| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Resuscitation Research Laboratories, Departments of Anesthesiology and Physiology, University of Texas Medical Branch, Galveston, Texas 77555-0801
 |
ABSTRACT |
|---|
 TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Isoflurane-anesthetized sheep were transfused with packed red blood cells (pRBCs) or diaspirin cross-linked hemoglobin (DCLHb) for treatment of intraoperative hemorrhage. A rapid 15-min hemorrhage with lactated Ringer (LR) infusion maintained filling pressure at baseline and reduced blood hemoglobin (Hb) to ~5 g/dl. Sheep received 2 g/kg Hb, DCLHb (n = 6), or pRBCs (n = 7); control group received LR alone (n = 6). After 2 h, anesthesia was discontinued; sheep were monitored in the animal intensive care unit for 48 h. DCLHb expanded blood volume more, but increased total blood Hb less, than pRBCs. Lower Hb and increased methemoglobin resulted in lower arterial oxygen content compared with the pRBCs. DCLHb caused pulmonary hypertension (from 13 to 30 mmHg) and elevated filling pressure (from 6 to 15 mmHg). Cardiac outputs (CO) were similar for all groups during anesthesia; however, during recovery CO increased only in the LR and packed pRBCs groups. DCLHb may limit the reflex ability to increase CO after volume expansion. Hemodynamic effects of DCLHb may be exaggerated when infused after large-volume LR.
hemorrhage; red blood cell substitutes; hemoglobin; fluid resuscitation; shock; packed red blood cells; diaspirin cross-linked hemoglobin
| |
INTRODUCTION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
SUBSTANTIAL EFFORTS HAVE BEEN MADE toward the development of an effective transfusion substitute for packed red blood cells (pRBCs) (9, 36). Most efforts have focused on free acellular hemoglobin (Hb) solutions with the Hb molecule modified chemically or genetically to optimize stability, intravascular retention, and a near-normal oxygen Hb dissociation (9, 14, 22, 35). There are large differences between the Hb content of pRBCs and free Hb solutions. A hematocrit (Hct) typical of pRBCs is 60-75, with a corresponding Hb concentration of 20-25 g/dl (22). Most Hb-based red blood cell (RBC) substitutes have a Hb concentration of 10-15%, and such Hb solutions are typically hyperoncotic colloids and expand blood volume more than the volume of the infused solution. For example, a recent study showed that 10% diaspirin cross-linked hemoglobin (DCLHb) expanded blood volume by ~130% of infused volume (7). On the other hand, pRBCs are suspended in a crystalloid anticoagulant saline buffer solution and would be expected to expand blood volume slightly less than the infused volume, because the saline solution would distribute throughout the entire extracellular space (7). Furthermore, most Hb solutions produce additional pharmacological effects such as vasoconstriction and gastrointestinal dysmotility, whereas pRBCs are largely devoid of such effects (12, 13, 21).
The very different Hb concentrations, volume expansion effects, and pharmacological properties of pRBCs compared with Hb-based pRBCs substitutes would be expected to have significant consequences on oxygen-carrying capacity and hemodynamics after transfusion. Despite this expectation, there are few data available on how RBC substitutes compare directly to pRBCs in clinically relevant conditions of normovolemic anemia. Most preclinical studies of RBC substitutes have focused on infusion in normal animals (top loading), exchange transfusion, or resuscitation of hemorrhagic hypovolemia (4, 5, 10, 25, 28). Furthermore, control groups are typically conventional crystalloids or colloids or whole blood and not the more clinically relevant pRBCs (4, 15, 19). Clinical trials have shown that RBC substitutes can reduce the need for RBC transfusions in some patients; however, direct comparisons between groups have been difficult because of patient variability and varied transfusion doses (8, 17).
In the present study, we compared equal-dose Hb transfusions of pRBCs vs. DCLHb in anesthetized, surgically stressed sheep subjected to a major hemorrhage and a resulting anemia due to blood loss and lactated Ringer (LR) infusion. Sheep were not allowed to go into hypovolemic shock during the hemorrhage, because LR was aggressively infused soon after the start of the hemorrhage, similar to the actions of a diligent anesthesiologist after an emergency intraoperative hemorrhage. The control group received only continued volume support with LR. Hemodynamic measurements were made for 2 h after transfusion in the anesthetized animals and for 2 days of postoperative recovery during which the sheep were monitored in a large-animal intensive care unit.
We hypothesized that the potent volume expansion properties of DCLHb would limit the increase in total Hb concentration compared with transfusion of pRBCs. A key question was to define the hemodynamic consequences resulting from the different properties of DCLHb, pRBCs, and LR in a model of major surgical stress combined with a large hemorrhage treated initially with crystalloid to stabilize cardiovascular function before randomization to treatment group.
| |
MATERIALS AND METHODS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
The experimental protocol was reviewed and approved by the Animal Care and Use Committee of the University of Texas Medical Branch at Galveston, with adherence to National Institutes of Health Guide for Care and Use of Laboratory Animals [DHHS Publication (NIH) 86-23].
Animals. All RBC transfusions were autologous donations made 8-10 days before the experiment. Nineteen adult merino sheep, weighing 24-42 kg (mean 31.1 ± 1.2 kg), were bled 18-22 ml/kg from a percutaneously placed jugular catheter (Insyte-W16GA2In IV catheter/needle unit, Becton Dickson Vascular Access, Sandy, UT). This volume of blood was collected to provide a source for the 2 g/kg dose of pRBCs stored in two citrate phosphate dextrose adenine (CPDA) blood collection bags containing 63 ml of citric acid (Teruflex blood bag system; CPDA-1 solution, Terumo, Tokyo, Japan). The DCLHb provided by Baxter Hemoglobin Therapeutics was the same product used in their clinical trials with physical properties previously described (23, 29, 30). In brief, DCLHb is a 10 g/dl solution of purified human Hb with alpha chains cross-linked with bis(3,5-dibromosalicyl)fumarate. Methemoglobin (metHb) content is <5%. We measured it by using a IL-482 cooximeter and found it to be 2.5-4%. The solution is made isotonic at 300 mosmol/kgH2O in a sodium lactate mixture.
