Therefore, in this analysis we preemptively made the decision to be more rigorous and considered inconclusive results as non-neoplastic lesions in effort to not overestimate sensitivity and attempt to improve generalizability of these results. Furthermore, this was done to simulate real-world management because it is impossible to exclude a neoplastic disease with an inconclusive result.

As expected from sampling diagnostic modalities, the specificity and PPV were high for both techniques, illustrating that a positive result for a neoplastic lesion is very reliable. Interestingly, FNB showed a statistical superiority when compared with FNA regarding PPV (100% versus 96.43%). However, in both groups the NPV were low, suggesting a negative result cannot exclude a neoplastic diagnosis. As the +LR measures how well a test diagnoses a lesion, the higher the +LR, the better the test performs in identifying the precise diagnosis. In our study, because specificity for EUS-FNB was 100%, the +LR could not be calculated. EUS-FNB had a significantly lower −LR compared with FNA. Therefore, FNB appears to be more reliable than FNA in excluding a suspect neoplastic lesion.

Current literature regarding the advantage of ROSE remains controversial regarding improvement in diagnostic accuracy for solid lesions (33,34). In all institutions included in this multi-center, retrospective study, ROSE was not uniformly available for all cases. For this reason, ROSE was used only in selected cases, typically those cases that were more technically challenging (ie, cases that had failed prior sampling). Given this, ROSE was performed on only 7.47% of total samples with a comparable percent of samples being adequate for evaluation (FNA 80.95% versus FNB 86.67%). Although the number of passes for ROSE adequacy was smaller for FNB as compared with FNA (3.85 ±1.34 versus 4.53 ±1.28), this was not statistically significant. Despite our results, previous studies have demonstrated a significantly lower number of passes required for FNB (6,19). This outcome may be translated into shorter procedure time, less risk of adverse events, and more operational efficiency for both endoscopy and cytopathology units.

In this study, we also compared both techniques with and without ROSE, demonstrating that EUS-FNB alone was superior to EUS-FNA with ROSE. This result is similar to a previous study which showed that EUS-FNB without ROSE provides a similar diagnostic yield than EUS-FNA with ROSE in SELs (34). In our study, FNA with ROSE was typically performed for difficult to diagnosed lesions leading to lower than anticipated results. Additionally, when comparing EUS-FNB alone with EUS-FNB with ROSE no statistical difference was found. Furthermore, in our study, FNB enables a diagnostic yield (cell-block) of 96.49% versus 80.87% for FNA with offsite assessment. Importantly, these results suggest that use of FNB alone may eliminate the need for ROSE entirely.

Unique to our study, a multivariable logistic regression analysis was performed to determine an association between several variables, including age, gender, needle type, needle size, and application of ROSE, cell-block, and IHC on diagnostic accuracy. No significant predictor for better accuracy was found – similar to a prior study evaluating EUS-FNB in patients with gastric SELs which showed no independent predictors for unsuccessful EUS-FNB with non-diagnostic or suggestive results (36). On the other hand, a previous study evaluating EUS-FNA in gastric SELs using different variables (ie, heterogeneous echo pattern, lesion size, long axis location within the stomach, number of needle passes, and layer of origin) showed that only heterogenous echo pattern was an independent predictor for obtaining a sufficient sample (3).

The safety of EUS-tissue sampling is well established with very few adverse events reported in the literature. Serious adverse events are even more rare (18,19,21,37,38). In our study, similar to a study by Kim et al (18) one minor adverse event was reported. Bleeding in this previous study was treated conservatively; however, in our study minor hemorrhage after EUS-FNA was treated with epinephrine injection and hemostatic clip placement. Several studies showed no adverse events related to EUS-tissue sampling in the diagnosis of SELs (13,19,38).

Despite this being the largest study exclusively evaluating SELs, we recognize some important limitations. First, this is a retrospective study with the inherent limitations expected with such a design. These include the potential for selection bias, lack of randomization, and potential for confounders. It would be clinically relevant to prospectively compare a large number of patients with SELs in effort to confirm the superiority of EUS-FNB over EUS-FNA. Additionally, SELs that were diagnosed but not sampled were not included in this analysis. Frequently, SELs <2 cm are not sampled (5,27,39); however, in this study we included smaller lesions, which may have resulted in lower accuracy rates for both sampling modalities. Subgroup analyses including esophageal and duodenal group included were also performed; however, the results are likely underpowered given the small number of patients. Furthermore, in effort to simulate clinical practice, multiple available needles sizes were used. This study included 3 FNB needles types (Acquire, SharkCore, and ProCore). The SharkCore needle was the most commonly used needle (>92%). Given the limited use with other needles, the numbers were too small to perform an appropriate comparative analysis. Although, there was no difference between groups regarding needle size, we cannot discount heterogeneity of our results or fail to acknowledge interoperator variability using these different needle sizes. Reassuringly, a previous meta-analysis including only high-quality randomized controlled trials, did not show significant difference between varied needles sizes (40). It is important to note ROSE was not available in all cases and the determination of adequacy was solely at the discretion of the cytopathologist and their ability to provide a meaningful diagnosis. For cases without ROSE, number of passes were determined by individual endoscopist and provider preference. Finally, we did not perform a cost-analysis. However, a previous randomized study has demonstrated that EUS-FNB was cost saving compared with EUS-FNA over a wide range of cost and outcome probabilities (41).

