|
|
With introduction of the ventricular
puncture and catheter insertion for central ICP monitoring by Adson and
Lillie in 1927, a reliable assessment was possible without the risk of
herniation.(1) Later, Guillaume and Janny employed a mechanic-electrical
pressure transducer for this measurement.
More public interest was evoked
when Lundberg published his fundamental work on ICP analysis in the 1960’s
and almost at the same time Langfitt presented his concept of cerebrospinal
compliance.(56,60,68) The breakthrough for the clinical application
of ICP monitoring brought the development of reliable and stable pressure
sensors.(29)
2. Principles of ICP monitoring
Pressure (P)/ ICP describes
the force (F) acting on a defined area (A):
One must distinguish between
the absolute pressure and the physiological pressure used
in medicine, which is the differential pressure between the measured
pressure and the atmospheric pressure. Only in fluids pressure is exactly
defined. In inhomogeneous tissues of unfixed shape a high distribution
of various force vectors occurs. Therefore, pressure is defined
as the force vector which acts homogeneously vertical on a defined measuring
area.(30) Taking this into account, the measured data obtained in various
locations is of limited value and a comparison might be difficult.
The intracranial space is divided into three large compartments
by the falx cerebri and the tentorium cerebelli. The supratentorial space
is further divided by the sphenoid wings. These structures are quite resistant
to pressure. Therefore, the measured pressure can differ supra- and infratentorially
as well as between both cerebral hemispheres. Also, there can occur a
difference in intraparenchymal measuring in relation to location and position
of the sensors to the main vector of the acting force. Clinically important
syndromes are falcine and tentorial herniation of the brain parenchyma
across the pressure gradient. The latter can also occur in the opposite
direction with infratentorial lesions (inverse tentorial herniation).
3. Definition of ICP and normal values
The ICP is defined as
the fluid pressure of the CSF that is measured at the level of the foramen
of Monro; the assessment must be performed in supine position for comparison
with “normal” references. The normal values are age dependent ( Table
1).
|
Age
Group
|
ICP
normal
|
|
Infant
|
<7.5
mm Hg
|
|
Child
|
<10
mm Hg
|
|
Adult
|
<15
mm Hg
|
4. Pressure sensors used
for ICP monitoring
4.1.
External pressure transducer:
A typical application for an external pressure transducer is the adaptation
of a fluid pressure sensor to a ventricular or spinal catheter. The advantage
of this method is its inexpensiveness, but it also possesses several pitfalls.
Very important is the exact selection of the neutral level (level of the
foramen of Monro). This projects at the level of the external acoustic
meatus in supine patient and at the level of the root of the nose in lateral
position. Therefore, with each positional change a new calibration is
required if the pressure sensor is not head-fixed.
Another problem is present when silicon catheters are used, which remarkably
reduce or distort the transmitted wave amplitude by dumping or by resonance.
There may be records of inadequate low ICP values with occlusion of the
catheter with maintained pulse waves due to the compressible silicon tube.
Therefore, with this method, a short transmitting distance and the use
of pressure stable (incompressible) tubes are important.
4.2. Fibreoptic systems: Common applications of the fibreoptic
system are intraparenchymal and intraventricular pressure monitoring.
The mode of assessment consists of reflection of a light ray at a pressure
sensitive membrane within the catheter tip. The reflected light is transferred
via light-wave cable to an external monitor, where the pressure is calculated.(16)
This procedure was introduced by Gaab and Dietrich (1975) and is commercially
available since 1985 (Camino Laboratories); it has found great acceptance.(22)
Most studies dealing with this method report high accuracy with a low
complication rate.(32,73,83,87) Others observed a progressive drifting
of the measured values with measuring time or a frequent fibreglass breaking.(16,96)
The largest published study demonstrated plausible data in only 85% of
cases and dysfunction in 11% due to cable breaking or catheter dislocation.(70)
4.3. Piezoresistive systems: This method is based on the
phenomenon that some crystals polarise under pressure, resulting in a
pressure dependent change of the electric conduction capability. The employment
of piezoresistive pressure transducers enabled a far-reaching miniaturisation
of the sensors.
