13 new and 19 explanted hydrocephalus shunt valves were tested under perfusion for two weeks, using a modified ASTM test. Some manufacturers seem to have problems with quality control. Our results show large deviations in pressure-flow-characteristics in different valves, in some cases leading to an extremely high pre-shunt pressure. Test results almost corresponding to manufacturer's specifications were found in diaphragm valves and in some ball and spring valves. Results from slit valves usually varied widely, some being good, but others bad (much too high a pressure) or dangerous with even possibly fatal characteristics. Whilst laboratory performance may not correlate with clinical results, due to different impact of cerebrospinal fluid (CSF) pressure and flow on hydrocephalus shunting systems, manufacturers should provide devices that meet their supposed specifications. We developed an apparatus allowing simulation of different CSF production rates, CSF outflow resistances, CSF pressure waves caused by arterial and venous pulsation, breathing and coughing, and hydrostatic differential pressure caused by body position changes. This model enables testing of shunt values under almost physiological conditions, thus closing the gap between conventional bench test results and clinical performance of hydrocephalus shunt valves.
