Building a Better Fluid for Emergency Resuscitation ofTraumatic Brain InjuryBruce A. Crookes, MD, Stephen M. Cohn, MD, FACS, Harry Bonet, MD, Elizabeth A. Burton, BA, Jacob Nelson, BS,Matthias Majetschak, MD, Albert J. Varon, MD, Joel M. Linden, PhD, and Kenneth G. Proctor, PhD
Hextend (HEX) is a colloid solution thatis FDA-approved for volume expansion dur-ing surgery. ATL-146e is a novel adenosineA2A receptor agonist that has anti-inflamma-tory, neuroprotective, and coronary vasodila-tor properties. Three series of experimentswere designed to evaluate the therapeutic po-tential of HEX�ATL-146e for emergency re-suscitation from traumatic brain injury (TBI)� hemorrhagic hypotension.
Methods: In the first two studies invivo, anesthetized, ventilated pigs (30–45kg) received a fluid percussion TBI, 45%arterial hemorrhage, and 30 minutes shockperiod. In Series 1, resuscitation consisted ofunlimited crystalloid (n � 8) or HEX (n �8) to correct systolic arterial pressure>100 mm Hg and heart rate <100 bpm forthe first 60 minutes (“emergency phase”),and then maintain cerebral perfusion pres-sure (CPP) > 70 mm Hg for 60–240 min-utes. In Series 2 (n � 31), resuscitation con-
sisted of a 1 L bolus of HEX � ATL-146e(10 ng/kg/min, n � 10) or HEX �placebo (n� 10) followed by crystalloid to the sameendpoints. In Series 3 in vivo, the hemody-namic response evoked by 0, 10, 50, or 100ng/kg/min ATL-146e was measured beforeor 60 minutes after HEX resuscitation from45% hemorrhage.
Results: Following TBI�hemorrhage,there were 4/22 deaths in series 1 and 11/31deaths in series 2. In those alive at 30 minutes,mean arterial pressure, cardiac index, mixedvenous O2 saturation, and cerebral venous O2
saturation were all reduced by 40–60%, whileheart rate and lactate were increased 2–5 fold.With no resuscitation (n � 2), there was min-imal hemodynamic compensation and pro-gressive acidosis. Upon resuscitation, thesevalues corrected but intracranial pressureprogressively rose from <5 mm Hg to 15–20mm Hg. Series 1: With HEX (n � 8) versuscrystalloid (n � 8), CPP was less labile, acid/
base was maintained, and the fluid require-ment was reduced by 60% (all p < 0.05) Series2: With ATL-146e (n � 10) versus placebo(n � 10), stroke volume and cardiac outputwere improved by 40–60%, and the fluid re-quirement was reduced by 30% (all p < 0.05).Series 3: ATL-146e caused a dose-related in-crease (p < 0.05) in stroke volume after, butnot before, hemorrhage. The effects on pre-load, afterload, and heart rate were similarbefore and after hemorrhage.
Conclusions: HEX alone is a safeand efficacious low volume alternative toinitial crystalloid resuscitation after TBI. Anadenosine A2A agonist combined with 1 Lof HEX safely and effectively counteracted adecrease in cardiac performance noted afterTBI�hemorrhage without causing hypoten-sion or bradycardia.
Each year in the United States, there are approximately1.6 million cases of traumatic brain injury (TBI), whichlead to 80,000 severe, permanent neurologic disabili-
ties,and 52,000 deaths.1 The morbidity and mortality ratesfrom TBI on the battlefield are even worse.2
The fundamental goals of resuscitation of the TBI patientare the restoration of circulating volume, blood pressure, and
oxygenation, but there are no evidence-based guidelines forthe optimal fluid type and/or endpoint for resuscitation.1 Werecently reported improved cardiopulmonary performance af-ter low volume resuscitation from severe chest trauma withHextend (HEX) supplemented with ATL-146e.3 HEX rapidlyrestored hemodynamics while ATL-146e improved early car-diac performance without causing hypotension orbradycardia.3 The overall aim of this study was to determinewhether HEX with or without ATL-146e had salutary actionsafter severe TBI.
HEX is a high molecular weight hydroxyethyl starch inbuffered electrolyte dextrose solution that is FDA-approvedas a plasma volume expander for treatment of hypovolemiaduring elective surgery. Because of its favorable profile of
Care, University of Miami School of Medicine, Miami, Florida (B.A.C.,S.M.C., E.A.B., J.N., M.M., K.G.P.), Department of Anesthesiology, Uni-versity of Miami School of Medicine and Department of Medicine, Miami,Florida (H.B., A.J.V., K.G.P.), University of Virginia School of Medicine(J.M.L.), Charlottesville, Virginia.
