ARTIFICIAL CELLS, BLOOD SUBSTITUTES, AND BIOTECHNOLOGY Vol. 32, No. 2, pp.

189207, 2004

Blood Substitute Resuscitation as a Treatment Modality for Moderate Hypovolemia

Anthony T. W. Cheung,1,* Bernd Driessen,3 Jonathan S. Jahr,4 Patricia L. Duong,1 Sahana Ramanujam,1 Peter C. Y. Chen,5 and Robert A. Gunther2
1 Department of Medical Pathology and Department of Surgery, University of California, Davis School of Medicine, Sacramento, California, USA 3 Department of Clinical Studies, University of Pennsylvania School of Veterinary Medicine, Kennett Square, Pennsylvania, USA 4 Department of Anesthesiology, University of California, Los Angeles David Geffen School of Medicine, Los Angeles, California, and Department of Anesthesiology, Charles Drew University of Medicine and Science, Los Angeles, California, USA 5 Department of Bioengineering, University of California, San Diego, La Jolla, California, USA 2

*Correspondence: Anthony T. W. Cheung, Professor of Clinical Pathology, Department of Medical Pathology, UC Davis School of Medicine, Research-III Building (Suite 3400), UC Davis Medical Center, 4645 Second Ave., Sacramento, CA 95817, USA; Fax: (916) 734-2698; E-mail: atcheung@ucdavis.edu. 189
DOI: 10.1081/BIO-120037827 Copyright & 2004 by Marcel Dekker, Inc. 1073-1199 (Print); 1532-4184 (Online) www.dekker.com

ORDER

REPRINTS

190

Cheung et al.

ABSTRACT
Blood substitute resuscitation as a treatment modality for moderate hypovolemia (40% blood loss) in a canine model has been evaluated using Oxyglobin (Biopure Hemoglobin Glutamer-200/ Bovine; a hemoglobin-based oxygen-carrier) and Hespan (6% hetastarch; a nonoxygen-carrier) as resuscitants. Autologous (shed) blood served as control. Nine dogs were studiedafter splenectomy, each dog was hemorrhaged (3236 mL/kg; MAP 50 mmHg) and randomly assigned to the three resuscitation groups. Microvascular, systemic function and oxygenation characteristics were monitored and/or measured simultaneously in prehemorrhagic (baseline), posthemorrhagic and postresuscitation phases for correlation real-time microvascular changes in the bulbar conjunctiva were noninvasively measured via computer-assisted intravital microscopy and systemic function and oxygenation changes were monitored and/ or measured via instrumentation and devices incorporated into our bioengineering station in an operating room setting. Blood chemistry was also studied for relevant measurements. Prehemorrhagic microvascular characteristics were similar in all animals (venular diameter 41 12 mm, A:V ratio 1:2, red-cell velocity 0.5 0.3 mm/s). All animals also showed similar prehemorrhagic systemic function and oxygenation measurements comparable to a previous study and were consistent with normal measurements in dogs. At the completion of hemorrhaging to achieve moderate hypovolemia (40% blood loss with MAP at 50 mmHg), all nine animals showed similar significant (P < 0.01) posthemorrhagic microvascular changes, including 17% decrease in diameter (34 7 mm), A:V ratio variable, and 80% increase in velocity (0.9 0.5 mm/s). All animals also showed similar significant (P < 0.01) posthemorrhagic systemic function and oxygenation changes, with decreases in Hct, aHbtotal, MPAP, MAP, SAP, DAP, CO, SVI, CaO2, and CvO2 and increases in HR and lactic acidosis. Shed blood (control) resuscitation restored posthemorrhagic microvascular changes close to prehemorrhagic values (diameter 39 6 mm, A:V ratio 1:2, velocity 0.6 0.4 mm/s). Oxyglobin and Hespan restored microvascular changes in similar manner close to prehemorrhagic values (Oxyglobin : diameter 38 3 mm, A:V ratio 1:2, velocity 0.6 0.4 mm/s; Hespan : diameter 38 7 mm, A:V ratio 1:2, velocity 0.5 0.4 mm/s). After resuscitation, shed blood (control) restored all systemic function and oxygenation changes close to prehemorrhagic values. However, both Oxyglobin and Hespan resuscitation restored systemic function changes, but not oxygenation changes, to prehemorrhagic values. This was an interesting finding because of the different oxygen-carrying capability

ORDER

REPRINTS

Treatment Modality for Moderate Hypovolemia of Oxyglobin (oxygen-carrying) and Hespan (nonoxygen-carrying). The result suggests that either volume replenishment alone (and not oxygen-carrying capability) is needed to treat moderate hypovolemia or oxygenation measurements obtained by standard methods (oximetry, blood chemistry) may not reflect tissue oxygenation levels.

