Contrast-enhanced Ultrasound (CEUS) and Shear Wave Elastography (SWE) Features for Characterizing Serous Microcystic Adenomas (smas): in Comparison to Pancreatic Neuroendocrine Tumors (Pnets)

Xiao-Fan Tian,Ling-Yun Yu,Dao-Hui Yang,Dan Zuo,Jia-Ying Cao,Ying Wang,Zi-Yi Yang,Wen -Hui Lou,Wen -Ping Wang,Wei Gong,Yi Dong
DOI: https://doi.org/10.1016/j.heliyon.2024.e25185
IF: 3.776
2024-01-01
Heliyon
Abstract:Objectives: Serous microcystic adenoma (SMA), a primary benign pancreatic tumor which can be clinically followed-up instead of undergoing surgery, are sometimes mis-distinguished as pancreatic neuroendocrine tumor (pNET) in regular preoperative imaging examinations. This study aimed to analyze preoperative contrast-enhanced ultrasound (CEUS) and shear wave elastography (SWE) features of SMAs in comparison to pNETs. Material and methods: In this retrospective study, patients with imaging-diagnosed pancreatic lesions were screened between October 2020 to October 2022 (ethical approval No. B2020-309R). Performing by a Siemens Sequoia (Siemens Medical Solutions, Mountain View, CA, USA) equipped with a 5C-1 curved array transducer (3.0–4.5 MHz), CEUS examination was conducted to observe the microvascular perfusion patterns of pancreatic lesions in arterial phase, venous/late phases (VLP) using SonoVue® (Bracco Imaging Spa, Milan, Italy) as the contrast agent. Virtual touch tissue imaging and quantification (VTIQ) – SWE was used to measure the shear wave velocity (SWV, m/s) value to represent the quantitative stiffness of pancreatic lesions. Multivariate logistic regression was performed to analyze potential ultrasound and clinical features in discriminating SMAs and pNETs. Results: Finally, 30 SMA and 40 pNET patients were included. All pancreatic lesions were pathologically proven via biopsy or surgery. During the arterial phase of CEUS, most SMAs and pNETs showed iso- or hyperenhancement (29/30, 97 % and 31/40, 78 %), with a specific early honeycomb enhancement pattern appeared in 14/30 (47 %) SMA lesions. During the VLP, while most of the SMA lesions remained iso- or hyperenhancement (25/30, 83 %), nearly half of the pNET lesions revealed an attenuated hypoenhancement (17/40, 43 %). The proportion of hypoenhancement pattern during the VLP of CEUS differed significantly between SMAs and pNETs (P = 0.021). The measured SWV value of SMAs was significantly higher than pNETs (2.04 ± 0.70 m/s versus 1.42 ± 0.44 m/s, P = 0.002). Taking a SWV value > 1.83 m/s as a cutoff in differentiating SMAs and pNETs, the area under the receiver operating characteristic curve (AUROC) was 0.825, with sensitivity, specificity and likelihood ratio (+) of 85.71 %, 72.73 % and 3.143, respectively. Multivariate logistic regression revealed that SWV value (m/s) of the pancreatic lesion was an independent variable in discriminating SMA and pNET. Conclusion: By comprehensively evaluating CEUS patterns and SWE features, SMA and pNET may be well differentiated before the operation. While SMA typically presents as harder lesion in VTIQ-SWE, exhibiting a specific honeycomb hyperenhancement pattern during the arterial phase of CEUS, pNET is characterized by relative softness, occasionally displaying a wash-out pattern during the VLP of CEUS.
What problem does this paper attempt to address?