Cause of fluctuations in ICP

The body has many excellent compensatory mechanisms that keep the ICP at a relatively constant level within a narrow range. However, the position of the patient’s body (lying down/standing up) has a physiological effect on ICP. The pressure (relative to atmospheric pressure) is slightly positive in the supine position, and slightly negative when standing. Actions such as coughing, sneezing and REM (rapid eye movement) sleep can also result in temporary physiological fluctuations in ICP. This is completely normal and healthy.

However, various diseases and injuries, such as traumatic brain injury caused by an accident, strokes, intracranial bleeding, inflammation of the brain or even brain tumours, can result in increased ICP and life-threatening situations.

Elevated ICP levels as well as dizziness, vomiting and impaired consciousness are also frequently observed in patients with hydrocephalus, a condition that causes disruptions to the CSF (cerebrospinal fluid) dynamics, in which either too much fluid builds up in the brain, the intracerebral circulation is disrupted or not enough fluid is resorbed. The most common treatment for patients with hydrocephalus is the implantation of a shunt in order to normalise the pressure in the head by draining CSF e.g. into the abdomen.

To better understand the mechanisms that cause elevated ICP, the Monro-Kellie hypothesis can be used. According to this theory, the total volume of three components, brain tissue, blood and CSF, within the skull must remain the same in order to keep cerebral pressure constant.

Monro-Kellie hypothesis

The total volume within the skull is made up of the three components - brain tissue, blood and CSF. Ideally, they exist in mutual equilibrium and can influence each other if this equilibrium becomes disrupted: if one component increases, another must be displaced from the solid skull where possible. As a result, the increase in volume of one component or the introduction of a new volume (e.g. a brain tumour) can lead to increased cerebral pressure. If the body’s own compensatory mechanisms become exhausted, the ICP can rise above a critical level. This reduces circulation in the brain, which can lead to a lack of oxygen, the death of nerve fibres or even the death of the patient. 

Current methods of measuring ICP

Knowing the ICP is important for deciding which treatments to use, such as when treating patients after a stroke, bleeding or a traumatic brain injury. The aim is to use suitable intensive care medical treatments to normalise the elevated ICP in order to prevent complications and subsequent damage. Monitoring the ICP is therefore a key component of neurointensive care.

The most common method of measuring ICP is placing tube-shaped probes into the skull that invasively determine the intracranial pressure. These are connected to an external machine that calculates the pressure and displays the levels. Due to the invasive nature of this procedure, the measurements can only be taken at the patient’s bedside in a hospital ward.

Depending on the location of the pressure-sensitive component in the tube-shaped probe, different types of invasive pressure sensors can be used. If piezoelectric and fibre-optic catheters are used, the pressure is measured right at the tip of the probe that is placed in the skull. Alternatively, the catheter may contain a filler, such as air or fluid as a pressure transmission medium, which transmits the pressure changes at the tip of the catheter to a pressure-sensitive component in an external reader. An example of this is external ventricular drains with an integrated pressure sensor.

Another option is using “telemetric” probes, which are fully implanted and allow the pressure readings to be obtained wirelessly, i.e. using radio waves (non-invasive).

Fully non-invasive sensors that can be attached to the head from the outside do not require implantation. These systems can sense the pressure in the cranial cavity using e.g. an ultrasound. However, the quantitative measurement of ICP cannot currently (in 2020) be achieved with these technologies.

Requirements for stable, reliable ICP sensors

In order to make important conclusions for diagnostics and improving treatment for each patient, it is essential for ICP sensors to ideally have the following key properties and functions:


1. Time monitoring

In general medical practice, measurements are usually taken during examinations. These measurements taken at selected points (e.g. using a lumbar puncture) only provide information about a specific moment. As a result, excessively high or low levels and critical trends may not be detected until later on e.g. after clinical symptoms have developed. Furthermore, it cannot be ruled out that the readings may be distorted due to so-called white coat hypertension and the stress of the examination situation.

In neurointensive care, tube-shaped ICP sensors are inserted for a period of one to two weeks and can continuously measure the pressure. This is referred to as time monitoring. Telemetric ICP sensors are an important advancement, particularly for outpatients, who can record the readings continuously throughout the day from home. The parameters and influencing factors that the treating physician derives from these readings can be vital for effectively treating patients.


2. High precision, long-term stability and low drift

In production, the sensors are calibrated, during which the readings from the sensor are compared with reference values. The differences are then compensated to correct production-related deviations so that each sensor can calculate the same pressure readings with high precision. It is also useful to eliminate possible temperature dependence of the pressure signal due to calibration.

In order to take reliable, long-term measurements, the pressure sensors must remain stable throughout their service life. When the ICP sensors are implanted, the electronics are exposed to a challenging environment that can greatly impact the long-term functionality. For example, there is the risk of fluid entering the sensor, resulting in corrosion or short circuits that can cause the sensor to fail. Drift of the sensor signal is a common observation, so the measured value no longer equates to the actual value. The sensor’s electronics can be protected from negative influences by hermetic encapsulation, which achieves long-term stability and functionality (Yu, Kim, Meng1, 2014). Metallic housings are an established solution for implantable medical devices due to their very low permeability to gases and liquids (Jiang , Zhou2, 2009). High long-term stability is critical, especially for the monitoring described above.


3. Miniaturisation

In order to facilitate the implantation and explantation with minimal trauma to the surrounding tissue, the placed pressure probes must be as small as possible. This is fairly simple for the tube-shaped ICP sensors, as the external reader supplies the energy and processes the readings.

