The effect of dexamethasone on vascular permeability ... - Springer Link

Acta Neuropathologica

Acta Neuropathol (Bed) (1986) 69:288-294

O Springer-Verlag 1986

The Effect of Dexamethasone on Vascular Permeability of Experimental Brain Tumours*' ** P. J. Luthert, J. Greenwood, P. L. Lantos, and O. E. Pratt Dept. of Neuropathology,Institute of Psychiatry,De CrespignyPark, London SE5 8AF, UK

Summary. The vessels of experimental gliomas show an abnormally high permeability to small polar molecules, such as mannitol. To establish whether this change in vessel permeability is modified by treatment with the corticosteroid dexamethasone, the kinetics of [14C]mannitol transfer into rat astrocytomas were estimated in both steroid- and saline-treated, tumourbearing animals. This was achieved by injecting [~4C]mannitol i.v., using a specially devised technique, so as to maintain a constant concentration of tracer in the blood plasma. In separate experiments steady levels of the tracer were maintained in the circulation from 1 to 30 rain. Mean plasma and tumour radioactivity were measured, and the apparent transfer constant of mannitol across the vascular endothelium and the size of the extravascular extracellular mannitol space in the tumours were calculated. Despite a significant clinical improvement in the treated animals and adequate circulating levels of dexamethasone at the time of the permeability studies, no difference in either the apparent transfer constant for the movement of mannitol into the tumours or the fractional extracellular mannitol space was detected between these animals and the controls. With steroid treatment both tumour-bearing and non-tumour bearing animals lost weight, and in the latter there was no consistent change in routine biochemical or haematological parameters. It was concluded that under these conditions it is unlikely that clinical improvement with dexamethasone therapy was due to a non-specific reduction in tumour vessel permeability to polar substances. * A preliminary account of this work was presented at the Sixty-Eighth Meeting of the British Neuropathological Society,January 1985 ** Supported by the WellcomeTrust and the MedicalResearch Council Offprint requests to: Dr. P. J. Luthert (address see above)

Key words: Experimental glioma - Vascular permeability - Dexamethasone - Mannitol

Introduction Glucocorticoids are extensively used in the management of patients with cerebral tumours. Although the dramatic clinical improvement often seen may, in part, be independent of any reduction in cerebral oedema (Pappius and McCann 1969; Gurcay et al. 1971; Pappius 1980) it is generally believed that this is the prime mode of action (Yamada et al. 1979, 1983; Matsuoka and Hossmann 1981). Of the many possible mechanisms involved in a reduction of vasogenic cerebral oedema by steroids, one mode of action, suggested by several investigators, is a lowering of the abnormally high permeability of the cerebral vessels to a wide range of blood-borne solutes (Rovit and Hagan 1968; Yamada et al. 1983). The majority of these studies have employed qualitative macromolecular tracers in assessing permeability and have not studied the kinetics of tracer transfer from blood to tissue. Furthermore, in oedema models producing extensive tissue damage, such as cold injury, steroids may act indirectly, by reducing the release of vasoactive inflammatory products through inhibition of the arachidonic acid cascade, rather than on the vessels directly, making it difficult to extrapolate to models, such as gliomas, where necrosis is not always a predominant feature. It has been reported recently that the apparent rate of transfer of tracer [14C]mannitol into an experimental rat glioma exceeds that for normal brain by almost two orders of magnitude (Deane et al. 1984). The aim of the present study was to ascertain, firstly, whether the glucocorticoid dexamethasone improved

P. J. Luthert et al.: Effects of Dexamethasone on Glioma Vascular Permeability the clinical c o n d i t i o n o f the a n i m a l s and, secondly, w h e t h e r a n y such i m p r o v e m e n t was associated with a r e d u c t i o n i n the high rate o f transfer o f 1 4 C - m a n n i t o l into these t u m o u r s . A l t h o u g h a r e d u c t i o n in p e r m e a b i l i t y to p o l a r molecules is j u s t one o f m a n y possible m e c h a n i s m s c o n t r i b u t i n g to the beneficial a c t i o n o f d e x a m e t h a sone, it is o f p a r t i c u l a r interest because o f the c o n c e r n that steroid t r e a t m e n t m a y impede the delivery o f water-soluble cytotoxic agents to p a t i e n t s with cerebral t u m o u r s ( N e u w e l t et al. 1982).