In summary, EUS-FNB was superior to EUS-FNA in the diagnosis of SELs. Additionally, EUS-FNB required a lower number of passes compared with EUS-FNA and allowed for more consistent cell-block and immunohistochemistry evaluation. EUS-FNB was superior to EUS-FNA with ROSE with no significant difference of FBNB with the addition of ROSE (ie, EUS-FNB with or without ROSE). These results strongly suggest that EUS-FNB should be considered a preferred modality in the diagnosis of SELs and may further imply a reduced role for ROSE in routine clinical practice. However, a large randomized controlled trial is required to confirm our findings. Only with improvement in diagnostic yield and correct identification of SELs, will gastroenterologists and surgeons be able to improve upon the treatment of these complex and difficult to manage lesions.

Supplementary Material

1

Acknowledgments

Potential Conflicts of Interest:

Diogo T. H. de Moura has no conflicts to disclose.

Thomas R. McCarty has no conflicts to disclose.

Pichamol Jirapinyo has no conflicts to disclose.

Igor Braga Ribeiro has no conflicts to disclose.

Victor K. Flumignan has no conflicts to disclose

Marvin Ryou is a consultant for Medtronic, GI Windows, EnteraSense, FujiFilm, Boston Scientific, Olympus (research grant), and Pentax.

Linda S. Lee has no conflicts to disclose

Christopher C. Thompson is a consultant for Boston Scientific – Consultant (Consulting fees), Medtronic – Consultant (Consulting Fees), USGI Medical – Consultant (Consulting Fees)/Advisory Board Member (Consulting fees)/Research Support (Research Grant), Olympus – Consultant (Consulting Fees)/Research Support (Equipment Loans); Apollo Endosurgery – Consultant/Research Support (Consulting fees/Institutional Research Grants) GI Windows – Ownership interest, Aspire Bariatrics – Research Support (Institutional Research Grant), Fractyl – Consultant/Advisory Board Member (Consulting Fees), Spatz – Research Support (Institutional Research Grant) EndoTAGSS – Ownership Interest, GI Dynamics – Consultant (Consulting Fees)/ Research Support (Institutional Research Grant)

ACRONYMS

EUS-FNA

endoscopic ultrasound fine-guided fine needle aspiration

EUS-FNB

endoscopic ultrasound fine-guided fine needle biopsy

EUS

endoscopic ultrasound

SELs

subepithelial lesions

GIST

gastrointestinal stromal tumors

EGD

esophagogastroduodenoscopy

ROSE

rapid on-site evaluation

FNA

fine needle aspiration

FNB

fine needle biopsy

IHC

immunohistochemistry

TCB

trucut biopsy

ESGE

European Society of Gastrointestinal Endoscopy

STARD

STAndards for the Reporting of Diagnostic accuracy studies

GI

gastrointestinal

G

gauge

PPV

positive predictive value

NPV

negative predictive value

LR−

negative likelihood ratio

LR+

positive likelihood ratio

Appendix 1.

