H ypertonic solutions in the treatment ofhypovolemic shock: A prospective, randomizedstudy in patients admitted to the emergencyroomRiad N. Younes, MD, Frederico Aun, MD, FACS, Claudia Q. Accioly, MD,Luciano P. L. Casale, MD, Israel Szajnbok, MD, and Dario Birolini, MD, FACS,sao Pauta, Brazit

Background. The inJusion 01 small volumes 01 hypertonic saline solution or hypertonicsaline plus dextran 70 is remarkably effective in restoring adequate hemodynamicconditions alter hypovolemic shock. This prospective double-blind study compares theimmediate hemodynamic effects 01 a bolus inJusion 01 7.5% NaCl or 7.5% NaCl plus6% dextran 70 (both 2400 mOsmjL) in severe hypovolemia.Methods. One hundred five adult patients admitted in hypovolemic shock (systolicblood pressure <80 mm Hg) were revived on arrival to the emergency room andadministration 01 a 250 ml intravenous bolus 01 hypertonic saline solution (n = 35),hypertonic saline plus dextran (n = 35), or isotonic saline solution (n = 35). ThisinJusion u,'as immediately 1011owed by standard CT)'stalloid and blood replacement untilS),stolic pressure reached 100 mm Hg. Mean arterial pressure (MAP) was measuredevery 5 minutes, and all intravenous inJusions were registered. Plasma volumeexpansion was calculated from plasma protein concentration measurements. Patientswere 1011owed up throughout their hospital course, and results 01 treatment wererecorded.Results. At the end 01 the inJusion period, and 5 and 10 minutes after inJusion, MAPwas significantly higher in patients receiving either hypertonic solution, compared withthe group receiving isotonic solution. All groups showed similar trends towardrestoration 01 hemodynamic parameters thereafter. The calculated plasma volumeexpansion, immediately after the bolus inJusion, was significantly higher(24.1% :t 1.8% and 24.9% :t 1.1%) in the hypertonic groups, compared with isotonic

groups (7.9% :t 1.3%). Significantly greater volumes 01 fiuids were required to restoresystolic pressure in the patients receiving isotonic saline solution than in the groupsreceiving hypertonic solution. There were no significant differences between the groupsreceiving hypertonic solutions. The incidence 01 complications was low, and them orta lit y Tale was similar in all groups.Conclusions. InJusion 01250 ml hypertonic saline solution in patients with severehypovolemia was not related to any comPlications, nor did it affect mortality Tales,. itimproved MAP significantly, acutely expanded plasma volume by 24%, and reducedsignificantiy the volumes 01 crystalloids and blood required in their resuscitation.

(SURGERY 1992,.111:380-5.)

From the Trauma Service, Department o/ Surgery, Uniuersity o/ Sao Pauta School o/ Medicine,Sao Pauta, Brazil

THE INFUSION OF hypertonic solution has been reportedto be effective in the treatment of hypovolemia inexperimental and patient studies.l, 2 Hyperosmotic 7.5%

sodium chloride solutions (2400 mOsm/L) have beenshown previously to improve the hemodynamic stateafter hemorrhagic shock, leading to a high survival rateeven when given in small volumes (approximately 4

Supported in part by Laboratorios B. Braun.

Presented at the Surgical Forum, the Annual Meeting of theAmerican College of Surgeons, Chicago, ni., Oct. 23-28, 1988.

Accepted for publication Feb. 16, 1991.

Reprint requests: Dario Birolini, MD, Rua Olegário Mar-

iano, 671, 05612, Sao Paulo, SP Brazil.

11/56/32532

380 SURGERY

(/-h-t.

~;,~

-

~

Volume 111Number 4

Hypertonic solutions in hypovolemic shock 381

Table l. Patient characteristics at entry into theprotocol

Age (yr) 31 :j: 10

Male/female (n) 26/9Cause of hypoyolemia (n)

TraumaBlunt 18/35Gunshot 11/35Stab 4/35Other 2/35

Trauma score 8 :j: 2

20/3510/354/351/35

9 :t 2

20/3512/352/351/35

9:1:3

mljkg body weight) and as the sale treatment.3 Thesebeneficial effects are attributed to the increase in myo-cardial contractility,4 widespread precapillary dilation,5fluid redistribution,6 and a neurogenic cardiovascularreflex originated in the lungs.7,8