191

INTRODUCTION In both human and veterinary medicine, resuscitation with allogenic blood has long been the mainstay in the treatment of acute and severe blood loss despite serious concerns associated with its use, including transmittable diseases, immunological incompatibility, transport, short shelf-life, short half-life, storage difficulties and supply shortage (Klein, 1995, 2000; Wallace et al., 1998). It is obvious that there are high hopes and expectations for the development of safer alternatives to blood transfusion or resuscitation, especially in out-of-hospital settings (battlefield or rural/wilderness environment). Recently, allogenic and xenographic, stroma-free, and ultrapurified artificial blood substitutes have been developed in an attempt to address these concerns (Hughes et al., 1996; Rabinirici et al., 1994; Winslow, 1999). Of interest to us are the hemoglobin-based oxygen carriers (HBOCs)the most common blood substitute being tested in the marketand nonoxygen-carrying colloids or crystalloids (NOCCs) and the development of a hypovolemia animal model with which to study their effects and functional efficacy. According to Harvey G. Klein in his 2000 New England Journal of Medicine editorial, the ideal blood substitute should be one which can carry and deliver oxygen, causes few undesirable effects (i.e., nontoxic), requires no compatibility, remains stable under prolonged periods of time during storage, persists in the circulation, easily constituted, and be available at a reasonable cost (Klein, 2000). Despite the high hopes and expectations, it has been surprisingly difficult to demonstrate the clinical effectiveness of HBOCs and NOCCs in an experimental settingas in the prevention or amelioration of tissue hypoxia to prevent circulatory collapse and fatality in hemorrhagic shock in an animal model. To develop a safe and efficacious HBOC is a formidable task that has been hindered by the business model of pharmaceutical and biotechnology companies. As discussed by Marcos Intaglietta in the 1999 U.S. Microcirculatory Society Eugene Landis Award acceptance lecture, . . . the primary focus was to create solutions by modification of the hemoglobin molecule that have identical oxygen carrying properties of

ORDER

REPRINTS

192

Cheung et al.

blood (e.g., concentration, oxygen affinity, Hb-oxygen dissociation curve). In the hands of physiologists and clinicians however, these products from biochemistry based research have shown to be detrimental and have not given satisfactory results because they were based on an inadequate understanding of how the physical properties of blood affect microvascular function, the raison detre of the vasculature (Intaglietta, 1999). Without a clear understanding on the hemodilution effects of HBOCs on the microcirculation and the detrimental/paradoxical effects when the hemodilution caused the hematocrit to drop below the transfusion trigger (12%), utilization of a leading HBOC met with an unfortunate fatal outcome when tested in human subjects during a phase-III clinical trial (Intaglietta, 1999). Restoration of systemic function and oxygenation variables has been a goal of pharmaceutical and biotechnology companies. However, there has been little or no emphasis placed on studying the effects of a HBOC solution in the microcirculation in an in vivo real-time setting. In addition, the clinical benefits and oxygenation effects of a HBOC are difficult to be, and have rarely been, demonstrated in an experimental setting. Clearly, the oxygenation changes in hemorrhagic shock and subsequent HBOC resuscitation treatment should be characterized and correlated with changes in systemic function and the real-time microcirculation. To achieve this goal, we have developed a canine hypovolemia model to serve as a research platform to study hemorrhagic shock and resuscitation treatment. We have also developed an integrated bioengineering station equipped with all needed instrumentations and real-time technologies to monitor and simultaneously measure the microvascular/systemic function/oxygenation variables in an operating room setting. The emphasis of this article is the establishment of this canine hypovolemia model and the bioengineering station for hemorrhagic shock and resuscitation studies, which allows the unique acquisition of microvascular data previously not emphasized in large animal hemorrhagic shock/resuscitation research.

MATERIALS AND METHODS Materials Animals Nine dogs (healthy adult, male or female, 3035 kg) were used. They were prepared, instrumented, splenectomized, hemorrhaged

ORDER

REPRINTS

Treatment Modality for Moderate Hypovolemia

193

(removal of 3236 mL/kg/h of blood to induce moderate hypovolemia/ 40% blood loss), and resuscitated as performed in a previously reported study and described in detail in Methods (Cheung et al., 2001b). The dogs were randomly assigned to three study groups (Oxyglobin , n 3; Hespan , n 3; and autologous (shed) blood as control; n 3) and studied as described in Methods.

Resuscitants Oxyglobin (Bovine Hemoglobin Glutamer-200/HB200; Biopure Corporation, Cambridge, MA) is a HBOC which has been approved by the Food and Drug Administration for canine use. Hespan (6% hetastarch; Abbott Laboratories, Chicago, IL) is a NOCC which is commonly used as a plasma expander in fluid replenishment. Autologous (shed) blood is anti-coagulated blood collected during hemorrhaging for use as resuscitation control.

