Relationship Between Scalp Blood Flow and Intracranial Pressure in Rabbits

H. Eskandary MD, H. Reihani MD, H. Najafipour MD

Neuroscience Research Center, Kerman University of Medical Sciences, Kerman, Iran

Introduction

Since the introduction of intracranial pressure monitoring (ICP) in 1951 by Guillaume and Janny¹ and the major work of Lundberg in 1960,² numerous methods of ICP monitoring have been devised. Unfortunately, as the majority of these methods are invasive,^2-7 the use of ICP monitoring has not become a routine procedure.² From 1959 to 1981, several authors have reported non-invasive methods for ICP monitoring in the newborn and infant,^7-17 but all of these techniques require an open fontanel and are therefore not applicable in older children and adults. Skalur et al. in 1979 and York et al. in 1984^18,19 have used visual evoked potential to indirectly evaluate ICP variations. Kast, in 1985,²⁰ recorded short-term fluctuations of CSF pressure non-invasively by the use of impedance audiometry and the measurement of the mechanical tension of the tympanic membrane. This technique, however, requires the tympanic membrane and ossicles to be intact.

Therefore, the search for non-invasive methods of ICP monitoring continues. In an attempt to overcome complications associated with the invasive nature of these methods,^4,15 we evaluated scalp blood flow (SBF) variables in the rabbit, in response to changes in ICP. SBF was measured by laser Doppler flowmetry technique, which is non-invasive and safe for blood flow monitoring in the human.

Materials and Methods

Experiments were performed on six New Zealand white rabbits of both sexes (2-3 kg body weight). Animals were initially anesthetized with Hypnorm (Janseen Co) 0.5 ml/3 kg, IM) and diazepam (1.7 mg/kg, IP).

When deemed necessary, complementary doses of sodium pentobarbital were injected via marginal vein during the experiment. Tracheotomy was performed to recover the animal from probable respiratory arrest due to anesthesia. The left femoral artery was cannulated and connected to a blood pressure transducer and the first channel of a Beckman Physiograph to record the arterial blood pressure during the experiment. The forehead was shaved to the line connecting the posterior part of the ears. An 0.5 cm vertical incision was made on the back part of the shaved area and the scalp was dissected to gain access to the skull. A fine burr hole, of 2 mm in diameter, was created in the skull until the CSF poured out. A heparinized saline-filled cannula of suitable diameter was left in the subdural space, and sealed around by bone wax and glue to prevent leakage. This cannula was connected to one arm of a Y-shaped plastic tube.

ICP was monitored continuously by connecting the other arm of the Y-shaped tube to another pressure transducer and connected to the second channel of the physiograph. Changes in ICP were induced by alteration in the height of a saline container connected to the end of the Y-shaped tube. A pressure transducer showed the pressure induced by the saline container as was connected to the container through the Y-shaped tube. A special skin-attached laser probe connected to a double channel laser Doppler blood flow monitor (Moor Instruments-England) was used to monitor scalp blood flow. The probe was fixed on the forehead skin in the midline between the eyes. Changes of blood flow were digitally monitored on the blood flow monitor and also simultaneously recorded by the third channel of the physiograph. The second laser probe was fixed on a shaved area of the leg skin (over the gastrocnemius muscle) to monitor alterations of the leg skin blood flow (LSBF) due to changes in ICP. This variable was recorded on the fourth channel of the physiograph. ICP was increased from its base line value step-by-step, with a 2mm Hg increment each time. Three minutes were allowed to elapse in order to allow blood flow to stabilize to its new value after each increment of ICP. The room temperature was the same in all experiments. The mean arterial pressure (MAP) was calculated from the physiography record by the following formula: MAP = Pd+1/3 (Ps-Pd), where Ps and Pd are systolic and diastolic pressure values, respectively.²¹ The values on the graphs are means ±SEM, and are compared with their controls using student paired t-test. The correlation between changes in ICP and blood flow was assessed using the Pearson correlation coefficient. The P values <0.05 were interpreted as significant.

Results

Figure 1 shows a tracing recorded from one of the animals during the experiment. The different base line values in this animal were: ICP, 2 mm Hg; MAP, 90 mm Hg; scalp and leg skin blood flow, 140 and 123 arbitrary units, respectively. When ICP was finally increased to 20 mm Hg, in nine steps of 2mm Hg increment during 27 minutes, the new values were: AMP, 126 mm Hg; scalp and leg skin blood flow values, 45 and 32 arbitrary units respectively. Normal ICP ranged from 1 to 5 mm Hg (2.3+0.6 mm Hg, mean ±SE) in the six animals. Base-line MAP, scalp and leg skin blood flow were 94.6±5.1 mm Hg, 139±9 and 173.3±28.3 arbitrary units respectively. It is noticeable that there was a short delay (about 4-10 seconds) between the elevation of ICP and initiation of reduction in scalp blood flow.

