Simultaneous continuous intracranial pressure (ICP) signals from two different ICP sensors (Codman ICP MicroSensor; Johnson & Johnson, Raynham, MA) placed within the brain parenchyma and the epidural space were compared in ten patients with idiopathic normal pressure hydrocephalus (iNPH). Comparisons were made at the single ICP wave level by determining differences in mean pressure, and pulse pressure amplitude (dP) and latency (dT, i.e. rise time). Differences for the parameters mean ICP, mean ICP wave amplitude and mean ICP wave latency were also determined during consecutive 6-s time windows. The ICP sensors located within the brain parenchyma and epidural space showed marked differences in mean pressure. On the contrary, there were minor differences between sensors in 8 of 10 patients regarding single wave pulse pressure amplitude (dP) and single wave latency (dT, i.e. rise time), and also with regard to the parameters mean ICP wave amplitude and mean ICP wave latency. The levels of mean ICP wave amplitudes used for predicting shunt response in iNPH patients were independent of sensor location in 9 of 10 patients. These results extend previous data that epidural ICP monitoring cannot be used to reliably determine mean ICP, however, epidural ICP monitoring is very useful for determining ICP waveform parameters such as pulse pressure amplitude (dP) and mean ICP wave amplitude. It is suggested that epidural ICP monitoring with determination of mean ICP wave amplitude can be used to predict shunt response in iNPH, though an ICP sensor designed for epidural ICP monitoring would be preferable.