Lyons, T. P., and M. L. Riedesel.hyperhydration: its effectsLife Sciences 53: L719'L787.

Life Sciences, Vol. 53, pp.1779-t787Printed in the USA

L993. GlYcerol-inducedon fluid comPartments in the rat '

Pergamon Press

GIJYCEROIJ-INDUCED HYPERHYDRATION: ITS EFFECTSON FIJUID COMPARTMENTS IN THE RAT

T.P. Lyons and M.L. Riedesel

Department of Biology, University of New MexicoAlbuquerque, New Mexico 87131-

(Received in final form October 4,1993)

Sufiunary

Glycerol solutions are a safe, effective means toachieve a long-term state of hyperhydration. The effectsof glycerol-induced hyperhydration (GIH) on the total bodywaLer (TBW) , extracel-l-ul-ar fluid (ECF), plasma vo]ume(PV), intracellular fluid (ICF), and interstitial- flui-d(ISF) were investigated in the rat. Ten rats wereintragastrically adminisLered either 20 mI/kg of water or5? glycerol. The TBW, ECF, and PV were determined by 3HrO,14C-inul-in and dye diluLion, respectively at 2 h post-ingestion. Urine volumes and fluid retention weremeasured throughout the experimenL. Fluid retention wasincreased by 50% wiLh t.he 5Z glycerol solution whencompared to water. When the fluid compartment data at. 2h are expressed in terms of percent of body weight, theTBW and ICF of the glycerol-treated rats exceeds that ofthe water-treated rats, and the values for t.he ECF, PV,and ISF were slmilar after the glycerol solution and waterregimens. Glycerof solutions appear to have a greatereffect in expanding TBW and ICF Lhan equal volumes ofwater.

The concept of increasing total body waLer (TBW) aboveeuhydrated levels, which wifl be defined as hyperhydration, hasreceived some attention in the fiterature (6, 7, 13, 15). Themajority of these studies are concerned with cardiovascul-ar andthermoregulatory responses to different hydration states and fail- toprovide adequate data on fluid balance, comparLmenLs, ordistribution (6,7, L3, l-5). A common approach to hyperhydrationhas been the ingest.ion of large volumes of fluid. However, thistechnique has limitations and has not been overly successful (6, 7,13, 15). The primary problem being that volumes of fluid in excessof euhydrat.ion tend to be rapidly excret.ed by the renal system,thus, the expansion of TBW by most oral- vol-ume loading procedures isinsignificant and short in durat-ion. However, our laboratory hasreported that fluid retenLion is enhanced and subsequently a stateof hyperhydration is achieved if glycerol solutions are ingested inplace of water or diluLe (0.18) saline (10, 11, 19). This glycerol-induced hyperhydration (GIH) resulted in a hyperosmoLic serum, butit did not effect other blood chemistries (10, 19). The GIH alsoproved to be a physiological advantage during exercise in the heatCPT Timothy Lyons, USARIEM, 10 Kansas St, Natick MA, 01760-5007

0024-320s/93 $6.00 + .00

Copyright o 1993 Pergamon Press Ltd All rights reserved

1780 Glycerol Solutions Vol. 53, No.23, 193

by increasing sweatelevations (10).

rates and attenuating core temperature

The ability to achieve a prolonged hyperhydration withoutdisturbance to body homeostasis could prove to be beneficial in manysj-tuations of fluid imbalance. ft could have the capability toincrease one's performance and tolerance to adverse conditions suchas heat stress, cofd environments, hypoxia, and micrograviLy. Amore direct application to the cl-inical setting would involveinducing hyperhydration prior to the initiation of procedures whichmay disturb fluid balance. rn addition, solutions which act ashydrating agents also have the potential to reso1ve the commonmedical dilemma of rehydration.

The present study was conducted in order. to determine thevolume expansion of each fluid compartmenL (TBW, ECF, ICF, PV, ISF)after volume loading regimens with equal volumes of waLer or 5?gIycerol. The hypothesis was that in order to maintain osmoticequilibrium the dlstribution of the glycerol would dictate thedist.ribution of the excess fluid during GIH.

Methods

Ten male adult Sprague-Dawley rats (250-350 g) were obtainedfrom the New Mexico BioJ-ogy Dept. Animal Resource Facility.Protocols were approved by the UniversiLy of New Mexico Animal Careand Use Committee and the NIH Animal Care and Use guidelines werestrictly adhered to.

