How useful is contrast enhancement after a normal unenhanced computed tomography brain scan?

An audit was performed of contrast use in 400 CT brain scans performed with and without contrast enhancement; all 400 scans were re-read by a second reader who did not know the result of the original reading. From this initial series 184 scans were identified in which the unenhanced series had been read as normal by both observers. Of these 184 studies, only one enhanced scan showed an abnormality, and this lesion is thought to be benign. In conjunction with published literature it is concluded that it is reasonable not to give contrast if there is a low expectation of a lesion being present, and if the unenhanced scan is normal. Protocols are presented for the use of intravenous contrast in CT brain scans.