Supported by Grants #10035 and #10339 from the Office of Naval Research.Portions of this work were presented in abstract form at the 14th Annual
Fellow, Resident, and Medical Student Surgical Research Forum, South FloridaChapter of the American College of Surgeons, Miami, Florida, May 2003; the 32ndCritical Care Congress, Society of Critical Care Medicine, San Antonio, Texas,January 2003; the Florida Committee on Trauma Resident Paper Competition,Tampa, Florida, November 2002; the 62nd Annual Meeting of American Associa-tion for the Surgery of Trauma at Orlando, Florida, September 2002.
SMC is currently WB Russ Professor and Chairman, Department ofSurgery, University of Texas Health Sciences Center, San Antonio, Texas.BAC is currently affiliated with the Department of Surgery, University ofVermont School of Medicine, Burlington, Vermont.
Address for reprints: Kenneth G. Proctor, PhD, Divisions of Traumaand Surgical Critical Care, Daughtry Family Department of Surgery, Uni-versity of Miami School of Medicine, Ryder Trauma Center, 1800 NW 10thAvenue Suite 215, Miami, FL 33136: email: [email protected].
DOI: 10.1097/01.TA.0000135162.85859.4C
The Journal of TRAUMA� Injury, Infection, and Critical Care
Volume 57 • Number 3 547
side effects relative to other intravenous fluids,4–6 HEX hasbeen selected for use as a field resuscitant by the U.S. mili-tary. ATL-146e is an adenosine A2A receptor-selective ago-nist that protects the spinal cord from reperfusion injury.7–12
It also has, anti-inflammatory,13–14 and coronary vasodilatorproperties15–16 and is in phase 2 clinical trials as a coronarydilator for pharmacological stress imaging.
A low volume resuscitation fluid supplemented with acompound that has well-described protective properties17–19
could offer logistical advantages for trauma patients in pre-hospital field conditions or whenever there is limited medicalresources or prolonged transport times.20 These actions couldbe life-saving in the critical minutes after TBI, when the brainis especially vulnerable to secondary hypoxia orhypotension.21–22 Three series of experiments were designedto evaluate the therapeutic potential of HEX � ATL-146e foremergency resuscitation from TBI � hemorrhagic hypoten-sion. The first series compared unlimited HEX alone to un-limited saline alone. The second series examined whethercardiac performance was enhanced when HEX was supple-mented with ATL-146e. The third series tested the dose-related myocardial effects of ATL-146e before and after fluidresuscitation. This is the first study to examine the actions ofHEX or ATL-146e after severe TBI.
METHODSMaterials
ATL146e (4-[3-[6-Amino-9-(5-ethylcarbamoyl-3,4-di-hydroxy-tetrahydro-furan-2-yl)-9H-purin-2-yl]-prop-2-ynyl]-cyclohexanecarboxylic acid methyl ester) was provided byDr. Jason Rieger of Adenosine Therapeutics, LLC, Char-
lottesville, VA. Hextend was provided by Dr. Paul Segall atBioTime, Inc, Berkeley, CA.
Study PopulationMale and female, prepubertal, farm-raised cross-bred
pigs (30–45 kg, n � 58) were housed in the university animalcare facility, and maintained in exact accordance with allpertinent animal welfare regulations. All data were collectedunder general anesthesia with ketamine (Fort Dodge AnimalHealth, Fort Dodge, IA), xylazine (Fort Dodge AnimalHealth) and fentanyl citrate (Abbott Labs, North Chicago,IL). The Animal Care and Use Committee of the Universityof Miami approved all procedures.
InstrumentationThree series of experiments were performed. The overall
study design is outlined in Figure 1. In anesthetized swine,the trachea was intubated, and mechanical ventilation wasinitiated in the supine position. One liter of normal saline wasadministered and the ventilator was adjusted to maintainoxygen saturation (SaO2) of 100%, an inspired oxygen con-centration (FiO2) of 40%, and an end-tidal CO2 (ETCO2) of40 mm Hg. A left external jugular catheter was placed foradministration of supplemental fluid. The right heart andpulmonary artery were catheterized via the right externaljugular vein. The left femoral artery was catheterized forpressure measurements and for hemorrhage. Flow-throughtransducers were used to obviate the need for heparin. Fol-lowing in vivo calibration (NOVA Ultra M, Waltham, MA),continuous measurements were made of all pressures, oxygensaturation, and cardiac output (Oximetrix3 SO2/CO com-
Fig. 1. Experimental design.
The Journal of TRAUMA� Injury, Infection, and Critical Care
548 September 2004
puter, Abbott Labs, Abbott Park, IL). EKG was routinelymonitored.
