- Email: [email protected]
Utility of EUS elastography in the diagnosis of gastric subepithelial tumors: a pilot study (with video)
Journal Pre-proof Utility of EUS elastography in the diagnosis of gastric subepithelial tumors: a pilot study (with video) Sun Hwa Kim, MD, PhD, In Kyung Yoo, MD, PhD, Chang-Il Kwon, MD, PhD, Sung Pyo Hong, MD, PhD, Joo Young Cho, MD, PhD PII:
S0016-5107(19)32245-X
DOI:
https://doi.org/10.1016/j.gie.2019.09.012
Reference:
YMGE 11753
To appear in:
Gastrointestinal Endoscopy
Received Date: 2 April 2019 Accepted Date: 4 September 2019
Please cite this article as: Kim SH, Yoo IK, Kwon C-I, Hong SP, Cho JY, Utility of EUS elastography in the diagnosis of gastric subepithelial tumors: a pilot study (with video), Gastrointestinal Endoscopy (2019), doi: https://doi.org/10.1016/j.gie.2019.09.012. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. 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. Copyright © 2019 by the American Society for Gastrointestinal Endoscopy
Utility of EUS elastography in the diagnosis of gastric subepithelial tumors: a pilot study (with video) Sun Hwa Kim MD, PhD, In Kyung Yoo MD, PhD, Chang-Il Kwon MD, PhD, Sung Pyo Hong MD, PhD, Joo Young Cho MD, PhD Department of Gastroenterology, Cha Bundang Medical Center, Cha University College of Medicine, Seongnam-si, Korea
Address Correspondence and Reprint requests to: In Kyung Yoo, M.D.PhD Professor of Medicine Director of Department of Gastroenterology Cha Bundang Medical Center 59 Yatapro, Bundang-gu, Seongnam-si, Korea Tel: (+82) 31-780-5641 Fax: (+82) 32-780-5005 [email protected]
Abstract Background and study aims: EUS elastography is a real-time imaging technique that analyzes the tissue elasticity. The aim of this study was to investigate the applicability of the quantitative EUS elastography in the differential diagnosis of the gastric subepithelial tumors (SETs). Methods: We prospectively registered 31 consecutive patients with gastric SETs and measured their strain ratios with EUS elastography. The strain ratios of gastric SETs were compared with the histopathological diagnosis. Results: Thirty patients (mean age 51.4 ± 12.6 years) were included in the analysis. The mean size of SETs was 2.3 ± 0.7 cm. Lipoma was measured at the lowest strain ratio of 1.6 (1.1-2.0), followed by leiomyoma 6.0 (2.0-29.0), ectopic pancreas 11.8 (1.7-29.3), gastrointestinal stromal tumors (GIST) 51.1 (29.0-67.0), and schwannoma 62.0. With the cutoff value of 22.7, EUS elastography could differentiate GIST from leiomyoma with the sensitivity and specificity of 100 % and 94.1 %, respectively (p=0.001; 95 % CI, 0.9791.000). Conclusions: EUS elastography could be a promising diagnostic adjunct for the assessment of gastric SETs, especially in differentiating GISTs from leiomyoma. Key words; Elastography, Endoscopic ultrasound, Gastric subepithelial tumor
Introduction EUS has been recommended for the differential diagnosis of gastric subepithelial tumors
(SETs) because it can provide information on the characteristics, including their size, margin, internal echogenicity, and the layer of origin. 1, 2 However, despite being considered as the basic diagnostic tool for gastric SETs, it is frequently inconclusive when differentiating the subtypes of gastric SET. Moreover, it is challenging to decide whether or not to perform invasive cytopathological confirmation, and unfortunately, obtaining sufficient tissue is technically difficult especially for those, which are located in the deep layer.3 EUS elastography is a real-time imaging technique that displays the tissue elasticity differences between diseased and normal tissues on the conventional B-mode ultrasound images.1 It analyzes the degree of tissue deformation under compression and reflects this information as a hue color spectrum. The current applications of elastography were extended in diverse diseases including those related to the pancreas,4-6 breast,7 thyroid gland,8 liver,9 prostate,10 and lymph node 5, 6. Several studies have demonstrated that it improved the diagnostic accuracy of conventional ultrasonography. Giovannini et al5, 6 reported the potential of EUS elastography in the differentiation of pancreatic masses and malignant lymph nodes with good sensitivity. In regard to gastric SETs, a previous study11 reported that EUS elastography could be helpful in differentiating gastrointestinal stromal tumors (GISTs) from other gastric SETs by color-based qualitative grading because GISTs have relatively firm mechanical properties than others. However, as this is a visual color-based qualitative method, interobserver bias is possible. In contrast, elastographic method with calculated strain ratio would be reproducible and yield more objective results.1 Most studies on quantitative EUS elastography investigated it in the context of pancreatic diseases and proposed a cut-off strain value for malignant lesions.12 Instead, to date, few studies have evaluated gastric SMTs according to their tissue elasticities using EUS elastography. Therefore, we aimed to investigate the feasibility of EUS
elastography in the differential diagnosis of the gastric SETs based on strain ratio. Methods Patients From April 2018 to December 2018, we prospectively recruited 41 consecutive patients with gastric SET diagnosed by esophagogastroduodenoscopy at Bundang CHA Medical Center. The strain ratios of gastric SETs were measured using EUS elastography and this was followed by histopathological confirmation after lumpectomy (n = 31). The exclusion criteria were as follows: gastric SET <1 cm, located in esophagus or duodenum, severe coagulopathy, cardiopulmonary disease, chronic kidney disease, and unwillingness to undergo lumpectomy. The study was approved by the ethics review board of Bundang CHA Medical Center (CHAMC 2019-01-015-003), and written informed consent was obtained from all patients. EUS guided elastography and measurement of the strain The procedures were carried out by an experienced endoscopist (JY Cho) using an EUS endoscope (EG-3670URK; Pentax Lifecare Division, Hoya Co, Ltd, Tokyo, Japan and GF TYPE UCT 260; Olympus Corporation, Tokyo, Japan) combined with diagnostic ultrasound imaging systems (HI VISION Ascendus; Hitachi Aloka Medical Ltd, Tokyo, Japan and EUME2; Olympus Corporation). First, conventional B-mode endoscopic ultrasound images were assessed to characterize the size of gastric SET, the relation of gastric SET to the layers of the stomach wall, and the internal echogenicity. Second, the elasticity of gastric SET was quantitatively measured using strain ratios (Fig. 1). They were measured when the compression wave reached the negative peak indicating that the probe has just released from the gastric SET. The average strain value within region of interest (ROI) of gastric SET was measured as A, and the strain value of
