4: Blood Technologies

4 Blood Technologies; 82 83 86 86

...... 87 ...... 88

...... 89 ...... 90

Page ...... 81 ...... 84

...... ‘,. . 90 4 Blood Technologies;

PART 1: TECHNOLOGIES FOR BLOOD COLLECTION AND PROCESSING

Introduction ing potentially fatal hemolytic transfusion reactions. Modern blood bank technology began less than 100 years ago, when several technological ob- In 1945, a new test was described (Coombs) stacles were overcome. First, it was necessary to which made possible the recognition of previously arrest the blood clotting mechanisms which nor- undetectable antigens and led to the identification mally convert liquid blood into a gel in a matter of more than 400 new red blood cell types over of minutes. Second, physicians needed tools by the subsequent 30 years (284). However, prob- which blood could be collected from one individ- ably fewer than 30 of these types are of clinical ual and infused into another with minimum risk concern in routine transfusion practice (220). to donor and recipient. Third, there was a lack As the demand for blood increased, the larger of understanding of the immune system and the blood banks began to consider adaptations of clin- potential for incompatibility between donor red ical chemistry analyzers which could perform blood cells and red blood cell antibodies produced multiple tests on a single blood sample to remove by recipients. some of the tedium and manual labor from mass The coagulation problem was solved by the ad- blood-typing. These systems have evolved into dition of citrate into freshly collected blood (344); computerized testing devices which deliver sam- citrate compounds are still the basis for stored ples, interpret results, and flag any spurious find- blood anticoagulants today. Preservatives such ings for further investigation. as glucose were soon added to extend the shelf In the beginning, donor blood collection and life of stored red cells from several days to sev- testing were conducted in the same setting as eral weeks. blood transfusion; each hospital would screen Devices used in collecting, storing, and ad- donors on the basis of specific patient needs, and ministering blood have ranged from quill and tub- donors were often a patient’s family and friends. ing contraptions to syringes (used currently for World War II resulted in the creation of donor newborn exchange transfusions), to reusable centers across the United States to collect large vacuum glass bottles, still the standard in parts amounts of plasma for use in treating combat of the world today. In the United States, glass bot- casualties, and intensive research into improving tles have been replaced by flexible plastic con- the storage life and function of red blood cells. tainers with attached satellite bags to hold sepa- When the war ended, community blood centers rate blood components. began to appear under the auspices of American Red Cross chapters, community service clubs, and Landsteiner’s description of ABO blood groups local medical societies. These centers acted as stag- in the early 1900s and other researchers’ elucida- ing areas from which mobile operations could be tion of antigen/antibody relationships led to sim- sent for blood collections, after which the blood ple laboratory tests for compatibility between was returned to the center for processing and donors and recipients, but 15 percent of transfu- subsequent distribution to hospitals. sion recipients remained at risk until the Rhesus or Rh blood types were discovered (331,342). Blood collection and processing in a commu- From that time to the present, determination of nity blood center can function as an assembly- the donor and recipient ABO and Rh blood types line operation because of the need for repeating has been the most important procedure in prevent- the same battery of tests on a number of donor

