Introduction

Trauma is the leading cause of death in the United States in persons younger than 44 years, with traumatic hemorrhage and trauma-induced coagulopathy contributing to mortality in the first 24 h [1]. Ongoing research in trauma resuscitation has revealed that goal-directed resuscitation with a protocolized balanced blood product transfusion strategy results in improved early survival and less deaths due to exsanguination [2–5].

Regardless of transfusion strategy , the need for red blood cells (RBCs) and their oxygen-carrying capacity is critical to survival. An American Association of Blood Banks (AABB) Blood Survey Report from 2013 showed a 12% decrease in whole blood and RBC collection when compared to 2011, even though the utilization of whole blood and RBCs decreased by 5% [6, 7]. Even with improvement in hemorrhage control strategies in trauma, other means of storing RBCs need to be explored due to the limited shelf life of liquid-packed RBCs and the development of storage lesions over time. One example of this is frozen RBCs (fRBCs).

History

The first published study of successful cryopreservation of RBCs was by Dr. Audrey Smith in 1950, who discovered that human blood diluted with equal parts of 30% glycerol-saline and frozen to −80 °C did not undergo the usual hemolysis caused by freezing and thawing [8]. She observed that glycerol, a sugar alcohol that is permeable to cell membranes, acted as a cryoprotective agent (CPA) during the freeze-thaw process. It was thought that the principle protective action was their ability to prevent the excessive concentration of electrolytes and other substances that otherwise occurs with freezing [9]. Although other CPAs such as dimethyl sulfoxide and hydroxyethyl starch have been used to prevent cryoinjury, glycerol is the more commonly studied and utilized CPA [9, 10]. Mollison and Sloviter published the first successful transfusion of thawed fRBCs and found a post-thawing preservation of 80–90% of the original cellular volume [11].

This resulted in enthusiasm for the process, as it was seen that cryopreservation could allow the long-term storage and stockpiling of large quantities of blood for prolonged periods, thus alleviating the issue of seasonal or unexpected catastrophic shortages and establishing greater availability of rare blood types.

Cryopreservation

Glycerol is a very effective cryoprotectant for cells in which they can diffuse into fairly quickly, such as human erythrocytes. The uptake of glycerol by human erythrocytes are both an active and passive process [12–15]. Once intracellular, glycerol will form firm hydrogen bonds with intracellular water, inhibiting it from turning into ice. In turn, this will suppress the rise in sodium chloride concentration to prevent extreme hypertonicity [16, 17].

Currently, two methods are commonly used to cryopreserve RBCs: the high glycerol concentration (HGC) method with 40% weight/volume (W/V) or the low glycerol concentration (LGC) method with 20% W/V.

The HGC method allows for an initial slow freezing rate (~1–3 °C/min) and storage in −80 °C freezers but adds to the thawing time. The LGC method requires rapid cooling (>100 °C/min) to −140 °C in liquid nitrogen but reduces the thawing time [18]. Advantages of the HGC method include storage up to 37 years and the relative ease of transportation in freezers [19]. One advantage of the LGC method is that it results in less hemolysis upon thawing than the HGC method [20]. Prior to transfusion, frozen RBCs require deglycerolization due to increased hemolysis caused by a more rapid endosmosis of water than exosmosis of glycerol [21].

The deglycerolization phase involves washing the glycerolized RBCs with hypertonic sodium chloride solution. The diluted RBCs are then placed in a centrifuge and the supernatant fluid is removed. The RBCs are washed twice with isotonic sodium chloride solution. Historically, the entire process required 1 l of wash solution and took almost 3 h [22].

Multiple methods were developed to speed up the process, and in 1963, Dr. Charles Huggins invented the cytoglomerator, which uses a dilution/agglomeration technique. It first involves removal of glycerol by adding non-electrolyte solutions contacting glucose and fructose to glycerolized RBCs while stirring. This decreases the ionic strength of the environment, causing RBCs to clump together, or agglomerate, once stirring stops. After decantation of the supernatant , isotonic saline is added to the agglomerated RBCs for disaggregation [23]. The average time was 50 min with 6.8 l of washing fluids used. In 1967, Haemonetics developed a blood processor called the M115 cell processor with a polycarbonate bowl with an attached shaker for mixing of the washing solutions. This process took 35 min and only used 2 l of washing fluid [24]. Both of these systems, however, are open systems, and the RBCs must be administered within 24 h of post-thaw and washing [25–27]. Haemonetics later developed a closed-system automated cell processor in 1998; the ACP 215 is able to use thawed pre-wash blood stored at 1–6 °C for up to 14 days and produce acceptable-quality post-wash RBCs that can be stored up to 7–14 days [28, 29]. This device utilizes sterile connecting devices, in-line 0.22 micron bacterial filters for solution delivery, and a disposable polycarbonate bowl to deglycerolize RBCs. The post-wash RBCs are then stored in standard blood storage solutions, either additive solution (AS)-3 (saline, adenine, glucose, citrate, and phosphate) or saline-adenine-glucose-mannitol (SAGM). The entire freeze-thaw-wash process takes about 2 h and is currently the most prevalent method of processing fRBCs [30].