water filled in the EUS scope balloon was measured as B for the reference. The ratio of A to B (B/A quotient), strain ratio, was calculated by Aloka diagnostic ultrasound imaging systems. The strain ratio was measured higher, in case the tumor was hard. We checked the strain values of the lesion (A) and the reference (B) 3 times repeatedly and considered the average elasticity for analysis to ensure precise measurement. Endoscopic resections and pathologic diagnosis All endoscopic resections were performed by a single experienced endoscopist (J.Y.C.). Capassisted ESD technique using APC was performed for endoscopic resection. First, snare mucosectomy was performed and revealed SET. Careful dissection was performed after submucosal injection. For submucosal dissection, we used a dual knife (KD-650L, Olympus Medical Systems, Tokyo, Japan) or APC and then used the cap to peel the muscularis propria layer along the capsule of the lesion. We termed this process as “cap-assisted ESD technique using APC," because spray-coagulation is currently used for dual knife or APC to dissect the lesion, and the cap is used to remove the lesion using mechanical force. The transparent cap, which is attached onto the tip of the endoscope, applies tension to the connective tissues during dissection. This technique helps to peel the capsule of the diseased site more easily and safely from the muscularis propria layer. The gastric wall defects were not routinely closed after endoscopic resection; however, two cases of small perforations were successfully managed by clips with a detachable snare. Histopathologic diagnosis was made and subsequently compared to the strain ratio. Statistical analysis Data are presented as means, standard deviations, or ranges, as appropriate. Intergroup differences were compared using the Kruskal–Wallis test for continuous data and chi-squared
test or linear by linear association for categorical data. A P value of < 0.05 was considered to be statistically significant. The optimal cut-off values were selected as the points of the receiver operating characteristic (ROC) curves where the sum of the sensitivity and the specificity become maximum. SPSS software (version 25.0; SPSS Inc, Chicago, Ill, USA) was used to analyze the data.
Results Patients characteristics A total of 41 patients who were diagnosed with gastric SET on esophagogastroduodenoscopy were enrolled. Six patients were excluded as the size of the gastric SMT was less than 1 cm and 4 patients did not want to undergo lumpectomy. Finally, 31 patients were included in the analysis. The mean age of the patients was 51.4 ± 12.6 (range 31-76), and 20 (64.5%) of them were male. The mean size of gastric SETs was 2.3 ± 0.7 cm (1.0-3.5). The most common final pathological diagnosis was leiomyoma in 17 patients, followed by GISTs in 7, ectopic pancreas in 4, lipoma in 2, and schwannoma in 1. The gastric SETs were located in the body (n = 11), cardia (n = 9), fundus (n = 4), antrum (n = 4), and angle (n = 3) of the stomach (Table 1). Elasticity analysis for the subtypes of the gastric SETs We compared the results of the final histopathological diagnosis of 31 patients with their respective strain ratios (Table 2). Lipoma was measured at the lowest strain ratio of 1.6 ± 0.7 (range, 1.1 2.0), presenting with hyperechogenic mass arising from submucosal layer in EUS. The most common gastric SET was leiomyoma with a mean strain ratio of 6.0 ± 6.9 (2.0 -
29.0). In EUS examinations, leiomyoma appeared as hypoechoic lesions arising from muscularis mucosa in 10 cases, and from muscularis propria in 8. Patients with ectopic pancreas showed a mean strain ratio of 11.8 ± 12.1 (1.7 - 29.3). There was considerable overlap in the strain ratio of leiomyoma and that of lipoma (p = 0.38) and between the strain ratio of leiomyoma and that of ectopic pancreas (p = 0.21). Therefore, the data showed no statistical difference in strain ratio between the groups. On the other hand, in those with GIST, it presented as hypoechoic mass in muscularis propria in EUS and showed higher mean strain ratio of 51.1 ± 11.8 (29.0 – 67.0). Six of them were classified as very low risk GISTs, and only one case was confirmed as intermediate risk with the highest strain value of 67.0. The mean strain value of GIST was significantly higher (p < 0.001) than that of other gastric subepithelial tumors except schwannoma (Fig. 2). Schwannoma, an irregular hypoechoic mass arising from muscularis propria in EUS, showed a strain ratio of 62.0. ROC analysis of the strain ratio yielded an area under the curve of 0.961 (p=0.001; 95 % CI, 0.896-1.000) for GIST. With the cut-off value of 22.7, the sensitivity and specificity of strain ratio measured by EUS elastography for distinguishing GIST from other gastric SETs were 100.0% and 87.5%, respectively (Fig. 3A). In addition, ROC analysis on subgroups of leiomyoma versus GIST was assessed. The data yielded an area under the curve of 0.996 (p=0.001; 95 % CI, 0.979-1.000) and the cut-off value of 22.7 demonstrated the sensitivity and specificity, 100 % and 94.1 %, respectively (Fig. 3B). Discussion In this study, we investigated the benefit of quantitative EUS elastography in the diagnosis of gastric SETs using strain ratio. Elastography, first described in the 1990s, is an attractive
imaging technology that is based on the difference of tissue mechanical properties.13 The differential diagnosis among gastric SETs during EUS examination is sometimes challenging, and thereby requiring EUS-guided fine-needle aspiration (EUS-FNA)14,
15
or EUS-guided
