Intracranial pressure waveform analysis: computation of pressure transmission and waveform shape indicators.

We studied transmission of arterial blood pressure to intracranial pressure by observing how the two pressure waveforms varied from baseline conditions to after postural change or jugular compression. Such experiments may lead to pressure waveform-based estimates of intracranial compliance. Using a single database of arterial blood pressure, central venous pressure, and intracranial pressure waveforms collected during baseline, jugular compresison, and head-elevated conditions from six Yucatan minipigs, we computed several numerical indicators of waveform shape to find an estimator of intracranial compliance. Of these indicators, two were based on the Fourier-decomposition of all three waveforms, and one was based on a new method for approximating the systolic slope of the intracranial pressure waveform. We computed amplitude transfer functions for the first six harmonics of the Fourier spectrum, treating intracranial pressure as system output and independently treating arterial blood pressure and central venous pressure as system inputs. Using these same inputs and outputs, we computed a single quotient based on the Fourier coefficients of the first six harmonics of the input and output waveforms. Finally, applying a Gaussian high-pass filter, we computed systolic slope approximations for all intracranial pressure wave cycles contained in a single respiratory cycle. Our third indicator was the mean-normalized variation of the slope approximations over a respiratory cycle. We studied how each composite at baseline varied with baseline mean intracranial pressure and how each composite changed from baseline as a result of a physical manipulation. Our analysis suggests that the composite based on respiratory variation of systolic slope approximations was positively correlated with mean intracranial pressure during baseline. The quotient based on Fourier coefficients with arterial blood pressure input seemed to increase from baseline to jugular compression. Composites that treated central venous pressure as input were both less correlated with mean intracranial pressure during baseline and exhibited less predictable changes from baseline to a physical manipulation than their counterparts that used arterial blood pressure as input. However, none of these apparent trends was statistically significant. The lack of statistically significant results may be due to the nature of the composites and/or the small sample size (n = 6). However, we hope this study stimulates further investigation of both central venous pressure-to-intracranial pressure (in addition to arterial blood pressure-to-intracranial pressure) transfer and automated computation of intracranial pressure waveform systolic slope. Such research may lead to noninvasively determined estimators of intracranial compliance.