STAndards for the Reporting of Diagnostic accuracy studies (STARD) checklist

Section & Topic No Item Reported on page #
TITLE OR ABSTRACT
1 Identification as a study of diagnostic accuracy using at least one measure of accuracy (such as sensitivity, specificity, predictive values, or AUC) 1,4
ABSTRACT
2 Structured summary of study design, methods, results, and conclusions (for specific guidance, see STARD for Abstracts) 4
INTRODUCTION
3 Scientific and clinical background, including the intended use and clinical role of the index test 5
4 Study objectives and hypotheses 5
METHODS
Study design 5 Whether data collection was planned before the index test and reference standard were performed (prospective study) or after (retrospective study) 5
Participants 6 Eligibility criteria 5
7 On what basis potentially eligible participants were identified (such as symptoms, results from previous tests, inclusion in registry) 5,6
8 Where and when potentially eligible participants were identified (setting, location and dates) 5,6
9 Whether participants formed a consecutive, random or convenience series 5,6
Test methods 10a Index test, in sufficient detail to allow replication 6,7
10b Reference standard, in sufficient detail to allow replication 6,7
11 Rationale for choosing the reference standard (if alternatives exist) 6,7
12a Definition of and rationale for test positivity cut-offs or result categories of the index test, distinguishing pre-specified from exploratory 6,7
12b Definition of and rationale for test positivity cut-offs or result categories of the reference standard, distinguishing pre-specified from exploratory 6,7
13a Whether clinical information and reference standard results were available to the performers/readers of the index test 6,7
13b Whether clinical information and index test results were available to the assessors of the reference standard 6,7
Analysis 14 Methods for estimating or comparing measures of diagnostic accuracy 7
15 How indeterminate index test or reference standard results were handled 7
16 How missing data on the index test and reference standard were handled 7
17 Any analyses of variability in diagnostic accuracy, distinguishing pre-specified from exploratory 7
18 Intended sample size and how it was determined -
RESULTS
Participants 19 Flow of participants, using a diagram 8
20 Baseline demographic and clinical characteristics of participants 8
21a Distribution of severity of disease in those with the target condition 8,9
21b Distribution of alternative diagnoses in those without the target condition 8,9
22 Time interval and any clinical interventions between index test and reference standard 8,9
Test results 23 Cross tabulation of the index test results (or their distribution) by the results of the reference standard 8–12
24 Estimates of diagnostic accuracy and their precision (such as 95% confidence intervals) 8–12
25 Any adverse events from performing the index test or the reference standard 9,10
DISCUSSION
26 Study limitations, including sources of potential bias, statistical uncertainty, and generalisability 15,16
27 Implications for practice, including the intended use and clinical role of the index test 13–16
OTHER INFORMATION
28 Registration number and name of registry -
29 Where the full study protocol can be accessed -
30 Sources of funding and other support; role of funders 2
Open in a new tab

Footnotes

Article Statement: The manuscript has been read and approved by all the authors. The requirements for authorship have been met. Each author believes that the manuscript represents honest work.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

