Blood Substitutes

Updated: Feb 03, 2015
  • Author: Sara J Grethlein, MD; Chief Editor: Emmanuel C Besa, MD  more...
  • Print


The attempt to develop a viable blood substitute spans more than 7 decades. These efforts have essentially focused on the ability of red blood cells to carry oxygen. Hence, most of the products that are in advanced-phase clinical trials are derivatives of hemoglobin and are known as hemoglobin-based oxygen carriers (HBOCs). [1]

Today, an increase in the number of elective surgeries and the still prevalent but small risk of transmission of blood-borne pathogens such as HIV have served as a stimulus to develop a synthetic substitute for human blood, more specifically for development of a red blood cell substitute.

However, to date, no oxygen-carrying blood substitutes are approved for use by the US Food and Drug Administration. This fact highlights not only the challenges that exist in formulating an effective blood substitute but also the immense potential that exists in this field.

The physiology of oxygen transport can be described as follows:

One gram of adult hemoglobin binds to 1.39 milliliters of oxygen. The oxyhemoglobin dissociation curve (see the image below) has a characteristic sigmoid shape due to the cooperative effect that exists between the multiple oxygen binding sites on the hemoglobin molecule. Factors that modify the ability of hemoglobin to bind oxygen include body temperature, pH of blood and the concentration of 2,3-diphosphoglycerate (2,3-DPG).

The oxyhemoglobin dissociation curve plots the pro The oxyhemoglobin dissociation curve plots the proportion of hemoglobin in its saturated form on the vertical axis against the prevailing oxygen tension on the horizontal axis. Image courtesy of Wikimedia Commons.

Characteristics of an Ideal Blood Substitute

The most important adverse effects of blood transfusion relate to the antigenicity of donor blood and its ability to transmit infections. Hence, an ideal blood substitute should lack antigenicity and eliminate, or at least substantially reduce, the ability to transmit infections. In addition, it should be readily available, should have a long half-life, and should be capable of being stored at room temperature. The biologic properties of an ideal blood substitute should include a reasonable amount of oxygen delivery, when compared to normal human red blood cells. [2]


Potential Uses

If such a blood substitute were indeed developed, it would presumably play a major role in the setting of trauma care and for elective surgeries. It would also benefit patients with medical conditions who are in need of long-term blood transfusions, such as patients with myelodysplastic syndrome and aplastic anemia. These products could also be used as organ preservatives to prevent or decrease reperfusion injury to donor organs. Religious and ethnic groups with concerns regarding the use of human derived blood products may accept these substances, which would provide a significant patient care advance.


First-Generation Products

Perfluorocarbon emulsions

Perfluorocarbons are chemically inert molecules containing primarily, as the name suggests, fluorine and carbon atoms. They are capable of dissolving large amounts of many gases, including oxygen. These molecules are hydrophobic in nature, and hence have to be emulsified prior to intravenous administration. [3]

After intravenous administration, the droplets of this emulsion are taken up by the reticuloendothelial system and then slowly broken down. They are then transported to blood, where they are bound to lipids and move to the lungs. In the absence of significant in vivo metabolism, perfluorocarbons are removed from the body by exhalation.

Perfluorocarbons demonstrate a linear oxygen dissociation curve in contrast to the sigmoid dissociation curve of blood. Hence, elevated arterial partial pressures of oxygen enhance oxygen transport by these molecules. However, this linear relationship can also work as a disadvantage since most of the oxygen is released prior to the arrival of the oxygen-laden molecule in the capillary network where the partial pressure of oxygen is lower, and hence the need for oxygen is greater.

Among the various perfluorocarbon emulsions that have been developed are the molecules perfluorooctyl bromide (perflubron), perfluorodecyl bromide, and perfluorodichlorooctane. [4]

The first product to be marketed contained perfluorodecalin and perfluorotripropylamine emulsified with Pluronic F-68 and called Fluosol-DA. It was manufactured by the Green Cross Corporation (Japan) and Alpha Therapeutics (Grifols USA, Los Angeles, Calif). It was approved by the FDA for use in percutaneous transluminal coronary angioplasty. However, due to marginal efficacy, a short effective half-life, temperature instability, low oxygen-carrying capacity, and adverse effects such as acute complement activation and disruption of pulmonary surfactant, this product has since been withdrawn from the market.

