- Email: [email protected]
Hextend–perfluorocarbon cocktail inhibits mean arterial pressure response in a rabbit shock model
Accepted Manuscript Hextend-Pfc Cocktail Inhibits Map Response in A Rabbit Shock Model P.S. Reynolds, PhD, B.D. Spiess, MD PII:
S0022-4804(15)00495-3
DOI:
10.1016/j.jss.2015.04.063
Reference:
YJSRE 13328
To appear in:
Journal of Surgical Research
Received Date: 23 February 2015 Revised Date:
13 April 2015
Accepted Date: 17 April 2015
Please cite this article as: Reynolds P, Spiess B, Hextend-Pfc Cocktail Inhibits Map Response in A Rabbit Shock Model, Journal of Surgical Research (2015), doi: 10.1016/j.jss.2015.04.063. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT REVISED 4/09/2015
HEXTEND-PFC COCKTAIL INHIBITS MAP RESPONSE IN A RABBIT SHOCK MODEL
RI PT
PS Reynolds PhD, and BD Spiess MD Department of Anesthesiology, Virginia Commonwealth University Medical Center Richmond VA 23298
SC
P. S. Reynolds, Dept. Anesthesiology, Virginia Commonwealth University Medical Center, Richmond, VA 23298-0695; Tel 804-628-1965; email: psreynolds@ vcu.edu
M AN U
B. D. Spiess. Dept. Anesthesiology, Virginia Commonwealth University Medical Center, Richmond, VA; 23298-0695; Tel 804-828-2267; email: [email protected]
DISCLOSURES:
TE D
Corresponding author: P. S. Reynolds
This work was funded by Entegrion Inc. under subcontract from the Office
of Naval Research, Grant Number N00014-10-C-0333-VCU-MRF. Sponsors had no role in project design, data collection, analysis, or interpretation, or in the writing of the manuscript. The
EP
authors declare no other conflicts of interest
AC C
AUTHOR CONTRIBUTIONS: P.S.R. and B.D.S are responsible for the study concept. P.S.R. provided oversight for the experiments, and is responsible for study design and data analysis. P.S.R drafted the manuscript, and P.S.R. and B.D.S interpreted the data and performed critical revision of the manuscript for important intellectual content. P.S.R. and B.D.S had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
ACCEPTED MANUSCRIPT 2
ABSTRACT BACKGROUND: Hextend (HEX) is standard of care resuscitation fluid for combat-related traumatic hemorrhage. Because HEX has limited oxygen-carrying capacity, combination
RI PT
therapy with oxygen therapeutics could improve oxygen delivery following hemodynamic shock. We hypothesized that addition of perflurocarbon (PFC) to HEX would improve
hemodynamics and oxygen delivery marker response in a rabbit model of hemorrhagic shock.
SC
METHODS: Anesthetized New Zealand rabbits (n = 23) were randomly allocated to
resuscitation with fresh whole blood (FWB), HEX, or HEX plus PFC (HEX+PFC) following 60
M AN U
min of hemorrhagic hypotension. Mean arterial pressure (MAP) was sampled every 2-3 min for 120 min post-infusion; MAP profiles were modelled by a one-compartment pharmacokinetic model to determine peak MAP (Pmax), time to peak MAP (tmax), and post-infusion MAP persistence. Arterial blood was sampled every 15 min to examine pH, blood gases PO2 and
TE D
pCO2, metabolites lactate and glucose, methemoglobin (metHb), and electrolytes. RESULTS: Compared to FWB and HEX, HEX+ PFC administration resulted in delayed peak MAP, and less persistent (p < 0.0001) MAP elevation; metHb was significantly elevated (p <
EP
0.0001) compared to FWB and HEX. There were no significant differences in PO2, pCO2, or pH. Glucose, Hct, and Hb of both HEX and HEX+PFC were significantly lower relative to FWB.
AC C
Lactate clearance was modest and transient for all treatments; base deficit was significantly more negative for HEX+PFC.
CONCLUSIONS: Addition of PFC to HEX did not improve hemodynamics or acidosis. Further dose- and volume-range studies are required to test efficacy of PFC in combination with HEX for hemorrhagic shock.
KEY WORDS: PFC; lactate; hydroxyethyl starch; pharmacokinetic model
ACCEPTED MANUSCRIPT 3
1.1 Introduction Hextend (HEX) is the current combat casualty care field standard for fluid resuscitation
RI PT
following hypovolemic hemorrhagic shock. Current military/Tactical Field Care guidelines
recommend a 500-mL bolus of Hextend to restore or maintain field vital signs (palpable radial pulse, improved mentation) until definitive care; the bolus is repeated once if no improvement is
SC
noted in 30 minutes1. This represents a fluid load of roughly 12-14 mL/kg over 30 min for a 7580 kg patient. However, although HEX is an excellent volume expander, it has limited oxygen-
M AN U
carrying capacity. These low volumes may result in patient under-resuscitation if they are not sufficient to correct any persistent oxygen delivery/demand mismatch (and resulting oxygen debt) following hemorrhage.
Combinational therapy consisting of a reduced-volume therapeutic such as HEX with
TE D
additional agents targeted towards improved tissue oxygen delivery could result in increased efficacy; in addition a therapeutic that could be administered as a single bolus would have considerable logistic advantages in the far-forward environment3. Perfluorocarbon (PFC)
EP
emulsions are intravascular oxygen therapeutics that temporarily enhance tissue oxygenation
AC C
because of increased solubility of physiological gases relative to plasma4,5. Increased tissue O2 delivery is expected to minimize the potential for ischemic tissue damage5, and therefore the risk of multiple organ failure. Stand-alone PFC emulsions cannot be used as large-volume replacement fluids as they have no oncotic potential6; administration volumes have been limited to a maximum of 6 mL/kg in prior human testing. However, when PFCs are combined with standard plasma expanders animal data suggest that both systemic O2 delivery and extraction are increased, and in far greater proportions to those expected from the actual volume of PFC infused 7. Addition of PFC to resuscitation fluids, together with supplementary oxygen, has been
ACCEPTED MANUSCRIPT 4
demonstrated to result in modest improvements in both regional/tissue oxygen delivery6, 8 and systemic oxygenation6 in animal models of hemorrhagic shock.
RI PT
To be a useful adjunct, ideally PFC should act synergistically with HEX to compensate for the poor oxygen delivery capacity of HEX and improve hemodynamic support. However effects of PFC on hemodynamics are rarely evaluated directly, although improvements in systemic
11
SC
hemodynamics have been reported incidental to the primary outcomes of tissue oxygenation9, 10, .
M AN U
In this study we used a rabbit model of hemorrhagic shock to directly determine effects of combination therapy with PFC on hemodynamics, acidosis, and lactate clearance, a surrogate measure of oxygen debt. We hypothesized that animals treated with a combination of PFC and HEX should show increased benefit (in terms of improved MAP response, reduced acidosis, and
TE D
increased speed of lactate clearance) compared to HEX alone. We defined clinical effectiveness in terms of immediacy of effect (increase in MAP support to > 60 mmHg within 20-30 min of infusion), persistence of effect (hemodynamic support maintained for at least 2 hr post-
EP
resuscitation), and clinically-relevant outcomes (life support for 2 hr, and resolution of clinical markers of tissue hypoperfusion by 2 hr post-infusion). Materials and Methods
AC C
2
2.1 Animals and animal husbandry. This study was approved in advance by the Institutional Care and Use Committee of
Virginia Commonwealth University (VCU protocol number AD20285) and conforms to the Public Health Service Policy on Humane Care and Use of Laboratory Animals (2002). Twentyfour male SPF New Zealand White rabbits (Robinson Services Inc., Mocksville, NC) were housed individually in standard raised-flooring cages. Toys and grass hay were provided for
ACCEPTED MANUSCRIPT 5
enrichment. Animals were maintained at approximately 25oC and 12-h light/12-h dark for a minimum of one week in the animal colony before experimentation. Animals were fed
RI PT
commercial pelleted rabbit chow supplemented with grass hay and water ad lib.
2.2. Surgical methods
SC
Rabbit body weight averaged 3.24 (SD 0.29) kg on the day of experiment. Animals were not fasted prior to surgery. Animals were pre-medicated with ketamine/xylazine (35/3-4 mg/kg)
M AN U
for surgical preparation. Heart rate and pulse oximetry (sPO2, %) were monitored by a portable veterinary oximeter with the sensor placed on the tongue (PulseSense™, NONIN Medical Inc., Plymouth MN). Airway support was provided with a veterinary supra-glottic airway (Rabbit Vgel R3, Docsinnovent, UK). Animals were pre-oxygenated with 100% O2 until respiratory quality was adequate and sPO2 > 96%. Anesthetic plane was subsequently maintained by
TE D
isoflurane (1-3%, balance 100% O2), with positive end-expiratory pressure of 5-8 cmH2O, and flow rate 0.5-0.6 L/min; anesthetic plane was assessed every 3-5 min by vital signs and reflex
EP
response (corneal reflex, toe pinch, limb retraction).
AC C
Arterial and venous catheters were implanted by surgical isolation and cannulation of the left carotid artery, jugular vein, femoral artery, and femoral vein. Catheters were polyethylene PE-90 tubing and flushed with sterile heparinized normal saline (3 U heparin/mL NS) as needed in volumes sufficient to clear the dead space (< 100 uL) to minimize clotting in the lines while avoiding systemic heparinization. The carotid artery and jugular vein were used to monitor MAP and central venous pressure (CVP) respectively. During surgery, core temperature was monitored continuously with a rectal probe and maintained at 36.5-38oC with a thermostaticallycontrolled feedback heating blanket (Harvard Apparatus, Holliston MA). Surgical plane of
ACCEPTED MANUSCRIPT 6
anesthesia was assessed at least every 5 min by vital signs and reflex response (corneal reflex, toe pinch, limb retraction). Two or three blood samples (100 µL) were obtained during surgery to determine arterial pO2 and pCO2 for assessment of oxygenation and ventilation adequacy.
RI PT
Animals were allowed to stabilize for approximately 1 hour (median 61 min, IQR 56-70 min) past the end of surgery; physiological stabilization was defined as MAP > 50 mmHg, lactate < 2
SC
mmol/L, pO2 > 400 mmHg and pCO2 40-55 mmHg.
2.3 Hemorrhage protocol.
M AN U
Animals were hemorrhaged from the femoral artery. We performed a controlled stepwise hemorrhage by withdrawing blood with a programmable syringe pump (PHD22/2000 Series, Harvard Apparatus, Holliston MA), at 3 mL/kg/min until MAP declined to 40 mmHg, then continued at 1 mL/kg/min to maintain hypotension (MAP 35-40 mmHg) for 60 min. Hemorrhage
TE D
was paused if MAP remained between 35-40 mmHg. If MAP <30 mmHg autologous shed blood was returned immediately at 3 mL/kg if MAP < 30 mmHg and hemorrhage resumed when MAP
EP
> 40 mmHg.