Protocol: The rats were surgically implanted with chronic centralvenous and arterial cannulae by an aseptic technique and allowed twodays to recover from the surgery. Prior to the starL of theexperiment the animals were fasted for 12 h and allowed ad libi-Lumwater int.ake. A11 food and waLer was withheld durinq the experiment.itself, excepL for fluids required by experimental design. Animalswere administered equal vol-umes (20 ml/kg) of either waLer or a 58glycerol sol-ution (1 g/kgr body weight) by gastric intubation witha 3" feeding needfe. Five of the rats received water and the otherfive the 5? glycerol solution at time zero. Fol-lowing f luidingestion the animals were immediately placed in plastic metabol-iccaqes where they remained throughout the experiment. Fluidcompartments were measured 2 h post-ingestion by the methodsdescribed be1ow.

Fl-uid Compart.ment Measurement Prot.ocol: The method for determiningfluid compartment volumes was similar to that reported by Durkot eLaf. (l-987). This technique was selected because it allowedsimultaneous analysis of t.he TBW, ECF, ICF, PV and ISF. The venouscannula was util-ized as a injection site and the arterial cannu1a asa blood sampling site throughout. A limited volume of fluid (1.5m1) was both injected and removed in order to minimize disturbancesof the fluid compartments.

The TBW was measured by injecting 800 nCi of 3HrO (NewEngland Nuclear) in 0.2 ml physiological saline via the centralvenous cannula 30 mi-n after fluid ingestion. An equil-ibrium time of90 min was all-owed at which time a 0.2 ml blood sample was obtainedfrom t.he arterial cannula, and approximately 750 nCi of 14C-inulin

Vol.53, No. 23, 1993 Glycerol Solutions

(New England Nuclear) in 0.2 ml saline and 2 mg of t.ricarbocyaninedye (Indocyanine green, Becton lickinson) in 1 ml of saline wasinjected. The 14C-inulin was utilized in determining ECF by Laking0.15 ml blood samples at 2, 4, 6,30, 60 and 90 min after inject.ionin order to calculate decay characteristics. PV was measured bydetermining the decay characteristics of the dye with 0.2 ml bl_oodsamples t.aken at 3, 5, and 7 min after injection. The TCF and ISFvolumes were calculated mathematically. A detail_ed description ofthe calculations are provided by Durkot et al. (1989).

Analvtic Methods: Blood samples were col-l-ected in heparinized tubes,centrifuged at 2500 RPM for 10 min, and refrigerated. The plasmawas analyzed for 3HrO and 14c-inufin with a liquidscintillation bet.a count.er (Tri-Carb series, Model 1500, packardInstrument Co. ) utilizing a standard program for dual labeling.Tricarbocyanine dye concentration was deLermined by measuring plasmaabsorbance at 800 nm with a Beckman DU 64 Spectrophotometer. Agravimetric method was utilized to determine urine volume. Thefl-oor of the cages were lined with absorbenL paper liners, whichwere weighed to the nearest 0.001 g prior to the experiment. Theliners were reweighed and replaced aL 30 min intervals. Thedifference in weight, after accounting for fecal- material-, wasrecorded as urine output. The urine output was accumulated andutil-ized in conjunction with the fluid intake to calculate overallfluid balance at each hour.

Stat,istical Analvsis: The data is presented as means + sLandarderrors. One-way ANOVA was utifized to determine t.he effect of thefluid regimens on fluid compartment volumes. Differences in urinevolume and fluid retention were anal-yzed by two-way Anova with arepeated measures model. A standard Tukey's post hoc Lest wasapplied to further characterize these differences. A p<0.05resul-ted in rejection of the null hypot.hesis.

ResultsThe mean weights and ingesLed fluid volumes were similar among

the t.wo groups (Table 1) .

TABLE IBody weiqht and ffuid compartment volumes.

WATER (20 ml/ks) 5? GLYCEROL (20 m]/kg)n=5 n=5

Weight, gFluid Volume, m1

TBW, mlECF, mlPV, m1

ICF, mlISF. ml 45-B t 1 .8

362.3 + 8.17'7+O?

245.L + 4.563.3 + 2.016.4 + 0.2

181.8 + 2.8

356.8 + 6.27.t + 0.2

248.3 + 6.360.6 + 2-B15.6 + 0.5

LBt .7 + 3.744.9 + )..a

Fluid comparLment analysis at 2-h post-ingestion. Datapresented as means t SE.

Mean accumul-ated urine volume and mean volume of fluid retained arepresented in Figures 1 and 2. The 5? glycerol sofution resulted inreductions in accumulaLed urine outpuL at 1, 2 and 3 h when comparedto ingestion of 20 ml/kg water (p<0.01) (Fig. 1). A 508 decrease was

L782 Glycerol Solutions Vol.53, No.23, 1993

observed after 2 h (nig. 1). The attenuated urine outputcontributed to an increased fluid retention with the glycerolsolution over that of water (Fig. 2). Differences j-n retention were1.9 mI at t h, 2 mI aL 2 h, and 2.4 ml at 3 h (p<0.01)(Fig. 2).

