Elevated intracranial pressure can cause an inadequate supply to the brain and ischemia as well as the nerves and brain cells to die as a result of the reduced cerebral blood flow. Furthermore, it can cause the dangerous displacement of brain tissue in the skull, e.g. displacement of parts of the cerebellum towards the foramen magnum. This is referred to as herniation.    

The aims of treating the patient to reduce elevated intracranial pressure is therefore to restore or maintain the blood supply to the brain and prevent the displacement of brain tissue. Treatment should also improve the well-being of the patient and reduce or, ideally, relieve symptoms, such as headaches and nausea.

In the past, treatment focused on lowering the ICP to below a defined limit. The diagnostic basis for this is invasive pressure sensors, which are used to determine the intracranial pressure e.g. in neurointensive care units. These devices display the measured mean ICP.

However, the past few years in particular have shown that looking solely at an ICP limit is not enough to effectively treat patients (Carrera, Kim, Castellani, 2010⁷). There are many reasons for this: Reducing the ICP to a single figure results in the loss of a lot of valuable information about the cerebrospinal regulation processes. Such limits overly simplify the very complex and heterogeneous pathologies (Le Roux, 2016⁸). Limit values are also typically calculated in studies with a large number of participants, so the focus is not on the individual patient. Such limit values generally only allow the treating physician to react if the given intracranial pressure has been exceeded (Le Roux, 2016⁹).

To further explain this, below is an example comparison of two patients whose intracranial pressure is above the physiological, i.e. healthy, range at the time the measurement was taken: Patient A has high compliance (Le Roux, 2016¹⁰) , i.e. small changes to the intracranial conditions do not cause a significant change in ICP. The elevated pressure readings may be well tolerated by this patient. On the other hand, patient B shows reduced compliance. Small pathological increases in volume lead to a significantly elevated ICP and critical reactions in this patient. Both patients must be treated differently for the increase in ICP, focusing on an ICP limit is not sufficient in these cases. Other parameters, such as compliance, must be taken into account when treating each patient and when considering different treatment options. But how can these parameters be determined?

Advances in measurement technology and data processing have now made it possible to evaluate larger data sets in real-time (e.g. while the measurements are being taken). This provides the treating physician with new information that would not have been accessible in the past without the support from technology. As a result, analysing the time of the ICP curves is becoming more and more prevalent in the field of intracranial pressure sensors. It could be shown that the dynamics of ICP, its waveform and the resulting parameters provide useful information that increases the quality of the patient’s treatment and proactively counteracts critical trends (Le Roux, 2016¹¹) (Czosnyka, Smielewski, 2007¹²).

The ICP waveform can have three main components (Czosnyka, Smielewski, 2007¹³):

1. Pulse waveform, associated with the heartbeat (typical heart rate 50-180 bpm)
2. Respiratory waveform, associated with breathing (typical respiratory rate of 8-20 cycles/minute)
3. Slow vasogenic waveform, i.e. from the blood vessels
       for example so-called Lundberg A and B waves can be observed in the pressure signal (typically 0.3-3 cycles/minute)(Czosnyka, Smielewski, 2007¹⁴)

An ICP curve over time produces the following example image (modified according to Czosnyka, Smielewski, 2007¹⁵):

The pulse (1) and respiratory (2) waveforms show significantly smaller pressure changes compared to the vasogenic (3) Lundberg A and B waves.

The pulse waveform is caused by the pulsation of blood due to the contractions of the heart and spreads through the blood vessels to the skull. The resulting pulsating volume changes lead to pressure pulses in the cranial cavity. This causes regular pressure peaks in the ICP, with the time between the maxima being determined by the heart rate.

So-called pulse amplitudes (AMP) are determined in diagnostics as a quantitative measure of the intensity of the pressure pulses. These refer to the pressure difference between the maximum and the two adjacent minima of each pulse. The pulse amplitudes are also referred to as the mean ICP wave amplitude (MWA) in the literature.

Figure according to (TeachMeSurgery, 2020¹⁷)

This correlation allows information about compliance to be obtained by quantifying the pulse amplitudes. It is known that, for example, amplitudes greater than 4 mmHG can indicate pathological changes with reduced compliance (Eide, 2016¹⁸).

