[CANCER RESEARCH 48, 7102-7106. December 15, 1988)

Role of Hypovolemic Hemoconcentration in Dose-dependent Flow Decline

Observed in Murine Tumors after Intraperitoneal Administration ofGlucose or MannitolPeter W. Vaupel1 and Paul G. Okunieff

Department of Radiation Medicine, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts 02114

ABSTRACT

Responses of tumor microcirculation (RBC flux) to i.p. glucose ormannitol injections were studied in early generation isotransplants of aspontaneous C3Hf/Sed mouse fibrosarcoma (FSalI). RBC flux in superficial tumor microregions was assessed using laser Doppler flowmetry.After administration of glucose or mannitol (a nonmetabolized sugaralcohol), a dose-dependent reduction in laser Doppler flow, and a dose-dependent increase in systemic hematocrit occurred concurrently. Maximum flow reductions induced by i.p. glucose or mannitol were statisticallyindistinguishable for equal osmotic load. Maximum decreases in RBCflux for glucose or mannitol were 20 and 25% (1.25 mg/g i.p.), 42 and48% (2.5 mg/g ¡.p.),72 and 60% (5 mg/g i.p.), and 80 and 75% (10 mg/g i.p.), respectively. Maximum increases in systemic hematocrit rangedfrom 18% (1.25 mg/g glucose i.p.) to 33% (10 mg/g glucose i.p.).Examination of RBC count, blood hemoglobin concentration, and fluidaccumulation in the abdominal cavity after glucose or mannitol administration were all compatible with a significant shift of intravascular/extracellular water into the abdominal cavity with resultant systemichypovolemic hemoconcentration. RBC volume and mean hemoglobincontent of RBC remained unchanged with glucose loading. The datasuggest that reductions in laser Doppler flow are predominantly causedby hypovolemic hemoconcentration following i.p. administration of hy-perosmolar sugar solutions. Changes in laser Doppler flow due to specificglucose-mediated or glucose-related phenomena are probably of minorimportance in the murine tumor system investigated. Future studies onmurine tumors, examining for specific effects of glucose on metabolismand/or therapy, should not use i.p. administration of hyperosmolar solutions.

INTRODUCTION

Hyperglycemia has been shown to decrease blood flow inrodent tumors (1-9). The pathogenetic mechanisms which leadto vascular prestasis or even stasis in malignant tumors areprobably quite complex: the reduction of tumor blood flow afterglucose loading may be due to both systemic [e.g., reduction incardiac output (4), increase in blood viscosity (2), glucose-mediated RBC rigidity (9, 10), swelling of RBC (11)] and localeffects [e.g., decreased percentage of cardiac output to tumortissue (4, 9), increase in RBC rigidity during tumor passage ofthe erythrocytes due to an intensified tumor tissue acidosiscaused by lactic acid accumulation (7)]. In contrast, past studiesexamining the effect of glucose given i.v. failed to show anysignificant blood flow decline in DS - carcinosarcomas growings.c. or as a tissue-isolated preparation in Sprague-Dawley rats(12, 13). In the latter studies hyperglycemia was accompaniedby a slight hypervolemic hemodilution without any significantchanges in the arterial blood pressure. The tumor blood flowdecline observed after i.p. glucose (1, 2, 5) therefore may beassociated with a significant osmotic water shift from the vascular compartment into the abdominal cavity thus inducing ahypovolemic hemoconcentration.

Received 3/22/88; revised 6/2/88, 8/8/88; accepted 9/16/88.The costs of publication of this article were defrayed in part by the payment

of page charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

' To whom requests for reprints should be addressed.

The present investigation was designed to test the impact ofhypovolemic hemoconcentration on RBC flux in murine tumorsfollowing i.p. administration of various doses of glucose (ormannitol). The experimental data suggest that flow declines atthe tumor periphery in a murine fibrosarcoma during hyperglycemia are predominantly caused by hypovolemic hemoconcentration rather than by acidosis-induced intratumor RBC rigidityand/or glucose-mediated RBC rigidity.

