Sir.—McConnell et al1 are to be congratulated on their serial magnetic resonance imaging study of children before and after surgery for congenital heart disease. Their findings provide support for the concerns about cerebral injury with congenital heart disease and as a consequence of cardiopulmonary bypass. However, the true incidence of persistent cerebral changes may be less than that suggested in their report.
Muraoka et al2 performed computed tomographic scans before and after cardiopulmonary bypass and also found signs of ventricular enlargement and dilation of the subarachnoid spaces. These changes occurred in fewer children (six of 57) and were transient in all cases, resolving 6 to 11 months after surgery. This difference may be explained by the fact that Muraoka et al scanned their infants at a mean of 29 days after surgery compared with the mean of 10 days in the study of McConnell et al, thus