2.4 Resuscitation protocol
AC C
Animals were resuscitated with 15 mL/kg test fluids via the femoral vein at a rate of
approximately 3 mL/min; median infusion time was 13 (SD 4) min. Animals received one of three treatment interventions: autologous fresh whole blood (FWB; gold standard control), Hextend® (HEX; TCCC standard control for resuscitation), or HEX combined with 3 mL/kg of 60% (w/v) perflubron (PFC) emulsion (Oxygent™, Sanguine/New Alliance Pharmaceutical Corp., Atlanta and San Diego, CA). HEX consists of 6% high-molecular weight (670 kDa) hydroxyethyl starch in lactated electrolyte solution (lactate 28 mEq/L, dextrose 0.99 g/L; Na+
ACCEPTED MANUSCRIPT 7
143 mEq/L, Cl− 124 mEq/L, Ca2+ 5 mEq/L, K+ 3 mEq/L; [package insert Hospira, Lake Forest, IL; 2004]). Animals were monitored continuously over the post-resuscitation period for 120 min, and then euthanized under deep anesthesia with sodium pentobarbital (Euthasol, 390 mg/mL
RI PT
pentobarbital sodium, 40 mg/kg IV, Virbac AH, Inc., Fort Worth, TX).
2.5. Hemodynamics
SC
Arterial systemic pressure was continuously monitored via the carotid artery with
pressure transducers (Transpac®, Abbott Critical Care Systems, Abbott Laboratories, North
M AN U
Chicago, IL) calibrated before each trial against a mercury sphygmomanometer. Signals were amplified (DA100C and TCD104A interface, BIOPAC Systems, Goleta, CA), and analog-digital conversions of signals to pressure readings were performed on-line at 500 Hz. Mean arterial pressure (MAP, mmHg) was calculated online from the smoothed arterial pressure average. All
TE D
data were collected and converted using AcqKnowledge™ software (Version 4.1, BIOPAC Systems, Goleta, CA). For analysis, data were captured over 20 s at 2-min intervals during
period.
EP
hemorrhage and resuscitation, and over 30 s at 2-5 min intervals during the post-resuscitation
AC C
2.6. Blood chemistry
Blood was sampled in 100 uL capillary tubes every 15 min during the hemorrhage
period, at the end of hemorrhage, and every 15 min for 2 hr post-resuscitation. Blood gases pO2 and pCO2 (mmHg), pH, hemoglobin [Hb] (g/dL), met-hemoglobin (metHb, %), hematocrit (Hct, %), base deficit (BD); metabolites lactate (mmol/L) and glucose (mmol/L), were determined using an ABL 800 analyzer (Radiometer America, Westlake, OH).
ACCEPTED MANUSCRIPT 8
2.7. Statistical methods The experimental design was a randomized block design; resuscitation fluid (FWB, HEX) and presence/absence of PFC were the fixed effects (treatments), and blocking was by
RI PT
animal batch (animals entered the holding colony in batches of six). Blocking minimizes
variation resulting from different holding times and time-related growth or size differences. Experimental run order was generated and randomized in PROC OPTEX (SAS 9.4). Power
SC
calculations were performed by simulation12, with MAP as the primary outcome variable. These calculations suggested that a sample size of 5-10 per group was adequate to detect a 20%
M AN U
specified difference between any two MAP trajectories in treatment, time, and (treatment x time) effects, with power > 0.8. Twenty-four animals were randomized to treatment by ID number using a computerized random-number algorithm (www. randomization.com). One animal died shortly following hemorrhage and was excluded from analysis; a total of 5 rabbits were assigned
TE D
to FWB control treatment, and 9 rabbits to each of the two HEX treatments. Interventions were not blinded to operators, but treatments were recoded and blinded for analysis.
EP
Post-infusion MAP data were modelled using a one-compartment pharmacokinetic model13 in PROC NLMIXED (SAS 9.4):
AC C
MAP(t) = b0 + b1 · t + b2 · δk · t · exp(−b3 · δk · t) + Z(t)
where t is time, b0 is a multiple point “baseline” describing pre-infusion MAP during the latter portion of the controlled hemorrhage period, b1 is the overall slope and represents the overall persistence of fluid effect, b2 controls the magnitude of peak change from baseline, and b3 models the “recovery” rate. In this study, ‘baseline’ refers to the controlled hemorrhage period when MAP is at a minimum and relatively constant, as the metric of interest is the MAP response to resuscitation. Pre-hemorrhage values are of interest only as a second control on
ACCEPTED MANUSCRIPT 9
physiological state pre-shock. Because treatment interventions were categorical, they were assigned dummy variables xi, coded 0 or 1 to discriminate between them14. The resuscitation fluid (treatment factor 1) with two levels (FWB, HEX) was assigned the dummy x1 , and
RI PT
presence/absence of PFC (factor 2) was assigned x2 with two levels (none, PFC added), such that FWB was represented by (x1, x2) = (0, 0), HEX = (1, 0), and HEX+PFC = (1, 1). Therefore the
‘effect size’δk for each treatment combination:
δk = γ0 + γ1·x1 + γ2·x2
SC
contribution of each component in the combination fluid could be evaluated by calculating
M AN U
where γ1 and γ2 are the regression coefficient estimates for x1 and x2 respectively, and measure the deviation of each component from the FWB standard (estimated by γ0). Treatment differences between curve shape estimates were assessed by contrasts15 on MAP peak change Pmax, estimated by (b2/b3) · e−1, and time to achieve peak MAP tmax, calculated from the first
TE D
derivative, where tmax =1/b3. Individual subject effects were modeled as random effects Z(t). The random-effect model is commonly used to estimate subject-specific effects, as it discriminates between observations within and between subjects, and accounts for inter-dependencies among
AC C
EP
responses within subjects15.
Blood chemistry data were analyzed by repeated-measures mixed model ANOVA in
PROC MIXED (SAS 9.4), with individual subject effects modeled as random effects within treatment group, and values of each variable at the end of hemorrhage as the baseline covariate. The best-fit covariance model for correlated errors was an autoregressive process ar(1) within group; this accounted for time-dependencies and heterogeneous variance between groups. Planned comparisons were made at 60 and 120 min, and family-wise error rates were controlled by Bonferroni’s correction, with statistical significance at p = 0.01. Selection bias was minimized
ACCEPTED MANUSCRIPT 10
because animals were properly randomized to both treatment and experimental run order, and baseline imbalances were controlled statistically; therefore no inferential statistical tests were performed on baseline (pre- and post- hemorrhage) characteristics16. Descriptive statistics are
RI PT
reported as means and standard deviations (SD) for normally-distributed continuous data; parameter estimates and effect sizes are reported as means and 95% confidence intervals.
SC
3. Results
Group descriptive statistics for pre- and post-hemorrhage volume and blood chemistry
M AN U
values are shown in Table 1. Average shed blood volume was 51 mL (SD 17 mL), or 16 (SD 5) mL/kg; this represents approximately 35% estimated total blood volume for a 3.3 kg rabbit17. Lactate and base deficit at the end of hemorrhage averaged 3.6 (SD 1.9) and -4.5 (SD 2.9) mmol/L respectively, representing a modest shock insult18,19. Average resuscitation volumes
TE D
were 51 (SD 3) mL, representing between 85-102% of shed blood volume.
Estimates of MAP trajectory were strongly affected by PFC (Table 2, Fig. 1). Addition of
EP
PFC to HEX resulted in a 10-min delay in peak MAP response to resuscitation, and a 4-8 mmHg lower peak MAP. Post-infusion MAP ‘persistence’, estimated from b1, was inhibited by PFC (p
AC C
< 0.0001), averaging an increase of approximately 4 mmHg improvement over 60 min for HEX+PFC, as compared to 13-14 mmHg for FWB and HEX.
Blood chemistry values for each treatment group at 60 and 120 min are given in Table 3.
There were no statistically-significant treatment or (treatment x time) effects for pH (p = 0.14; 0.23), pO2 (p = 0.65; 0.14), pCO2 (p = 0.90; 0.89), [Na+] (p = 0.14; 0.38) and [Ca2+] (p = 0.22; 0.11). HEX + PFC animals were clinically acidotic (pH < 7.35) at both time periods relative to
ACCEPTED MANUSCRIPT 11
control groups. In comparison to controls, HEX+PFC animals showed elevated [Cl− ] and reduced [K+ ] and [HCO3−]. MetHb for HEX+PFC was significantly elevated over the entire post-resuscitation period relative to both FWB and HEX alone (p = 0.013) averaging 1.2%,
RI PT
compared to 0.5% for both FWB and HEX (Fig. 2). Lactate (Fig. 3) increased for the duration of the infusion period, showed modest clearance to 2.9 (SD 1.5) mmol/L for approximately 60 min post-infusion then increased over the next 60 min to 3.6 (SD 1.6), mmol/L, levels exceeding
SC
those observed pre-infusion. However there were no significant differences between treatments at designated time points (p > 0.5). In contrast, base excess became progressively more negative
M AN U
for HEX+PFC animals, exceeding < -6 mmol/L (the so-called ‘critical shock threshold’19 ) at both designated post-resuscitation time points. Glucose of both HEX and HEX+PFC was depressed relative to FWB (p = 0.004; Fig. 4). Following hemodilution, as expected, Hct/Hb in both HEX treatments were significantly lower relative to those laboratory measurements in FWB
4. Discussion
TE D
(p < 0.001).
EP
Combination therapy involves two or more pharmacologic agents administered either separately or in a single fixed-dose combination, with the therapeutic goals of optimizing
AC C
response and simplifying administration. Previous studies have indicated that PFC improves both whole body and regional oxygenation, and is thought to be beneficial for treating decompression sickness20,21 and acute dilutional anemia22. As adjuncts to conventional fluid resuscitation, PFCs have the potential to maintain or restore tissue oxygenation in the critical time window following injury and before bleeding control and definitive care6. However to be a useful adjunct, addition of PFC should, at the least, not detract from the hemodynamic support conferred by volume expander fluids; ideally, PFC should act synergistically with HEX both to compensate for the
ACCEPTED MANUSCRIPT 12
poor oxygen delivery capacity of HEX and improve hemodynamic support. Scattered data suggested that PFC augmentation of resuscitation fluids for treating hemorrhagic hypovolemic
RI PT
shock may result in improved MAP support as well as regional tissue oxygenation 8, 10.
However, in this preclinical trial evaluating hemodynamic support as a primary outcome, addition of 3 mg/kg PFC to HEX was of no discernable benefit in correcting hemodynamics
SC
compared to either FWB or HEX alone. Addition of PFC to HEX was associated with more prolonged times to MAP resolution and reduced post-infusion MAP persistence. As reduction in
M AN U
pH is predominantly driven by changes in strong ions and pCO2, we had also predicted that addition of PFC would correct or blunt shock-induced acidosis through increased CO2 scavenging from the blood, thus driving changes in electrolytes. However, at the doses and administration rates used in this study, treatment differences were clinically modest for arterial
TE D
pCO2, strong ions, lactate, and [HCO3̶ ]; other acid-base disruptions noted were clinicallysignificant metHb elevation, negative base excess below the critical shock threshold, mild to
EP
modest hyperchloremia, and moderate to poor glucose dynamics.
Elevated metHb is indicative of RBC oxidative stress and oxidation of Hb, and reduces
AC C
Hb capacity for oxygen transfer. The elevated metHb levels noted for the PFC group, in spite of supplementary O2, suggests a strong association with PFC administration, rather than anesthesia agents or HEX alone, and may be related to PFC-induced nitric oxide (NO) release23. The depression in Hct and Hb noted for both HEX treatments is not surprising, given the degree of hemodilution involved in replacing almost 35% shed blood volume with non-blood fluid24. However, compared to HEX controls, the clinically and statistically significant negative base excess and reduction in [HCO3-] noted for HEX+PFC animals indicates a metabolic acidosis not
ACCEPTED MANUSCRIPT 13
accounted for by the dilution effects of HEX alone. Unmeasured changes in other ions and weak acids (such as albumin and phosphates) may account for these differences, and indicate the
RI PT
potential for PFC-induced renal acidosis.
