A new method for processing of continuous intracranial pressure signals.

This paper describes a new method for processing of continuous pressure signals. Continuous intracranial pressure (ICP) signals were sampled at 100 Hz, converted into digital data and processed during 6s time windows. According to a new algorithm, cardiac beat-induced single ICP waves were identified; pressure waves caused by noise in the signal were rejected for further analysis. The amplitude and latency values of the accepted single ICP waves were determined. For accepted 6s time windows, the mean ICP wave was computed as mean ICP wave amplitude and mean ICP wave latency. Mean ICP for every time window was computed according to current practice as sum of pressure levels divided by number of samples. The mean ICP wave parameters provide information about the single ICP waves that is not given by mean ICP. The method has been implemented in software to be used during online ICP monitoring, revealing mean ICP wave amplitude, mean ICP wave latency and mean ICP as numerical values every 6s. The values are presented in trend plots. Verification of correct single ICP wave identification can be done during online ICP monitoring. The clinical significance of the method was illustrated in four patients by observations that mean wave amplitudes corresponded better to the acute clinical state than the mean ICP; mean wave amplitudes could be elevated despite a normal mean ICP. In one patient with ICP and arterial blood pressure (ABP) signals monitored simultaneously with identical time reference, there was a weak correlation between mean ICP and ABP wave amplitudes. It is tentatively suggested that the mean ICP wave parameters are related to intracranial pressure-volume compensatory reserve capacity (compliance).