79 80 ● Blood Policy and Technology samples, ranging from 200 per day in medium- Figure 9 shows an example of this labeling for- size centers to over 1,000 per day in large centers. mat. The labeling scheme has been endorsed by On the other hand, transfusion services repeat the FDA (182) and appears on 60 to 70 percent of entire workflow upon receipt of physician orders products distributed by U.S. blood banks. How- which, except in cases of standing orders for elec- ever, not all blood centers who label their prod- tive surgeries, are unpredictable and often urgent. ucts with bar codes have the equipment to make Therefore, batch or assembly-line processing is use of the machine-readable information, and few not as feasible, except for verification of donors’ transfusion services have the capability. Never- blood types and elective surgery workups. theless, eye-readable uniformity and simplicity are universally appreciated aspects of the commonality Blood is inherently variable, with each collected label, given a history of multiple color-coded la- unit representing a unique “lot. ” Because it is im- bel schemes and superfluous information on possible to test each unit for effectiveness before earlier blood product label formats. The uniform transfusion, the use of scientifically proven pro- label has allowed developers of automated testing tocols and regular checks on representative prod- systems and management information systems to ucts are important. design equipment and software around a single The use of computers for donor resources and machine-readable format for use in the United inventory management at the community blood States and several countries in Europe and the Far center, and standard policies for surgical blood East. needs at the transfusion service have improved As yet untapped, but applicable, technologies blood utilization, decreased outdating, reduced include: recruiting by computerized telecommuni- charges to the patient, and improved management cations linkages, donor and physician education of donor resources. Computer management of via cable health networks, and robotics labora- discrete functions has been accomplished at many tories to perform certain repetitive tasks (209). U.S. blood centers, but often lack coordination The feasibility of these alternatives is unproven, among donor recruitment, collection, laboratory, but studies are under way. inventory, and financial management departments. One comprehensive information system has been designed as a joint venture between two Blood Collection community blood centers (Blood Center of South- Technologies involved with collection of blood eastern Wisconsin and the South Florida Blood from donors address three basic functions: 1) ac- Service) and Arthur Andersen & Co. The goals curate identification of donors, 2) determination of all such systems are to minimize manual docu- of donor suitability, and 3) collection of blood. mentation and subjective analysis, and provide useful statistics for operational management and Donor identification is accomplished in most planning. blood banks by the use of a cardboard or plastic A number of automated data management sys- card embossed with name, address, permanent identification number, and blood type. Most tems throughout the world utilize uniform bar blood banks also have rosters of “special” donors coding technology for the identification of donor against which the donor’s identification is com- units and samples for production and inventory pared. “Special” donors include individuals whose management. This uniformity, which has the blood is extraordinarily rare (for special needs), advantage of standardized machine-readable as well as simplified eye-readable features, is the re- as well as those whose blood is unacceptable for transfusion due to laboratory findings on previ- sult of a special task force of the American Blood ous donations. Commission, whose efforts were funded by the National Heart, Lung, and Blood Institute and en- Donor suitability is determined by an assess- dorsed by the International Society for Blood ment of current health status (temperature, blood Transfusion. pressure, screening test for anemia, etc. ) and an Ch. 4—Blood Technologies . 81

