N K Rollins1, P Nisen, K N Shapiro. 1. Department of Radiology, University of Texas Southwestern Medical Center and Neurosurgeons for Children, Dallas, USA.
Abstract
PURPOSE: We compared visibility of residual juvenile cerebellar pilocytic astrocytomas (JPAs) on early postoperative and follow-up MR studies to determine whether early postoperative MR imaging has a valid role as a baseline study. METHODS: We reviewed the MR images of 21 consecutive children who had undergone resection of cerebellar JPA. The diagnosis of residual tumor was made on the basis of nodular enhancement that corresponded to enhancing tumor on the preoperative MR studies and/or nonenhancing nodular T2 signal that corresponded to nonenhancing tumor. Because no patient received chemotherapy or radiation therapy, abnormal T2 signal or enhancement on the early postoperative study that resolved on the follow-up study was presumed to be due to peritumoral edema and/or surgical manipulation. Nodular T2 signal and/or enhancement in the tumor bed not seen on the initial postoperative MR study but present on the subsequent MR study and unchanged on serial follow-up MR studies was presumed to represent residual tumor rather than tumor that had recurred. RESULTS: Compared with follow-up studies, the initial postoperative MR images were true-positive for residual tumor in six patients, false-positive in five, equivocal for residual tumor in four, true-negative in five, and false-negative in one. Residual tumor did not consistently enhance, and peritumoral edema and changes resulting from surgical manipulation tended to mask or simulate residual tumor. CONCLUSION: Early postoperative MR imaging is not accurate in differentiating residual JPA from postoperative changes, and the role of early postoperative MR imaging as a baseline study for comparison with further studies is questionable.
PURPOSE: We compared visibility of residual juvenile cerebellar pilocytic astrocytomas (JPAs) on early postoperative and follow-up MR studies to determine whether early postoperative MR imaging has a valid role as a baseline study. METHODS: We reviewed the MR images of 21 consecutive children who had undergone resection of cerebellar JPA. The diagnosis of residual tumor was made on the basis of nodular enhancement that corresponded to enhancing tumor on the preoperative MR studies and/or nonenhancing nodular T2 signal that corresponded to nonenhancing tumor. Because no patient received chemotherapy or radiation therapy, abnormal T2 signal or enhancement on the early postoperative study that resolved on the follow-up study was presumed to be due to peritumoral edema and/or surgical manipulation. Nodular T2 signal and/or enhancement in the tumor bed not seen on the initial postoperative MR study but present on the subsequent MR study and unchanged on serial follow-up MR studies was presumed to represent residual tumor rather than tumor that had recurred. RESULTS: Compared with follow-up studies, the initial postoperative MR images were true-positive for residual tumor in six patients, false-positive in five, equivocal for residual tumor in four, true-negative in five, and false-negative in one. Residual tumor did not consistently enhance, and peritumoral edema and changes resulting from surgical manipulation tended to mask or simulate residual tumor. CONCLUSION: Early postoperative MR imaging is not accurate in differentiating residual JPA from postoperative changes, and the role of early postoperative MR imaging as a baseline study for comparison with further studies is questionable.