The interpretation of data from continuous monitoring of intracranial pressure (ICP) in patients with suspected normal pressure hydrocephalus (NPH) is the subject of controversy. Despite the fact that overnight ICP monitoring is widely used for the diagnosis of NPH, normative criteria are poorly defined. The present study demonstrates that there is a relationship between the relative frequency, the absolute amplitude, the wavelength and the morphology of B-waves and different sleep stages. Intraventricular intracranial pressure was recorded continuously overnight in 16 patients with suspected normal pressure hydrocephalus. Simultaneous polysomnography was performed to investigate the relation of spontaneous ICP oscillations to different sleep stages. A correlative analysis was done with the data of 13 patients. Three patients were excluded, one who was awake throughout the night and two in whom polysomnography was incomplete due to technical reasons. The mean resting cerebrospinal fluid (CSF) pressure was 12.87 cm CSF. B-waves were observed in the ICP recordings of all patients. They were present for a mean of 72% of the total recording time. The relative frequency of B-waves was higher during REM sleep and sleep stage 2 as compared to wakefulness (87.8% and 83.2% vs. 56. p < 0.05). The absolute amplitude was higher during REM sleep than in wakefulness (9.56 vs. 3.44 cm CSF, p < 0.05). Wavelengths were longer in REM sleep than in wakefulness and stages 1 and 2 (62.4 vs. 42, 40.7 and 44.8 sec, p < 0.05). The morphology of B-waves was also related to different sleep stages. Ramp-type B-waves were associated with REM sleep in six patients, however, were also present in sleep stage 2 in three of them. Knowledge of the relation of spontaneous ICP oscillations to different sleep stages may help to establish physiological foundations and alterations. Furthermore, polysomnography may be useful to avoid erroneous interpretation of ICP recordings due to sleep stage related variability.