MATERIALS AND METHODS

Animals and Tumors. Experimental animals were 10- to 12-week-oldC3Hf/Sed mice derived from our defined flora mouse colony (14).Animals were provided with sterilized animal pellets and acidified andvitamin K-fortified water ad libitum. Early generation isotransplants ofa poorly differentiated fibrosarcoma (FSalI) which arose spontaneouslyin a female C3Hf/Sed mouse were used. Single cell suspensions wereprepared by trypsinization and transplanted s.c. into the dorsum of theright hind foot. The tumor grows rapidly with a volume-doubling timeof 2 days at a volume of 100 to 200 mm3 and is very weakly immuno-

genic (15). All experiments were performed on conscious mice. Inaddition to FSall tumors, in a few experiments another spontaneouslyarisen murine tumor (mammary carcinoma MCalV isotransplantedinto the hind foot dorsum) or human tumor xenografts (U87 humanglioblastoma heterotransplanted into NCR-nu/nu mice) were used tolargely exclude cell line specified effects of glucose loading on tumorblood flow. Tumor volumes on the day of study were calculated by anellipsoid approximation using the 3 diameters ( V = ?r/6 x d\ x di xd,).

Laser Doppler Flowmetry. The LASERFLO Blood Perfusion Monitor 403A (TSI Inc., St. Paul, MN) was used for this study. Power atthe end of the optical fiber was 1.5 mW (wavelength, 780 nm). Thislaser Doppler flowmeter provides a stable, reproducible, and noninva-sive method for continuously monitoring of tissue perfusion in superficial tissue areas. LDF2 is integrated over a hemisphere of approximately 2 mm3 (16, 17). Laser Doppler flow signals were obtained from

central locations on a tumor. The fiberoptic probe was placed above(but not in contact with) the tumor tissue under study using a flexiblemetal stand. Hence, tissue compression was avoided and there was nodisturbance of the microcirculation due to fiberoptic probe manipulations.

The microprocessor of this flowmeter computes several variables(RBC flux, RBC velocity, and number of moving RBC) which wererecorded simultaneously on a multichannel chart recorder (type 6514;Linseis, Selb, West Germany). Data were expressed as relative unitswhich represent percentage values of full scale deflection on the instrument meter (18).

Laser Doppler Flowmetry was also performed in normal tissue (skinat hind foot dorsum). In these studies, RBC flux in the skin wasmeasured after i.p. injection of saline (0.5 ml) or glucose (10 mg/g).

Experimental Protocol. After placement on a Styrofoam pad, themice were immobilized and the tumor-bearing foot was taped to thepad to minimize movement artifacts during LDF measurements. A 22-gauge, 0.75-inch Teflon catheter (Angio-set, Deseret Medical, Sandy,UT) was placed i.p. for the administration of glucose, mannitol, orsaline. Once the RBC flux reached a constant baseline reading, theagent was administered via bolus injection. As a rule, 0.5 ml of aqueous

2The abbreviation used is: LDF, laser Doppler flow.

7102

Research. on September 8, 2018. © 1988 American Association for Cancercancerres.aacrjournals.org Downloaded from

LASER DOPPLER FLOW IN MURINE TUMORS DURING HYPERGLYCEMIA

glucose Solutions (6.25 to 50%) was given over a 2-min period yieldingdifferent i.p. glucose doses (1.25, 2.5, 5, and 10 mg/g). Mean tumorvolume in the glucose-treated animals was 239 ±19 mm3 (n = 50).

Mannitol, a sugar alcohol which is not metabolized, was injectedusing comparable doses and fluid volumes with the exception of thehighest dose (1.0 ml of a 25% mannitol solution at 10 mg/g). Forcontrol studies, 0.5 ml of a 0.9% NaCl solution (saline) was injectedrather than glucose or mannitol to examine for the effects of i.p.injection of isotonic fluid into the mice. Upon administration of thedifferent agents, LDF signals were recorded for 90 min. Mean tumorvolume in the mannitol-treated animals was 242 ±22 mm3 (n = 40).

Blood Glucose Concentration. In a separate set of experiments glucosewas injected i.p. using the above-described concentrations and volumes.Blood glucose concentrations were measured spectrophotometricallyusing a glucose oxidase test kit (Glucose Procedure No. 510; SigmaDiagnostics, St. Louis, MO). Blood samples (50 /<! microsamplingpipets) were withdrawn from the ophthalmic plexus by using tumor-free mice. Blood sugar level was measured before and 15, 30, 60, 90,and 180 min postinjection. For each glucose dose, blood glucoseconcentration was measured in 10 animals.