REFERENCES

  • 1.Hedenström P, Marschall HU, Nilsson B, et al. High clinical impact and diagnostic accuracy of EUS-guided biopsy sampling of subepithelial lesions: a prospective, comparative study. Surg Endosc. 2018;32:1304–1313. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Brunaldi VO, Coronel M, Chacon DA, et al. Subepithelial rectal gastrointestinal stromal tumor - the use of endoscopic ultrasound-guided fine needle aspiration to establish a definitive cytological diagnosis: a case report. J Med Case Rep. 2017;11:59. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Mekky MA, Yamao K, Sawaki A, et al. Diagnostic utility of EUS-guided FNA in patients with gastric submucosal tumors. Gastrointest Endosc. 2010;71:913–9. [DOI] [PubMed] [Google Scholar]
  • 4.Ji JS, Lee BI, Choi KY, et al. Diagnostic yield of tissue sampling using a bite-on-bite technique for incidental subepithelial lesions. Korean J Intern Med. 2009;24:101–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Chaves DM, Meine GC, Moura DT, et al. Endoscopic ultrasound in the evaluation of upper subepithelial lesions. Arq Gastroenterol. 2015;52:186–9. [DOI] [PubMed] [Google Scholar]
  • 6.Han JP, Lee TH, Hong SJ, et al. EUS-guided FNA and FNB after on-site cytological evaluation in gastric subepithelial tumors. J Dig Dis. 2016;17:582–587. [DOI] [PubMed] [Google Scholar]
  • 7.Khoury T, Sbeit W, Ludvik N, et al. Concise review on the comparative efficacy of endoscopic ultrasound-guided fine-needle aspiration vs core biopsy in pancreatic masses, upper and lower gastrointestinal submucosal tumors. World J Gastrointest Endosc. 2018;10:267–273. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Fernández-Esparrach G, Sendino O, Solé M, et al. Endoscopic ultrasound-guided fine-needle aspiration and trucut biopsy in the diagnosis of gastric stromal tumors: a randomized crossover study. Endoscopy. 2010;42:292–9. [DOI] [PubMed] [Google Scholar]
  • 9.Philipper M, Hollerbach S, Gabbert HE, et al. Prospective comparison of endoscopic ultrasound-guided fine-needle aspiration and surgical histology in upper gastrointestinal submucosal tumors. Endoscopy. 2010;42:300–5. [DOI] [PubMed] [Google Scholar]
  • 10.Sepe PS, Moparty B, Pitman MB, et al. EUS-guided FNA for the diagnosis of GI stromal cell tumors: sensitivity and cytologic yield. Gastrointest Endosc. 2009;70:254–61. [DOI] [PubMed] [Google Scholar]
  • 11.Akahoshi K, Sumida Y, Matsui N, et al. Preoperative diagnosis of gastrointestinal stromal tumor by endoscopic ultrasound-guided fine needle aspiration. World J Gastroenterol. 2007;13:2077–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Inoue T, Okumura F, Sano H, et al. Impact of endoscopic ultrasound-guided fine-needle biopsy on the diagnosis of subepithelial tumors: A propensity score-matching analysis. Dig Endosc. 2019;31:156–163. [DOI] [PubMed] [Google Scholar]
  • 13.Iwai T, Kida M, Imaizumi H, et al. Randomized crossover trial comparing EUS-guided fine-needle aspiration with EUS-guided fine-needle biopsy for gastric subepithelial tumors. Diagn Cytopathol. 2018;46:228–233. [DOI] [PubMed] [Google Scholar]
  • 14.Polkowski M, Jenssen C, Kaye P, et al. Technical aspects of endoscopic ultrasound (EUS)-guided sampling in gastroenterology: European Society of Gastrointestinal Endoscopy (ESGE) Technical Guideline - March 2017. Endoscopy. 2017;49:989–1006. [DOI] [PubMed] [Google Scholar]
  • 15.Pepe MS. The Statistical Evaluation of Medical Tests for Classification and Prediction. Section 9.3. Oxford University Press, 2003. [Google Scholar]
  • 16.Campbell I Chi-squared and Fisher-Irwin tests of two-by-two tables with small sample recommendations. Stat Med. 2007;26:3661–75. [DOI] [PubMed] [Google Scholar]
  • 17.Bang JY, Kirtane S, Krall K, et al. In memoriam: Fine-needle aspiration, birth: Fine-needle biopsy: The changing trend in endoscopic ultrasound-guided tissue acquisition. Dig Endosc. 2019;31:197–202. [DOI] [PubMed] [Google Scholar]
  • 18.Kim GH, Cho YK, Kim EY, et al. ; Korean EUS Study Group. Comparison of 22-gauge aspiration needle with 22-gauge biopsy needle in endoscopic ultrasonography-guided subepithelial tumor sampling. Scand J Gastroenterol. 2014;49:347–54. [DOI] [PubMed] [Google Scholar]
  • 19.El Chafic AH, Loren D, Siddiqui A, et al. Comparison of FNA and fine-needle biopsy for EUS-guided sampling of suspected GI stromal tumors. Gastrointest Endosc. 2017;86:510–515. [DOI] [PubMed] [Google Scholar]
  • 20.Kim GH, Ahn JY, Gong CS, et al. Efficacy of Endoscopic Ultrasound-Guided Fine-Needle Biopsy in Gastric Subepithelial Tumors Located in the Cardia. Dig Dis Sci 2019. Ahead of print. [DOI] [PubMed] [Google Scholar]
  • 21.Obuch J, Wani S. EUS-guided tissue acquisition in GI stromal tumors. Gastrointest Endosc 2017;86:516–518. [DOI] [PubMed] [Google Scholar]