Methods Animal Preparation and Instrumentation Healthy, adult mongrel dogs (n 9) were used. Approval (Animal Welfare Assurance #A-3433-01) by the UC Davis Animal Care and Use Committee was obtained for this study. The experimental protocol in this study was in compliance with the Guide for the Care of Laboratory Animals (National Institutes of Health publication #86-23, revised 1985). Each dog was premedicated with IM oxymorphone (0.02 mg/kg) and atropine (0.02 mg/kg), followed by percutaneous catheterization of the cephalic vein for continuous infusion of lactated Ringers solution (LRS) at a rate of 10 mL/kg/h throughout the initial preparation and instrumentation period and administration of drugs. Anesthesia was induced with IV propofol (24 mg/kg) and diazepam (0.5 mg/kg), followed by orotracheal intubation. Anesthesia was maintained following a balanced protocol, using isoflurane and fentanyl to minimize potential confounding hemodynamic effects (Ilkiw, 1999; Lemmens, 1995). During animal preparation and instrumentation, isoflurane in oxygen was delivered at an end-tidal concentration of 0.81.2%, and fentanyl infused at a rate of 0.7 mg/kg/min following an initial bolus of fentanyl (10 mg/kg). The dog was mechanically ventilated with an anesthesia ventilator

ORDER

REPRINTS

194

Cheung et al.

(Model 2000; Hallowell EMC, Pittsfield, MA), using tidal volumes (VT) of 1520 mL/kg and a respiratory rate of 912 breaths per minute to ensure an arterial partial pressure of carbon dioxide (PaCO2) in the range of 3545 torr (4.66.0 kPa). End-tidal partial pressure of CO2 (PETCO2), end-tidal concentration of isoflurane (ISOET) and inspired O2 concentration (FiO2) were continuously monitored using a Datex airway gas monitor (Datex 254; Helsinki, Finland). Each dog was instrumented initially in dorsal recumbency and then placed on its side. Further monitoring included continuous recording of the electrocardiogram (ECG; Monitor Model 78353B, Hewlett Packard, Andover, MA) and arterial O2 saturation (SaO2; Model N-180 pulse oximeter, Nellcor Inc., Hayward, CA). Further instrumentation included placement of catheters into the dogs femoral artery for arterial blood withdrawal, and determinations of systemic arterial pressures using membrane transducers (Model 1290A, Hewlett Packard, Watham, MA). An 8-fr balloon-tipped flow-directed thermodilution pulmonary arterial catheter (OptiQTM, Abbott Laboratories, Chicago, IL) was also inserted via the jugular vein and floated into the pulmonary artery under direct monitoring of pressure traces for measurements of central venous pressure (CVP), pulmonary occlusion pressure (POP), mean arterial pressure (MAP), systolic arterial pressure (SAP), diastolic arterial pressure (DAP), mean pulmonary arterial pressure (MPAP), core body temperature and cardiac output (CO). The pulmonary arterial catheter was connected to a cardiac output computer (Critical Care System QVUE, Oximetrix 3, Abbott Laboratories, Chicago, IL) for continuous CO monitoring. Cardiac output was also assessed by thermodilution in triplicate using 10 mL of saline at room temperature. Body temperature was maintained between 3839 C by means of a heating pad and circulating warm air blanket (Bair Hugger Model 505, Augustine Medical Inc., Eden, MN). After preparation and instrumentation, the dog was splenectomized following a midline laparotomy to prevent release of sequestered blood during sympathetic stimulation (Cheung et al., 2001b). After splenectomy, the dog went through a stabilizing period of 45 min, after which the dog was ready for hemorrhaging.

Measurement of Variables Our bioengineering station is capable of measuring numerous variables. Of interest to this study, the measured variables included ECG, MAP, SAP, DAP, MPAP, POP, CVP, heart rate (HR), stroke

ORDER

REPRINTS

Treatment Modality for Moderate Hypovolemia

195

volume index (SVI), and CO. Arterial and mixed-venous heparinized blood samples were collected intermittently (whenever appropriate) from the femoral artery and the right atrium, respectively. Immediately after collection, blood samples were sealed and stored on ice. Subsequently, arterial and mixed-venous total hemoglobin (aHbtotal; vHbtotal), arterial plasma hemoglobin (aHbplasma) and methemoglobin (Met-Hb) concentrations, and arterial and mixed-venous O2 saturation (SaO2; SvO2) were measured in these samples, using a co-oximeter (Model 482, Instrumentation Laboratories, Lexington, MA). Arterial and mixed-venous O2 content (CaO2; CvO2) were directly measured in triplicate using an oxygen-specific electrode (LEXO2CON-K, Hospex Fiberoptics, Chestnut Hill, MA). Arterial and mixed-venous lactate (lactatea; lactatev) concentrations were determined in duplicate by means of a lactate analyzer (Model 1500, YSI Inc., Yellow Springs, OH). Arterial and mixed-venous pH (pHa; pHv) and partial pressures of O2 (PaO2; PvO2) and CO2 (PaCO2; PvCO2) were analyzed with a blood gas analyzer (Rapidlab Model 248, Bayer Corporation Diagnostics Division, Norwood, MA). Blood gas values were corrected for the body temperature of the animal at the time of sampling. Arterial and mixedvenous standard base excesses (SBEa; SBEv) and bicarbonate levels (aHCO3; vHCO3) were computed by the blood gas analyzer. Other variables, including body surface area (BSA in m2), systemic vascular resistance (SVR), etc, were calculated using standard equations.