Modifications of Systemic I li-matomi after i.p. Glucose or Mannitol.

Quantitative data on the effect of different glucose doses i.p. on systemichematocrit were obtained in another separate series of experiments.Hematocrit values in blood withdrawn from the ophthalmic plexus (in20-Ml capillary tubes) of tumor-free animals were determined beforeand 15, 30, 60, and 90 min postinjection using a standard microhe-matocrit technique (capillary tube centrifugation). In some experimentshematocrit was additionally measured using an ELT-80 counter (OrthoDiagnostic Systems Div., Becton Dickinson, Braintree, MA).

Statistical Analyses. Values given in this study are means ±SE ifnot otherwise stated. The double tail / test and Scheffé'stest were used

to determine statistically significant differences.

RESULTS

Blood Glucose Concentrations during Hyperglycemia. Theaverage results obtained from the blood glucose measurementsare shown in Fig. 1 (each symbol represents the mean value ±SD of 10 animals at a given time). After i.p. glucose injection,the blood glucose concentration peaked at t = 15 min postinjection in a dose-dependent manner. Blood glucose level thendecreased to the baseline in 30 min at 1.25 and 2.5 mg/g i.p.

50 r

40

30

20

10

\ GLUCOSE\ DOSE

\ (mg/g i.p.)

ÃŽ1.25'"I

20 40 60

TIME (mini

80 100 180

in approximately 90 min at 5 mg/h, and after 180 to 360 minat 10 mg/g.3 No significant changes in blood glucose levels

were obtained after bolus injection of saline (0.5 ml i.p.).Changes in Hematocrit after i.p. Glucose Loading. The average

results obtained from the hematocrit determinations are shownin Fig. 2 (each symbol represents the mean ±SE of 10 animalsat a given time). Following i.p. glucose injection, there aresignificant increases in the blood hematocrit values at all dosesused. These results clearly indicate that a dose-dependent hy-povolemic hemoconcentration was present during the totalduration of the LDF experiments. Following i.p. injection ofmannitol, there are significant increases in blood hematocritvalues comparable to those obtained with hyperosmolar glucosesolutions. For example, maximum hematocrit was 58.0 ±0.8%upon glucose and 56.5 ±0.4% after mannitol (at doses of 5mg/g i.p.).

Modifications in Laser Doppler Flow after Glucose Loading.Whereas no significant change in RBC fluxes were observedafter i.p. administration of saline (0.5 ml), glucose at dosess* 2.5 mg/g i.p. were followed by significant reductions in LDF.As a rule the drop in LDF was due to both a decrease of themean velocity of RBCs and, to a somewhat lesser extent, adecrease of the number of moving RBC. The higher the dosechosen, the steeper was the flow decline and more pronouncedthe inhibition of the RBC flux (see Fig. 3). In general, the dose-dependent flow reduction reached maximum values between 50min (2.5 mg/g i.p.), and 75 min (10 mg/g i.p.). After 1.25 mg/g i.p. glucose, LDF tended to decrease slightly, reached a nadirvalue of 0.8 at i = 50 min, and then came up to baseline valuesafter 75 min (see Fig. 5).

To test whether there are any significant differences in LDFbetween small (tumor volume, 76 ±7 mm3) and larger FSalltumors (tumor volume, 300 ±33 mm3), flux measurements

were performed at 5 mg/g glucose i.p. From Fig. 4 there isclear evidence that changes in RBC flux are identical in super-

GLUCOSEDOSE

(mg/g i.p.)

I1

62

se

54

50

46

44

10

A {, V, ._.-.-/ .f- -L --... *//l '-•-..L

II I I SA. •¿�(ft

20 40 60 80 100

TIME (min)Fig. 2. Averagehematocritvaluesversustime after i.p. glucoseinjection;bars,

SE; \ I(Imicefor each glucosedoseat each time intervalafter glucoseloading.Hematocrit values did not change after saline injection (0.5 ml i.p.) in control

Fig. 1. Averageinjection;bars, SD;glucoseloading.

blood glucose concentration versus time after i.p. glucose experiments.N = \0 mice for eachglucosedoseat eachtime intervalafter

3M. Urano, unpublishedresults.

