Model-Based Noninvasive Estimation of Intracranial Pressure from Cerebral Blood Flow Velocity and Arterial Pressure

Science Translational Medicine  11 Apr 2012:
Vol. 4, Issue 129, pp. 129ra44
DOI: 10.1126/scitranslmed.3003249

You are currently viewing the editor's summary.

View Full Text
As a service to the community, AAAS/Science has made this article free with registration.

Under Pressure

Measuring the pressure inside your head is no trivial task. It currently requires an invasive probe or catheter, inserted through a hole in the skull, directly into the brain or surrounding space. Monitoring intracranial pressure (ICP) is important for patients with severe neurological trauma, but because the process is so invasive, ICP measurements are avoided in routine assessment of other neuropathologies that could benefit from knowing this vital sign. To make ICP monitoring available to a larger patient population, Kashif et al. have developed a noninvasive method of estimating the pressure using tools already available in the clinic.

The authors created a simplified model of what’s inside our heads: brain tissue, blood vessels, and cerebrospinal fluid. The variables involved in this model included arterial blood pressure (ABP), cerebral blood flow (CBF), and ICP. Using an electrical circuit as an analog, Kashif and colleagues were able to describe ABP and CBF as the respective voltage and current, the cerebral vasculature as the resistor, and the compliance of the vasculature and brain tissue as the capacitor. Hospital records of 37 patients with traumatic brain injury contained data for ABP (taken from the radial artery) and CBF velocity (from transcranial ultrasound), which were then plugged into the authors’ model to estimate ICP. These estimated values were compared to the patient’s actual invasive ICP waveforms, correctly identifying elevated ICP (>20 mmHg) with high sensitivity and specificity.

The beauty of the model described by Kashif et al.—other than its ability to estimate ICP noninvasively—is that it does not require calibration or training before application. This feature makes it amenable to use in the clinic, immediately upon examining a patient. Such a model-based approach will find use in many clinical situations where real-time estimation of ICP could spare patients the pain of and recovery from more invasive methods of measurement.