Experimental surgery.
Sheep were fasted 2 days before the experiment. The surgical placement
of vascular lines and a 2-h abdominal surgery were used as an
intraoperative model of surgical stress. On the day of the experiment,
sheep were sedated with ketamine (10 mg/kg im, Ketaset, Fort Dodge
Animal Health, Fort Dodge, IA) for induction of anesthesia. Sheep were
surgically prepared in a sterile operating environment, orotracheally
intubated (8- or 10-mm-ID cuffed tracheal tube, Mallinckrodt, St.
Louis, MO), and mechanically ventilated (Narkomed 2A, North American
Drager, Telford, PA) under 1.5-2.5% isoflurane (Abbott
Laboratories, Chicago, IL) anesthesia in 50% oxygen. Tidal volume and
respiratory rates were initially set at 14 ml/kg and at 15 breaths/min,
and, thereafter, respiratory rate was adjusted to maintain an arterial
PCO2 between 30 and 35 Torr. This began a 3- to
4-h period of extensive sterile surgery, including placement of
vascular catheters, a major abdominal surgery, and abdominal organ
manipulation. During the first hour, four vascular catheters were
inserted in the right and left femoral arteries and veins to access the
abdominal aorta and the inferior vena cava, respectively. On the left
side, large femoral catheters were used, fabricated from an intravenous
extension set cut down to 24 in. (Baxter Healthcare, Deerfield, IL). On
the right side, smaller femoral Intracath 16-gauge 24-in. intravenous
catheters (Becton Dickinson) were placed. This allowed simultaneous
measurement of arterial pressure, performance of hemorrhage, blood
sampling, and infusions. A pulmonary arterial catheter (7-Fr Oximetrix
Opticath, Abbott Laboratories) for measurement of mixed venous oxygen
saturation (S
Protocol. A key goal of the study, and the focus of this analysis, was to determine the acute and sustained systemic effects of transfusion of DCLHb or pRBCs to correct anemia in sheep subjected to major abdominal surgery and intraoperative hemorrhage. Data from the visceral instrumentation are beyond the focus of the present analysis and are not presented.
After the 3- to 4-h surgical preparation, data collection was started as part of a 3.5-h simulation of a rapid intraoperative hemorrhage and its subsequent treatment. The intraoperative simulation was divided into three periods: a 1-h baseline; a 30-min hemorrhage period with initial treatment of LR only; and a 2-h posthemorrhage period of pRBCs, DCLHb, or a third control group that continued to receive LR only.Baseline period. Hemodynamic parameters were recorded, and blood samples were taken during a 1-h baseline period. The LR infusion was continued as needed to maintain filling pressure and CO at steady levels during surgery and the baseline period.
Hemorrhage period.
After the last baseline measurement, all sheep were submitted to a
rapid 10- to 15-min hemorrhage, with bled volume calculated by the
following equation to reduce Hb to 5 g/dl
|
Transfusion period and treatment groups. Sheep were then randomized to be maintained with LR only (LR group, n = 6) or infused with pRBCs (n = 7) or DCLHb (n = 7). One investigator kept randomization, whereas the anesthesiologists, surgeons, and technicians performing intraoperative care were not informed of the group assignment until shortly before transfusion. This prevented alterations in the preparation or pretreatment care in anticipation of a specific treatment. The Hb dose for the pRBCs and DCLHb groups was the same (2 g/kg of body wt), and both were delivered over the first 30 min of a 2-h intraoperative treatment period.
Postoperative recovery.
During the last 30-45 min of anesthesia, all surgical packing was
removed, and the abdominal incision was carefully closed using 0-Vicryl
(Ethicon, Somerville, NJ) and 0-Prolene (Ethicon). Anesthesia was
discontinued, and the sheep were allowed to recover. After each sheep
was spontaneously breathing, it was transferred to a large-animal
intensive care unit for postoperative care, including pain management
and hemodynamic monitoring for 48 h. During postoperative care,
the fluid-infusion rates were adjusted to maintain filling pressures
and a urine output >1
ml · kg
1 · h1.
Buprenorphine (0.3 mg im) was administered during recovery from anesthesia and then every 6-12 h to minimize postsurgical pain. We
have learned to recognize behavioral signs when sheep are uncomfortable and assess the need for additional pain medication. These are a
lowering of the head, drooping of ears, grinding of teeth, and narrowed eyes.
Measured variables. Data collected included mean arterial pressure (MAP), pulmonary arterial pressure (Ppa), right atrial pressure (RAP), and pulmonary artery occlusion pressure (PAOP), commonly called pulmonary wedge pressure. CO was measured in triplicate by the thermodilution technique, and results were averaged. Blood-gas analysis was performed on both arterial and mixed venous (pulmonary artery) blood samples (System 1302, Instrumentation Laboratory, Lexington, MA). Total Hb, metHb, and percent oxygen saturation were measured with a cooximeter (IL-482, Instrumentation Laboratory) calibrated for human blood. Hct was determined for each blood sample by capillary tube centrifugation.
Calculated variables.
Systemic oxygen delivery (
O2) and
systemic oxygen consumption (
O2) were
calculated from the following formulas
|
|
|
|
|
Statistical methods. All of the outcomes in this study were continuous measures with approximately normal distributions. We used analysis of variance, followed by Tukey's Studentized range test, to assess differences between the three treatment groups at each time point. Because the Studentized range test controls the experiment-wise error rate, declarations of statistical significance were set at the alpha level of 0.05. To assess within-group changes across time, for example during the anesthetized period vs. recovery period, we used paired t-tests.
| |
RESULTS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
All animals required substantial volume support with LR during the surgical preparation to maintain filling pressure and CO. During the surgical preparation and through the end of the baseline period, the sheep required 185.5 ± 19.7, 175.3 ± 18.9, and 207.3 ± 13.1 ml/kg (means ± SE) of LR for the pRBCs, LR, and DCLHb groups, respectively. This large volume of fluid support likely reflects several factors, including a lengthy major abdominal surgery and the sheep being supine, which is an abnormal position for sheep and can impair venous return. Also, all animals experienced intraoperative hypothermia as body temperatures fell from normal awake values of ~39°C to 36-37°C by the baseline period, with no significant differences between groups.
