Intracranial pressure waveforms
Intracranial pressure waveforms include distinct periodic components: heart pulse waves, respiratory waves and quasi-periodic slow vasogenic waves (Lundberg B waves). Every waveform has its characteristic frequency (heart rate 50-180 bpm, respiratory waves 8-20 cycles min-1 and slow waves 0.3-3 cycles min-1), and can be identified using spectral analysis . By definition, a frequency spectral analysis examines the spectral compositions and intensities of sinusoidal waveforms. he pulse and the respiratory waveform have a fundamental amplitude and several harmonic components . he amplitude of the fundamental component of the pulse waveform (AMP) is useful for the evaluation of various indices describing cerebrospinal pressure dynamics. It correlates positively with mean ICP, a finding that can be explained by the decrease in compliance in the steep part of the volume-pressure curve, as seen in Fig. 4.1. With rising ICP, every stroke volume ejected by the heart transiently increases intra-cranial volume and leads to an increase in AMP. he exponential shape of the pressure-volume relationship is not the only factor influencing the magnitude of ICP pulse waves. he delay between arterial inflow and venous or CSF outflow profiles (pulsatile CSF flow through aqueduct cerebri) varies with mean ICP and hence also shapes the ICP pulse waveform. Further modulatory factors include the elastic properties of cerebral arteries, actively modulated by CPP and PaCO2 , and the increase in pulsatility of arterial blood inflow with rising ICP.
Morphological composition of the pulse wave and system analysis both try to describe similarities and discrepancies between the pulse waveform of arterial pressure and ICP. Some authors have proposed a classification of three distinctive ‘peaks’ seen during heart evolution in ICP: P1, transmitted ‘passively’ from systole of ABP, and P2-P3, associated with pressure-volume compensation of arterial blood inflow (Fig. 4.6). Whether such a classification has deeper physiological relevance or clinical application remains to be seen.
Respiratory waves are almost always present in ICP recordings. he pressure signal itself is complex, as both arterial and venous factors contribute to the respiratory waves seen in ICP.
All components that have a spectral representation within the frequency limits of 0.05-0.0055 Hz (20 s-3 min) can be classified as slow or Lundberg B waves and nowadays are not defined as precisely as in the original Lundberg thesis. Slow waves occur due to fluctuations of CBF that lead to changes in intra-cranial blood volume and hence ICP. he origin and reason for the presence of slow waves is still under debate. As slow fluctuations of pressure also occur in ABP, ICP slow waves can be interpreted as a response to ABP variations, depending on the state of cerebral autoregulation. Another theory suggests that slow waves are triggered by a ‘central pacemaker’ due to a cyclic demand of brain metabolites. hese waveforms can also be seen in healthy subjects, but in this context they are much smaller amplitude (<3 mmHg). Increases in the amplitude of these slow waves above 8 mmHg and more suggest reduced intracranial compliance and imply intracranial pathology. On the other hand, a complete absence of slow waves is also a bad predictor in head-injured patients.
Fig. 4.6. Morphology of the pulse waveform of intracranial pressure (ICP), showing the distinctive peaks P1, P2 and P3. P1 is usually associated with systole of arterial blood pressure (ABP). Peaks P2 and P3 (delayed) are associated with blood volume transport. The arterial blood volume (CaBV) curve may be estimated by time integration of flow velocity (FV) detected with transcranial Doppler ultrasonography. Peaks P2 and P3 seem to be associated with arterial blood volume changes and the pressure-volume compensatory reserve. a.u., Arbitrary units.
Lundberg A waves, known as ‘plateau waves’, are pathological slow vasogenic waves that m ay lead to ICP rising above 40 and up to 80 mmHg. hey dramatically reduce CPP in minutes and cause cerebral ischaemia. Plateau waves can be observed in about 25% of patients with head injury. hey develop more often in younger patients and do not affect the outcome unless they last for a very long period.
Finally, Lundberg C waves occur at a rate of 4-8 min-1 and are of limited duration and amplitude; therefore, they are probably of little pathological significance.
Secondary indices derived from intracranial pressure
The method to calculate secondary indices from ICP waveform is based on the ‘moving correlation coefficient’ method. his method was developed in order to examine the degree of correlation between two factors within a time series where the number of paired observations is large. In a moving correlation window (3-10 min), time-averaged values from each factor (6-10 s) are plotted as an x-y scattergram. Calculation of the correlation coefficient, which ranges from maximal -1 (negative correlation) to maximal +1 (positive correlation), is renewed every 6-10 s or at longer intervals. he ‘moving correlation coefficient’ may be presented and analysed as a time-dependent variable, responding to dynamic events such as ICP increase, such as that seen in a plateau wave, or arterial hypo-and hypertension .
Pressure-volume compensatory reserve
A secondary ICP index that describes the pressure-volume curve is called RAP (correlation coefficient (R) between the amplitude of the fundamental component (A) of ICP and mean ICP (P)).his index indicates the degree of correlation between AMP and mean ICP over a defined time period (usually 3-5 min). A lack of synchronization between changes in the amplitude of the fundamental component and mean ICP indicates a RAP close to 0 and indicates a good pressure-volume compensatory reserve. In this phase, the intracranial compartment can cope with the increase in volume and produces no or very little change in ICP. he AMP varies directly with mean ICP, when the RAP increases to +1. his indicates that the focus of the pressure-volume curve shifts to the steep part. At this stage, any further increase in volume may produce a rapid increase in ICP. he RAP is usually close to +1 following head injury and subsequent brain swelling. When ICP increases further, AMP decreases and the RAP value falls below 0. his may occur when the cerebral autoregulatory capacity is exhausted and the pressure-volume relationship flattens again. In this situation, the capacity of cerebral arterioles to dilate in response to a CPP decrement is exhausted (they tend to collapse passively). Normally, a negative RAP and an ICP >20 mmHg indicates a terminal cerebrovascu-lar disturbance with deterioration in pulse pressure transmission from the arterial bed to the intracranial compartment (Fig. 4.7). Following decompressive cra-niotomy, a decrease in RAP to 0 indicates recovery of the pressure-volume compensatory reserve.