Figure 2 shows the results of changes in SBF and LSBF following alterations in ICP in the 108 experiments. It is clear that there is a reverse relationship between changes in ICP and SBF up to the ICP of about 18 mm Hg (Fig. 2a). After that the blood flow reached a plateau, and following a small and transient increase, it did not recover despite a reduction of ICP towards its normal base line value during 24 minutes.

To answer the question whether the changes of SBF secondary to the change in ICP were a specific or a generalized phenomenon, the LSBF at a point far from the head was also measured. Figure 2b shows the result of this measurement in the six animals. The changes of LSBF showed relatively the same trend of changes as SBF, though not in smooth increments.

Figure 3 shows the regression lines between changes in ICP, SBF and LSBF when ICP was plotted on the X axis as an independent variable and blood flow on the Y axis as a dependent variable. There is a reverse relationship between them. The Pearson correlation coefficient for SBF was -0.839 to ICP of 23 mm Hg (Fig. 3a) which is related to values up to minute 27 in Fig. 3a. Overall, the correlation coefficient was -0.519 (P<0.018) when all individual values of SBF were interpolated against their related values of ICP to the end of the experiment (minute 51 in Fig. 1a). The correlation coefficient for LSBF is -0.39 to ICP of 23 mm Hg (Fig. 3b) which is related to values up to minute 27 in Fig. 1b.

Figure 4 shows the changes in MAP with changes in ICP in the six animals. There was a direct relationship between ICP and MAP in all of the experiments. Overall MAP reached the mean value of 107.63 mm Hg from the mean base value of 94.6 mm Hg when IP reached its peak mean value of 19.83 mm Hg. However, the peak mean value of 109.3 for MAP was observed when ICP reduced to the mean value of 17.5 mm Hg which shows a delay in changes in MAP compared to the changes in ICP (Fig. 4) (R-0.819, P<0.000).

Discussion

ICP monitoring devices in use include epidural and subdural monitors, subarachnoid screws and bolts, as well as ventricular and intraparenchymal catheters.^3,4,22-25 The problems of ICP monitoring include complications due to the invasive nature of procedures such as infection and intracranial hemorrhage.²⁶ In the non-invasive techniques, the lack of these complications is impressive.

In the present study, in an attempt to find a non-invasive and indirect method of ICP evaluation, we used laser Doppler flowmetry to measure the blood flow, and consequently to predict changes in ICP. We found that SBF showed a rapid response to ICP manipulations, correlating well up to pressures of about 18 mm Hg (Fig. 2a). We did not increase ICP to values of more than 23 mm Hg as this would cause the animal to become conscious, and lead to a disturbance of blood flow measurement. On the other hand, SBF reached its minimum ICP value of about 18.5 and did not follow its decreasing trend; therefore, it was not necessary to continue the experiment with deeper anesthesia which might have caused respiratory or heart complications. Autoregulatory mechanisms probably did not allow the SBF to decrease to

lower values with increasing ICP, as this might interfere with the provision of minimum metabolic needs for skin cells.

The mechanism of changes in SBF due to changes in ICP is unclear at present. At first, one may speculate that blood flow changes are due to an alteration in blood pressure (BP). However, BP monitoring revealed that when ICP was raised the BP increased and SBF decreased (Fig. 4).The reverse relationship indicates that changes in BP are not responsible for decreased SBF due to elevation of ICP. BP elevation is probably a compensatory mechanism to correct brain blood flow impairments induced by high ICP values.

We know that intracranial and extracranial circulations are interrelated by emissary veins as well as other collateral veins and arteries.²⁷ The same direct effect of elevated ICP on brain venous system could be responsible for the changes in SBF. However, leg skin blood flow also showed parallel changes to SBF (Fig. 2b). This indicates that skin blood flow reduction is not limited to the scalp but can also be recorded at other sites. This is of clinical importance as in head-injured patients with scalp injury, this and other skin areas can be substituted for scalp blood flow monitoring. The lower correlation coefficient between LSBF and ICP compared with SBF and ICP is due to the higher variations in LSBF values around the mean (Fig. 2a,b). The probe for SBF was placed on a distinct anatomic site (see Methods) but such precise location could not be found for LSBF probe due to large surface of the leg skin. ICP manipulation reduced SBF by 51.7% (from the mean 139 to 67) and LSBF by 44% (from the mean 173 to 97) (Fig. 2a,b), does not reflect such a high difference in correlation coefficients. As the changes in blood flow due to changes in ICP were found to be generalized, other mechanisms, such as hormonal or neural, may be involved. What seems to be the main mechanisms are the systemic catecholamine release and an increase in mean arterial pressure and vasoactive motor function due to increased ICP. An increase in systemic vascular resistance is manifested in the scalp and systemically associated with a reduction in blood flow as the mean arterial pressure rises. Unfortunately, the blood flow did not recover when ICP was returned towards its basal value (Fig. 2a,b) indicating impairment of skin blood flow regulation mechanisms due to relatively long-lasting high ICP values. If this phenomenon occurs in the brain, it could probably lead to serious and irreversible damages to the brain function. Unfortunately, we did not continue the experiment to see if the blood flow restituted after recovering of ICP.

