This study was performed to determine if a correlation exists between pancreatic parenchymal enhancement or lack thereof on contrast-enhanced CT and surgical evidence of pancreatic necrosis. Accurate CT assessment of pancreatic vascular perfusion would be helpful in preoperative planning before pancreatic debridement and necrosectomy. The CT scans and medical records were blindly and retrospectively reviewed in 13 patients with the operative diagnosis of pancreatic necrosis. In all cases, CT examinations preceded surgery within 72 hr. Surgical and pathologic findings served as the gold standards for the diagnosis of pancreatic necrosis. Seven (54%) of 13 patients had a region or regions of parenchymal necrosis within the head, body, and/or tail of the pancreas seen during surgery. All seven of these patients had no pancreatic enhancement on CT in at least one region of the pancreas (CT sensitivity = 100%). In four of the seven, two regions were found to be necrotic at surgery, but only one of the two segments did not enhance on CT scans. In three of the seven patients, lack of contrast enhancement on CT (no enhancement of the pancreatic head and body in two patients and throughout the gland in another) correlated with necrosis in the same regions at surgery. The remaining six (46%) patients, who had peripancreatic or small, focal, and/or superficial areas of pancreatic necrosis at surgery, had normal pancreatic enhancement on CT. Our results show that regions of pancreatic necrosis found surgically correlate with lack of enhancement of pancreatic parenchyma on CT. Peripancreatic necrosis and minor areas of focal or superficial parenchymal necrosis were not detected on CT.