Computer-Assisted Intravital Microscope (CAIM) A CAIM system was anchored along the front edge of the operating table for noninvasive videotaping of the real-time conjunctival microcirculation in the eye of the dog.a CAIM was originally designed to study real-time microvascular changes in vivo in diabetic and sickle cell anemia patients (Cheung et al., 1999; 2001a; 2001c; 2002a; 2002b; submitted). To incorporate CAIM as part of the instrumentations in the bioengineering station, it has been substantially redesigned and modified to study the microcirculation in the bulbar conjunctiva of the dog so that real-time
a

As described above, measurements were made on numerous systemic function and oxygenation variables in this study. Majority of the measurements were specifically made to monitor the physiological status of the animal during experimentation and will not be reported as experimental data. Only relevant measurements will be presented below as results and tabulated for correlation.

ORDER

REPRINTS

196

Cheung et al.

microvascular changes in the conjunctival microcirculation could be videotaped simultaneously with the measurements of systemic functions and oxygenation in the same animal for critical correlation (see Fig. 1). CAIM was macro-optic based and has an optical magnification of 4.5 and an on-screen magnification of 125. The optical magnification of CAIM was fixed because of its macro design; this nonchangeable magnification is important as it assured that all microvascular measurements made in various phases of the experimentation are quantified on the same basis without a magnification variable. An anti-red (#58 Wratten green filter) fiber-optic light source was used to illuminate the

Figure 1. Bioengineering station setup with instrumentation and intravital microscope. (View this art in color at www.dekker.com.)

ORDER

REPRINTS

Treatment Modality for Moderate Hypovolemia

197

conjunctival microcirculation and to enhance vessel visualization. The angle and level of the front element of CAIM were adjusted to provide the flattest possible surface for focusing. A COHU video-camera (Model 2622-100; -inch monochrome CCD-format, San Diego, CA) was used to noninvasively videotape the real-time conjunctival microcirculation via a high-resolution video-recorder. Focus of CAIM was centered on the microcirculation in the bulbar conjunctiva of the left eye of the dog. The microvascular activities were videotaped and subsequently analyzed via computer-assisted image analysis for microvascular morphometry and dynamics using in-house developed imaging software VASCAN and VASVEL (Cheung et al., 1999; 2001a; 2001b; 2001c; 2002a; 2002b; submitted). The imaging software can objectively quantify over 20 parameters of microvascular characteristics. However, only a few specific morphometric (vessel distribution presented as arteriole:venule/A:V ratio and venular diameter) and dynamic (red-cell velocity) characteristics, which we consider relevant based on a previous investigation, were objectively quantified and reported in this study (Cheung et al., 2001b). Most conjunctival vessels have unique shapes and forms and are easily recognizable. In this study, the vessels of interest were identified and relocated for videotaping. During various phases of the experimentation in this study, the same vessels were longitudinally followed so that relevant measurements (e.g., vessel diameter and red-cell velocity) were made for comparison, with each vessel serving as its own reference control (see Figs. 2 and 3).

Moderate Canine Hypovolemia Model Each dog was allowed to stabilize after the splenectomy. At the end of a 45-min stabilization period, prehemorrhagic (baseline) measurements on systemic function and oxygenation were made (Baseline/ Prehemorrhagic phase). In addition, videotape sequences were made via CAIM on the conjunctival microcirculation simultaneously. Immediately after the baseline measurements were made, approximately 40% (32 36 mL/kg) of the dogs blood volume (85 mL/kg body weight) was withdrawn from the lateral saphenous and femoral veins until a MAP of 50 mmHg was reached. The hemorrhage method of attaining a MAP of 50 mmHg as a clinical criterion was used to ensure the induction of acute but nonlethal hypovolemia. The hemorrhage procedure to achieve moderate hypovolemia (40% blood loss) was normally completed in 3045 min. Shed blood was collected, anticoagulated and used in

ORDER

REPRINTS

198
Pre-hemorrhage Post-hemorrhage

Cheung et al.
Post-resuscitation

2A

2B

Figure 2. (A) Frame-captured images of the conjunctival microcirculation of the dog during 3 experimental phases in Oxyglobin resuscitation. (B) Same framecaptured images as in Figure 2A with identically-spaced white lines superimposed to illustrate the changes in venular diameter.