trucut biopsy (EUS-TCB).16 In addition, these invasive procedures may result in insufficient or negative cytopathologic diagnosis depending on the tissue acquisition technique with a limited diagnostic adequacy of 38% to 89%.15-17 Therefore, the ability to differentiate the etiology of gastric SETs with less-invasive diagnostic technique is clinically desirable. Several studies have been conducted to distinguish GISTs from other SETs to increase the preoperative diagnostic rate and to avoid unnecessary surgery.18-20 An analysis of 62 patients revealed that GISTs present with hyperenhancement on contrast-enhanced EUS (CEEUS) with a sensitivity and specificity of 98% and 100%, respectively, and that 88% of the GISTs contain avascular areas.19 Another study that used digital image analysis software demonstrated that GISTs have a higher echogenicity and heterogeneity within a tumor, as compared with other benign SETs.20 To improve the diagnostic rate of gastric SETs, we analyzed whether noninvasive elastography during EUS examination would be useful in the clinical diagnosis of gastric SETs. Our result demonstrated that lipoma and leiomyoma had lower strain ratios, whereas GIST and schwannoma showed relatively higher strain ratios, indicating that the latter have harder tissue properties. Among gastric SETs, lipoma presents as hypodense mass in abdominal CT, and in endoscopy, it is easily compressed and lifted up by biopsy forceps with yellow colored appearance. Ectopic pancreas is strongly suggested by its specific feature of umbilicated surface corresponding to a draining duct. In contrast, it is occasionally difficult to differentiate between leiomyomas and GISTs in clinical practice due to identical appearance on EUS; therefore, deciding whether to follow up conservatively or to undergo subsequent tumor resection is challenging.21 A previous study22 reported that gastric GISTs tend to
present with heterogenicity, hyperechoic spots, marginal halo and higher echogenicity on EUS but it is not always possible to predict GISTs according to these features. A preliminary study by Antonini et al23 demonstrated that EUS elastography with the strain ratio improved the differential diagnosis of SETs, in that strain ratios of GISTs were higher than those of other benign GI lesions. However, this preliminary study permitted results from EUS-fine needle biopsy (EUS-FNB) or clinical follow-up, whereas we made the final diagnosis only by histological confirmation obtained from endoscopic resection. Also, although they regarded a healthy gastrointestinal wall as the reference, we introduced a more reliable measurement using the water level in the EUS endoscope balloon. We decided to use water as our reference value because measurements of multiple layers of the thin gastric wall may vary each time. Based on our results, if the lesion is suspicious for either leiomyoma or GIST on EUS interpretation, the strain ratio more than 22.7 favors GIST rather than leiomyoma with the sensitivity of 100% and specificity of 94.1%. This suggests that additional elastography might enhance the diagnostic accuracy of EUS on the differential diagnosis of gastric SETs. The strain ratios were found to be significantly higher in GIST than in leiomyoma. In our data, GISTs and leiomyomas were of similar size; therefore, the strain ratios were not affected by lesion size. To maximize the quality of elastography, we measured the strain ratios of the lesion 3 times repeatedly. Before measuring the strain ratios, we checked the elastic color distribution on EUS images and tried to include most of the elastic heterogeneity in ROI. In addition, to acquire reproducible and objective data, we checked the compression wave displayed at ultrasound monitor to measure the strain value when the probe was just released against gastric SET.1 The elastography procedure time took about 5 to 10 minutes with no procedure-related adverse events, highlighting that it is a relatively convenient and safe method.
There are some limitations in our study. First, the patients recruited in the study were small in number. It might be necessary to recruit more patients to conclude more reliable results. Second, strain ratios were measured 3 times during each procedure. Although there was some variability between the readings, differences were minimal (Appendices). The mean value was calculated and used as the final result for each patient. Additionally, the strain value was measured by one experienced endoscopist (J.Y.C.) while the patient was sedated, and the strain ratio at the most appropriate video sequences was chosen, supporting the adequacy of elastographic measurements. Third, patients diagnosed with GIST were mostly in the verylow-risk group, and only 1 patient was in the intermediate risk group; therefore, we could not accurately evaluate the elastic difference in the risk groups. Although there was a trend of a higher strain ratio in intermediate risk GISTs than in very-low-risk cases, even in the small number of cases, the strain ratio alone would not substitute cytopathology in assessing the mitotic activity or immunohistochemistry features. Further studies with larger sample size and inclusion of other subtypes of SETs are necessary to enhance the accuracy of the cut-off value of strain ratio in the differential diagnosis of gastric SETs. In conclusion, this study demonstrated that EUS elastography could be a promising diagnostic technique for the assessment of gastric SETs. Disclosure The authors have no conflicts of interest to disclose.
Table 1. Patient demographics and clinicopathological characteristics of gastric subepithelial tumors. Characteristic
Total (n=31)
Sex, male/female, n (%)
20/11 (64.5/35.5)
Age, mean (SD), years
51.4 (12.6)
Tumor location, n (%) Cardia
9 (29.0)
Fundus
4 (12.9)
Body
11 (35.5)
Antrum
4 (12.9)
Angle
3 (9.7)
Tumor size on EUS, mean (SD), cm
2.3 (0.7)
1 ≤size <2 cm, n (%)
14 (45.2)
2 ≤size <3 cm, n (%)
12 (38.7)
≥3 cm, n (%)
5 (16.1)
Pathological diagnosis, n (%) Leiomyoma
17 (54.8)
GIST
7 (22.6)
Ectopic pancreas
4 (12.9)
Lipoma
2 (6.5)
Schwannoma
1 (3.2)
Table 2. The strain ratios of gastric subepithelial tumors according to the final histopathologic diagnosis.
