Oral Presentations

Current Status of Blood Substitute Research: Toward a New Paradigm

R. M. Winslow

Sangart, Inc. and University of California, San Diego

Over 50 years have passed since Amberson warned that progress in the development of a hemoglobin‐based blood substitute would be slow unless the quality and uniformity of products could be improved. He also identified the key issues that have been the central focus of research: renal and hemodynamic impairment. Today, highly purified, characterized solutions are available, but generally not to unaffiliated academic researchers. Although progress has been slow, basic research has moved in surprising directions. Old assumptions about fundamental properties such as oxygen affinity, viscosity, oncotic pressure and oxygen capacity are being questioned and new solutions are being formulated that would have seemed counterintuitive to Amberson. These solutions are not designed to duplicate the properties of the red blood cell, but to optimize their ability to oxygenate tissue, taking into account local vascular regulation of perfusion. As these solutions become available, early assumptions about their clinical applications and clinical trial design must also be reexamined.

Safety and Efficacy of Both Hb‐Vesicles and Albumin‐Heme

E. Tsuchida

Advanced Research Institute, Waseda University, Tokyo 169‐8555, Japan

Over 2 decades of research on routinely available O₂ infusion has specified the design of the best materials and its dynamic function that meet the requisites: no blood type and pathogen/viscosity and osmotic pressure are adjusted comparable with blood/low toxicity and prompt metabolism after infusion of 2–3 lit. per a person/long‐term storage over a year and low cost. After their preclinical study, the clinical study is now in preparation.

The Hb‐vesicle (HbV) encapsulates purified conc. Hb with a phospholipid bilayer membrane (dia. 200 nmϕ), and its solution properties can be adjusted comparable with blood. Surface modification of HbV with (OE)_n ensures stable dispersion state and storage over a year at r.t. In vivo tests have clarified the influences on biodistribution, metabolism in RES, serum clinical chemistry, blood coagulation, etc. HbV does not induce vasoconstriction thus maintains blood flow and tissue oxygenation. Significant efforts have been made to realize the production of HbV with a facility of GMP standard, preclinical and finally clinical trials.

Albumin‐heme, which does not contain globin, is promising. Oncotic pressure is adjustable with oligomeric albumin. Albumin‐heme binds NO as Hb does, however vasoconstriction due to its low pI (4.8) and low permeability across the vascular wall (1/100 of Hb). Enforced property of the lipidheme structure prevents electron transfer and prolongs the functional life of the O₂‐heme complex. Safety issues such as biodistribution and metabolism are now being clarified, and the optimal molecule will be specified soon, it does not induce.

The Need for Blood Substitiutes in Anesthesiology and Intensive Care

R. G. Hahn

Department of Anesthesiology, Karolinska Institute at Söder Hospital, Stockholm, Sweden

The most apparent life‐saving effects of blood substitutes are likely to be found in prehospital care. Much blood may then be needed within a short period of time, but logistic problems makes restoration of the blood volume by crystalloid and colloid fluid the only feasible treatment. Even at hospital, erythrocytes may be available only after a delay, and again the blood volume is first restored while transfusion is given lower priority. With artificial blood, however, the blood volume and its oxygen‐carrying capacity could always be restored at the same time. Replacement of blood loss with “clear fluids” is normally performed in a well controlled fashion during surgery, and in the presence of normovolemia erythrocytes are transfused when the Hb concentration has decreased to a certain level called the “transfusion trigger”. Until recently, Hb 10 g/dL was the accepted transfusion trigger, while we now debate to what extent concentrations down to 5 g/dL are “safe”. The trigger point should be adjusted according to the health status of the patient, the chief determinants being the capacity for increasing cardiac output, maintaining a high Hb saturation, and the degree of arteriosclerosis. In intensive care, the Hb is also maintained at 10 g/dL while slightly lower limits have recently been proposed. In most cases of surgery and intensive care, cost and the risk of infection are the prime incentives for the use of blood substitutes. Additional benefit may be identified operations where the blood loss varies greatly or is difficult to predict. For example, the median blood loss during transurethral resection of the prostate is only 300 ml while transfusion of erythrocytes may still be needed in as much as 10% of the patients.

Normovolemic Hemodilution with Hemoglobin Vesicles Improves Oxygenation in Ischemic Hamster Flap Skin

R. Wettstein,¹ S. Schramm,¹ A. Banic,¹ M. Leunig,¹ H. Sakai,² S. Takeoka,² E. Tsuchida,² and D. Erni^1,*

¹Depts. of Orthopedic, Plastic and Hand Surgery, Inselspital University Hospital, 3010 Berne, Switzerland

²Polymer Chemistry, ARISE, Waseda University, Tokyo 169‐8555, Japan

The aim of this study was to test if oxygenation in ischemic tissue may be improved by normovolemic hemodilution with artificial oxygen‐carrying blood substitutes. To this end, a skin flap model in hamsters (anesthesia: pentobarbital 100 mg/kg ip, euthanasia: 1000 mg/kg iv) was used, which consists of an anatomically perfused and an extended, ischemic part. The microhemodynamics were investigated with intravital microscopy. Partial tissue oxygen tension was measured with a Clark microprobe. Hemodilution was performed by exchanging 50% of the total blood volume with liposome encapsulated human hemoglobin (Hb vesicles, HbV) dissolved in 8% human albumin or 6% dextran 70. The size of the vesicles was about 250 nm, the P50 22 mmHg, and the Hb concentration of the solutions was 7.5 g/dl. Hemodilution with HbV dissolved in dextran lead to an increase in microvascular blood flow in the ischemic flap tissue by 32 to 61% (p < 0.05 vs. baseline and control), whereas blood flow remained virtually unchanged after hemodilution with HbV dissolved in human albumin. ptO2 was improved from 12.6 to 17.9 mmHg (p < 0.05) after hemodilution with HbV dissolved in dextran, but only from 14.3 to 14.8 (ns) if HbV was dissolved in human albumin. Our results suggest that normovolemic hemodilution with artificial oxygen‐carrying blood substitutes may be used therapeutically to improve the oxygenation in local tissue ischemia. The effect depends on the rheological formulation of the oxygen‐carrying solution.

HBOC‐201: Pivotal, Multicenter, Multinational, Phase III Study in Orthopedic Surgery Patients, Efficacy and Safety Results

J. S. Jahr¹ and M. S. Gawryl^2,*

¹University of California, Los Angeles, CA 90095, USA

²Biopure Corporation, Cambridge, MA 02141, USA

HBOC‐201 (Hemopure, hemoglobin glutamer‐ 250, bovine), given as an alternative to red blood cells (RBC), was evaluated in a randomized, controlled, single blind study of patients undergoing orthopedic surgery. Following IRB approval and informed consent, patients were randomized 1:1 to receive either HBOC‐201 or RBC at the time of the first perioperative allogeneic RBC transfusion decision. Subjects, 18 years and older, had not received erythropoietin or undergone PAD and were expected to require at least 2 units of RBC transfusion. Efficacy was determined by the proportion of patients in the HBOC‐201 treatment group who did not receive any transfusions of allogeneic RBC during the study following the initiation of treatment with HBOC‐201 and safety was determined by assessing physical condition, vital signs, and clinical laboratories. Of the 688 patients that received study treatment, 350 (50.9%) were in the HBOC‐201 treatment group and 338 (49.1%) were in the RBC treatment group. At randomization, mean total hemoglobins were 9.1 g/dL and 9.2 g/dL for the HBOC‐201 and RBC treatment groups respectively. In the HBOC‐201 treatment group, the efficacy endpoint was exceeded: 337 (96.3%) patients on Day 1, 246 (70.3%) patients on Day 7 and 208 (59.4%) patients at follow up (6 weeks) avoided transfusion with allogeneic RBC. Blood pressure was increased initially in the HBOC‐201 group but returned to baseline and was not considered clinically important. The data from this study reveals HBOC‐201 to be effective and well tolerated.

Effects of Small Volume Resuscitation on Brain Tissue Oxygen and Cerebral Perfusion: A Randomized Comparison of HSD, LR, and the Hemoglobin‐Based Oxygen Carrier, HBOC‐201

G. T. Manley,^1,2 N. Derugin,¹ D. Morabito,^1,3 and M. M. Knudson^1,3

University of California San Francisco, ¹San Francisco Injury Center, and the Departments of ²Neurological Surgery and ³Surgery

This study compared the efficacy and durability of small volume resuscitation on brain tissue oxygen tension (PbrO₂) and cerebral perfusion pressure (CPP) using a hemoglobin based oxygen carrier (HBOC‐201), hypertonic saline with dextran (HSD), or lactated Ringer's solution (LR) in a model of hemorrhagic shock. The protocol simulated a pre‐hospital clinical situation. Brain oxygen probes were placed in 30 swine to measure PbrO₂ directly. Swine were hemorrhaged and then randomized to HBOC‐201 (6 cc/kg), LR (12 cc/kg), or HSD (4 cc/kg). PbrO2 declined to 11 ± 5 mm Hg and CPP decreased to 32 ± 3 mm Hg following hemorrhage. Small volume resuscitation with HBOC resulted in a significantly (p = 0.009) higher CPP (63 ± 18 mm Hg) over the 2 hour observation period versus HSD (52 ± 18 mm Hg) and LR (45 ± 6 mm Hg). The amount of time that PbrO₂ remained below 15 mm Hg following fluid resuscitation was only 0.5 ± 1.4 min. for HBOC‐201 as compared to 12 ± 35 min. for HSD and 18 ± 43 min. for LR. Although all 3 solutions resulted in improvements in CPP and PbrO₂, HBOC‐201 appears to be more efficacious and has greater durability than HSD or LR. HBOC‐201 may be a promising cerebral resuscitation fluid, particularly in the pre‐hospital arena where a small‐volume bolus may provide oxygen and perfusion support during transport and initial hospital management.