7103

Research. on September 8, 2018. © 1988 American Association for Cancercancerres.aacrjournals.org Downloaded from

LASER DOPPLER FLOW IN MURINE TUMORS DURING HYPERGLYCEMIA

1.5

1.0

0.8

06

5!

0.3

0.2

0.1L

SALINE(0.5ml)

(10)

GLUCOSEDOSE

(mg/g ¡.p.)

1.25

•¿�5

ñ O 10

20 40 60 80 100

TIME (min)

Fig. 3. Laser Doppier flow in murine FSall tumors versus time after i.p.glucose loading at varying doses (1.25 to 10 mg/g). Time- and dose-dependentI I»Ideclines take place with glucose doses ¿--2.5mg/g. No significant changesin LDF are observed after saline (0.5 ml). Following 1.25 mg/g i.p. glucose, flowtends to decrease and then recovers reaching baseline values at the end of theobservation period. Numbers in parentheses, number of tumors investigated.

1.5

1.0

0.8

0.6

0.4

Control (soline)

0.2O.1(I

T T "T TNs10Glucose(5mg/gip.)

o VT = 300 ±33 mm3

•¿�VT= 76± 7mm3s±s.E.M.)

2040 60 80 100

TIME (min)

Fig. 4. Laser Doppler flow in superficial tissue areas of small and large FSalltumors versus time after glucose loading (5 mg/g i.p.). Values are means ±SE(bars); N, number of tumors investigated. Means and 95% confidence limits (r./.)are presented for control experiments (saline, 0.5 ml i.p.; 10 tumors).

flcial tissue areas of large and small murine tumors.When the maximum LDF inhibitions following glucose load

ing at different i.p. doses are compared for statistical analysis,P< 0.01 with the exception of 1.25 mg/g versus saline, 2.5 mg/g versus 1.25 mg/g, and 10.0 mg/g versus 5.0 mg/g. The changesdescribed are not cell line specific since LDF measurements onMCalV murine tumors and human tumor xenografts in NCR-nn/nu mice showed comparable results. RBC fluxes in normalskin marginally increased after i.p. administration of saline (0.5ml, n = 5). Glucose at doses that induced a significant drop ofLDF in tumors also led to a shutdown of skin microcirculation.Maximum flow reduction in skin was comparable to that intumors (e.g., 83 versus 80% at 10 mg/g glucose i.p.), indicatingthat a glucose-mediated systemic cardiovascular effect rather

than a glucose-specific mechanism on tumor microcirculationis the prevailing event underlying the observed flow changes.

Modifications in Laser Doppler Flow after Mannitol Loading.As was the case with glucose loading, after mannitol administration at doses 5*2.5 mg/g i.p. RBC fluxes were reduced in atime- and dose-dependent manner (see Fig. 5). Here again, a

positive correlation between the dose administered and thesteepness of the flow decline as well the maximum flow inhibition was observed. As a rule, nadir flow values occurredbetween 50 min (2.5 and 10 mg/g i.p.) and 60 min (5 mg/g i.p.)

When the maximum flow inhibition upon i.p. mannitol injection at different doses are compared for statistical analysis,P< 0.01 with the exception of 1.25 mg/g versus saline, 5.0 mg/g versus 2.5 mg/g, and 10.0 mg/g versus 5.0 mg/g. Comparingmaximum LDF reductions upon i.p. glucose loading with thoseobserved following i.p. mannitol there is clear evidence that theeffects at the same doses (similar osmotic load) are comparable(P = 0.81). This indicates that the flow inhibition at tumor

periphery in FSall tumors does not preferentially result fromglucose-related mechanisms, such as intratumor lactacidosis(low pH), RBC rigidity, or glucose-mediated RBC rigidity, butrather from (systemic) hypovolemic hemoconcentration. Similar to the maximum hematocrit adjusted after i.p. glucoseloading, the maximum LDF inhibition is dose dependent forboth agents (see Fig. 6).

Intertumor Variability. In all experimental series, pronouncedintertumor variability in LDF inhibition following i.p. glucoseor mannitol was a common finding. For example, maximumLDF reduction varied between 37 and 87% in FSall tumorsfollowing glucose administration at a dose of 5 mg/g i.p. NadirRBC fluxes were observed 30 to 90 min postinjection. Noclearcut correlation was found between the extent of the flowdeclines and tumor size or baseline RBC flux before administration of the agent.

li-J

1.0 r

0.8 -

0.6-

0.4

0.3

0.2

0.1

MANNITOLDOSE

(mg/g i.p.)