Pathological diagnosis
Strain ratio mean (SD)
range
Lipoma (n=2)
1.6 (0.7)
1.1-2.0
Leiomyoma (n=17)
6.0 (6.9)
2.0-29.0
Ectopic pancreas (n=4)
11.8 (12.1)
1.7-29.3
GIST (n=7)
51.1 (11.8)
29.0-67.0
Schwannoma (n=1)
62.0
62.0
Figure legends Figure 1. EUS elastography images. The strain ratio was measured when the compression wave (green) reached the negative peak indicating the probe has just been released from gastric subepithelial tumor. A, The hypoechoic tumor with the strain ratio of 3.7 was finally diagnosed with leiomyoma. B, Whereas, the hypoechoic lesion with the strain ratio of 52.3 was diagnosed with GIST. Figure 2. The differences between mean strain ratios of GIST were statistically significant compared to those of lipoma (P < .002), leiomyoma (P < .001), and ectopic pancreas (p = 0.001). P < .05 was considered statistically significant. (Schwannoma was excluded in the analysis because of the small number.) Figure 3. Receiver operating characteristic curve analysis of the strain ratio for the differentiation of gastrointestinal stromal tumor (GIST) from other gastric subepithelial tumors (A) and from leiomyoma (B). References 1. 2. 3.
4.
Sigrist RMS, Liau J, Kaffas AE, et al. Ultrasound Elastography: Review of Techniques and Clinical Applications. Theranostics 2017;7:1303-1329. Cui XW, Chang JM, Kan QC, et al. Endoscopic ultrasound elastography: Current status and future perspectives. World J Gastroenterol 2015;21:13212-24. Hwang JH, Saunders MD, Rulyak SJ, et al. A prospective study comparing endoscopy and EUS in the evaluation of GI subepithelial masses. Gastrointest Endosc 2005;62:202-8. Kawada N, Tanaka S. Elastography for the pancreas: Current status and future perspective. World J Gastroenterol 2016;22:3712-24.
5.
6.
7. 8.
9. 10. 11.
12.
13. 14. 15.
16.
17.
18. 19.
20.
21.
22. 23.
Giovannini M, Hookey LC, Bories E, et al. Endoscopic ultrasound elastography: the first step towards virtual biopsy? Preliminary results in 49 patients. Endoscopy 2006;38:344-8. Giovannini M, Thomas B, Erwan B, et al. Endoscopic ultrasound elastography for evaluation of lymph nodes and pancreatic masses: a multicenter study. World J Gastroenterol 2009;15:1587-93. Bayat M, Denis M, Gregory A, et al. Diagnostic features of quantitative comb-push shear elastography for breast lesion differentiation. PLoS One 2017;12:e0172801. Cosgrove D, Barr R, Bojunga J, et al. WFUMB Guidelines and Recommendations on the Clinical Use of Ultrasound Elastography: Part 4. Thyroid. Ultrasound Med Biol 2017;43:4-26. Friedrich-Rust M, Poynard T, Castera L. Critical comparison of elastography methods to assess chronic liver disease. Nat Rev Gastroenterol Hepatol 2016;13:402-11. Clinical practice guidelines for ultrasound elastography: prostate. J Med Ultrason (2001) 2016;43:449-55. Tsuji Y, Kusano C, Gotoda T, et al. Diagnostic potential of endoscopic ultrasonography-elastography for gastric submucosal tumors: A pilot study. Dig Endosc 2016;28:173-8. Carrara S, Di Leo M, Grizzi F, et al. EUS elastography (strain ratio) and fractal-based quantitative analysis for the diagnosis of solid pancreatic lesions. Gastrointest Endosc 2018;87:1464-1473. Gennisson JL, Deffieux T, Fink M, et al. Ultrasound elastography: principles and techniques. Diagn Interv Imaging 2013;94:487-95. 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. 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. 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. Fernandez-Esparrach G, Sendino O, Sole 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. Chu CS. Gastrointestinal Stromal Tumors: How to Increase the Preoperative Endoscopic Ultrasonography Diagnostic Rate. J Med Ultrasound 2018;26:177-180. Ignee A, Jenssen C, Hocke M, et al. Contrast-enhanced (endoscopic) ultrasound and endoscopic ultrasound elastography in gastrointestinal stromal tumors. Endosc Ultrasound 2017;6:55-60. Kim GH, Kim KB, Lee SH, et al. Digital image analysis of endoscopic ultrasonography is helpful in diagnosing gastric mesenchymal tumors. BMC Gastroenterol 2014;14:7. Yegin EG, Duman DG. Small EUS-suspected gastrointestinal stromal tumors of the stomach: An overview for the current state of management. Endosc Ultrasound 2016;5:69-77. Kim GH, Park DY, Kim S, et al. Is it possible to differentiate gastric GISTs from gastric leiomyomas by EUS? World J Gastroenterol 2009;15:3376-81. Antonini F, Fusaroli P, Frazzoni L, et al. EUS ELASTOGRAPHY STRAIN RATIO IN THE DIFFERENTIAL DIAGNOSIS OF GASTROINTESTINAL SUBEPITHELIAL
LESIONS: PRELIMINARY RESULTS OF A MULTICENTER STUDY. Endoscopy 2018;50:OP170.
