The evaluation of patients with a suspected pituitary microadenoma: computer tomography compared to magnetic resonance imaging

Abstract

To perform a prospective study to compare the ability of magnetic resonance imaging (MRI) and computer assisted tomography (CT) to predict the position of a tumour within the pituitary fossa and to assess the clarity of the image generated by the two modalities in patients with suspected pituitary microadenomas. Nineteen patients were investigated with CT and MRI scans and underwent trans-sphenoidal exploration of the pituitary fossa. Both scans were assessed blind and independently by three neuroradiologists, A, B, and C; A and B examined the CT scans and B and C the MRI scans. The predicted site was then compared to the size and position of the tumour at surgery. Twenty-five patients were recruited and had both CT and MRI scans, 19 subsequently underwent trans-sphenoidal exploration of the pituitary fossa and formed the study population. The joint opinions of A and B of the CT scans were correct in 10/19, and those of B and C of the MRI scans were correct in 17/19 cases, P = 0.008. Observer agreement was used to assess the clarity of the image. A and B agreed about the site of the tumour on the CT scan in 14 of 19 cases (Kappa statistic 0.556); B and C agreed on the MRI scans in 19/19 cases (Kappa statistic 1), P = 0.025. After the exclusion of five cases found to be macroadenomas at surgery, the joint opinions of A and B of the CT scans were correct in 8/14, and those of B and C of the MRI scans were correct in 12/14 cases, P = 0.133. A and B agreed about the site of the tumour on the CT scan in 12/14 cases (Kappa statistic 0.653), B and C agreed on the MRI scans in 14/14 cases (Kappa statistic 1, no significant difference). In the assessment of suspected pituitary microadenomas MRI is the imaging technique of choice in view both of its greater diagnostic accuracy and the avoidance of radiation exposure.