The experimental groups were very similar at baseline with Hb values of
8-9 g/dl. Figure 1A
displays mean ± SE blood Hb. This value is slightly low for
normal sheep levels of ~10 g/dl but reflects the predonation of RBCs
8-10 days earlier and the effects of the surgical preparations and
the volume therapy during this period. The hemorrhage volume was
calculated to further reduce blood Hb to ~5.0 g/dl from the formula
described in MATERIALS AND METHODS, and this goal was well
achieved. This required removal of 31.4 ± 2.4, 32.3 ± 3.4, and 32.0 ± 1.8 ml/kg blood and infusion of 57.5 ± 7.9, 61.4 ± 10.8, and 54.6 ± 7.4 ml/kg of LR during the 30-min
period of hemorrhage in the pRBCs, LR, and DCLHb groups, respectively.
Hemorrhage reduced blood Hb to 5.1 ± 0.3, 5.0 ± 0.4, and
5.5 ± 0.3 g/dl for the pRBCs, LR, and DCLHb groups, respectively.
|
Despite infusion of an identical mean dose of Hb (2 g/kg) in the pRBCs and DCLHb groups, the pRBCs group had a higher mean level of blood Hb (8.3 ± 0.5 g/dl) 120 min (T120) after treatment started compared with the DCLHb group's mean Hb level (7.2 ± 0.4 g/dl). However, this difference was not statistically significant. The LR group had a low Hb level of 5.3 ± 0.3 g/dl, which was virtually unchanged from the hemorrhage level and was significantly lower than the other two groups. During postoperative recovery, all groups exhibited slight decreases in the Hb level; however, at 48 h of postoperative recovery, the Hb levels were still significantly higher in the pRBCs and DCLHb groups, 7.7 ± 0.5 and 6.7 ± 0.2, respectively, compared with the LR group 4.8 ± 0.4 g/dl.
Figure 1B displays the mean values of Hct during the experiment for all three groups. At baseline, groups had similar levels of Hct, ranging from 23 to 26. These values are slightly low for normal healthy sheep (typically 28-32); again, these values reflect the predonation of pRBCs and surgical reparation. Hct levels below normal are not an uncommon finding in surgical patients. After the hemorrhage and LR infusion, the Hct levels fell to a level of 15-17. During treatment, only the pRBCs group showed a rapid and significant increase of Hct to near baseline levels of ~24 during the 120-min posttreatment and only a slightly lower sustained level of 22-23 through the postoperative recovery period. The LR group maintained a stable Hct at the low level of ~15, from the end of hemorrhage through the 48-h postoperative recovery. On the other hand, the DCLHb group showed a significant decrease in the Hct levels to 11.5 ± 0.7 at 30 min after the treatment started. However, Hct returned to the pretreatment hemorrhage levels (15.2 ± 0.5 g/dl) at the first postoperative measurement. This suggests a transient period of plasma volume expansion after the DCLHb infusion that ended early during the postoperative recovery. After 10 h of recovery (R10) and until the end of the experiment, Hct levels recovered slightly to ~16.5 g/dl in the DCLHb group, suggesting a further slight loss of plasma volume.
Figure 2A displays MAP. All
three groups show similar curves from baseline until the end of
hemorrhage with no significant differences detected between groups.
Despite the LR infusions that maintained cardiac filling pressure, at
15 min posthemorrhage MAP decreased ~25 mmHg below baseline. MAP
slightly improved ~15 mmHg below baseline before treatment started at
30-min posthemorrhage. After treatment, the DCLHb group displayed a
rapid and significant increase in MAP from 75 to 115 mmHg at 30 min
after treatment started (T30). The pRBCs and LR groups had slower and
smaller increases during the intraoperative period. During
postoperative recovery, MAP significantly increased in both the LR and
pRBCs groups, such that there were few significant differences between the three groups. During postoperative recovery, MAP was sustained at
100-120 mmHg with the highest level apparent in the DCLHb group; however, this difference was not significant.
|
Figure 2B displays Ppa. The groups were similar at baseline (13-15 mmHg) and only fell slightly at the end of the hemorrhage (12-13 mmHg). During treatment, the DCLHb group showed a rapid, large, and significant increase of nearly 30 mmHg at T30. This Ppa remained elevated through the second hour of recovery (R2) and significantly higher compared with the other groups. The pRBCs group showed a transient and moderate increase of ~8 mmHg at T30, whereas the LR group had no significant alterations, and both the LR and pRBCs groups were at baseline level at the end of treatment. During postoperative recovery, there were only small variations, and at 48 h of recovery (R48) all groups showed similar mean levels that were ~5.0 mmHg higher than baseline levels.
Figure 3A displays RAP. The
experimental groups were very similar at baseline until the end of
hemorrhage. They displayed no significant differences between groups
and no apparent fall after hemorrhage per experimental design because
of the administered LR. However, after treatment the DCLHb group showed
a significantly large transient increase of ~15 mmHg at T30, which
subsided to 13 and 7 mmHg at T60 and T120, respectively. The pRBCs and
LR groups maintained baseline levels of RAP after treatment. During postoperative recovery, all groups maintained near-baseline levels with
no statistically significant differences between groups.
|
The PAOP is displayed in Fig. 3B and exhibits patterns similar to RAP for each group. The peak PAOP in the DCLHb group was measured at T30, with a value of 20 mmHg. From this point on PAOP values declined over the next 2 h. During postoperative recovery, the mean PAOP of each group were not significantly different from each other, although the pRBCs and especially the DCLHb groups exhibited trends to increase over time.
Figure 4A displays CO
expressed as percent of baseline level (l/min). Baseline CO values were
4.5 ± 0.4, 5.2 ± 0.6, and 5.2 ± 0.6 in the pRBCs, LR,
and DCLHb groups, respectively. On a per weight unit basis sheep have
higher COs than humans, which results from their relatively high normal
heart rates and large hearts. All three groups showed little change in
CO and no significant group differences during baseline, hemorrhage, or
during treatment. However, with postoperative recovery CO significantly
increased in both the LR and pRBCs groups but not in the DCLHb group.