Appendix 1 Range of strain ratio for 3 measurements
Pathological diagnosis
Strain ratio mean (range of 3 measurements) 2.2 (1.7-2.8) 2.0 (1.8-2.2) 3.0 (2.4-3.7) 4.0 (3.7-4.4) 9.2 (8.5-10.4) 5.2 (3.8-7.4) 2.8 (2.3-3.5) 3.5 (2.3-6.3)
Leiomyoma (n=17)
5.0 (3.7-6.7) 4.5 (3.5-6.0) 3.8 (3.2-4.2) 2.0 (1.1-3.2) 16.4 (14.0-18.2) 2.5 (1.7-3.4) 6.0 (5.2-6.8) 2.5 (2.1-2.9) 29.0 (17.1-35.1) 52.3 (43.0-60.5) 29.0 (23.5-37.9) 42.5 (39.5-54.0)
GIST (n=7)
56.7 (52.3-59.4) 47.8 (40.2-54.3) 62.3 (58.6-65.3) 67.0 (63.3-69.2)
Abbreviations: SET, subepithelial tumor; EUS, endoscopic ultrasound; GIST, gastrointestinal stromal tumor; ROI, region of interest; APC, argon plasma coagulation; ROC, receiver operating characteristic; EUS-FNA, EUS-fine needle aspiration; EUS-TCB, EUS guided trucut biopsy; IT knife, insulation-tipped knife; CEEUS, contrast-enhanced EUS; EUS-FNB, EUS fine needle biopsy
S0016-5107(19)32245-X
DOI:
https://doi.org/10.1016/j.gie.2019.09.012
Reference:
YMGE 11753
To appear in:
Gastrointestinal Endoscopy
Received Date: 2 April 2019 Accepted Date: 4 September 2019
Please cite this article as: Kim SH, Yoo IK, Kwon C-I, Hong SP, Cho JY, Utility of EUS elastography in the diagnosis of gastric subepithelial tumors: a pilot study (with video), Gastrointestinal Endoscopy (2019), doi: https://doi.org/10.1016/j.gie.2019.09.012. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. 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. Copyright © 2019 by the American Society for Gastrointestinal Endoscopy
Utility of EUS elastography in the diagnosis of gastric subepithelial tumors: a pilot study (with video) Sun Hwa Kim MD, PhD, In Kyung Yoo MD, PhD, Chang-Il Kwon MD, PhD, Sung Pyo Hong MD, PhD, Joo Young Cho MD, PhD Department of Gastroenterology, Cha Bundang Medical Center, Cha University College of Medicine, Seongnam-si, Korea
Address Correspondence and Reprint requests to: In Kyung Yoo, M.D.PhD Professor of Medicine Director of Department of Gastroenterology Cha Bundang Medical Center 59 Yatapro, Bundang-gu, Seongnam-si, Korea Tel: (+82) 31-780-5641 Fax: (+82) 32-780-5005 [email protected]
Abstract Background and study aims: EUS elastography is a real-time imaging technique that analyzes the tissue elasticity. The aim of this study was to investigate the applicability of the quantitative EUS elastography in the differential diagnosis of the gastric subepithelial tumors (SETs). Methods: We prospectively registered 31 consecutive patients with gastric SETs and measured their strain ratios with EUS elastography. The strain ratios of gastric SETs were compared with the histopathological diagnosis. Results: Thirty patients (mean age 51.4 ± 12.6 years) were included in the analysis. The mean size of SETs was 2.3 ± 0.7 cm. Lipoma was measured at the lowest strain ratio of 1.6 (1.1-2.0), followed by leiomyoma 6.0 (2.0-29.0), ectopic pancreas 11.8 (1.7-29.3), gastrointestinal stromal tumors (GIST) 51.1 (29.0-67.0), and schwannoma 62.0. With the cutoff value of 22.7, EUS elastography could differentiate GIST from leiomyoma with the sensitivity and specificity of 100 % and 94.1 %, respectively (p=0.001; 95 % CI, 0.9791.000). Conclusions: EUS elastography could be a promising diagnostic adjunct for the assessment of gastric SETs, especially in differentiating GISTs from leiomyoma. Key words; Elastography, Endoscopic ultrasound, Gastric subepithelial tumor
Introduction EUS has been recommended for the differential diagnosis of gastric subepithelial tumors
(SETs) because it can provide information on the characteristics, including their size, margin, internal echogenicity, and the layer of origin. 1, 2 However, despite being considered as the basic diagnostic tool for gastric SETs, it is frequently inconclusive when differentiating the subtypes of gastric SET. Moreover, it is challenging to decide whether or not to perform invasive cytopathological confirmation, and unfortunately, obtaining sufficient tissue is technically difficult especially for those, which are located in the deep layer.3 EUS elastography is a real-time imaging technique that displays the tissue elasticity differences between diseased and normal tissues on the conventional B-mode ultrasound images.1 It analyzes the degree of tissue deformation under compression and reflects this information as a hue color spectrum. The current applications of elastography were extended in diverse diseases including those related to the pancreas,4-6 breast,7 thyroid gland,8 liver,9 prostate,10 and lymph node 5, 6. Several studies have demonstrated that it improved the diagnostic accuracy of conventional ultrasonography. Giovannini et al5, 6 reported the potential of EUS elastography in the differentiation of pancreatic masses and malignant lymph nodes with good sensitivity. In regard to gastric SETs, a previous study11 reported that EUS elastography could be helpful in differentiating gastrointestinal stromal tumors (GISTs) from other gastric SETs by color-based qualitative grading because GISTs have relatively firm mechanical properties than others. However, as this is a visual color-based qualitative method, interobserver bias is possible. In contrast, elastographic method with calculated strain ratio would be reproducible and yield more objective results.1 Most studies on quantitative EUS elastography investigated it in the context of pancreatic diseases and proposed a cut-off strain value for malignant lesions.12 Instead, to date, few studies have evaluated gastric SMTs according to their tissue elasticities using EUS elastography. Therefore, we aimed to investigate the feasibility of EUS
elastography in the differential diagnosis of the gastric SETs based on strain ratio. Methods Patients From April 2018 to December 2018, we prospectively recruited 41 consecutive patients with gastric SET diagnosed by esophagogastroduodenoscopy at Bundang CHA Medical Center. The strain ratios of gastric SETs were measured using EUS elastography and this was followed by histopathological confirmation after lumpectomy (n = 31). The exclusion criteria were as follows: gastric SET <1 cm, located in esophagus or duodenum, severe coagulopathy, cardiopulmonary disease, chronic kidney disease, and unwillingness to undergo lumpectomy. The study was approved by the ethics review board of Bundang CHA Medical Center (CHAMC 2019-01-015-003), and written informed consent was obtained from all patients. EUS guided elastography and measurement of the strain The procedures were carried out by an experienced endoscopist (JY Cho) using an EUS endoscope (EG-3670URK; Pentax Lifecare Division, Hoya Co, Ltd, Tokyo, Japan and GF TYPE UCT 260; Olympus Corporation, Tokyo, Japan) combined with diagnostic ultrasound imaging systems (HI VISION Ascendus; Hitachi Aloka Medical Ltd, Tokyo, Japan and EUME2; Olympus Corporation). First, conventional B-mode endoscopic ultrasound images were assessed to characterize the size of gastric SET, the relation of gastric SET to the layers of the stomach wall, and the internal echogenicity. Second, the elasticity of gastric SET was quantitatively measured using strain ratios (Fig. 1). They were measured when the compression wave reached the negative peak indicating that the probe has just released from the gastric SET. The average strain value within region of interest (ROI) of gastric SET was measured as A, and the strain value of