Lactate clearance is a surrogate for both the magnitude and duration of tissue hypoxia and oxygen debt 2, 19, 25. We observed that lactate clearance was modest and transient for all
SC
resuscitation treatments in this study, including blood. This suggests that all animals were underresuscitated by the single 15 mL/kg bolus used in this study, even though resuscitation volumes
M AN U
approximated the shed blood volume. This is of practical interest as failure to correct persistent shock-related hypoperfusion within a limited time window may ultimately contribute to multiple organ failure and late death2, and combat casualty care fluid resuscitation guidelines suggest initial resuscitation volumes substantially less than those used in this study1. However the
TE D
discrepancy between the magnitude of post-resuscitation lactate and base excess values for the PFC group suggest that clinical interpretation of these values may not be straightforward. In clinical populations, hyperchloremic acidosis may be associated with abnormal base excess
EP
values without profound lactate acidosis, either from changes in absolute [Cl-], changes in [Cl-] relative to Na+]26, or excessive elimination of [HCO3-] from the kidneys27. In this study, group
AC C
differences in [Na+] and [Cl-] were not particularly large; nevertheless [HCO3-] were greatly affected by the addition of PFC. These data further support the warnings against over-reliance on base deficit as a single surrogate index of tissue oxygenation without consideration of other components of acid-base balance.26
4.1. Limitations and adverse events. There are two limitation of this study. First we used healthy, young, and anesthetized animals
ACCEPTED MANUSCRIPT 14
subjected to controlled hemorrhage. This model does not closely resemble likely clinical situations, namely polytrauma with uncontrolled hemorrhage, or alternatively, high-risk surgical patients. However, because the shock insult is induced to a controlled and repeatable endpoint,
RI PT
this model does provide a standardized, controlled and reliable platform to assess physiological response to the test interventions28, which was the primary objective of this study. Second, this study was primarily designed to evaluate fundamental and clinically-relevant markers of
M AN U
or specific side effects associated with either PFC or HEX.
SC
response to resuscitation, namely MAP and blood chemistry, and not tissue oxygenation per se
The issue of adverse events occurring with both hydroxyethyl starch solutions and therapeutic adjuncts, such as PFC, is of increasing concern. HEX is falling out of favor for fluid replacement therapy, and at present no PFC is approved for human clinical use. Use of hydroxyethyl starch
TE D
solutions for fluid replacement has been associated with increased mortality, renal failure, refractory pruritus, and especially coagulation abnormalities29, 30. In rabbit models, hemodilution with HEX has been associated with reductions in clot propagation and strength, primarily
EP
because of disruptions in F-XIIIa-fibrin polymer crosslinking24. Documented adverse events of PFCs include cardiovascular and central nervous system effects, macrophage and complement
AC C
activation, cytokine release, and platelet reduction31. However synergisms between PFC emulsions and hydroxyethyl starch solutions have been noted that may result in exacerbated coagulation abnormalities. Aggregation formation induced by PFC emulsions in the presence of red blood cells and hydroxyethyl starch solutions32, or between PFC and dextran33, result in clumping of leukocytes and/or erythrocytes, endothelial cell damage, impaired rheology, restriction of capillary flow, and capillary collapse. Hyperchloremia and excessive bicarbonate elimination has been demonstrated to affect eicosanoid release in renal tissue, resulting in
ACCEPTED MANUSCRIPT 15
vasoconstriction, reduction of the glomerular filtration rate, and hypotension27. It is highly possible that synergistic interactions between hydroxyethyl starch present in HEX and PFC emulsions induced coagulation, complement and biolipid inflammatory mediator abnormalities
RI PT
that contributed to the aberrations in MAP and blood chemistry observed in this study.
5. Conclusions
SC
PFC has promise as an oxygen therapeutic adjunct and HEX has had common clinical use as a volume expander. However, adverse events have been documented with use of these products in
M AN U
both preclinical and clinical assessments. In this study, significant disruptions of clinicallyrelevant ‘markers’ of response to resuscitation, namely MAP and blood chemistry, were observed with the use of PFC in conjunction with HEX. Before these products can be reconsidered for therapeutic use in hemorrhagic shock, both dose- and volume-range preclinical
TE D
studies should be performed in the context of combination trials. Evaluation of candidate agents and solutions will require the application of sophisticated experimental designs such as mixture designs (where the response or outcome is a function of the proportions of the therapeutic
EP
components, rather than volume per se)34, and designs incorporating different absolute total volumes and timing of administration relative to point of injury. A wider variety of outcomes
AC C
related to possible adverse safety profiles, such as coagulation profiles, complement, and end organ cell damage must be assessed rigorously in preclinical models.
AUTHORSHIP
P.S.R. and B.D.S are responsible for the study concept. P.S.R. provided oversight for the experiments, and is responsible for study design and data analysis. P.S.R drafted the manuscript, and P.S.R. and B.D.S interpreted the data and performed critical revision of the manuscript for
ACCEPTED MANUSCRIPT 16
important intellectual content. P.S.R. and B.D.S had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
RI PT
DISCLOSURE
Financial support for this research was provided by Entegrion Inc. Sponsors had no role in project design, data collection, analysis or interpretation, or in the writing of the manuscript. The authors
M AN U
ACKNOWLEDGMENTS
SC
declare no other conflicts of interest.
AC C
EP
TE D
We thank AR Morris, M Ellenberg, K Bass, and S Khan for excellent technical assistance.
ACCEPTED MANUSCRIPT 17
REFERENCES [1]
Salomone JP, Pons PT. Tactical field care. In: Salomone JP, Pons PT, eds. PHTLS
RI PT
Prehospital Trauma Life Support: Military Version. St. Louis MO: Mosby-Elsevier; 2007: 516-537.
Barbee RW, Reynolds PS, Ward KR. Assessing shock resuscitation strategies by oxygen debt repayment. Shock 2010; 33 :113-122.
[3]
SC
[2]
Dubick MA, Atkins JL. Small-volume fluid resuscitation for the far-forward combat
[4]
M AN U
environment: current concepts. J Trauma 2003;54: S43-S45.
Riess JG. Perfluorocarbon-based oxygen delivery. Artif Cells Blood Subs Biotechn 2006;34:567-580.
[5]
Spiess BD. Perfluorocarbon emulsions as a promising technology: a review of tissue and
[6]
TE D
vascular gas dynamics. J Appl Physiol 2009; 106: 1444 –1452, 2009. Kemming GI, Meisner FG, Wojtczyk CJ, Packert KB, Minor T, Thiel M, Tillmanns J, Meier J, Bottino D, Keipert PE, Faithfull S, Habler OP. OxygentTM as a top load to colloid
EP
and hyperoxia is more effective in resuscitation from hemorrhagic shock than colloid and hyperoxia alone. Shock 2005; 24:245–254 Padnick LB, Linsenmeier RA, Goldstick TK. Perfluorocarbon emulsion improves
AC C
[7]
oxygenation of the cat primary visual cortex. J Appl Physiol 1999; 86:1497–1504. [8]
Audonnet-Blaise S, Krafft M-P, Smania Y, Mertes P-M, Marie P-Y, Labrude P, Longrois D, Menu P. Resuscitation of severe but brief haemorrhagic shock with PFC in rabbits restores skeletal muscle oxygen delivery and does not alter skeletal muscle metabolism. Resuscitation 2006;70:124-132.
ACCEPTED MANUSCRIPT 18
[9]
Johnson EC, Erickson BK, Podolsky A, Birks EK, Keipert PE, Faithfull NS, Wagner PD. Effects of a perfluorocarbon emulsion for enhanced O2 solubility on hemodynamics and O2 transport in dogs. J Appl Physiol 1995; 79:1777–1786.
RI PT
[10] Paxian M, Keller SA, Huynh TT, Clemens MG. Perflubron emulsion improves hepatic microvascular integrity and mitochondrial redox state after hemorrhagic shock. Shock, 2003; 20: 449–457.
SC
[11] Zhou Z, Sun D, Levasseur JE, Merenda A, Hamm RJ, Zhu J, Spiess BD, Bullock MR. perfluorocarbon emulsions improve cognitive recovery after lateral fluid percussion brain
M AN U
injury in rats. Neurosurgery 2008;63:799–807.
[12] Littell R, Milliken GA, Stroup WW, Wolfinger RD, Schabenberger O. SAS for mixed models. Cary NC: SAS Institute Inc.; 2006.
[13] Sánchez BN, Wu M, Raghunathan TE, Diez-Roux AV. Modeling the salivary cortisol
2012; 176: 918-928
TE D
profile in population research: The multi-ethnic study of atherosclerosis. Amer J Epidemiol
[14] Draper NR, Smith H: Applied regression analysis, 3rd ed, John Wiley & Sons, New York,
EP
pp. 299-317, 1998
[15] Brown H, Prescott R: Applied mixed models in medicine. Chichester UK: John Wiley &
AC C
Sons, Ltd., pp. 455, 2006.
[16] Altman DG: Practical statistics for medical research. London UK: Chapman & Hall. pp 611, 1991.
[17] Mott JC. The relation of blood volume to body weight and arterial haemoglobin levels in rabbits. J Physiol 1967;191:131-140.
ACCEPTED MANUSCRIPT 19
[18] Rixen D, Raum M, Holzgraefe B, Sauerland S, Nagelschmidt M, Neugebauer E, Shock and Trauma Study Group. A pig hemorrhagic shock model: oxygen debt and metabolic acidemia as indicators of severity. Shock 2001;16:239-244
RI PT
[19] Rixen D, Siegel JH: Bench-to-bedside review: oxygen debt and its metabolic correlates as quantifiers of the severity of hemorrhagic and post-traumatic shock. Crit Care 2005; 9:441453.
SC
[20] Smith CR, Parsons JT, Zhu J, Spiess BD. The effect of intravenous perfluorocarbon
emulsions on whole-body oxygenation after severe decompression sickness. Diving and
M AN U
Hyperbaric Medicine 2012;42(1):10-17.
[21] Zhu J, Hullett JB, Somera L, Barbee RW, Ward KR, Berger BE, Spiess BD. Intravenous perfluorocarbon emulsion increases nitrogen washout after venous gas emboli in rabbits Undersea and Hyperbaric Medicine 2007;34(1): 7-20.
TE D
[22] Cabrales P, Briceño JC. Delaying blood transfusion in experimental acute anemia with a perfluorocarbon emulsion. Anesthesiology 2011; 114(4): 901–911 [23] Ash-Bernal R, Wise R, Wright SM. Acquired methemoglobinemia: a retrospective series of
[24]
EP
138 cases at 2 teaching hospitals. Medicine. 2004; 83(5):265-73. Nielsen VG. Effects of
AC C
Hextend hemodilution on plasma coagulation kinetics in the rabbit: role of factor XIIImediated fibrin polymer crosslinking. Journal of Surgical Research 2006;132, 17–22. [25] Nguyen HB, Rivers EP, Knoblich BP, Jacobsen G, Muzzin A, Ressler JA, Tomlanovich MC. Early lactate clearance is associated with improved outcome in severe sepsis and septic shock Crit Care Med 2004; 32:1637–1642. [26] Brill SA, Stewart TR, Brundage SI, Schreiber MA. Base deficit does not predict mortality when secondary to hyperchloremic acidosis. Shock 2002; 17( 6): 459–462
ACCEPTED MANUSCRIPT 20
[27] Bullivant EMA, Wilcox CS, Welch WJ: Intrarenal vasoconstriction during hyperchloremia: role of thromboxane. Am J Physiol 1989, 256:152-157.
animal models. Shock 2005; 24: 33-39.
