Objective: Endoscopic ultrasonography [EUS] has been claimed to be superior to computed tomography in detecting pancreatic cancer. However, there have been only few studies comparing multi-detector computed tomography [MDCT] with EUS. This study aims to compare the detection rate of EUS with that of MDCT for pancreatic cancer.

Materials and Methods: This retrospective study took place in January 2009 to January 2014 at King Chulalongkorn Memorial Hospital, Bangkok, Thailand. Patients who underwent both EUS and MDCT for suspected or known pancreatic cancer were included. Patients with insufficient data, cystic pancreatic lesion, benign masses and/or other malignant masses were excluded. Only patients with pathologically proven pancreatic ductal adenocarcinoma were included for analysis. The detection rates of EUS and MDCT for pancreatic cancer were compared.

Results: Sixty-eight patients with pathologically proven pancreatic adenocarcinoma (mean age 63.2+10.9 years) were analyzed. The mean maximal diameters of lesions identified (n = 60) and missed (n = 8) by MDCT were 43.7+15.6 mm and 18.4+2.9 mm, respectively. EUS and MDCT detection rate for pancreatic cancer was 98.5% and 88.2%, respectively. In pancreatic tumors 2 cm in diameter, EUS and MDCT detection rate was 83.3% and 0%, respectively. The pancreatic tumors that were missed by EUS were associated with baseline calcified chronic pancreatitis.

Conclusion: EUS had a higher detection rate than MDCT for detection of pancreatic cancer, particularly for small pancreatic tumors. However, it missed pancreatic tumor in a patient with calcified chronic pancreatitis.

Keywords: Endoscopic ultrasonography, Endoscopic ultrasound, Endosonography, EUS, Pancreatic cancer, Pancreatic neoplasm, Solid pancreatic lesion