water filled in the EUS scope balloon was measured as B for the reference. The ratio of A to B (B/A quotient), strain ratio, was calculated by Aloka diagnostic ultrasound imaging systems. The strain ratio was measured higher, in case the tumor was hard. We checked the strain values of the lesion (A) and the reference (B) 3 times repeatedly and considered the average elasticity for analysis to ensure precise measurement. Endoscopic resections and pathologic diagnosis All endoscopic resections were performed by a single experienced endoscopist (J.Y.C.). Capassisted ESD technique using APC was performed for endoscopic resection. First, snare mucosectomy was performed and revealed SET. Careful dissection was performed after submucosal injection. For submucosal dissection, we used a dual knife (KD-650L, Olympus Medical Systems, Tokyo, Japan) or APC and then used the cap to peel the muscularis propria layer along the capsule of the lesion. We termed this process as “cap-assisted ESD technique using APC," because spray-coagulation is currently used for dual knife or APC to dissect the lesion, and the cap is used to remove the lesion using mechanical force. The transparent cap, which is attached onto the tip of the endoscope, applies tension to the connective tissues during dissection. This technique helps to peel the capsule of the diseased site more easily and safely from the muscularis propria layer. The gastric wall defects were not routinely closed after endoscopic resection; however, two cases of small perforations were successfully managed by clips with a detachable snare. Histopathologic diagnosis was made and subsequently compared to the strain ratio. Statistical analysis Data are presented as means, standard deviations, or ranges, as appropriate. Intergroup differences were compared using the Kruskal–Wallis test for continuous data and chi-squared
test or linear by linear association for categorical data. A P value of < 0.05 was considered to be statistically significant. The optimal cut-off values were selected as the points of the receiver operating characteristic (ROC) curves where the sum of the sensitivity and the specificity become maximum. SPSS software (version 25.0; SPSS Inc, Chicago, Ill, USA) was used to analyze the data.
Results Patients characteristics A total of 41 patients who were diagnosed with gastric SET on esophagogastroduodenoscopy were enrolled. Six patients were excluded as the size of the gastric SMT was less than 1 cm and 4 patients did not want to undergo lumpectomy. Finally, 31 patients were included in the analysis. The mean age of the patients was 51.4 ± 12.6 (range 31-76), and 20 (64.5%) of them were male. The mean size of gastric SETs was 2.3 ± 0.7 cm (1.0-3.5). The most common final pathological diagnosis was leiomyoma in 17 patients, followed by GISTs in 7, ectopic pancreas in 4, lipoma in 2, and schwannoma in 1. The gastric SETs were located in the body (n = 11), cardia (n = 9), fundus (n = 4), antrum (n = 4), and angle (n = 3) of the stomach (Table 1). Elasticity analysis for the subtypes of the gastric SETs We compared the results of the final histopathological diagnosis of 31 patients with their respective strain ratios (Table 2). Lipoma was measured at the lowest strain ratio of 1.6 ± 0.7 (range, 1.1 2.0), presenting with hyperechogenic mass arising from submucosal layer in EUS. The most common gastric SET was leiomyoma with a mean strain ratio of 6.0 ± 6.9 (2.0 -
29.0). In EUS examinations, leiomyoma appeared as hypoechoic lesions arising from muscularis mucosa in 10 cases, and from muscularis propria in 8. Patients with ectopic pancreas showed a mean strain ratio of 11.8 ± 12.1 (1.7 - 29.3). There was considerable overlap in the strain ratio of leiomyoma and that of lipoma (p = 0.38) and between the strain ratio of leiomyoma and that of ectopic pancreas (p = 0.21). Therefore, the data showed no statistical difference in strain ratio between the groups. On the other hand, in those with GIST, it presented as hypoechoic mass in muscularis propria in EUS and showed higher mean strain ratio of 51.1 ± 11.8 (29.0 – 67.0). Six of them were classified as very low risk GISTs, and only one case was confirmed as intermediate risk with the highest strain value of 67.0. The mean strain value of GIST was significantly higher (p < 0.001) than that of other gastric subepithelial tumors except schwannoma (Fig. 2). Schwannoma, an irregular hypoechoic mass arising from muscularis propria in EUS, showed a strain ratio of 62.0. ROC analysis of the strain ratio yielded an area under the curve of 0.961 (p=0.001; 95 % CI, 0.896-1.000) for GIST. With the cut-off value of 22.7, the sensitivity and specificity of strain ratio measured by EUS elastography for distinguishing GIST from other gastric SETs were 100.0% and 87.5%, respectively (Fig. 3A). In addition, ROC analysis on subgroups of leiomyoma versus GIST was assessed. The data yielded an area under the curve of 0.996 (p=0.001; 95 % CI, 0.979-1.000) and the cut-off value of 22.7 demonstrated the sensitivity and specificity, 100 % and 94.1 %, respectively (Fig. 3B). Discussion In this study, we investigated the benefit of quantitative EUS elastography in the diagnosis of gastric SETs using strain ratio. Elastography, first described in the 1990s, is an attractive
imaging technology that is based on the difference of tissue mechanical properties.13 The differential diagnosis among gastric SETs during EUS examination is sometimes challenging, and thereby requiring EUS-guided fine-needle aspiration (EUS-FNA)14,
15
or EUS-guided
