The physiological basis of intracranial pressure change with progressive epidural brain compression

An experimental evaluation in cats

Full access

✓ Sequential cerebrospinal fluid (CSF) pressure-volume studies were carried out in seven cats during the expansion at a constant rate of an epidural balloon. The same studies were performed in three control cats. Beginning after 20 minutes of inflation and continuing to the point of pupillary dilatation there was a progressive increase in the pressure-volume index (volume required to change intracranial pressure (ICP) by tenfold). During the course of balloon inflation, there was also a progressive increase in CSF elastance (instantaneous ICP change per unit change in CSF volume). At the point of pupillary dilatation there was a marked, abrupt increase in the pressure-volume index and an equally dramatic decrease in CSF elastance. The CSF outflow resistance increased to a variable extent during balloon inflation. The plot of the CSF pressure versus balloon volume (the mass lesion pressure-volume curve) was of the classical configuration with an initial relatively flat segment and a final steep segment. A hypothesis is presented that interprets the shape of the mass lesion pressure-volume curve in terms of changes occurring in the elastic properties of the tissues surrounding the CSF space and the volume of the CSF space. It is proposed that this hypothesis will explain most of the commonly observed variations in CSF pressure. Confusion regarding the ICP-volume relationships has arisen because of lack of specificity regarding which anatomical spaces are being perturbed.

Abstract

✓ Sequential cerebrospinal fluid (CSF) pressure-volume studies were carried out in seven cats during the expansion at a constant rate of an epidural balloon. The same studies were performed in three control cats. Beginning after 20 minutes of inflation and continuing to the point of pupillary dilatation there was a progressive increase in the pressure-volume index (volume required to change intracranial pressure (ICP) by tenfold). During the course of balloon inflation, there was also a progressive increase in CSF elastance (instantaneous ICP change per unit change in CSF volume). At the point of pupillary dilatation there was a marked, abrupt increase in the pressure-volume index and an equally dramatic decrease in CSF elastance. The CSF outflow resistance increased to a variable extent during balloon inflation. The plot of the CSF pressure versus balloon volume (the mass lesion pressure-volume curve) was of the classical configuration with an initial relatively flat segment and a final steep segment. A hypothesis is presented that interprets the shape of the mass lesion pressure-volume curve in terms of changes occurring in the elastic properties of the tissues surrounding the CSF space and the volume of the CSF space. It is proposed that this hypothesis will explain most of the commonly observed variations in CSF pressure. Confusion regarding the ICP-volume relationships has arisen because of lack of specificity regarding which anatomical spaces are being perturbed.

A central question in neurosurgery is: What does an elevation of intracranial pressure really mean? With the advent of continuous monitoring of intracranial pressure (ICP) in patients, many paradoxes have evolved. In patients with similar levels of ICP, small volumes of fluid introduced into the lateral ventricles may produce quite different changes in ICP.17 Patients with the same elevation of ICP may vary dramatically in degree of neurological impairment.20 There does not seem to be a direct relationship between brain shift (angiographically) and ICP in patients with mass lesions.6,19 An explanation for such seemingly paradoxical situations is important in the management of intracranial hypertension. A comprehensive theory relating ventricular fluid pressure to fundamental properties of the brain and its surroundings is needed.

In this report we relate the pressure-volume index,12,14 cerebrospinal fluid (CSF) elastance,11,12,14 CSF outflow resistance, and the equilibrium volume of the CSF space to the ventricular fluid pressure seen as a function of the volume of a progressively expanding epidural balloon in lightly anesthetized, ventilated cats. Based on our data, a hypothesis is presented that relates the level of ICP under any circumstances to bulk flow CSF dynamics and the elastic properties of the CSF space. We regard the hypothesis presented here as a first step toward a comprehensive theory relating alterations in ventricular fluid pressure to fundamental properties of the brain and its surroundings.

Mathematical Model

Marmarou12 has investigated the elastic properties of the intracranial and intraspinal contents in the cat under normal physiological conditions. In this work, Marmarou has studied the changes in ICP induced by the injection of varying, small volumes (ΔV) of fluid into the cisterna magna of the cat, and has demonstrated the following relationship:

mu1
where: Po = the baseline ICP, and Pp = the peak ICP in response to an intracisternal injection of fluid volume ΔV. The constant, k, in Equation 1 is referred to as the pressure-volume index (PVI). The PVI determines the logarithmic base (101/PVI) that, under normal physiological conditions, defines the shape of the CSF pressure-volume curve (CSF pressure plotted as a function of CSF volume). In physical terms the PVI is the volume required to raise the ICP by tenfold.

Equation 1 has been demonstrated to describe correctly the CSF pressure-volume interrelationships in seven hydrocephalic infants.22 In these seven patients the fluid volume was injected into the lateral ventricles. Furthermore, the PVI of brain-injured patients has been employed as an index of the status of intracranial volumetric compensation.13

Equation 1 may be generalized as follows so that it applies to any disturbance of ICP:

mu2
where tPeq = the equilibrium pressure of the CSF space at time t, tPcsf = the CSF pressure at time t, tVT = the total volume of the CSF space at time t, tVeq = the equilibrium volume of the CSF space at time t, and tΔVL = the volume loading of the CSF space at time t.

The equilibrium pressure of the CSF space has been defined as follows:

mu3
where tFcsf = the rate of CSF formation at time t, Pv = the opening pressure of the CSF drainage system, and tRo = CSF outflow resistance at time t.

Utilizing Equation 3, we can rewrite Equation 2 as:

mu4
so that CSF pressure at a given time is expressed as a function of CSF production, CSF outflow resistance, the opening pressure of the CSF drainage system, any deviation of CSF volume from equilibrium, and the pressure-volume index.

The elastance of the CSF space (Ecsf) may be thought of as the change in ICP occurring with a unit change in CSF volume at a particular point on the CSF volume-pressure curve. The volume-pressure response (VPR) (ΔPcsf/ΔVcsf for very small injection volumes9,17) is an estimate of CSF elastance. The CSF compliance (Ccsf) is the inverse of Ecsf, and may be thought of as the change in CSF volume required to produce a unit change in ICP at a given point on the volume-pressure curve for the CSF space. The relationship between ICP, PVI, Ecsf, Ccsf, and VPR may be defined from Equation 2 as follows:

mu5
and
mu6

The decay of the Pcsf following a single, rapid fluid injection into the CSF space can be thought of as a reflection of the CSF outflow resistance (Ro). Figure 1 illustrates the pattern of decay of Pcsf following a rapid intraventricular fluid injection. The equation relating Ro to the decay of Pcsf after a single intraventricular fluid injection is as follows:

mu7
where Pt = the CSF pressure at time t after injection (Fig. 1), and Po and Pp are defined as in Equation 1.

Fig. 1.
Fig. 1.

Polygraph recording illustrating the response of the ICP, BP, and EEG to a bolus intraventricular injection 0.3 ml of normal saline. The peak ICP (Pp) was taken as the first trough following intraventricular saline injection (Pp = 35.5 mm Hg). The baseline ICP (Po) was taken as the first diastolic ICP, which immediately preceded injection and was in the same part of the ICP respiratory cycle as Pp, (Po = 10.0 mm Hg). Pt was considered to be the 1-minute, postinjection, diastolic ICP that occurred in the same part of the ICP respiratory cycle as Pp (Pt = 20.5). Mean ICP before injection was 11 mm Hg. From Equation 1, Equation 5, and Equation 7, PVI = 0.55 ml, Ecsf = 46 mm Hg/ml, and Ro = 125 mm Hg/ml/min. Note that there is no change in the EEG or in the blood pressure with the rapid intraventricular injection of 0.3 ml of saline.

To derive Equation 7,12,14 the CSF elastance (dP/dV), Equation 6, is multiplied by the rate of change of CSF volume (dV/dt = Fcsf−Acsf where Acsf = the rate of CSF absorption). This multiplication yields a first order, nonlinear differential equation expressing the rate of change of CSF pressure (dP/dt) as a function of ICP, Ro, PVI, and Peq. The solution of this differential equation may be solved for Ro yielding Equation 7.12

If the equations described by Marmarou and his associates12–14,22 relating CSF pressure and CSF volume remain valid under pathological conditions, the changes in ICP under these pathological circumstances can be viewed as a consequence of changes in the following parameters: CSF outflow resistance (Ro), formation of CSF (Fcsf), drainage pressure of the CSF system (Pv), the pressure-volume index (PVI), the CSF volume (Vcsf), and the equilibrium volume of the CSF space (eqVcsf). In this report the changes in ICP as a consequence of brain compression are related to sequential changes in the PVI, Veq, Ro, and Ecsf.

Materials and Methods
Calculation of CSF Pressure-Volume Data

To determine PVI (Equation 1), Ro (Equation 7), and Ecsf (Equation 6), intraventricular injections of saline were made through a stopcock interposed in the tubing connecting a ventricular needle to a Statham P23 pressure transducer.* This stopcock was rigidly fixed to a stereotaxic frame so that any movement artifact associated with injection would be eliminated. All of the rapid intraventricular injections were made at a rate of approximately 0.1 ml/sec.

Figure 1 shows a polygraph recording demonstrating the ICP changes at the time of one of the rapid intraventricular saline injections made in these experiments. The peak pressure, Pp, (Equation 1) in response to rapid intraventricular injection of saline was taken to be the first clearly identifiable diastolic CSF pressure trough after the saline injection. The baseline pressure, Po (Equation 1), was taken as the diastolic CSF pressure located in the same part of the ICP respiratory cycle as Pp. The diastolic CSF pressure at 1 minute after intraventricular saline injection was taken as Pt (Equation 6) for calculation of Ro. The Pt was chosen so that it occurred at the same point in the ICP respiratory cycle as Po and Pp. When there were fluctuations in the arterial blood pressure, Po and Pt were picked so that they not only occurred in the same part of the ICP respiratory cycle as Pp but also so that the arterial blood pressure associated with Po and Pt would coincide as nearly as possible with the blood pressure level associated with Pp. For the baseline periods in Groups 1 and 2 (see below) Ecsf (Equation 6) was calculated for each baseline value of the PVI using the mean equilibrium ICP before balloon inflation in Group 1 or before the mock-inflation period in Group 2. During balloon inflation or mock inflation (see below), Ecsf was calculated from the mean ICP immediately before each intraventricular saline injection.

