Article Text
Abstract
Introduction Recent data suggest that QEUSE, a novel technique that allows real-time quantification of tissue stiffness, can accurately differentiate benign from malignant solid pancreatic masses (area under the receiver operating curve [AUROC]=0.98).1 External validation of the diagnostic utility of this technique has not been carried out.
Methods 101 patients with CT and/or EUS-proven solid pancreatic masses underwent 108 QEUSE procedures using the Hitachi EUB-7500 or Preirus ultrasound workstation and Pentax linear echoendoscopes. Multiple elastographic measurements of the mass lesion (A) and soft tissue references areas (B) were undertaken and the corresponding strain ratios (B/A) were calculated. Final diagnosis was based on EUS-fine needle aspiration (EUS-FNA) cytology, biliary brushings and/or resection specimen histology. The diagnostic accuracy of QEUSE for discriminating malignant from benign pancreatic masses was assessed.
Results The median lesion size was 3 cm. The final underlying diagnoses were primary pancreatic carcinoma (71.3%), neuroendocrine tumour (9.9%), metastatic cancer (2%) and pancreatitis (16.8%). Malignant pancreatic masses had a higher strain ratio (p=0.002) and lower mass elasticity (p=0.003) than inflammatory ones. However, the AUROC for the detection of pancreatic malignancy was only 0.74 for the strain ratio and only 0.73 for the mass elasticity. Similarly, the diagnostic accuracy of QEUSE for detecting pancreatic malignancy in our cohort was less favourable than those reported recently (see Abstract OC-109 table 1), with lower strain ratio (4.62 vs 6.04) and higher pancreatic mass elasticity cutoffs (0.27 vs 0.05) providing the highest accuracy.
Conclusion In the largest single-centre study of QEUSE of the pancreas reported to date, we found this technology to be less accurate and specific for differentiating pancreatic masses than recently reported, suggesting that it may only complement rather than substitute the role of pancreatic EUS-FNA in the future.
Competing interests None declared.
Reference 1. Iglesias-Garcia, et al. Gastroenterology 2010.