The First Clinical Studies of the Perfluorochemical Emulsion Perftoran (PF) in Russia

V. V.Moroz,¹ N. L. Krylov,¹ N. A. Onishchenko,² A. N. Kaidash,³ E. I. Maevsky,^4,* and G. R. Ivanitsky⁴

¹Research Institute of General Reanimatology & Main Military Clinical Hospital, Moscow

²Scientific Transplantology Center, Moscow

³Vishnevsky Surgery Institute, Moscow

⁴Institute of Theoretical and Experimental Biophysics RAS, 142290, Pushchino, Moscow Reg., Russia

After the experimental and first clinical trials of PF, manufactured by OJSC SPC “Perftoran”, had been completed the second and third phases of clinical studies were carried out on 947 patients in Russia by 1995. The main participants of clinical studies were Main Military Clinical Hospital, Cardio‐vascular Research Institute, Military Medical Academy, Vishnevsky Surgery Institute, Center of Transplantology, Dnepropetrovsk Medical Academy and Surgery Scientific Center. PF was administered in dosages from 4 to 30 ml/Kg BW depending on the disease. Doses of PF as a blood substitute were from 1000 to 5300 ml, in particular, for the treatment of severe anemia (traumatic and surgical bleeding), hemorrhagic, traumatic and toxic‐infection shock, craniocerebral trauma, polytrauma. As PF does not contain polymers it must be used with some plasma expanders for treating massive blood losses. PF infusions and breathing in the air enriched with O₂ (FiO₂ ∼ 0.5), as a rule, improved central and peripheral hemodynamics, reduced edema and thromboembolic damages, increased O₂ delivery to tissues, shortened the reanimation period. The patients` conditions with hypoxic encephalopathy and obliteration injury of vessels improved after 2 or 3 infusions of PF, 200 or 400 ml per day. Cardioplegia with use of PF ensured a spontaneous, quick and fuller restoration of electric and contractive activity in the heart at reperfusion. PF infusions into kidney donors reduced the rejection rate and increased the amount of long‐living transplanted kidneys from 48% to 81%.

Systemic and Microvascular Responses to Hemorrhagic Shock and Resuscitation with Hb‐Vesicles (HbV) as Oxygen Carriers

H. Sakai,¹ R. Wettstein,² A. G. Tsai,² S. Takeoka,¹ M. Intaglietta,² and E. Tsuchida¹

¹Advanced Research Institute for Sci. & Eng., Waseda University, Tokyo 169‐8555, Japan

²Dept. Bioengineering, University of California, San Diego, La Jolla, CA 92093‐0412, USA

Hb‐vesicle (HbV) can provide O₂ carrying capacity to plasma expanders. Its ability to restore systemic and microcirculatory condition after hemorrhagic shock was evaluated in conscious hamsters dorsal skinfold window preparation. The HbV was suspended in 8% HSA, and the [Hb]was either 3.8 g/dL (HbV(3.8)/HSA) and 7.6 g/dL (HbV(7.6)/HSA). Shock was induded by 50% blood withdrawal, and MAP at 40 mmHg was maintained for 1 hr by the additional bleeding. The hamsters receiving either of the HbV(3.8)/HSA and HbV(7.6)/HSA suspensions restored MAP to 93 ± 14 and 93 ± 10 mmHg, respectively, similar with those receiving the shed blood (98 ± 13 mmHg), and these are significantly higher than HSA alone (62 ± 12 mmHg). Only the HSA group maintained hyperventilation and negative BE. During the shock the microvascular blood flow reduced to about 10–20% of the baseline, and re‐infusion of the shed blood restored it to about 60–80%, not to the original level. This is mainly due to the sustained constriction of the small artery A₀ (143 ± 29 µm ϕ). The HbV/HSA groups show better microvascular blood flow and tissue oxygenation, especially for HbV(3.8)/HSA group, than the HSA group. These results indicate that the addition of the O₂ carrying capacity to HSA improves restoration of blood pressure and metabolic acidocis, and tissue hypoxia. However, in the skin microcirculation, dilation of constricted A₀ should be required for the complete recovery of blood flow and tissue oxygenation.

The Response of Human Brain Capillary Endothelial Cells to a Novel Hemoglobin‐Adenosine‐Glutathione Based Red Cell Substitute

J. Simoni,¹ G. Simoni,¹ D. E. Wesson,² J. A. Griswold,¹ and M. Feola¹

Departments of ¹Surgery and ²Internal Medicine, Texas Tech University Health Sciences Center, Lubbock, Texas 79430, USA

The free hemoglobin (Hb)‐based blood substitutes under current testing have met with limited success. The problems revolve around blood vessel constriction and the pro‐oxidant and pro‐inflammatory properties of Hb. Toward a resolution of these problems, we have developed a novel Hb modification procedure. This red cell substitute is composed of Hb crosslinked intramolecularly with o‐ATP and intermolecularly with o‐adenosine, and conjugated with reduced glutathione (GSH). In a recent study, we investigated the direct neurotoxic potential of our red cell substitute using cultures of human brain neurons and astrocytes (Art. Cells Blood Subs. Immob. Biotech. 29(2):128,2001). While this research demonstrated a lack of neurotoxicity, the obtained data failed to answer questions about the possible indirect neurotoxic effects of this product via activation of the brain capillary endothelial cells (EC). Therefore, this study compared the effects of our novel red cell substitute and those of unmodified (U) Hb, using plasma as a control, on normal and GSH depleted EC. Confluent normal and GSH depleted cells were incubated overnight with 0.4 mM Hb solutions or plasma. After treatment, cells were tested for permeability by determining the diffusion rate of ¹²⁵I‐albumin across the monolayer, evaluated for early and advanced apoptosis, and for expression of the adhesion molecules. The pro‐oxidant effect of Hb was examined by measurement of intracellular GSH, conjugated dienes and 8‐isoprostane. Results indicate that the UHb increases EC permeability, initiates cell shrinkage, produces apoptotic events and inflammatory responses. These effects are aggravated in GSH depleted cells. Contrarily, our red cell substitute did not appear to induce apoptosis, shrinkage and oxidative stress nor to increase inflammatory reactions of normal and GSH depleted EC. The diffusion rate of ¹²⁵I‐albumin was similar to that of the control in both tested groups. The effect of this red cell substitute can be linked with the type of modification procedure that lowers Hb pro‐oxidant potential and the anti‐inflammatory and cytoprotective properties of adenosine.

Effect of Oxycyte Treatment on Permanent Focal Cerebral Ischemia in Rats

J. Woitzik, T. Karutz, and L. Schilling

Department of Neurosurgery, University Hospital Mannheim, University of Heidelberg, 68167 Mannheim, Germany

Oxycyte is a new perfluorocarbon‐based emulsion to function as an artificial oxygen carrier. The present study was conducted to evaluate the neuroprotective effect of Oxycyte treatment in focal cerebral ischemia in male rats (Sprague‐Dawley, 270–350 g body weight). A permanent middle cerebral artery (MCA) occlusion was induced by positioning of an intravascular suture. Animals received an intravenous infusion of 10 ml/kg body weight Oxycyte or isotonic saline at the time of MCA occlusion and were allowed to breath pure oxygen (normobaric hyperoxygenation, nbHO) or normal air. Animals were sacrificed 8 hours after MCA occlusion and the necrotic volume was calculated from silver nitrate‐stained serial slices. In addition, immunohistochemical staining for peroxynitrite formation was performed. Ischemic volume was 353 ± 21 mm³ in saline‐nbHO animals and 370 ± 31 mm³ in Oxycyte/air‐treated animals as compared to 335 ± 36 mm³ in control (saline–air) animals. However, animals treated with Oxycyte breathing pure oxygen had a significantly reduced infarct size (309 ± 44 mm³) as compared to saline‐treated rats breathing oxygen and animals receiving Oxycyte emulsion breathing air. The intensity of peroxynitrite staining was most pronounced in saline treated animals breathing air whereas it appeared markedly suppressed with Oxycyte treatment. Therefore, treatment with Oxycyte commencing early after onset of brain ischemia may be of therapeutic efficacy.