Experimental Groups

Ten cats of either sex, weighing from 2.6 to 5.3 kg, were divided into two groups.

Group 1 (Balloon-Inflation Group). In each of these seven cats baseline calculations of PVI, Ro, and Ecsf were made on the basis of five to six rapid injections of 0.2, 0.3, or 0.4 ml of saline into the lateral ventricle. After each injection, the pressure was allowed to return to near the CSF equilibrium pressure. Upon completion of the final baseline intraventricular injections, CSF pressure was allowed to stabilize at the equilibrium value. Slow constant rate (mean inflation rate = 0.023 ± 0.001 ml/min (SD)) inflation of a left occipital balloon was then begun. During balloon inflation sequential values for PVI, Ro, and Ecsf were calculated from sequential, rapid intraventricular injections of 0.2, 0.3, or 0.4 ml of saline at 10-minute intervals. Mean CSF pressure was noted before each intraventricular saline injection. The 10-minute intervals between intraventricular saline injections during epidural balloon inflation proved sufficient time for the ICP to reach a stable level. Experiments were terminated when the cats had fixed, dilated pupils and a flat electroencephalogram (EEG).

Group 2 (Control Group). In these three animals the epidural balloon was placed but not inflated. The same sequence of measurements of PVI, Ro, and Ecsf was carried out.

Preparation of Animals

Each animal was anesthetized intravenously with a solution of 2 to 4 ml of 1% sodium Brevital in normal saline titrated to the point at which the blink reflex was just suppressed. This level of anesthesia was maintained with intermittent intravenous doses of 1% Brevital (1 to 2 ml) until all operative manipulations were completed. Anesthesia was supplemented by infiltration of 1% Xylocaine into all surgical sites and into the external auditory canals.

A cannula for medication was placed in the inferior vena cava via the femoral vein. The femoral artery was cannulated to monitor blood pressure. Tracheostomy was performed and all of the animals were paralyzed with a continuous intravenous infusion of 0.2% succinylcholine chloride. The animals were maintained on controlled ventilation with air supplemented with O2 so that arterial blood gases were maintained within the following limits: pCO2 = 33 ± 3 mm Hg, pO2 > 100 mm Hg, and pH = 7.35 ± 0.05. Body temperature was regulated at 37° ± 0.5° C by a heating blanket. After placing the animal in a stereotaxic frame, the scalp and temporal muscles were reflected and appropriate holes were drilled in the skull to accommodate: 1) two ventricular needles in the left and right frontal areas; 2) two paracentral electroencephalographic (EEG) screws; and 3) an epidural balloon in the left occipital area. Puncture of both lateral ventricles was accomplished stereotaxically with the continuous infusion method used to identify ependymal puncture. The EEG screws were placed by counting turns according to the thickness of bone so that the tips of the screws were just in contact with the dura. The epidural balloon was constructed by attaching a finger cot to the end of a plastic “Y” connector. The balloon assembly was checked for leaks by inflation to 5 cc before implantation and mounted so that its surface was flush with the dura.

After ICP and EEG recordings had been established and the epidural balloon correctly positioned, the whole assembly was fixed in place with dental acrylic applied slowly in layers with copious irrigation to dissipate heat. One of the ventricular needles was connected to a separate stopcock secured to the stereotaxic frame for injection of small volumes of fluid (Fig. 1).

At the conclusion of each experiment, the brains (including the upper cervical spinal cord) were removed and suspended for at least 24 hours in 10% buffered formalin. The cerebral hemispheres were sectioned coronally at 5 mm intervals and the brain stem and spinal cords cut sagittally.

Data Collection, Calibration, and Analysis

Blood pressure and ventricular fluid pressure were measured by Statham P23 strain gauge transducers calibrated against a water standard and displayed on the chart paper of an 8-channel Gould physiologic recorder. Calibration of the blood pressure and CSF pressure transducers was repeated at the end of each experiment and the amount of transducer drift was recorded and was less than ± 1 mm Hg for any animal. The single channel of EEG was monitored via a Gould EEG Coupler calibrated against an internal standard of 50 µV. The EEG was recorded at a sensitivity of 10 to 20 µV/division with the low-frequency filter at 0.1 Hz and the high-frequency filter at 60 Hz. Arterial blood gases were analyzed on an IL blood gas analyzer, calibrated against 5% and 10% CO2, 12% O2, and standard buffers of pH 6.84 and 7.34.

For each animal in Group 1 a plot of ICP versus balloon volume was constructed (Fig. 2). The inflection point (Table 1) of the ICP versus epidural balloon volume curve was determined by an extrapolation to the curve of the intersection point of two lines representing the slopes of the two major parts of the curve.

Fig. 2.
Fig. 2.

Plot of the intracranial mass lesion pressure-volume curves (ICP versus balloon volume) for each the seven Group 1 animals. * = the point at which the left pupil dilates. ↑ = the break point of the curve. These curves are of the classic configuration having an initial relatively flat segment and a second relatively steep segment. Pupillary dilatation is a late event and significantly large volumes of saline have accumulated in the epidural balloon by the time the break point is reached.

TABLE 1

Relationship between pupillary dilatation and the inflection point

Cat No.BreakpointPupil Dilates%
t(min)BV(ml)ICP(mm Hg)t(min)BV(ml)ICP(mm Hg)  
1661.521962.25769
2481.124801.89260
3561.325952.28559
4591.426852.06269
5671.530952.27471
6481.116831.96258
7531.227852.05862
mean ± SD57 ± 81.3 ± 0.224 ± 588 ± 72.0 ± 0.270 ± 1464 ± 5

* t = time; BV = balloon volume; ICP = intracranial pressure.

% = The % of BV at the time of pupillary dilatation achieved at the time of the breakpoint.

Based on the four to six rapid intraventricular saline injections made into each animal before balloon inflation, means and standard deviations for the PVI, Ro, and Ecsf were determined to represent baseline conditions for each animal. Changes in PVI, Ro, and Ecsf noted during balloon inflation have been displayed graphically (Figs. 3 right, 4 right, and 6 right) so that the sequential alterations in each variable are related to the means and standard deviations noted for that variable during the preinflation, baseline period. These plots representing the sequential changes in PVI, Ro, and Eo were constructed by calculating the number of standard deviations by which a particular variable (PVI, Ro, Eo) deviated from the baseline mean for that variable as a function of the time of balloon inflation. Further analysis of the changes in Ro during balloon inflation in Group 1 was made by dividing the Group 1 animals according to the general trend noted in Ro (Table 5).

Fig. 3.
Fig. 3.

Alterations in the pressure-volume index (PVI), Left: Plot of sequential PVI's in each control animal (Group 2). Dashed line indicates the baseline standard deviation. Heavy line indicates the baseline mean PVI. Note that during the mock-inflation period the PVI does not exceed the baseline mean by greater than one standard deviation. Right: Plot indicating sequential changes in the pressure-volume index (PVI) seen in each Group 1 animal during the balloon-inflation period. The number of standard deviations (baseline sample) by which a given measurement of PVI exceeds the baseline mean (ΔPVI/SD) is plotted for each animal as a function of balloon-inflation time. Dashed lines indicate the maximum and minimum values of ΔPVI/SD in control groups. * = the point at which the left pupil dilates. ↑ = the inflection point of the intracranial mass lesion pressure-volume curve. After the balloon has been inflated for 20 minutes there is a progressive increase in PVI with increasing balloon volume. Large increases in PVI occur at the time of pupillary dilatation.

Fig. 4.
Fig. 4.

Alterations in elastance of the CSF space (Ecsf). Left: Plot of sequential Ecsf values in each control animal. Dashed lines indicate the baseline standard deviation. Heavy lines indicate the baseline mean Ecsf. During the mock-inflation period, Ecsf exceeds the baseline mean Ecsf by a maximum of 5.5 standard deviations (max ΔE/SD = 5.5). Right: Plot indicating sequential changes in Ecsf seen in each Group 1 animal during balloon inflation. The number of standard deviations (baseline sample) by which a given measurement of Ecsf exceeds the baseline mean (ΔEcsf/SD) is plotted for each animal as a function of balloon-inflation time. Dashed lines indicate the maximum value of ΔEcsf/SD seen in the control group. * = the point at which the left pupil dilates. There is a progressive increase in Ecsf with increasing balloon volume. There is a marked decline in Ecsf associated with dilatation of the left pupil.

Fig. 6.
Fig. 6.

Alterations in CSF outflow resistance (Ro). Left: Plot of sequential Ro values in each control animal. Dashed lines indicate the baseline standard deviation. Heavy lines indicate the baseline mean Ro. During the mock-inflation period, Ecsf exceeds the baseline mean Ecsf by a maximum of 8.5 standard deviations (max ΔE/SD = 8.5). Right: Plot indicating sequential changes in Ro seen in each Group 1 animal during the balloon-inflation period. The number of standard deviations (baseline sample) by which a given measurement of Ro exceeds the mean baseline Ro (ΔRo/SD) is plotted for each animal as a function of balloon-inflation time. Dashed line indicates the maximum value of ΔE/SD seen in the control group. Notice that only three Ro values from the balloon-inflation period fall below the baseline mean value.