The effects of ingestion of water (20 ml/kg) and a 58 glycerolsol-ution (1 g/kg) on the fluid spaces are presented in Tables 1 and2. The absolute volumes of fluid compartments at 2 h were notdifferent between the two experimental groups (Table 1-). If these

I

rrME (h)

FIG ]-

Accumul-ated urine vofume. Inqestion of 20 m1/kg ofwater or 5? glycerol solution (L g/kg) at time zero.Significance between waLer and glycerol at t h, 2 h,and 3 h, p<0.01

daLa are cafcul-ated as a percent of the t.otal body weight, there areincreases j-n TBW and ICF with the ingestion of glycerol whencompared to the same volume of water (p<0.05) (Table 2). The TBW was67.8 + 0.3 I and 69.7 + 0.4 ? of the total body weight foll-owingingestion of water and 5% glycerol, respectively (p<0.01-) (Tab1e 2).The ICF was 50.3 + 0.5 3 with water and 52.8 + 0.7 I with glycerol(p<0.05) (Table 2). The vafues for the other fluid compartmenLs,ECF, PV, and fSF, were similar at 2 h for the two groups (Table 2)'

bDv4ErDtr-

ot'lz tt

I wnrnn

flsz cLYcERoL

Vol. 53, No. 23, 1993 Glycerol Solutions L783

rrME (h)

FIG. 2

Volume of fluid retained. Same fluid ingestion regimenas Figure 1. Significance beLween water and glycerol att h, 2 h and 3 h, p<0.01.

TABLE TI

Fluid comparLment volumes (? of bodv weiqht).

WATER (20 ml/ks)-_trLT_ J

5Z GLYCEROL (20 ml/kg)n=5

TBW %

ECF ZPVZICF ?ISF ? L2.9 ! 0.2Data presented as means + SE, * p50.05, ** p50.01.

Discussion

Oral administ.ration of a 5? glycerol solution enhanced fluidretention and attenuated urine output for 3 h in the rat. whencompared to the ingestion of water afone. This positive fluid

!

4af-1

zF3t'l

t4

,to

67.8 r 0.311 .4 t 0.14.5 r 0.1

50.3 r 0.5

69.1 t 0.4**17.0 r 0.34.4 ! 0.L

52.8 + 0.7*12.6 ! O-3

! wnran

f-l sz cLYcERoL

Glycerol Solutions Vol. 53, No.23, 193

balance at 2 h post-ingestion resulted in a condition of increasedTBW, which to i considerable extent resulLs from a ICF expansion(Figrs. 1 and 2, Table 2). No differences in ECF, PV, or ISF wereobserved between Lhe two fluid loading condiLions at 2 h (Table 2).

The changes in fluid balance attributed Lo Lhe GIH are in closeagreement wiCh data previously observed in boLh humans and ra|s,lnder similar protocols (10, 1'1', 19 ) ' Riedesel et al . (19 ) andLyons et al. lro) reported urine volume reductions in humans of 40Lo 52 ? after the ingestion of glycerol sol-utions when compared toan equivalent vol-ume of water. The 50% decrease in the presentstudy is within the range of 45 to 53? observed in a recentinveitigation with the rat subject to identical hyperhydrationregimeni (11-). The increased fluid retention also support.s the factthat a glycerol*induced hyperhydration was achieved (Fiq. 2) - Ifthe volumes of fluid retained are calculated as a percenL of totalfluid intake comparisons with our human data are more meaningful.These values from the current study are B3?, 722, and 648 at t h, 2

h, and 3 h following the 5? glycerol solution and 55?, 442, and 30?after ingestion of water at time zero. These data are similar tothe 83?, JOeo, and 63? at 1, 2, and 3 h with glycerol and 51?' 412,and 25? without glycerol observed in previous work with GIH in rats(11). Humans have demonstrated retentions of B2z at 1.5 h, 74% aL2.5 h and 68? at 3.5 h with glycerot and 403 at 1.5 h, 382 aL 2'5h, and 35? at 3.5 h with water (10). The maintenance of a highlypositive fluid bal-ance for 3 h supports the fact. that a GIH similarLo that observed in previous rat and human investigations wasachieved in the current series of experimenLs.