Additional ProceduresIn Series 1 and 2 experiments only, animals were placed
in the prone position. A 1 cm diameter craniotomy was madeat the intersection of the transverse and sagittal suturesthrough which the superior sagittal sinus was cannulated. Afiberoptic probe (Integra Neurosciences, Camino, Plainsboro,NJ) was inserted into the subdural space through the midlinecraniotomy for measuring intracranial pressure (ICP). A sec-ond 1 cm craniotomy was made over the left fronto-parietalcortex, approximately 1 cm lateral and 2 cm anterior tobregma. A fluid percussion device was attached to a multi-port, hollow bolt inserted into the second craniotomy. Finally,bispectral electroencephalogram (EEG) probes were attachedand continuously monitored (Aspect BIS, Natick, MA).These procedures have been described in several previouspublications from this lab.23–26
In Series 1 and 2 experiments only, after a 30–60 min-utes stabilization period following instrumentation, a total offive separate measurements of the physiologic variables de-scribed below were made in pre-injury conditions at 10 min-utes intervals. The first three measurements were made toverify baseline stability, then CO2 was infused into the ven-tilatory circuit, and titrated until the ETCO2 was 60–80 mmHg. After 10 minutes of increased FiCO2, data were col-lected. The ventilator was returned to FiCO2 �0; FiO2 �0.4and data were collected. A CO2 challenge was repeated at 60minutes intervals to evaluate cerebrovascular reactivity andcompliance, as previously described.23–24
Dose Response ProtocolIn Series 3, there were no CO2 challenges. After a 30–60
minutes stabilization period, three separate measurements ofthe physiologic variables described below were made at 10minutes intervals to verify baseline stability, then an infusionat 10 mL/hr of 10, 50, or 100 ng/kg/min ATL-146e in normalsaline was administered. Each dose was administered for 10minutes, then data were collected. After the dose-responsecurve was completed, there was a 20 minutes washout periodfollowed by injury/resuscitation.
Injury/Resuscitation ProtocolIn Series 1 and 2, a fluid percussion pulse (approximately
8.5 atmospheres for 10 msec) was administered to the intactdura.23–26 Immediately after this injury, 45% of the estimatedblood volume was withdrawn from the femoral artery cath-eter over 30 minutes. In Series 3, there was a 45% hemor-rhage with no fluid percussion. During the course of thehemorrhage, the ventilator was switched to room air (FiO2 �0.21) and no intravenous fluids were administered. Followingthe 30 minutes hemorrhage period, the FiO2 was increased to0.4 and resuscitation was initiated.
In Series 1, resuscitation consisted of either HEX ornormal saline that was titrated to a systolic arterial pressure(SAP) �100 mm Hg and heart rate (HR) �100 bpm for thefirst 60 minutes (“emergency phase”). This 60 minutes periodwas designed to simulate the transport time and the time thatthe newly-arrived TBI patient spends in the emergency roombefore receiving his ventriculostomy (i.e. initial resuscitation,time obtaining and interpreting radiographic studies, and timeto place a ventriculostomy catheter). After 60 minutes, eitherHEX or normal saline was titrated to maintain cerebral per-fusion pressure (CPP) � 70 mm Hg. An unlimited volume ofeither fluid was administered to achieve the resuscitationendpoints. (see Fig. 1). In n � 2, no resuscitation fluid wasadministered.
In Series 2, either ATL-146e (10 ng/kg/min) or normalsaline was infused at 10 mL/hr starting at 25 minutes afterTBI. At 30 minutes after TBI, the resuscitation consisted of 1L of HEX followed by lactated Ringers that was titrated toSAP �100 mm Hg and HR �100 bpm for the “emergencyphase.” After 60 minutes, lactated Ringers was titrated tomaintain CPP � 70 mm Hg. An unlimited volume of lactatedRingers was administered to achieve the resuscitation end-points. (see Fig. 1)
In Series 3, at 30 minutes after hemorrhage, resuscitationconsisted of 1 L of HEX followed by lactated Ringers titratedto SAP �100 mm Hg and HR �100 bpm. At 60 minutes, thedose-response curve for ATL-146e was repeated. (see Fig. 1)
Physiologic VariablesThe following experimental data were collected: volume
of intravenous fluid administered, blood temperature, HR,SAP, MAP, pulmonary artery pressure, pulmonary capillarywedge pressure, central venous pressure, cardiac output,mixed venous oxygen saturation (SvO2), SaO2, ETCO2, FiO2,bispectral EEG, EEG burst suppression ratio, sagittal sinuspressure, and intracranial pressure. Blood gases and electro-lytes were drawn from the arterial line and from the sagittalsinus catheter at the 30 minutes intervals. A complete bloodcount was performed every 30 minutes (Abbott Cell-Dyne1600).
StatisticsData are expressed as mean � SE. Differences were
compared with two-way analysis of variance and non-para-metric tests (Fisher PLSD and Scheffe F test) at the 95%confidence interval.