The LR group showed the highest CO level (150-160% of baseline
levels) during postoperative recovery time. The DCLHb group had the
lowest CO [~80% of baseline level at 5 h into the
postoperative recovery period (R5)], whereas the pRBCs group had an
intermediate position (130% of baseline). The DCLHb group's CO was
significantly lower than that of the other two groups at R5 and lower
than the LR group at R10. At the end of the experiment, all groups
displayed an increased CO of 155, 145, and 125% of baseline level for
LR, pRBCs, and DCLHb groups, respectively. At this final time point, group differences were not statistically significant.
|
There were no significant differences between treatment groups for either heart rate or stroke volume. Mean baseline heart rates were ~110 beats/min, and mean baseline stroke volumes were ~45 ml. Heart rates and stroke volumes did not change significantly during the intraoperative phase. DCLHb infusion resulted in a lowered heart rate for the first hour posttransfusion, presumably because of transient hypertension. During recovery, heart rate displayed increased variance in all groups with a not significant trend of an increase in the LR group and a decrease in the pRBCs group. Stroke volume increased with postoperative recovery in all three groups with no group differences as CO increased.
Figure 4B displays extracellular base excess as calculated from the blood gases. Baseline base excess shown in Fig. 4 was approximately zero and reflects the effects of surgical stress and anesthesia. Healthy conscious sheep have a base excess level of +3 to +6, which reflects their normal pH of 7.45-7.55. All three groups exhibited a slight decrease in base excess, 1-2 meq below their baseline intraoperative levels, during the hemorrhage. During recovery, all groups recovered to above baseline intraoperative levels of base excess. Differences between groups and difference of groups over time were not significant.
Figure 5 displays the calculated
O2 and O2
of oxygen. The O2 had a decrease during
hemorrhage of ~40% in all groups, directly proportional to the
reduced Hct. During the 2 h intraoperative treatment period,
O2 in the pRBCs (P < 0.05) and DCLHb (P > 0.05) groups steadily improved
after transfusion, but values remained slightly decreased
~10-15% below baseline, whereas the LR group exhibited no
improvement compared with the LR O2
during the hemorrhage. During the postoperative recovery, the
O2 of the pRBCs group significantly
increased and was the highest O2 of any
group and near normal. The O2 of the LR
group also significantly increased, but the DCLHb group did not have a
statistically significant change. Group differences after the
intraoperative treatment were only statistically significant at R5,
with the O2 for the pRBCs group being
significantly greater than the DCLHb group.
|
Figure 5B displays O2, which
shows no apparent change or significant group differences during
baseline hemorrhage and treatment, maintaining the anesthetized
baseline levels of 100 ml/min for all groups. During the awake and
recovery phase, O2 levels significantly increased approximately twofold from the anesthetized baseline levels. This increase was maintained until the experiment ended.
Table 1 displays PaO2,
P
|
Morbidity and mortality. Recovery from anesthesia usually required 1-2 h after anesthesia was stopped before the animals would spontaneously breath and could be extubated. It was typically 4-8 h before animals would stand. Compared with chronic animal models that the senior authors have had experience with, these postoperative recoveries were long and difficult (16, 37). The investigators' subjective evaluation was that the DCLHb animals were less alert and took longer to recover; however, these data were not objectively graded or recorded. During recovery, some sheep had reduced levels of arterial and mixed venous oxygen tension. In these cases blow-by oxygen was administered. Blow-by oxygen was administered for 1-4 h in two of seven pRBCs animals, six of seven DCLHb animals, and zero of six LR animals. Animals were euthanized during recovery if they became hypotensive and hemodynamically unstable despite volume support. All animals in the pRBCs (n = 7) and the LR (n = 6) groups survived 6 postoperative days without any significant morbidity and were then euthanized. One animal (1/6) in the DCLHb group (n = 6) died. This animal was hemodynamically similar to other DCLHb animals during the early postoperative recovery period but died at 20 h during the recovery period. The CO and blood pressure were slightly higher than in the other DCLHb animals, but this animal displayed intermittent decreases in filling pressure despite the fluid therapy, whereas both the base excess and mixed venous oxygen saturation declined. Another DCLHb animal had representative hemodynamics but could not stand because of dysfunctional hindlimbs. In two pilot studies using a larger dose of 4 g/kg of DCLHb, both animals died during the first 24 h of recovery.
| |
DISCUSSION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Animals were subjected to a major surgical stress in that they were anesthetized 7-8 h and had two abdominal incisions with major procedures performed and three peripheral incisions for catheter placement. Additional cardiovascular stress was likely caused by the sheep being in the supine position for most of the intraoperative period and becoming hypothermic despite efforts to maintain body temperature. Such stresses and the resultant volume requirements are not representative of a simple or even a complex elective surgical procedure that proceeds routinely. In the authors' opinion, these changes are more representative of a difficult elective surgical case such as esophagectomy or spine surgery with complications. Perhaps the combined hypothermia and surgical stress we imposed are the most representative of physiological challenges in emergency trauma followed by surgery with significant blood loss.
We hypothesized that the potent volume expansion properties of DCLHb would dilute out the Hb concentration in blood compared with pRBCs. The two specific and clinically relevant questions we hoped to address in this study were 1) How does treatment with an equivalent Hb dose of either pRBCs or DCLHb compare in a model of surgical stress, hemorrhage, and induced anemia with respect to the intraoperative and postoperative hemodynamics? and 2) How are oxygen content, delivery, and consumption affected by these two treatments?
Blood volume expansion. The Hct and Hb changed in the same direction, both in the pRBCs and LR groups. The LR group maintained a steady reduced level (anemia) in both Hct and Hb, whereas the pRBCs group restored both Hb and Hct to near baseline. On the other hand, the DCLHb group caused a rise in Hb and a fall in Hct; these changes are due to a combination of the plasma volume expansion (7, 14) and plasma concentration of Hb after DCLHb infusion.