Benefits

Clinical Use

Throughout history , wars have played an important role in the development and advancement of medical care. The first documented successful blood transfusion took place in the US Civil War, when, in 1864, Dr. Fryer transfused 16 ounces to a soldier who underwent an above-knee amputation [58, 59]. By the first World War, knowledge of citrate as an anticoagulant was available, but not utilized for quite some time [60]. Rous and Turner in the first year of World War I created a mixture of 5.4% glucose and 3.8% sodium citrate to protect RBCs from hemolysis for 4 weeks [61]. In 1917, a military physician named Oswald Robertson designed an icebox with glass containers filled with whole blood mixed with the Rous-Turner solution, thus becoming the world’s first blood banker [62]. By the World War II, whole blood was able to be fractionated into plasma and was used to treat burn victims in Pearl Harbor after the Japanese attack [63]. During the early years of World War II, the general belief was that plasma was enough to compensate for hemorrhagic shock [64]. As the war raged on, focus was turned back to whole blood, which carried through the Korean War. Noticeably, the mortality rate of wounded soldiers after reaching a hospital decreased from 10% in World War I to 2.6% in the Korean War [65, 66].

In the time between the Korean and Vietnam Wars, the US Navy commissioned the Blood Research Laboratory in 1956, later renamed the Naval Blood Research Laboratory (NBRL) in 1965, which was tasked with developing long-term preservation of RBCs, especially for use on naval ships. That same year, the NBRL established the first frozen blood bank at Chelsea Naval Hospital in Massachusetts, adopting the HGM [53].

The first clinical trial utilizing fRBCs was performed by Haynes et al. at Chelsea Naval Hospital, where more than 1000 units of fRBCS stored up to 44 months were transfused to more than 355 patients. In addition to equivalent clinical results as compared to liquid RBCs, they found a decreased rate of transfusion-related hepatitis and adverse febrile reactions [21].

In 1966, a frozen blood bank at the Navy Station Hospital in Da Nang, Republic of South Vietnam, was established. Their objective was to receive a limited supply of frozen Group O, Rh-negative blood from the United States for use in selected casualties, using a Huggins cytoglomerator for the processing of fRBCs. A total of 307 units of fRBCs were transfused. In vitro studies showed a 27% red cell loss, with a final volume of 210 mL with a hematocrit of 87%. In vivo studies showed an immediate posttransfusion mean hemoglobin increase of 3.68 mg/100 mL as compared to 0.72 mg/100 mL for liquid RBCs. Measured serum creatinine and bilirubin level were acceptable. The authors concluded that fRBCs during wartime are an alternative to the walking donor system, which is fraught with possible logistical complications such as transportation, communication, personnel, blood-borne pathogen transmission, and donor safety [67]. During the 1991 Gulf War, approximately 7000 frozen units were available on two US Naval hospital ships, but none were used. The Joint Trauma System Performance Improvement Branch analyzed data from the Department of Defense Trauma Registry and Massive Transfusion Database in Afghanistan and found 63 patients between January 2010 and September 2011 who received massive transfusions that required the use of fRBCs. When compared to a control population of 525 patients who did not receive fRBCs during their massive transfusions, there were no significant differences in complications including transfusion reactions, coagulopathy, renal failure, deep vein thrombosis , or respiratory failure [68]. Currently, the US Naval Hospital Ship Mercy carries a stock of 2850 fRBC units at all times, and in 5 months, over 200 deglycerolized units were transfused to patients in various settings [69]. The US Military Joint Trauma System Clinical Practice Guideline regarding fRBCs supports the use of fRBCs, with the primary indication as a supplement to liquid RBCs during periods of increased transfusion requirements in order to decrease hemorrhagic morbidity and mortality in casualties [68].

Internationally, the Laboratory of the Dutch Red Cross Blood Service began freezing RBCs in the early 1960s. Using the LGM, they initially began freezing phenotypically uncommon or rare RBCs. They also froze O-positive and O-negative units to increase their inventory in anticipation of the possible shortages in the summer and winter months [64]. During the first Gulf War in 1991, the Netherlands Military Blood Bank realized that shipment of liquid RBCs was cost-ineffective and would not guarantee availability at all times, unlike fRBCs. To test this, they froze and sent 1360 units of fRBCs and frozen platelets to Iraq in 2005. They learned that a military hospital blood bank facility can be deployed without regular shipments of liquid blood products and can meet the needs of a surgical team by thawing and washing a certain amount weekly, creating a hybrid liquid-frozen blood bank, although it is not known how busy the surgical team was during that time [69].