Benefit and Risk Perceptions in Transfusion Medicine: Blood and Blood Substitutes

K. C. Lowe,¹ E. Ferguson,² K. Farrell,² and V. James³

¹Life & Environmentalciences and

²Psychology, University of Nottingham, Nottingham NG7 2RD,

³National Blood Service, Sheffield S5 7JN, UK

Blood transfusion is a remarkably safe, routine procedure in clinical medicine. However, little attention has focused on the perceptions of risk associated with the receipt of blood, blood products or ‘blood substitutes’. It is pertinent to ask 1) what key stakeholder groups know about transfusion, 2) how safe they perceive blood/blood products to be, 3) how the latter information might influence their own and others' perceptions of risk linked to transfusion, and 4) the extent to which approved blood substitutes might be preferred over autologous or donor blood. An appreciation of what stakeholders perceive to be the benefits and risks of the receipt of blood and blood substitutes will inform future transfusion strategies. To obtain such information, a programme of research has been initiated involving a combination of qualitative macro‐scale and micro‐scale approaches to data collection. Surveys have targeted key stakeholder groups, namely UK adult blood donors and non‐donors, anaesthetists, general practitioners and healthcare journalists of both genders. Message framing and cueing studies have also been performed with undergraduate students. Data from the latter studies will underpin the communication of relevant and accurate information to stakeholders. This will improve misunderstandings about current issues associated with blood donation and transfusion against the backdrop of changing public trust of healthcare professionals and attitudes and expectations on blood safety and benefits of blood substitutes.

Hyperbaric Correction of the Blood Cellular Composition in Critical States

Sh. Z. Kasimov, Sh.Kh. Narzikulova, Ye. N. Dudura

Research Center of Surgery, Tashkent, Republic of Uzbekistan

Destruction of the erythrocytes cell membrane happens in the traumatic shock, that reduces the activity of the oxygen transport function of the blood and accelerates the tissue hypoxia. For the correction of this disturbances the hyperbaric oxygenation (HBO) has been used.

Methods: The blood was investigated in 7 patients. HBO was performed in BLKS‐3 altitude chamber in 1.8–2.0 PA mode, 45–60 min. exposure, from 7 to 10 runs. The blood was investigated by scanning microscope.

Results: It was established that the number of erythrocytes with disturbed cell membrane‐echinocytes, stomatocytes and spherocytes increased in the traumatic shock. The erythrocytes were presented by diskocytes in control group (donors). In calculation of the erythrocytes formula it was revealed the diskocytes reduction up to 29.8 ± 1.7% before HBO. The diskocytes content increased up to 66.7 ± 2.4% and the transformed forms of the erythrocytes diminished considerably after HBO.

Conclusion: The HBO use allowed to qualitative improve the erythrocytes formula and hence, the oxygen transport function and reological properties of the blood in complex treatment of the traumatic shock.

Continuous, Non‐invasive, Multi‐wavelength Pulse Oximetry Measurement of Methemoglobin (MHb) in the Arterial Blood

M. T. Huiku,¹ S. L. Mioc,² S. L. Baskin,³ G. A. Rockwood,⁴ A. V. Moran,⁴ K. R. Armstrong,⁵ B. J. Lukey,⁴ D. W. Kahler,⁴ D. L. Burman,⁴ and C. M. Arroyo⁴.

¹Datex‐Ohmeda, Helsinki, Finland

²Datex‐Ohmeda, Louisville, CO 80027, USA

³Pharmacology, ⁴Drug Assessment and ⁵Comparative Medicine Divisions, US Army Medical Research Institute of Chemical Defense, Aberdeen Proving Grounds, MD 21010, USA

MHb is a biproduct of first generation blood substitutes, and is currently assessed by invasive techniques requiring the use of co‐oximeters. In this study, continuous, non‐invasive MHb measurement is presented. Five rhesus monkeys (Macaca mulatta) were injected intravenously with sodium nitrite, a MHb former, and the evolution of MHb was continuously measured by a prototype non‐invasive multi‐wavelength pulse oximeter. Excellent agreement with co‐oximeter values was obtained. Electron Paramagnetic Resonance measurements of the blood revealed concomitant increases in nitrosylhemoglobin. Measuring the MHb non‐invasively could improve the safe administration of blood substitutes.

Selectron Transfer Between Hemoglobin Subunits

M. C. Marden and L. Kiger

U. Inserm 473, 84, rue du Général Leclerc, 94276 Le Kremlin‐Bicêtre, France

Electron transfer reactions occur between the subunits of hemoglobin. If a sample of hemoglobin is initially with one type of subunit, alpha or beta, in the oxidized state (valency hybrids), incubation under anaerobic conditions tends to randomize the type of subunit that is oxidized. Within a few hours at pH 7, 25°C, a 0.1 mM Hb solution approaches a form with about 60% of beta chains reduced, indicating a faster transfer rate in the direction alpha to beta.

The electron transfer occurs predominantly between deoxy and aquo‐met subunits, both being high spin species. There was little observable electron transfer for samples saturated with oxygen or with CN bound to the oxidized subunits. The kinetics showed little dependence on the quaternary state of hemoglobin.

Incubation of oxidized cross‐linked HbA (DCL Hb from Baxter) with deoxy HbS also displayed electron transfer, implying a mechanism via inter‐tetramer collisions. The rate of transfer increases with increasing Hb concentration, confirming a mechanism based on collisions. These results suggest that in vivo, collisions between the Hb tetramers will help distribute electrons furnished by the reductase system present in the erythrocyte.

Production and Properties of Hemospan™, a Novel Malemide‐Pegylated Human Hemoglobin for Use as a Blood Substitute

K. D. Vandegriff, K. W. Chapman, J. M. Lohman, A. Malavalli and R. M. Winslow

Sangart, Inc., San Diego, CA 92121, USA

Hemospan™ is a polyethylene glycol (PEG)‐conjugated hemoglobin. The process involves site‐specific modification of human hemoglobin in a two‐step chemical reaction: 1) Thiolation using iminothiolane to add − SH groups on the surface of hemoglobin at primary amines, followed by 2) PEGylation using a maleimide‐activated PEG (MW = 5,000). Hemospan™ is being manufactured in a GMP‐approved facility at Sangart, Inc. Manufacturing is relatively simple; the entire production process is carried out aerobically, and does not require any chromatographic purification steps. There is an average of 5 PEG polymers per hemoglobin tetramer, giving a calculated average MW of 90 kDa. The product displays a single, homogeneous peak by size‐exclusion chromatography. Hemospan™ is formulated at a hemoglobin concentration of ∼ 4 g/dL such that the solution is hyperoncotic (COP ∼ 50 mm Hg) and has a viscosity approaching that of blood (∼ 2.5 vs 4 cP, respectively). It has high O₂ affinity, compared to unmodified human hemoglobin or blood, and is non‐cooperative: P50 ∼ 5 torr; Hill coefficient ∼ 1.2. The half‐life of Hemospan™ is ∼ 24 hours in rats. It does not elicit a pressor response, and has been shown to maintain normal acid‐base status in rats during complete exchange transfusions down to undetectable hematocrits. Hemospan™ has been tested in several pre‐clinical models for safety and efficacy, and is now in early‐phase human clinical trials.

Improved Survival Time of Double Insult Hemorrhagic Shock in Swine After Resuscitation with Hemospan™

D. Drobin,¹ J. Lohman,⁴ E. Malm,³ J. Bursell,³ L. Gottschalk,⁴ B. T. Kjellström,^3,2 and R. M. Winslow²

Departments of ¹Anesthesiology and ²Surgery, Karolinska Institute at Söder Hospital, Stockholm, Sweden; ³The Swedish Defense Research Agency, Stockholm, Sweden; ⁴Sangart Inc., San Diego, CA, USA

Oxygen therapeutics has been developed over decades. The clinical usefulness has been disputed as it was demonstrated that vasoconstriction and bradycardia seriously impeded oxygen delivery and survival. A novel product, Hemospan™, based on polyethylene‐glycol‐conjugated hemoglobin (PEG‐Hb) was tested in an animal model of severe, double insult hemorrhagic shock. Juvenile pigs (n = 19) were subjected to a 40% hemorrhage and resuscitated to normovolemia with either of four solutions (vide infra). Subsequently, animals were hemorrhaged again. Defined fractions of the calculated blood volume (15%, 15%, 10%, 10%, 5%, 5%, and 2,5%) were withdrawn with 15 minutes intervals, until death occured. Central and systemic vascular pressures and cardiac output were obtained, and analysis included blood chemistry and ‐gases. Test substances were PEG‐Hb 4% (MP4) or PEG‐Hb 4% in Pentastarch 5% (HS4). Pentastarch 10% (PS) or the animal's own shed blood served as controls. 70% of shed volume was infused during 30 minutes. Cardiac output, systemic and pulmonary pressure, and lactate sustained markedly longer after resuscitation with HS4 and MP4. The survival time was 49 minutes longer in the HS4 compared to the blood group (p < 0.03) and was in the order: HS4 > MP4 > PS > Blood.

To conclude, PEG‐Hb significantly increases the survival in this surgical hemorrhage model.

