CT-guided needle biopsy of the pancreas: a retrospective analysis of diagnostic accuracy.

The purpose of this study was to evaluate the accuracy of percutaneous needle biopsy of the pancreas and to compare its sensitivity with that of simple computerized tomographic (CT) prediction of malignancy. The records of 41 patients who underwent CT-guided percutaneous needle biopsies of the pancreas during a 4-yr period were reviewed retrospectively. Pre-biopsy CT scans were evaluated independently by two experienced radiologists in a blinded fashion. Needle biopsy yielded a correct diagnosis of cancer (sensitivity) in only 45% of patients ultimately found to have a malignancy, whereas the radiologic interpretation alone correctly predicted cancer in 76% of such patients. The negative predictive value of needle biopsy was no better (43%) than the radiologist's CT interpretation alone. Our findings confirm previous observations that percutaneous needle biopsy is an insensitive method of diagnosing pancreatic cancer. In our experience, needle biopsy proved less accurate than expert radiologic interpretation of CT scans alone.