Based on the encouraging data originating from ex-perimental studies that confirmed the safety and efficacyof the infusion of hypertonic solutions in the treatmentof hypovolemic shock, and from experimental and pre-liminary clinical studies from the University of Califor-nia, Davis,9, 10 reporting that the addition of hyperon-

cotic 6% dextran 70 to the hypertonic saline solutionimproves the effectiveness of the resuscitation fromhemorrhagic shock, we initiated this randomized, dou-ble-blind study at the Hospital das Clinicas of the Uni-versity of Sao Paulo School of Medicine, Sao Paulo,Brazil. In this study "re evaluate the immediate hemo-dynamic effects of the bolus infusion of hypertonic sa-line solution or hypertonic saline plus dextran 70administered as the initial therapy in patients with se-vere hypovolemia compared with standard isotoniccrystalloid resuscitation. The other objectives of thisstudy were to assess electrolyte and protein alterationsafter resuscitation with either solution and to evaluatethe relative plasma expansion and impact of the hyper-tonic solutions on fluid requirements during the resus-citation process. We algo assessed the morbidity associ-ated with resuscitation, looking at the eventual compli-cations related to the shock state, the hypertonic sodiumload, and the infusion of dextran. In addition, we eval-uated the effect of the infusion of these solutions on theoutcome of the patients included in this protocolo

MATERIAL AND METHODS

One hundred five patients admitted to the emergencyroom in hypovolemic shock were included in the study.The patients eligible for the protocol were adults (aged>18 years), admitted ,vith hemorrhagic hypovolemia(with systolic pressure <80 mm Hg) with a palpablepulse or positive electrocardiogram, nonpregnant, andwith no previous history of cardiac or metabolic diseases.The patients included were randomized in a double-blind fashion to receive a 250 mI intravenous bolus ofeither hypertonic 7.50;0 NaCI solution (HS; n = 35),

hypertonic 7.50;0 NaCI plus 60;0 dextran 70 solution(H8D; n = 35), or isotonic 0.9% NaCI solution (18;n = 35). These solutions were prepared in similar and

unmarked bottles and coded ,vith randomly assignednumbers. The solutions were infused during a 2- to3-minute period, immediately followed by standardcrystalloid (0.90;0 NaCI) and blood replacement untilsystolic pressure was higher than 100 mm Hg (hemo-dynamic end point). Blood transfusion was routinely

indicated by the Trauma Service staff in our hospitalwhenever clinically obvious anemia was present or he-matocrit levels reached 30% or less, with ongoing hem-orrhage. Blood samples for electrolyte, protein, osmo-larity, and complete blood count determinations weredrawn before and 15 to 30 minutes after the infusion ofthe test solution. All intravenousvolumes administeredduring the resuscitation process were registered by theinvestigators. Plasma volume expansion was calculatedfrom plasma protein measurements according to theFick principIe. The alteration of the plasma volume wasestimated from the initial (protin) and final (protfin)plasma protein concentrations, through the followingequation: Plasma expansion = (Protin -Protfin)/

Protfin.The follow-up period consisted of the patients' entire

hospital stay. During the infusion period, we registeredany complication that could be related to the adminis-tration of highly concentrated solutions (particularlycardiac arrhythmias and neurologic alterations), as wellas any coagulopathy. During follow-up, the patientswere evaluated daily for the occurrence of pulmonarycomplications (evidenced by clinical, radiologic, or lab-oratory alterations consistent with respiratory failure),renal complications (evidenced by alterations of urinaryoutput, increase of blood urea nitrogen and creatinineconcentrations, and decrease of glomerular filtrationrate), cardiac complications (arrhythmias, myocardialinfarction, or congestive heart failure), or infectiouscomplications (confirmed by purulent discharge or pos-itive bacterial culture). When referred to in the Resultssection of this study, morbidity describes the occurrenceof a complication not associated with death. Patientswho had complications that ultimately resulted in deathare included only under mortality.

Statistical analysis was done by analysis of variance,unpaired Student's t test, and X2 test, with a = 0.05.

Results are presented as mean ::t SEM.

Younes el al. SurgeryAPril1992

382

95~w(/)

+1

'@:I:EÉ:,

C-

<x;~

85

75

65

55

45o 5 10

Time (min)

15 20

+7.5% NaCI ~ 7.5% NaCI+ Dex --4- 0.9% NaCI

* p<0.01 -** p<0.05 (0.9% vs 7.5% & 7.5%+ Dex)

Fig. 1. Hypertonic solutions in shock: MAP. Dex, Dextran.