  • 22.Moura DTH, de Moura EGH, Matuguma SE, et al. EUS-FNA versus ERCP for tissue diagnosis of suspect malignant biliary strictures: a prospective comparative study. Endosc Int Open. 2018;6:E769–E777. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.De Moura DTH, Coronel M, Chacon DA, et al. Primary adenosquamous cell carcinoma of the pancreas: the use of endoscopic ultrasound guided - fine needle aspiration to establish a definitive cytologic diagnosis. Rev Gastroenterol Peru. 2017;37:370–373. [PubMed] [Google Scholar]
  • 24.De Moura DT, Chacon DA, Tanigawa R, et al. Pancreatic metastases from ocular malignant melanoma: the use of endoscopic ultrasound-guided fine-needle aspiration to establish a definitive cytologic diagnosis: a case report. J Med Case Rep. 2016;10:332. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.De Moura DTH, Coronel M, Ribeiro IB, et al. The importance of endoscopic ultrasound fine-needle aspiration in the diagnosis of solid pseudopapillary tumor of the pancreas: two case reports. J Med Case Rep. 2018;12:107. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Varadarajulu S, Bang JY, Holt BA et al. The 25-gauge EUS-FNA needle: Good for on-site but poor for off-site evaluation? Results of a randomized trial. Gastrointest Endosc 2014; 80:1056–63. [DOI] [PubMed] [Google Scholar]
  • 27.Schlag C, Menzel C, Götzberger M, et al. Endoscopic ultrasound-guided tissue sampling of small subepithelial tumors of the upper gastrointestinal tract with a 22-gauge core biopsy needle. Endosc Int Open. 2017;5:E165–E171. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Lee JH, Choi KD, Kim MY, et al. Clinical impact of EUS-guided Trucut biopsy results on decision making for patients with gastric subepithelial tumors ≥ 2 cm in diameter. Gastrointest Endosc. 2011;74:1010–8. [DOI] [PubMed] [Google Scholar]
  • 29.Lee JH, Cho CJ, Park YS, et al. EUS-guided 22-gauge fine needle biopsy for the diagnosis of gastric subepithelial tumors larger than 2 cm. Scand J Gastroenterol 2016;51:486–93. [DOI] [PubMed] [Google Scholar]
  • 30.Hoda KM, Rodriguez SA, Faigel DO. EUS-guided sampling of suspected GI stromal tumors. Gastrointest Endosc 2009;69:1218–23. [DOI] [PubMed] [Google Scholar]
  • 31.Watson RR, Binmoeller KF, Hamerski CM, et al. Yield and performance characteristics of endoscopic ultrasound-guided fine needle aspiration for diagnosing upper GI tract stromal tumors. Dig Dis Sci 2011;56:1757–62. [DOI] [PubMed] [Google Scholar]
  • 32.Na HK, Lee JH, Park YS, et al. Yields and Utility of Endoscopic Ultrasonography-Guided 19-Gauge Trucut Biopsy versus 22-Gauge Fine Needle Aspiration for Diagnosing Gastric Subepithelial Tumors. Clin Endosc. 2015;48:152–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Kong F, Zhu J, Kong X, et al. Rapid On-Site Evaluation Does Not Improve Endoscopic Ultrasound-Guided Fine Needle Aspiration Adequacy in Pancreatic Masses: A Meta-Analysis and Systematic Review. PLoS One. 2016;11:e0163056. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Matynia AP, Schmidt RL, Barraza G, et al. Impact of rapid on-site evaluation on the adequacy of endoscopic-ultrasound guided fine-needle aspiration of solid pancreatic lesions: a systematic review and meta-analysis. J Gastroenterol Hepatol. 2014;29:697–705. [DOI] [PubMed] [Google Scholar]
  • 35.Rodrigues-Pinto E, Jalaj S, Grimm IS, Baron TH. Impact of EUS-guided fine-needle biopsy sampling with a new core needle on the need for onsite cytopathologic assessment: a preliminary study. Gastrointest Endosc. 2016;84:1040–1046. [DOI] [PubMed] [Google Scholar]
  • 36.Lee M, Min BH, Lee H, et al. Feasibility and Diagnostic Yield of Endoscopic Ultrasonography-Guided Fine Needle Biopsy With a New Core Biopsy Needle Device in Patients With Gastric Subepithelial Tumors. Medicine (Baltimore). 2015;94:e1622. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Mohan BP, Shakhatreh M, Garg R, et al. Efficacy and safety of EUS-guided liver biopsy: a systematic review and meta-analysis. Gastrointest Endosc. 2019;89:238–246.e3. [DOI] [PubMed] [Google Scholar]
  • 38.Iglesias-Garcia J, Poley JW, Larghi A, et al. Feasibility and yield of a new EUS histology needle: results from a multicenter, pooled, cohort study. Gastrointest Endosc 2011;73: 1189–96. [DOI] [PubMed] [Google Scholar]
  • 39.Demetri GD, von Mehren M, Antonescu CR, et al. NCCN Task Force report: update on the management of patients with gastrointestinal stromal tumors. J Natl Compr Cancer Netw 2010;8:S1–41; quiz S2–4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Guedes HG, Moura DTH, Duarte RB, et al. A comparison of the efficiency of 22G versus 25G needles in EUS-FNA for solid pancreatic mass assessment: A systematic review and meta-analysis. Clinics (Sao Paulo). 2018;73:e261. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Aadam AA, Wani S, Amick A, et al. A randomized controlled cross-over trial and cost analysis comparing endoscopic ultrasound fine needle aspiration and fine needle biopsy. Endosc Int Open. 2016;4:E497–505. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS1593895-supplement-1.pdf (24.2KB, pdf)

RESOURCES