1.25

2.5

~0~— HOIS o -o -o 10

20 40 60 80 100

TIME (min)Fig. 5. Laser Doppler flow in murine FSall tumors versus time after i.p.

mannitol loading at varying doses (1.25 to 10 mg/g). Time- and dose-dependentLDF declines occur after mannitol loading »2.5 mg/g. No significant LDFdecrease is observed after 1.25 mg/g. Numbers in parentheses, number of tumorsinvestigated.

7104

Research. on September 8, 2018. © 1988 American Association for Cancercancerres.aacrjournals.org Downloaded from

LASER DOPPLER FLOW IN MURINE TUMORS DURING HYPERGLYCEMIA

1101

i GLUCOSE^_~-—* MANNITOL

(1OI

10

DOSE (mg/g i.p.)

Fig. 6. Maximum LDF reduction observed after i.p. injection of glucose ormannitnl as a function of dose [means ±SE (bars)]. The injection volume wasalways 0.5 ml of aqueous solutions with the exception of mannitol at 10 mg/g(1.0 ml). Numbers in parentheses, number of FSall tumors investigated.

DISCUSSION

Methods. In the present paper, flow changes in murine fibro-sarcomas following i.p. glucose or mannitol loading at differentdoses could continuously and noninvasively be monitored bylaser Doppler flowmetry. LDF has been proposed as a usefuland valid method for measuring microcirculatory function insmall, discrete tissue areas (for reviews see Refs. 19-22). Themeasured flux predominantly represents the RBC flux throughthe microvasculature from the arterial to the venous side. It isdefined as the number of RBC (= volume fraction of RBC)moving in the measured volume times the mean RBC velocity.RBC flux is integrated over all the vessels running within theilluminated volume without regard of flow directions. In general, LDF provides a good estimate of tissue blood flow asevidenced by using electromagnetic flowmeters (20), radiola-beled microspheres (20), isolated perfused organ preparations(16), Krypton-85 clearance technique (18), plethysmography(22), and [MC]iodoantipyrine (23). Furthermore, there was a

significant and linear relationship between LDF and local microcirculatory flow assessed by well known techniques such asH2 clearance (21) and video microscopy (17). There is evidencethat RBC flux in superficial tumor areas measured with laserDoppler flowmetry can reflect blood flow changes in tumorsmonitored with the Krypton-85 clearance technique (18).

The major limitations of the laser Doppler flowmetry usedare that: (a) it measures only RBC fluxes in superficial tissueareas; (b) a small portion of the Doppler beat notes can sometimes be seen although the blood flow is stopped, most probablydue to cellular or intracellular movements or to extracellularbulk flow (24); (c) it is quite sensitive to movement artifacts;(d) it is generally difficult to calibrate LDF in absolute terms(21) because there appears to be no universal calibration factor(20); and (e) there are certain limitations of the LDF techniquewhen assessing heterogeneous tumor blood flow, i.e., one localmeasurement on the tumor surface is not necessarily a sufficientbasis for conclusions about the whole tumor.

In our study the tumor replaced the subcutis and the coriumcompletely and invariably reached the avascular epidermis. Thetissue overlaying the tumors had an average thickness of 34 ±0.3 (SE) ¡im.Due to this small thickness, a removal of the

avascular epidermis to measure blood flow in the s.c. tumorswas not necessary and thus any possible disturbance of theblood supply to the tumor could be avoided.

Results. A number of investigators have shown that hypergly-cemia leads to a reduction in tumor blood flow and pH (for arecent publication reviewing the literature available, see Ref.9), a reduction of the fraction of clonogenic cells (25), and anincrease in size of the acutely and chronically hypoxic cellfractions (26). These observations have led to the use of combined hyperthermia and hyperglycemia in the treatment ofcancer (for a review see Ref. 10). Changes in tumor microcirculation, in hemorrheological properties, in cardiac output, andin the rate of aerobic and anaerobic glycolysis have been shownto be involved in the pathophysiological changes observed intumors after glucose loading.