It is possible to estimate the initial blood volume expansion due to the infusion of 20 ml/kg DCLHb. We can assume that, during the 30-min period of hemorrhage and the LR infusion, normovolemic hemodilution was achieved because filling pressures were maintained. The normal blood volume for sheep is 65 ml/kg. If Hct were decreased from 15.5 to 11.5 g/dl, then the new calculated blood volume due to plasma volume expansion would be 88 ml/kg. This represents a ~23 ml/kg expansion or 115% of the 20 ml/kg infusion. The ability of DCLHb to expand blood volume more than infused volume has been reported to be ~130% of infused volume in a sheep model of simple hypovolemia (7). A lesser expansion in the present study may reflect a greater transvascular loss of fluid, perhaps because of a capillary leak induced by the surgical trauma. Urine output levels were similar in all three groups during the intraoperative period. Enhanced volume expansion properties of RBC substitutes and volume expanders may be considered a desirable feature. Improved volume expansion is the main rationale for the use of colloids over crystalloids. Enhanced volume expansion largely explains the effectiveness of small volume hypertonic formulations (31, 32). However, enhanced volume expansion is also a limitation for a RBC substitute. Plasma volume expansion limits the ability to improve the oxygen content of blood because both the preinfusion Hct is reduced and the concentration of infused Hb itself is diluted. Indeed, Hct was significantly reduced from 15.5 to 11.6 in the DCLHb group. Furthermore, 2 h after an equal-Hb dose of 2 g/kg of DCLHb or pRBCs, the resultant concentration of total Hb increased 3.2 ± 0.3 g/dl after pRBCs and only 1.7 ± 0.2 g/dl after DCLHb. These data support our hypothesis that DCLHb potent volume expansion properties contribute to a lower Hb concentration. This higher mean of the Hb concentration contributed to the statistically significant increase inO2
after treatment and again after recovery in the pRBCs group, whereas
the lower mean Hb in the DCLHb contributed to the lack of significant
increase with DCLHb at these time points. An ideal blood substitute
formulation might have a low to normal colloid osmotic pressure with a
high Hb concentration. For reasons not entirely clear to us, few RBC
substitutes in development have such properties.
Hemodynamics. As stated in MATERIALS AND METHODS, the directions to the anesthesiologist during the hemorrhage were to aggressively correct the volume deficit by infusing LR to restore and maintain filling pressures and CO. This goal was accomplished as the RAP, PAOP, and CO did not fall significantly after hemorrhage. However, despite the infusions, MAP was reduced 15-25 mmHg and Ppa was reduced 2-3 mmHg. This may reflect the reduced viscosity of whole blood due to hemodilution.
The 30-min hemorrhage period and LR volume infusion could be described as acute normovolemic hemodilution, which is often reported to be associated with increased CO instead of the unchanged CO and reduced arterial pressure we observed. However, the effects of anesthesia and surgical stress may blunt the ability of CO to increase during hemodilution (24). More importantly, CO levels significantly increased in both the LR and pRBCs groups during postoperative recovery but not in the DCLHb group. The DCLHb group maintained CO levels during recovery similar to intraoperative levels with one prominent low point at R5. Infusion of several Hb-based RBC substitutes have been shown to cause lower CO levels than in the control groups treated with traditional volume expanders (7, 14, 35). For example, Fischer et al. (7) found that CO levels increased similarly in both normovolemic and hemorrhaged sheep after infusions of either a large volume of crystalloid (60 ml/kg LR) or a smaller volume of colloid (20 ml/kg human albumin). On the other hand, 20 ml/kg DCLHb expanded blood volume more than either a large volume of LR or equal volumes of albumin. Furthermore, DCLHb increased RAP more, but caused only a modest increase in CO in hemorrhaged sheep and reduced CO in normovolemic sheep. In the present study, the DCLHb treatment expanded blood volume better than the pRBCs or the LR on the basis of greater reductions in Hct and higher filling pressures. Despite increased blood volume after DCLHb infusion, CO levels were similar in all three groups immediately after the intraoperative treatment. Also, CO levels increased in the postoperative recovery period, both in the LR and pRBCs groups, but did not in the DCLHb group. Increased metabolism and CO levels are an appropriate and expected response during postoperative warming and recovery from anesthesia. We suggest that DCLHb may have impaired CO because of either an increased afterload from vasoconstriction or a direct cardiac systolic or diastolic dysfunction or a combination of these factors. Such effects may be a limitation of DCLHb and perhaps of other first-generation RBC substitutes. The effects of
--Hb (US
Army), which is the same molecule as DCLHb (Baxter) suspended in a
different buffer, caused coronary vasoconstriction (18).
Decreased CO levels have also been reported after infusion with a
bovine Hb-based RBC substitute (14, 35). On the other
hand, studies have shown an increase in total coronary blood flow after
DCLHb treatment (11), and no deleterious cardiac effects
have been reported in clinical trials (17).
It is interesting to speculate that an impaired ability to increase CO
may be related to Hb binding of nitric oxide (NO). Comparisons between
NO binding, Hb solutions and new second-generation non-NO-binding RBC
substitutes are needed to compare their effects on CO and cardiac
function in clinically relevant animal models. It has been suggested
that a reduced CO after infusion of a RBC substitute is an
"appropriate" response, as there is less need for increased CO and
work because both blood oxygen content and O2 are increased (7).
However, pRBCs also increase oxygen content, whereas other volume
expanders such as LR and albumin generally increase both CO and
O2. Clearly, DCLHb and some of the other
first-generation Hb-based RBC substitutes are unique volume expanders,
as they appear to have a specific effect to inhibit CO despite
significant volume expansion (7, 14, 35).
Oxygen content and delivery.