Table 11. Laboratory results for the patients duringthe study

HS HSD IS

Sodium (mEqjL)Before infusion 143:t 4 147 :t 3 144 :t 415 min 158 :t 3* 155 :t 4* 143 :t 430 min 146 :t 4 147 :t 2 141 :t 5

Osmolality (mOsmjL)Before infusion 305:t 6 308 :t 4 306 :t 415 min 348 :t 13* 340:t 16* 305:t 930 min 321 :t 13 315:t 15 304 :t 7

Data are mean :t SEM..p < 0.01 compared with preinfusion levels.

RESULTS

Patient characteristics and baseline values (Table 1)for the three groups were similar, with no significantdifferences among them. Four patients with hypo-volemia of nontraumatic origin were included: one pa-tient with spontaneous hemothorax and three patientswith upper gastrointestinal bleeding from a benignpeptic ulcer. Trauma patients had severe injuries oftheabdomen, thorax, or extremities, with bleeding occur-ring mostly from severed majar vessels. These vascularinjuries were the result of lesions of intraparenchymalblood vessels or a majar artery or vein. The average es-timated blood loss in the patients included in OUT study,in all three groups, was in excess of 2000 mI. There wereno differences among groups when the incidence of iso-lated majar vessel injuries was compared (HS, 5; HSD,3; and IS, 6). Most of OUT patients had multiple organ

injuries,. with bleeding from various sites simulta-

neously.The mean arterial pressure (MAP) increased signif-

icantly in the patients who received HS and HSD so-lutions (Fig. 1). At the end of the bolus infusion, theMAP was significantly higher in the groups given hy-pertonic solutions compared ,vith the group givenisotonic solution. AII groups sho,ved similar trends to-,vard restoration of hemodynamic parameters thereaf-ter, because they received routine resuscitation untilsystolic pressure was restored. Plasma sodium concen-tration was increased after the bolus infusion in bothgroups given hypertonic solutions and returned to pre-treatment levels 30 minutes later (Table 11). There ,vereno significant alterations in the IS group. The changesobserved in the serum sodium levels were paralleled bythe changes in osmolarity (Table 11).

The calculated plasma volume expansion after thebolus infusion was significantly higher in the groupsgiven hypertonic solutions (HS, 24.1 % :t 1.8%; HSD,24.9% :t 1.1 %) compared with the group given isotonicsolution (7.9% :t 1.3%).

Significantly greater volumes of crystalloids andblood were required to restore systolic pressure in thegroup given isotonic solution than in the groups givenhypertonic solutions during the resuscitation process(Fig. 2). The median of crystalloid volume infusion re-quired in HS, HSD, and IS was 1000, 1500, and 2000mI, respectively. The median blood volume transfusionrequired in HS, HSD, and IS was O, O, and 500 mI, re-

spectively.We did not observe any sigñ:ificant differences in

overall complication and mortality rates in the three

J*

-

Hypertonic solutionsinhypovolemic shock 383Volume 111

Number.4

~

Group

-Crystalloid ~ Blood

* p<0.01 (0.9% vs. 7.5% /7.5% + Dex)

Fig. 2. Volumes required during resuscitation process (mean :t SEM). Dex, Dextran.

Table 111. Morbidity and mortality rates of the

patients included in the protocol

Mortality (nJComPlications (n)

2/352/353/35

HSHSDIS

groups (Table 111). Only one patient died in the emer-gency room during the study. This patient received iso-tonic resuscitation. No complications related to the in-fusion of hypertonic solutions were observed.

DISCUSSIONThe rapid transfer of injured patients from the scene

of trauma to a medical center has been proved to be ofthe utmost importance.ll Isotonic fluidsusually admin-istered in the prehospital setting areoften insufficient tomeet the needs to compensate for the blood loss. In re-sponse to this concern, the use of small volumes of hy-pertonic sodium chloride solutions has been evaluatedfor the treatment of hemorrhagic shock. Early studieshave shown their effectiveness in reversing the hemody-namic disturbances and decreasing mortality rates ofhypovolemic conditions. In 1980 Velasco et al.3 reported.1 OOU¡o survival of anesthetized dogs infused with 7.SU¡oNaCl in a volume equal to 10U¡o of the shed blood aftera fixed pressure hemorrhage. Several subsequent ex-perimental studies showed that the administration ofhypertonic sodium chloride dramatically increased bloodpressure12 and cardiac output13 and restored acid-basebalance. It algo increases myocardial contractility4 andthe mean circulatory filling pressure,14 reflecting a de-crease in overall venous capacitance and leading to amore effective and adequate adaptation of the vascularcapacity to the decreased blood volume. At the cellularlevel the infusion of 2400 mOsmfL NaCl improvedcellular resting membrane potential and decreasedintracellular water content after shock.15 Renal bloodflow16 and urine output11 were shown to in crease sig-

