Intracranial pressure (ICP) is an important invasive monitoring parameter in management of patients with acute brain injury and compromised compliance. This study aimed to compare waveforms obtained from standard ICP monitoring and noninvasive ICP monitoring (nICP) methods.We analyzed continuous arterial blood pressure (ABP) waves, ICP (with standard monitoring), and nICP recorded simultaneously. All signal recordings were sliced into data chunks, each 1 min in duration, and from the mean pulse, we determined the time to peak (Tp) and the ratio between tidal and percussion waves (P2/P1). We also calculated the Isomap projection of the pulses into a bidimensional space-K1 and K2. The defined nICP and ICP parameters were compared using a unilateral Wilcoxon-Mann-Whitney test. The Pearson correlation coefficient and normalized mutual information were used to verify the association between parameters.In total, 1504 min of monitoring from ten patients were studied. Nine of the patients were male. The mean age of the patients was 58.4 ± 10.4 years, and they had an initial Glasgow Coma Scale of 9 ± 4, a mean Simplified Acute Physiology Score (SAPS II) of 45.6, and an intensive care unit stay of 44 ± 45 days. With the exception of Tp, all parameters showed a weak linear association but presented a strong nonlinear association.Mutual information analysis and a bigger sample size would be helpful to build more refined models and to improve understanding of the waveform relationships.
Keywords: Arterial blood pressure; Dimensionality reduction; Intracranial pressure; Noninvasive; Waveform comparison.