If telemetric pressure probes are used, the technological challenges to be overcome are immense. The placement of the implant entirely within the cranial cavity or subcutaneously and the spatial circumstances there do not allow for the use of batteries with sufficient storage capacity and therefore significant dimensions for operating the sensor. For telemetric sensors, the energy must be transferred from outside to the implant wirelessly via induction. Particularly with higher sampling rates (see below), it is essential when transferring data to maintain a continuous stream of data without disruptions to the communication so that the pressure signal can be recorded in a useable form.  


4. Low infection rate

The tube-shaped pressure sensors are inserted transcutaneously (i.e. through the skin and the skull bone) into the cranial cavity. As with all invasive procedures, there is a risk of infection that exponentially increases after just a few days. The movements of the patient can also cause the probe to become dislocated (i.e. shift). This issue can be solved using so-called bolts, in which the pressure probe is secured to the skull with a screw, while also sealing the area through which the implant was inserted from any pathogens entering. To reduce the risk of infection, antimicrobial-impregnated catheters can also be used. However, the challenges above mean that the tube-shaped pressure sensors are only intended for use in hospitals for a few days. The telemetric pressure sensors above are also an alternative approach as these can be read from outside telemetrically and therefore non-invasively.


5. Ambient pressure compensation

With ICP, it is important to understand that it is given and interpreted as the differential pressure to the ambient pressure (air pressure). It is the difference to the ambient pressure that indicates possible critical cerebral pressure phases and not the actual absolute pressure in the ventricular system. The physiological range of this differential pressure is between approx. -5 and +15 cmH2O. In patients with hydrocephalus, small ICP changes of just a few cmH2O can decide whether a patient experiences symptoms or not. By contrast, the air pressure can fluctuate by more than 50 cmH2O if the weather changes. Height relative to sea level also has a major influence on ambient pressure. When ascending a mountain, the rule of thumb is that for every 1000 m of height, the air pressure drops by around 100 cmH2O.

The figure shows the close physiological window for normal ICP as well as the variance of the ambient pressure depending on the location or weather.



In order to determine the ICP, the pressure measurement systems used must include suitable methods for compensating for the ambient pressure. With tube-shaped sensors, the pressure can be compensated by their design, so the pressure can be measured relative to the ambient pressure straight through the insertion site of the pressure sensor. However, if telemetric pressure sensors are used, two absolute pressure measurements must be taken - the absolute intracranial pressure and the absolute ambient pressure from outside. The two measured pressures are automatically subtracted to calculate the ICP that is required for diagnostics and treatment.


6. Sampling rate

In the past, medical practice focused on measuring average ICP levels, which were used as a basis for selecting suitable treatment to normalise the intracranial pressures. However, current ICP research is looking more and more at analysing the dynamics of the ICP signal. The parameters that can be read from the time-dependent curves, such as amplitudes, can be used to assess compliance and the remaining intracranial compensatory mechanisms of the patient. It is currently assumed that amplitudes greater than 4 mmHG may indicate pathological changes that require therapeutic measures (Schuhmann3, 2008). Amplitude errors would misdirect patient management and are therefore unacceptable from a clinical perspective (Holm4, 2009).

In order to record the rather complex dynamics of intracranial pressure and correctly determine these parameters, the pressure sensors used must have a suitably high sampling rate. The sampling rate defines the number of samples taken per second. If the time-dependent pressure fluctuation is dynamic, the sampling rate must be large enough to properly display the curve.

Nyquist–Shannon sampling theorem

Nyquist–Shannon sampling theorem states that the sampling rate must be at least twice the frequency of the signal to be sampled. The impact of the sampling rate on the presentation of the complex ICP curves that are comprised of various single periodic signals can be shown as follows. At a sampling rate of e.g. 100 Hz (i.e. 100 measurements per second), all of the information from signals ranging between 0 and 50 Hz can be retrieved. Individual signals with higher frequencies cannot be represented. By reducing the sampling rate from 100 to 25 Hz, the individual signals with frequencies of 12.5 to 50 Hz can no longer be determined. Reducing the sampling rate can therefore cause the fine structure of the original complex signals to no longer be adequately represented, resulting in a loss of information (Holm5, 2009).


Sampling rate using sound as an example

The role of the sampling rate can be easily described based on the following practical example:

The human ear can hear frequencies of up to 20 kHz. Nyquist–Shannon sampling theorem states that the signal must be sampled at twice the frequency. In this case, the sampling rate must be at least 2x 20 kHz = 40 kHz in order to adequately represent the signal. The sampling rate used in the industry for audio files on CDs is 44.1 kHz.

On the right-hand side, two example sounds are presented with very different sampling rates to better illustrate this. The higher the sampling rate, the higher the quality of the resulting sound. If the sampling rate is low, information is lost, which can be easily heard.

Tests on the minimally acceptable sampling rate for intracranial pressure curves show that with measurements under 25 Hz, the curves cannot be realistically recorded and the resulting ampli-tudes are incorrect.


Figure according to Holm et al., Medical Engineering & Physics - data from table 1

With sampling rates from 50 Hz and higher, the curves are represented accurately (Holm6, 2009).


Measuring and monitoring ICP are important tools for deciding the most suitable treatment of various illnesses and trauma for each patient using effective diagnostics. Despite the sometimes high technological requirements of ICP sensors, the treating physician has various systems available to determine the intracranial pressure.

The requirements may vary depending on the case. According to current technology and for the most challenging cases of long-term non-invasive measurements, it can be said that the following requirements are particularly important for high-quality ICP measurement: a high sampling rate of more than 25 Hz, telemetric – i.e. non-invasive – measurement for a low risk of infection, automatic compensation of the ambient pressure, a small structure that transfers high-quality data as well as hermetic encapsulation of the electronics in a metal casing for a high service life and low drift.

Beyond the technical requirements, it is also interesting to look at the curves for ICP measurements.
What causes the curves? How are they read? What do they tell us?