Use of Perflubron‐Based Emulsion (Oxygent™) as an Alternative to Intraoperative Blood Transfusion

P. E. Keipert

Alliance Pharmaceutical Corp., San Diego, CA 92121, USA

Oxygent™ is a 60% w/v perfluorochemical emulsion based on perflubron emulsified with lecithin, that has a median particle diameter of < 0.2 µm, and can be stored refrigerated for up to 2 years. Oxygent is currently in late stage clinical development as an intravascular oxygen carrier in patients undergoing general surgery, and is being developed in conjunction with Baxter Healthcare Corporation. Previous Phase 2 studies have demonstrated the ability of Oxygent to reverse physiological triggers and delay the need for blood transfusion (Spahn et al., [1999]), and to decrease allogeneic blood transfusion in cardiac surgery (Hill et al., [2002]). When used in conjunction with acute normovolemic hemodilution (ANH) in a multicenter European Phase 3 study in major general surgery, Oxygent was able to significantly reduce and avoid red cell transfusion through hospital discharge in patients experiencing an estimated surgical blood loss of > 10 mL/kg (Spahn et al., [2002]). In 2001, a parallel Phase 3 study in cardiac surgery had to be terminated early, due to imbalances in certain adverse events that appear to have resulted from overly aggressive autologous blood harvesting just prior to cardiopulmonary bypass. Alliance and Baxter are currently preparing to launch an international Phase 3 program in general surgery, using Oxygent to avoid the need for allogeneic blood without the need for any autologous blood harvesting or ANH.

Hemoglobin and Cell Signaling Mechanisms

L. H. Yeh and A. I. Alayash^*

Center for Biologics Evaluation and Research (CBER), Food and Drug Administration, Bethesda, Maryland 20899, USA

Cell‐free hemoglobins (Hb) developed as oxygen therapeutics are designed to correct oxygen deficit due to ischemia in a variety of clinical settings. Direct cytotoxic effects associated with Hb have been ascribed to redox reactions between Hb and biological peroxides (i.e. H₂O₂, LOOH, and ONOO⁻) in vitro. Biological peroxides have been implicated as regulators of redox sensitive cell signaling pathways. The effects of reactions between Hb and biologically relevant peroxides in the context of cell signaling have not been explored. Transcriptional activator hypoxia‐inducible factor (HIF‐1α) functions as global oxygen sensor and heme oxygenase, involved in heme degradation, were both expressed in endothelial cells under hypoxic conditions. Diaspirin cross‐linked Hb, an oxygen carrier introduced with hypoxia induces changes in the levels of these proteins, the extent of which is strongly dependent on the oxygen carrying and redox state of this Hb. Cell‐free Hb modulates key cell‐signaling pathways as it converts oxygen signal to a transduction signal that activates HIF‐1α, which may result in alterations of important physiological mediators.

Animal Models on Blood Substitutes Research: Advantages and Disadvantages

J. C. Briceño,^1,2 I. E. Rincón,¹ J. F. Vélez,¹ I. Castro,¹ P. J. Cabrales,² D. M. Gaitán,^1,2 E. Chávarro,^1,2 P. J. Bernal,^1,2 and C. E. Vásquez²

¹Fundación CardioInfantil Instituto de Cardiología and ²University of Los Andes, P.O. Box 4976 Bogotá, Colombia

As part of a Blood Substitutes Research Program several animal models have been implemented. The main purpose of these models is to evalaute efficacy or safety of Hemoglobin‐based (Hb‐BSs) and PFC‐based blood substitutes (PFC‐BSs). Given the ethical and economical restrictions currently adapted on animal research, it appears useful to review the advantages and limitations of these models.

Small animal models are very useful in testing safety and toxicity, specially in hemorrhagic shock and hemodilution models. Ethical restrictions on small animal models are less strict, and animal procurement, handling and care is less expensive. However, validation of results from these species to human beings is not straightforward; anesthetic, surgical and instrumentation procedures are more complex; and oxygen delivery and consumption studies are cumbersome due to difficult vascular access and sample volume restrictions.

Canine and porcine models are suitable for efficacy evaluation, mainly in hemodilution and cardiopulmonary bypass procedures. Physiological and hemodynamical similarity with human beings is remarkable; anesthetic, surgical and instrumentation procedures, are very similar to those performed in man, as it is oxygen consumption studies. On the other hand, ethical restrictions are stronger, and animal procurement and care is fairly complicated.

Effect of Hb‐Encapsulation with Vesicles Against H₂O₂ Reaction and Lipid Peroxidation

S. Takeoka, Y. Teramura, T. Atoji, and E. Tsuchida

Advanced Research Institute for Sci. & Eng., Waseda University, Tokyo 169‐8555, Japan

Hemoglobin (Hb) binds oxygen reversibly in the ferrous state and was gradually oxidized to metHb (Fe^III), generating reactive oxygen species(ROS). Hb was also reacted with those ROS to form metHb and ferrylHb (Fe^IV = O), followed by the release of ferric ion from the denatured Hb. Those products are considered to cause unfavorable side effects of Hb‐based oxygen carriers. In the reaction of a hemoglobin (Hb) solution with hydrogen peroxide (H₂O₂), metHb (Fe^III) and ferrylHb were produced, and H₂O₂ was decomposed by catalase‐like reaction of Hb. The aggregation of discolored Hb products due to heme degradation was accompanied with the release of iron. On the other hand, the concentrated Hb within the Hb vesicle was reacted with H₂O₂ permeated through the lipid membrane, and the same products as the Hb solution were formed within the vesicle. However, there were no turbidity change and no particle diameter change of the vesicles after the reaction with H₂O₂. No free iron was detected outside the vesicle, though iron was released from the denatured Hb inside the vesicle, indicating the barrier effect of the bilayer membrane against the permeation of ferric ion. When egg york lecitin (EYL) vesicles were added to the reaction mixture of Hb and H₂O₂, the peroxidation of EYL by ferrylHb and hydroxy radical generated from reaction of the ferric iron with H₂O₂ were observed. Whereas, no lipid peroxidation was observed in the case of the Hb vesicle dispersion because the saturated lipid membrane of the Hb vesicle should prevent the interaction of the ferrylHb and ferric iron with the EYL.

Permeability of Polyelectrolyte Microcapsules (PEMC)

H. H. Bäumler,¹ R. Georgieva,¹ S. Moya,² and H. Kiesewetter¹

¹Institute of Transfusion Medicine, Charité, Humboldt University of Berlin, D‐10098

²Max Planck Institute of Colloids and Interfaces, Golm, D‐14476, Germany

Microcapsules (PEMC) in the 100nm to 10 µm size range are of both scientific and technological interest, since they have potential applications as nano‐ and micro‐containers. Liposomes represent a well‐known example of spherical closed thin films, which in addition to their numerous applications as model biological membranes, have already been employed as drug delivery systems in pharmaceutics and cosmetics (Lasic, [1993]). PEMC fabricated from polyelectrolytes offer advantages in that they are permeable to small polar molecules and are extremely stable against chemical and physical influences (Bäumler et al., [2000]; Caruso et al., [1998]; Klitzing and Möhwald, [1995]).

The wall thickness determined by Atomic Force Microscopy as well as the bending modulus (BM) of the PEMC determined by micropipette was depended on the number of layers and the used polyelectrolytes. The thickness was in the range between 7 and 28 nm and the BM was between 3.5 · 10^{− 15} and 1.8 · 10^{− 14}, 4 to 5 magnitudes larger than the BM of red blood cells (Evans, [1983]). The permeability of the PEMC can be varied in the M_w range of a few Dalton up to some kDa.

Liposome‐Encapsulated Hemoglobin Provides Better Oxygenation in Ischemic, Collateralized Hamster Flap Tissue Than Cellular Hemoglobin

S. Schramm,¹ C. Contaldo,¹ H. Sakai,² S. Takeoka,² E. Tsuchida,² M. Leunig,¹ A. Banic,¹ and D. Erni¹

Depts. of ¹Orthopedic, Plastic and Hand Surgery, Inselspital University Hospital, 3010 Berne, Switzerland

²Polymer Chemistry, ARISE, Waseda University, Tokyo 169‐8555, Japan

In previous experiments, we have shown that the oxygenation in ischemic and hypoxic, collateralized hamster flap tissue could be improved by normovolemic hemodilution with a lipsosome encapsulated hemoglobin (Hb vesicles, HbV) solution. The aim of this study was to investigate the effect of HbV at various degrees of hemodilution. To this end, hamsters were exposed to a gradual normovolemic hemodilution of 20%, 40% and 60% blood exchange with 6% dextran 70 (Dx70) and HbV dissolved in Dx70 (HbV‐Dx70, Hb 7.5 g/dl, P50 = 15 mmHg). Microvascular blood flow (measured by intravital microscopy) continuously increased to approximately 150% in both groups. Hemodilution with Dx70 led to a transient increase in tissue oxygen tension from 10.4 ± 3.4 mmHg to 12.2 ± 3.0 mmHg at the 40% blood exchange (Hb 9.7 ± 1.2 g/dl, hematocrit 29.6 ± 5.6%), whereas tissue oxygen tension gradually increased with every step of hemodilution with HbV‐Dx70 from 8.2 ± 3.1 mmHg to 16.2 ± 5.3 mmHg (60% blood exchange, Hb 8.8 ± 0.9 g/dl, hematocrit 24 ± 5%). We conclude that the hypoxic, collateralized tissue may benefit from normovolemic hemodilution with HbV‐Dx70 even at a high degree of blood exchange, and that HbV delivers more oxygen to that tissue than cellular Hb.