The reported data are sugqestive of the fact that the TBW andthe ICF comparLment undergo a greater expansion following ingestionof a 5? glycerol solution than with an equal volume of water. Thisstudy by aLsign only a1lows characterization of the differences influib -distributiorr between the two hyperhydrating regimens;therefore, absolute changes in fluid compartment volume frombaseline to post-ingestion for a given fluid treatment are notknown. It iJ plausible that both of the fluid protocols increasethe ECF volume, but the GIH also results in an additional expansionof the ICF compartmenL, which accounts for the greater TBW. Tn thiscase it would be expected that with GIH the volume changes for theTBW and ICF would be sj-milar. However, when compared to wateringestion the GIH resul-ted in increases of 3.2 and 5.9 ml in TBW andtCF, respectively. This discrepancy between the volume of expansionof t.he TBW and ICF will be further discussed.

The increased fluid retention with the GIH should be comparablein magnitude to the volume expansion observed in the TBW. Thediffefence in fluid retention between water and the 5? glycerol was2.0 ml aL 2 \; however, the difference in TBW between the two fluidregimens was 3.2 m] A previous study with Lhe rat estimated thatth; cIH should result in a theoretical fluid retention of 3.7 mI,which would be distributed inLo the TBW (11).

The two previously mentioned inconsistencies in the data arenot readily eiplained. The methods employed are well accepted andthe results are comparable to other techniques (9, 16, L'1,23). Thepossibility of experimental error in some of the procedures exists,and in addition, the vol-umes involved in GIH in an animal- model,

Vol.53, No. 23, 1993 Glycerol Solutions

such as the rat, are relatively small. Therefore, interpretation ofLhe actual fluid compartment volume data should be ]imited, and therelative direction of change may be more important. A11 the animalsin this study demonstrated a similar response to GIH, primarily.anincrease in tew and rCF (Table 2). The acLual volume of expansionmay be questionable for the reasons mentioned, but the trend wasconsj-stent and statistically significant.

Both the osmoLic properLies and distribution of the glycerolappear to play a major rote in the GIH. The hyperosmolality due tothe elevat-ed serum glycerol levefs is present up to 2.5 h in humans,but the enhanced flui-d retention is maintained for 4 to 8.5 h (10,19). Thj-s difference demonstrates that the GIH in previous studiesis not merely an aftermath of a hyperosmotic vascul-ar compartment.As a result, it is unlikely Lhat Lhe GIH j-s so1ely dependent on ahormonal response, such as an elevated ADH release.

A plausible hypothesis for GIH is based on the fact thatglycerol fras been sho-wn to distribute quite readily throughout theier{ fZZ, 24). In order to maintain isoLonicity, fluid shifUs wouldaccompany ihe movement of glycerot into the fl-uid spaces. . Tlt.fluid being distributed intraCellularly with GIH woul-d minimize thestimulation of volume and osmotic receptors, and it would Seem thata large compartment, such as the ICF, has a much greater potenLialto act as a fluid reservoir.

The increase in TBW by ICF expansion would also support theidea that the fluid load is being distributed out of the vascufarspace. our ]aboratory has not detected changes in PV at 1 to 8 hpost-inqestion of glycerol (1 g/kg') plus a large volume of water'(21-.4 *1 Zt g) in man

-or the rat (10, t1', 1-9 ) . The absence of ahemodilution after volume loading has al-so been observed in oLherstudies (15). Nadel et al. (15) reported that. a study in humansinvolving exog.enous ADH and 2,000 ml of water resulted in anincreased fluid retention and hyperhydration, but PV was notchanged, which might signify an ext.ravascular fluid distribution(15 ) .

fn conLrast to our findingis Murray et al. (:-4) reported thatthe ingestion of concentrated glycerol solutions during exercise didnot reiult in hyperhydration, but did induce a short-term expansionof the pV. Simillr findings were observed with pre-exerciseglycerol feedings (5). In these studies the increases in PV wereitiributed to ai el-evat.ed plasma osmolality, which was a result ofthe high serum glycerol concentrations. However, it is difficul-t torelate the resulls of these studies with those of our laboratorybecause the protocols did not involve fLuid loading or a GIH. - Theelevated plasma osmolality after glycerol infusion has also beenshown to elicit a hemodilution response (1, 8, 12).

A GIH similar to that observed in studies with humans wasachieved in this experiment with the rat. In addition, it seemsthat glycerol solulions may be more effective than water inexpana-in-g TBW because of the added ability to expand intracellularspaces. More extensive studies involving glycerol distribution andtire corresponding fluid shifts need to be conducted in order tofurther characterize the mechanisms of GIH.