trucut biopsy (EUS-TCB).16 In addition, these invasive procedures may result in insufficient or negative cytopathologic diagnosis depending on the tissue acquisition technique with a limited diagnostic adequacy of 38% to 89%.15-17 Therefore, the ability to differentiate the etiology of gastric SETs with less-invasive diagnostic technique is clinically desirable. Several studies have been conducted to distinguish GISTs from other SETs to increase the preoperative diagnostic rate and to avoid unnecessary surgery.18-20 An analysis of 62 patients revealed that GISTs present with hyperenhancement on contrast-enhanced EUS (CEEUS) with a sensitivity and specificity of 98% and 100%, respectively, and that 88% of the GISTs contain avascular areas.19 Another study that used digital image analysis software demonstrated that GISTs have a higher echogenicity and heterogeneity within a tumor, as compared with other benign SETs.20 To improve the diagnostic rate of gastric SETs, we analyzed whether noninvasive elastography during EUS examination would be useful in the clinical diagnosis of gastric SETs. Our result demonstrated that lipoma and leiomyoma had lower strain ratios, whereas GIST and schwannoma showed relatively higher strain ratios, indicating that the latter have harder tissue properties. Among gastric SETs, lipoma presents as hypodense mass in abdominal CT, and in endoscopy, it is easily compressed and lifted up by biopsy forceps with yellow colored appearance. Ectopic pancreas is strongly suggested by its specific feature of umbilicated surface corresponding to a draining duct. In contrast, it is occasionally difficult to differentiate between leiomyomas and GISTs in clinical practice due to identical appearance on EUS; therefore, deciding whether to follow up conservatively or to undergo subsequent tumor resection is challenging.21 A previous study22 reported that gastric GISTs tend to
present with heterogenicity, hyperechoic spots, marginal halo and higher echogenicity on EUS but it is not always possible to predict GISTs according to these features. A preliminary study by Antonini et al23 demonstrated that EUS elastography with the strain ratio improved the differential diagnosis of SETs, in that strain ratios of GISTs were higher than those of other benign GI lesions. However, this preliminary study permitted results from EUS-fine needle biopsy (EUS-FNB) or clinical follow-up, whereas we made the final diagnosis only by histological confirmation obtained from endoscopic resection. Also, although they regarded a healthy gastrointestinal wall as the reference, we introduced a more reliable measurement using the water level in the EUS endoscope balloon. We decided to use water as our reference value because measurements of multiple layers of the thin gastric wall may vary each time. Based on our results, if the lesion is suspicious for either leiomyoma or GIST on EUS interpretation, the strain ratio more than 22.7 favors GIST rather than leiomyoma with the sensitivity of 100% and specificity of 94.1%. This suggests that additional elastography might enhance the diagnostic accuracy of EUS on the differential diagnosis of gastric SETs. The strain ratios were found to be significantly higher in GIST than in leiomyoma. In our data, GISTs and leiomyomas were of similar size; therefore, the strain ratios were not affected by lesion size. To maximize the quality of elastography, we measured the strain ratios of the lesion 3 times repeatedly. Before measuring the strain ratios, we checked the elastic color distribution on EUS images and tried to include most of the elastic heterogeneity in ROI. In addition, to acquire reproducible and objective data, we checked the compression wave displayed at ultrasound monitor to measure the strain value when the probe was just released against gastric SET.1 The elastography procedure time took about 5 to 10 minutes with no procedure-related adverse events, highlighting that it is a relatively convenient and safe method.
There are some limitations in our study. First, the patients recruited in the study were small in number. It might be necessary to recruit more patients to conclude more reliable results. Second, strain ratios were measured 3 times during each procedure. Although there was some variability between the readings, differences were minimal (Appendices). The mean value was calculated and used as the final result for each patient. Additionally, the strain value was measured by one experienced endoscopist (J.Y.C.) while the patient was sedated, and the strain ratio at the most appropriate video sequences was chosen, supporting the adequacy of elastographic measurements. Third, patients diagnosed with GIST were mostly in the verylow-risk group, and only 1 patient was in the intermediate risk group; therefore, we could not accurately evaluate the elastic difference in the risk groups. Although there was a trend of a higher strain ratio in intermediate risk GISTs than in very-low-risk cases, even in the small number of cases, the strain ratio alone would not substitute cytopathology in assessing the mitotic activity or immunohistochemistry features. Further studies with larger sample size and inclusion of other subtypes of SETs are necessary to enhance the accuracy of the cut-off value of strain ratio in the differential diagnosis of gastric SETs. In conclusion, this study demonstrated that EUS elastography could be a promising diagnostic technique for the assessment of gastric SETs. Disclosure The authors have no conflicts of interest to disclose.
Table 1. Patient demographics and clinicopathological characteristics of gastric subepithelial tumors. Characteristic
Total (n=31)
Sex, male/female, n (%)
20/11 (64.5/35.5)
Age, mean (SD), years
51.4 (12.6)
Tumor location, n (%) Cardia
9 (29.0)
Fundus
4 (12.9)
Body
11 (35.5)
Antrum
4 (12.9)
Angle
3 (9.7)
Tumor size on EUS, mean (SD), cm
2.3 (0.7)
1 ≤size <2 cm, n (%)
14 (45.2)
2 ≤size <3 cm, n (%)
12 (38.7)
≥3 cm, n (%)
5 (16.1)
Pathological diagnosis, n (%) Leiomyoma
17 (54.8)
GIST
7 (22.6)
Ectopic pancreas
4 (12.9)
Lipoma
2 (6.5)
Schwannoma
1 (3.2)
Table 2. The strain ratios of gastric subepithelial tumors according to the final histopathologic diagnosis.
Pathological diagnosis
Strain ratio mean (SD)
range
Lipoma (n=2)
1.6 (0.7)
1.1-2.0
Leiomyoma (n=17)
6.0 (6.9)
2.0-29.0
Ectopic pancreas (n=4)
11.8 (12.1)
1.7-29.3
GIST (n=7)
51.1 (11.8)
29.0-67.0
Schwannoma (n=1)
62.0
62.0
Figure legends Figure 1. EUS elastography images. The strain ratio was measured when the compression wave (green) reached the negative peak indicating the probe has just been released from gastric subepithelial tumor. A, The hypoechoic tumor with the strain ratio of 3.7 was finally diagnosed with leiomyoma. B, Whereas, the hypoechoic lesion with the strain ratio of 52.3 was diagnosed with GIST. Figure 2. The differences between mean strain ratios of GIST were statistically significant compared to those of lipoma (P < .002), leiomyoma (P < .001), and ectopic pancreas (p = 0.001). P < .05 was considered statistically significant. (Schwannoma was excluded in the analysis because of the small number.) Figure 3. Receiver operating characteristic curve analysis of the strain ratio for the differentiation of gastrointestinal stromal tumor (GIST) from other gastric subepithelial tumors (A) and from leiomyoma (B). References 1. 2. 3.