289

rapidly out of the vessels by a forced intra-arterial injection of isotonic saline (Pratt 1985), the head was cut off, the brain removed rapidly and the leptomeninges and accessible choroid plexus stripped from the surface of the brain which was then frozen in hexane cooled to - 70~C. Tissue samples were removed immediately and divided as follows: the cerebellum and brain stem were separated from the forebrain and each was divided into two parts to give duplicate samples. Slices 1 - 2 mm thick were cut coronally through the cerebral hemispheres. At the level of the tumour at least two samples were taken from the central region using a brain biopsy needle (i.d. 2.5 mm). Cerebrum, remote from the tumour, was also sampled in duplicate. Tumours less than 3 mm in diameter and those growing non-invasively over the surface of the brain were excluded from this study. The frozen tissue samples were weighed in tared scintillation vials.

Materials and Methods Animals, Tumour Model and Treatment

Radioactivity Counting

Adult BD IX rats of either sex, weighing between 250 and 450 g, were used. Water and food were given without restriction. Rats were inoculated in the right cerebral hemisphere with cultured neoplastic A15A5 cells, cloned from a mixed N-ethylN-nitrosourea-induced glioma (Lantos et al. 1976). The transplanted tumours have previously been characterised as invasive malignant astrocytomas, with associated oedema and occasional small areas of necrosis (Davaki and Lantos 1981). Three to 6 weeks after inoculation, when the animals' condition began to deteriorate, they were treated with either dexamethasone or saline on 3 consecutive days. Dexamethasone (0.8 mg-kg-l) was administered i.p. at 24-h intervals, whilst control tumour-bearing animals were treated with an equivalent volume of saline (0.9% w/v). The permeability measurements were started 30 min after the third dose. On each day of treatment, the animals were weighed and clinically graded on an arbitrary scale between 1 and 5: grade 5, if they were apparently normal; grade 4, if they were slightly slow; grade 3 if they could just climb a 45~ gradient; grade 2 if they could stand but not climb, and grade 1, if they were unable to stand.

Blood samples were taken into numbered syringes, the cells spun down and the plasma removed. The tissue and duplicate 10-~tl plasma samples were dissolved in a solution of an organic base (Soluene-350, Packard Instruments). The radioactivity in the samples was measured in an automatic scintillationspectrometer (Tricarb, Model 2409, Packard Instruments) after the addition of a toluene-based scintillationcocktail. When necessary, quench corrections were done by means of an external standard. As described previously corrections were made for the slight drop in mannitol concentration across the vascular bed (Deane et al. 1984).

Measurement of the Penetration of [14C]Mannitol into the Tissue Under sodium pentobarbitone anaesthesia (BPC, 50 rag-kg-1 b. wt. i.p.), two Nylon catheters (Portex, Hythe, Kent, UK) were inserted: one (1.02 mm ext. diam.) into a femoral artery for withdrawing blood samples and for flushing out the vascular system with saline and one (0.63 mm ext. diam.) into a femoral vein for giving the programmed i.v. injection. Animals were hepariuised to prevent blood clotting, especially at the cannula tip and during removal and centrifugation of the samples. Using an i.v. injection protocol, as described previously (Deane et al. 1984), [14C]manuitol was injected so as to replace the tracer as fast as it left the circulation (Pratt 1985) enabling steady levels to be maintained within the circulation for up to 30 min. The mean percentage deviation from the mean plasma radioactivity during the maintenance of a steady level was 6.4%. The [14C]mannitol, which had a specific activity of between 1.5 and 2.2MBq Ixmo1-1, was given at a concentration of 0.170 mmol. 1-1 in isotonic NaC1 solution. The rate of penetration of [14C]mannitol into tumour and non-tumour tissue, of both steroid- and saline-treated animals, was measured by calculating the ratio of tracer in the tissue at the end of the experiment to its mean level in the blood plasma over the period during which the programmed injection of the tracer was given. The experimental period ranged from I to 30 rain, and at the end of this period the blood was washed