The pulse amplitudes can be measured straight from the ICP curve. The readings depend on possible drift effects of the pressure sensors.

In order to determine the pulse amplitudes, the measurement system must have a sufficiently high sampling rate in order to be able to precisely determine these values. Good software is also required that can objectively and automatically analyse the ICP waveforms (Pennacchietti, 2020¹⁹).

The detailed analysis of the ICP volume curve shows a distinct exponential rise. When the intracranial pressure reaches a certain level, a turning point in the curve is observed where the graph flattens out as the ICP increases. Thus, the pulse amplitudes rise as the ICP increases up to the turning point when they drop again. Therefore, small pulse amplitudes on the left (high compliance with low ICP) as well as on the right (low compliance with high ICP) can be observed from the turning point. When considered together, the ICP and pulse amplitudes therefore do not clearly describe the part of the ICP volume curve in which the patient’s intracranial state is.

In order to make a clear statement on compliance and the compensatory reserve for each patient, another parameter - the RAP index - was introduced in addition to the above parameters. The RAP index describes the statistical correlation between the mean ICP and AMP. R in the term RAP stands for the correlation coefficient, A is the abbreviation for amplitude and P is the mean pressure. The value of the RAP index can be from -1 to +1 (Czosnyka , Smielewski, Timofeev, 2007²⁰).

What general statements can be made with a correlation coefficient? The correlation coefficient generally describes the dependency of variable A on variable B:

Example 1: The values of variable A increase while the values of variable B increase linearly. In this case, the correlation coefficient of both variables is +1, which is referred to as a positive linear relationship.

Example 2: The values of variable A increase while the values of variable B decrease linearly. The correlation coefficient takes the value of -1, in this case, a negative linear relationship exists between the two variables.

A correlation coefficient of 0 indicates no correlation, i.e. changes to variable A do not result in changes to variable B. Values between 0 and 1 or 0 and -1 indicate a correlation between both variables that is not fully linear (Ratner, 2009²¹).

If this is applied to the RAP index, the following statements can be made: When intracranial pressure readings are small, small volume changes on the left part of the ICP volume curve do not cause the pulse amplitudes to increase. Since there is no correlation between these two values, the RAP index is 0. 

When the intracranial pressure readings are higher, the RAP index takes the value of >0 around the steep rise of the ICP volume curve, which means the rise in ICP leads to an increase in the pulse amplitudes. The compensatory reserve is low in this range, so small volume changes lead to a sharp increase in ICP. If the ICP further increases, the pulse amplitude drops again from the turning point of the ICP volume curve. Here, the RAP index is <0. At the turning point, the RAP therefore changes sign. This identifies a critical range from which elevated ICP values can lead to ischemia, irreversible brain damage and herniation (Jin, Choi, Kim,2019²²). Ultimately, pulsations are no longer observed in the pressure signal when the pressure readings are very high, the RAP index once again takes the value of 0. 

From a practical perspective, the RAP index is relevant e.g. in neurointensive medicine. In most patients who are admitted to hospital with traumatic brain injuries, a good compensatory reserve is observed within the first few hours (RAP is 0), however this deteriorates when cerebral oedema occurs. The RAP then consistently shows values close to +1 (Czosnyka, Smielewski, Timofeev, et al., 2007²³).

On the one hand, a reduction or increase in the RAP index may be due to the change in position of the patient’s craniospinal system on the ICP volume curve. On the other hand, pathological effects may also lead to a change in the shape of this curve over time and a different RAP index being calculated.   

In summary, it can be stated that the RAP index has proven to be a reliable measure for the compensatory reserve (Czosnyka, Steiner, Balestreri, et al., 2005²⁴) and compliance (Varsos, Kasprowicz, Smielewski, Czosnyka, 2014²⁵).

In order to determine the RAP index, the pressure measurement system must have low drift and high precision as well as a suitable sampling rate of the pressure signal in order to calculate the RAP based on the calculated mean ICP and pulse amplitudes.

A detailed analysis of the curve of the pulsatile wave over time reveals a characteristic fine structure. It shows that every pressure pulse is comprised of at least three single pulses. These are identified on the time axis from left to right with P1, P2 and P3. Where do they come from and can they be used for a diagnosis?

