Clinical applications of a non-invasive ICP monitoring method

Abstract

Background and purpose: Until now the assessment of intracranial pressure (ICP) requires invasive methods. A previously introduced mathematical model allowed the non-invasive estimation of ICP (nICP) from arterial blood pressure (ABP) and blood flow velocity (FV). In various studies we have investigated the accuracy of this method and possible clinical applications. Methods and results: Selected hemodynamic parameters, calculated from the cerebral blood FV and the ABP curves, were used to express the relationship between ABP input and ICP output by linear transformation rules. In several clinical studies the accuracy and possible benefits of this method of non-invasive ICP (nICP) assessment were investigated. Assessment of ICP plateau waves: In 17 severely head injured patients we verified this model by comparison of nICP and measured ICP during generation of plateau waves, recorded in seven of these patients. In all simulations plateau elevations of ICP were well replicated. The correlation coefficient between increase of nICP and real ICP was R=0.98; P<0.001. Lumbar infusion tests: Twenty one hydrocephalic patients were studied. Parallel increases in real ICP and nICP during lumbar infusion tests were evidently visible. Resistance of cerebrospinal fluid outflow (Rcsf) was computed using nICP and compared with Rcsf computed from real ICP. The mean error between real and non-invasive Rcsf was 4.1±2.2 mmHg min/ml. Cerebral autoregulation: One hundred and forty five patients were studied after severe head injuries. The state of autoregulation was assessed by moving correlation of cerebral perfusion pressure (CPP=ABP−ICP) and FV (Mx index). nICP instead of ICP was used to continuously estimate the state of autoregulation and to dynamically adapt the nICP procedure to this state. A median error between ICP and nICP of 6.0 mmHg was observed. Directly and non-invasively assessed Mx indices correlated highly significantly ( R=0.9; P<0.001). Conclusions: The results demonstrate that the nICP assessment model constitutes a reliable method to monitor ICP and may therefore provide various useful clinical applications.