Materials The D-l-[14C]mannitol was obtained from Amersham International PLC (Amersham, Bucks, UK) dissolved in sterile isotonic NaC1, and stored at - 20~C until required. Dexamethasone monosodium phosphate was supplied by Merck Sharp and Dohme (Hoddesdon, UK) and was diluted, prior to use, to a concentration of 0.8 mg. ml-1 with sterile saline (0.9% w/v). All other chemicals were of analytical quality if available.

Assessment of Non-specific Effects of Dexamethasone and Measurement of its Plasma Concentration In eight female animals dexamethasone was administered i.p. on 2 consecutive days at a dose of 0.8 mg. kg- 1. Twenty-four hours later, the animals were anaesthetised and exsanguinated by intracardiac puncture following no further treatment or 30 min after a third dose of steroid. Dexamethasone was extracted and measured according the modified procedure of Houghton et al. (1981). Following three doses, at 24-h intervals, routine haematological and biochemical measurements were made on eight dexamethasone-treated and four saline-treated animals. Changes in body weight were also recorded in these rats.

Results General Effects o f Dexamethasone As d e x a m e t h a s o n e p r o d u c e s weight loss in n o r m a l a n i m a l s (Fig. 1), there was a possibility t h a t the clinical i m p r o v e m e n t was s e c o n d a r y to d e h y d r a t i o n . H o w ever, the t u m o u r - b e a r i n g a n i m a l s treated with steroids

290

P.J. Luthert et al.: Effectsof Dexamethasoneon GliomaVascular Permeability ----0----

Saline

no tumour

(n-4)

D e x a m e t h a s o n e no tumour ( n = 8 ) Saline

J~

tumour

Table 1. The percentage of tumour-bearing animals improving, deteriorating or remaining the same over the first and second 24-h periods of treatmentwith saline or dexamethasone Percentage

(n=15)

D e x a m e t h a s o n e tumour

Saline-treated (n=15)

Dexamethasonetreated (n=30)

Day0-1 D a y l - 2

Day0-1 D a y l - 2

0 27 73

60 33 7

(n=30)

+

Improved No change Deteriorated

J~ o~

0 40 60

47 33 2

.$ g

Effects of Dexamethasone upon the Kinetics of [14C]Mannitol Transfer

r

8 -

I

I

1

2

Day

Fig. 1. Mean cumulative percentage weight change (4- SEM)

over the first and second 24-h periods of treatment in tumour and non-tumour-bearinganimals treated with saline or dexamethasone

lost significantly less weight over the second 24-h period of treatment than those given saline (P = 0.01 ; Student's t-test). In addition, no striking changes were found in plasma urea, electrolytes, haematocrit or blood sugar. Most tumour-bearing animals treated with dexamethasone improved dramatically, even following a single dose. The clinical score for saline-treated animals dropped from 4.2 + 0.4 (mean + SD) to 2.8 • 0.7 over the 48-h period of treatment, but the dexamethasone treated animals improved from a mean score of 3.6 _+ 0.7 to 4.5 _+ 0.7 over the same period. Table 1 shows the percentage of animals deteriorating, remaining the same or improving, according to this arbitrary scale, over the first and second 24-h periods of treatment. While approximately half the steroidtreated animals improved over each day, the control rats either showed no change or deteriorated. The mean dexamethasone plasma concentration prior to the last dose was 35_+28ng m1-1 ( m e a n + S E M ; n = 4 ) which increased to 747 • 88 rig" ml- 1 (n -- 4) 30 min after the third and final dose.