CaO2 was significantly higher after the pRBCs
transfusion than after the DCLHb transfusion for two reasons. Mean Hb
was lower, albeit not significantly, and SaO2 was
significantly lower after the Hb DCLHb treatment because of an
increased metHb. After intraoperative treatment,
O2 was only significantly increased in
the RBC group. After recovery, O2 was
increased in both the LR and the pRBCs group but not in the DCLHb
group. Despite these differences in the group responses, over time
there was a treatment effect for O2 at
only one time point, R5, when O2 was
significantly higher for the pRBCs group vs. the DCLHb group. Despite
differences in O2, there were no
apparent treatment effects on either O2 or base excess. Thus it could be viewed that the DCLHb treatment was as
physiologically effective as the other two treatments. On the other
hand, we have no data to suggest that the DCLHb treatment offered any
physiological benefit compared with the pRBCs treatment in this animal
model. Likewise, it can be viewed that there was no apparent advantage
in the pRBCs group compared with the LR group, given that all survived
in both groups and O2 and base excess
were not significantly different. It may require a greater hemorrhage
and/or a more severe anemia to demonstrate a significant difference and
need for transfusion. Our study may be one demonstration of the
difficulties involved in showing a physiological advantage to a RBC
substitute. As with different volume expanders and the use of pulmonary
artery catheters, it may be extremely difficult to prove a
statistically significant comparative clinical benefit of a RBC
substitute vs. human blood of pRBCs.
O2 and O2
values are calculated from SaO2 and
SO2 values should be accurate, because at
high saturations there is little difference between the human and sheep
oxygen dissociation curves. The SO2 values. However, because
S--Hb (34). Nevertheless, oxygen saturation of
patients administered DCLHb is typically measured with standard human
cooximeter settings on a clinical cooximeter. Future cooximeters may
require settings to allow for measuring oxygen saturation with
different mixtures of RBC-Hb and blood substitute-Hb.
Clinical implications. Most of the early animal studies suggested some physiological benefit to the DCLHb treatments (4, 5, 7, 25). Furthermore, early clinical studies suggested that the DCLHb treatment was clinically safe, delivered and unloaded oxygen, and could reduce RBC requirements in surgical operations (17, 20, 21). However, the commercial development of DCLHb was abruptly halted in 1998. An interim safety analysis of a Food and Drug Administration phase III, multicenter US emergency room study of DCLHb treatment of severe trauma showed an increased mortality in the DCLHb groups compared with the control patients receiving standard care of crystalloid and pRBCs treatment. An extensive analysis of the trial data has failed to provide a satisfactory explanation of the increased mortality (29). Nor do the present study and our data provide a satisfactory explanation for the one death and significant morbidity (an inability to stand) that was observed in another sheep of the DCLHb group. There is little that can be concluded statistically from our mortality and morbidity data. Such outcomes may or may not have been related to the treatment.
Although there were clear physiological differences between the treatments, our data do not prove a deleterious outcome effect of DCLHb. However, our data suggest at least a potential limitation of DCLHb, which occasionally may affect clinical outcome when combined with major physiological stress conditions. We suggest that the US trauma trial of DCLHb may have exhibited such deleterious effects when patients suffering from severe traumatic injuries and hemorrhage were administered DCLHb infusion. Furthermore, we suggest that this same limitation may have accounted for the mortality in our study as we used an animal model of severe surgical stress and hemorrhage. We speculate on two possible theories for DCLHb treatment limitation: 1) large-volume infusions of LR combined with the DCLHb treatment caused greater cardiovascular stress due to combined hypervolemia and vasoconstriction than with either treatment alone, and 2) DCLHb treatment impairs the ability of the heart to increase CO under conditions of increased preload or increased demands for greaterO2.
The combinations of large-volume LR and DCLHb can cause hypervolemia
and aggravate the increased afterload due to vasoconstriction and the
increased preload due to hypervolemia and perhaps also due to a direct
cardiac effect. Nearly all preclinical animal studies and clinical
trials evaluated infusions of DCLHb alone vs. various control volume
expanders. There have been few studies in which large-volume LR was
infused along with DCLHb as done in the present study and also reported
for most patients in the US trauma trial of DCLHb. Many, if not most,
of the severe trauma patients in the DCLHb trauma trial received
substantial volumes of LR, either in the field or the emergency room
before receiving DCLHb. Interestingly, in a European prehospital trauma
trial in which DCLHb was used alone as the initial treatment and
without LR an increased mortality was not apparent in the DCLHb groups (29).
We speculate that if the myocardium has already been compromised,
because of either preexisting disease or the effects of trauma and
shock, then the added stress of hypervolemia, increased afterload, and
increased cardiac work could lead to an immediate cardiac failure or a
cardiac depression, which subsequently manifests during recovery. We
recently reported that combinations of large volume of LR and DCLHb
aggravate the systemic and particularly the pulmonary hypertension of
DCLHb (1-3). In the present study, systemic pressures
did not increase to levels of clinical concern, but both left and right
filling pressures and the pulmonary pressure were exceedingly high
particularly during the first 30-60 min after administration of
DCLHb. Trauma patients are not routinely monitored for either filling
pressure or pulmonary pressure in the emergency room. Excessive filling
or pulmonary pressures could have occurred in some of the patients in
the DCLHb trauma trials.
Despite volume expansion, DCLHb and other RBC substitutes have a
consistent record of causing only mild increases in CO levels. Fischer
et al. (7) found that CO increased less after DCLHb than
with either equal volumes of albumin or large volumes of LR despite
greater volume expansion with DCLHb. In normovolemia, DCLHb and other
RBC substitutes often cause decreases in CO (14, 35). We
suggest the possibility that DCLHb limits the ability of the heart to
respond appropriately to increases in either preload or metabolic
demands. The overall safety and efficacy record of DCLHb, in both
animal and human studies, indicates that such a limitation was rarely
life threatening. However, an inability to increase CO levels
appropriately has been correlated with increased morbidity in
critically ill trauma patients after surgical stress or accident
(27). In the present study, CO was increased during postoperative recovery as metabolism increased in both the LR and the
pRBCs groups but not in the DCLHb group.