RESULTSSeries 1
TBI�40% hemorrhage produced 4/22 deaths within30–40 minutes. Those that died (31.8 � 3.5 kg) were hem-orrhaged 958 � 107 mL. In n � 2 survivors, no resuscitationfluid was administered. Both remained alive for the 240minutes observation period but were moribund (e.g.MAP�40 mm Hg, HR �120 bpm, arterial lactate � 8
Emergency Resuscitation for Traumatic Brain Injury
Volume 57 • Number 3 549
mmol/L, ischemic EEG changes, fixed-dilated pupils). Thestudy population was comprised of 2 fluid resuscitatedgroups. The saline group (32.6 � 1.8 kg, n � 8) was hem-orrhaged 966 � 44 mL. The HEX group (32.6 � 2.1 kg, n �8) was hemorrhaged 981 � 64 mL. In the two groups, MAPand HR were indistinguishable in baseline and shock condi-tions, and were restored to similar values during the “emer-gency” phase, and throughout the “CPP management” phase(Fig. 2). By 240 minutes, 6/8 were alive with HEX comparedwith 5/8 with saline.
Cardiac output was 2.5–3.0 L/min before hemorrhage(70–90 mL/min/kg), and dropped to 1.0–1.5 L/min (30–40mL/min/kg) post-hemorrhage, but then recovered to 3.5–4.5L/min (80–120 mL/min/kg) with resuscitation. Arterial lac-tate increased from �2 mmol/L at baseline to �5.5 mmol/Lafter shock, but gradually corrected with fluid. SvO2 de-creased from 66–74% to 42 - 48% during shock, and thenonly partially corrected to 50–60%. In contrast, cerebralvenous O2 saturation fell less during shock (from 35–45% to20–30%) and fully recovered to baseline with resuscitation.Intra-cranial pressures rose progressively during the resusci-tation period, eventually reaching 15–20 mm Hg by the endof observation. However, there were no treatment-relateddifferences between these values, so the data are not shown.
Cerebrovascular compliance (as reflected by the CO2
-evoked change in ICP) and cerebrovascular reactivity (asreflected by the CO2-evoked change in cerebral venous O2
saturation) were both significantly impaired after TBI, how-ever, there were no treatment-related differences betweenthese values, so the data are not shown.
Figure 3 shows the total amount of fluid to achieve theresuscitation endpoints in the two groups. With saline, 161 �17 mL/kg (5443 � 535 mL, n � 5) was required comparedwith 56 � 7 mL/kg (1880 � 249 mL, n � 6) with HEX (p �0.05). Despite these differences, filling pressures, hemat-ocrits, and plasma Na� and K�, were similar between groups,so those data are not shown.
Figure 4 shows that CPP and acid/base were better main-tained with HEX versus saline. By 240 minutes post TBI,CPP was extremely labile and completely fluid dependentwith saline, but was relatively stable at �70 mm Hg in theHEX group. This difference was significant (p � 0.05). WithHEX, there was an arterial base excess at the end of theobservation which was near the pre-injury baseline (3–6
Fig. 2. Series 1 Experiments: For 30–90 minutes after TBI (“emer-gency phase”), an unlimited volume of either HEX or saline wasadministered to maintain SAP�100 mm Hg and HR�100 bpm. Forthe remainder of the 240 minutes post TBI observation period, CPPwas maintained �70 mm Hg with as much of either fluid as neces-sary. These data show that MAP and HR were virtually identicalbetween the groups at all time points.
Fig. 4. Series 1 Experiments: CPP and Acid-base status were bothbetter maintained with HEX versus saline and the differences weresignificant (both p � 0.05).
Fig. 3. Series 1 Experiments: The total amount of HEX required toachieve the resuscitation endpoints was 60% less than the totalamount of saline and the difference was significant (p � 0.05).
The Journal of TRAUMA� Injury, Infection, and Critical Care
550 September 2004
mEq/L). However, with saline, a progressive acidosis devel-oped and by 240 minutes post TBI, there was a base deficit of7–9 mEq/L. This difference was significant (p � 0.05). Therewas a similar trend with ICP, but that apparent difference wasnot statistically significant.
Series 2In this series, TBI�45% hemorrhage produced 11/31
deaths within 30–40 minutes. Those that died (38.1 � 1.6kg) were hemorrhaged 988 � 34 mL. In the placebo group(39.1 � 1.3, n � 10), the hemorrhage volume was 1006 � 32mL. In the ATL-146e group (39.2 � 1.9, n � 10), thehemorrhage volume was 1006 � 50 mL. All that survived topoint of resuscitation survived to the end of the 240 minutesobservation period.