In all of the non-tumour areas sampled in both dexamethasone and saline-treated animals the ratio of the radioactivity in the tissue to that in the plasma (R,/ Rp) increased in direct proportion to the injection period over the range studied (up to 30 rain). This data could be fitted adequately by linear regression (Deane et al. 1984) and, using analysis of covariance with time as the covariant, no dexamethasone-related effect was determined in any of the three non-tumour areas studied. The gradients of these lines give the first order rate constants and hence transfer constants (see below), for mannitol infux into non-tumour brain tissue (Table 2). The intercepts of these lines are positive, demonstrating what has previously been described as the fast rnannitol space (Sisson and Oldendorf 1971). The ratio of radioactivity'g -1 tumour tissue/ radioactivity' ml-1 plasma (Rt/Rp) was calculated for each experiment and plotted against time (Fig. 2). This ratio increased almost in proportion to the length of the period of injection for the first few minutes. As the period was prolonged further, the ratio increased less rapidly, until after about 20 min no further increase was seen. It was concluded that the efflux of the tracer back into the blood from the extracellular space of the tumour was small for periods up to 2 min, whereas after 20 rain the extracellular concentration had apparently reached that of the plasma and efflux had become equal to influx. The control and experimental data for tumour tissue (Figs. 2, 3) were fitted separately to a twocompartment model using the method of maximum likelihood (Bard 1974) as described previously (Deane et al. 1984). This enables calculations to be made for the first order rate constant (k) and the fraction of the tissue occupied by the extravascular extracellular space (Se). The best estimates of these parameters derived from the control data of a previous study (Dearie et al. 1984), from a group of control experiments in the present work and from the pooled control data from the two series are shown in Table 3. The

P. J. Luthert et al. : Effects of Dexamethasone on Glioma Vascular Permeability 2. Regression line parameters for Rt/Rp on time for three brain areas and their means in both dexamethasone-treated (n = 28) and control (n = 28) animals. The intercepts correspond to the so-called fast mannitol space and the gradients to the apparent transfer constants for mannitol across cerebral capillary endothelium Table

Regression line parameters ( x 10-4) SD

Gradient (min - 1)

SD

Cerebrum Cerebellum Brain stem Mean Dexamethasone

27.9 28.7 32.3 29.6

7.00 5.81 7.78 5.40

4.46 4.01 4.70 4.39

0.80 0.46 0.72 0.51

0.873 0.889 0.862 0.916

Cerebrum Cerebellum Brain stem Mean

54.1 41.9 36.4 44.1

7.81 7.25 9.33 9.33

3.61 4.33 3.90 3.95

0.76 0.53 0.69 0.69

0.745 0.814 0.716 0.746

Control

:

9

0.1

b

10

20 Time

30

(rain)

Fig. 2. The accumulation of radioactivity in glioma and nontumour tissue in control animals from a series of experiments in which a steady level of [14C]mannitol was maintained in the blood plasma for periods up to 30 min. The ratio Rt/Rp is the radioactivity in the tissue at the end of the experiment divided by the mean radioactivity in the blood plasma during the experiment. The continuous line has been fitted to the pooled, tumour data points (0) using a two-compartment model and the interrupted line represents the fit by linear regression Of the ratio upon time for saline-treated forebrain tissue free of tumour (A)

0.3

0,2 R! 0.1

/

I

I

I

10

20

30

Time

Correlation coefficient

Intercept

0-3

Or

291

(min)

Fig. 3. The data points represent the tumour Rt/Rp ratios for dexamethasone-treated animals and the fit to this data is shown by the continuous line. The interrupted line represents the fit to the control tumour data as shown in Fig. 2

estimates of the parameters for the dexamethasonetreated animals lie between or close to those for the two control groups and close to the pooled control parameters. The product of Se and k gives an estimate of the unidirectional flux of the tracer out of the blood across the blood-tissue barrier, i.e. o f the apparent transfer constant for diffusion o f mannitol across the barrier. In situations where blood flow greatly exceeds the transfer constant the latter approximates to the permeability-surface area (PS) product (Blasberg et al. 1981). The mean surface area fraction for central parts of this tumour model has been measured recently (Luthert and Lantos 1985) permitting the estimation of the permeability coefficient of glioma endothelium for mannitol (Table 3).