In summary, we have compared infusions of equal Hb doses of the DCLHb
treatment and the pRBCs treatment for hemorrhagic anemia with a LR
control group in an animal model of major surgical stress. There were
no definitive deleterious or beneficial effects established with any of
the treatments. Furthermore, the data suggest significant physiological
stress, hypervolemia, and elevated preload and afterload pressure with
DCLHb treatment. These combined effects may cause an impairment of the
ability to augment CO when large volumes of LR are administered before
DCLHb. Such physiological stresses and cardiac impairments may account
in part for the deleterious mortality outcomes reported in the US
multicenter trauma trial of DCLHb.
| |
ACKNOWLEDGEMENTS |
|---|
We thank Mary Townsend for manuscript preparation and editing, Craig Hartman for criticisms and discussions, and James Grady for statistical support.
| |
FOOTNOTES |
|---|
This research was supported in part by a contract from Baxter Hemoglobin Therapeutics.
Address for reprint requests and other correspondence: G. C. Kramer, Resuscitation Research Laboratories, Dept. of Anesthesiology, UTMB, Galveston, TX 77555-0801 (E-mail: gkramer{at}utmb.edu).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Received 28 November 2000; accepted in final form 7 September 2001.
| |
REFERENCES |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
1.
Brauer, KI,
Prough DS,
Traber DL,
and
Kramer GC.
Lactated Ringer's solution (LR) exacerbates increased minute cardiac work after diaspirin cross-linked hemoglobin (DCLHb) (Abstract).
Shock
11:
67,
1999.
2.
Brauer KI, Prough DS, Traber DL, Traber LD, and Kramer GC. Volume
expansion and hemodynamic interactions between Lactated Ringer's (LR)
and diaspirin cross-linked hemoglobin (DCLHb) in hemorrhaged sheep
(Abstract). Ann Trauma Anesth Crit Care Symp (ATACCS) and World
Expos 12th Chicago, IL 1999.
3.
Brauer, KI,
Prough DS,
Traber DL,
Traber LD,
and
Kramer GC.
Large volume infusion of lactated Ringer's (LR) exacerbates hemoglobin (DCLHb) induced hypertension (Abstract).
Am J Respir Crit Care Med
159:
A521,
1999.
4.
Cohn, SM,
and
Farrell TJ.
Diaspirin cross-linked hemoglobin resuscitation of hemorrhage: comparison of a blood substitute with hypertonic saline and isotonic saline.
J Trauma
39:
210-216,
1995[ISI][Medline].
5.
DeAngeles, DA,
Scott AM,
McGrath AM,
Korent VA,
Rodenkirch LA,
Conhaim RL,
and
Harms BA.
Resuscitation from hemorrhagic shock with diaspirin cross-linked hemoglobin, blood, or hetastarch.
J Trauma
42:
406-414,
1997[ISI][Medline].
6.
Dehring, DJ,
Traber DL,
Fader RC,
Traber L,
and
Doty S.
Dose dependent hemodynamic response to live pseudomonas in sheep.
In: Perspectives in Shock Research, edited by Passmore JC.. New York: Liss, 1988, p. 349-354.
7.
Fischer, SF,
Burnet M,
Traber DL,
Prough DS,
and
Kramer GC.
Plasma volume expansion with solutions of hemoglobin, albumin, and Ringer lactate in sheep.
Am J Physiol Heart Circ Physiol
276:
H2194-H2203,
1999
8.
Gould, SA,
Moore EE,
Moore FA,
Haenel JB,
Burch JM,
Sehgal H,
Sehgal L,
DeWoskin R,
and
Moss GS.
Clinical utility of human polymerized hemoglobin as a blood substitute after acute trauma and urgent surgery.
J Trauma
43:
325-331,
1997[ISI][Medline].
9.
Gould, SA,
and
Moss G.
Current perspectives on blood substitutes.
Curr Surg
44:
279-281,
1987[Medline].
10.
Greenberg, AG.
Clinical implications of blood substitutes.
Art Organs
22:
47-49,
1998.
11.
Gulati, G,
Sharma AC,
and
Burhop KE.
Effect of stroma-free hemoglobin and diaspirin cross-linked hemoglobin on the regional circulation and systemic hemodynamics.
Life Sci
55:
827-837,
1994[ISI][Medline].
12.
Hartman, JC,
Argoudelis G,
Doherty D,
Lemon D,
and
Gorczynski R.
Reduced nitric oxide reactivity of a new recombinant human hemoglobin attenuates gastric dysmotility.
Eur J Pharmacol
363:
175-178,
1998[ISI][Medline].
13.
Hess, JR,
MacDonald VW,
and
Brinkley WW.
Systemic and pulmonary hypertension after resuscitation with cell-free hemoglobin.
J Appl Physiol
74:
1769-1778,
1993
14.
Kasper, SM,
Walter M,
Grune F,
Bischoff A,
Erasmi H,
and
Buzello W.
Effects of a hemoglobin-based oxygen carrier (HBOC-201) on hemodynamics and oxygen transport in patients undergoing preoperative hemodilution for elective abdominal surgery.
Anesth Analg
83:
921-927,
1996[Abstract].
15.
Keipert, PE,
Gomez CL,
Gonzales A,
MacDonald VW,
Hess JR,
and
Winslow RM.
Diaspirin cross-linked hemoglobin: tissue distribution and long-term excretion exchange transfusion.
J Lab Clin Med
123:
701-711,
1994[ISI][Medline].
16.
Kramer, GC,
Sibley L,
Aukland K,
and
Renkin EM.
Wick sampling of interstitial fluid in rat skin: further analysis and modifications of the method.
Microvasc Res
32:
39-49,
1986[ISI][Medline].
17.
Lamy, ML,
Daily EK,
Brichant JF,
Larbuisson RP,
Demeyere RH,
Vandermeersch EA,
Lehot JJ,
Parsloe MR,
Berridge JC,
Sinclair CJ,
Baron JF,
and
Przybelski RJ.
Randomized trial of diaspirin cross-linked hemoglobin solution as an alternative to blood transfusion after cardiac surgery. The DCLHb Cardiac Surgery Trial Collaborative Group.
Anesthesiology
92:
646-656,
2000[ISI][Medline].
18.
Macdonald, VW,
and
Motterlini R.
Vasoconstrictor effects in isolated rabbit heart perfused with bis(3,5-dibromosalicyl)fumarate cross-linked hemoglobin (alpha alpha Hb).