Civilian usage of fRBCs began with the invention of the cytoglomerator by Dr. Huggins, making Massachusetts General Hospital the first civilian center to use fRBCs on a large scale [13]. Between 1971 and 1972, his blood bank froze over 15,000 units, and 14,406 units were transfused in the subsequent years, with Huggins claiming that no cases of hepatitis occurred [55]. Gerald Moss, a student of Dr. Huggins, ran the Cook County Blood Bank in Chicago using almost exclusively fRBCs. By 1975, they were freezing and thawing more than 10,000 units of RBC a year [55].

Drawbacks/Future Directions

One of the biggest drawbacks of fRBCs is the preparation. Whereas liquid RBCs are ready to use after removal from a 1–6 °C cooler, fRBCS need to be deglycerolized and prepared, which takes at least 50 min even with the new ACP 215 [70]. This detracts from the utility of fRBCS in emergent scenarios. This problem can be mitigated by maintaining a portion of the RBCs thawed for immediate use. A second drawback is cost, which is about three times the cost of liquid preservation, with the majority due to instruments used in processing, disposables, and solutions [71]. This ties into the third drawback, which is the in vivo recovery rate of 75% at 24 h after transfusion of fRBCs [72]. Although this is at the threshold of the FDA’s current standard of transfusion recovery rate, the extra cost associated without the benefit of increased recovery can cause reservations in widespread adaptation of this technique of blood storage. However , the prospective randomized multicenter trial revealed that hematocrits performed 12 h after transfusion were identical in the liquid and fRBC groups. This suggests that the senescent cells that are removed in the deglycerolization process are similarly removed by the body after transfusion resulting in equivalent loss in vivo. A recent paper by Chang et al. found that after deglycerolization, cryopreserved blood developed storage lesions at a faster rate when compared with liquid never-frozen pRBCs, thus negating one of the benefits of fRBCs if they are not transfused within 14 days after deglycerolization [73].

The processing and usage of fRBCs peaked in 1978/79, where the majority of their use was in dialysis centers for potential cadaveric renal transplant recipients. It was originally thought that sensitization to donor histocompatibility could affect graft survival and fRBCs had a lower amount of leukocytes than liquid RBCs to provoke this immune response. However, Opelz and Terasaki in 1978 published a study of 1360 cadaver donor transplants comparing 4-year graft survival in never-transfused patients and those who had over 20 pretransplant transfusions. The results showed that graft survival was 30% in those who never received transfusions as compared to over 65% in those who received transfusions, suggesting that prior exposure to donor histocompatibility, antigens enhanced graft survival [74]. As a result of this study, utilization of fRBCs began their steady decline. The demand for fRBCs also declined after the discovery of the human immunodeficiency virus (HIV) in 1983. By 1992, 9621 cases of acquired immunodeficiency syndrome attributed to blood transfusions were identified, and it was not until 1985 when a screening process for HIV was created. Since the cryopreserved blood was not tested for HIV, fear of transmission has curtailed their use [75].

Regarding rare blood use, with the advent of longer liquid RBC storage, the Dutch Red Cross now averages transfusing 30 units per year. Even the American Red Cross Rare Donor Registry decreased their inventory size from 18,000 to 9800 between 1981 and 1990 [55].

In the wake of the terrorist attack on the World Trade Center in New York City, New York, and the Pentagon in Arlington, Virginia, a total of 2800 people were killed and 4000 were injured. Despite the New York Blood Center having between 18,000 and 22,000 units available and the Washington DC metropolitan area having 12,000 units available during that time, only a mere 258 units of RBCs were transfused [63]. A few hours after the attacks, the Red Cross ceased distribution of blood from its regional centers, forcing local hospitals to look elsewhere for blood. Around the same time, the National Institutes of Health opened up their blood bank, and ceased collections within 24 h due to adequate filling of its inventory. In addition, blood donors were lining up outside of local donation centers to contribute to the blood supply inventory. With no evidence of blood shortage, the Red Cross did not cease donations for weeks, with their justification being the creation of a “National Blood Reserve” of a 7- to 10-day supply of liquid RBCs followed by fRBCs, resulting in collection of 287,000 extra units of blood [53]. The Red Cross purchased 70 Haemonetics ACP 215 machines, only to find out that many of the extra units of blood were preserved in additive solutions that were not licensed to be used with the machine. Only 9500 of the planned 100,000 units were frozen, and many of the extra units were sold or given away as they neared their expiration date. In total, almost 600,000 extra units of blood were collected with over 300,000 units destroyed. Due to the high wastage of donated blood, the state of New York experienced a decrease in the number of donors due to their perception that their donated blood was wasted [53].