4.
Sigrist RMS, Liau J, Kaffas AE, et al. Ultrasound Elastography: Review of Techniques and Clinical Applications. Theranostics 2017;7:1303-1329. Cui XW, Chang JM, Kan QC, et al. Endoscopic ultrasound elastography: Current status and future perspectives. World J Gastroenterol 2015;21:13212-24. Hwang JH, Saunders MD, Rulyak SJ, et al. A prospective study comparing endoscopy and EUS in the evaluation of GI subepithelial masses. Gastrointest Endosc 2005;62:202-8. Kawada N, Tanaka S. Elastography for the pancreas: Current status and future perspective. World J Gastroenterol 2016;22:3712-24.
5.
6.
7. 8.
9. 10. 11.
12.
13. 14. 15.
16.
17.
18. 19.
20.
21.
22. 23.
Giovannini M, Hookey LC, Bories E, et al. Endoscopic ultrasound elastography: the first step towards virtual biopsy? Preliminary results in 49 patients. Endoscopy 2006;38:344-8. Giovannini M, Thomas B, Erwan B, et al. Endoscopic ultrasound elastography for evaluation of lymph nodes and pancreatic masses: a multicenter study. World J Gastroenterol 2009;15:1587-93. Bayat M, Denis M, Gregory A, et al. Diagnostic features of quantitative comb-push shear elastography for breast lesion differentiation. PLoS One 2017;12:e0172801. Cosgrove D, Barr R, Bojunga J, et al. WFUMB Guidelines and Recommendations on the Clinical Use of Ultrasound Elastography: Part 4. Thyroid. Ultrasound Med Biol 2017;43:4-26. Friedrich-Rust M, Poynard T, Castera L. Critical comparison of elastography methods to assess chronic liver disease. Nat Rev Gastroenterol Hepatol 2016;13:402-11. Clinical practice guidelines for ultrasound elastography: prostate. J Med Ultrason (2001) 2016;43:449-55. Tsuji Y, Kusano C, Gotoda T, et al. Diagnostic potential of endoscopic ultrasonography-elastography for gastric submucosal tumors: A pilot study. Dig Endosc 2016;28:173-8. Carrara S, Di Leo M, Grizzi F, et al. EUS elastography (strain ratio) and fractal-based quantitative analysis for the diagnosis of solid pancreatic lesions. Gastrointest Endosc 2018;87:1464-1473. Gennisson JL, Deffieux T, Fink M, et al. Ultrasound elastography: principles and techniques. Diagn Interv Imaging 2013;94:487-95. 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. 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. 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. Fernandez-Esparrach G, Sendino O, Sole 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. Chu CS. Gastrointestinal Stromal Tumors: How to Increase the Preoperative Endoscopic Ultrasonography Diagnostic Rate. J Med Ultrasound 2018;26:177-180. Ignee A, Jenssen C, Hocke M, et al. Contrast-enhanced (endoscopic) ultrasound and endoscopic ultrasound elastography in gastrointestinal stromal tumors. Endosc Ultrasound 2017;6:55-60. Kim GH, Kim KB, Lee SH, et al. Digital image analysis of endoscopic ultrasonography is helpful in diagnosing gastric mesenchymal tumors. BMC Gastroenterol 2014;14:7. Yegin EG, Duman DG. Small EUS-suspected gastrointestinal stromal tumors of the stomach: An overview for the current state of management. Endosc Ultrasound 2016;5:69-77. Kim GH, Park DY, Kim S, et al. Is it possible to differentiate gastric GISTs from gastric leiomyomas by EUS? World J Gastroenterol 2009;15:3376-81. Antonini F, Fusaroli P, Frazzoni L, et al. EUS ELASTOGRAPHY STRAIN RATIO IN THE DIFFERENTIAL DIAGNOSIS OF GASTROINTESTINAL SUBEPITHELIAL
LESIONS: PRELIMINARY RESULTS OF A MULTICENTER STUDY. Endoscopy 2018;50:OP170.
Appendix 1 Range of strain ratio for 3 measurements
Pathological diagnosis
Strain ratio mean (range of 3 measurements) 2.2 (1.7-2.8) 2.0 (1.8-2.2) 3.0 (2.4-3.7) 4.0 (3.7-4.4) 9.2 (8.5-10.4) 5.2 (3.8-7.4) 2.8 (2.3-3.5) 3.5 (2.3-6.3)
Leiomyoma (n=17)
5.0 (3.7-6.7) 4.5 (3.5-6.0) 3.8 (3.2-4.2) 2.0 (1.1-3.2) 16.4 (14.0-18.2) 2.5 (1.7-3.4) 6.0 (5.2-6.8) 2.5 (2.1-2.9) 29.0 (17.1-35.1) 52.3 (43.0-60.5) 29.0 (23.5-37.9) 42.5 (39.5-54.0)
GIST (n=7)
56.7 (52.3-59.4) 47.8 (40.2-54.3) 62.3 (58.6-65.3) 67.0 (63.3-69.2)
Abbreviations: SET, subepithelial tumor; EUS, endoscopic ultrasound; GIST, gastrointestinal stromal tumor; ROI, region of interest; APC, argon plasma coagulation; ROC, receiver operating characteristic; EUS-FNA, EUS-fine needle aspiration; EUS-TCB, EUS guided trucut biopsy; IT knife, insulation-tipped knife; CEEUS, contrast-enhanced EUS; EUS-FNB, EUS fine needle biopsy