Figure 5 shows that throughout the observation period,MAP and HR were essentially superimposed with or withoutATL-146e. At 60 minutes after the initiation of resuscitation,HR was 73 � 6 bpm and 74 � 4 bpm in the other. Systolicblood pressure, at the same time, was 118 � 5 mm Hg in oneand 120 � 4 mm Hg in the other.
Several other variables, including arterial lactate, basedeficit, SvO2 and pH, showed the same basic pattern as Series1 with shock and resuscitation. However, there were notreatment-related differences, so those data are not shown.
Figure 6 shows the total amount of fluid required tomaintain systemic hemodynamics. The placebo � HEX �
LR group required 6085 � 582 mL (or about 150 mL/kg),while the ATL-146e � HEX� LR group received 4360 �513 mL (or about 115 mL/kg). This difference was signifi-cant (p � 0.05).
Figure 7 shows that, despite the relative volume deficitwith HEX�ATL-146e versus HEX alone, cardiac output andstroke volume corrected faster and were maintained through-out the resuscitation period. Cardiac output, which was 3.0–3.4 L/min (75–80 mL/min/kg) before hemorrhage, droppedby half after hemorrhage. After initiation of resuscitation,stroke volume and cardiac index were increased by 20–40%with HEX�ATL-146e. Cardiac output after resuscitationrose from 1.23 � 0.08 L/min (32 � 2 mL/min/kg) post-hemorrhage to 3.50 � 0.28 L/min (91 � 9 mL/min/kg) withHEX alone at 60 minutes, and to 4.80 � 0.85 L/min (122 �19 mL/min/kg) at the same time with the A2A agonist. The
Fig. 5. Series 2 Experiments: At 25 minutes after TBI, a 10 mL/hrinfusion of either ATL-146e (10 ng/kg/min) or placebo was started.The infusion continued for the entire 240 minutes post TBI obser-vation. Five min later (i.e. at 30 minutes after TBI), each received aone liter bolus of HEX. For 30–90 minutes post TBI (“emergencyphase”), an unlimited volume of lactated Ringers was administeredto maintain SAP�100 mm Hg and HR�100 bpm. At 90 minutes postTBI, and for the remainder of the 240 minutes post TBI observationperiod, CPP was maintained �70 mm Hg with as much lactatedRingers as necessary. These data show that MAP and HR werevirtually identical between the groups at all time points.
Fig. 6. Series 2 Experiments: In the ATL-146e group, the totalamount of fluid (HEX �LR) required to achieve the resuscitationendpoints was about 30% lower than the total amount of fluid in theplacebo group and the difference was significant (p � 0.05).
Fig. 7. Series 2 Experiments: With ATL-146e versus placebo, car-diac output and stroke volume corrected faster and were bothsustained at elevated levels. By the end of the observation, bothcardiac output and stroke volume were significantly different in thetreatment group (both p � 0.05).
Emergency Resuscitation for Traumatic Brain Injury
Volume 57 • Number 3 551
elevation in cardiac output was sustained: the value at 240minutes with HEX�A2A was 5.77 � 0.81 L/min (148 � 20mL/min/kg) versus 3.70 � 0.54 L/min (96 � 15 mL/min/kg).These differences were significant (p � 0.05).
The bottom panel of Figure 7, shows that there was aATL-146e treatment-related difference in stroke volume, thatmimicked the effect on cardiac output.
Series 3In this series (37.5 � 2.8 kg, n � 5), 45% hemorrhage
with no TBI produced 0/5 deaths. Table 1 shows the dose-related effects of ATL-146e on determinants of myocardialperformance before and after hemorrhage. The changes inpre-load, afterload and heart rate were similar before andafter hemorrhage. ATL-146e produced moderate hypotension(15–20 mm Hg decrease) with little effect on pre-load beforeand after hemorrhage. However, there was a difference incardiac output and stroke volume. Before hemorrhage, car-diac output did not increase until the dose was 50–100 ng/kg/min (i.e. 5–10� the dose used in Series 2 experiments),but this was accompanied by increases in heart rate, so therewas no net change in stroke volume. In contrast, after hem-orrhage, there were significant increases in cardiac output andstroke volume at 10, 50, and 100 ng/kg/min.
DISCUSSIONThe results from this experimental study on brain trauma
confirm and extend our recent experimental study on severechest trauma,3 which showed that low volume resuscitationwith HEX � ATL-146e enhanced cardiopulmonary perfor-mance. In both injury models, for the sake of realism, softtissue injury was superimposed on a life-threatening hemor-rhage insult, and the control group received “standard ofcare” for 60 minutes, to simulate transport to the traumacenter, initial evaluation, stabilization, radiographic studies,etc. In both studies, comparatively low volumes of HEXrapidly corrected the volume deficit, while the adenosineA2A agonist, ATL-146e, enhanced cardiac output and strokevolume without producing the well-described hypotension orbradycardia expected with its parent compound, adenosine.