RI PT
[28] Lomas-Niera JL, Perl M, Chung C-S, Ayala A: Shock and hemorrhage: an overview of
[29] Wiedermann CJ. Hydroxyethyl starch – can the safety problems be ignored? Wien Klin Wochenschr. 2004;116(17-18):583-594.;
SC
[30] Hartog CS, Welte T, Schlattmann P, Reinhart K: Fluid replacement with hydroxyethyl starch in critical care—a reassessment. Deutsch Arztebl Int 2013; 110(26): 443−50.
M AN U
[31] Castro CB, Briceno JC. Perfluorocarbon-based oxygen carriers: review of products and trials. Artificial Organs 2010; 34(8):622–634
[32] Vásques, DM, Ortiz D, Alvarez O, Briceñ JC, Cabrales P. Hemorheological implications of perfluorocarbon based oxygen carriers, interactions with colloid plasma expanders and
TE D
blood. Biotechnol Prog 2013;29: 796-807.
[33] Nolte D, Pickelmann S, Lang M, Keipert P, Messmer K. Compatibility of different colloid plasma expanders with perflubron emulsion; an intravital microscope study in the hamster.
EP
Anesthesiology 2000; 93:1261-70.
[34] Cornell JA. A primer on experiments with mixtures. Hoboken NJ: John Wiley & Sons,
AC C
2011
ACCEPTED MANUSCRIPT 21
FIGURE LEGENDS Fig. 1. Average MAP profiles of rabbits following 60 min of hemorrhagic hypotension and resuscitation with FWB, HEX, or HEX +PFC, estimated from a one-compartment
achieve peak MAP.
RI PT
pharmacokinetics model of MAP over post-infusion time. Dotted lines represent time to
Fig. 2. Average arterial metHb (%) of rabbits following resuscitation with FWB, HEX, or HEX
SC
+ PFC. Error bars represent SE at each time point.
Fig. 3. Arterial lactate and base deficit (mmol/L) of rabbits following resuscitation with FWB,
M AN U
HEX, or HEX +PFC. Error bars represent SE at each time point.
Fig. 4. Arterial glucose (mmol/L) of rabbits following resuscitation with FWB, HEX, or HEX
AC C
EP
TE D
+PFC. Error bars represent SE at each time point.
ACCEPTED MANUSCRIPT 1
Table 1. Pre- and post-hemorrhage characteristics of rabbits in each study group (FWB, HEX, HEX+ PFC) HEX n=9
Mean
SD
Mean
Body mass (kg)
3.1
0.1
3.4
Shed blood volume (mL)
54
12
59
Resuscitation volume (mL)
47
2
50
Pre-hemorrhage pH 7.342
SD
Mean
SD
0.2
3.2
0.3
16
50
9
3
52
5
SC
Blood chemistry
HEX+PFC n=9
RI PT
FWB n=5
0.143
7.363
0.047
7.344
0.048
49
4
53
9
54
21
pO2 (mmHg)
467
55
499
37
498
40
Hb (g/dL)
10.5
1.0
11.6
0.8
11.2
1.0
metHb (%)
0.4
0.4
0.2
0.4
0.0
0.4
Hct (%)
32.4
2.9
35.5
2.4
34.4
3.0
K+ (mmol/L)
2.7
0.5
2.8
0.6
3.0
0.6
142
3
140
3
142
3
Ca (mmol/L)
2.5
0.2
2.6
0.2
2.6
0.2
Cl ̶ (mmol/L)
106
6
104
5
103
5
Glucose (mmol/L)
9.9
3.7
13.0
4.7
11.4
2.1
Lactate (mmol/L)
1.6
0.7
1.3
0.4
1.7
0.7
Base excess (mmol/L)
2.8
4.2
2.4
2.8
2.5
3.7
28.2
3.2
27.2
2.4
28.2
3.9
0.115
7.347
0.061
7.337
0.055
TE D
Na+ (mmol/L)
AC C
EP
2+
M AN U
pCO2 (mmHg)
HCO3- (mmol/L)
End of hemorrhage pH 7.323
pCO2 (mmHg)
50
19
41
9
37
6
pO2 (mmHg)
516
41
516
50
476
31
Hb (g/dL)
9.0
1.2
9.7
1.4
9.4
0.0
metHb (%)
0.5
0.5
0.2
0.4
0.0
0.3
Hct (%)
28.1
3.5
29.3
3.6
29.0
3.7
ACCEPTED MANUSCRIPT 2
K+ (mmol/L)
1.0
3.4
0.9
2.9
0.3
Na (mmol/L)
137
4
137
4
142
2
Ca2+ (mmol/L)
2.6
0.3
2.4
0.3
2.3
0.1
Cl- (mmol/L)
104
8
106
8
112
3
Glucose (mmol/L)
16.6
3.7
16.2
6.6
11.1
4.1
Lactate (mmol/L)
3.9
1.5
4.0
1.7
3.0
2.0
Base excess (mmol/L)
-3.4
2.8
-3.6
3.0
-5.9
2.1
HCO3- (mmol/L)
24.5
4.7
21.2
3.1
18.6
2.0
AC C
EP
TE D
M AN U
SC
RI PT
3.9
+
ACCEPTED MANUSCRIPT 3
Table 2. Estimates for post-infusion MAP ‘persistence’ (b1), peak MAP Pmax, and time to reach Pmax (tmax) calculated from a one-compartment pharmacokinetics model of MAP over post-infusion time in rabbits hemorrhaged to 35% TBV and then resuscitated with FWB (n = 5), HEX (n = 9), or HEX+PFC (n = 9). Difference from
FWB
HEX
95% CI Estimate
Lower,
Upper
|t|
FWB
0.21
0.17
0.24
.
HEX
0.23
0.20
0.26
0.79
HEX+PFC
0.06
0.03
0.09
FWB
20
HEX
22
HEX+PFC
17
tmax (min) 16
HEX
AC C
p
.
.
.
0.38
.
.
<0.0001
69.15
<0.0001
24
.
.
.
.
19
25
1.09
0.29
.
.
14
20
1.24
0.23
2.78
0.011
13
18
.
.
.
.
18
16
20
1.50
0.15
.
.
26
22
30
4.64
0.0001
3.93
0.0007
EP
HEX+PFC
|t|
16
TE D
FWB
47.25
M AN U
Pmax (mmHg)
p
SC
b1 (mmHg/min)
RI PT
Difference from
ACCEPTED MANUSCRIPT 4
Table 3. Descriptive statistics for blood chemistry of hemorrhaged rabbits resuscitated with fresh whole blood (FWB); Hextend alone, or Hextend augmented with 3 mL/kg PFC, at (A) 60 min post-resuscitation; (B) 120 min post-resuscitation. Statistical differences between standard FWB and each HEX group are indicated by p-values.
HEX n=9
FWB n=5
RI PT
A. 60 min post-resuscitation HEX+PFC n=9
HEX vs HEX+PFC
SD
Mean
SD
p
Mean
SD
p
p
7.357
0.039
7.365
0.019
0.85
7.309
0.025
0.32
0.09
pCO2 (mmHg)
41
5
41
3
0.97
38
2
0.67
0.51
pO2 (mmHg)
487
25
490
15
0.91
500
27
0.72
0.75
Hb (g/dL)
9.8
0.6
7.0
0.3
<0.0001
7.5
0.4
0.002
0.36
metHb (%)
0.46
0.25
0.55
0.14
0.77
1.44
0.22
0.010
0.002
Hct (%)
30.3
1.7
21.8
0.8
<0.0001
23.3
1.3
0.002
0.35
K+ (mmol/L)
3.0
0.4
3.0
0.2
0.96
2.6
0.1
0.279
0.07
Na+ (mmol/L)
141
1
141
1
0.93
142
1
0.44
0.45
Ca2+ (mmol/L)
2.3
0.1
2.3
0.1
0.59
2.2
0.1
0.44
0.81
Cl- (mmol/L)
109
Glucose (mmol/L)
13.5
Lactate (mmol/L)
3.2
Base excess (mmol/L)
-2.9
-
M AN U
TE D
21.8
AC C
HCO3 (mmol/L)
3
110
2
0.89
114
2
0.146
0.08
1.6
10.8
1.7
0.25
8.5
0.8
0.009
0.22
0.6
3.0
0.5
0.86
2.7
0.5
0.57
0.64
1.6
-2.5
1.1
0.83
-6.4
1.1
0.075
0.017
1.6
22.2
1.1
0.83
18.8
1.1
0.12
0.035
EP
pH
SC
Mean
ACCEPTED MANUSCRIPT 5 (b) 120 min post-resuscitation
Mean
SD
p
Mean
SD
p
HEX vs HEX+PFC p
7.365
0.022
0.81
7.252
0.031
0.05
0.005
40
3
0.74
0.75
481
36
0.83
0.57
Mean
SD
42
5
39
3
0.57
pO2 (mmHg)
490
25
457
19
0.30
Hb (g/dL)
10.2
0.6
7.3
0.7
<0.0001
7.1
0.5
0.0002
0.84
metHb (%)
0.44
0.25
0.46
0.17
0.94
1.16
0.26
0.06
0.032
31.5
1.7
21.9
1.0
K (mmol/L)
3.2
0.4
3.0
0.3
+
141
1
141
1
2+
2.6
0.1
2.2
-
Cl (mmol/L)
108
3
Glucose (mmol/L)
11.8
1.6
Lactate (mmol/L)
3.9
0.6
Base excess (mmol/L)
-2.0
1.6
HCO3-(mmol/L)
22.8
1.6
AC C
EP
Ca (mmol/L)
22.3
1.6
0.0002
0.81
0.61
2.5
0.2
0.08
0.11
0.85
144
2
0.27
0.21
0.1
0.03
2.2
0.1
0.007
0.88
112
2
0.35
116
2
0.04
0.19
8.3
2.0
0.18
8.5
1.1
0.09
0.94
3.2
0.6
0.46
3.6
0.5
0.69
0.69
-3.4
1.4
0.50
-9.1
1.5
0.002
0.008
21.2
1.4
0.44
16.8
1.4
0.005
0.030
TE D
Hct (%) +
Na (mmol/L)
<0.0001
M AN U
pCO2 (mmHg)
SC
pH 7.354 0.039
HEX+PFC
RI PT
HEX
FWB
ACCEPTED MANUSCRIPT
70
HEX FWB
RI PT
60
SC
55 50
HEX +PFC
45 40 35 30 20
40
60
TE D
0
M AN U
MAP mmHg
65
80
AC C
EP
Time (min)
100 120
ACCEPTED MANUSCRIPT
2.0
RI PT
1.0
SC
0.5
FWB HEX HEX+PFC
M AN U
metHb %
1.5
0.0
TE D
-0.5
AC C
EP
Time
ACCEPTED MANUSCRIPT
5.0
RI PT
4.0 3.5 3.0
2.5
SC
2.0
FWB
1.5
HEX
M AN U
Lactate, mmol/L
4.5
1.0
AC C
EP
TE D
0.5
HEX+PFC
ACCEPTED MANUSCRIPT
A.
FWB HEX
2
HEX+PFC
0
RI PT
-2 -4 -6
-8 -10
SC
Base excess, mmol/L
4
M AN U
-12
Time
25
TE D
30
FWB
HEX HEX+PFC
EP
Bicarbonate [HCO3-] mmol/L
B.