TABLE 5

Changes in CSF outflow resistance (Ro) secondary to progressive epidural balloon inflation

Cat No.Baseline Mean Ro ± SD Ro Response*tα = 0.05
175 ± 41C70
299 ± 11C70
359 ± 6B80
4118 ± 26C
573 ± 4A40
666 ± 16A60
761 ± 4B80
mean ± SD67 ± 15

A = progressive increase in Ro with epidural balloon inflation; B = elevated Ro but no progressive changes with epidural balloon inflation; C = final moderate-to-large elevation in Ro with epidural balloon inflation.

tα 0.05 = epidural balloon inflation time required for Ro to differ from baseline Ro at significance level = 0.05.

In order to test whether a given value of PVI, Ro, or Ecsf from the balloon inflation period in Group 1 differed from the pre-inflation determinations, the method of Dixon4 was employed to calculate significance limits (Tables 3 and 5) for the value in question relative to the baseline values of that particular variable. This particular statistical method provides a test of whether a given value of PVI, Ro, or Ecsf from the balloon inflation period belongs to the same population of measurements as values of PVI, Ro, and Ecsf calculated during the baseline period. This statistical method is generally used to evaluate extreme values of a given sample, but its application is valid under the conditions of our experiments.

TABLE 3

Changes in pressure-volume index (PVI) secondary to progressive epidural balloon inflation*

Cat No.Pressure-Volume Index (ml)Significance Level
Baseline mean ± SDPPDAPDΔPVItα = 0.05 (min)tα = 0.01 (min) 
10.69 ± 0.021.43.42.04050
20.52 ± 0.022.05.33.35070
30.76 ± 0.021.32.61.37070
40.66 ± 0.031.02.91.95060
50.64 ± 0.021.26070
60.60 ± 0.051.56060
70.63 ± 0.030.771.60.838080
mean ± SD1.3 ± 0.43.2 ± 1.41.9 ± 0.959 ± 1366 ± 10

PPD = measurement immediately before pupillary dilatation; APD = measurement immediately after pupillary dilatation; ΔPVI = PVI immediately after pupillary dilatation minus PVI immediately before pupillary dilatation; tα0.05 = balloon inflation time required for PVI to differ from baseline at a significance level of 0.05; and tα0.01 = balloon inflation time required for PVI to differ from baseline at a significance level of 0.01.

Changes in PVI, Eo, and Ro in Group 2 (control) during the mock-inflation period were compared to baseline values by plotting the actual value of the variables as a function of observation number during the baseline period, and as a function of time during the mock-inflation period (Figs. 3 left, 4 left, and 6 left). For comparison with Group 1 data the maximum value of ΔX/SD (where ΔX = the deviation of PVI, Ro, or Eo from the baseline mean value, and SD = the standard deviation of the baseline measurements for that particular variable) was calculated in Group 2 for the mock-inflation period (Figs. 3 left, 4 left, and 6 left).

Results
Mass Lesion Pressure-Volume Curve

Figure 2 shows the plot of the mass lesion pressure-volume curve (ICP = F(VM) where VM = mass volume) for each of the seven animals in Group 1. Each of the mass lesion pressure-volume curves demonstrated the classic configuration of an initial relatively flat segment, a break point, and a final relatively steep segment. The break points of the ICP versus mass volume plots occurred at a mean duration of balloon inflation of 57 ± 8 min (SD) and a mean balloon volume of 1.3 ± 0.2 ml (SD) (Table 1). The mean ICP at the break point was 24 ± 5 mm Hg (SD). The ICP at the break point of the mass lesion pressure-volume curve was never greater than 30 mm Hg. Mean ICP at the time of pupillary dilatation was 70 ± 14 mm Hg (SD) (Table 1). In Group 2 during the mock-inflation period ICP did not vary more than a few mm Hg.

The percentage of balloon volume at the time when the break point of the mass lesion pressure-volume curve had been reached, relative to the total balloon volume at the time of pupillary dilatation, averaged 64% ± 5% (SD) (Table 1). At the break point, epidural balloon volume had reached about 6% of estimated total intracranial volume; by the time of pupillary dilatation epidural balloon volume had reached about 10% of estimated total intracranial volume.

Pressure-Volume Index

In Group 2 (control) the mean baseline PVI's varied from 0.50 ± 0.03 to 0.60 ± 0.03 ml (SD) (Table 2). The mean baseline PVI's in Group 1 (balloon inflation) varied from 0.52 ± 0.02 to 0.76 ± 0.02 ml (SD) (Table 3).

TABLE 2

Baseline mean pressure volume data in control animals*

Cat No.PVI ± SD (ml)Ecsf ± SD (mm Hg/ml)Ro ± SD (mm Hg/ml/min)
10.60 ± 0.0339 ± 293 ± 6 
20.59 ± 0.0266 ± 2131 ± 9 
30.50 ± 0.0342 ± 2114 ± 21 

PVI = pressure-volume index; Ecsf = elastance of the CSF space; Ro = CSF outflow resistance; SD = standard deviation.

Figure 3 demonstrates the changes in PVI associated with epidural balloon inflation in Group 1 and the changes in PVI during the mock-inflation period in Group 2. Figure 5 shows the typical sequential changes in PVI in a single animal from Group 1 during the balloon-inflation period. During the first 20 minutes of epidural balloon inflation, initial PVI values in all but two of the Group 1 animals were below the baseline mean PVI. None of these initial sub-baseline PVI values differed from the mean baseline PVI at a level of significance of ≤0.05. Only six of the initial 18 sub-baseline PVI's seen in Group 1 during epidural balloon inflation differed from the baseline PVI's by more than the difference that was observed during mock-inflation in Group 2 (Fig. 3 right). However, in Group 1 after 20 minutes of balloon inflation, there was a progressive elevation of the PVI with increasing balloon volume such that the PVI's obtained during the balloon-inflation period differed from baseline PVI values at the 0.05 level of significance by 50 ± 13 min (SD) (balloon volume = 1.2 ± 0.3 ml). The 0.01 level of significance for the balloon inflation PVI's relative to the baseline PVI's in Group 1 is reached at 66 ± 10 min (SD) of inflation time (balloon volume = 1.5 ± 0.2 ml) (Table 3). No trend was noted in sequential PVI's determined during the mock-inflation period in Group 2 (Fig. 3 left).

Fig. 5.
Fig. 5.

Plot of ICP, PVI, and Ecsf as a function of balloon volume from a single Group 1 animal. With increasing balloon volume there is a progressive rise in both Ecsf and PVI. The intracranial mass lesion pressure-volume curve is of the classic configuration.

In every Group 1 animal in which dilatation of the left pupil was followed by a PVI measurement (five animals), pupillary dilatation was associated with an abrupt elevation of PVI (Fig. 3 right) from 1.3 ± 0.4 to 3.2 ± 1.4 ml (SD) (Table 3). The mean change in PVI associated with pupillary dilatation was 1.9 ± 0.9 ml (SD) (Table 3).

Among the 10 animals in this study there was no relationship between the level of mean arterial blood pressure (MABP) or pCO2 during the baseline period and the level of the PVI. The pCO2 remained constant during inflation of the balloon.

Cerebrospinal Fluid Elastance

The mean baseline Ecsf in control animals varied from 39 ± 2 to 66 ± 2 mm Hg/ml (SD) (Table 2, Fig. 4 left). The mean baseline Ecsf in Group 1 (balloon inflation) varied from 15 ± 1 to 44 ± 2 mm Hg/ml (SD) (Table 4).

TABLE 4

Changes in elastance of the CSF space (Ecsf) secondary to progressive epidural balloon inflation*

Cat No.Ecsf (mm Hg/ml)Significance Level
Baseline mean ± SDPPDAPDΔEtα = 0.05 (min)tα = 0.01 (min) 
127 ± 17944−853040
244 ± 28140−413050
339 ± 19770−272040
435 ± 29260−322060
536 ± 11242030
615 ± 1972022
732 ± 212998−312020
mean ± SD100 ± 2061 ± 27−33 ± 523 ± 537 ± 5

PPD = immediately before pupillary dilatation; APD = immediately after pupillary dilatation; ΔE = Ecsf immediately after pupillary dilatation minus E immediately before pupillary dilatation; tα 0.05 = balloon inflation time required for baseline elastance to differ from baseline mean at a significance level of 0.05; tα 0.01 = balloon inflation time required for baseline elastance to differ from baseline mean at a significance level of 0.01.

Figure 4 shows the changes in Ecsf associated with epidural balloon inflation in Group 1 as well as the changes in Ecsf during the mock-inflation period in Group 2. Figure 5 shows the typical sequential changes in Ecsf seen in a single Group 1 animal. In Group 1 there was a progressive elevation of the Ecsf associated with increasing epidural balloon volume. The Ecsf values obtained in Group 1 during the balloon-inflation period differ from the baseline Ecsf values at the 0.05 level of significance when the mean balloon inflation time equals or exceeds 23 ± 5 min (balloon volume = 0.5 ± 0.1 ml) (Table 4). The 0.01 level of significance for the Group 1, balloon inflation Ecsf values is reached when the mean balloon-inflation time equals or exceeds 37 ± 15 min (balloon volume = 0.8 ± 0.3 ml) (SD) (Table 4). Two of the control animals (Fig. 4 left) experienced a moderate elevation of Ecsf during the mock-inflation period. However, in these two control animals the elevations in Ecsf during mock inflation were insignificant when compared to the large changes in Ecsf occurring in Group 1 with balloon inflation (Fig. 4).

Dilatation of the left pupil was followed by an impressive, abrupt diminution of Ecsf from 100 ± 20 to 61 ± 27 mm Hg/ml (SD) (Fig. 4, Table 4). The mean change in Ecsf in Group 1 associated with pupillary dilation was 33 ± 5 mm Hg/ml (SD) (Table 4).