Figure 9.— Uniform Labeling Format for Blood Products

Form 7203 (1/81) EXPIRES F 01255 I I II II I II

\\ \ \ \ 9 Unit Number EXPIRES

I \ ------

1 hlrmd

LOT NO CODE

SOURCE: E Rosslter, “Technologies for Blood Collection, ” 1984

interview covering past and present illnesses, im- (598). There are also concerns among blood bank munizations, hospital procedures, or other con- physicians about the adequacy of current donor ditions (including travel outside the United States, criteria when used for frequent, long-term donors lifestyle, and exposure to other individuals) which (562). could make donating blood unsafe for the donor or transfusion unsafe for the recipient. While some Pre-donation cell counts or hemoglobin meas- of the donor acceptability criteria are straightfor- urements generally assure that donating will not ward, most are subject to varying degrees of inter- be unduly risky for the donor and that an ade- pretation at the time of screening, In the absence quate number or volume of the desired compo- of factual information to support all the decisions nent can be collected. Findings of diminished that affect donor suitability, intuition is often used platelet effectiveness in individuals taking aspirin 82 . Blood Policy and Technology has led to deferral of such donors in platelet- donor who subsequently has a traumatic veni- apheresis programs (565). (See ch. 2, pt. 3, for puncture in the collecting process, only red blood a discussion of the risks to donors. ) cells and plasma may be prepared. The empty, extra satellite container, which represents a cost One measure of the safety of the collection of $4.52 (based on differences in the price of dou- process is the frequency of moderate to severe re- ble and triple packs with a 5-day platelet storage actions in which donors faint or lose conscious- 0.3 container in 1984) is lost. The extent of this prob- ness during or after donation. A rate of per- lem has been estimated by some to affect 7 to 10 cent or less is regarded as acceptable (44). percent of collected units (472). Collection of whole blood is accomplished by Successful adaptation of sterile docking devices the use of pre-sterilized collection sets which con- to routine component production operations sist of a needle, tubing and container pre-filled could eliminate the need for multiple bag con- with anticoagulant/preservative solution. The figurations; all blood could be collected into single main collection bag may have one or more smaller, packs (at $3.18 each in 1984) and those units satellite bags attached to it by tubing. After needed for component production could be coupled venipuncture, blood flows by gravity into the with separate transfer packs (at $2.25 each in main container, mixes with the anticoagulant/ 1984) via sterile docking technology (projected preservative and remains there until an internal cost of $0.50 per component) (3)0 Savings of $0.50 closure is opened to the satellite containers for to $1.00 per component would be possible. Fur- component preparation. Table 26 lists the usual ther savings could be realized on less expensive types of components that can be made from each storage containers for products designated for fur- container configuration. ther manufacturing (e.g., plasma fractionation). Underutilization of the multiple bag set can re- During the last 10 years, a new technology sult when changes in blood orders or incidents originating in the United States called “apheresis” during the collection process necessitate prepara- has become popular for the collection of large tion of fewer or different components than the amounts of a specific component from a single ones for which the configuration was designed. donor. (The therapeutic use of this technology For example, a bad (traumatic) venipuncture was reviewed in a July 1983 OTA case study, The results in decreased platelet and cryoprecipitate Safety, Efficacy, and Cost Effectiveness of Ther- yields; therefore, if a triple pack is assigned to a apeutic Apheresis. ) Plasmapheresis can be accomplished by man- Table 26.—Standard Blood Collection ual techniques, which involve collection con- Container Systems tainers and techniques similar to those for whole blood collection. After centrifugation and separa- a b Bag configuration Components prepared Cost per pack tion of the plasma, the red cells, white cells, and Single pack Whole blood $3.18 Double pack Red blood cells $6,52 platelets are reinfused into the donor. The entire Plasma process is then repeated, resulting in two final Triple pack Red blood cells $11 .04’ plasma products and involving a second reinfu- Cryoprecipitate or platelets Plasma sion of the donor’s residual cells. The procedure Ouadruple pack Red blood cells $14.55C takes approximately 90 minutes. Plasmapheresis Platelets is a relatively uncommon procedure in commu- Cryoprecipitate Plasma nity blood centers and transfusion services; for Quintuple pack Pediatric doses of red $17.78 example, less than 0.1 percent of the total plasma blood cells and plasma collections in the Red Cross in 1983 were by aproduct$ listed do not include modified whole blood end resulting byproducts, or components prepared by breaking original (sterile) seals. plasmapheresis (48). bList prices for CPDA.I systems from Fenwal Division, Baxter-Travenoi, Deer- fieid, iL. Plasma may be collected as a byproduct of cprice iS for 5-day piatelet product; 3-day platelet or Cryoprecipitate container price is reduced by $1.44. automated cytapheresis (cells) procedures for SOURCE: E. Rossiter, “Technologies for Biood Collection,” 1984. platelets or white blood cells. Current automated Ch. 4—Blood Technologies Ž 83 technologies are not cost effective compared to Current plateletapheresis technologies are sum- manual methods for routine plasmapheresis (2), marized in table 27. The costs of all systems are but are useful nonetheless in collecting large comparable (1984 prices); hardware is approx- amounts of antibody-rich plasma from specially imately $30,000 and collection container systems immunized donors for reagent use or further man- $65-85, except for the 5-day container system. ufacturing into injectable products, such as hyper- Platelets collected by apheresis (single-donor immune anti-Rh(D) for prevention of hemolytic platelets) have been proven to be as safe and ef- disease in the newborn. fective as platelets prepared from whole blood Membrane filtration technology under devel- (random-donor platelets) (301, 501), but there is opment may provide a cost-effective alternative controversy surrounding the extent to which to automated plasmapheresis. One developer pro- plateletapheresis products should be used in lieu jects a 20- to 35-minute procedure in which oper- of platelets from whole blood collections. Critics ator intervention is minimized by microprocessor of “routine” plateletapheresis cite economic rea- control. sons (pheresis platelets costs ranged from $2OO to $350 per dose; manually collected platelets, from All automated apheresis devices work on the $125 to $200 in 1984), the higher prevalence of principle of centrifuging whole blood to enable donor complications, and lack of clinical data to the separation of components, transfer of the support widespread and indiscriminate use of desired component to a storage container, and plateletapheresis technology (483). Nevertheless, reinfusion of the remaining components. The first few blood bankers doubt that apheresis technol- apheresis devices were largely mechanical proc- ogies could someday replace manual platelet col- essors, but later machines are microprocessor- lections. controlled with multiple modes for different component schemes. Operator experience and tech- The 24-hour shelf life of apheresis platelets nique have been shown to substantially affect (compared to the 3- or 5-day shelf life of platelets equipment performance. separated from whole blood collections, because the collection system in apheresis is opened in the Collection techniques for plateletapheresis are separation and reinfusion of the donor’s cells) has fairly new. Effectiveness and safety data show been a major problem in inventory management some variation in end-products, mainly in platelet with these products. But one company, Fenwal, yield and undesirable contamination by extra- is now marketing a closed system for the produc- neous white blood cells and red blood cells (198). tion of apheresis platelets with a 5-day shelf life.