AC C
20
15
Time
S0022-4804(15)00495-3
DOI:
10.1016/j.jss.2015.04.063
Reference:
YJSRE 13328
To appear in:
Journal of Surgical Research
Received Date: 23 February 2015 Revised Date:
13 April 2015
Accepted Date: 17 April 2015
Please cite this article as: Reynolds P, Spiess B, Hextend-Pfc Cocktail Inhibits Map Response in A Rabbit Shock Model, Journal of Surgical Research (2015), doi: 10.1016/j.jss.2015.04.063. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT REVISED 4/09/2015
HEXTEND-PFC COCKTAIL INHIBITS MAP RESPONSE IN A RABBIT SHOCK MODEL
RI PT
PS Reynolds PhD, and BD Spiess MD Department of Anesthesiology, Virginia Commonwealth University Medical Center Richmond VA 23298
SC
P. S. Reynolds, Dept. Anesthesiology, Virginia Commonwealth University Medical Center, Richmond, VA 23298-0695; Tel 804-628-1965; email: psreynolds@ vcu.edu
M AN U
B. D. Spiess. Dept. Anesthesiology, Virginia Commonwealth University Medical Center, Richmond, VA; 23298-0695; Tel 804-828-2267; email: [email protected]
DISCLOSURES:
TE D
Corresponding author: P. S. Reynolds
This work was funded by Entegrion Inc. under subcontract from the Office
of Naval Research, Grant Number N00014-10-C-0333-VCU-MRF. Sponsors had no role in project design, data collection, analysis, or interpretation, or in the writing of the manuscript. The
EP
authors declare no other conflicts of interest
AC C
AUTHOR CONTRIBUTIONS: P.S.R. and B.D.S are responsible for the study concept. P.S.R. provided oversight for the experiments, and is responsible for study design and data analysis. P.S.R drafted the manuscript, and P.S.R. and B.D.S interpreted the data and performed critical revision of the manuscript for important intellectual content. P.S.R. and B.D.S had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
ACCEPTED MANUSCRIPT 2
ABSTRACT BACKGROUND: Hextend (HEX) is standard of care resuscitation fluid for combat-related traumatic hemorrhage. Because HEX has limited oxygen-carrying capacity, combination
RI PT
therapy with oxygen therapeutics could improve oxygen delivery following hemodynamic shock. We hypothesized that addition of perflurocarbon (PFC) to HEX would improve
hemodynamics and oxygen delivery marker response in a rabbit model of hemorrhagic shock.
SC
METHODS: Anesthetized New Zealand rabbits (n = 23) were randomly allocated to
resuscitation with fresh whole blood (FWB), HEX, or HEX plus PFC (HEX+PFC) following 60
M AN U
min of hemorrhagic hypotension. Mean arterial pressure (MAP) was sampled every 2-3 min for 120 min post-infusion; MAP profiles were modelled by a one-compartment pharmacokinetic model to determine peak MAP (Pmax), time to peak MAP (tmax), and post-infusion MAP persistence. Arterial blood was sampled every 15 min to examine pH, blood gases PO2 and
TE D
pCO2, metabolites lactate and glucose, methemoglobin (metHb), and electrolytes. RESULTS: Compared to FWB and HEX, HEX+ PFC administration resulted in delayed peak MAP, and less persistent (p < 0.0001) MAP elevation; metHb was significantly elevated (p <
EP
0.0001) compared to FWB and HEX. There were no significant differences in PO2, pCO2, or pH. Glucose, Hct, and Hb of both HEX and HEX+PFC were significantly lower relative to FWB.
AC C
Lactate clearance was modest and transient for all treatments; base deficit was significantly more negative for HEX+PFC.
CONCLUSIONS: Addition of PFC to HEX did not improve hemodynamics or acidosis. Further dose- and volume-range studies are required to test efficacy of PFC in combination with HEX for hemorrhagic shock.
KEY WORDS: PFC; lactate; hydroxyethyl starch; pharmacokinetic model
ACCEPTED MANUSCRIPT 3
1.1 Introduction Hextend (HEX) is the current combat casualty care field standard for fluid resuscitation
RI PT
following hypovolemic hemorrhagic shock. Current military/Tactical Field Care guidelines
recommend a 500-mL bolus of Hextend to restore or maintain field vital signs (palpable radial pulse, improved mentation) until definitive care; the bolus is repeated once if no improvement is
SC
noted in 30 minutes1. This represents a fluid load of roughly 12-14 mL/kg over 30 min for a 7580 kg patient. However, although HEX is an excellent volume expander, it has limited oxygen-
M AN U
carrying capacity. These low volumes may result in patient under-resuscitation if they are not sufficient to correct any persistent oxygen delivery/demand mismatch (and resulting oxygen debt) following hemorrhage.
Combinational therapy consisting of a reduced-volume therapeutic such as HEX with
TE D
additional agents targeted towards improved tissue oxygen delivery could result in increased efficacy; in addition a therapeutic that could be administered as a single bolus would have considerable logistic advantages in the far-forward environment3. Perfluorocarbon (PFC)
EP
emulsions are intravascular oxygen therapeutics that temporarily enhance tissue oxygenation
AC C
because of increased solubility of physiological gases relative to plasma4,5. Increased tissue O2 delivery is expected to minimize the potential for ischemic tissue damage5, and therefore the risk of multiple organ failure. Stand-alone PFC emulsions cannot be used as large-volume replacement fluids as they have no oncotic potential6; administration volumes have been limited to a maximum of 6 mL/kg in prior human testing. However, when PFCs are combined with standard plasma expanders animal data suggest that both systemic O2 delivery and extraction are increased, and in far greater proportions to those expected from the actual volume of PFC infused 7. Addition of PFC to resuscitation fluids, together with supplementary oxygen, has been
ACCEPTED MANUSCRIPT 4
demonstrated to result in modest improvements in both regional/tissue oxygen delivery6, 8 and systemic oxygenation6 in animal models of hemorrhagic shock.
RI PT
To be a useful adjunct, ideally PFC should act synergistically with HEX to compensate for the poor oxygen delivery capacity of HEX and improve hemodynamic support. However effects of PFC on hemodynamics are rarely evaluated directly, although improvements in systemic
11
SC
hemodynamics have been reported incidental to the primary outcomes of tissue oxygenation9, 10, .
M AN U
In this study we used a rabbit model of hemorrhagic shock to directly determine effects of combination therapy with PFC on hemodynamics, acidosis, and lactate clearance, a surrogate measure of oxygen debt. We hypothesized that animals treated with a combination of PFC and HEX should show increased benefit (in terms of improved MAP response, reduced acidosis, and
TE D
increased speed of lactate clearance) compared to HEX alone. We defined clinical effectiveness in terms of immediacy of effect (increase in MAP support to > 60 mmHg within 20-30 min of infusion), persistence of effect (hemodynamic support maintained for at least 2 hr post-
EP
resuscitation), and clinically-relevant outcomes (life support for 2 hr, and resolution of clinical markers of tissue hypoperfusion by 2 hr post-infusion). Materials and Methods
AC C
2
2.1 Animals and animal husbandry. This study was approved in advance by the Institutional Care and Use Committee of
Virginia Commonwealth University (VCU protocol number AD20285) and conforms to the Public Health Service Policy on Humane Care and Use of Laboratory Animals (2002). Twentyfour male SPF New Zealand White rabbits (Robinson Services Inc., Mocksville, NC) were housed individually in standard raised-flooring cages. Toys and grass hay were provided for
ACCEPTED MANUSCRIPT 5
enrichment. Animals were maintained at approximately 25oC and 12-h light/12-h dark for a minimum of one week in the animal colony before experimentation. Animals were fed
RI PT
commercial pelleted rabbit chow supplemented with grass hay and water ad lib.
2.2. Surgical methods
SC
Rabbit body weight averaged 3.24 (SD 0.29) kg on the day of experiment. Animals were not fasted prior to surgery. Animals were pre-medicated with ketamine/xylazine (35/3-4 mg/kg)
M AN U
for surgical preparation. Heart rate and pulse oximetry (sPO2, %) were monitored by a portable veterinary oximeter with the sensor placed on the tongue (PulseSense™, NONIN Medical Inc., Plymouth MN). Airway support was provided with a veterinary supra-glottic airway (Rabbit Vgel R3, Docsinnovent, UK). Animals were pre-oxygenated with 100% O2 until respiratory quality was adequate and sPO2 > 96%. Anesthetic plane was subsequently maintained by
TE D
isoflurane (1-3%, balance 100% O2), with positive end-expiratory pressure of 5-8 cmH2O, and flow rate 0.5-0.6 L/min; anesthetic plane was assessed every 3-5 min by vital signs and reflex
EP
response (corneal reflex, toe pinch, limb retraction).
AC C
Arterial and venous catheters were implanted by surgical isolation and cannulation of the left carotid artery, jugular vein, femoral artery, and femoral vein. Catheters were polyethylene PE-90 tubing and flushed with sterile heparinized normal saline (3 U heparin/mL NS) as needed in volumes sufficient to clear the dead space (< 100 uL) to minimize clotting in the lines while avoiding systemic heparinization. The carotid artery and jugular vein were used to monitor MAP and central venous pressure (CVP) respectively. During surgery, core temperature was monitored continuously with a rectal probe and maintained at 36.5-38oC with a thermostaticallycontrolled feedback heating blanket (Harvard Apparatus, Holliston MA). Surgical plane of
ACCEPTED MANUSCRIPT 6
anesthesia was assessed at least every 5 min by vital signs and reflex response (corneal reflex, toe pinch, limb retraction). Two or three blood samples (100 µL) were obtained during surgery to determine arterial pO2 and pCO2 for assessment of oxygenation and ventilation adequacy.
RI PT
Animals were allowed to stabilize for approximately 1 hour (median 61 min, IQR 56-70 min) past the end of surgery; physiological stabilization was defined as MAP > 50 mmHg, lactate < 2
SC
mmol/L, pO2 > 400 mmHg and pCO2 40-55 mmHg.
2.3 Hemorrhage protocol.
M AN U
Animals were hemorrhaged from the femoral artery. We performed a controlled stepwise hemorrhage by withdrawing blood with a programmable syringe pump (PHD22/2000 Series, Harvard Apparatus, Holliston MA), at 3 mL/kg/min until MAP declined to 40 mmHg, then continued at 1 mL/kg/min to maintain hypotension (MAP 35-40 mmHg) for 60 min. Hemorrhage
TE D
was paused if MAP remained between 35-40 mmHg. If MAP <30 mmHg autologous shed blood was returned immediately at 3 mL/kg if MAP < 30 mmHg and hemorrhage resumed when MAP
EP
> 40 mmHg.
2.4 Resuscitation protocol
AC C
Animals were resuscitated with 15 mL/kg test fluids via the femoral vein at a rate of
approximately 3 mL/min; median infusion time was 13 (SD 4) min. Animals received one of three treatment interventions: autologous fresh whole blood (FWB; gold standard control), Hextend® (HEX; TCCC standard control for resuscitation), or HEX combined with 3 mL/kg of 60% (w/v) perflubron (PFC) emulsion (Oxygent™, Sanguine/New Alliance Pharmaceutical Corp., Atlanta and San Diego, CA). HEX consists of 6% high-molecular weight (670 kDa) hydroxyethyl starch in lactated electrolyte solution (lactate 28 mEq/L, dextrose 0.99 g/L; Na+
ACCEPTED MANUSCRIPT 7
143 mEq/L, Cl− 124 mEq/L, Ca2+ 5 mEq/L, K+ 3 mEq/L; [package insert Hospira, Lake Forest, IL; 2004]). Animals were monitored continuously over the post-resuscitation period for 120 min, and then euthanized under deep anesthesia with sodium pentobarbital (Euthasol, 390 mg/mL
RI PT
pentobarbital sodium, 40 mg/kg IV, Virbac AH, Inc., Fort Worth, TX).