Discussion

Despite the fact that dexamethasone has gained widespread use in the treatment of tumour-related oedema over the last 20 years, its mechanism of action remains unclear. The reduction in cerebral oedema that it generally produces is of undoubted importance in controlling the symptoms of raised intra-cranial pressure, but it is possible that the rapid improvement in clinical symptoms in patients and changes in E E G activity in experimental animals precede or are independent of this reduction (Pappius and M c C a n n 1969). As well as possibly affecting the formation, spread and resolution of vasogenic oedema, steroids may modulate neuronal function (Pappius 1980) or the growth of the tumour itself (Gurcay et al. 1971; Wilson et al. 1972; Shapiro and Posner 1974). It has been suggested that steroids m a y reduce the formation o f tumour-related cerebral oedema by lowering the abnormally high permeability of the vessels to polar solutes (Yamada et al. 1983). Recently,

292

P.J. Luthert et al.: Effects of Dexamethasone on Glioma Vascular Permeability

Table 3. Values for extravascular mannitol space (Se), first order rate constants (k), transfer factors and permeability coefficient for [14C]mannitol in control and dexamethasone-treated experimental gliomas

s,

k (min - 1)

Transfer factor

gm

SEM

gm

SEM

ml/min/g

SEM

Permeability coefficient (cm/s'10-5)

Control data (Deane et al. 1984)

0.217

0.010

0.300

0.050

0.0660

0.0140

1.26

Control data (Present study)

0.242

0.011

0.177

0.031

0.0428

0.0094

0.81

0.0536

0.0085

1.02

0.0534

0.0086

1.02

Pooled control data

0.229

0.008

0.234

0.029

Dexamethasone-treated

0.216

0.006

0.247

0.033

in an experimental glioma, the increased permeability of the vessels to mannitol, a substance normally excluded from the brain, has been quantified (Deane et al. 1984). The fact that this abnormality was not reversed by the administration of dexamethasone, despite striking clinical improvements, suggests that a reduction in vessel permeability is not the prime mode of action of dexamethasone in the treatment of cerebral turnouts. An advantage of using a relatively inert tracer, such as mannitol, when investigating the effect of steroids is that of avoiding possible modification of the binding, cell uptake or metabolism of the tracer (Kostyo 1965; Duval et al. 1983). Moreover, by following the increasing ratio of tissue to plasma radioactivity (Rt/Rp) at different time points it is possible to estimate not only the transfer factors, which can be obtained by single time point studies (Groothuis et al. 1984c), but also the fractional extracellular space. Even though a definite beneficial effect of dexamethasone was detected, the possibility must be considered that the steroids were acting on a different time scale than that of the permeability study. Steroids may act upon cellular metabolism either directly, or indirectly via the genome following binding to a cytoplasmic receptor (Duval et al. 1983). Direct effects are generally apparent within minutes, so as adequate plasma levels were detected at the time of the study and as plasma concentrations of steroids underestimate those within brain (Withrow and Woodbury 1972), gliomas (Yamada et al. 1979) and oedematous brain (Kostron and Fischer 1983), it is probable that the dosage was adequate if the mechanism of action was direct. However, genomic effects take longer to develop, but may continue for many hours following the disappearance of dexamethasone from the circulation (Duval et al. 1983). Therefore, the first two doses in this study should have initiated and maintained any genomic mechanism. Previous investigations have demonstrated effects of steroids upon the permeability of cerebral vessels