Pre-hemorrhage

Post-hemorrhage

Post-resuscitation

3A

3B

Figure 3. (A) Frame-captured images of the conjunctival microcirculation of the dog during 3 experimental phases in Hespan resuscitation. (B) Same framecaptured images as in Figure 3A with identically-spaced white lines superimposed to illustrate the changes in venular diameter. (View this art in color at www.dekker.com.)

ORDER

REPRINTS

Treatment Modality for Moderate Hypovolemia

199

autologous blood resuscitation to serve as control. Sequestered blood in the spleen was collected to compute for the total amount of blood loss. In similar manner, posthemorrhagic measurements and videotape sequences were made (Posthemorrhagic phase). At the end of posthemorrhagic measurements, the dog was randomly assigned to one of the three resuscitation groups (autologous/shed blood at 30 mL/kg, Oxyglobin at 10 mL/kg and Hespan at 30 mL/kg). Systemic function and oxygenation measurements were again made at the completion of the resuscitation process (Resuscitation 1 phase). Videotape sequences were also made simultaneously. Another set of measurements and videotape sequences were made 3-h postresuscitation (Resuscitation 2 phase).

Statistics All results were averaged and reported as mean SD. Analysis of variance, students t-test and post hoc Bonferroni corrections were used whenever appropriate. A 0.05 significance level was used in this study. P values <0.01 (e.g., P 0.000045 or P 5.793 104) were presented as P < 0.01 for simplicity.

RESULTS All prehemorrhagic (baseline) microvascular, systemic function and oxygenation measurements were similar in all nine dogs, with no statistical difference between the randomly assigned autologous/shed blood, Oxyglobin and Hespan groups, and were within the normal range reported for dogs in veterinary medicine. At the time when MAP was reduced to 50 mmHg, moderate hypovolemia (40% of blood loss) normally resulted. Compared with prehemorrhagic microvascular characteristics (venular diameter 41 12 mm, A:V ratio 1:2, red-cell velocity 0.5 0.3 mm/s), significant (P < 0.01) posthemorrhagic changes occurred in the conjunctival microcirculation of all nine animals, including 17% decrease in diameter (34 7 mm), A:V ratio extremely variable, and 80% increase in velocity (0.9 0.5 mm/s). At the same time, all nine dogs showed similar significant (P < 0.01) posthemorrhagic changes in systemic function and oxygenation measurements, with decreases in hematocrit (Hct), aHbtotal, aHbplasma, MPAP, MAP, SAP, DAP, CO, SVI, CaO2 and CvO2

ORDER

REPRINTS

200 Table 1.

Cheung et al. Relevant systemic function and oxygenation measurements.

Resuscitation Resuscitation PrePostResuscitation hemorrhage (baseline) hemorrhage 1 2 group 99 7 95 13 78 17 140 16 127 20 141 15 95 18 80 12 84 9 12 2 10 3 13 4 64 34 63 43 21 36 114 19 115 10 82 18 29 7 26 6 43 11 20 1 20 1 20 2 16 2 16 2 15 2 55 2 43 6 55 4 92 22 63 18 87 7 41 7 31 6 35 4 93 95 10 4 51 12 23 34 14 11 215 9 211 13 192 20 62 61 12 3 18 2 15 2 16 2 62 51 71 84 11 93 8 79 1 119 10 131 18 134 9 64 9 70 10 53 3 17 3 11 3 18 7 93 21 10 1 55 55 64 148 24 153 14 98 7 26 1 15 1 61 2 19 2 17 1 16 4 15 2 81 71 104 4 77 24 70 35 152 9 156 75 154 33 82 2 50 40 52 3 13 1 96 19 4 60 1 4 64 41 03 42 100 10 187 58 89 8 36 4 10 8 53 4 20 1 16 2 13 5 16 2 93 71

Variables MAP (mmHg)

Control Oxyglobin Hespan SAP (mmHg) Control Oxyglobin Hespan DAP (mmHg) Control Oxyglobin Hespan MPAP (mmHg) Control Oxyglobin Hespan POP (mmHg) Control Oxyglobin Hespan CVP (mmHg) Control Oxyglobin Hespan HR (min1) Control Oxyglobin Hespan SVI Control (mL/m2/beat) Oxyglobin Hespan CaO2 Control Oxyglobin (volume %) Hespan CvO2 Control Oxyglobin (volume %) Hespan

and increases in HR and lactic acidosis. Pre- and posthemorrhagic microvascular measurements are illustrated in Figs. 2 and 3 while relevant systemic function, oxygenation and blood chemistry measurements are summarized in Tables 1 and 2; all the measurements are similar to results from a previous HBOC/shed blood study (Cheung et al., 2001b).