The experimental results presented in this paper unequivocally show that the dose-dependent flow reduction in a murinetumor system is mostly due to a hypovolemic hemoconcentra-tion following i.p. administration of hyperosmolar glucose ormannitol solutions. No significant specifically glucose-mediated or glucose-related changes were observed in the murinefibrosarcomas investigated. The i.p. administration of glucose(or mannitol) is associated with a significant osmotic water shiftfrom the vascular (extracellular) compartment into the abdominal cavity. In 6 animals the total fluid volume accumulatedwithin 45 min after glucose loading at 10 mg/g i.p. was approximately 1.8 ml. Taking into account the volume of theinjected glucose solution (0.5 ml), -1.3 ml of fluid must havebeen transferred from the extracellular space into the abdominalcavity within that time period. Assuming a total blood volumeof 8% of body weight (27) and a volume of the extracellularspace of 20-21% of body weight (28, 29), it is concluded that~25% of the total extracellular fluid volume is shifted into theabdominal cavity (in these estimates an intracellular-extracel-lular water transfer is neglected). It has to be noted that thiswater shift (25% of the extracellular fluid volume) is accompanied by a 25% increase in RBC (9.55 x IO6 versus 7.65 x

lOV^l) and in hemoglobin concentration (165 versus 121 g/liter) and a 29% increase in hematocrit (58 versus 45%; measurements of RBC parameters were performed with an ELT-80counter). Mean RBC volume (61 versus 59 fl), mean RBChemoglobin content (17.7 versus 17.1 pg), and mean RBChemoglobin concentration (286 versus 288 g/liter) were notaffected by this fluid shift indicating that the increase in hematocrit was not due to cell swelling following glucose loading.

Possible mechanisms responsible for the flow reduction afteri.p. administration of hypertonic glucose (or mannitol) solutions are summarized in Fig. 7. When glucose is given i.V., thepathogenetic mechanisms following the osmotic water shiftfrom the intravascular space into the abdominal cavity and theperitoneal irritation are missing. In consequence, the effect ofglucose i.v. on tumor blood flow should be less pronouncedand/or delayed. Furthermore, the time course of hyperglycemia(i.e., peak concentrations immediately after i.v. bolus injectionversus a somewhat delayed increase in the blood glucose concentration following i.p. injection) may also play a role in thiscontext.

The fluid accumulation observed in the abdominal cavityafter glucose (or mannitol) loading not only leads to a reductionof the circulating blood volume but most probably also leads toother circulatory disturbances. For example, the abdominalfluid accumulation causes a rise of the intraabdominal pressurewhich may impair venous return. As a consequence, a reductionin cardiac output and a redistribution of the circulating blood

7105

Research. on September 8, 2018. © 1988 American Association for Cancercancerres.aacrjournals.org Downloaded from

LASER DOPPLER FLOW IN MURINE TUMORS DURING HYPERGLYCEMIA

,^j Hypovolemio |•¿�Osmotic

Water Shift (QJ"V)HemoconcentrationRBC

Flexibility ( ],Tissue

pH)Lactic

Acid t|*Tumor

Glycolysis 1^~"

ViscosityTUMOR•^~r;Cordioc

iOutput'Blood

,Pressure'BLOOD

FLOWI1

| VenousDrainage I

Fig. 7. Pathogenetic mechanisms possibly responsible for the blood flowreduction in tumors following i.p. administration of hyperosmolar glucose solutions.

volume (due to peripheral vasoconstriction) may result. Furthermore, a decrease in the mean arterial blood pressure and incardiac output may also result from a reflex bradycardia following vagai stimulation after i.p. administration of hypertonicglucose (6.25 to 50%) or mannitol (6.25 to 25%) solutions,since stimulation of the peritoneum (and of the viscera) oftenevokes a depressor response (30).

According to the data presented, cell line-related effects ofglucose (or mannitol) loading on tumor blood flow can largelybe excluded. Whether the tumor growth site chosen affects themagnitude of the flow changes was not systematically studied.However, from pilot studies using s.c. tumors at the thigh thereis evidence that our observations generally hold true for s.c.tumors. Neoplasms implanted into tissues other than the sub-cutis are expected to exhibit flow reductions upon glucoseloading, too, although the degree of changes might be differentfrom those observed with s.c. tumors.