in a variety of models of vasogenic oedema (Rovit and Hagan 1968; Yamada et al. 1983). Some of these, such as cold injury (Pappius and McCann 1969), involve the production of considerable tissue damage. It is well-recognised that steroids inhibit the production of vasoactive metabolites of arachidonic acid (Wolfe 1982), and it may well be that the reduction of permeability noted under these circumstances is not brought about by a direct action upon endothelial cells. However, it has been suggested that dexamethasone does not affect the metabolism of arachidonic acid in the central nervous system (Pappius and Wolfe 1983). That tissue damage is an important consideration when investigating the possible modes of action of steroids is supported by a recent study (Groothuis et al. 1984a), in which a reduction in the transfer constant for c~-aminoisobutyric acid (AIB) was reported, particularly in the necrotic portions of the tumour. Although central portions of the tumour were sampled in this study, necrotic regions only amount to some 5% - 10% of the total central turnout mass in this model (Luthert and Lantos 1985). In contrast to our findings, Yamada et al. (1983), also using an experimental glioma model, have shown a reduction in permeability to AIB on treatment with methyl prednisolone. These findings may not be due to a reduction in vessel permeability but to steroid-induced alteration of AIB transport into tumour cells. Alternatively, they may be the consequence of using both a different turnout cell line and steroid and, possibly, the trauma produced by implanting a l-ram 3 piece of tumour tissue into the cerebral hemisphere. More recently, other studies have failed to show any steroid-mediated reduction in tumour vessel permeability to radioiodinated serum albumin (RISA) (Blasberg et al. 1984) or to horseradish peroxidase (Matsuoka and Hossmann 1981 ; Hossmann et al. 1983). It is possible that important changes were taking place at the periphery of the tumour and not in the central portions sampled, or that pentobarbitone anaesthesia may have interfered with some steroid-

P. J. Luthert et al.: Effects of Dexamethasone on Glioma Vascular Permeability mediated phenomena. In addition, the technique may have been too insensitive, but the rapidity and magnitude of the therapeutic effect might suggest that, if a permeability change was the prime underlying mechanism, it too should have been easily detectable. The effect of dexamethasone on tumour vessel permeability is not only of interest from the point of the mechanisms of oedema reduction and therapeutic effect. Estimates of transfer constants for water-soluble substances in both patients (Groothuis et al. 1984b) and experimental animals (Groothuis et al. 1984c), in conjunction with mathematical modelling (Levin et al. 1980), have demonstrated the significance of blood-tissue transfer as a limiting factor in drug delivery. The failure o f dexamethasone to decrease the permeability in our tumour model (Table 3) suggests that steroid treatment would not necessarily impair the transfer of water-soluble drugs into tumours. N o change in mannitol permeability was found in non-tumour areas (Table 2). The permeability of cerebral endothelium to mannitol is extremely low and of the same order of magnitude as that for lipid bi-layers (Ohno et al. 1978). A small change in permeability would not necessarily be detectable so close to the limits of the technique and generally steroids are associated with a change in permeability of normal membranes to water (Reid et al. 1983) and ions (Duval et al. 1983). However, it has been suggested that dexamethasone may reduce the permeability of normal mouse cerebral arterioles to horseradish peroxidase (Hedley-Whyte and Hau 1980). The weight loss seen in normal animals treated with dexamethasone (Fig. 1; Lowy and Yim 1980) might lead one to suppose that the turnout-bearing animals improved through a generalised dehydration and subsequent shrinkage o f the brain. This possibility is, however, unlikely as steroid-treated tumourbearing animals lost less weight than their salinetreated counterparts which rarely showed any signs o f improvement. The failure of plasma electrolytes and haematocrit to be altered significantly by dexamethasone is consistent with similar findings under slightly different circumstances (Renaudin et al. 1973) although changes in total body electrolytes have been demonstrated (Shenkin and Gutterman 1969). Little is known about steroid-induced oedema reduction within experimental or human gliomas (Gurcay et al. 1971), the emphasis having been on peritumoural white matter. However, if oedema reduction has taken place, the constancy of the extravascular mannitol space and hence the fractional extracellular space would suggest that any oedema reduction has affected both intra- and extra-cellular compartments equally. This supports the findings of

293

Herrmann et al. (1972) who found both compartments reduced in brain slices, but goes against the interpretation of the limited spread of RISA from an oedema forming focus resulting from a tightening of extracellular space secondary to a shift of fluid into the intracellular compartment (Blasberg et al. 1984). It also supports the suggestion that dexamethasone has actions remote from the vascular endothelium. In conclusion dexamethasone is not altering the permeability of glioma vasculature to mannitol in this model, despite a significant clinical improvement. Concern has been expressed that dexamethasone may impair drug delivery to gliomas (Neuwelt et al. 1982), but this study suggests that steroids are unlikely to effect drug delivery by reducing resting vascular permeability to polar substances.