In conclusion, based on the results of this study, the scalp blood flow measurement may be used as an indicator of ICP. As many blood flow measurement techniques have the disadvantage of invasiveness and therefore are not clinically applicable, we propose the laser Doppler flowmetery method which has been frequently used in the human studies.²⁸ Prior to this, more studies should be performed to determine the base line values of the scalp blood flow in human and then the relationship between ICP and blood flow values should be investigated. The clinical profile of this technique of ICP monitoring is currently being evaluated.

References

1 Guillaume J, Janny P. Manometrie intracranienne continue. Rev Neurol 1951;84:131-42.

2 Levin AB. The use of a fiberoptic intracranial pressure monitor in clinical practice. Neurosurg 1977;1:266-271.

3 Allen R. Intracranial pressure: a review of clinical problems, measurement techniques and monitoring methods. J Medical Engineering Technology 1986;10:299-320.

4 Aucoin PJ, Kotilainen IIR, Gantz NM, et al. Intracranial pressure monitors: Epidemiologic study of risk factors and infections. Am J Med 1986;89:369-76.

5 Gambardella G, Avella D, Tomasello R. Monitoring of brain tissue pressure with a fiberoptic device. Neurosurgery 1992;31:918-22.

6 Rosner MJ, Beeker DP. ICP Monitoring complication and associated factors. Clin Neurosurg 1976;23:494-519.

7 Vidyasagar D, Rayu TNK. A simple non-invasive technique of measuring intracranial pressure in the newborn. Pediatrics 1977;59:957-61.

8 Bertt EM, Measurements of cerebrospinal fluid pressure in infants without puncture. Dev Med Child Neurol 1966;8:207-10.

9 Dearden NM, McDowall DG, Gibson RM. Assessment of Leeds device for monitoring intracranial pressure. J Neurosurg 1984;60:123-9.

10 Edwards J. An intracranial pressure tonometer for use on neonates: preliminary report. Dev Med Child Neurol 1974;16:38-9.

11 Heifetz MD, Weiss M. Detection of skull expansion with increased intracranial pressure. J Neurosug 1981;55:811-12.

12 Horbar JD, Yeager S, Philip AGS, Lucey JF. Effect of application force on non-invasive measurements of intracranial pressure. Pediatrics 1980;66:455-7.

13 Hill A, Volpe JJ. Measurement of intracranial pressure using the Ladd intracranial pressure monitor. J Pediatr 1981;98:974-6.

14 Philip AGS. Non-invasive monitoring of intracranial pressure: a new approach for neonatal clinical pharmacology. Clin Perinatol 1979;6:123-37.

15 Robinson RO, Sutton PRP. Non-invasive method for measuring intracranial pressure in normal newborn infants. Dev Med Child Neurol 1997;19:305-7.

16 Salmon JH, Hayyar W, Bada HS. The fontogram: A non-invasive intracranial pressure monitor. Pediatrics 1977;60:721-5.

17 Wealthall SR, Smallwood R. Pressure via the fontanelle without puncture. J Neurol Neurosug Psychiatry 1974; 37:88-96.

18 Sklar FH, Ehle AL, Clark WK. Visual evoked potentials: A non-invasive technique to monitor patients with shunted hydrocephalus. Neurosurgery 1979;4:529-34.

19 York D, Legan M, Berner S, Watts C. Further studies with a non-invasive method of intracranial pressure estimation. Neurosurgery 1984;14:456-61.

20 Kast R. A new method for non-invasive measurement of short-term cerebrospinal pressure change in humans. J Neurol 1985;232:260-1.

21 Bern RM, Levy MN. Cardiovascular physiology, 5th ed St Louis: Mosby: 1992.

22 Vries JK, Becker DP, Yourg HF. A subarachnoid screw for monitoring intracranial pressure. J Neurosurg 1973; 39:416-19.

23 Basauri L. Intracranial pressure monitoring: cost-benefit analysis of ventricular catheters and fiberoptic systems. Critical Reviews in Neurosurgery 1997;3:165-8.

24 Artru F: Monitoring of intracranial pressure. Cah Anesthesiol 1996;44:87-90.

25 Shapiro S, Bowman R, Callahany WC: The fiberoptic intraparanchymal cerebral pressure monitor in 244 patients. Surg Neurol 1996;45:278-82.

26 Jensen RL, Hahn YS, Ciro E. Risk factors of intracranial pressure monitoring in children with fiberoptic devices: a critical review. Surg Neurol 1997;47:16-22.

27 Hoffman WY. Scalp injuries and their management In: Youmans JR. Neurological Surgery, 4th ed. WB Saunders 1966;P:1916.

28 Shepherd AP, Oberg PA. Laser Doppler blood flowmetry, Boston, Kluwer Academic Publisher. 1990;PP:141-74.

AIM Home|Table of Contents