BackgroundInvasive tests measuring resistance to cerebral spinal fluid (CSF) outflow and the effect of temporary drainage of CSF are used to select candidates affected by idiopathic normal pressure hydrocephalus (iNPH) for shunt surgery. Neither test, however, completely excludes patients from treatment. Perfusion and diffusion magnetic resonance imaging (MRI) are non-invasive techniques that might be of value in selecting patients for surgical treatment and understanding brain changes in iNPH patients. The aim of this study was to understand the role of perfusion and diffusion MRI in selecting candidates for shunt surgery and to investigate the relationship between cerebral perfusion and possible microstructural changes in brain tissue before and after invasive tests, and after ventricular-peritoneal (VP) shunt implantation, to better clarify pathophysiological mechanisms underlying iNPH.MethodsTwenty-three consecutive patients with probable iNPH were included in this study. Patients underwent a clinical and neuroradiological evaluation before and after invasive tests, and after surgery. Only patients who showed a positive result in at least one of the invasive tests were submitted for VP shunt implantation. Perfusion and diffusion magnetic resonance imaging (MRI) was performed before and after invasive tests and after shunt surgery.ResultsThirteen patients underwent surgery and all showed clinical improvement after VP shunt implantation and a significant increase in perfusion in both periventricular white matter (PVWM) and basal ganglia (BG) regions. The 10 patients that did not have surgery showed after invasive tests, a significant reduction in perfusion in both PVWM and BG regions. Comparing the changes in perfusion with those of diffusion in positive patients we found a significant positive correlation in BG and a significant inverse correlation in PVWM area.ConclusionsPerfusion MRI is a non-invasive technique that could be useful together with invasive tests in selecting patients for surgical treatment. Furthermore, the relationship between perfusion and diffusion data could better clarify pathophysiological mechanisms underlying iNPH. In PVWM area we suggest that interstitial edema could reduce microvascular blood flow and interfere with the blood supply to these regions. In BG regions we suggest that a chronic hypoxic insult caused by blood hypo-perfusion produces a chronic cytotoxic edema. Both in PVWM and in BG regions, pathophysiological mechanisms could be modified after VP-shunt implantation.