The first series of experiments basically showed that,when resuscitating to the systemic hemodynamic endpointscommonly used in patients (SAP�100 mm Hg, HR�100bpm for first 60 minutes) with unrestricted volume, HEX wasequivalent to crystalloid, except less volume was required(Fig 3; 1.9 � 0.2 L or 56 � 7 mL/kg versus 5.4 � 0.5 L or161 � 17 mL/kg) and CPP and base deficit were bettermaintained (Fig 4).
This finding is important because the brain is particularlyvulnerable to secondary hypotensive or hypoxic insults dur-ing transport and during the diagnostic work-up at the traumacenter. The newly-arrived patient may be resuscitated for 1–2hours before receiving a ventriculostomy (i.e. time to initialresuscitation, time obtaining and interpreting radiographicstudies, and time to prepare and place a ventriculostomycatheter). Furthermore, decreasing the volume of resuscita-tion fluid could provide a logistical advantage in scenarios inwhich space and weight are at a premium, such as on thebattlefield, during emergency transport, or during resuscita-tion in a rural setting. In addition, CPP resuscitative strate-gies, with their emphasis on maintaining intravascular vol-ume, may lead to hypervolemia when crystalloids are utilizedfor resuscitation. This undesired consequence has been pre-viously associated with an increased risk of pulmonary com-plications, including Acute Respiratory Distress Syndrome.1
Determining the exact mechanism for the protective ef-fect of HEX after TBI is beyond the scope of this study,however if is probably not related to tonicity per se. ThemOsm/L of NS, HEX, and LR are 308, 307, and 275 respec-tively. Furthermore, we measured blood osmolarity and thevalues ranged from 280–290 and did not differ significantlybetween the HEX and NS groups. There are several studieswhich have clearly shown benefits of hypertonic saline afterTBI in patients27–29 Whereas that mechanism is probablyrelated to reduced endothelial swelling30 due to the hyperos-molarity, a similar physical effect probably could not explainthe actions of HEX.
It is clear that in patients, HEX expands intravascularvolume 2–3� more than crystalloid, for up to 48 hours. Evenafter as much as 8 L, HEX is apparently not associated with
Table 1 Series 3 Experiments: Dose-Related Actions of A2A Agonist, ATL-146e, on Determinants of MyocardialPerformance
The Journal of TRAUMA� Injury, Infection, and Critical Care
552 September 2004
the coagulopathy often seen with other synthetic colloids.4–6
However, one problem with the design of series 1 is that insome field situations, or during emergency medical transport,unlimited volumes of fluid are not available.
The second series of experiments was intended to mimica situation where resources are limited. In this design, a oneliter bolus of HEX was administered after initial treatmentwith a placebo or ATL-146e (10 ng/kg/min). These datashowed improved early myocardial performance with no un-desired hemodynamic effects (Figs 5–7).
The third series of experiments suggests that cardiacoutput was increased because of an inotropic effect of ATL-146e. It is important to emphasize that in any given set ofcircumstances an increase in cardiac output could be attrib-uted to an increase in pre-load (Frank-Starling effect), adecrease in afterload, an increase in heart rate, or an increasein contractility. Table 1 shows that ATL-146e had qualita-tively similar, dose-related effects on pre-load, afterload,heart rate, and cardiac output before or after hemorrhage.However, there was a dose-related increase in stroke volume,after (but not before) hemorrhage, which is consistent withincreased contractility. This increase is most likely due to adirect A2A mediated coronary vasodilation15–16 or to an A2Aeffect to counteract the action of inflammatory mediators thatinhibit myocardial contractility.7–14 In vitro, ATL-146e is apotent dilator (ED50 � 30 � 4 nM) of pig coronary arteries,with a low affinity (EC50 � 440 � 40 nM) at A1receptors.33–34 An inotropic action could explain the furtherreduction in the amount of intravenous fluid required toresuscitate animals receiving ATL-146e (Fig 6). Further in-vestigation is needed to determine whether this short-termaction during the immediate resuscitation period has anypractical significance for survival or long-term neurologicoutcome.
Adenosine is a normal constituent of all body fluids thataccumulates during inflammation, hypoxia, and ischemia andserves to modulate the pathologic responses associated withreperfusion. For almost 50 years, various investigators havereported benefits of fluids supplemented with ATP, adeno-sine or one of its related purine metabolites. Unfortunately,the putative benefits of most adenosine-related drugs31,32 areoffset by short duration of action, and powerful vasodilatorand bradycardic actions that are absolutely contraindicatedduring hypotensive shock states. The explanation is that, inaddition to its metabolic actions inside the cell, the extracel-lular actions of adenosine are dependent upon on at least 4sub-types of cell-surface receptors (A1, A2A, A2B, and A3),which vary widely in distribution between species and celltype.19 With the molecular cloning of these receptors, and thesynthesis of several new agonists33,34 and antagonists,35 therehas been a resurgent interest in the theoretical potential ofadenosine analogs to target individual receptors to protectagainst inflammation and reperfusion injury, particularly inthe central nervous system7–12,36. However, it should be em-
phasized that the beneficial effects in spinal cord reperfusioninjury may not directly apply to TBI.