Sensors of this type possess a very small drift of the measured values.(76)
Most sensors can easily be hooked up to the usual patient ICU monitors.
5. Modes of ICP monitoring
5.1. Measuring localisations:
As reported by Unterberg 1995, exclusively epidural, intraventricular
and intraparenchymal pressure monitoring systems are used in Germany.
Very rarely, there is reference to the application of fontanometry in
the literature.
5.1.1. Fontanometry: In infants, there is the possibility
of non-invasive pressure measurement through the open frontal fontanel
with the "aplanation principle". (31) For this, it is important to take
into account that the elastic membrane develops during the 23rd week and
the subcutaneous fat layer within the 28th week of pregnancy.(89) The
main problem is the inability to measure an exact pressure at the beginning.
The procedure is suited for non-invasive screening and trend evaluation,
but not for exact evaluation of the ICP.(50)
5.1.2. Epidural pressure monitoring: As early as 1948, Riechert
(cited by Gaab) suggested this method of epidural pressure monitoring.
It was first performed by Gerlach in 1952.(30,33) The method currently
used is based on the fundamental experimental work of Gaab.(30) In his
work, he for the first time used an epidural pressure transducer with
correct implantation technique (sufficient loosening of the dura mater
of 1 cm around the sensor and coplanar implantation of the probe) to exactly
measure the ICP. For the exact evaluation of the ICP, the sensor is epidurally
implanted through a burr hole or a present craniotomy after loosening
the dura. Main mistakes are the insufficient loosening of the dura with
consecutive assessment of false high pressures due to dural tension and
the disregard of the coplanar measuring principle (roughness or scarring
of the dura, tilted implantation of the sensor).
The advantage of this method lies in the integrity of the cerebrospinal
space; this allows a consecutive reliable evaluation of the cerebrospinal
compliance and of the CSF outflow resistance (Fig. 1).
 |
| Figure 1 - Epidural ICP sensor (type Mammendorf Accurate plus) |
5.1.4. Parenchymal measuring: This relatively new method has come into increasingly wide use because of its simple implantation technique. A piezoresistive or fibreoptic sensor is implanted through a frontal burr hole or a screw. Signal transduction occurs either simply via the pressure input of the usual patient monitors, or by additional interface boxes.
For the interpretation of the data, it is important to note that vectors and not a “pressure” are measured, and that due to the compartmentalisation of the intracranial space an uneven pressure distribution exists. Nevertheless, there is a good correlation between ventricular and parenchymal “pressure” values.(32)
Infection incidence and post-operative bleeding complications are between 0-5%. This complication rate is less than with ventricular pressure monitoring (Fig. 2).(32,78)
 |
| Figure 2 - Intraparenchymal ICP probe |
Advantages of this method are low costs combined with the possibility of permanent or intermittent CSF drainage for ICP reduction and repeated calibration. Disadvantages are the increasing infection rates with time progression, the risk of the ventricular puncture itself, susceptibility to artefacts (distortion, blocking of the catheter, dislocation of the systems; mean rate of dysfunction 3%), and mismeasurement (dislocation, hydrostatic measuring mistakes, pressure suppression by air bubbles, resonance phenomena, slit ventricles with loss of fluid adaptation with massive increased ICP and positioning of the patients).(39) Particularly with small or shifted ventricles, puncture of the lateral ventricles can be difficult.(30,39,58) The rate of missed punctures is around 6%.(39) We ourselves found, in a prospective investigation of 121 patients with ventricular pressure monitoring (traumatic head injury and spontaneous haemorrhage) only in 2 cases a manifest meningitis, and one clinically relevant haemorrhage but a relatively high number of technical dysfunctions due to catheter dislocation (n=7) or occlusion (n=6). There was a particularly high complication rate with regard to bleeding; intracerebral haemorrhages associated with percutaneous needle puncture occurred in up to 40% of patients.(37) The infection rate sharply increases after 5 days, and is especially high (30%) in haemorrhagic CSF (eg. SAH, IVH). The use of a pressure transducer integrated into the catheter tip gives a better measuring quality and reduces the incidence of false low data due to catheter occlusion and dislocation, but also leads to higher costs (Fig. 3).