The P1 wave, also referred to as the “percussion wave”, is the result of the arterial pulse that spreads to the cranial cavity. P1 is the result of the direct expansion of the arterial walls due to the pulse (so-called windkessel effect) that is transferred to CSF and other intracranial components and can therefore be identified in the pressure signal (Czosnyka, Czosnyka, 2020²⁶).

By comparing the time curves of the arterial blood pressure and the ICP, other components in the intracranial pressure signal can be identified. P3, also referred to as the “dicrotic wave”, is likely to originate from the veins.  

Figure according to (March, Hickey, 2016²⁷) and (Czosnyka, Czosnyka, 2020²⁸)

For P2, however, no direct association to a part of the arterial pressure curve could be found. Examinations show that P2 may be linked to the pulsatile arterial intracranial blood volume (Czosnyka, Czosnyka, 2020²⁹). These pulsatile increases in volume cause pulsatile pressure changes in the skull as a response (see chapter “The diagnostic value of pulse amplitudes”). The amplitude of P2 therefore depends on the elastic properties of the cranial cavity and therefore on compliance.

Figure according to (Stettin, 2008³⁰)

If compliance is high, the arterial pulse will be cushioned by the cranial cavity, the intensity of the pulse is P1 > P2 > P3 (see point (1) in the curve). With high compliance, the cranial cavity acts like a sponge that absorbs the pulses. If intracranial compliance is reduced, intracranial components cause the pulse to be transferred more strongly, which leads to an increase in the intensity of P2 so that the intensities shift: P1< P2 > P3 (see points (2) and (3) in the curve). In this case, the elastic properties of the cranial cavity act like a stone with low compliance. The fine structure of the individual pressure pulses is therefore dependent on compliance.

In summary, in addition to the pulse amplitudes AMP, the fine structure of the individual pressure pulses also contains diagnostic information about compliance.

In order to determine the fine structure, the pressure measurement system used must have a sufficiently high sampling rate in order to be able to determine the intensities of the three components in the individual pressure pulses.

In addition to the pulsatile waveform, morphologies can also be observed in the ICP curve that originate from respiration. This is due to the following effects: During inhalation, the intrathoracic pressure drops, which causes the cerebral venous pressure to reduce. The outflow of the blood increases and the ICP drops. The opposite effect when exhaling causes a sinusoidal pattern in the ICP curve in line with the respiratory rate. An increase of the intrathoracic pressure and therefore the increase of the ICP can also be observed during the Valsalva manoeuvre (e.g. during bowel movements) and when coughing, sneezing and vomiting (Peate, Wild, Nair, 2018³¹).

As described above, pulsatile and respiratory components can overlap in intracranial pressure curves as follows:

Important parameters can be derived from the ICP curves over time and the mean ICP (here approx. 5 mmHg) can be determined. The mean is shown in the figure in blue. The peaks of the pulsatile component show a gap of approx. 0.7 seconds (purple marking), i.e. the associated contraction of the heart has a frequency of 85/min. The pulse amplitude of the pulsatile component is high in the example at approx. 5 mmHg, which may indicate reduced compliance (red marking).

The peaks of the respiratory component show a gap of approx. 7 seconds, i.e. the breathing rate is approx. 9 breaths/minutes (green marking). The observed amplitudes of the respiratory component are significantly smaller than the amplitudes of the pulsatile component in the example shown.

When measuring the ICP, slower changes to the ICP signal can also be observed in addition to high frequent pulsatile and respiratory components. These are morphologies of vasogenic (from the blood vessels) origin.

According to Lundberg, three waveforms A, B and C can be identified (Lundberg, 1960³²). The most important characteristics of Lundberg A and B waves are described below. Explanations of the Lundberg C waves will be omitted due to their unclear clinical relevance.

Lundberg A waves

Lundberg A waves are also referred to as plateau waves. They have the following shape: Starting with a certain ICP, the ICP increases sharply, followed by a plateau lasting for 5 to 20 minutes and pressures over 50 mmHg are reached. After this time, the pressure readings drop dramatically.