Different Efficacy In Vitro of Hemoglobin Based Oxygen Carriers and Red Cells

E. Bucci,¹ T. L. Watts,² H. E. Kwansa,¹ and A. Fasano²

¹Department of Biochemistry and Molecular Biology, and ²Center for Vaccine Development, University of Maryland, School of Medicine Baltimore, MD 21201, USA

Samples of rabbit ileal mucosa were mounted in Ussing cuvettes perfused either with Ringer solution alone or with Ringer containing either cell free sebacoyl crosslinked hemoglobin A(DECA) or bovine red cells. DECA is human hemoglobin intramolecularly crosslinked with a sebacoyl residue. Its oxygen affinity has a P50 near 30 mmHg and an oxygen binding cooperativity with the Hill parameter n near 2.0. Bovine red cells were choosen because their oxygen affinity properties are very similar to those of DECA Classically in order to maintain tissue metabolism the perfusates are equilibrated with 95% oxygen, 5% carbon dioxide. However when 3 g/dl of DECA were added to the Ringer, transport could be maintanied with only 30% oxygen, while perfusates containing a suspension of bovine red cells, diluted to a hemoglobin concentration of 3 gr/dl, failed to maintain transport.

It appears that while oxygen transport by red cells is regulated by oxygen partial pressure, oxygen affinity, rate of oxygen release from hemoglobin, rate of diffusion of molecular oxygen, rate of oxygen consumption, instead, oxygen transport by cell free hemoglobin is a phenomenon practically unregulated, subject only to the rate of metabolic consumption. This encourages a revision of the design of red cell substitutes, so far based solely on the attempt of mimicking the oxygen affinity characteristics of blood.

Vascular Response to Infusions of a Non‐extravasating Hemoglobin Polymer

E. Bucci,¹ B. Matheson,² H. E. Kwansa,¹ A. Rebel,³ T. Mito,³ and R. C. Koehler³

¹Dept of Biochemistry, Medical Sch.,

²Dept of Physiology, Dental Sch., Univ. of Maryland, Maryland, USA

³Dept. of Anesthesiology and Critical Care Medicine, Johns Hopkins University, Baltimore, MD, USA

The mechanism for the pressor response to hemoglobin infusions is thought to be the intravascular scavenging of nitric oxide by cell‐free hemoglobin. Pressor response is also concomitant with short intravascular retention times and the extravasation into hilar lymph of kidneys . We investigated this relationship by infusing a hemoglobin polymer which would not extravasate. A polymer of bovine hemoglobin was developed by coupling directly the amino groups of one hemoglobin molecule with the carboxyl groups of an adjacent molecule, i.e.“zero link”. The polymer has a P50 near 4 mmHg, and radius R_h = 500 Å. After infusion, the zero‐link polymer did not appear in the hilar lymph of the rat and did not produce the increase of mean arterial pressure in the cat and in the rat These data indicate that as long as infused hemoglobin does not extravasate, it does not affect the mean arterial pressure. After 40% exchange transfusion in cats, cerebral blood flow was maintained at pre‐transfusion levels, in contrast to the increase with albumin transfusion, thereby indicating active regulation by oxygen delivery. When transfused in mice during middle cerebral artery occlusion, the brain infarct volume was smaller with the zero‐link polymer exchange transfusion than with albumin transfusion or controls.

These data suggest that the high oxygen affinity of the polymer did not prevent its oxygen delivery in vivo.

Comparison of the Effect(s) of Hemodilution with Two Hemoglobin‐Based Oxygen Carrying Solutions on Neutrophil and Platelet Activation in the Rat

M. Toussaint‐Hacquard,¹ V. Latger‐Cannard,² P. Menu,¹ T. Lecompte,² and C. Vigneron^1,2

¹Department of Hematology‐Physiology, University Henri Poincaré‐Nancy 1, 54001 Nancy;

²Department of Biological Hematology, CHU Nancy, 54500 Vandoeuvre les Nancy, France

The aim of this study was to evaluate and to compare the potential effect(s) on the polymorphonuclear leukocytes (PMNs) and platelets behaviour of two differently chemically modified HBOCs versus a medium molecular weight hydroxyethylstarch. Whole blood samples were withdrawn from an anaesthetized rat which undergone a normovolemic exchange transfusion of 20% of the total blood volume with Hb‐Dex‐BTC (Hb cross‐linked with Dextran; PM 400 kDa,) or αα‐Hb (intra‐crosslinked Hb; PM 64 kDa) or hydroxyethylstarch (Elohes®; PM 200 kDa). Rat blood samples collected before the injection and after 1 and 3 hours, were assessed for 1) the expression of adherence receptors CD62L, CD18 and CD11b, which reflected the activation state of the neutrophils by an indirect immunofluorescence method and by flow cytometry analysis and 2) whole blood impedance aggregometry using the collagen (5 µg/mL) as agonist. The expressions of the three leukocyte adherence receptors were similar for each solution, at each time point, to those noted before the hemodilution, and the aggregation pattern of platelets in citrated whole blood did not change. We concluded that the presence of large dose of HBOCs in the bloodstream, irrespective of the nature of their chemical modifications, did not appear to activate PMNs nor modify platelet aggregation pattern in the rat

Does Modified Hemoglobin Penetrate Inside Polymorphonuclear Neutrophil? Evaluation of the Potential Effect of Dex‐BTC‐Hb or αα‐Hb in the Hemodiluted Rat

M. Toussaint‐Hacquard,¹ V. Latger‐Cannard,² P. Menu,¹ B. Faivre‐Fiorina,¹ T. Lecompte,² and C. Vigneron^1,2

Departments of ¹Hematology‐Physiology, University Henri Poincaré, 54001 Nancy, France

²Biological Hematology, CHU Nancy, 54500 Vandoeuvre, France

To check the safety of cell‐free hemoglobin (Hb) in bloodstream, we have previously shown that Hb solutions did not appear to induce potential activation of the polymorphonuclear leukocytes (PMNs) using adherence receptor expression. However, as recently reported 1) a potentialisation of Escherichia Coli infection in the Hb presence, 2) a Hb penetration inside endothelial cells and 3) a receptor to the haptoglobin‐Hb complex (CD163) on some mononuclear cells surface, we hypothesized that Hb may penetrate PMNs and could hence interfere with their phagocyte functions. On anesthetized rats, 20% hemodiluted with Dex‐BTC Hb (Hb cross‐linked with Dextran; PM 400 kDa,) or αα‐Hb (intra‐crosslinked Hb; PM 64 kDa), were collected blood samples before and 0.25, 1 and 3 hours after exchange transfusion. The PMNs, isolated by density gradient, were analyzed by spectrophotometry (540 nm), flow cytometry and immuno histochemistry. After PMNs lyses, the suspension medium spectrophotometric analysis showed no Hb traces inside cytosol. Flow cytometric results showed a lack of antibody fixation against Hb neither on the membrane, nor inside the PMN. At each time points, immunohistochemistry data confirmed the absence of Hb inside PMN. In conclusion, whatever irrespective of their chemical modifications, Hb did not seem to penetrate inside PMNs, nor to bind to their membrane, shortly or later after Hb infusion in the rat.

Oxygen Carrying Plasma Expander – Potential Roles in Clinical Practice

B. Fagrell

Karolinska Institute at Department of Medicine, Karolinska Hospital, SE‐171 76 Stockholm, Sweden

Since several decades scientist have been trying to develop compounds that act like blood in delivering oxygen to the living tissues. The research has so far been very disappointing because of more or less severe side effects of the compounds so far tested. One of the more serious problems has been vasoconstriction, most probably cause by a direct effect on the endothelium of small vessels. A new compound – Hemospan™ – has now been tested for the first time in humans, and no side effects whatsoever have been noticed. This product is composed of hemoglobin encapsulated in PEG (polyethylene glycol) and this composition seems to prevent the serious side effect of vasoconstriction. In animal studies it has not shown any tendency to vasoconstriction, but rather a vasodilation. It has also been found to increase oxygenation of tissues, most probably due to a recruitment of nutritional capillaries. This opens up some very intriguing possibilities to use this substance not only as a plasma expander, but also to improve the oxygenation of ischemic tissues. Clinical conditions where improvement of tissue oxygenation is of vital importance is e.g. myocardial ischemia, peripheral critical limb ischemia and ischemic stroke. It has been shown in animal preparations that the oxygen extraction across the capillary bed seems to be improved by Hemospan™, and this opens up the possibility to deliver more oxygen to ischemic tissues, which may have great clinical implications in many clinically severe conditions.