Both RAP and PVI describe the pressure-volume curve, but generally there is no correlation between the two indices. Whereas PVI precisely describes the steepness of the pressure-volume curve over a defined segment, RAP indicates the focal point on the pressure-volume curve. he RAP also correlates with CBF, suggesting that severe cerebrovascular disturbance is associated with inadequate brain perfusion.
An alternative method to assess the cerebrospinal volume compensatory reserve can be done by a special system, such as the Spiegelberg brain compliance monitor. his has been developed to measure brain compliance by measuring the ICP response to a known small increase in volume by inflating and deflating the air pouch at the end of the catheter. It may act as an early warning system before decompensation, but correlation with outcome has not yet been demonstrated .
Cerebrovascular pressure reactivity
The cerebrovascular pressure reactivity index (PRx) is another ICP-derived index for assessing cerebrovas-cular reactions by observing the response of ICP to slow spontaneous fluctuations in ABP. In other words, cerebrovascular pressure reactivity reflects the ability of smooth muscle tone in the walls of cerebral arteries and arterioles to react to changes in transmural pressure. With increasing CPP, intact cerebrovascular pressure reactivity will lead to vasoconstriction and a reduction in CBV and hence ICP. Slow waves of ABP are almost always present and are of sufficient magnitude to provoke a vasomotor response. Taking advantage of this fact, cerebrovascular pressure reactivity can be determined continuously without manipulation of ABP by monitoring the response of ICP to these normal fluctuations in mean ABP. he PRx is determined by calculating the ‘moving correlation coefficient’ between time-averaged values of ICP and ABP. A positive PRx signifies a positive gradient of the regression line between the slow components of ABP and ICP, which is associated with passive behaviour of a pathological, non-reactive arterial vasculature. A negative value of PRx reflects normal reactive cerebral vessels, as ABP waves provoke inversely correlated waves of ICP.
The PRx value is an indicator of cerebral autoregu-lation, although these two terms should not be used synonymously as they describe slightly different concepts that operate over slightly different ranges of cerebrovascular physiology – vasodilation reaches its maximum at arterial pressures below the lower threshold for constant CBF. However, PRx shows a good correlation with cerebral autoregulation indices assessed by TCD or positron emission tomography. It correlates also with ICP and CPP. During ICP plateau waves, PRx consistently increases from near-zero to positive values (Fig. 4.8). Similarly, during episodes of arterial hypo- or hypertension that exceed the physiological limits of autoregulation, PRx increases to positive values.
Fig. 4.7. Example of continuous recording of cerebral compensatory reserve using RAP index in a patient after traumatic brain injury: the entire period of recording covers approximately 4 days. During the initial period, even with elevation of intracranial pressure (ICP; mean 22 mmHg), RAP is low, indicating a good compensatory reserve. Later, ICP increases to 30-40 mmHg and is unstable, and RAP increases to +1, indicating a poor compensatory reserve. In the final period, ICP remains elevated, but cerebral perfusion pressure (CPP) decreases; RAP also decreases, indicating a final derangement of regulation of cerebrovascular tone. The patient died.
Intracranial pressure and outcome following head injury
Many studies have demonstrated the detrimental role of intracranial hypertension and dysautoregulation on outcome after head injury. An averaged ICP above 25 mmHg over the whole monitoring period increases the risk of death twofold in severe head injury. Furthermore, reductions in the amplitude of slow waves and averaged RAP and PRx are strong predictors of fatal outcome. Whereas ICP and RAP only differentiate patients with fatal outcome from those who survive at 6 months, PRx distinguishes among patients with good outcome, moderate disability, severe disability and death. he outcome impact of indices such RAP and PRx suggest that good vascular reactivity is an important element of brain hom-oeostasis, enabling the brain to protect itself against uncontrollable elevations in intracranial volume and hence ICP.
Fig. 4.8. Stable intracranial pressure (ICP) disturbed by a single plateau wave, which is associated with a decrease in cerebral perfusion pressure (CPP) and an increase in cerebrovascular pressure reactivity index (PRx), with the latter continuously monitored as a variable changing in time. On the bottom graph, green denotes good reactivity and red disturbed reactivity. Periods of disturbed reactivity extend for a period past the elevation of ICP, indicating that the ischaemic insult must have affected normal vascular tone for a longer period than the wave itself. ABP, arterial blood pressure.
It is important to emphasize that mean ICP correlates strongly with both AMP and RAP. In a multi-variate analysis of outcome, increased ICP and positive PRx are independent variables associated with a worse outcome. As mean CPP is an actively controlled variable in CPP-oriented protocols, it has lost its predictive power for outcome in more contemporary statistical evaluations. However, there is evidence that systemic hypotension independently increases morbidity and mortality. It is currently proposed that the critical threshold for CPP is between 50 and 60 mmHg, so management nowadays aims to keep CPP above 65-70 mmHg with judicious use of vasopressors .