A more stable emulsion containing 58% weight per volume perfluorooctyl bromide and 2% weight per volume perfluorodecyl bromide emulsified in 3.6% weight per volume egg yolk phospholipid concentrated to 60% weight per volume was then developed by Alliance Pharmaceutical Corporation (San Diego, Calif). With greater stability at room temperature than Fluosol-DA and with an increased oxygen carrying capacity (4-5 times greater than Fluosol-DA) this compound, called Oxygent, showed great promise initially. However, phase III trials showed an increased incidence of stroke in treated patients compared to controls and trials have been halted. [5]

Other PFC emulsions that have been developed include Perftoran (Perftoran, Moscow, Russia; also registered and in use in Mexico under the trade name Perftec), Oxycyte (Synthetic Blood International, Costa Mesa, Calif; entering phase II trials), and Oxyfluor (HemaGen/perfluorocarbon, St Louis, Mo; discontinued due to safety issues). [6]

A study by Johnson et al indicated that dodecafluoropentane may be an effective oxygen-delivery agent for the treatment of hypoxic medical conditions. [4] In an in vitro comparison of 2% weight per volume emulsions of dodecafluoropentane, perfluorodecalin, and perfluorooctyl bromide, the investigators found that at 21º C, dodecafluoropentane absorbed 3 times more oxygen than did the other 2 agents, and that at 37º C, it absorbed 7 times more oxygen. [4]


Stroma-Free Hemoglobin

Stroma-free hemoglobin has been investigated as an oxygen carrier since the 1940s, when researchers realized that native hemoglobin is not antigeneic. A solution containing stroma-free hemoglobin has many advantages over red blood cells, including the ability to withstand sterilization and a shelf life of approximately 2 years at room temperature for some products.

Solutions of acellular hemoglobin are not as effective at oxygenation as packed red blood cells because of their high affinity for oxygen. Red blood cells have adapted to release oxygen at an oxygen half saturation pressure of hemoglobin (p-50) of approximately 26.5 mm Hg. This is due to the allosteric effect of 2,3-bisphosphoglycerate (2,3-BPG), which shifts the oxyhemoglobin dissociation curve to the right. In the absence of 2,3-BPG, stroma-free hemoglobin has a p-50 of 12-14 mm Hg.

Unmodified free hemoglobin, when infused rapidly, splits into dimers and is cleared by glomerular filtration and uptake by the reticuloendothelial system.

When free hemoglobin was used initially, it caused a substantial increase in oncotic pressure because of its hyperosmolarity. Unfortunately, the initial attempts at transfusing stroma-free hemoglobin produced renal dysfunction, coagulopathy, and hypertension. Adverse effects were attenuated by various modifications to the hemoglobin molecule to prevent glomerular filtration and to stabilize the molecule to withstand heat and chemical purification during production. Hypertension induced by infusion of these products was out of proportion to the volume infused and has been more difficult to prevent. It is thought to result from hemoglobin binding to nitric oxide, which is a potent vascular endothelial relaxant.

Several approaches have been tried to decrease the avidity with which hemoglobin binds to oxygen. These adaptations include the addition of organic phosphate to serve the function of 2,3-BPG and adenosine triphosphate, cross-linking dimers of hemoglobin tetramers and polymerizing the tetramers to decrease oncotic pressure and prevent glomerular filtration. Hypertension has remained a significant adverse effect of stroma-free hemoglobin.

Diaspirin cross-linked hemoglobin (DCLHb) is the prototype molecule of this category of blood substitutes. It consists of cross-linking between the two alpha chains that lend stability to the molecule. The cross-linking agent is bis (dibromosalicyl) fumarate (DBBF). DCLHb made from outdated human blood has a shelf life of approximately 9 months when frozen and 24 hours when refrigerated. The intravascular half-life is 2-12 hours and is dose dependant. Clinical trials proved that this compound did indeed raise blood oxygen content. However, it also caused intense vasoconstriction resulting in increased systemic pressure, reduced cardiac output, and increased vascular resistance. Hence, no net benefit was derived from this product.