Cerebrospinal Fluid Outflow Resistance

The mean baseline Ro in the three control animals varied from 93 ± 6 to 131 ± 9 mm Hg/ml/min (SD) (Table 2). The mean baseline Ro in Group 1 varied from 59 ± 6 to 118 ± 26 mm Hg/ml (SD) (Table 5).

Figure 6 shows the changes in Ro associated with balloon inflation in Group 1 and during the mock-inflation period in Group 2. In the Group 1 animals, Ro showed no definite single trend associated with increasing balloon volume (Fig. 6 right). However, of the 52 measurements of Ro in Group 1 during balloon expansion only three fell below the baseline mean value. In control animal No. 1 (Fig. 6 left), the Ro appeared to reset to a new, higher value during the mock-inflation period. Only 17 of the 52 measurements of Ro made during balloon inflation in Group 1 exceeded the level of the maximum spontaneous upward variation in Ro (shown as the dotted line in Fig. 4 right) noted in the control animals. The general patterns of change in Ro during epidural balloon inflation have been summarized in Table 5. In one animal of Group 1 during the balloon-inflation period, Ro did not reach a value that differed from the baseline Ro value at a level of significance of ≤0.05. In the remaining six Group 1 animals, Ro during balloon inflation differed from baseline Ro at the 0.05 level of significance when balloon-inflation time reached a mean of 67 ± 15 min (SD) (balloon volume = 1.5 ± 0.3 ml) (Table 5).

Blood Pressure

Baseline MABP's in the control group varied from 142 ± 5 to 166 ± 5 mm Hg (SD). In the three control animals there was no change in blood pressure during the mock-inflation period. The blood pressure changes in the balloon-inflation animals are shown in Table 6. Baseline MABP's varied from 113 ± 3 to 182 ± 4 mm Hg (SD). Four patterns of blood pressure changes were noted during epidural balloon inflation in Group 1 animals. Before dilation of the left pupil, five of seven Group 1 animals experienced a decline in MABP amounting to 31 ± 13 mm Hg (SD). At the time of dilation of the left pupil, MABP was 127 ± 22 mm Hg (SD). Five Group 1 animals demonstrated a terminal Cushing response. In these five animals all but one demonstrated a decline in blood pressure from the point at which dilation of the pupil began to the point at which the Cushing response began. The mean blood pressure at the onset of the Cushing response was 103 ± 42 mm Hg (SD). The time from onset of pupillary dilatation to the beginning of the Cushing response was a mean of 10 ± 7 min (SD).

TABLE 6

Blood pressure changes secondary to epidural balloon inflation*

Animal No.Baseline mean MABP ISP (mm Hg)PatternMABP (mm Hg) PPDMABP (mm Hg) PCRMABP (mm Hg) CRΔt (min) PD → CR
1166 ± 2A1357730811
2140 ± 4A115
3182 ± 4C1335012518
4188 ± 3B12010730017
5157 ± 16D1701862800
6134 ± 5A1159530010
7113 ± 3B100

Pattern of blood pressure change: A = progressive decline in blood pressure; B = moderate increase in blood pressure with later decrease in blood pressure before pupil dilates; C = no initial trend in blood pressure with late blood pressure decline before pupil dilates; D = no trend. PPD = measurement immediately before pupillary dilatation; PCR = measurement immediately before Cushing response; CR = measurement at onset of Cushing response; Δt(min) PD → CR = the time elapsed between pupillary dilatation and the onset of the Cushing response.

Pupillary Responses and EEG

In all but one Group 1 animal the left pupil became dilated before the right; in the remaining animal both pupils dilated simultaneously. The mean time of balloon inflation before pupillary dilatation was 88 ± 7 min (SD). Pupils were usually noted to narrow 10 to 20 minutes before dilatation of the left pupil. At the time of the Cushing response the pupils were usually fixed and dilated. The EEG was usually active at the time of pupillary dilatation and became flat before, or at the onset of, the Cushing response.

Pathological Findings

All Group 1 animals demonstrated the following macroscopic lesions within the central nervous system: 1) focal depression in the left occipital lobe immediately beneath the epidural balloon measuring approximately 1.5 cm in diameter and 1 to 2 mm deep; 2) subarachnoid hemorrhage around the margin of this depression especially posteriorly; 3) white matter swelling in the left cerebral hemisphere with midline shift and compression of left lateral ventricle; and 4) petechial hemorrhages in the cortex and adjacent white matter under the balloon.

In addition to these lesions, three Group 1 animals demonstrated scattered petechial hemorrhages in the left diencephalon, and one animal had petechial hemorrhages in the midline of the midbrain. The brains of the control animals were macroscopically normal.

Conclusions

Our results support the following conclusions concerning the physiological sequelae of constant rate epidural brain compression in cats:

  1. The plot of ICP versus epidural mass volume has two clearly definable zones: an initial flat segment and a final steep segment (Fig. 2).

  2. Significant volume accumulates in the epidural balloon (about 64% of the volume necessary to dilate the pupil) before the steep segment of the ICP versus epidural mass volume curve. The ICP may be relatively low (< 30 mm Hg) at this stage.

  3. Pupillary dilatation is a late sign of decompensation from an expanding epidural mass lesion.

  4. After 20 minutes of epidural balloon inflation there is a progressive increase in the pressure-volume index (PVI).

  5. Large increases in PVI occur at the time of transtentorial herniation.

  6. The CSF elastance (Ecsf) increases progressively with increasing epidural mass volume up to the time of pupillary dilatation.

  7. There is an abrupt decrease in Ecsf associated with transtentorial herniation.

  8. Epidural brain compression may elevate CSF outflow resistance (Ro).

Discussion

Before these experiments we expected that brain compression would produce a decrease in PVI and an increase in Ecsf. A decrease in PVI was anticipated because this change would reflect a steeper CSF pressure-volume relationship (Equation 2). Elevation of Ecsf would also be a reflection of the decreased craniospinal volume buffering capacity. Much to our surprise PVI was stable initially and then began to increase dramatically (Fig. 3 right). Simultaneous to these changes in PVI there was a steady increase in Ecsf (Fig. 4 right). One is, therefore, faced with two seeming contradictions. Since PVI determines the shape of the Pcsf versus Vcsf plot and since Ecsf defines the instantaneous slope of this same curve; how could Ecsf increase and PVI remain stable? Furthermore, how could PVI increase (a flatter curve) while Ecsf is also increasing (a steeper curve)? Since Ecsf, which depends on PVI (Equation 6), behaved in cats exactly the same during epidural balloon inflation as the VPR (a direct estimate of Ecsf not depending on PVI) had in baboons,9 we were led to suspect that the above discrepancies were not due to failure of the equations to adequately represent the system but were due to inadequacies in our understanding of the physiological basis of CSF pressure change. We were forced, therefore, to critically review our understanding of the physiology of CSF pressure change.

Speculation about intracranial pressure-volume interrelationships has been largely shaped by the Monro-Kellie doctrine and its later modifications.2,7,21,23,24 The Monro-Kellie theory is based upon the assumption that the skull is rigid and its contents, basically fluid, are incompressible. Any change in the volume of an intracranial component corresponds, therefore, to an equal and opposite change in the volume of one or more of the other intracranial components. If one includes the contents of the spinal canal as well as the venous blood in the spinal epidural plexus, the Monro-Kellie theory remains acceptable today.

According to the Monro-Kellie theory, compensation for an expanding intracranial mass lesion occurs by displacement of venous blood and spinal fluid from the intracranial and spinal system. Langfitt, et al.,8 have plotted the time course of the ICP during slow, constant-rate expansion of an extradural balloon. The initial segment of the curve shows only a modest increase in ICP with time, then the curve breaks sharply so that any further expansion of the mass results in a large increase in ICP. Thus, the ICP may remain relatively low at a point in time when an intracranial mass has reached a significant size. Our results confirm that for constant rate expansion of an epidural balloon, the plot of ICP versus mass volume has two segments: a slowly ascending segment, and a steep segment (Fig. 2). We also confirm that significant mass volume accumulates before large increases in ICP occur. In our experiments, by the time the epidural mass has reached the break point of the ICP versus mass volume curve, the mean ICP was only 24 ± 5 mm Hg (SD) while the epidural balloon volume had reached 64% ± 5% (SD) of the balloon volume that ultimately caused pupillary dilatation (Fig. 2 and Table 1).

To avoid confusion with other pressure-volume functions relating to the craniospinal axis we term the plot of ICP versus volume of an expanding intracranial mass the “intracranial mass lesion pressure-volume curve.” There may be several different intracranial mass lesion pressure-volume curves, each with its own peculiar mathematical properties, depending upon the type of mass lesion considered. For example, there is no reason to think, as is usually assumed, that the intracranial mass lesion pressure-volume curve should have the same mathematical properties for an intraaxial mass lesion as for an extraaxial mass lesion. The nature of the intracranial mass lesion pressure-volume curves for mass lesions other than epidural balloons expanding at a constant rate remain to be determined.

The usual description of the initial segment of the intracranial mass lesion pressure-volume curve for constant rate expansion of an epidural mass emphasizes the flat nature of this part of the curve. This emphasis obscures the fact that a significant CSF pressure increase does occur during this part of the curve (Fig. 2). If the Monro-Kellie doctrine holds and the volume of an intracranial mass is always compensated by the loss of the same volume of CSF and venous blood, why should the CSF pressure increase at all? The classic description of the steep segment of the intracranial mass lesion pressure-volume curve for progressive epidural brain compression tends to obscure the fact that this steep segment is far from vertical (Fig. 2). The decline in blood pressure noted during epidural balloon inflation in our animals may have caused the mass lesion pressure-volume curves to be less steep than if constant blood pressure had been maintained throughout epidural balloon inflation. Considering the magnitude of blood pressure change in our animals we do not believe this effect is significant. If the Monro-Kellie doctrine holds and the rapidly ascending part of the intracranial mass lesion pressure-volume curve represents the point where intracranial volumetric compensation fails, then why does not the slope of the second portion of the intracranial mass lesion pressure-volume curve more nearly approach a vertical asymptote?