There are at least three elements of the experimentaldesign that limit the practical relevance of these findings tothe clinical situation. First, the observation period post TBIwas relatively short; Second, there was no behavioral para-digm to test whether these short term physiologic changesimproved neurologic outcome. Third, the pigs received fluidalone to address CPP during resuscitation from severe TBI �hemorrhage, whereas a TBI patient could receive mannitol,pressors, and/or transfusions. All of these issues have beenaddressed in a series of follow-up experiments.37
In summary, these present observations substantiate ourearlier work3 that HEX and ATL-146e have no obviousharmful effects after trauma. In addition, the use of HEX forTBI reduced the volume of intravenous fluid required toobtain standard, physiologic endpoints of resuscitation. Sup-plemental ATL-146e further reduced the volume required forresuscitation, by increasing myocardial contractility. Furtherstudies are warranted to investigate the long-term effects ofHEX and ATL-146e on intracranial pressure, cerebral inflam-matory response to injury, and neurologic outcome.
ACKNOWLEDGMENTSWe thank Dr. Jason Rieger of Adenosine Therapeutics, LLC (Char-
lottesville, VA) for providing the ATL146e; Cindy Elidrissi, RN, of AspectMedical Systems (Natick, MA) for providing the bispectral EEG monitor;Paul Segall, PhD, of BioTime, Inc. (Berkeley, CA) for providing the Hex-tend; Pam Lehman of Arrow International, Inc. (Reading, PA) for providingthe arterial catheters and introducers; Terry Shirey, PhD, of Nova Biomedical(Waltham, MA) for providing the Stat Ultra Blood Gas and ElectrolyteAnalyzer.
REFERENCES1. Brain Trauma Foundation. Guidelines for the Prehospital
Management of Traumatic Brain Injury. Guidelines for theManagement and Prognosis of Severe Traumatic Brain Injury.Guidelines for the Surgical Management of Traumatic Brain Injury.Available at: www.braintrauma.org.
2. Bowen TE, Bellamy RF, eds. Emergency War Surgery. SecondUnited States Revision of the Emergency War Surgery NATOHandbook. Washington, DC: United States Government PrintingOffice; 1988.
3. Kelly ME, Miller PR, Greenhaw JJ, Fabian TC, Proctor KG. Novelresuscitation strategy for pulmonary contusion after severe chesttrauma. J Trauma. 2003;55:94–105.
5. Wilkes NJ, Woolf R, Mutch M, et al. The effects of balanced versussaline-based hetastarch and crystalloid solutions on acid–base andelectrolyte status and gastric mucosal perfusion in elderly surgicalpatients. Anesth Analg. 2001;93:811–816.
6. Gan TJ, Bennett-Guerrero E, Phillips-Bute B, et al. Hextend, aphysiologically balanced plasma expander for large volume use inmajor surgery: a randomized phase III clinical trial. Hextend StudyGroup. Anesth Analg. 1999;88:992–998.
7. Cassada DC, Tribble CG, Long SM, et al. Adenosine A2A agonistreduces paralysis after spinal cord ischemia: correlation with A2A
Emergency Resuscitation for Traumatic Brain Injury
Volume 57 • Number 3 553
receptor expression on motor neurons. Ann Thorac Surg. 2002;74:846–849, discussion 849–850.
8. Cassada DC, Tribble CG, Young JS, et al. Adenosine A2A analogueimproves neurologic outcome after spinal cord trauma in the rabbit.J Trauma. 2002;53:225–229, discussion 229–31.
9. Cassada DC, Tribble CG, Long SM, et al. Adenosine A2A analogueATL-146e reduces systemic tumor necrosing factor-alpha and spinalcord capillary platelet-endothelial cell adhesion molecule-1expression after spinal cord ischemia. J Vasc Surg. 2002;35:994–998.
10. Cassada DC, Gangemi JJ, Rieger JM, et al. Systemic adenosine A2Aagonist ameliorates ischemic reperfusion injury in the rabbit spinalcord. Ann Thorac Surg. 2001;72:1245–1250.
11. Cassada DC, Tribble CG, Laubach VE, et al. An adenosine A2Aagonist, ATL-146e, reduces paralysis and apoptosis during rabbitspinal cord reperfusion. J Vasc Surg. 2001;34:482–488.
12. Cassada DC, Tribble CG, Kaza AK, et al. Adenosine analoguereduces spinal cord reperfusion injury in a time-dependent fashion.Surgery. 2001;130:230–235.