In conclusion, i.p. administration of hypertonic glucose (ormannitol) solutions lead to a dose-dependent flow inhibition insuperficial tissue areas of s.c. murine tumors due to hypovo-lemic hemoconcentration. In deeper tissue areas with increasingacidosis, hypoxia, and ischemia, other factors may also contribute to the decline of total tumor blood flow. The changingmetabolic milieu of the tumor center may produce hyperglycémieblood flow responses different from those measured atthe tumor periphery (for a recent compilation of mechanismspossibly involved see Ref. 31). However, since perfusion pressure is probably the most important parameter governing bloodflow through tumors, hypovolemic hemoconcentration (andthus hypotension) should also play a substantial role in the flowchanges observed in central tumor areas after i.p. administration of hypertonic glucose or mannitol solutions. The observeddecrease in flow is caused by a tremendous increase in wholeblood viscosity and most probably by a drop in the perfusionpressure. Hence, when glucose-specific effects on blood flowthrough murine tumors are to be evaluated, i.p. administrationof hypertonic glucose solutions must be avoided.

REFERENCES

1. Algire, G., and Legallais. F. Vascular reactions of normal and malignanttissues in vivo. J. Nati. Cancer Inst., 12: 399-408. 1951.

2. Calderwood, S., and Dickson. }. Effect of hyperglycemia on blood flow, pH,

and response to hyperthermia (42") of the Yoshida sarcoma in the rat. CancerRes., 40:4728-4733, 1980.

3. Calderwood, S., and Dickson, J. Inhibition of tumor blood flow at high bloodsugar levels: effects of tumor pH and hyperthermia. Nail. Cancer Inst.Monogr.,6/: 221-223, 1982.

4. DiPette, D. }., Ward-Hartley, K. A., and Jain, R. K. Effect of glucose onsystemic hemodynamics and blood flow rate in normal and tumor tissues inrats. Cancer Res., 46:6299-6304, 1986.

5. Jain, R. K., Shah, S. A., and Finney, P. L. Continuous non-invasive monitoring of pH and temperature in rat Walker 2S6 carcinoma during normo-glycemia and hyperglycemia. J. Nati. Cancer Inst., 73:429-436, 1984.

6. Osinsky, S. P., Sidorenko, M. V., and Nikolaev, V. G. Local blood flow inthe tumour during induced hyperglycemia. Exp. Oncol., 7: 51-53, 1985.

7. von Ardenne, M., and Reitnauer, P. G. Selective occlusion of cancer tissuecapillaries as the central mechanism of the cancer multistep therapy. Jpn. J.Clin. Oncol., 10: 31-48, 1980.

8. von Ardenne, M., and Reitnauer, P. G. Die manipulierte selektive Hemmungder MikroZirkulation im Krebsgewebe. J. Cáncer Res. Clin. Oncol., 103:269-279, 1982.

9. Ward-Hartley, K. A., and Jain, R. K. Effect of glucose and galactose onmicrocirculatory flow in normal and neoplastic tissues in rabbits. CancerRes.. 47:371-377, 1987.

10. Traykov, T. T., and Jain, R. K. Effect of glucose and galactose on red bloodcell membrane deformability. International Journal of Microcirculation —¿�Clinical and Experimental, 6: 35-44, 1987.

11. von Ardenne, M., and Reitnauer, P. G. Die übersäuerungsbedingte,durchSchwellung der Erythrozyten verursachte Venolenaufhellung im Bereich derMikroZirkulation. Folia Haematol. (Leipz.), 114: 273-281, 1987.

12. Vaupel, P., Guenther, H., and Grote, J. Einfluss einer Hyperglykämieaufdie Sauerstoff- und Glucoseaufnahme von Implantationstumoren (DS-Car-cinosarkom) in vivo. Z. Krebsforsch, 77: 17-25, 1972.

13. Vaupel, P., Okunieff, P., and Kluge, M. Response of tumor red blood cellflux to hyperthermia and/or hyperglycemia. Int. J. Hyperthermia, in press,1989.

14. Sedlacek, R. S., and Mason, K. S. A simple and inexpensive method formaintaining a defined flora mouse colony. Lab. Animal Sci., 27: 667-670,1977.

15. Suit, H. D., Sedlacek, R. S., Silver, G., and Dosoretz, D. Pentobarbitalanesthesia and the response of tumor and normal tissue in the C3Hf/Sedmouse to radiation. Radiât.Res., 104: 47-65, 1985.