2.5. Hemodynamics
SC
Arterial systemic pressure was continuously monitored via the carotid artery with
pressure transducers (Transpac®, Abbott Critical Care Systems, Abbott Laboratories, North
M AN U
Chicago, IL) calibrated before each trial against a mercury sphygmomanometer. Signals were amplified (DA100C and TCD104A interface, BIOPAC Systems, Goleta, CA), and analog-digital conversions of signals to pressure readings were performed on-line at 500 Hz. Mean arterial pressure (MAP, mmHg) was calculated online from the smoothed arterial pressure average. All
TE D
data were collected and converted using AcqKnowledge™ software (Version 4.1, BIOPAC Systems, Goleta, CA). For analysis, data were captured over 20 s at 2-min intervals during
period.
EP
hemorrhage and resuscitation, and over 30 s at 2-5 min intervals during the post-resuscitation
AC C
2.6. Blood chemistry
Blood was sampled in 100 uL capillary tubes every 15 min during the hemorrhage
period, at the end of hemorrhage, and every 15 min for 2 hr post-resuscitation. Blood gases pO2 and pCO2 (mmHg), pH, hemoglobin [Hb] (g/dL), met-hemoglobin (metHb, %), hematocrit (Hct, %), base deficit (BD); metabolites lactate (mmol/L) and glucose (mmol/L), were determined using an ABL 800 analyzer (Radiometer America, Westlake, OH).
ACCEPTED MANUSCRIPT 8
2.7. Statistical methods The experimental design was a randomized block design; resuscitation fluid (FWB, HEX) and presence/absence of PFC were the fixed effects (treatments), and blocking was by
RI PT
animal batch (animals entered the holding colony in batches of six). Blocking minimizes
variation resulting from different holding times and time-related growth or size differences. Experimental run order was generated and randomized in PROC OPTEX (SAS 9.4). Power
SC
calculations were performed by simulation12, with MAP as the primary outcome variable. These calculations suggested that a sample size of 5-10 per group was adequate to detect a 20%
M AN U
specified difference between any two MAP trajectories in treatment, time, and (treatment x time) effects, with power > 0.8. Twenty-four animals were randomized to treatment by ID number using a computerized random-number algorithm (www. randomization.com). One animal died shortly following hemorrhage and was excluded from analysis; a total of 5 rabbits were assigned
TE D
to FWB control treatment, and 9 rabbits to each of the two HEX treatments. Interventions were not blinded to operators, but treatments were recoded and blinded for analysis.
EP
Post-infusion MAP data were modelled using a one-compartment pharmacokinetic model13 in PROC NLMIXED (SAS 9.4):
AC C
MAP(t) = b0 + b1 · t + b2 · δk · t · exp(−b3 · δk · t) + Z(t)
where t is time, b0 is a multiple point “baseline” describing pre-infusion MAP during the latter portion of the controlled hemorrhage period, b1 is the overall slope and represents the overall persistence of fluid effect, b2 controls the magnitude of peak change from baseline, and b3 models the “recovery” rate. In this study, ‘baseline’ refers to the controlled hemorrhage period when MAP is at a minimum and relatively constant, as the metric of interest is the MAP response to resuscitation. Pre-hemorrhage values are of interest only as a second control on
ACCEPTED MANUSCRIPT 9
physiological state pre-shock. Because treatment interventions were categorical, they were assigned dummy variables xi, coded 0 or 1 to discriminate between them14. The resuscitation fluid (treatment factor 1) with two levels (FWB, HEX) was assigned the dummy x1 , and
RI PT
presence/absence of PFC (factor 2) was assigned x2 with two levels (none, PFC added), such that FWB was represented by (x1, x2) = (0, 0), HEX = (1, 0), and HEX+PFC = (1, 1). Therefore the
‘effect size’δk for each treatment combination:
δk = γ0 + γ1·x1 + γ2·x2
SC
contribution of each component in the combination fluid could be evaluated by calculating
M AN U
where γ1 and γ2 are the regression coefficient estimates for x1 and x2 respectively, and measure the deviation of each component from the FWB standard (estimated by γ0). Treatment differences between curve shape estimates were assessed by contrasts15 on MAP peak change Pmax, estimated by (b2/b3) · e−1, and time to achieve peak MAP tmax, calculated from the first
TE D
derivative, where tmax =1/b3. Individual subject effects were modeled as random effects Z(t). The random-effect model is commonly used to estimate subject-specific effects, as it discriminates between observations within and between subjects, and accounts for inter-dependencies among
AC C
EP
responses within subjects15.
Blood chemistry data were analyzed by repeated-measures mixed model ANOVA in
PROC MIXED (SAS 9.4), with individual subject effects modeled as random effects within treatment group, and values of each variable at the end of hemorrhage as the baseline covariate. The best-fit covariance model for correlated errors was an autoregressive process ar(1) within group; this accounted for time-dependencies and heterogeneous variance between groups. Planned comparisons were made at 60 and 120 min, and family-wise error rates were controlled by Bonferroni’s correction, with statistical significance at p = 0.01. Selection bias was minimized
ACCEPTED MANUSCRIPT 10
because animals were properly randomized to both treatment and experimental run order, and baseline imbalances were controlled statistically; therefore no inferential statistical tests were performed on baseline (pre- and post- hemorrhage) characteristics16. Descriptive statistics are
RI PT
reported as means and standard deviations (SD) for normally-distributed continuous data; parameter estimates and effect sizes are reported as means and 95% confidence intervals.
SC
3. Results
Group descriptive statistics for pre- and post-hemorrhage volume and blood chemistry
M AN U
values are shown in Table 1. Average shed blood volume was 51 mL (SD 17 mL), or 16 (SD 5) mL/kg; this represents approximately 35% estimated total blood volume for a 3.3 kg rabbit17. Lactate and base deficit at the end of hemorrhage averaged 3.6 (SD 1.9) and -4.5 (SD 2.9) mmol/L respectively, representing a modest shock insult18,19. Average resuscitation volumes
TE D
were 51 (SD 3) mL, representing between 85-102% of shed blood volume.
Estimates of MAP trajectory were strongly affected by PFC (Table 2, Fig. 1). Addition of
EP
PFC to HEX resulted in a 10-min delay in peak MAP response to resuscitation, and a 4-8 mmHg lower peak MAP. Post-infusion MAP ‘persistence’, estimated from b1, was inhibited by PFC (p
AC C
< 0.0001), averaging an increase of approximately 4 mmHg improvement over 60 min for HEX+PFC, as compared to 13-14 mmHg for FWB and HEX.
Blood chemistry values for each treatment group at 60 and 120 min are given in Table 3.
There were no statistically-significant treatment or (treatment x time) effects for pH (p = 0.14; 0.23), pO2 (p = 0.65; 0.14), pCO2 (p = 0.90; 0.89), [Na+] (p = 0.14; 0.38) and [Ca2+] (p = 0.22; 0.11). HEX + PFC animals were clinically acidotic (pH < 7.35) at both time periods relative to
ACCEPTED MANUSCRIPT 11
control groups. In comparison to controls, HEX+PFC animals showed elevated [Cl− ] and reduced [K+ ] and [HCO3−]. MetHb for HEX+PFC was significantly elevated over the entire post-resuscitation period relative to both FWB and HEX alone (p = 0.013) averaging 1.2%,
RI PT
compared to 0.5% for both FWB and HEX (Fig. 2). Lactate (Fig. 3) increased for the duration of the infusion period, showed modest clearance to 2.9 (SD 1.5) mmol/L for approximately 60 min post-infusion then increased over the next 60 min to 3.6 (SD 1.6), mmol/L, levels exceeding
SC
those observed pre-infusion. However there were no significant differences between treatments at designated time points (p > 0.5). In contrast, base excess became progressively more negative
M AN U
for HEX+PFC animals, exceeding < -6 mmol/L (the so-called ‘critical shock threshold’19 ) at both designated post-resuscitation time points. Glucose of both HEX and HEX+PFC was depressed relative to FWB (p = 0.004; Fig. 4). Following hemodilution, as expected, Hct/Hb in both HEX treatments were significantly lower relative to those laboratory measurements in FWB
4. Discussion
TE D
(p < 0.001).
EP
Combination therapy involves two or more pharmacologic agents administered either separately or in a single fixed-dose combination, with the therapeutic goals of optimizing
AC C
response and simplifying administration. Previous studies have indicated that PFC improves both whole body and regional oxygenation, and is thought to be beneficial for treating decompression sickness20,21 and acute dilutional anemia22. As adjuncts to conventional fluid resuscitation, PFCs have the potential to maintain or restore tissue oxygenation in the critical time window following injury and before bleeding control and definitive care6. However to be a useful adjunct, addition of PFC should, at the least, not detract from the hemodynamic support conferred by volume expander fluids; ideally, PFC should act synergistically with HEX both to compensate for the
ACCEPTED MANUSCRIPT 12
poor oxygen delivery capacity of HEX and improve hemodynamic support. Scattered data suggested that PFC augmentation of resuscitation fluids for treating hemorrhagic hypovolemic
RI PT
shock may result in improved MAP support as well as regional tissue oxygenation 8, 10.
However, in this preclinical trial evaluating hemodynamic support as a primary outcome, addition of 3 mg/kg PFC to HEX was of no discernable benefit in correcting hemodynamics
SC
compared to either FWB or HEX alone. Addition of PFC to HEX was associated with more prolonged times to MAP resolution and reduced post-infusion MAP persistence. As reduction in
M AN U
pH is predominantly driven by changes in strong ions and pCO2, we had also predicted that addition of PFC would correct or blunt shock-induced acidosis through increased CO2 scavenging from the blood, thus driving changes in electrolytes. However, at the doses and administration rates used in this study, treatment differences were clinically modest for arterial
TE D
pCO2, strong ions, lactate, and [HCO3̶ ]; other acid-base disruptions noted were clinicallysignificant metHb elevation, negative base excess below the critical shock threshold, mild to
EP
modest hyperchloremia, and moderate to poor glucose dynamics.
Elevated metHb is indicative of RBC oxidative stress and oxidation of Hb, and reduces
AC C
Hb capacity for oxygen transfer. The elevated metHb levels noted for the PFC group, in spite of supplementary O2, suggests a strong association with PFC administration, rather than anesthesia agents or HEX alone, and may be related to PFC-induced nitric oxide (NO) release23. The depression in Hct and Hb noted for both HEX treatments is not surprising, given the degree of hemodilution involved in replacing almost 35% shed blood volume with non-blood fluid24. However, compared to HEX controls, the clinically and statistically significant negative base excess and reduction in [HCO3-] noted for HEX+PFC animals indicates a metabolic acidosis not
ACCEPTED MANUSCRIPT 13
accounted for by the dilution effects of HEX alone. Unmeasured changes in other ions and weak acids (such as albumin and phosphates) may account for these differences, and indicate the
RI PT
potential for PFC-induced renal acidosis.