Acknowledgement. The authors are deeply grateful to Miss J. Martin for culturing the neoplastic cells and to Dr. M. Moss and Dr. E. Houghton of Racecourse Security Services' Laboratories, P.O. Box 15, Newmarket, Suffolk (UK) for the plasma dexamethasone assay. References Bard Y (1974)Nonlinear parameter estimation. Academic Press, New York Blasberg RG, Kobayashi T, Patlak CS, Shinohara M, Miyoaka M, Rice JM, Shapiro WR (1981) Regional blood flow, capillary permeability, and glucose utilization in two brain tumor models: preliminary observations and pharmacokinetic implications. Cancer Treat Rep [Suppl 2] 65:3 - 12 Blasberg RG, Nakagawa H, Patlak CS, Groothuis DR (1984) Dexamethasone reduces tumor and brain extracellular space: effects on edema propagation. J Neuro-oncol 2:277 [Abstr 60] Davaki P, Lantos PL (1981) The development of brain tumours produced in rats by the intracerebral injection of neoplastic glial cells: a fine structural study. Neuropathol Appl Neurobiol 7:49 - 61 Deane BR, Greenwood J, Lantos PL, Pratt OE (1984) The vasculature of experimental brain tumours. IV. The quantification of vascular permeability. J Neurol Sci 65: 59- 68 Duvai D, Durant S, Homo-Delarche F (1983) Non-genomic effects of steroids. Interactions of steroid molecules with membrane structures and functions. Biochim Biophys Acta 737: 409 - 442 Groothuis D, Fisher J, Bigner D, Vick N (1984a) The effects of dexamethasone on blood flow and capillary permeability in an experimental brain tumor. J Neuro-oncol 2:274 [Abstr 45] Groothuis D, Vriesendorp F, Mikhael M, Blasberg R, Patlak C, Vick N (1984b) In vivo measurement of capillary permeability in brain tumors. J Neuro-oncol 2:273 [Abstr 43] Groothuis DR, Molnar P, Blasberg RG (1984c) Regional blood flow and blood-to-tissue transport in five brain tumor models. In: Homburger F (ed) Progress in experimental tumor research, vol 27. Karger, Basel, pp 132-153 Gurcay O, Wilson C, Barker M, Eliason J (1971) Corticosteroid effect on transplantable rat glioma. Arch Neurol 24:266269

294

P.J. Luthert et al. : Effects of Dexamethasone on Glioma Vascular Permeability

Hedley-Whyte ET, Hau DW (1980) Effect of dexamethasone on normal blood-brain barrier. Fed Proc 39: 774 Herrmann H-D, Neuenfeldt D, Dittmann J, Palleske H (1972) The influence of dexamethasone on water content, electrolytes, blood-brain barrier and glucose metabolism in cold injury edema. In: Reulen H J, Schurmann K (eds) Steroids and brain edema. Springer, Berlin Heidelberg New York, pp 7 7 - 85 Hossmann K-A, Hurter T, Oschlies U (1983) The effect ofdexamethasone on serum protein extravasation and edema development in experimental brain tumors of cat. Acta Neuropathol (Berl) 60: 223 - 231 Houghton E, Dumasia MC, Wellby JK (1981) The use of combined high performance liquid chromatography negative ion chemical ionization mass spectrometry to confirm the administration of synthetic corticosteroids to horses. Biomed Mass Spect 8 : 558 - 564 Kostron H, Fischer J (1983) Regional, cellular, and subcellular distribution of [aH]dexamethasone in rat brain edema. Surg Neurol 20: 48 - 54 Kostyo JL (1965) In vitro effects of adrenal steroid hormones on amino acid transport in muscle. Endocrinology 76: 6 0 4 613 Lantos PL, Roscoe JP, Skidmore CJ (1976) Studies of the morphology and tumorigenicity of experimental brain tumours in tissue culture. Br J Exp Pathol 57:95-104 Levin VA, Wright DC, Landahl HD, Patlak CS, Csejtey J (1980) In situ drug delivery. Br J Cancer [Suppl 4] 41:74-78 Lowy MT, Yim GKW (1980) Selective reduction by dexamethasone of stress-related hyperphagias. Life Sci 27:2553 - 2558 Luthert PJ, Lantos PL (1985) A morphometric study of the microvasculature of a rat glioma. Neuropathol Appl Neurobiol 11 :461 - 474 Matsuoka Y, Hossmann K-A (1981) Corticosteroid therapy of experimental turnout oedema. Neurosurg Rev 4:185 - 190 Neuwelt EA, Barnett PA, Bigner DD, Frenkel EP (1982) Effects of adrenal cortical steroids and osmotic blood-brain barrier opening on methotrexate delivery to gliomas in the rodent: The factor of the blood-brain barrier. Proc Natl Acad Sci USA 79 :4420- 4423 Ohno K, Pettigrew KD, Rapoport SI (1978) Lower limits of cerebrovascular permeability to nonelectrolytes in the conscious rat. Am J Physiol 235 :H299-I-I307 Pappius HM, McCann WP (1969) Effects of steroids on cerebral edema in cats. Arch Neurol 20: 2 0 7 - 216 Pappius HM (!980) Mapping of cerebral functional activity with radioactive deoxyglucose: Application in studies of traumatized brain. In: Cerv6s-Navarro J, Ferszt R (eds) Advances in neurology, vol 28. Raven Press, New York, pp 271 - 279