1785

17K Glycerol Solutions Vol.53, No.23, 1993

Acknowledqements

Thi-s study was taken in part from a dissertation submitted byT.P. Lyons in partial fulfillment for a PhD degree in 1991 and wassupported by NASA Research Grant NAG-9-401-Basic.

References

1. L. CHEN, Y. WANG, and J. WANG, Cfinical Science'77 535-539(1989).

2. M.J. DURKOT, O. MARTTNEZ, D. BROOKS-MCQUADE and R.P.FRANCESCONI, rT. Appl . Physiol . 61 1031--34 (1986 ) .

3. B. EKBLOM, C..]. GREENLEAF, J.E. GREENLEAF, ANdL. HERMANSEN, Acta Physiol. Scand. 19 415-483 (1970).

4. S.M. FORTNEY, E. R. NADEL, C. B. WENGER, and J. R.,BOVE,,J. App1. Physiol.: Respirat. Environ. Exercise PhysioI.50 292-98 (1981).

5. M. GLEESON, R.,f. MAUGHAN, and P.L. GREENHAFF, Eur..T. Appl .

Physiol. 55 645-653 (1985) .

6. J.E. GREENLEAF and B.L. CASTLE, J. App1. physiol. 30 847-53(1971-).

7. R. GRUCZA, M. SZCZYPACZEWSKA, and S. KOZLOWSKI , Eur. ,J. Appl .

Physiol. 56 605-607 (1987).B. A. KOBAYASHI, H. HANDA, M. ISIKAWA, W. TAKI, Y YANEKAWA,

S. TANADA, Y. YANEKURA, M. SENDA, T. FUJITA, K. TORIZUKA,H. FUKUYAMA, N. FUJIMOTO, K HARADA, M. KAMEYAMA, BTainEdema,Y.Inaba, I. KJ-aLzo, and M. Spatz (eds), 273-276, Springer-Verlag, Berlin, HiedelberS (1985).

9. M. LARSSON, L. ,JOHNSON, c. NYLANDER, and V. OHMAN,Acta. Physiol. Scand. 110 53-57 (l-980).

10. T.P. LYONS, M. L. RIEDESEL, L. E. MEULT, and T. W. CHTCK,Med. Sci. Sports Exerc. 2 417-483 (1990).

11. T.P LYONS, PhD Dissertation, University of New Mexico,(1991).

1-2. J.S. MEYER, Y. TTOTH, S. KAMORTO, K.M. WELCH and MATTHEWERICSSON, Circulation 51 70L-1-2 (1975).

13. S.V. MOROFF, and D.E. BASS, J. App1. Physio1.20 267-270(196s).

1-4. R. MURRAY, D. E. EDDY, G. L. PAUL, J. G. SEIFFERT, and GEORGEA. HALABY, ,J. Appl. Physiol . , 71 144-149 (l-991-) .

15. NADEL, E. R., S. M. FORTNEY, and C. B. WENGER, J. Appl. Physiol.49(4) 7t5-721, (1980).

16. H. NOSE, T. MORIMOTO, and K. OGURA, Jpn. J. Physiol.,33 101_9-1029 (1983 ) .

17. H. NOSE, T. YAWATA, and T. MORIMOTO, Am. .I .

Physiol. 249 (Regulat.ory fntegrative Comp. Physiol.18) R166-R171 (198s) .

18. H. NOSE, M. MORITA, T. YAWATA, T. MORIMOTO, Am. J. physiol.251 (RegulaLory rntegrative Comp. Physiol. 20) R492-R498(1986).

19. M.L. RIEDESEL, D.Y. ALLEN, c.T. PEAKE, and K. AL-QATTAN,J. Appl. Physiol. 63(6) 2262-68 (1987).

20. M.N. SAWKA, R.W. HUBBARD, R.P. FRANCESCoNI and D.H.HORSTMAN, Eur. J. Appt. Physiol. 51 303-312 (1983).

21. M.N. SAWKA, R. R. GONZALEZ, A. J. YOUNG, S.R. MUZA, K.B.PANDOLF, W.A. LATZKA, R.C. DENNIS and C.R. VALERT, Am. ,J.Physiol. 255 (Regulat.ory Integrative Comp. Physiol. 24)R456-R463 (1988).

Vol.53, No.23, 1993 Glycerol Solutions

22. A..T. Wade, Biochem. 'J. 196 547-556 (l-981) .

23. W.M. WALLACE, K. GOLDSTEIN, A. TAYLOR, and T. M. TEREE,Am. J. Physiol. 2I9 1544-1548,(1970).

24. B. WINKLER, R. STEELE, and N. ALTSZULER. Amer. ,J. Physiol .

216 1_9L-196 (1969).

r787