In an attempt to determine to what extent the measurement of cerebrospinal fluid (CSF) outflow resistance (R(out)) can distinguish patients with normal-pressure hydrocephalus (NPH) who respond to shunt surgery from patients who fail to do so, the authors undertook a controlled trial. Eighteen patients with the diagnosis of NPH entered the study. In connection with the shunt operation all patients who entered the study underwent R(out) measurement by means of the pressure-volume index (PVI) method of bolus manipulation (Marmarou et al.). As a safety measure patients above 70 of age and patients with dementia as the only clinical symptom underwent continuous intraventricular pressure measurement before they entered the final study. As a result a total of fourteen patients underwent PVI test and shunt surgery during the same procedure. Patients with normal R(out) thus served as controls, since all patients underwent shunt surgery irrespective of the result of the R(out) measurement. Eight patients improved following surgery and the outcome did not correlate with result of PVI nor R(out) determinations.