nificantly, along with a redistribution of arterial bloodflow preferentially to the splanchnic and renal circula-tion and the subsequent amelioration of renal cortex

cellular metabolism.18One of the major concerns in the patients with mul-

tiple injuries, blood loss, and hypotension is that the re-suscitation with large volumes of intravenous fluids mayworsen the effect of a concomitant head injury by con-tributing to cerebral edema. Compared with isotonicadministration, the HS solution decreased intracranialpressurel9; even in the presence of an intracranial masslesion, resuscitation with hypertonic (3%) saline solu-tion is accompanied by lower intracranial pressure val-ues and less cerebral edema than with isotonic salinesolution or colloid resuscitation.20 In light of these stud-ies, we, as well as others,10 included patients with head

injuries in the protocoloHypertonic solutions have been shown to be effective

in the management of patients in the intensive rafeunit,21 in thoracoabdominal aortic operations!2-24 and

during the transportation of trauma patients.10Despite the relative safety of small-volume infusion

of the hypertonic solutions in experimental models, few

11

384 Younes el al.Surg

4p

/92

that resuscitation with hyperosmotic solutions requiressmaller volumes of additional fIuids to maintain hemo-dynamic parameters.2S, 29

In the emergency room setting we did not observe any

major problems of rebleeding from injured blood vesselsafter the restoration of blood pressure in the patients ofthe three groups. The same observations ,vere reportedpreviously by Holcroft et al. 10

When we compared the HS solution with the HSDsolution, we did not see any differences in the incidenceof complications related to the infusion of dextran, nordid we see any significant immediate hemodynamic dif-ferences in the patients who received either solution.Although many theoretic6 and experimental models9, 30showed clear advantages of the hypertonicjhyperoncoticassociation over the hypertonic solution alone, ,ve couldnot show these advantages in OUT study. One explana-tion might be that resuscitation ,vas carried on contin-uously after the bolus infusion, and the decline in bloodpressure and cardiac output, described in the experi-mental models6 to Occur 15 to 30 minutes after the endof the infusion of HS solution, ,vas masked by the ad-

ministration of supplemental fIuids. Nevertheless, thevolumes administered in the emergency room to either

hypertonic group were similar, suggesting that the he-modynamic advantages of hypertonicjhyperoncotic as-sociation over hypertonic sodium chloride alone, if at allpresent, must be minimal in this setting.

We conclude that the bolus infusion of 250 mI hyper-tonic saline solution in the initial treatment of patients,vith hypovolemia is effective in immediately restoringMAP to normal levels, acutely expanding the plasmavolume by 24%, and reducing significantly the volumesof crystalloids and blood required in the resuscitation. Inaddition, this infusion is not related to any complica-tions, nor does it affect the mortality rates when givenin the emergency room. This study presents significantsupport and justifies the ongoing evaluation of the useof small-volume hypertonic solutions in the resuscita-tion of patients with hypovolemia in the field beforetransportation to a medical center, as well as in theemergency room before definitive treatment.

We are indebted to Mauricio Rocha e Silva, MD, PhD, COIhis assistance and orientation, Ms. Vanda M. Yoshida andMs. Elizabete Minami for their expert technical assistance,and the staff of the Trauma Service for their support.

studies are available related to the reactions of patientswith hypovolemia to the highly hypertonic sodiumload.25, 26 Our study was performed in the emergency