A great deal of work on DCLHb was initially performed by the US Army. They contracted Baxter Healthcare (Deerfield, Ill) to produce DCLHb (HemAssist) on a large scale. However, due to significant adverse effects associated with it, the army discontinued further development of this molecule. Subsequently, Baxter Healthcare halted further development of DCLHb in 1998 after the product failed trials in patients with stroke and trauma. [6]

The first recombinant hemoglobin product, rHb 1.1 (Optro, Somatogen and Eli Lilly) was a genetically engineered variant of human hemoglobin, Hemoglobin Presbyterian, with modifications to decrease its oxygen affinity. The product was produced in Escherichia coli and had an intravascular half-life of 2-19 hours and a shelf life of 18 months when refrigerated. The adverse effect profile was similar to DCLHb and consisted of vasoconstriction, GI distress, fever, chills, and backache. Currently, the production of this compound has been discontinued due to significant adverse effects. Similarly, rHb 2.0, a recombinant hemoglobin produced by Baxter Healthcare had an adverse safety profile that resulted in discontinuation of production.

The development of polymerized hemoglobin was also a result of significant adverse effects caused by intravenous infusion of stroma-free hemoglobin. The cross-linking agent in this case was glutaraldehyde. However, it was discovered that the Hb-glutaraldehyde polymerization reaction was difficult to control and resulted in the formation of products of various molecular sizes.

PolyHeme (Northfield Laboratories Inc., Evanston, Ill) is a first-generation pyridoxylated polymerized hemoglobin made from outdated human blood. It is one of the few products currently being evaluated in a phase III clinical trial that is enrolling patients. It has a half-life of 24 hours, a shelf life longer than 12 months when refrigerated, and a p-50 of 28-30 mm Hg. In a phase II trial in 44 patients with acute trauma, PolyHeme reduced the required the number of allogenic red blood cell transfusions. Results of a trial published in 2002 showed that 75% of patients with red cell hemoglobin levels less than 1 gm% survived traumatic injury after receiving PolyHeme as compared to 16% of historical controls at the same hemoglobin level. [7]

Northfield laboratories was involved in a 720-patient phase III trial in trauma patients in which subjects were randomized to receive either PolyHeme or standard of care at the time of injury. On reaching the hospital, patients in the control arm receive blood as indicated whereas patient in the PolyHeme arm continued to receive PolyHeme for 12 hours, and then received blood, as indicated. Preliminary results indicated that 46 of 349 patients treated with PolyHeme died, whereas 35 of 363 patients in the control group died. The difference in mortality between the 2 groups at 30 days was not significant. Therefore, the trial seemed to show that although PolyHeme was not inferior to standard of care, it was not superior to it either.

Other first-generation polymerized hemoglobin products include HbOC-201 (Hemopure manufactured by Biopure Corporation, Cambridge, Mass) and HemoLink (Hemosol Corporation, Mississauga, Canada).

Hemopure is a polymerized form of bovine hemoglobin with a p-50 of 30 mm Hg that is closer to human hemoglobin than stroma-free hemoglobin. It has an intravascular half-life of 8-23 hours and a shelf life of 36 months at room temperature. Hemopure is approved in South Africa for the treatment of adult surgical patients who are acutely anemic with the intention of eliminating or reducing the need for allogenic red blood cell transfusions. In the United States, phase II trials have been put on hold due to safety issues.

In December 2006, the Blood Products Advisory Committee of the FDA voted against recommending that the US Navy proceed with late-phase clinical trials of Hemopure. The main reason for this was the adverse effect profile of this compound since previous studies had shown that Hemopure could increase the risk of strokes and myocardial infarction. Biopure is currently addressing the FDA's questions regarding safety and efficacy of the product.

A comparative study of HBOC-201 versus allogeneic red blood cell (RBC) transfusions found that up to 7 units of HBOC-201 infused over the course of 6 days resulted in RBC transfusion avoidance in 43% of patients. [8]


Next-Generation Blood Substitutes

Polyethylene glycol (PEG) hemoglobin (Enzon Pharmaceuticals, Bridgewater, NJ) was a conjugated bovine hemoglobin that was evaluated for use in cancer therapy to increase tumor oxygenation and enhance the efficacy of radiation and chemotherapy. However, at present, production of this product has been discontinued.