The Theory and Application of Intravascular Microbubbles as an Ultra‐effective Means of Transporting Oxygen and Other Gases

C. E. G. Lundgren, G. W. Bergoe, and I. Tyssebotn

Center for Research and Education in Special Environments, School of Medicine and Biomedical Sciences, University at Buffalo, Buffalo, NY, USA

Theoretically, volume‐stabilized microbubbles may effectively support gas exchange between lungs and tissues (Van Liew and Burkard, [1996]). Subcapillary sized bubbles can be generated by i.v. injection of an emulsion of dodecafluoropentane (DDFP, boiling temp 29°C). At body temperature, the DDFP particles evolve into bubbles initially composed of DDFP gas and then equilibrate with O₂ and CO₂ tensions in surrounding tissues. Thus, O₂ is transported from the lungs and CO₂ from the tissues. The feasibility of this method for life‐sustaining oxygen supply has now been demonstrated in erythrocyte deprived normovolemic rats both anesthetized (Tyssebotn et al., [1999a]) (100% mortality in controls) and awake (Tyssebotn et al. [1999b]), in erythrocyte depleted normoxic pigs (Tyssebotn et al., [2001]) and in treatment of severe experimental right‐to‐left shunts in pigs (Lundgren et al., [1999]). Injecting hematologically normal rats with DDFP emulsion provides for markedly increased tissue (muscle) O₂ tensions. Moreover, the emulsion greatly enhanced the elimination of tissue nitrogen in O₂ breathing pigs (Lundgren et al., [2001]). Finally, at this meeting we report on the life‐sustaining effect of the emulsion in a severe hemorrhagic shock in pigs. Theoretically, 1 ml of emulsified DDFP provides for the O₂ consumption of a resting O₂ breathing adult person. The efficiency of the bubbles for O₂ transport is underscored by the extremely small doses (0.002–0.014 ml/kg body weight) of DDFP, in the form of a 2% emulsion, which were used in the referenced experimental studies.

Intravascular Microbubbles: Successful Treatment of Potentially Fatal Hemorrhagic Shock in Pigs

I. Tyssebotn, G. W. Bergoe, and C. E. G. Lundgren

Center for Research and Education in Special Environments, University at Buffalo (SUNY), Buffalo, New York, USA

Infusion (i.v.) of a 2% dodecafluoropentane emulsion (DDFPe) forming microbubbles which provide O₂ transport and volume expansion preserves vital functions in a potentially lethal hemorrhagic shock model in pigs. Pentobarbital‐anesthetized pigs (n = 8) were artificially ventilated with air plus a small amount of O₂ to normalize PaO₂ at 100 mm Hg. Blood loss from an artery was standardized at 38 ± 1 (SD) ml/kg over 40 min. After bleeding, the systolic blood pressure (SAP) was 71 ± 2 mm Hg. Twenty min later control (C) (n = 4) and treatment (T) pigs (n = 4) received 30 ml of lactated Ringer’s solution i.v. plus 0.3 ml/kg of blank in Cs and 0.3 ml/kg of DDFPe in Ts. After sham treatment, the Cs displayed steadily declining arterial blood pressures and muscle PO₂ and died in 67 ± 39 min. By contrast, in Ts the muscle PO₂ increased during treatment and in 3 of them the SAP increased to above 100 mm Hg and never declined below 70 mm Hg during a 6 hr (post‐treatment) observation period (intentional termination). One T lived with a blood pressure above 70 mm Hg for close to 3 hrs, then became hypotensive and died. On autopsy, its kidneys showed a large number of cysts and minimal volume of (macroscopically) normal tissue. The other animals had normal intrathoracic and abdominal organs. These preliminary results suggest that i.v. administration of a 2% DDFPe, in extremely small doses, may provide effective first‐line treatment of otherwise lethal hemorrhagic shock.

Study supported by US Army Medical Research Acquisition Activity Grant (ID: DAMD170110778) and by Sonus Pharmaceuticals, Inc.

Low‐Concentrated Perfluorochemical Emulsion (LPFCE) Transfers Oxygen Together with the Remaining Red Blood Cells (RBC) After Massive Blood Replacement

E. I. Maevsky,¹ G. R. Ivanitsky,¹ I. N. Kusnetzova,² B. I. Islamov,¹ and V. V. Moroz³

¹Institute of Theoretical and Experimental Biophysics RAS, 142290, Pushchino, Moscow Reg.

²Russian Research Institute of Hematology and Transfusiology, S‐Petersburg, Russia

³Research Institute of General Reanimatology RHAS, Moscow, Russia

After a massive blood loss and hemodilution a hematocritis and RBC content reduce to 30–40% of the “norm” level. Therefore O₂ capacity of the remaining RBC is sufficient to transfer the necessary amount of O₂. The problem lies not in complete replacement of the lost RBC but in facilitating O₂ delivery to tissues from the remaining RBC. Since only 25–35% of O₂ connected with Hb can be utilized in tissues the volume and the rate of O₂ release from the remaining RBC are insufficient, and a low O₂ utilization becomes a critical parameter at hemodilution when cardiac output grows considerably. Under these conditions LPFCE can increase utilization of O₂. We have developed a LPFCE of 10 vol. % named Perftoran with a low viscosity and good rheological properties. It has the much higher rate of saturation and desaturation with O₂ than that of RBC. Due to a small size of particles (< 0.1 mcm) with fluorocritis of 0.03–0.05 a diffusion exchange surface participating in O₂ transfer is by the order higher than in RBC. Finally, small PFCE particles can flow along narrowed vessels that are inaccessible for RBC. To make the remaining RBC and RFCE work together it is enough to breathe a mixture of O₂ and air at FiO₂ about 0.5. The experiments on animals with blood replacement and clinical studies testify that joint functioning of LPFCE and the remaining RBC increase O₂ delivery, preserve energy supply function of mitochondria, decrease lactate/pyruvate and increase ATP/ADP in tissues.

New Stable Fluorocarbon Emulsions with Small Particle Sizes

M. P. Krafft, J. L. Thomas, S. Marie Bertilla, and P. Marie

Chimie des Systèmes Associatifs, Institut Charles Sadron (CNRS). 6 rue Boussingault, 67 083 Strasbourg Cedex, France

For fluorocarbon emulsions, small droplet sizes translate into prolonged intravascular persistence and reduced side effects. We report here on the obtaining of highly stable, small‐sized and narrowly dispersed perfluorooctyl bromide (PFOB) emulsions stabilized by combinations of egg yolk phospholipids (EYP) and semifluorinated alkane C₆F₁₃C₁₀H₂₁ (F6H10). After 6 months at 25°C, the mean droplet diameter was only ∼ 80 nm, as compared to ∼ 180 nm for the reference emulsion stabilized by phospholipids alone, as assessed by photosedimentation and quasielastic light scattering spectroscopy. In parallel, a co‐surfactant effect has been demonstrated for C₈F₁₇C₁₆H₃₃ (F8H16) at the water/PFOB interface using pendant drop interfacial activity measurements, thus supporting the hypothesis that semifluorinated alkanes are preferentially located in the phospholipid interfacial film.

A series of new (perfluoroalkyl)alkylpoly(oxyethylene) derivatives: C_nF_{2n + 1}(CH₂)_mC(O)O(CH₂CH₂O)_xCH₃ (FnHmPOE), with n = 4, 6, 8, 10; m = 2, 4, 10, and x = 16, 45 and 113, was also synthesized. Some of these FnHmPOEs were found to stabilize PFOB emulsions as efficiently as FnHm diblocks. The average diameter of emulsion droplets stabilized by equimolar mixtures of EYP and FnHmPOE (x = 45) was only ∼ 90 nm, with a narrow particle size distribution, after 6 months at 25°C. Such fluorinated POE‐derived surfactants may also help increase the intravascular persistence of fluorocarbon emulsions.

The Development and Manufacture of Perfluorochemical Emulsion Perftoran for Blood Replacement, Ischemic Diseases Treatment and Transplantology

G. R. Ivanitsky,^1,2 K. N. Makarov,³ L. L. Gervitz,³ E. I. Maevsky,^1,2 B. I. Islamov,¹ G. M. Kulakova,^1,2 RYa Senina,² SYu Puskin,² and I. A. Maslennikov²

¹Institute of Theoretical and Experimental Biophysics RAS and

²OJS SPC Perftoran, 142290, Pushchino, Moscow Reg.

³Institute of Elementorganic Compounds RAS, Moscow, Russia

Perftoran (PF) is an emulsion of 10 vol.% of perfluorochemicals (PFCs), the main two being perfluorodecalin (PFD) and perfluoromethylcyclohexylpiperidine (PFMCP) in ratio 7:3. The half‐time for removing PFD and PFMCP from the organism is about 2 and 12 weeks respectively. PF contains a small amount of other PFCs whose physical‐chemical properties are between those of PFD and PFMCP. Morphological studies revealed no damages or residual effects of PFCs after temporary occupation of the tissues. As a surfactant we used Proxanol‐268 resembling Pluronic F68 but having lower toxicity. PF is manufactured with the use of an original procedure permitting to obtain narrow particles distribution with average size of 0.07 mcm and to prevent a peroxide formation in the surfactant. Thanks to all that the reactogenicity of PF measured with leucopoenia index in rabbits is rather small. It was found that the presence of particles bigger than 0.12–0.14 mcm was one of the causes for reactogenicity. P contains no particles > 0.14 mcm. PF is compatible with albumin and different crystalloid solutions. PF is stored at –18°C for 3 years or at 4°C for 3 weeks. Experimental studies have shown that PF is harmless, not toxic, not cancerogenic, not mutagenic, has small influence on hematopoesis, improves hemodynamics, catabolism of foreign compounds and organ preservation.

PF is highly effective in kidney transplantation, cardioplegia, massive blood replacement and treatment of circulation disorders.