The obvious partial answer to these questions is that the shift in intracranial volumes caused by a growing mass lesion is not in itself the primary cause of the changes in ICP. How then does an expanding intracranial mass lesion act to produce an elevation of ICP? An analogous situation is the inflation of a balloon in a water bath where the total volume of the container is kept constant by runoff or influx of water to accommodate for changes in balloon volume and where pressure is measured in the balloon. The analogy is even more applicable if one considers the hypothetical balloon to be inflated with saline at a constant rate (Fcsf), and if one considers the hypothetical balloon to have a hole in it so that fluid exits from the balloon against a specific outflow resistance (Ro), the outflow hole in the balloon also having a specific opening pressure (Pv). The pressure inside the balloon at any time is determined by the volume inside the balloon, and by the elastic properties of the walls of the balloon (tendency of the balloon walls to squeeze in on the space inside the balloon). If the elastic properties of the walls of the balloon remain stable, the rate of change of pressure inside the balloon is determined by the rate of change of volume inside the balloon. The factors that determine the rate of change of volume inside the hypothetical balloon are: 1) the rate of fluid input (Fcsf), 2) the resistance to fluid outflow (Ro), and 3) the opening pressure of the hole in the balloon (Pv). The importance of the elastic properties of the walls of the hypothetical balloon in our analogy is illustrated by comparing two such balloons, one with a wall of thick rubber and the other with a wall of thin rubber. Clearly these two balloons will have different internal pressures at some time (t) when Fcsf, Ro, Vcsf, and Pv for the two balloons are equal.

The anatomical structures that interact to form the elastic walls of the CSF space are: 1) the easily distensible spinal dural sac with its surrounding, readily collapsible epidural venous plexus; 2) the pia and ependyma; 3) the collapsible subarachnoid and intraparenchymal veins; 4) the more rigid scaffolding of cerebral arteries and arterioles that run like struts through the cerebral parenchyma (vascular pressure gradients may effect the elastic properties of the walls of the CSF space); 5) the brain tissue itself; and 6) the cranial dura reinforced by the bony calvaria.

The elastic properties of a system such as our hypothetical balloon or the CSF space are defined by the mathematical expression giving the pressure inside the system as a function of the volume inside the system. We term the plot of CSF pressure versus CSF volume “the CSF pressure-volume curve.” Any mathematical function defining the CSF pressure as a function of CSF volume we term a “CSF elastic property function.” There is certainly no reason to assume that the complex anatomical structures that interact to produce the elastic responses of the CSF space should be represented under all circumstances by a single mathematical function or even a single class of mathematical functions.

Equation 4 (see mathematical model) does seem, however, to be the valid CSF elastic property function under normal physiological conditions in the cat,12 and for some conditions in man,22 indicating that CSF pressure depends on just the properties that were predicted from the balloon analogy, namely, Fcsf, Ro, Pv, Vcsf, and the elastic properties of the surroundings represented by PVI and Veq.

If Equation 4 is the correct CSF elastic property function under conditions of epidural brain compression (an assumption now under investigation in our laboratory), one would predict that the reason for the shape of the intracranial mass lesion pressure-volume curve could be fully understood by evaluating the effect of brain compression upon Fcsf, Ro, Pv, Vcsf, PVI, and Veq. We have assumed that Fcsf and Pv did not change under the conditions of our experiments.

With Fcsf and Pv constant, the equilibrium pressure (Peq) of the system depends only on the outflow resistance of the system, Ro (Peq = RoFcsf + Pv).3,5 In other words, Peq is the pressure that must be achieved so that the rate of fluid influx into the system equals the rate of fluid efflux from the system. The equilibrium volume, Veq, is that volume inside the system necessary to achieve the Peq. Thus, Veq depends on the Ro and on the structural properties of the walls of the CSF space. When the surroundings squeeze in tightly on the CSF space, a small Veq will be necessary to achieve the Peq. However, when the surroundings do not squeeze in so tightly on the CSF space, a large Veq will be necessary to achieve the same Peq. Since the PVI reflects only the pressure change in the CSF space associated with a given volume change of the CSF space, the PVI depends only on the structural properties of the surroundings of the CSF space. Thus, both PVI and Veq are constants that reflect the structural properties of the surroundings of the CSF space. As such, one would expect that any pathological process such as cerebral edema, brain compression, or intracerebral hemorrhage, that would alter the mechanical properties of brain parenchyma would also produce changes in Veq and PVI (provided that Equation 4 is the correct CSF elastic property function under the pathological circumstances in question).

Figure 7 upper left shows the ways in which structural changes in the surroundings of the CSF space or changes in the Ro (assuming Fcsf and Pv constant) would alter the graph of the CSF elastic property function under consideration (Equation 4). A decrease in PVI makes the CSF pressure-volume curve steeper, while an increase in PVI flattens it. An increase in Ro results in an increase in both Peq and Veq, while a decrease in Ro produces a decrease in Peq and Veq. Primary structural changes affecting only Veq shift the whole CSF pressure-volume curve either to the right or to the left on the volume axis.

Fig. 7.
Fig. 7.

The CSF pressure-volume curves under conditions of progressive epidural brain compression. Upper Left: Diagrammatic representation of the factors responsible for altering the shape and location of the graph of the CSF elastic property function given by Equation 4 (mathematical model). Elevating or decreasing the PVI makes the curve respectively flatter or steeper. Increasing or decreasing the equilibrium pressure of the CSF space, Peq, respectively raises or lowers the curve on the pressure axis. Changing the equilibrium volume of the CSF, Veq, causes a parallel horizontal shift of the curve along the volume axis. Changes in Ro cause parallel changes in both Veq and Ro. Upper Right: Diagrammatic representation of the hypothetical reason why ICP falls when epidural balloon inflation is stopped. oPVI, oVeq, and oPeq = the values of these variables before epidural balloon inflation is started and tPVI, tVeq, tPeq, and tVT = the values of PVI, Veq, Peq and VT when epidural balloon inflation is stopped at time t. Since the decrease in Veq with epidural balloon inflation has exceeded the decrease in VT, the ICP declines to tPeq when balloon inflation has stopped. Lower: Diagrammatic representation of the alterations in the CSF pressure-volume curves hypothesized to occur with progressive epidural brain compression. PVIo, Eo, oVeq, oVT, and oPeq represent the values of these variables before epidural balloon inflation and PVIn, nEo, nVT, nVeq, and nPeq represent the values of these variables at the nth determination (where n = 1.....5) of the pressure-volume status during epidural balloon inflation. Initially the shift in Veq is small, but later becomes large, compared with the change in PVI. When transtentorial herniation occurs, the CSF pressure-volume curve may become very flat and Ecsf may actually decrease.

In this communication we have reported a progressive increase in both PVI and Ecsf, and a somewhat variable increase in Ro associated with constant-rate epidural balloon inflation in the cat. Variations within the cerebrovascular system may have accounted for the variability in Ro by altering the decay of ICP following an intraventricular fluid injection. We did not evaluate Veq in this series of experiments. In other experiments we have noted that when constant-rate epidural balloon inflation has been continued to a certain point and then stopped, the CSF pressure declines to a value less than the CSF pressure at the moment when balloon inflation is stopped. Since total CSF volume almost certainly decreases with progressive brain compression, a decrease in CSF pressure when epidural balloon inflation is stopped is most compatible with decreasing Veq as epidural balloon volume increases (Fig. 7 upper right).

We propose that the shape of the intracranial mass lesion pressure-volume curve is a reflection of the balance between the progressive right-to-left shift of Veq, the progressive flattening of sequential CSF pressure-volume curves (due to increasing PVI), and the fact that total CSF volume decreases more slowly than Veq. The changes in PVI and Veq are secondary to structural changes in brain parenchyma caused by the enlarging epidural mass lesion (Fig. 7 lower). The CSF pressure-volume curves are simultaneously shifted upward depending on the degree of increase in Ro with balloon inflation. Increase in Ro may be due simply to mechanical occlusion of the subarachnoid space. The relatively flat segment of the intracranial mass lesion pressure-volume curve (Fig. 2) reflects the fact that little right-to-left shift of Veq has occurred relative to the degree of flattening of the CSF pressure-volume curve. The steep part of the intracranial mass lesion pressure-volume curve, on the other hand, represents the situation where the right-to-left shift of the Veq is large compared to the flattening of the CSF pressure-volume curve (Fig. 7 lower).

The progressive increase in CSF elastance (Ecsf) occurs because there is a shift to a steeper part of a generally flatter CSF pressure-volume curve. The decline in Ecsf at the time of transtentorial herniation (indicated by pupillary dilatation) occurs because the structural changes in the brain parenchyma at that time are such that the CSF pressure-volume curve is very flat. In spite of the right-to-left shift of Veq, the CSF system changes transiently at this time to a less steep part of a flatter CSF pressure-volume curve. The changes in Ecsf in our experiments parallel the changes in VPR noted by Leech and Miller9 during the constant-rate inflation of an epidural balloon in baboons where decline in VPR was associated with the final stages of epidural balloon inflation. These parallel changes in Ecsf and VPR emphasize the fact that VPR is simply a practical method of estimating Ecsf (see Equation 5). Löfgren and Zwetnow11 also noted a decrease in CSF elastance as ICP approached MABP during the final phases of epidural balloon expansion in dogs.