\vard, \vith all the patients under direct medical obser-vation for the entire duration of the protocolo The claseobservation in the emergency room, as opposed to the

prehospital setting, enabled us to draw blood samplesbefore and after the test infusion andprovided impor-tant biochemical data on the patients included in the

protocoloThe MAP was rapidly restored to normallevels in the

patients \vho received the small-volume infusions of ei-ther hypertonic saline solution or hypertonic saline plusdextran solutions. At the end of the infusion period,blood pressure was significantly higher in the hyper-tonic groups, compared with the patients randomized toreceive normal saline solution. AII groups sho\ved sim-ilar trends toward restoration of hemodynamic param-eters thereafter, beca use they received routine resusci-tation until the end point (systolic pressure > 1 00 mmHg) \vas reached. This is in accord with previouslypublished data on trauma patients receiving hyperos-

moticjhyperoncotic solutions during transportation.10Serum sodium concentration and osmolality increased

significan tI Y in the hypertonic groups after the infusionof the hyperosmolar solutions. The levels observed inour patients are comparable to those reported in previ-ous studies.10, 26 Like\vise, we did not observe any com-

plications related to hypernatremia and hyperosmolal-ity (e.g., arrhythmias or central nervous system adversereactions). On the contrary, many patients were morealert immediately after the infusion of hyperosmolar

solutions, possibly reHecting better cerebral perfusionand oxygenation. The levels of sodium and osmolalitydecreased to\vard normal values as resuscitation pro-gressed, probably beca use of dilution and natriuresis.

The plasma volume expansion estimated in thepatients of HS and HSD (24%) mar represent a shiftof 450 to 800 mI into the intravascular space. This rapidHuid redistribution, associated with a decrease in sys-temic venous capacitance, is of paramount importancein increasing the effective circulatory volume of patients\vith hypovolemia. Prior experimental6, 27 and clinical

studies26 have algo shown prompt expansion of plasmavolume after the administration of hypertonic solutions.This is mainly beca use of an osmotically induced Huidshift from cellular to extracellular compartments.

Significantly greater volumes of crystalloids andblood \vere required to restore and maintain the hemo-

dynamic parameters in the isotonic group than in thehypertonic groups during the resuscitation process.These results confirm experimental studies showing

REFERENCESBaue AE, Tragus E'and hypertonic solutiin the initial treatmenl

1967;7:743-56.

lrkins WM. A compar',n blood flow and oxyge

of hemorrhagic shoc

m of isotonic

k

Trauma

Hypertonic solutions in hypovolemic shock 385Volume 777Number 4

Beneficial effects of a hypertonic solution for resuscitation in the

presence of acute hemorrhage. Am J Surg 1987;154:585-8.18. Maksoud JG, Younes RN, Ayoub AAR. Tissue surface pH:

difference between muscular andrena1 reperfusion in responseto hypertonic resuscitation (NaCI 7.5%) from hemorrhagic

shock. Circ Shock 1987;21:372.19. Prough DS, Johnson JC, Stump DA, Stullken EH, Poole GV,

Ho",ard G. Effects of hypertonic saline versus lactated Ringer's

solution from hemorrhagic shock. J Neurosurg 1986;64:627-32.20. Gunnar W, Jonasson O, Merlotti G, Stone J, Barret J. Head

injury and hemorrhagic shock: studies of the blood-brain barrierand intracranial pressure after resuscitation with normal saline

solution, 3'7. saline solution, and dextran-40. SURGERY 1988;

103:398-407.21. De Felippe J Jr, Timoner J, Velasco IT, Lopes OU, Rocha E,

Silva M Jr. Treatment of refractory hypovolemic shock by so-

dium 7.5'7. chloride injections. Lancet 1980;2:1002-4.22. Younes RN, Bechara MJ, Langer B, et al. Prevention of aortic

declamping shock with hypertonic 7.5'7. NaCI solution inabdominal aneurysmectomy of the aorta [Abstract]. Braz J Med

Biol Res 1986;19:497.23. Auler JOC, Pereira MHC, Gomide Amaral RV, Stolf NG,

Jatene AD, Rocha e Silva M Jr. Hemodynamic effects of hy-pertonic sodium chloride during surgical treatment of aortic an-

eurysms. SURGERY 1987;101:594-601.24. Shackford SR, Fortlage DA, Peters RM, Fridlund PH, Sise

MJ. Serum osmolar and electrolyte changes associated with

large infusions of hypertonic sodium lactate for intravascular,.olume expansion of patients undergoing aortic reconstruction.

Surg Gynecol Obstet 1987;164:127-36.25. Younes RN, Aun F, Birolini D, Ka",ahara NT, Oliveira MR.

Hypertonic sodium chloride (7.5'7.) improves the hemodynam-ics of hypovolemic patients [Abstract]. Braz J Med Biol Res

1986;19:498.26. Younes RN, Aun F, Birolini D, et al. The initial treatment of

hypovolemic patients: use of the hypertonic 7.5'7. NaCI solution.