Research on Future Generations of Red Blood Cell Substitutes

T. M. S. Chang

Dept. of Physiology, Medicine & Biomedical Engineering, Faculty of Medicine, McGill University, Montreal, Quebec, Canada H3G 1Y6

Diacids (Chang, Science 1964) and glutaraldehyde (Chang BBRC 1972) have been used to cross‐link hemoglobin (Hb) into Polyhemoglobin (PolyHb). The modern PolyHbs from Northfield, Biopure and Hemosol are in the final phases of Phase III clinical trials for peri‐operative uses. For some other uses, one has to consider the potential for ischemia‐reperfusion injuries due to sustained ischemia in hemorrhagic shock or other causes. In an ischemia‐reperfusion intestinal rat model, after 90 minutes of ischemia, reperfusion with our laboratory prepared PolyHb resulted in significant increases in oxygen radicals, but minimal when using our 2nd generation glutaraldehyde crosslinked PolyHb‐superoxide dismutase‐catalase (PolyHb‐SOD‐CAT) (D'Agnillo & Chang, Nature Biot 1998). Using a transient global ischemia‐reperfusion rat brain model (Powanda & Chang, Art Cell Blood Sub Imm.Biot 2002, 30:25–41) we recently showed that there was no significant ischemia‐reperfusion effect if ischemia was 30 minutes or less. However, after 60 minutes of global cerebral ischemia, except for PolyHb‐SOD‐CAT, the other solutions tested including PolyHb, SF‐Hb, oxygenated saline & SF‐Hb containing SOD & CAT in solution, resulted in disruptions of blood brain barrier and brain edema. This will be discussed in details. Our continuing study on a 3rd generation biodegradable polymeric nanocapsules containing Hb and rbc enzymes (Chang & Yu, 2001) has substantially increased their circulation time.

Oxygen Delivery to Tissue vs. Microvascular Function in Blood Replacement with Oxygen Carrying Plasma Expanders: Who Has the Priority?

M. Intaglietta

Department of Bioengineering, University of California, San Diego, La Jolla, California 92093, USA

Microvascular function consonant with tissue survival requires a threshold of capillary pressure needed for maintaining functional capillary density (FCD). Arrival at the transfusion trigger using conventional plasma expanders causes blood viscosity to be about half normal, and if blood pressure is normal, the necessary vascular resistance is consequent to vasoconstriction, which regulated capillary pressure. Thus correction of further blood losses also requires maintenance of blood viscosity in order to avoid further constrictor responses. Molecular hemoglobin solutions used as oxygen carrying plasma expanders present a gamut of vasoconstrictor responses and plasma viscosities. This is due to their effective molecular dimensions that regulate: 1) Facilitated oxygen diffusion, the determinant of autoregulatory vasoconstriction; 2) Plasma viscosity, the determinant of FCD; and, 3) Overall vasoconstriction that determines the oxygen consumption of the microcirculation. In this scenario tissue survival following the use molecular hemoglobin results from the synergy of oxygen delivery and blood/plasma viscosity, where minimal support of oxygen carrying capacity is sufficient if FCD is maintained.

Research supported by USPH NHLBI Grants R24‐HL64395, HL 62354 and HL62318.

The Free Volume Theory of Mechanochemical Transduction

J. A. Frangos

La Jolla Bioengineering Institute; La Jolla, CA, USA

The ability of cells to respond to hydrodynamic stimuli is ubiquitous amongst all cells and organisms, and has been implicated in a number of physiological and pathological processes. While many of the biochemical transduction pathways have been characterized, the primary mechanoreceptor(s) remain(s) unknown. It is our hypothesis that hydrodynamic shear destabilizes the plasma membrane, leading to a decrease in membrane microviscosity, or more precisely, an increase in membrane free volume. Mechanochemical transduction is proposed to occur when membrane‐associated signaling proteins are activated by the increase intramolecular mobility.

A number of studies have implicated a role of heterotrimeric G proteins in the mediation of cellular responses to fluid shear stress and stretch. Studies from our lab demonstrate that heterotrimeric G proteins are rapidly activated by hydrodynamic shear, representing the earliest known biochemical response to mechanical stimulation presented. Furthermore, both fluid shear stress and membrane fluidizing agents activate these G proteins in the absence of classical G protein coupled receptors. Using fluorescent molecular rotors it was recently shown that hydrodynamic shear increases membrane free volume. Taken together, these results demonstrate that hydrodynamic shear stress stimulates cellular responses by increasing membrane fluidity and activating heterotrimeric G proteins.

The presentation will discuss these results in the context of the membrane free volume theory of mechanochemical transduction.

Microvascular Oxygen Distribution and Capillary Perfusion During Induced Hyperoxia and Hypoxia

A. G. Tsai,¹ P. Cabrales,^1,2 H. Sakai,³ and M. Intaglietta¹

¹Dept. of Bioengineering, University of California, San Diego, La Jolla, California, USA

²Dept. of Mechanical Engineering, Universidad de los Andes, Bogotá, Colombia

³Dept. of Polymer Chemistry, ARISE, Waseda University, Tokyo, Japan

Microvascular perfusion and local tissue pO₂ distribution was characterized in the subcutaneous skeletal muscle of the unanesthetized hamster skinfold preparation during hypoxia (10% O₂/90% N₂) and hyperoxia (100% O₂). Arterial blood sampled from the carotid artery had pO₂ of 477 and 20 mmHg after 15 min exposure to high and low oxygen mixtures (baseline = 60 mmHg). Both groups did not exhibit any significant change in mean arterial blood pressure and heart rate. Hyperoxia caused large feeding (A1) and transverse arterioles (A2) to vasoconstrict by 10 and 16% respectively. During hypoxia, A1 was statistically unchanged and A2 dilated by 35%. Hyperoxic exposure resulted in feeding arterioles pO₂ and interstitial tissue pO₂ to be 97 and 32 mmHg respectively. Hypoxic conditions resulted in feeding arterioles pO₂ and interstitial tissue pO₂ to be 13 and 3 mmHg respectively. Baseline pO₂ distribution in feeding arterioles and interstitial tissue pO₂ were 58 and 21 mmHg respectively. Functional capillary density (FCD) was reduced in hyperoxia by 26% but remained unchanged from baseline during hypoxia. Results suggest that without a change in the perfusion pressure, arteriolar vasoconstriction in responses to high pO₂ due to hyperoxia is a likely mechanism leading to the observed decreased in FCD.

Research supported by NIH R24‐HL64395, HL62354 and HL62318.

Molecular Dimensions of Hb‐Based O₂ Carriers Determine Constriction of Resistance Arteries and Hypertension in Conscious Hamster Model

H. Sakai,¹ A. G. Tsai,² S. Takeoka,¹ E. Tsuchida,¹ and M. Intaglietta²

¹Advanced Research Institute for Sci. & Eng., Waseda University, Tokyo 169‐8555, Japan

²Dept. of Bioengineering, University of California, San Diego, La Jolla, CA 92093‐0412, USA

The effect of molecular size of Hb‐based O₂ carriers on the diameter and blood flow of resistance arteries (A₀, diameter, 158 ± 21 µm) and mean arterial pressure (MAP) were studied in the conscious hamster dorsal skinfold model. Intra‐molecularly crosslinked Hb (XLHb), PEG‐conjugated pyridoxalated Hb, hydroxyethylstarch‐conjugated XLHb, glutaraldehyde‐polymerized XLHb and PEG‐conjugated Hb‐vesicles (PEG‐HbV) were synthesized. Their molecular diameters were 7, 22, 47, 68 and 224 nm, respectively. The top load infusion of 7 ml/kg of XLHb (5 g/dl) caused the immediate increase of MAP, which was 34 ± 13 mmHg higher 3 hrs after infusion. There was a simultaneous decrease in diameter of A₀ (79 ± 8% of basal value) that caused blood flow to decrease throughout the microvascular network. The diameter of smaller arterioles did not change significantly. Infusion of O₂ carriers of greater molecular size resulted in lesser vasoconstriction and hypertension, PEG‐HbV showing the smallest changes. Infusion of albumin produced no microvascular or systemic effects. Constriction of resistance arteries was correlated to the level of hypertension, and the responses proportional to the molecular dimensions of Hb‐based O₂ carriers. The underlying mechanism is likely that the effects are related to the diffusion properties of the different Hb molecules.

Increased Oxygen Consumption by the Microvascular Wall Due to Vasoactive Hemoglobin Solutions

B. Friesenecker,^1,4 A. G. Tsai,⁴ R. Wettstein,^2,4 P. Cabrales,^3,4 and M. Intaglietta⁴

¹Dept. of Anesthesia and Critical Care Medicine, The Leopold Franzens University of Innsbruck, Innsbruck, Austria. ²Division of Plastic and Reconstructive Surgery, Inselspital University Hospital, Berne, Switzerland. ³Dept. of Bioengineering, Universidad de los Andes, Bogotá, Colombia. ⁴Dept. of Bioengineering, University of California, San Diego, La Jolla, California, USA

In normal conditions the microvascular wall consumes as much as 25% of the total oxygen delivery to the tissue according to estimates based on the measurement of oxygen loss from the arteriolar microvessels in the awake hamster skin fold model as evidenced by the pO₂ gradient across the arteriolar wall. In 50–60 µm diameter arterioles the wall gradient is 18.5 mm Hg, of which 12.0 mmHg are due to the geometry of the system and the remainder 6.5 mmHg results from oxygen consumption. Isovolemic exchange of αα‐Hb (15 g/dl) caused the arteriolar wall gradient to increase to 26.6 mmHg and reduced tissue pO₂ to 10.8 mmHg (vs. 22.0 mmHg normal). By contrast the infusion of prostaglandin PGE1 caused the arteriolar wall gradient to decrease to 15.6 mmHg and tissue pO₂ to rise to 31.9 mmHg. In both instances blood oxygen carrying capacity was the same. These findings lead to a conceptual model for oxygen partition in the tissue showing that vasoconstriction significantly impairs microvascular function because of the decreased oxygen supply and the increased oxygen consumption by the microcirculation.