Lactate clearance is a surrogate for both the magnitude and duration of tissue hypoxia and oxygen debt 2, 19, 25. We observed that lactate clearance was modest and transient for all
SC
resuscitation treatments in this study, including blood. This suggests that all animals were underresuscitated by the single 15 mL/kg bolus used in this study, even though resuscitation volumes
M AN U
approximated the shed blood volume. This is of practical interest as failure to correct persistent shock-related hypoperfusion within a limited time window may ultimately contribute to multiple organ failure and late death2, and combat casualty care fluid resuscitation guidelines suggest initial resuscitation volumes substantially less than those used in this study1. However the
TE D
discrepancy between the magnitude of post-resuscitation lactate and base excess values for the PFC group suggest that clinical interpretation of these values may not be straightforward. In clinical populations, hyperchloremic acidosis may be associated with abnormal base excess
EP
values without profound lactate acidosis, either from changes in absolute [Cl-], changes in [Cl-] relative to Na+]26, or excessive elimination of [HCO3-] from the kidneys27. In this study, group
AC C
differences in [Na+] and [Cl-] were not particularly large; nevertheless [HCO3-] were greatly affected by the addition of PFC. These data further support the warnings against over-reliance on base deficit as a single surrogate index of tissue oxygenation without consideration of other components of acid-base balance.26
4.1. Limitations and adverse events. There are two limitation of this study. First we used healthy, young, and anesthetized animals
ACCEPTED MANUSCRIPT 14
subjected to controlled hemorrhage. This model does not closely resemble likely clinical situations, namely polytrauma with uncontrolled hemorrhage, or alternatively, high-risk surgical patients. However, because the shock insult is induced to a controlled and repeatable endpoint,
RI PT
this model does provide a standardized, controlled and reliable platform to assess physiological response to the test interventions28, which was the primary objective of this study. Second, this study was primarily designed to evaluate fundamental and clinically-relevant markers of
M AN U
or specific side effects associated with either PFC or HEX.
SC
response to resuscitation, namely MAP and blood chemistry, and not tissue oxygenation per se
The issue of adverse events occurring with both hydroxyethyl starch solutions and therapeutic adjuncts, such as PFC, is of increasing concern. HEX is falling out of favor for fluid replacement therapy, and at present no PFC is approved for human clinical use. Use of hydroxyethyl starch
TE D
solutions for fluid replacement has been associated with increased mortality, renal failure, refractory pruritus, and especially coagulation abnormalities29, 30. In rabbit models, hemodilution with HEX has been associated with reductions in clot propagation and strength, primarily
EP
because of disruptions in F-XIIIa-fibrin polymer crosslinking24. Documented adverse events of PFCs include cardiovascular and central nervous system effects, macrophage and complement
AC C
activation, cytokine release, and platelet reduction31. However synergisms between PFC emulsions and hydroxyethyl starch solutions have been noted that may result in exacerbated coagulation abnormalities. Aggregation formation induced by PFC emulsions in the presence of red blood cells and hydroxyethyl starch solutions32, or between PFC and dextran33, result in clumping of leukocytes and/or erythrocytes, endothelial cell damage, impaired rheology, restriction of capillary flow, and capillary collapse. Hyperchloremia and excessive bicarbonate elimination has been demonstrated to affect eicosanoid release in renal tissue, resulting in
ACCEPTED MANUSCRIPT 15
vasoconstriction, reduction of the glomerular filtration rate, and hypotension27. It is highly possible that synergistic interactions between hydroxyethyl starch present in HEX and PFC emulsions induced coagulation, complement and biolipid inflammatory mediator abnormalities
RI PT
that contributed to the aberrations in MAP and blood chemistry observed in this study.
5. Conclusions
SC
PFC has promise as an oxygen therapeutic adjunct and HEX has had common clinical use as a volume expander. However, adverse events have been documented with use of these products in
M AN U
both preclinical and clinical assessments. In this study, significant disruptions of clinicallyrelevant ‘markers’ of response to resuscitation, namely MAP and blood chemistry, were observed with the use of PFC in conjunction with HEX. Before these products can be reconsidered for therapeutic use in hemorrhagic shock, both dose- and volume-range preclinical
TE D
studies should be performed in the context of combination trials. Evaluation of candidate agents and solutions will require the application of sophisticated experimental designs such as mixture designs (where the response or outcome is a function of the proportions of the therapeutic
EP
components, rather than volume per se)34, and designs incorporating different absolute total volumes and timing of administration relative to point of injury. A wider variety of outcomes
AC C
related to possible adverse safety profiles, such as coagulation profiles, complement, and end organ cell damage must be assessed rigorously in preclinical models.
AUTHORSHIP
P.S.R. and B.D.S are responsible for the study concept. P.S.R. provided oversight for the experiments, and is responsible for study design and data analysis. P.S.R drafted the manuscript, and P.S.R. and B.D.S interpreted the data and performed critical revision of the manuscript for
ACCEPTED MANUSCRIPT 16
important intellectual content. P.S.R. and B.D.S had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
RI PT
DISCLOSURE
Financial support for this research was provided by Entegrion Inc. Sponsors had no role in project design, data collection, analysis or interpretation, or in the writing of the manuscript. The authors
M AN U
ACKNOWLEDGMENTS
SC
declare no other conflicts of interest.
AC C
EP
TE D
We thank AR Morris, M Ellenberg, K Bass, and S Khan for excellent technical assistance.
ACCEPTED MANUSCRIPT 17
REFERENCES [1]
Salomone JP, Pons PT. Tactical field care. In: Salomone JP, Pons PT, eds. PHTLS
RI PT
Prehospital Trauma Life Support: Military Version. St. Louis MO: Mosby-Elsevier; 2007: 516-537.
Barbee RW, Reynolds PS, Ward KR. Assessing shock resuscitation strategies by oxygen debt repayment. Shock 2010; 33 :113-122.
[3]
SC
[2]
Dubick MA, Atkins JL. Small-volume fluid resuscitation for the far-forward combat
[4]
M AN U
environment: current concepts. J Trauma 2003;54: S43-S45.
Riess JG. Perfluorocarbon-based oxygen delivery. Artif Cells Blood Subs Biotechn 2006;34:567-580.
[5]
Spiess BD. Perfluorocarbon emulsions as a promising technology: a review of tissue and
[6]
TE D
vascular gas dynamics. J Appl Physiol 2009; 106: 1444 –1452, 2009. Kemming GI, Meisner FG, Wojtczyk CJ, Packert KB, Minor T, Thiel M, Tillmanns J, Meier J, Bottino D, Keipert PE, Faithfull S, Habler OP. OxygentTM as a top load to colloid
EP
and hyperoxia is more effective in resuscitation from hemorrhagic shock than colloid and hyperoxia alone. Shock 2005; 24:245–254 Padnick LB, Linsenmeier RA, Goldstick TK. Perfluorocarbon emulsion improves
AC C
[7]
oxygenation of the cat primary visual cortex. J Appl Physiol 1999; 86:1497–1504. [8]
Audonnet-Blaise S, Krafft M-P, Smania Y, Mertes P-M, Marie P-Y, Labrude P, Longrois D, Menu P. Resuscitation of severe but brief haemorrhagic shock with PFC in rabbits restores skeletal muscle oxygen delivery and does not alter skeletal muscle metabolism. Resuscitation 2006;70:124-132.
ACCEPTED MANUSCRIPT 18
[9]
Johnson EC, Erickson BK, Podolsky A, Birks EK, Keipert PE, Faithfull NS, Wagner PD. Effects of a perfluorocarbon emulsion for enhanced O2 solubility on hemodynamics and O2 transport in dogs. J Appl Physiol 1995; 79:1777–1786.
RI PT
[10] Paxian M, Keller SA, Huynh TT, Clemens MG. Perflubron emulsion improves hepatic microvascular integrity and mitochondrial redox state after hemorrhagic shock. Shock, 2003; 20: 449–457.
SC
[11] Zhou Z, Sun D, Levasseur JE, Merenda A, Hamm RJ, Zhu J, Spiess BD, Bullock MR. perfluorocarbon emulsions improve cognitive recovery after lateral fluid percussion brain
M AN U
injury in rats. Neurosurgery 2008;63:799–807.
[12] Littell R, Milliken GA, Stroup WW, Wolfinger RD, Schabenberger O. SAS for mixed models. Cary NC: SAS Institute Inc.; 2006.
[13] Sánchez BN, Wu M, Raghunathan TE, Diez-Roux AV. Modeling the salivary cortisol
2012; 176: 918-928
TE D
profile in population research: The multi-ethnic study of atherosclerosis. Amer J Epidemiol
[14] Draper NR, Smith H: Applied regression analysis, 3rd ed, John Wiley & Sons, New York,
EP
pp. 299-317, 1998
[15] Brown H, Prescott R: Applied mixed models in medicine. Chichester UK: John Wiley &
AC C
Sons, Ltd., pp. 455, 2006.
[16] Altman DG: Practical statistics for medical research. London UK: Chapman & Hall. pp 611, 1991.
[17] Mott JC. The relation of blood volume to body weight and arterial haemoglobin levels in rabbits. J Physiol 1967;191:131-140.
ACCEPTED MANUSCRIPT 19
[18] Rixen D, Raum M, Holzgraefe B, Sauerland S, Nagelschmidt M, Neugebauer E, Shock and Trauma Study Group. A pig hemorrhagic shock model: oxygen debt and metabolic acidemia as indicators of severity. Shock 2001;16:239-244
RI PT
[19] Rixen D, Siegel JH: Bench-to-bedside review: oxygen debt and its metabolic correlates as quantifiers of the severity of hemorrhagic and post-traumatic shock. Crit Care 2005; 9:441453.
SC
[20] Smith CR, Parsons JT, Zhu J, Spiess BD. The effect of intravenous perfluorocarbon
emulsions on whole-body oxygenation after severe decompression sickness. Diving and
M AN U
Hyperbaric Medicine 2012;42(1):10-17.
[21] Zhu J, Hullett JB, Somera L, Barbee RW, Ward KR, Berger BE, Spiess BD. Intravenous perfluorocarbon emulsion increases nitrogen washout after venous gas emboli in rabbits Undersea and Hyperbaric Medicine 2007;34(1): 7-20.
TE D
[22] Cabrales P, Briceño JC. Delaying blood transfusion in experimental acute anemia with a perfluorocarbon emulsion. Anesthesiology 2011; 114(4): 901–911 [23] Ash-Bernal R, Wise R, Wright SM. Acquired methemoglobinemia: a retrospective series of
[24]
EP
138 cases at 2 teaching hospitals. Medicine. 2004; 83(5):265-73. Nielsen VG. Effects of
AC C
Hextend hemodilution on plasma coagulation kinetics in the rabbit: role of factor XIIImediated fibrin polymer crosslinking. Journal of Surgical Research 2006;132, 17–22. [25] Nguyen HB, Rivers EP, Knoblich BP, Jacobsen G, Muzzin A, Ressler JA, Tomlanovich MC. Early lactate clearance is associated with improved outcome in severe sepsis and septic shock Crit Care Med 2004; 32:1637–1642. [26] Brill SA, Stewart TR, Brundage SI, Schreiber MA. Base deficit does not predict mortality when secondary to hyperchloremic acidosis. Shock 2002; 17( 6): 459–462
ACCEPTED MANUSCRIPT 20
[27] Bullivant EMA, Wilcox CS, Welch WJ: Intrarenal vasoconstriction during hyperchloremia: role of thromboxane. Am J Physiol 1989, 256:152-157.
animal models. Shock 2005; 24: 33-39.