Artif Cells
22:
565-575,
1994.
19.
Malcolm, DS,
Hamilton IN,
and
Schultz SC.
Characterization of the hemodynamic response to intravenous diaspirin crosslinked hemoglobin solution in rats.
Artif Cells Blood Substit Immobil Biotechnol
22:
91-107,
1994[ISI][Medline].
20.
Przybelski, RJ,
Daily EK,
Kisicki JC,
Mattia-Goldberg C,
Bounds MJ,
and
Colburn WA.
Phase I study of the safety and pharmacologic effects of diaspirin cross-linked hemoglobin solution.
Crit Care Med
24:
1993-2000,
1996[ISI][Medline].
21.
Reah, G,
Bodenham AR,
Mallick A,
Daily EK,
and
Przybelski RJ.
Initial evaluation of diaspirin cross-linked hemoglobin (DCLHb) as a vasopressor in critically ill patients.
Crit Care Med
25:
1480-1488,
1997[ISI][Medline].
22.
Remy, B,
Deby-Dupont G,
and
Lamy M.
Red blood cell substitutes: fluorocarbon emulsions and haemoglobin solutions.
Br Med Bulletin
55:
277-298,
1999
23.
Rhea, G,
Bodenham A,
Mallick A,
Przybelski R,
and
Daily E.
Vasopressor effects of diaspirin cross-linked hemoglobin (DCLHb) in critically ill patients (Abstract).
Crit Care Med
24:
A39,
1996.
24.
Schou, H,
Perez de Sa V,
Larsson A,
Roscher R,
Kongstad L,
and
Werner O.
Hemodilution significantly decreases tolerance to isoflurane-induced cardiovascular depression.
Acta Anaesthesiol Scand
41:
218-228,
1997[ISI][Medline].
25.
Schultz, SC,
Powell CC,
Burris DG,
Nguyen H,
Jaffin J,
and
Malcolm DS.
The efficacy of diaspirin crosslinked hemoglobin solution resuscitation in a model of uncontrolled hemorrhage.
J Trauma
37:
408-412,
1994[ISI][Medline].
26.
Sharan, M,
and
Popel AS.
Algorithm for computing oxygen dissociation curve with pH, PCO2, and CO in sheep blood.
J Biomed Eng
11:
48-52,
1989[ISI][Medline].
27.
Shoemaker, W,
Appel PL,
and
Kram HB.
Prospective trial of supernormal values of survivors as therapeutic goals in high-risk surgical patients.
Chest
94:
1176-1186,
1988
28.
Siegel, JH,
Fabian M,
Smith JA,
and
Costantino D.
Use of recombinant hemoglobin solution in reversing lethal hemorrhagic hypovolemic oxygen debt shock.
J Trauma
42:
199-212,
1997[ISI][Medline].
29.
Sloan, EP,
Koenigsberg M,
Gens D,
Cipolle M,
Runge J,
Mallory MN,
and
Rodman G, Jr.
Diaspirin cross-linked hemoglobin (DCLHb) in the treatment of severe traumatic hemorrhagic shock: a randomized controlled efficacy trial.
JAMA
282:
1857-1864,
1999
30.
Sloan, EP,
and
Koenigsberg MD.
The efficacy trial of diaspirin cross-linked hemoglobin (DCLHb) in the treatment of severe traumatic hemorrhagic shock.
Acad Emerg Med
6:
379-380,
1999.
31.
Smith, GJ,
Kramer GC,
Perron PR,
Nakayama S,
Gunther RA,
and
Holcroft JW.
A comparison of several hypertonic solutions for resuscitation of bled sheep.
J Surg Res
39:
517-528,
1985[ISI][Medline].
32.
Tonnesen, AS.
Crystalloids and colloids.
In: Anesthesia (4th ed.), edited by Miller RD.. New York: Churchill Livingstone, 1994, p. 1595-1617.
33.
Torrington, KG,
McNeil JS,
Phillips YY,
and
Ripple GR.
Blood volume determinations in sheep before and after splenectomy.
Lab Anim Sci
39:
598-602,
1989[ISI][Medline].
34.
Vandegriff, KD,
Medina F,
Marini MA,
and
Winslow RM.
Equilibrium oxygen binding to human hemoglobin cross-linked between the alpha chains by bis(3,5-dibromosalicyl) fumarate.
J Biol Chem
264:
17824-17833,
1989
35.
Vlahakes, GJ,
Lee R,
Jacobs EE,
LaRaia PJ,
and
Austen WG.
Hemodynamic effects and oxygen transport properties of a new blood substitute in a model of massive blood replacement.
J Thorac Cardiovasc Surg
100:
379-388,
1990[Abstract].
36.
Winslow, RM.
Blood substitutes
a moving target.
Nat Med
1:
1212-1213,
1995[ISI][Medline].
37.
Wu, CH,
Lindsey CE,
Traber DL,
Herndon DN,
and
Kramer GC.
Measurement of bronchial blood flow with radioactive microspheres in awake sheep.
J Appl Physiol
65:
1131-1139,
1988
This article has been cited by other articles:
|
|
 |
|
  D. Drobin, B. T |
), LR (
), and diaspirin
cross-linked hemoglobin (DCLHb; 
). Total blood Hb was
significantly increased after treatment in both the pRBCs and DCLHb
groups but not in the LR group. Highest levels were apparent in the
pRBCs group; however, the between-group differences between pRBCs and
DCLHb groups were not statistically significant. Hb levels in the pRBCs
group were statistically higher than in the LR group at all
posttreatment time points, whereas the increased Hb in the DCLHb group
was not significant after 5 h into the postoperative (post op)
recovery period (R5). B: mean hematocrit (Hct) was
significantly increased by the pRBCs treatment and significantly
reduced by the DCLHb treatment. The Hct posttreatment was significantly
higher at all time points in the pRBCs group vs. the other 2 groups.
Group differences: *P < 0.05 DCLHb vs. LR;
P < 0.05 pRBCs vs. LR; 
P < 0.05 DCLHb vs. pRBCs.