Lundberg A waves are caused by the following effects: The ICP readings that are typically high to begin with cause a reduction in the cerebral perfusion pressure (CPP). This is a measure of circulation in the brain. The CPP is the difference between the mean arterial pressure (MAP) and the ICP.

If the cerebral perfusion pressure is no longer enough to meet the metabolic needs in the cranial cavity, this causes cerebral artery vasodilation, i.e. expansion of the blood vessels as part of so-called autoregulation. This increases the cerebral blood volume, which, in turn, causes an even higher ICP. The repeated cycle causes life-threatening ICP readings. Lundberg A waves are pathological and must be treated immediately.

Lundberg B waves

The description of Lundberg B waves is very inconsistent in the literature, various waveforms are described using this term (Martinez-Tejada, Arum, Wilhjelm, Juhler, Andresen, 2019³³). For example, there are reports that Lundberg B waves can be both sinusoidal and ramp-shaped, with the individual wave crests occurring at a frequency of 0.5 to 3/min.

It is described that these waves could indicate reduced compliance (Spiegelberg, Preuß, Kurtcuoglu, 2016³⁴). However, there is currently no consensus in the literature about the value of Lundberg B waves in diagnosing hydrocephalus and predicting whether a patient can benefit from a shunt.

The ramp-shaped Lundberg B waves appear to be linked to elevated pCO₂ readings (i.e. an increase in the concentration of CO₂ in the arterial blood) caused by snoring and sleep apnoea (Spiegelberg, Preuß, Kurtcuoglu, 2016³⁵ and Román, Jackson, Fung, et al., 2019³⁶).  On the other hand, sinusoidal Lundberg B waves may be related to changes in the arterial blood pressure, which are not associated with the breathing rhythm or arterial CO₂ (Spiegelberg, Preuß, Kurtcuoglu 2016³⁷).

In summary, various morphologies can occur in the curve of ICP readings. For example, the pulsatile and respiratory waves are physiological i.e. they also occur in healthy people. By contrast, the Lundberg A waves clearly indicate pathological changes and must be treated immediately due to the life-threatening high ICP readings.

Thanks to technological advances, a wide range of additional information for diagnostics can now be derived from the intracranial pressure over time. It therefore comes as no surprise that solely determining the mean pressure is gradually losing importance in a clinical setting. The parameters taken from the ICP over time, such as pulse amplitudes and RAP, can help to quantify the pulsatile waveform of the pressure curves and make important conclusions about the compliance and compensatory reserves. The detection of further curve morphologies, such as Lundberg A waves, is also very important.

Complying with technical requirements for measuring intracranial pressures is an important basis for selecting the suitable treatment for each patient from analysing the curve.     

Bibliography

Chapter 1:

1: Lawrence Yu, Brian J Kim, Ellis Meng; Chronically implanted pressure sensors: challenges and state of the field; Sensors, 2014 Oct 31;14(11):20620-44.

2: Jiang G., Zhou D.D. (2009) Technology Advances and Challenges in Hermetic Packaging for Implantable Medical Devices. In: Zhou D., Greenbaum E. (eds) Implantable Neural Prostheses 2. Biological and Medical Physics, Biomedical Engineering. Springer, New York, NY

3: Schuhmann et al., Pediatr. Neurosurg., 2008, 44, 269-279

4,5,6: Holm et al., Medical Engineering & Physics 31 (2009) 601–606

------------------

Chapter 2:

7: Carrera E, Kim DJ, Castellani G, et al. What shapes pulse amplitude of intracranial pressure?. J Neurotrauma. 2010;27(2):317-324. doi:10.1089/neu.2009.0951

8,9,10,11: Le Roux P. Intracranial Pressure Monitoring and Management. In: Laskowitz D, Grant G, eds. Translational Research in Traumatic Brain Injury. Boca Raton (FL): CRC Press/Taylor and Francis Group; 2016.