RI PT
[28] Lomas-Niera JL, Perl M, Chung C-S, Ayala A: Shock and hemorrhage: an overview of
[29] Wiedermann CJ. Hydroxyethyl starch – can the safety problems be ignored? Wien Klin Wochenschr. 2004;116(17-18):583-594.;
SC
[30] Hartog CS, Welte T, Schlattmann P, Reinhart K: Fluid replacement with hydroxyethyl starch in critical care—a reassessment. Deutsch Arztebl Int 2013; 110(26): 443−50.
M AN U
[31] Castro CB, Briceno JC. Perfluorocarbon-based oxygen carriers: review of products and trials. Artificial Organs 2010; 34(8):622–634
[32] Vásques, DM, Ortiz D, Alvarez O, Briceñ JC, Cabrales P. Hemorheological implications of perfluorocarbon based oxygen carriers, interactions with colloid plasma expanders and
TE D
blood. Biotechnol Prog 2013;29: 796-807.
[33] Nolte D, Pickelmann S, Lang M, Keipert P, Messmer K. Compatibility of different colloid plasma expanders with perflubron emulsion; an intravital microscope study in the hamster.
EP
Anesthesiology 2000; 93:1261-70.
[34] Cornell JA. A primer on experiments with mixtures. Hoboken NJ: John Wiley & Sons,
AC C
2011
ACCEPTED MANUSCRIPT 21
FIGURE LEGENDS Fig. 1. Average MAP profiles of rabbits following 60 min of hemorrhagic hypotension and resuscitation with FWB, HEX, or HEX +PFC, estimated from a one-compartment
achieve peak MAP.
RI PT
pharmacokinetics model of MAP over post-infusion time. Dotted lines represent time to
Fig. 2. Average arterial metHb (%) of rabbits following resuscitation with FWB, HEX, or HEX
SC
+ PFC. Error bars represent SE at each time point.
Fig. 3. Arterial lactate and base deficit (mmol/L) of rabbits following resuscitation with FWB,
M AN U
HEX, or HEX +PFC. Error bars represent SE at each time point.
Fig. 4. Arterial glucose (mmol/L) of rabbits following resuscitation with FWB, HEX, or HEX
AC C
EP
TE D
+PFC. Error bars represent SE at each time point.
ACCEPTED MANUSCRIPT 1
Table 1. Pre- and post-hemorrhage characteristics of rabbits in each study group (FWB, HEX, HEX+ PFC) HEX n=9
Mean
SD
Mean
Body mass (kg)
3.1
0.1
3.4
Shed blood volume (mL)
54
12
59
Resuscitation volume (mL)
47
2
50
Pre-hemorrhage pH 7.342
SD
Mean
SD
0.2
3.2
0.3
16
50
9
3
52
5
SC
Blood chemistry
HEX+PFC n=9
RI PT
FWB n=5
0.143
7.363
0.047
7.344
0.048
49
4
53
9
54
21
pO2 (mmHg)
467
55
499
37
498
40
Hb (g/dL)
10.5
1.0
11.6
0.8
11.2
1.0
metHb (%)
0.4
0.4
0.2
0.4
0.0
0.4
Hct (%)
32.4
2.9
35.5
2.4
34.4
3.0
K+ (mmol/L)
2.7
0.5
2.8
0.6
3.0
0.6
142
3
140
3
142
3
Ca (mmol/L)
2.5
0.2
2.6
0.2
2.6
0.2
Cl ̶ (mmol/L)
106
6
104
5
103
5
Glucose (mmol/L)
9.9
3.7
13.0
4.7
11.4
2.1
Lactate (mmol/L)
1.6
0.7
1.3
0.4
1.7
0.7
Base excess (mmol/L)
2.8
4.2
2.4
2.8
2.5
3.7
28.2
3.2
27.2
2.4
28.2
3.9
0.115
7.347
0.061
7.337
0.055
TE D
Na+ (mmol/L)
AC C
EP
2+
M AN U
pCO2 (mmHg)
HCO3- (mmol/L)
End of hemorrhage pH 7.323
pCO2 (mmHg)
50
19
41
9
37
6
pO2 (mmHg)
516
41
516
50
476
31
Hb (g/dL)
9.0
1.2
9.7
1.4
9.4
0.0
metHb (%)
0.5
0.5
0.2
0.4
0.0
0.3
Hct (%)
28.1
3.5
29.3
3.6
29.0
3.7
ACCEPTED MANUSCRIPT 2
K+ (mmol/L)
1.0
3.4
0.9
2.9
0.3
Na (mmol/L)
137
4
137
4
142
2
Ca2+ (mmol/L)
2.6
0.3
2.4
0.3
2.3
0.1
Cl- (mmol/L)
104
8
106
8
112
3
Glucose (mmol/L)
16.6
3.7
16.2
6.6
11.1
4.1
Lactate (mmol/L)
3.9
1.5
4.0
1.7
3.0
2.0
Base excess (mmol/L)
-3.4
2.8
-3.6
3.0
-5.9
2.1
HCO3- (mmol/L)
24.5
4.7
21.2
3.1
18.6
2.0
AC C
EP
TE D
M AN U
SC
RI PT
3.9
+
ACCEPTED MANUSCRIPT 3
Table 2. Estimates for post-infusion MAP ‘persistence’ (b1), peak MAP Pmax, and time to reach Pmax (tmax) calculated from a one-compartment pharmacokinetics model of MAP over post-infusion time in rabbits hemorrhaged to 35% TBV and then resuscitated with FWB (n = 5), HEX (n = 9), or HEX+PFC (n = 9). Difference from
FWB
HEX
95% CI Estimate
Lower,
Upper
|t|
FWB
0.21
0.17
0.24
.
HEX
0.23
0.20
0.26
0.79
HEX+PFC
0.06
0.03
0.09
FWB
20
HEX
22
HEX+PFC
17
tmax (min) 16
HEX
AC C
p
.
.
.
0.38
.
.
<0.0001
69.15
<0.0001
24
.
.
.
.
19
25
1.09
0.29
.
.
14
20
1.24
0.23
2.78
0.011
13
18
.
.
.
.
18
16
20
1.50
0.15
.
.
26
22
30
4.64
0.0001
3.93
0.0007
EP
HEX+PFC
|t|
16
TE D
FWB
47.25
M AN U
Pmax (mmHg)
p
SC
b1 (mmHg/min)
RI PT
Difference from
ACCEPTED MANUSCRIPT 4
Table 3. Descriptive statistics for blood chemistry of hemorrhaged rabbits resuscitated with fresh whole blood (FWB); Hextend alone, or Hextend augmented with 3 mL/kg PFC, at (A) 60 min post-resuscitation; (B) 120 min post-resuscitation. Statistical differences between standard FWB and each HEX group are indicated by p-values.
HEX n=9
FWB n=5
RI PT
A. 60 min post-resuscitation HEX+PFC n=9
HEX vs HEX+PFC
SD
Mean
SD
p
Mean
SD
p
p
7.357
0.039
7.365
0.019
0.85
7.309
0.025
0.32
0.09
pCO2 (mmHg)
41
5
41
3
0.97
38
2
0.67
0.51
pO2 (mmHg)
487
25
490
15
0.91
500
27
0.72
0.75
Hb (g/dL)
9.8
0.6
7.0
0.3
<0.0001
7.5
0.4
0.002
0.36
metHb (%)
0.46
0.25
0.55
0.14
0.77
1.44
0.22
0.010
0.002
Hct (%)
30.3
1.7
21.8
0.8
<0.0001
23.3
1.3
0.002
0.35
K+ (mmol/L)
3.0
0.4
3.0
0.2
0.96
2.6
0.1
0.279
0.07
Na+ (mmol/L)
141
1
141
1
0.93
142
1
0.44
0.45
Ca2+ (mmol/L)
2.3
0.1
2.3
0.1
0.59
2.2
0.1
0.44
0.81
Cl- (mmol/L)
109
Glucose (mmol/L)
13.5
Lactate (mmol/L)
3.2
Base excess (mmol/L)
-2.9
-
M AN U
TE D
21.8
AC C
HCO3 (mmol/L)
3
110
2
0.89
114
2
0.146
0.08
1.6
10.8
1.7
0.25
8.5
0.8
0.009
0.22
0.6
3.0
0.5
0.86
2.7
0.5
0.57
0.64
1.6
-2.5
1.1
0.83
-6.4
1.1
0.075
0.017
1.6
22.2
1.1
0.83
18.8
1.1
0.12
0.035
EP
pH
SC
Mean
ACCEPTED MANUSCRIPT 5 (b) 120 min post-resuscitation
Mean
SD
p
Mean
SD
p
HEX vs HEX+PFC p
7.365
0.022
0.81
7.252
0.031
0.05
0.005
40
3
0.74
0.75
481
36
0.83
0.57
Mean
SD
42
5
39
3
0.57
pO2 (mmHg)
490
25
457
19
0.30
Hb (g/dL)
10.2
0.6
7.3
0.7
<0.0001
7.1
0.5
0.0002
0.84
metHb (%)
0.44
0.25
0.46
0.17
0.94
1.16
0.26
0.06
0.032
31.5
1.7
21.9
1.0
K (mmol/L)
3.2
0.4
3.0
0.3
+
141
1
141
1
2+
2.6
0.1
2.2
-
Cl (mmol/L)
108
3
Glucose (mmol/L)
11.8
1.6
Lactate (mmol/L)
3.9
0.6
Base excess (mmol/L)
-2.0
1.6
HCO3-(mmol/L)
22.8
1.6
AC C
EP
Ca (mmol/L)
22.3
1.6
0.0002
0.81
0.61
2.5
0.2
0.08
0.11
0.85
144
2
0.27
0.21
0.1
0.03
2.2
0.1
0.007
0.88
112
2
0.35
116
2
0.04
0.19
8.3
2.0
0.18
8.5
1.1
0.09
0.94
3.2
0.6
0.46
3.6
0.5
0.69
0.69
-3.4
1.4
0.50
-9.1
1.5
0.002
0.008
21.2
1.4
0.44
16.8
1.4
0.005
0.030
TE D
Hct (%) +
Na (mmol/L)
<0.0001
M AN U
pCO2 (mmHg)
SC
pH 7.354 0.039
HEX+PFC
RI PT
HEX
FWB
ACCEPTED MANUSCRIPT
70
HEX FWB
RI PT
60
SC
55 50
HEX +PFC
45 40 35 30 20
40
60
TE D
0
M AN U
MAP mmHg
65
80
AC C
EP
Time (min)
100 120
ACCEPTED MANUSCRIPT
2.0
RI PT
1.0
SC
0.5
FWB HEX HEX+PFC
M AN U
metHb %
1.5
0.0
TE D
-0.5
AC C
EP
Time
ACCEPTED MANUSCRIPT
5.0
RI PT
4.0 3.5 3.0
2.5
SC
2.0
FWB
1.5
HEX
M AN U
Lactate, mmol/L
4.5
1.0
AC C
EP
TE D
0.5
HEX+PFC
ACCEPTED MANUSCRIPT
A.
FWB HEX
2
HEX+PFC
0
RI PT
-2 -4 -6
-8 -10
SC
Base excess, mmol/L
4
M AN U
-12
Time
25
TE D
30
FWB
HEX HEX+PFC
EP
Bicarbonate [HCO3-] mmol/L
B.
AC C
20
15
Time