ORDER

REPRINTS

Treatment Modality for Moderate Hypovolemia Table 2. Relevant blood chemistry measurements.

201

Variables Hct (%)

Resuscitation Resuscitation PrePostResuscitation hemorrhage (Baseline) hemorrhage 1 2 group 46 4 44 3 40 3 16 2 15 1 14 1 00 00 00 0.2 0.2 0.2 0.1 0.6 0.4 7.2 0.5 7.7 0.3 7.2 0.5 1.3 0.4 0.7 0.3 1.4 0.0 39 4 37 2 36 4 13 1 12 1 12 1 00 00 00 0.3 0.2 0.2 0.1 0.6 0.3 6.4 0.4 6.3 0.7 6.5 0.5 3.0 1.1 2.8 1.0 3.0 0.3 43 5 28 2 20 3 14 1 12 0 71 00 2.8 0.1 00 0.2 0.1 0.2 0.2 0.5 0.3 6.7 0.4 6.5 0.5 6.0 0.4 2.3 0.4 2.2 0.4 2.2 0.6 45 4 28 4 21 4 16 2 12 1 72 00 2.5 0.5 00 0.3 0.1 0.4 0.2 0.4 0.2 7.2 0.2 6.5 0.5 5.7 0.4 1.2 0.3 2.0 1.6 1.5 0.2

Control Oxyglobin Hespan aHbTotal (g/dL) Control Oxyglobin Hespan aHbPlasma (g/dL) Control Oxyglobin Hespan Met-Hb (%) Control Oxyglobin Hespan TS (g/dL) Control Oxyglobin Hespan Lactatea (mmol/L) Control Oxyglobin Hespan

At the end of resuscitation treatment (Resuscitation 1 phase), shed blood (control) resuscitation significantly (P < 0.01) restored all posthemorrhagic microvascular changes close to baseline values (diameter 39 6 mm, A:V ratio 1:2, velocity 0.6 0.4 mm/s; see Figs. 4 to 5). In addition, it also significantly (P < 0.01) restored all posthemorrhagic systemic function and oxygenation changes close to baseline values (see Tables 1 and 2). Oxyglobin and Hespan resuscitations, in similar manner, significantly (P < 0.01) restored all posthemorrhagic microvascular changes close to baseline values (Oxyglobin : diameter 38 3 mm, A:V ratio 1:2, velocity 0.6 0.4 mm/s; Hespan : diameter 38 7 mm, A:V ratio 1:2, velocity 0.5 0.4 mm/s; see Figs. 25). Similar to the effects of shed blood resuscitation, Oxyglobin and Hespan resuscitations also restored systemic function changes close to baseline values. However, unlike shed blood resuscitation (oxygen-carrying), Oxyglobin (HBOC; oxygen-carrying) and Hespan (NOCC; nonoxygen-carrying) resuscitations did not significantly restore oxygenation changes to baseline values despite their difference in oxygencarrying capability (see Tables 1 and 2).

ORDER

REPRINTS

202

Cheung et al.

Shed Blood
50 Venular Diameter 45 40
(m)

35 30 25 20 Pre-hemorrhage Post-hemorrhage Post-resuscitation

Oxyglobin
50 Venular Diameter 45 40
(m)

35 30 25 20 Pre-hemorrhage Post-hemorrhage Post-resuscitation

Hespan
50 Venular Diameter 45 40
(m)

35 30 25 20 Pre-hemorrhage Post-hemorrhage Post-resuscitation

Figure 4. Post-hemorrhagic and post-resuscitation changes in venular diameter. (View this art in color at www.dekker.com.)

DISCUSSION The goal of this study was to evaluate the effects of Oxyglobin and Hespan resuscitation as a treatment modality for moderate hemorrhagic

ORDER

REPRINTS

Treatment Modality for Moderate Hypovolemia

203

Shed Blood
1 Red-Cell Velocity 0.8
(mm/sec)

0.6 0.4 0.2 0 Pre-hemorrhage Post-hemorrhage Post-resuscitation

Oxyglobin
1 Red-Cell Velocity 0.8
(mm/sec)

0.6 0.4 0.2 0 Pre-hemorrhage Post-hemorrhage Post-resuscitation

Hespan
1 Red-Cell Velocity 0.8
(mm/sec)

0.6 0.4 0.2 0 Pre-hemorrhage Post-hemorrhage Post-resuscitation

Figure 5. Post-hemorrhagic and post-resuscitation changes in red-cell velocity. (View this art in color at www.dekker.com.)

shock (40% blood loss) using autologous/shed blood (the clinically accepted gold standard) as control. Oxyglobin (a HBOC) and Hespan (a NOCC) were chosen because of their difference in oxygen-carrying capability. An additional goal of this study was to demonstrate the

ORDER

REPRINTS

204

Cheung et al.