12, 13,14, 15: Czosnyka M, Smielewski P, Timofeev I, et al. Intracranial pressure: more than a number. Neurosurg Focus. 2007;22(5):E10. Published 2007 May 15. doi:10.3171/foc.2007.22.5.11

16: Czosnyka M, Pickard JD. Monitoring and interpretation of intracranial pressure. J Neurol Neurosurg Psychiatry. 2004;75(6):813-821. doi:10.1136/jnnp.2003.033126

17: TeachMeSurgery, 2020: teachmesurgery.com/neurosurgery/flow-and-pressure/monro-kellie-doctrine/ abgerufen am 7. September 2020

18: Eide PK. The correlation between pulsatile intracranial pressure and indices of intracranial pressure-volume reserve capacity: results from ventricular infusion testing. J Neurosurg. 2016;125(6):1493-1503. doi:10.3171/2015.11.JNS151529.

19: Pennacchietti, Valentina et al. “Single center experiences with telemetric intracranial pressure measurements in patients with CSF circulation disturbances.” Acta Neurochirurgica (2020): 1 - 11.

20: Czosnyka M, Smielewski P, Timofeev I, et al. Intracranial pressure: more than a number. Neurosurg Focus. 2007;22(5):E10. Published 2007 May 15. doi:10.3171/foc.2007.22.5.11

21: Ratner, B. The correlation coefficient: Its values range between +1/−1, or do they?. J Target Meas Anal Mark 17, 139–142 (2009). doi.org/10.1057/jt.2009.5

22: Jin SC, Choi BS, Kim JS. The RAP Index during Intracranial Pressure Monitoring as a Clinical Guiding for Surgically Treated Aneurysmal Subarachnoid Hemorrhage: Consecutive Series of Single Surgeon. Acute Crit Care. 2019;34(1):71-78. doi:10.4266/acc.2019.00437

23: Czosnyka M, Smielewski P, Timofeev I, et al. Intracranial pressure: more than a number. Neurosurg Focus. 2007;22(5):E10. Published 2007 May 15. doi:10.3171/foc.2007.22.5.11

24: Czosnyka M, Steiner L, Balestreri M, et al. Concept of "true ICP" in monitoring and prognostication in head trauma. Acta Neurochir Suppl. 2005;95:341-344. doi:10.1007/3-211-32318-x_70

25: Varsos GV, Kasprowicz M, Smielewski P, Czosnyka M. Model-based indices describing cerebrovascular dynamics. Neurocrit Care. 2014;20(1):142-157. doi:10.1007/s12028-013-9868-4

26: Czosnyka M, Czosnyka Z. Origin of intracranial pressure pulse waveform. Acta Neurochir (Wien). 2020;162(8):1815-1817. doi:10.1007/s00701-020-04424-4

27: nursekey.com/intracranial-hypertension-theory-and-management-of-increased-intracranial-pressure/ abgerufen am 07. September 2020

28, 29: Czosnyka M, Czosnyka Z. Origin of intracranial pressure pulse waveform. Acta Neurochir (Wien). 2020;162(8):1815-1817. doi:10.1007/s00701-020-04424-4

30: d-nb.info/1001032489/34 abgerufen am 7. September 2020

31: Ian Peate, Karen Wild, Muralitharan Nair: Nursing practice: knowledge and care, edited, 2018, Wiley

32: Lundberg N. Continuous recording and control of ventricular fluid pressure in neurosurgical practice. Acta Psychiatr Scand Suppl. 1960;36(149):1-193.

33: Martinez-Tejada I, Arum A, Wilhjelm JE, Juhler M, Andresen M. B waves: a systematic review of terminology, characteristics, and analysis methods. Fluids Barriers CNS. 2019;16(1):33. Published 2019 Oct 15. doi:10.1186/s12987-019-0153-6

34, 35: Andreas Spiegelberg, Matthias Preuß, Vartan Kurtcuoglu,B-waves revisited, Interdisciplinary Neurosurgery, Volume 6, 2016, Pages 13-17, doi.org/10.1016/j.inat.2016.03.004.

36: Román, G.C., Jackson, R.E., Fung, S.H. et al. Sleep-Disordered Breathing and Idiopathic Normal-Pressure Hydrocephalus: Recent Pathophysiological Advances. Curr Neurol Neurosci Rep 19, 39 (2019). doi.org/10.1007/s11910-019-0952-9

37: Andreas Spiegelberg, Matthias Preuß, Vartan Kurtcuoglu,B-waves revisited, Interdisciplinary Neurosurgery, Volume 6, 2016, Pages 13-17, doi.org/10.1016/j.inat.2016.03.004.