Hemospan (Sangart Inc., San Diego, Calif), also known as MP4OX, is a PEG-conjugated human hemoglobin currently undergoing clinical trials in the US and Europe. [9] In animal models Hemospan (MP4OX) has been shown to be effective in cases of hemorrhagic shock. [10] Adverse effects associated with the vasoactive properties of first-generation blood substitutes are not seen with Hemospan (MP4OX). At relatively low concentrations, Hemospan (MP4OX) is capable of transporting large amounts of oxygen. Sangart Inc. announced positive results from a phase II study for this product in November 2005.

The trial, conducted in Sweden, involved 90 patients undergoing hip arthroplasty. Patients were randomized to receive either Hemospan (MP4OX) or Ringer acetate (control) before induction of spinal anesthesia. Hemospan (MP4OX) was found to be well tolerated in the study group with no serious adverse effects attributed to the product during the trial period. The percentage of hypotensive episodes in the Hemospan (MP4OX) group was about 45% compared to 87% among controls. Incidence of intra-operative vasopressor use was about 15% in the Hemospan (MP4OX) group compared to 32% among controls. See Sangart Inc: Clinical Trials for more information on recent studies.

Pyridoxylated hemoglobin polyoxyethylene conjugate (PHP) is a conjugated hemoglobin developed by Apex Bioscience that completed a phase III trial in August 2009 in patients with shock associated with systemic inflammatory response syndrome. The study has been designed to evaluate the safety and efficacy of continuous IV infusion of PHP plus conventional vasopressor treatment versus continuous IV infusion of Plasma-lyte A plus conventional vasopressors as a treatment for restoring hemodynamic stability in SIRS patients with shock. See, Identifier: NCT00021502, for more information.

For hemoglobin-based oxygen carriers cross-linked with enzymes, there has been an effort to synthesize compounds that not only perform the function of carrying oxygen, as do the molecules mentioned previously, but also harbor some of the enzyme activity that normal red blood cells possess. Polymerized hemoglobin has been cross-linked with catalase and superoxide dismutase to form a compound that, in animal models can not only carry oxygen but also remove oxygen radicals that are responsible for ischemia reperfusion injuries. [11]

Polyhemoglobin has also been cross-linked with tyrosinase to form a soluble complex that can carry oxygen and decrease the systemic levels of tyrosine. This agent can help increase the efficacy of chemotherapy and radiation therapy in tumor tissue and in a melanoma model has been shown to delay tumor growth without having significant adverse effects. [11]

Efforts have been made to encapsulate hemoglobin within a lipid-membrane to create a compound capable of carrying oxygen while not being associated with significant vasoconstriction. These liposomes appear to be retained in plasma for a significant period. However, they are difficult to produce and can activate the reticuloendothelial system, the complement pathway, and platelets. At present, the only institutions working actively on this product are in Japan.

The ultimate RBC substitute would contain not only hemoglobin but also other contents of the red blood cell also encapsulated in an artificial membrane. However, production of such a product would be extremely challenging. Efforts have been made to use polyactide, a biodegradable polymer that is converted to lactic acid in the body, to create artificial red blood cells. These cells contain hemoglobin along with the RBC enzyme complement including superoxide dismutase, methemoglobin reductase, and catalase. [11]


Adverse Effects Related to Blood Substitutes

Adverse effects associated with hemoglobin-based oxygen carriers include hypertension, abdominal pain, skin rash, diarrhea, jaundice, hemoglobinuria, oliguria, fever, stroke, and laboratory anomalies such as an elevation in lipase levels. Although most of these side effects were transient and clinically asymptomatic, many clinical trials involving these agents have been discontinued or held due to the associated adverse effects.



Despite many years of research, the ideal blood substitute continues to elude researchers. Most of the initial attempts at synthesizing blood substitutes failed because of significant adverse effects. However, continued research has helped us better understand the physiology of red blood cells and the interactions of RBCs with their surrounding environment. This has helped in developing newer products that do not have significant vasoactive properties, as did the first-generation compounds.

Hopefully, as better blood substitutes are developed and enter routine clinical use, the need for blood transfusions in the operative and trauma settings will decrease. Large-scale production of blood substitutes would also help to meet the anticipated increase in demand for blood as the population ages and the blood donor pool diminishes.