 |
| Figure 3 - Venticular catheter with integrated ICP sensor (Medtronic) |
6. Data registration
The base for any analysis consists in the exact and continuous registration of the measured data. A minimum requirement is a paper based registration with a paper speed of 0.5 cm/min. The usual minute registration of the mean ICP values by ICU monitors is only sufficient for the evaluation of the cerebral perfusion pressure and leads to an unfortunate reduction of the data information. More appropriate appears the application of PC-based registration and analysis systems. Therefore, NEUROLAB* was developed by us, which allows a continuous evaluation of the ICP and CPP, as well as pulse, breathing and B-waves (Fig. 4).
|
| Figure 4 - On line ICP analysis with the NEUROLAB monitoring system. |
The compliance (C) describes the volume-dependent increase of the pressure in the craniospinal space and can be used for the assessment of the cerebrospinal space reserves. The inversion is termed elastance (E).
Figure 6 - Bolus injection test – clinical examples
Nevertheless, for the usual intensive care of patients the formula by
Marmarou is valid; it gives an estimation of the intracranial space reserve
of the actual compliance and elastance of the intracranial space.
8. CSF circulation parameters
For the description of the CSF circulation, the CSF formation rate (Iform),
the CSF outflow resistance (Rout), and conduction (Cout) (which is the
inverse of resistance) are used. For assessing these parameters several
methods exist:
8.1. CSF outflow resistance: The calculation of the
CSF outflow resistance allows an accurate assessment of resorptional dysfunctions.
This investigation depends on several conditions: Any spinal or ventricular
puncture in the time period immediately before the examination should
be avoided (about two weeks), nor must there be any large bone defect
in the skull. Due to the risk of herniation, craniospinal pressure gradients
are a contraindication for this procedure.
The CSF resorption can be assessed with isotope cisternography, the bolus
test or the constant volume test.(13,14,30,35,38,47,51,64,74,91)
The latter has first been described by Katzman in 1970.(51) It is a time
consuming test and a burden for the patients. Subsequent developments
include the constant pressure infusion test by Friden and Ekstedt and
the constant volume infusion test by Gjerris modified by Gaab (Fig. 7).
(11,27,30)
Figure 7 - Constant volume test – clinical examples
The PVI was aimed to shorten the examination time (Marmarou(64) 1978).
|
Author/Year |
Physiologic Cout in ml/min/mm Hg |
Method |
|
Ekstedt, 1978 (26) |
0.11 |
Constant pressure test |
|
Sklar, 1980 (86) |
0.13 |
Constant pressure test |
|
Tans &Poortvliet, 1984 (90) |
>0.11 |
Constant volume test |
|
Gjerris (35) |
>0.08 |
Constant volume test |
|
Albeck, 1991(2) |
0.11 |
Constant volume test |
8.2. CSF formation rate: This parameter for the description of the CSF circulation is difficult to assess in clinical practice. The procedure described by Marmarou suffers from similar restrictions as the method for CSF resorption assessment with the PVI.(64)
Table 3 - Normal value for CSF formation rate
|
Author
|
 |
|
Ekstedt
(25)
|
0.4
ml/min
|
|
Drake
(23)
|
0.15
ml/min (baby, 1 month)
0.25 ml/min (child, 8 years) |
These signals are partly generated extracranially like the pulse and breathing waves. Others like the B- and A- (plateau) - waves are generated within the intracranial vascular system.