Many of the known facts regarding the normal physiology and pathophysiology of alterations in volume pressure within the craniospinal system can be explained by considering how a given process alters the CSF pressure-volume curve. Clearly, any process that produces a change in ICP must do so by: altering the bulk flow dynamics of CSF and changing the volume inside the CSF space, that is, changing Ro, Peq, Veq, and ultimately VT; changing the elastic properties of the surroundings of the CSF space so that the environment squeezes in more or less tightly on the CSF space, that is, changing Veq and PVI when Equation 4 holds; or by simultaneously changing both CSF bulk flow dynamics and the CSF elastic property function.

As an example, the change in volume (and intravascular pressures) in the arterial side of the cerebral circulation occurring with each cardiac cycle is most likely associated with rhythmic alterations of the mechanical properties of brain parenchyma, thus altering the CSF elastic property function and leading to the familiar CSF pulsations synchronous with the cardiac cycle and with respirations. One would imagine that, under normal conditions, the CSF space is subjected to recurrent cyclic shifts among a family of CSF pressure-volume curves producing the respiratory and pulse pressure variations in ICP.

How would these concepts explain the well known fact that the pulse pressure of the ICP increases as ICP increases? Using Equation 4 we find that the CSF pressure-volume curves become more asymptotic (Fig. 7) as ICP increases so that the vertical distance between two similar curves that are separated on the volume and pressure axis increases with increasing ICP (Fig. 8). To maintain this relationship between CSF pressure-volume curves the magnitude of the right-to-left shift of Veq must balance out any increase in PVI, or a decrease in PVI must balance out any left-to-right shift in Veq. Whether the cardiac cycle acts primarily to change Veq, PVI, or to change both of the variables is as yet unknown. In other words, we hold that the change in volume in the arterial side of the cerebral circulation with cardiac ejection is immediately balanced by the loss of an equal volume of cerebral venous blood (the Monro-Kellie doctrine holds) and that CSF pressure is simultaneously increased because the scaffolding of cerebral arterioles throughout the brain has now stiffened, causing the cerebral parenchyma to push in more upon the CSF space (a change in the CSF elastic property function). We do not feel, as proposed by others,1,12 that each cardiac cycle loads the CSF space by a volume quantitatively equal to the volume change occurring in the arterial side of the cerebral circulation. If the CSF space is volume-loaded by each cardiac cycle, it is because the CSF pressure-volume curve shifts relative to the total volume of the CSF space, and the volume-loading of the CSF space under these conditions is not quantitatively equal to any volume change in the cerebral circulation.

Fig. 8.
Fig. 8.

Diagram illustrating the mechanism by which ICP is altered during a single cardiac cycle. As systole occurs, the pressure-volume index changes from dPVI to sPVI and Veq changes from dVeq to sVeq. As a result of these changes in the CSF elastic property function, ICP changes from dP1 to sP1 or from dP2 to sP2. The ICP pulse pressure increases with increasing baseline CSF pressure from ΔP1 to ΔP2.

We also propose that the changes in ICP noted with alterations in pCO2 and arterial blood pressure as well as spontaneous changes in ICP secondary to cerebral vasomotor activity (such as Lundburg waves) may all be understood by considering how these various alterations affect the CSF elastic property. Furthermore, any pathological process that elevates ICP must be evaluated according to how that particular process has altered the CSF elastic property function.

This communication is the natural extension of previous investigations dealing with ICP in situations where the intracranial contents have been loaded by an expanding mass lesion. Langfitt, et al.,8 described the general characteristics of the ICP versus epidural mass volume plot. When evaluating the VPR as a predictor of the degree of intracranial volumetric compensation, Miller and his coworkers9,16 concluded that the actual shape of the ICP versus mass volume plot was variable among patients. Studies of the effect of mannitol and steroids on ICP and VPR10,15,18 have led to the conclusion that the shape of a stylized “intracranial volume-pressure curve” was altered by these agents. We now point out that many possibly dissimilar volume-pressure curves may be defined for the craniospinal axis, depending upon the specific anatomic compartments considered. Furthermore, a given volume change in two different intracranial compartments may not produce the same effect on ventricular fluid pressure. In this report we study some of the interrelationships between two different intracranial pressure-volume functions and explain the shape of the intracranial mass lesion pressure-volume curve according to changes (similar to those hypothesized by Miller and Leech10,15,18 for a stylized “intracranial pressure-volume curve”) occurring in the CSF pressure-volume curve.

Future investigations in the area of ICP dynamics must evaluate the validity of Equation 4, both in animals and, when possible, in man, under a variety of pathological conditions and under circumstances where individual physiological variables such as BP, and pCO2, have been systematically altered. If Equation 4 is valid, the effect of pathological and physiological alterations upon the parameters of this equation must be carefully worked out. On the other hand, if Equation 4 is not valid under some pathological and physiological conditions, appropriate CSF elastic property functions must be determined. The proper parameters of the correct equations should be evaluated in patients and ultimately the data handling should be automated. Only when ICP and appropriate parameters of valid CSF elastic property functions can be easily measured and quickly analyzed in the clinical setting will we be in a position to understand the “meaning” of a given level of ICP in a particular patient.

Acknowledgments

We wish to acknowledge the technical assistance of Ms. Arletha T. Allen, Mr. William Young, and Ms. Sharon Read of the Division of Neurological Surgery, Virginia Commonwealth University, Medical College of Virginia, Richmond, Virginia.

References

  • 1.

    Avezaat CJJvan Eijndhoven JHMde Jong DAet al: A new method of monitoring intracranial volume/pressure relationship in Beks JWFBosch DABrock M (eds): Intracranial Pressure III. Berlin/Heidelberg/New York: Springer-Verlag1976 pp 308313Avezaat CJJ van Eijndhoven JHM de Jong DA et al: A new method of monitoring intracranial volume/pressure relationship in Beks JWF Bosch DA Brock M (eds): Intracranial Pressure III. Berlin/Heidelberg/New York: Springer-Verlag 1976 pp 308–313

  • 2.

    Burrows G: On Disorders of the Cerebral Circulation; and on the Connection Between Affections of the Brain and Diseases of the Heart. London: Longman, Brown, Green and Longman1846 pp 5556Burrows G: On Disorders of the Cerebral Circulation; and on the Connection Between Affections of the Brain and Diseases of the Heart. London: Longman Brown Green and Longman 1846 pp 55–56

  • 3.

    Cutler RWPPage LGalicich Jet al: Formation and absorption of cerebrospinal fluid in man. Brain 91:7077201968Cutler RWP Page L Galicich J et al: Formation and absorption of cerebrospinal fluid in man. Brain 91:707–720 1968

  • 4.

    Dixon WJ: Processing data for outliers. Biometrics 9:74891953Dixon WJ: Processing data for outliers. Biometrics 9:74–89 1953

  • 5.

    Heisey SRHeld DPappenheimer JR: Bulk flow and diffusion in the cerebrospinal fluid system of the goat. Am J Physiol 203:7757811962Heisey SR Held D Pappenheimer JR: Bulk flow and diffusion in the cerebrospinal fluid system of the goat. Am J Physiol 203:775–781 1962

  • 6.

    Johnston IHJennett B: The place of continuous intracrnial pressure monitoring in neurosurgical practice. Acta Neurochir 29:53631973Johnston IH Jennett B: The place of continuous intracrnial pressure monitoring in neurosurgical practice. Acta Neurochir 29:53–63 1973

  • 7.

    Kellie G: An account of the appearances observed in the dissection of two or three individuals presumed to have perished in the storm of the 3rd, and whose bodies were discovered in the vicinity of Leith on the morning of the 4th November, 1821, with some reflections of the pathology of the brain. Trans Med Chi Soc Edinb 1:841691824Kellie G: An account of the appearances observed in the dissection of two or three individuals presumed to have perished in the storm of the 3rd and whose bodies were discovered in the vicinity of Leith on the morning of the 4th November 1821 with some reflections of the pathology of the brain. Trans Med Chi Soc Edinb 1:84–169 1824

  • 8.

    Langfitt TWWeinstein JDKassell NF: Cerebral vasomotor paralysis produced by intracranial hypertension. Neurology 15:6226411965Langfitt TW Weinstein JD Kassell NF: Cerebral vasomotor paralysis produced by intracranial hypertension. Neurology 15:622–641 1965

  • 9.

    Leech PMiller JP: The intracranial volume-pressure relationships during experimental brain compression in primates. I. Pressure responses to changes in ventricular volume. J Neurol Neurosurg Psychiatry 37:109310981974Leech P Miller JP: The intracranial volume-pressure relationships during experimental brain compression in primates. I. Pressure responses to changes in ventricular volume. J Neurol Neurosurg Psychiatry 37:1093–1098 1974

  • 10.

    Leech PMiller JD: The intracranial volume-pressure relationships during experimental brain compression in primates. 3. The effect of mannitol and hyperventilation. J Neurol Neurosurg Psychiatry 37:110511111974Leech P Miller JD: The intracranial volume-pressure relationships during experimental brain compression in primates. 3. The effect of mannitol and hyperventilation. J Neurol Neurosurg Psychiatry 37:1105–1111 1974

  • 11.

    Löfgren JZwetnow NN: Influence of a supratentorial expanding mass on intracranial pressure-volume relationships. Acta Neurol Scand 49:5996121973Löfgren J Zwetnow NN: Influence of a supratentorial expanding mass on intracranial pressure-volume relationships. Acta Neurol Scand 49:599–612 1973

  • 12.

    Marmarou A: A theoretical and experimental evaluation of the cerebrospinal fluid system. Drexel UniversityPh.D. Thesis1973Marmarou A: A theoretical and experimental evaluation of the cerebrospinal fluid system. Drexel University Ph.D. Thesis 1973

  • 13.

