Objective: Various modern imaging procedures such as endoscopic retrograde cholangiopancreatography (ERCP), computed tomography (CT), and endoscopic ultrasonography (EUS) have been shown to be highly accurate in the diagnosis of specific disorders of the pancreas. However, prior information often causes bias in the interpretation of these results. Little information is available concerning the value of these examinations in the primary and differential diagnosis of suspected pancreatic disease-particularly in comparison with clinical evaluation, including laboratory tests and transabdominal ultrasound (TUS).
Methods: Clinical and imaging information (EUS, ERCP, and CT) was collected for 184 inpatients who were referred over a 5-yr period for evaluation of suspected pancreatic disease. On the basis of patient history, laboratory tests, and the results of routine TUS, one gastroenterologist, who was unaware of any of the other procedures or the final diagnosis, made a presumptive clinical diagnosis. CT and ERCP images and EUS videotapes were then analyzed by three different and independent examiners, who had the same clinical information except for the TUS results, but were completely blinded to the results of the other examinations and the patients' diagnoses. The final diagnoses were obtained by surgery, histology, and cytology, plus a follow-up of at least 1 yr (mean 35 months) in all noncancer cases.
Results: The final diagnoses were: normal pancreas (n = 36), chronic pancreatitis without a focal inflammatory mass (n = 53) or with a focal inflammatory mass (n = 18), and pancreatic malignancy (n = 77). Clinical evaluation, including ultrasonography, achieved a sensitivity for pancreatic disease of 94% but a specificity of only 35%. The figures for the sensitivity and specificity of the three imaging procedures were 93% and 94%, respectively, for EUS; 89% and 92% for ERCP; and 91% and 78% for CT (p < 0.05 for the specificity of clinical assessment vs all three imaging tests, p > 0.05 for comparison of the three imaging procedures). In the differential diagnosis between cancer and chronic pancreatitis as well as between malignant and inflammatory tumors, there was no difference among clinical assessment and the three imaging tests.
Conclusions: In a group of patients with a high suspicion of pancreatic disease, little additional sensitivity in the diagnosis of pancreatic disease is provided by sophisticated imaging procedures such as EUS, ERCP, and CT, in comparison with clinical assessment including laboratory values and TUS. However, the specificity can be substantially improved. To confirm the diagnosis, one of the three examinations is needed, depending on the suspected disease and local expertise. The imaging procedures should be performed in a stepwise fashion for specific purposes, such as exclusion of pancreatic disease and the planning of treatment in chronic pancreatitis and pancreatic cancer.