Abstract

Background: Typical clinical symptoms of chronic pancreatitis are vague and non-specific and therefore diagnostic tests are required, none of which provide absolute diagnostic certainly, especially in the early stages of disease. Ultrasonography (US) of the pancreas is challenging, given its retroperitoneal location with overlying structures and relatively small size. The quality and thereby the clinical usefulness of the pancreatic ultrasound imaging has rapidly advanced along with the technological progress.  Objectives: In this prospective observational cohort study, our aim was to evaluate the diagnostic accuracy of features detected by a high-end transabdominal US scanner compared with the CP diagnosis defined by a diagnostic score combining clinical and imaging features, the Mayo score. Results: A total of 25 eligible patients who fulfilled the Mayo score, the whole pancreas could be visualized in 16 patients (64%) (Visualization score #2 in all 3 segments), and sufficient visualization for inclusion was achieved in 22 patients (88%). The pancreatic tail was the part of the pancreas most frequently incompletely visualized (07 patients, 28%). In 03 patients, visualization of the entire pancreas was inadequate to determine a US score. We performed a sub-analysis on the group with minimal change CP represented by Mayo scores 0-6 (n=22). In this group we calculated a sensitivity of 0.58 (0.37-0.79) and a specificity of 0.99 (0.93-1) for the Rosemont score.  Summary & Conclusion: Abdominal ultrasound is a simple, non-invasive, widely available imaging tool. We found that the modality has good diagnostic accuracy and that the extent of sonographic changes is reflected by the grade of exocrine failure by the transabdominal approach. Its limitations are represented by the fact that it’s highly examiner dependent and patient-dependent.