Fifty-four shunt-responsive patients were selected from a prospective protocol directed to study patients with suspected normal pressure hydrocephalus (NPH). Patients with gait disturbances, dementia, non-responsive L-Dopa Parkinsonism, urinary or faecal incontinence and an Evans ratio greater or equal to 0.30 on the CT scan were included in the study. As a part of their work-up all patients underwent intracranial pressure monitoring and hydrodynamic studies using Marmarou's bolus test. According to mean intracranial pressure (ICP) and the percentage of high amplitude B-waves, patients were subdivided in the following categories: 1) Active hydrocephalus (mean ICP above 15 mmHg), which is in fact no tone normal pressure hydrocephalus; 2) Compensated unstable hydrocephalus, when mean ICP was below 15 mmHg and B-waves were present in more than 25% of the total recording time and 3) Compensated stable hydrocephalus when ICP was lower or equal to 15 mmHg and beta waves were present in less than 25% of the total recording time. The majority of the patients in this study (70%) presented continuous high or intermittently raised ICP (active or unstable compensated hydrocephalus group). Mean resistance to outflow of CSF (Rout) was 38.8 mm Hg/ml/min in active hydrocephalus and 23.5 mm Hg/ml/min in the compensated group (Students t-test, p less than 0.05). Higher resistance to outflow was found in patients with obliterated cortical sulci and obliterated Sylvian cisterns in the CT scan. No statistically significant correlation was found when plotting the percentage of beta waves against pressure volume index (PVI), compliance or Rout. An exponential correlation was found when plotting beta waves against the sum of conductance to outflow and compliance calculated by PVI method (r = 0.79). Patients with the so-called normal pressure hydrocephalus syndrome have different ICP and CSF dynamic profiles. Additional studies taking into consideration these differences are necessary before defining the sensitivity, specificity and predictive value of ICP monitoring and CSF studies in selecting appropriate candidates for shunting.