16. Shepherd, A. P., Riedel, G. L., Kiel, J. W., Haumschild, D. J., and Maxwell,L. C. Evaluation of an infrared laser-Doppler blood flowmeter. Am. J.Physiol., 252: G832-G839, 1987.

17. Tyml, K., and Ellis, C. G. Simultaneous assessment of red cell perfusion inskeletal muscle by laser Doppler flowmetry and video microscopy. Int. J.Microcirc. Clin. Exp., 4: 397-406, 1985.

18. Vaupel, P., Kluge, M., and Ambroz, M. C. Laser Doppler flowmetry insubepidermal tumours and in normal skin of rats during localized ultrasoundhyperthermia. Int. J. Hyperthermia, 4: 307-321, 1988.

19. Wilkin, J. K. Periodic cutaneous blood flow during postocclusive reactivehyperemia. Am. J. Physiol., 250: H765-H768, 1986.

20. Smits, G. J., Roman, R. J., and Lombard, J. H. Evaluation of laser-Dopplerflowmetry as a measure of tissue blood flow. J. Appi. Physiol., 61:666-672,1986.

21. Diresta, G. R., Kiel, J. W., Riedel, G. L., Kaplan, P., and Shepherd, A. P.Hybrid blood flow probe for simultaneous II clearance and laser-Dopplervelocimetry. Am. J. Physiol., 253: G573-G581, 1987.

22. Johnson, J. M., Taylor, W. F., Shepherd, A. P., and Park, M. D. Laser-Doppler measurement of skin blood flow: comparison with plethysmography.J. Appi. Physiol., 5«:798-803, 1984.

23. Rundqvist, I., Smith, Q. R., Michel, M. E., Ask, P., Oberg, P. A., andRapoport, S. I. Sciatic nerve blood flow measured by laser Doppler flowmetryand (MC) iodoantipyrine. Am. J. Physiol., 248: H311-H317, 1985.

24. Tenland, T. On laser Doppler flowmetry. Methods and microvascular applications. Linkóping Studies in Science and Technology Dissertations No. 83,LinköpingUniversity Medical Dissertations No. 136, 1982.

25. Jaehde, E., and Rajewsky, M. F. Tumor-selective modification of cellularmicroenvironment in vivo: effect of glucose infusion on the pH in normaland malignant rat tissues. Cancer Res., 42: 1505-1512, 1982.

26. Urano, M., Todoroki, T., Kahn, J., and Okunieff, P. The effect of hyperglycemia on the tumor response to irradiation given alone or in combinationwith hyperthermia. Radiât.Res., ///: 488-497, 1987.

27. Schroeder, L. Morphologische und physiologische Daten der Haus- undLaboratoriumstiere. Leipzig: Hirzel-Verlag, 1986.

28. Thews, G., and Vaupel, P. Autonomie Functions in Human Physiology.Berlin: Springer-Verlag, 1985.

29. Valtin. H. Renal Function, Ed. 2. Boston: Little, Brown & Co., 1983.30. Berne, R. M., and Levy, M. N. Cardiovascular Physiology, Ed. 5. St. Louis:

C. V. Mosby, 1986.31. Sevick, E. M., and Jain, R. K. Blood flow and venous pH of tissue-isolated

Walker 256 carcinoma during hyperglycemia. Cancer Res., 48: 1201-1207,1988.

7106

Research. on September 8, 2018. © 1988 American Association for Cancercancerres.aacrjournals.org Downloaded from

1988;48:7102-7106. Cancer Res   Peter W. Vaupel and Paul G. Okunieff  Administration of Glucose or MannitolFlow Decline Observed in Murine Tumors after Intraperitoneal Role of Hypovolemic Hemoconcentration in Dose-dependent

  Updated version

  http://cancerres.aacrjournals.org/content/48/24_Part_1/7102

Access the most recent version of this article at:

   

   

   

  E-mail alerts related to this article or journal.Sign up to receive free email-alerts

  Subscriptions

Reprints and

  .pubs@aacr.orgDepartment at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

  Permissions

  Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/48/24_Part_1/7102To request permission to re-use all or part of this article, use this link

Research. on September 8, 2018. © 1988 American Association for Cancercancerres.aacrjournals.org Downloaded from