Value of image enhancement of endoscopic ultrasound for diagnosis of gastrointestinal subepithelial lesions

Abstract

Objectives: Among subepithelial lesions (SELs), gastrointestinal stromal tumors (GISTs) should be identified and surgically treated at an early stage. However, it is difficult to diagnose SELs smaller than 20 mm. In recent years, endoscopic ultrasound (EUS) elastography (EUS-EG) and contrast-enhanced harmonic EUS (CH-EUS) have been reported to be useful for the diagnosis of SELs, although the diagnostic accuracy of a combination of EUS techniques with image enhancement is unknown.

Methods: Patients with SELs who underwent EUS-guided tissue acquisition, EUS shear-wave elastography (EUS-SWE), EUS strain elastography (EUS-SE), and CH-EUS from January 2019 to June 2023 were enrolled. To assess the diagnostic accuracy for differentiating GISTs from other SELs, shear-wave velocity on EUS-SWE, the strain ratio on EUS-SE, and vascularity on CH-EUS were determined and their diagnostic accuracies were compared.

Results: Forty-three patients were enrolled. When the cut-off value was set at 3.27 m/s, the sensitivity, specificity, and diagnostic accuracy of shear-wave velocity were 28.6%, 86.2%, and 34.9%, respectively. When the cut-off value was set at 3.79, the sensitivity, specificity, and diagnostic accuracy of the strain ratio were 93.1%, 64.3%, and 83.7%, respectively. The sensitivity, specificity, and diagnostic accuracy of CH-EUS were 79.3%, 92.3%, and 83.7%, respectively. When EUS-SE was combined with CH-EUS, the sensitivity and diagnostic accuracy were the highest among binary combinations of image enhancement modalities.

Conclusions: EUS-SE and CH-EUS are useful for differentiating GISTs from other SELs. Furthermore, the use of both modalities may further improve the identification of GISTs.

Keywords: EUS shear‐wave elastography; EUS strain elastography; GIST; contrast‐enhanced harmonic EUS; subepithelial lesions.

FIGURE 1

(a) Image of a subepithelial lesion (SEL) examined by endoscopic ultrasound shear‐wave elastography (EUS‐SWE). Shear‐wave velocity (Vs) was measured 10 times in the present study. (b) Image of a SEL examined by endoscopic ultrasound strain elastography (EUS‐SE). The strain ratio was measured twice in the present study. Evaluation of images near the apex of the graph waveforms is desirable, and as far as possible images of the apex of the graph were considered. (c) Image of a pathology specimen stained with Masson's trichrome (×400). A BZ‐800 microscope (Keyence Corporation) was utilized to measure the degree of fibrosis in resected specimens.

FIGURE 2

Shear‐wave velocity (Vs) results: (a) Box‐and‐whisker plot of Vs in the gastrointestinal stromal tumor (GIST) and non‐GIST groups. (b) Receiver operating characteristic (ROC) curve for the differential diagnosis of GISTs and non‐GISTs using Vs. (c) ROC curve for the differential diagnosis of GISTs and non‐GISTs smaller than 20 mm using Vs.

FIGURE 3

Strain ratio results: (a) Box‐and‐whisker plot of the strain ratio in the gastrointestinal stromal tumor (GIST) and non‐GIST groups. (b) Receiver operating characteristic (ROC) curve for the differential diagnosis of GISTs and non‐GISTs using the strain ratio. (c) ROC curve for the differential diagnosis of GISTs and non‐GISTs smaller than 20 mm using the strain ratio.

FIGURE 4

Correlations between the degree of fibrosis and tissue hardness values of gastrointestinal stromal tumors (GISTs). (a) Correlation between the degree of fibrosis and shear‐wave velocity (Vs) for overall GIST cases. (b) Correlation between the degree of fibrosis and the strain ratio for overall GIST cases.