Supported by NIH HL62318, HL 64395, HL62354, Jubiläumsfonds ÖNB #5526.

MalPEG‐Hb, a New Hemoglobin‐Based Red Cell Substitute That Does Not Produce Vasoconstriction

R. M. Winslow^1,2 and K. D. Vandegriff¹

¹Sangart, Inc. and ²University of California, San Diego, California, USA

The principal limitation to the development of hemoglobin‐based red cell substitutes has been their propensity to cause vasoconstriction, leading to hypertension, reduced O₂ delivery, and gastrointestinal side effects. The most popular explanation for these effects is the well‐known scavenging of NO by hemoglobin. However, our group reported that various modified hemoglobins with different degrees of vasoactivity had essentially the same NO binding affinity (Rohlfs et al., [1998]), which led us to formulate an alternative hypothesis. Since plasma hemoglobin delivers O₂ very efficiently to vascular walls (Page et al., [1998]), and arterioles are known to constrict in response to hyperoxia (Lindbom et al., [1980]), we suggested that vasoconstriction results from oversupply of O₂ to arteriolar walls when freely diffusive hemoglobin is present in the vascular space. Based on this theory, we designed a modified hemoglobin with reduced diffusive properties. In vitro studies in an artificial capillary (McCarthy et al., [2001]) have shown that MalPEG‐Hb releases O₂ in a fashion that is nearly identical to that of red blood cells, in spite of a very low P50 (∼ 6 mmHg). Animal studies demonstrate the absence of hypertension, and marked efficacy in protection during simulated hemorrhage. We believe MalPEG‐Hb has potential to become a safe and effective hemoglobin‐based red cell substitute.

Increased Tissue Oxygenation After Resuscitation with MalPEG‐Hemoglobin in Hemorrhagic Shock in Awake Hamsters

R. Wettstein,^1,2 D. Erni,² A. G. Tsai,¹ R. M. Winslow,^1,3 and M. Intaglietta^1,3

¹Department of Bioengineering, University of California, San Diego, California, USA

²Division of Plastic Surgery, Inselspital, Bern, Switzerland

³Sangart Inc., San Diego, California, USA

The effect of resuscitation with polyethylene glycol‐modified human hemoglobin (PEG‐Hb) with a p50 of 5.5 mmHg was compared to shed blood (SB).

Hamsters implemented with a skinfold chamber were hemorrhaged 50% of the estimated blood volume. Microvascular diameter, flow velocity, functional capillary density (FCD), pO₂ in vessels and extravascular tissue, as well as mean arterial pressure and base excess were analyzed for one hour after resuscitation.

PEG‐Hb increased FCD to 64% vs. 44% for SB. Microvascular flow increased 16% for MalPEG‐Hb relative to baseline and remained decreased by 44% for SB. Hb concentration was 10.4 (SB) and 7.5 g/dl (PEG‐Hb), tissue pO₂ 19 and 8 mmHg. The presence of 0.9 g/dl PEG‐Hb increases oxygen extraction per gram of red cell hemoglobin in the tissue analyzed compared to SB and improves blood flow and provides an oxygen reserve to anoxic regions.

Research supported by NIH R24‐HL64395, HL 62354 and HL62318.

Cardiopulmonary Distress Caused by Liposome‐Encapsulated Hemoglobin in Pigs: Mechanism and Strategies for Prevention

J. Szebeni,¹ L. Baranyi,¹ S. Savay,¹ M. Bodo,² R. Bünger,³ and C. R. Alving¹

¹Dept. Membrane Biochemistry and ²Resuscitation Medicine, Walter Reed Army Institute of Research, Washington DC 20307, USA

³Dept. Anatomy and Physiology, USUHS, Bethesda, MD 20814, USA

Intravenous injection of minute (milligram) amounts of liposome‐encapsulated hemoglobin (LEH) in pigs can cause massive hemodynamic changes, including maximal pulmonary hypertension, systemic hypotension, fall of cardiac output and, ultimately, death via anaphylactoid shock. The phenomenon is due to complement (C) activation by the phospholipid bilayer membrane of liposomes, which, in turn, is a consequence of the binding of natural anti‐phospholipid and anti‐cholesterol antibodies to the vesicles. The composition, size and administration method of liposomes all have significant influence on the reaction. Factors with adverse impact include the presence of hemoglobin and negatively charged phopsholipids on the vesicle surface, high (> 50%) amount of cholesterol in the membrane, large, heterogenous liposome size and bolus injection vs. slow infusion. The pulmonary hypertensive effect of liposomes showed strong correlation with thromboxane (TX)B₂ release into the blood, suggesting that TXA₂‐mediated pulmonary vasoconstriction is a key event in the pathomechanism. Consistent with the causal roles of C and TXA₂, inhibitors of C activation or action, as well as the cyclooxigenase blocker, indomethacin, were effective inhibitors of the above cardiopulmonary changes. We conclude that pigs provide a sensitive model to study the potential adverse biological effects of LEH that arise from C activation.

Possible Role for “Artificial Blood” in Military Medicine

B. T. Kjellström

The Swedish Defense Research Agency, Experimental Traumatology Unit, and Department of Surgery, Karolinska Institute, both at Söder Hospital, SE‐118 83, Stockholm, Sweden

Exsanguinating hemorrhage is the cause of death in almost 50% of battlefield trauma victims. Although many of these deaths occur within minutes and usually are due to severe damage to the heart or major blood vessels, a substantial number of lives could be saved if an oxygen carrying blood substitute could be administered early after trauma; preferably on or close to the scene. New thinking in resuscitation has led to approval of hypertonic saline (US Army) and hypertonic saline‐dextran (Swedish Armed Forces). However, none of these plasma expanders can carry oxygen. Typically, in military medicine, the earliest point where blood transfusions can be administered is at the field hospital or maybe the brigade aid station. Often the military medical organization has to depend on civilian sources of blood supply, a fact that generates logistic problems, especially in overseas operations. Another option, sometimes deployed, would be to recruit blood donors among own troops. This practice can be disputed from ethical standpoints.

Consequently, the logical step would be the development of oxygen carrying resuscitation formulas meeting military‐specific demands. Such formulas should initially be evaluated in pre‐clinical trauma studies. They should allow low volume resuscitation, be possible to store at ambient temperature and be administered by paramedics on or close to the battlefield. The possibility of on scenario manufacturing would be important, especially in developing countries and prolonged operations (cf. UN peace making/peace keeping).

Restoration of Tissue Oxygen Delivery and Extraction in the Microcirculation by Blood, Molecular Hemoglobin and Dextran After Hemorrhagic Shock

H. Kerger,¹ G. Groth,¹ A. G. Tsai,² and M. Intaglietta²

¹Department of Anesthesiology, University Hospital of Mannheim, D‐68135 Mannheim, Germany

²Department of Bioengineering, University of California, San Diego, La Jolla, CA 92093, USA

Oxygen delivery (DO₂) and consumption (VO₂) were evaluated 30 min after resuscitation from hemorrhagic shock (2 h; 40 mmHg) to determine the efficacy of o‐raffinose cross‐linked oligomerized hemoglobin solution (Hb_molecular, 10 g/dl) when compared to shed blood and dextran 70 kDa (6%). Studies were carried out in the hamster skin fold preparation. Resuscitation volume was 50% of shed blood. Microcirculatory parameters were arteriolar and venular blood flow and pO₂ and extravascular pO₂. This data combined with information on blood hemoglobin (Hb) content, the oxygen dissociation curve for hamster blood and Hb_molecular, was used to calculate the relative changes in DO₂ and VO₂ during resuscitation. Blood trans fusion leading to 11 g/dl Hb provided the largest DO₂ and VO₂. Resuscitation with Hb_molecular [5.5 g Hb/dl in red blood cells (RBC) and 3.5 g Hb/dl in plasma] caused DO₂ and VO₂ to be 65% relative to blood transfusion, while Dextran infusion (Hb 5.0 g/dl) yielded 33 and 41%. Although DO₂ and VO₂ were significantly diminished in Hb_molecular resuscitation, arterial pressure and pO₂, base excess and capillary perfusion attained the same level of recovery as blood transfusion. These results may in part be due to the pressor effect of Hb_molecular and the fact that the transfusion trigger for this species is < 5.5 g Hb/dl. Compared to RBC, effective DO₂ capacity of Hb_molecular is only 40%, a factor due to the high P₅₀ of 52.5 mmHg.