Pappius HM, Wolfe LS (1983) Functional disturbances in brain following injury. Search for underlying mechanisms. Neurochem Res 8 : 6 3 - 72 Pratt OE (1985) Continuous-injectionmethods for the measurement of flux across the blood-brain barrier: the steadystate, initial-rate method. In: Marks N, Rodnight R (eds) Research methods in neurochemistry, vol 6. Plenum Press, New York London, pp 1t 7 - 1 5 0 Reid AC, Teasdale GM, McCulloch J (1983) The effects of dexamethasone administration and withdrawal on water permeability across the blood-brain barrier. Ann Neurol 13:28-31 Renaudin J, Fewer D, Wilson CB, Boldrey EB, Calogero J, Enot KJ (1973) Dose dependency of Decadron in patients with partially excised brain tumors. J Neurosurg 39 : 302- 305 Rovit RL, Hagan R (1968) Steroids and cerebral edema: the effects of glucocorticoids on abnormal capillary permeability following cerebral injury in cats. J Neuropathol Exp Neurol 27: 2 7 7 - 299 Shapiro WR, Posner JB (1974) Corticosteroid hormones. Effects in an experimental brain turnour. Arch Neuro130:217- 221 Shenkin HA, Gutterman P (1969) The analysis of body water compartments in postoperative craniotomy patients. III. The effects of dexamethasone. J Neurosurg 31:400-407 Sisson WB, Oldendorf WH (1971) Brain distribution spaces of mannitol-3H, inulin-14C, and dextran-14C in the rat. Am J Physiol 221:214-217 Wilson CB, Barker M, Hoshino T, Oliver A, Downie R (1972) Steroid-induced inhibition of growth in glial tumors: a kinetic analysis. In: Reulen H J, Schurmann K (eds) Steroids and brain edema. Springer, Berlin Heidelberg New York, pp 95 - 100 Withrow CD, Woodbury DM (1972) Some aspects of the pharmacology of adrenal steroid and the central nervous system. In: Reulen H J, Schurmann K (eds) Steroids and brain edema. Springer, Berlin Heidelberg New York, pp 41-55 Wolfe LS (1982) Eicosanoids: Prostaglandins, thromboxanes, leukotrienes, and other derivatives of carbon-20 unsaturated fatty acids. J Neurochem 38:1 - 14 Yamada K, Bremer AM, West CR (1979) Effects of dexamethasone on tumor-induced brain edema and its distribution in the brain of monkeys. J Neurosurg 50:361-367 Yamada K, Ushio Y, Hayakawa T, Arita N, Yamada N, Mogami H (1983) Effects of methylprednisolone on peritumoral brain edema. J Neurosurg 59:612-619

Received July 5, 1985/Accepted October 7, 1985