9.1. Pulse wave: Under physiological conditions, pulse wave frequency correlated waves occur with an amplitude of 1 to 4 mm Hg or 10-30% of the mean ICP, respectively. A more detailed analysis of the pulse amplitude shows several wave peaks, which are consecutively termed P1-P5. P1-P3 peaks occur regularly and mainly correspond to pulse waves of 2nd class. Cardoso demonstrated a correlation between the P2- wave and the cerebral compliance.(15) Nevertheless, these waves underly multifactorially influences and their analysis cannot provide reliable information on compliance.
With decreasing compliance the pulse wave amplitude increases. This precedes the ICP increase. With hyperventilation and presence of vasospasm, the pulse wave decreases (Fig. 8).
Figure 8 - ICP pulse walve with demonstration of the P1-3 - maxima.
Figure 9 - ICP breathing curve.
9.3.1. Plateau-waves: Typical for plateau-waves (synonym: A-waves) are characterised by a quick increase in ICP from a previously slowly increasing mean pressure. Later, the ICP reaches a long persisting plateau of above 40 mm Hg or even more, with a duration between 5 to 30 min. Finally, there is a sudden steep decline of the ICP, much faster than its increase, often reaching a level below the starting pressure. In a decompensating intracranial space-occupying lesion, this decline might be skipped with series of permanently increasing A-waves occurring from higher and higher pressure levels, or there is a sudden terminal ICP-increase within one plateau-wave. Lundberg assumed an acute increase of the intracranial blood volume as the cause of the plateau-wave as early as 1969.(80) This was confirmed by consecutive studies.(45,65,81) First, Rosner introduced the theory that the plateau-wave represents an active reaction in the setting of a far-reaching intact autoregulation with increased ICP and limited CPP.(81) Hayashi demonstrated the induction of plateau-waves in dogs by stimulation of the medulla oblongata.(42-44) Our own results show a positive correlation between the occurrence of plateau-waves and the clinical outcome as an indicator for an intact regulation of the cerebral perfusion.(40)
Also discussed is the compression of parasinal bridging veins as a cause of the A-wave (Fig. 10).(3)
Figure 10 - A (Plateau-) wave after severe head injury.
Auer demonstrated a correlation of calibre changes of the pial vessels and B-waves.(4) Mautner-Hubert showed a simultaneous occurence of B-waves with changes in the flow velocity within the MCA in healthy volunteers.(66) Other authors supported these findings.(24,72) With the introduction of a continuous automated ICP-analysis, we found a continuous B-activity with small amplitudes also in healthy people.(69) In comparison to this, the lower assessing limit for the paper-supported registration lies at 1.7 mm Hg (Fig. 11).
Figure 11a -Sinus-like B-waves
Figure 11b - Ramp-like B-waves
9.4. Artifacts: Main cause for artifacts are positional changes of the patient (spontaneously or with nursing steps). Particularly susceptible is the external pressure transducer. Others causes are coughing, endotracheal suctioning, changing of the ventilation mode (hyperventilation), etc. Artifacts mostly show typical brief steep pressure increases and decreases. In diagnostics of patients with hydrocephalus, it is recommended to accept only almost artifact free registrations for the calculation of the mean pressure and the wave activity, as typically present during night sleep (Fig. 12).
Figure 12 - Typical artifacts of a restless patient.
With uncritical ICP values, unilateral internal jugular vein compression can be employed. Subsequent ICP increase can be used for sensor control.
10. Importance of ICP monitoring
With respect to complications, it is important to take into account that for head injured patients ICP monitoring supplies the decisive parameters for the subsequent treatment decisions. Particularly, these patients benefit from aggressive ICP treatment.(25)
The lowest complication rate is reported with epidural ICP monitoring, the highest with the intraventricular ICP catheter. With the latter, mainly infections are observed, although in the literature there is often no differentiation between positive bacteriological evidence and clinically manifest infection (Table 4).
| Table 4 - Complication rate of various ICP monitoring methods.
|
|
Author/Reference
|
Intraventricular
|
Intraparenchymal
|
Epidural
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
 |