    Marmarou AShapiro KShulman K: Isolation of factors leading to sustained elevations of the ICP in Beks JWFBosch DABrock M (eds): Intracranial Pressure III. Berlin/Heidelberg/New York: Springer-Verlag1976 pp 3336Marmarou A Shapiro K Shulman K: Isolation of factors leading to sustained elevations of the ICP in Beks JWF Bosch DA Brock M (eds): Intracranial Pressure III. Berlin/Heidelberg/New York: Springer-Verlag 1976 pp 33–36

  • 14.

    Marmarou AShulman KLaMorgese J: Compartmental analysis of compliance and outflow resistance of the cerebrospinal fluid system. J Neurosurg 43:5235341976Marmarou A Shulman K LaMorgese J: Compartmental analysis of compliance and outflow resistance of the cerebrospinal fluid system. J Neurosurg 43:523–534 1976

  • 15.

    Miller JD: Pressure and volume in the craniospinal axis. Clin Neurosurg 22:761051975Miller JD: Pressure and volume in the craniospinal axis. Clin Neurosurg 22:76–105 1975

  • 16.

    Miller JDGaribi J: Intracranial volume/pressure relationships during continuous monitoring of ventricular fluid pressure in Brock MDietz H (eds): Intracranial Pressure. Berlin/Heidelberg/New York: Springer-Verlag1972 pp 270274Miller JD Garibi J: Intracranial volume/pressure relationships during continuous monitoring of ventricular fluid pressure in Brock M Dietz H (eds): Intracranial Pressure. Berlin/Heidelberg/New York: Springer-Verlag 1972 pp 270–274

  • 17.

    Miller JDGaribi JPickard JD: Induced changes of cerebrospinal fluid volume, effects during continuous monitoring of ventricular fluid pressure. Arch Neurol 28:2652691973Miller JD Garibi J Pickard JD: Induced changes of cerebrospinal fluid volume effects during continuous monitoring of ventricular fluid pressure. Arch Neurol 28:265–269 1973

  • 18.

    Miller JDLeech PJ: Effects of mannitol and steroid therapy on intracranial volume-pressure relationships in patients. J Neurosurg 42:2742811975Miller JD Leech PJ: Effects of mannitol and steroid therapy on intracranial volume-pressure relationships in patients. J Neurosurg 42:274–281 1975

  • 19.

    Miller JDPickard JD: Intracranial volume/pressure studies in patients with head injury. Injury 5:2652681974Miller JD Pickard JD: Intracranial volume/pressure studies in patients with head injury. Injury 5:265–268 1974

  • 20.

    Miller JDStanek ALangfitt TW: Concepts of cerebral perfusion pressure and vascular compression during intracranial hypertension. Prog Brain Res 35:4114321972Miller JD Stanek A Langfitt TW: Concepts of cerebral perfusion pressure and vascular compression during intracranial hypertension. Prog Brain Res 35:411–432 1972

  • 21.

    Monro A: Observations on the Structure and Functions of the Nervous System. Edinburgh: Creech and Johnson1783 pp 56Monro A: Observations on the Structure and Functions of the Nervous System. Edinburgh: Creech and Johnson 1783 pp 5–6

  • 22.

    Shulman KMarmarou A: Pressure volume considerations in infantile hydrocephalus. Dev Child Neurol 13 ( Suppl 25):90951971Shulman K Marmarou A: Pressure volume considerations in infantile hydrocephalus. Dev Child Neurol 13 (Suppl 25):90–95 1971

  • 23.

    Weed LH: Some limitations of the Monro-Kellie hypothesis. Arch Surg 18:104910681929Weed LH: Some limitations of the Monro-Kellie hypothesis. Arch Surg 18:1049–1068 1929

  • 24.

    Weed LHMcKibben PS: Pressure changes in the cerebro-spinal fluid following intravenous injection of solutions of various concentrations. Am J Physiol 48:5125301919Weed LH McKibben PS: Pressure changes in the cerebro-spinal fluid following intravenous injection of solutions of various concentrations. Am J Physiol 48:512–530 1919

Statham P23 pressure transducer manufactured by Statham Laboratories, Inc., P. O. Box 1178, Huerto Rey, Puerto Rico.

Gould Physiologic Recorder and Gould Brush EEG Coupler Model 11–4307–02 manufactured by Gould Inc., Instrument System Division, 3631 Perkins Avenue, Cleveland, Ohio.

IL Blood Gas Analyzer Model 113 manufactured by Instrumental Laboratory, Inc., 113 Hartwell Avenue, Lexington, Massachusetts.

This research was supported in part by NIH Grant 1 PO1 NS12587, A.D.W. Grant 3558(544) and Biomedical Research Support Grant NIH 5 SO7RR05430–15.

If the inline PDF is not rendering correctly, you can download the PDF file here.

Article Information

Address reprint requests to: Humbert G. Sullivan, M.D., Box 758, MCV Station, Medical College of Virginia, Richmond, Virginia 23298.

© AANS, except where prohibited by US copyright law.

Headings

Figures

  • View in gallery

    Polygraph recording illustrating the response of the ICP, BP, and EEG to a bolus intraventricular injection 0.3 ml of normal saline. The peak ICP (Pp) was taken as the first trough following intraventricular saline injection (Pp = 35.5 mm Hg). The baseline ICP (Po) was taken as the first diastolic ICP, which immediately preceded injection and was in the same part of the ICP respiratory cycle as Pp, (Po = 10.0 mm Hg). Pt was considered to be the 1-minute, postinjection, diastolic ICP that occurred in the same part of the ICP respiratory cycle as Pp (Pt = 20.5). Mean ICP before injection was 11 mm Hg. From Equation 1, Equation 5, and Equation 7, PVI = 0.55 ml, Ecsf = 46 mm Hg/ml, and Ro = 125 mm Hg/ml/min. Note that there is no change in the EEG or in the blood pressure with the rapid intraventricular injection of 0.3 ml of saline.

  • View in gallery

    Plot of the intracranial mass lesion pressure-volume curves (ICP versus balloon volume) for each the seven Group 1 animals. * = the point at which the left pupil dilates. ↑ = the break point of the curve. These curves are of the classic configuration having an initial relatively flat segment and a second relatively steep segment. Pupillary dilatation is a late event and significantly large volumes of saline have accumulated in the epidural balloon by the time the break point is reached.

  • View in gallery

    Alterations in the pressure-volume index (PVI), Left: Plot of sequential PVI's in each control animal (Group 2). Dashed line indicates the baseline standard deviation. Heavy line indicates the baseline mean PVI. Note that during the mock-inflation period the PVI does not exceed the baseline mean by greater than one standard deviation. Right: Plot indicating sequential changes in the pressure-volume index (PVI) seen in each Group 1 animal during the balloon-inflation period. The number of standard deviations (baseline sample) by which a given measurement of PVI exceeds the baseline mean (ΔPVI/SD) is plotted for each animal as a function of balloon-inflation time. Dashed lines indicate the maximum and minimum values of ΔPVI/SD in control groups. * = the point at which the left pupil dilates. ↑ = the inflection point of the intracranial mass lesion pressure-volume curve. After the balloon has been inflated for 20 minutes there is a progressive increase in PVI with increasing balloon volume. Large increases in PVI occur at the time of pupillary dilatation.

  • View in gallery

    Alterations in elastance of the CSF space (Ecsf). Left: Plot of sequential Ecsf values in each control animal. Dashed lines indicate the baseline standard deviation. Heavy lines indicate the baseline mean Ecsf. During the mock-inflation period, Ecsf exceeds the baseline mean Ecsf by a maximum of 5.5 standard deviations (max ΔE/SD = 5.5). Right: Plot indicating sequential changes in Ecsf seen in each Group 1 animal during balloon inflation. The number of standard deviations (baseline sample) by which a given measurement of Ecsf exceeds the baseline mean (ΔEcsf/SD) is plotted for each animal as a function of balloon-inflation time. Dashed lines indicate the maximum value of ΔEcsf/SD seen in the control group. * = the point at which the left pupil dilates. There is a progressive increase in Ecsf with increasing balloon volume. There is a marked decline in Ecsf associated with dilatation of the left pupil.

  • View in gallery

    Alterations in CSF outflow resistance (Ro). Left: Plot of sequential Ro values in each control animal. Dashed lines indicate the baseline standard deviation. Heavy lines indicate the baseline mean Ro. During the mock-inflation period, Ecsf exceeds the baseline mean Ecsf by a maximum of 8.5 standard deviations (max ΔE/SD = 8.5). Right: Plot indicating sequential changes in Ro seen in each Group 1 animal during the balloon-inflation period. The number of standard deviations (baseline sample) by which a given measurement of Ro exceeds the mean baseline Ro (ΔRo/SD) is plotted for each animal as a function of balloon-inflation time. Dashed line indicates the maximum value of ΔE/SD seen in the control group. Notice that only three Ro values from the balloon-inflation period fall below the baseline mean value.

  • View in gallery

    Plot of ICP, PVI, and Ecsf as a function of balloon volume from a single Group 1 animal. With increasing balloon volume there is a progressive rise in both Ecsf and PVI. The intracranial mass lesion pressure-volume curve is of the classic configuration.