functional capability and applicability of our canine hypovolemia model coupled with the newly established bioengineering station to simultaneously measure microvascular, systemic function and oxygenation changes in the same animal during experimentation. From the results, both Oxyglobin and Hespan resuscitations were both effective in restoring microvascular as well as systemic function changes to prehemorrhagic/baseline values in similar manner. These efficacious effects are comparable to shed blood control values. This study confirms the value of the canine moderate hypovolemia model as well as the functional capability of the bioengineering system and its applicability to measure a broad spectrum of measurements simultaneously. In addition, we have also demonstrated that real-time microvascular characteristics (and changes) can be noninvasively, simultaneously and longitudinally studied with systemic function and oxygenation variables for correlation; this is the first time these variables were simultaneously measured in the same animal and correlated in hemorrhagic shock and resuscitation investigations. We have demonstrated that despite a difference in oxygen-carrying capability, both Oxyglobin (a HBOC/oxygen-carrying) and Hespan (a NOCC/nonoxygen-carrying) resuscitations did not restore posthemorrhagic oxygenation changes to prehemorrhagic/baseline values. This is interesting because of the difference in oxygen-carrying capability; Oxyglobin was expected to restore oxygenation changes in a similar manner to shed blood while Hespan was expected to function as a plasma expander for fluid/volume replenishment only with no expectation to restore oxygenation changes. Since the dogs from both Oxyglobin (n 3) and Hespan (n 3) resuscitation groups survived the moderate hypovolemic phase and responded well as evidenced by all posthemorrhagic changes (except oxygenation) being restored close to baseline values, an argument could be made that oxygen-carrying capability may not be needed to treat moderate hypovolemia. These results, though limited in the number of animals used (n 9 with n 3 in each experimental group), suggest that volume replenishment alone (in the case of Hespan ) could serve as a treatment modality for moderate hypovolemia. Whether a NOCC can successfully restore microvascular and systemic changes in severe hypovolemia (50% blood loss) is not known and is currently being investigated in a follow-up study. We were puzzled by the interesting results which similarly showed no difference in effect in the restoration of oxygenation changes despite a difference in oxygen-carrying capability between a HBOC and a NOCC. We decided to look further into this interesting but puzzling

ORDER

REPRINTS

Treatment Modality for Moderate Hypovolemia

205

situation. In a series of related studies using the same moderate hypovolemia canine model, we have demonstrated that Oxyglobin , when administered in a resuscitation volume:blood loss ratio 1:3, returned most systemic, regional (i.e., splanchnic) and microvascular parameters to prehemorrhagic levels, despite persistence of low cardiac output (Driessen et al., 2001a; 2001b; 2003). Though increases in arterial oxygen content and oxygen delivery following Oxyglobin infusion were relatively small, lactic acidosis resolved in the low-volume resuscitated dogs as much as in dogs in a control group receiving large-volume resuscitation with shed blood, indicating that even small volume resuscitation of Oxyglobin improved capillary oxygen transport efficiency and, thus, oxygen uptake into tissues sufficiently enough to allow aerobic cell metabolism to resume even in situations of moderately reduced arterial oxygen content. We suspect that the failure to show restoration of oxygenation changes to baseline level in this study might be due to the inability of the standard OR instrumentations and/or methods (oximetry, blood chemistry) to measure very small changes in oxygenation levels or the inability of the methods to measure tissue oxygenation levels. To clarify this situation, we have purchased a fiber-optic microsensor/thermocouple probe system using the OxyFloTM/OxyLiteTM technology (Oxford Optronix, Oxford, UK) to monitor and measure real-time local tissue blood flow, temperature and tissue oxygenation (tissue PO2) in a follow-up investigation to correlate with results from this study. The incorporation of CAIM, instrumentations and monitoring devices in an OR setting to establish an integrated bioengineering station provided us with the ability to simultaneously measure a broad spectrum of microvascular, systemic function and oxygenation changes. In reality, it served as an ideal data acquisition and monitoring system to study hemorrhagic shock and resuscitation treatment. Using CAIM in this study, we have also confirmed its usefulness in real-time longitudinal studies in microvascular research in vivo. CAIM can be used as an objective and quantitative tool to evaluate treatment modalities in general and clinical trials that are designed to confirm changes in the microcirculation in particular. The noninvasive nature of its applicability is crucial when used in this type of clinical trial. The fact that each conjunctival vessel can be identified and relocated to serve as its own reference control in a longitudinal setting significantly enhances and underscores the uniqueness of CAIM and its applicability in clinical trials which involve real-time changes in the microcirculation.

ORDER

REPRINTS

206

Cheung et al.