14. McPherson JA, Barringhaus KG, Bishop GG, et al. AdenosineA(2A) receptor stimulation reduces inflammation and neointimalgrowth in a murine carotid ligation model. Arterioscler Thromb VascBiol. 2001;21:791–796.
15. Glover DK, Ruiz M, Takehana K, et al. Pharmacological stressmyocardial perfusion imaging with the potent and selective A(2A)adenosine receptor agonists ATL193 and ATL146e administered byeither intravenous infusion or bolus injection. Circulation. 2001;104:1181–1187.
16. Riou LM, Ruiz M, Rieger JM, et al. Influence of propranolol,enalaprilat, verapamil, and caffeine on adenosine A(2A)-receptor-mediated coronary vasodilation. J Am Coll Cardiol. 2002;40:1687–1694.
17. Stone TW. Purines and neuroprotection. Adv Exp Med Biol. 2002;513:249–280.
18. Obata T. Adenosine production and its interaction with protection ofischemic and reperfusion injury of the myocardium. Life Sci. 2002;71:2083–2103.
19. Linden J. Molecular approach to adenosine receptors: receptor-mediated mechanisms of tissue protection. Annu Rev PharmacolToxicol. 2001;41:775–787.
20. Pope A, French G, Longnecker DE, eds. Fluid Resuscitation: Stateof the Science for Treating Combat Casualties and Civilian Injuries.Report from the Institute of Medicine. Washington, DC: NationalAcademy Press; 1999. Full text available on line at: http://www.nap.edu.
21. Bell MJ, Robertson CS, Kochanek PM, et al. Interstitial brainadenosine and xanthine increase during jugular venous oxygendesaturations in humans after traumatic brain injury. Crit Care Med.2001;29:399–404.
22. Robertson CS, Valadka AB, Hannay HJ, et al. Prevention ofsecondary ischemic insults after severe head injury. Crit Care Med.1999;27:2086–2095.
23. Fabian MJ, Proctor KG. Hemodynamic actions of acute ethanol afterresuscitation from traumatic brain injury. J Trauma. 2002;53:864–875.
24. Gibson JB, Maxwell RA, Schweitzer JB, Fabian TC, Proctor KG.Resuscitation from severe hemorrhagic shock after traumatic braininjury using saline, shed blood, or a blood substitute. Shock. 2002;17:234–244.
25. Glass TF, Fabian MJ, Schweitzer JB, Weinberg JA, Proctor KG. Theimpact of hypercarbia on the evolution of brain injury in the settingof traumatic brain injury and systemic hemorrhage. J Neurotrauma.2001;18:57–72.
28. Mattox KL, Maningas PA, Moore EE, et al. Prehospital hypertonicsaline/dextran infusion for posttraumatic hypotension. The U.S.A.Multicenter Trial. Ann Surg. 1991;213:482–491.
29. Wade CE, Grady JJ, Kramer GC, Younes RN, Gehlsen K, HolcroftJW. Individual patient cohort analysis of the efficacy of hypertonicsaline/dextran in patients with traumatic brain injury andhypotension. J Trauma. 1997;42(5 suppl):S61–S65.
30. Mazzoni MC, Borgstrom P, Intaglietta M, Arfors KE. Capillarynarrowing in hemorrhagic shock is rectified by hyperosmotic saline–dextran reinfusion. Circ Shock. 1990;31:407–418.
31. Menasche P, Jamieson WRE, Flameng W, Davies MK. Acadesine: anew drug that may improve myocardial protection in coronary arterybypass grafting: results of the first international multicenter study.J Thorac Cardiovasc Surg. 1995;110:1096–1106.
32. Mangano DT. Effects of acadesine on myocardial infarction, stroke,and death following surgery: a meta-analysis of the 5 internationalrandomized trials. The Multicenter Study of Perioperative Ischemia(McSPI) Research Group. JAMA. 1997;277:325–332.
34. Rieger JM, Brown ML, Sullivan GW, Linden J, Macdonald TL.Design, synthesis, and evaluation of novel A2A adenosine receptoragonists. J Med Chem. 2001;44:531–539.
35. Ongini E, Monopoli A, Cacciari B, Baraldi PG. Selective adenosineA2A receptor antagonists. Farmaco. 2001;56:87–90.
36. Dunwiddie TV, Masino SA. The role and regulation of adenosine inthe central nervous system. Annu Rev Neurosci. 2001;24:31–55.
37. King DR, Cohn SM, Proctor KG. Intracranial pressure, coagulationchanges, and neurologic outcome after resuscitation fromexperimental traumatic brain injury with Hetastarch. Surgery (inpress).
The Journal of TRAUMA� Injury, Infection, and Critical Care