  • View in gallery

    The CSF pressure-volume curves under conditions of progressive epidural brain compression. Upper Left: Diagrammatic representation of the factors responsible for altering the shape and location of the graph of the CSF elastic property function given by Equation 4 (mathematical model). Elevating or decreasing the PVI makes the curve respectively flatter or steeper. Increasing or decreasing the equilibrium pressure of the CSF space, Peq, respectively raises or lowers the curve on the pressure axis. Changing the equilibrium volume of the CSF, Veq, causes a parallel horizontal shift of the curve along the volume axis. Changes in Ro cause parallel changes in both Veq and Ro. Upper Right: Diagrammatic representation of the hypothetical reason why ICP falls when epidural balloon inflation is stopped. oPVI, oVeq, and oPeq = the values of these variables before epidural balloon inflation is started and tPVI, tVeq, tPeq, and tVT = the values of PVI, Veq, Peq and VT when epidural balloon inflation is stopped at time t. Since the decrease in Veq with epidural balloon inflation has exceeded the decrease in VT, the ICP declines to tPeq when balloon inflation has stopped. Lower: Diagrammatic representation of the alterations in the CSF pressure-volume curves hypothesized to occur with progressive epidural brain compression. PVIo, Eo, oVeq, oVT, and oPeq represent the values of these variables before epidural balloon inflation and PVIn, nEo, nVT, nVeq, and nPeq represent the values of these variables at the nth determination (where n = 1.....5) of the pressure-volume status during epidural balloon inflation. Initially the shift in Veq is small, but later becomes large, compared with the change in PVI. When transtentorial herniation occurs, the CSF pressure-volume curve may become very flat and Ecsf may actually decrease.

  • View in gallery

    Diagram illustrating the mechanism by which ICP is altered during a single cardiac cycle. As systole occurs, the pressure-volume index changes from dPVI to sPVI and Veq changes from dVeq to sVeq. As a result of these changes in the CSF elastic property function, ICP changes from dP1 to sP1 or from dP2 to sP2. The ICP pulse pressure increases with increasing baseline CSF pressure from ΔP1 to ΔP2.

References

1.

Avezaat CJJvan Eijndhoven JHMde Jong DAet al: A new method of monitoring intracranial volume/pressure relationship in Beks JWFBosch DABrock M (eds): Intracranial Pressure III. Berlin/Heidelberg/New York: Springer-Verlag1976 pp 308313Avezaat CJJ van Eijndhoven JHM de Jong DA et al: A new method of monitoring intracranial volume/pressure relationship in Beks JWF Bosch DA Brock M (eds): Intracranial Pressure III. Berlin/Heidelberg/New York: Springer-Verlag 1976 pp 308–313

2.

Burrows G: On Disorders of the Cerebral Circulation; and on the Connection Between Affections of the Brain and Diseases of the Heart. London: Longman, Brown, Green and Longman1846 pp 5556Burrows G: On Disorders of the Cerebral Circulation; and on the Connection Between Affections of the Brain and Diseases of the Heart. London: Longman Brown Green and Longman 1846 pp 55–56

3.

Cutler RWPPage LGalicich Jet al: Formation and absorption of cerebrospinal fluid in man. Brain 91:7077201968Cutler RWP Page L Galicich J et al: Formation and absorption of cerebrospinal fluid in man. Brain 91:707–720 1968

4.

Dixon WJ: Processing data for outliers. Biometrics 9:74891953Dixon WJ: Processing data for outliers. Biometrics 9:74–89 1953

5.

Heisey SRHeld DPappenheimer JR: Bulk flow and diffusion in the cerebrospinal fluid system of the goat. Am J Physiol 203:7757811962Heisey SR Held D Pappenheimer JR: Bulk flow and diffusion in the cerebrospinal fluid system of the goat. Am J Physiol 203:775–781 1962

6.

Johnston IHJennett B: The place of continuous intracrnial pressure monitoring in neurosurgical practice. Acta Neurochir 29:53631973Johnston IH Jennett B: The place of continuous intracrnial pressure monitoring in neurosurgical practice. Acta Neurochir 29:53–63 1973

7.

Kellie G: An account of the appearances observed in the dissection of two or three individuals presumed to have perished in the storm of the 3rd, and whose bodies were discovered in the vicinity of Leith on the morning of the 4th November, 1821, with some reflections of the pathology of the brain. Trans Med Chi Soc Edinb 1:841691824Kellie G: An account of the appearances observed in the dissection of two or three individuals presumed to have perished in the storm of the 3rd and whose bodies were discovered in the vicinity of Leith on the morning of the 4th November 1821 with some reflections of the pathology of the brain. Trans Med Chi Soc Edinb 1:84–169 1824

8.

Langfitt TWWeinstein JDKassell NF: Cerebral vasomotor paralysis produced by intracranial hypertension. Neurology 15:6226411965Langfitt TW Weinstein JD Kassell NF: Cerebral vasomotor paralysis produced by intracranial hypertension. Neurology 15:622–641 1965

9.

Leech PMiller JP: The intracranial volume-pressure relationships during experimental brain compression in primates. I. Pressure responses to changes in ventricular volume. J Neurol Neurosurg Psychiatry 37:109310981974Leech P Miller JP: The intracranial volume-pressure relationships during experimental brain compression in primates. I. Pressure responses to changes in ventricular volume. J Neurol Neurosurg Psychiatry 37:1093–1098 1974

10.

Leech PMiller JD: The intracranial volume-pressure relationships during experimental brain compression in primates. 3. The effect of mannitol and hyperventilation. J Neurol Neurosurg Psychiatry 37:110511111974Leech P Miller JD: The intracranial volume-pressure relationships during experimental brain compression in primates. 3. The effect of mannitol and hyperventilation. J Neurol Neurosurg Psychiatry 37:1105–1111 1974

11.

Löfgren JZwetnow NN: Influence of a supratentorial expanding mass on intracranial pressure-volume relationships. Acta Neurol Scand 49:5996121973Löfgren J Zwetnow NN: Influence of a supratentorial expanding mass on intracranial pressure-volume relationships. Acta Neurol Scand 49:599–612 1973

12.

Marmarou A: A theoretical and experimental evaluation of the cerebrospinal fluid system. Drexel UniversityPh.D. Thesis1973Marmarou A: A theoretical and experimental evaluation of the cerebrospinal fluid system. Drexel University Ph.D. Thesis 1973

13.

Marmarou AShapiro KShulman K: Isolation of factors leading to sustained elevations of the ICP in Beks JWFBosch DABrock M (eds): Intracranial Pressure III. Berlin/Heidelberg/New York: Springer-Verlag1976 pp 3336Marmarou A Shapiro K Shulman K: Isolation of factors leading to sustained elevations of the ICP in Beks JWF Bosch DA Brock M (eds): Intracranial Pressure III. Berlin/Heidelberg/New York: Springer-Verlag 1976 pp 33–36

14.

Marmarou AShulman KLaMorgese J: Compartmental analysis of compliance and outflow resistance of the cerebrospinal fluid system. J Neurosurg 43:5235341976Marmarou A Shulman K LaMorgese J: Compartmental analysis of compliance and outflow resistance of the cerebrospinal fluid system. J Neurosurg 43:523–534 1976

15.

Miller JD: Pressure and volume in the craniospinal axis. Clin Neurosurg 22:761051975Miller JD: Pressure and volume in the craniospinal axis. Clin Neurosurg 22:76–105 1975

16.

Miller JDGaribi J: Intracranial volume/pressure relationships during continuous monitoring of ventricular fluid pressure in Brock MDietz H (eds): Intracranial Pressure. Berlin/Heidelberg/New York: Springer-Verlag1972 pp 270274Miller JD Garibi J: Intracranial volume/pressure relationships during continuous monitoring of ventricular fluid pressure in Brock M Dietz H (eds): Intracranial Pressure. Berlin/Heidelberg/New York: Springer-Verlag 1972 pp 270–274

17.

Miller JDGaribi JPickard JD: Induced changes of cerebrospinal fluid volume, effects during continuous monitoring of ventricular fluid pressure. Arch Neurol 28:2652691973Miller JD Garibi J Pickard JD: Induced changes of cerebrospinal fluid volume effects during continuous monitoring of ventricular fluid pressure. Arch Neurol 28:265–269 1973

18.

Miller JDLeech PJ: Effects of mannitol and steroid therapy on intracranial volume-pressure relationships in patients. J Neurosurg 42:2742811975Miller JD Leech PJ: Effects of mannitol and steroid therapy on intracranial volume-pressure relationships in patients. J Neurosurg 42:274–281 1975

19.

Miller JDPickard JD: Intracranial volume/pressure studies in patients with head injury. Injury 5:2652681974Miller JD Pickard JD: Intracranial volume/pressure studies in patients with head injury. Injury 5:265–268 1974

20.

Miller JDStanek ALangfitt TW: Concepts of cerebral perfusion pressure and vascular compression during intracranial hypertension. Prog Brain Res 35:4114321972Miller JD Stanek A Langfitt TW: Concepts of cerebral perfusion pressure and vascular compression during intracranial hypertension. Prog Brain Res 35:411–432 1972

21.

Monro A: Observations on the Structure and Functions of the Nervous System. Edinburgh: Creech and Johnson1783 pp 56Monro A: Observations on the Structure and Functions of the Nervous System. Edinburgh: Creech and Johnson 1783 pp 5–6

22.

Shulman KMarmarou A: Pressure volume considerations in infantile hydrocephalus. Dev Child Neurol 13 ( Suppl 25):90951971Shulman K Marmarou A: Pressure volume considerations in infantile hydrocephalus. Dev Child Neurol 13 (Suppl 25):90–95 1971

23.

Weed LH: Some limitations of the Monro-Kellie hypothesis. Arch Surg 18:104910681929Weed LH: Some limitations of the Monro-Kellie hypothesis. Arch Surg 18:1049–1068 1929

24.

Weed LHMcKibben PS: Pressure changes in the cerebro-spinal fluid following intravenous injection of solutions of various concentrations. Am J Physiol 48:5125301919Weed LH McKibben PS: Pressure changes in the cerebro-spinal fluid following intravenous injection of solutions of various concentrations. Am J Physiol 48:512–530 1919

TrendMD

Metrics

Metrics

All Time Past Year Past 30 Days
Abstract Views 13 13 7
Full Text Views 156 156 60
PDF Downloads 74 74 18
EPUB Downloads 0 0 0

PubMed

Google Scholar