Rev Hosp Clin Fac Med Sao Paulo 1988;43:138-41.27. Kramer GC, Perron PR, Lindsey DC, et al. Smal\-volume re-

suscitation with hypertonic saline dextran solution. SURGERY

1986;100:239-46.28. Stanford GG, Patterson R, Payne L, Fabian TC. Hypertonic

saline resuscitation in a porcine model of severe hemorrhagic

shock. Arch Surg 1989;124:733-6.29. Traverso LW, Bel\amy RF, Hol\enbach SJ, Witcher LD. Hy-

pertonic sodium choride solutions: effects on hemodynamics and

survival after hemorrhage in swine. J Trauma 1987;27:32-9.30. Maningas FA, DeGuzman LR, Til\man FJ, et al. Smal\-vol-

ume infusion of 7.5'7. NaCl in 6'7. dextran 70 for the treatmentof severe hemorrhagic shock in swine. Ann Emerg Med

1986;15:1131-7.

2. Monafo WW, Chuntrasakul C, Ayvazian YH. Hypertonic so-dium solutions in the treatment of burn shock. Am J Surg

1973;126:778-83.3. Velasco IT, Pontieri Y, Rocha e Silva M Jr, Lopes OU. Hy-

perosmotic NaCI and severe hemorrhagic shock. Am J Physiol

1980;239:H664-73.4. Wi1denthal K, Mietz,,'iak DS, Mitchell JH. Acute effectsof

increased serum osmolality on left ventricular performance. Am

J PhysioI1969;216:898-904.5. Gazitua S, Scott JB, Chow CC, Haddy F J. Effects of osmolar-

ity on canine renal vascular resistance. Am J Physiol 1969;

217:1216-23.6. Mazzoni MC, Borgstrom P, Arfors KE, Intaglietta M. Dy-

namic fluid redistribution in hyperosmotic resuscitation of hy-

povolemic hemorrhage. Am J PhysioI1988;255:H629-37.7. Lopes OU, Pontieri V, Rocha e Silva M Jr, Velasco IT. Hy-

perosmotic NaCI and severe hemorrhagic shock: role of the in-

nervated lung. Am J PhysioI1981;241:H883-90.8. Younes RN, Aun F, Tomida RM, Birolini D. The role oflung

innervation in the hemodynamic response to hypertonic sodium

chloride solutions in hemorrhagic shock. SURGERY 1985;

98:900-6.9. Smith GJ, Kramer GC, Perron P, Nakayama S, Gunther RA,

Holcroft JW. A comparison of several hypertonic solutions for

resuscitation of bled sheep. J Surg Res 1985;39:517-28.10. Holcroft JW, Vassar MJ, Turner JE, Derlet R W , Kramer GC.

3% NaCI and 7.5% NaCI/dextran 70 in the resuscitation of se-

verely injured patients. Ann Surg 1987;206:279-88.11. Smith JP, Bodai BI, Hill AS, Frey CF. Prehospital stabiliza-

tion of critically injured patients: a failed concepto J Trauma

1985;25:65- 70.12. N akayama S, Sibley L, Gunther RA, Holcroft JW , Kramer GC.

Small volume resuscitation with hypertonic saline (2400 mOsm/liter) during hemorrhagic shock. Circ Shock 1984;13:149-59.

13. Layon J, Duncan D, Gallaghar TJ, Banner MJ. Hypertonicsaline as a resuscitation solution in hemorrhagic shock: effect onextravascular lung '\Tater and cardiopulmonary function. Anesth

Analg 1987;66:154-8.14. Lopes OU, Velasco IT, Guertzenstein PG, Rocha e Silva MJr,

Pontieri V. Hypertonic NaCI restores mean circulatory filling

pressure in severely hypovolemic dogs. Hypertension 1986;8

(suppl):II95-9.15. Nakayama S, Kramer GC, Carlsen RC, HolcroftJW. Infusion

of very hypertonic saline to bled rats: membrane potentials and

fluid shifts. J Surg Res 1985;38:180-6.16. Rocha e Silva M, Negraes FA, Soares AM, Pontieri V,

Lopponow L. Hypertonic resuscitation from severe hemor-

rhagic shock: patterns of regional circulation. Circ Shock 1986;

19:165-75.17. Gane JB, Wallace BH, Caldwell FT, Smith SD, Searcy R.