ACKNOWLEDGMENTS The research was partially funded by a NIH grant (HL67432) to ATWC, a University of California, Davis Professional Development Award to ATWC, discretionary gifts to the Biomedical Engineering Unit in the Department of Medical Pathology to ATWC, a UC Davis Medical Pathology Robert Stowell (Ph.D/M.D.) Scholarship to PLD, a Hugh Edmondson Summer Research Fellowship to PLD, and a Howard Hughes Medical Institute (S.H.A.R.P.) Research Fellowship to PLD. The assistance of Michelle Barbosa and Betty Hu is very much appreciated. Part of this article has been presented in a poster/discussion session in the 22nd Meeting of the European Society for Microcirculation (August 2002; Devon, UK). This paper has been presented as a workshop session lecture in the 9th International Symposium on Blood Substitutes and 10th Annual Meeting of the Society of Blood Substitutes (March 2003; Shinjuku/Tokyo, Japan).

REFERENCES Cheung, A. T. W., Chan, M. S., Duong, P. L., et al. (2003). The effects of poloxamar 188 (FlocorTM) on sickle cell vasoocclusive crisis: a computer-assisted intravital microscopy study. Blood Submitted. Cheung, A. T. W., Chen, P. C. Y., Larkin, E. C., et al. (2002a). Microvascular abnormalities in sickle cell disease: a computerassisted intravital microscopy study. Blood 99:39994005. Cheung, A. T. W., Harmatz, P., Wun, T., et al. (2001a). Correlation of abnormal intracranial vessel velocity (measured by transcranial Doppler ultrasonography) with abnormal conjunctival vessel velocity (measured by computer-assisted intravital microscopy) in sickle cell disease. Blood 97:34013404. Cheung, A. T. W., Jahr, J. S., Driessen, B., et al. (2001b). The effects of hemoglobin-based oxygen carrier Hemoglobin glutamer-200 (bovine) on the microcirculation in a canine hypovolemia model: a noninvasive computer-assisted intravital microscopy study. Anesth. Analges. 93:832838. Cheung, A. T. W., Perez, R. V., Chen, P. C. Y., et al. (1999). Improvements in diabetic microangiopathy after successful simultaneous pancreas-kidney transplantation: a computer-assisted intravital microscopy study on the conjunctival microcirculation. Transpl. 68:927932.

ORDER

REPRINTS

Treatment Modality for Moderate Hypovolemia

207

Cheung, A. T. W., Price, A. R., Duong, P. L., et al. (2002b). Microvascular abnormalities in pediatric diabetic patients. Microvas. Res. 63:252258. Cheung, A. T. W., Ramanujam, S., Greer, D. A., et al. (2001c). Microvascular complications in the bulbar conjunctiva of type 2 diabetes mellitus (T2DM) patients. Endocrine Prac. 7:358363. Driessen, B., Jahr, J. S., Lurie, F., et al. (2003). Arterial oxygenation and oxygen delivery after hemoglobin-based oxygen carrier infusion in canine hypovolemic shock: a dose-response study. Crit. Care Med. 31:17711779. Driessen, B., Jahr, J. S., Lurie, F., et al. (2001a). Effects of haemoglobinbased oxygen carrier hemoglobin glutamer-200 (bovine) on intestinal perfusion and oxygenation in a canine hypovolemia model. Brit. J. Anaesth. 86:683692. Driessen, B., Jahr, J. S., Lurie, F., et al. (2001b). Inadequacy of lowvolume resuscitation with hemoglobin-based oxygen carrier Hemoglobin blutamer-200 (bovine) in canine hypovolemia. J. Vet. Pharmacol. Therap. 24:6171. Hughes, C. S., Antal, E. J., Locker, P. J., et al. (1996). Physiology and pharmacokinetics of a novel hemoglobin based oxygen carrier in humans. Crit. Care Med. 24:756764. Ilkiw, J. E. (1999). Balanced anesthetic techniques in dogs and cats. Clin. Techniq. in Small Animal Prac. 114:2733. Intaglietta, M. (1999). Microcirculatory basis for the design of artificial blood. Microcir. 6:247258. Klein, H. G. (1995). Oxygen carriers and transfusion medicine. Art. Cell Blood Substit. Immobil. Biotech. 22:123135. Klein, H. G. (2000). The prospects for red-cell substitutes (editorial). N. Engl. J. Med. 342:838843. Lemmens, H. J. M. (1995). Pharmacokinetik-pharmacodynamic relationships for opioids in balanced anesthesia. Clin. Pharmacol. 29:231242. Rabinirici, R., Rudolph, A. S., Vernick, J., et al. (1994). Lyophilized liposome encapsulated hemoglobin: evaluation of hemodynamics, biochemical, and hematologic responses. Crit. Care Med. 22:480485. Wallace, E. L., Churchill, W. H., Surgenor, D. M., et al. (1998). Collection and transfusion of blood and blood components in the United States. Transfus. 38:625636. Winslow, R. M. (1999). New transfusion strategies: red cell substitutes. Ann. Rev. Med. 50:337353.