Table 27.—Plateletapheresis Technologiesa

Product Manufactured Year dating Technology model developed period Comments lntermittent- flow centrifuge ...... Haemonetics 30 1973 24 hrs. First system approved by FDA. Requires extra step to remove extraneous cell contamination. Haemonetics V-50 1983 24 hrs. Produces platelet-rich plasma. Continuous-flow centrifuge ...... Fenwal CS 3000 1979 24 hrs. Fewer extraneous red blood ceils. Lower extracorporeal volume. 1983 5 days Sterile docking device technology and a different type of storage container add $40 to collection container costs. Dual stage continuous- flow centrifuge ...... IBM 2997 1977 24 hrs. Manufacturer plans no further development of product Iine.b 84 ● Blood Policy and Technology

Leukapheresis (for collection of white blood a period of about 24 hours. The most dense cells, cells) can also be performed (independent of or red blood cells, settle out to fill the bottom half concomitant with platelet collection) with the of a container. Less dense white cells settle in a same technology; costs and techniques are simi- thin, visible layer on top of the red cell mass. lar to those for plateletapheresis. Although the Platelets are distributed among the white blood safety, effectiveness and efficiency of leukaphere- cells, with some platelets remaining suspended in sis technologies, and leukocytes as a transfusion the plasma layer. product have been under study for over a dec- This separation process can be accomplished ade, use of the technology remains controversial in a matter of minutes using high-speed centri- (90,405,468). Clinical data to support the use of fuges. The flexibility of the plastic collection con- white blood cell transfusions are extremely diffi- tainer allows removal of the top layer into a sterile cult to obtain because of the morbidity of the re- attached satellite bag. cipients; white blood cells have generally been used for individuals with overwhelming infections The original collection container configuration unresponsive to antibiotics. dictates the number and type of components which can be prepared from whole blood (see Blood Component Preparation table 26). Through a scheme of primary and secondary processing steps, a variety of end-products The proportion of whole blood donations that can be produced, either for direct transfusion or are separated into components varies among for further manufacturing into other products (fig. blood centers from around 50 to more than 99 10). Secondary processing adds to product cost percent, reflecting local medical preferences and and is generally limited to special applications. practices. The minimum yield from one unit of On average, 80 percent of whole blood collec- whole blood is one unit of red cells and one unit tions are processed into two or more components, of plasma. Additional components which are most commonly red blood cells and frozen or liq- commonly separated are platelets and cryopre- uid plasma. Platelets are also prepared from over cipitate. 30 percent of collections and have become the The plasma and cellular components of col- “driving force” behind scheduling and planning lected whole blood will separate by gravity over of component production. Platelet shelf life is the

Figure 10.—Derivation of Blood Components

Further manufacturing Red blood cells Washed None

None

Rejuvenated None

Leukocytes removed Byproduct white blood cells Whole

Platelets None (liquid)

Plasma Fractionating for plasma ( (liquid) protein products Fresh frozen factor concentrate

Frozen None Ch. 4—Blood Technologies ● 85 shortest of all whole blood components, except added after other components are separated. It for granulocytes; the product must be prepared is not known whether the additive system ap- within several hours of blood collection to pre- proach (still in the introductory phase in 1984) serve platelet function and survival. Because over will allow appreciable further reduction in red 60 percent of blood collection operations are off- blood cell outdating, but there are other advan- site periodic shuttles back to the blood center are tages to additive systems recognized from years necessary throughout the day. Special collection of use in Sweden and Australia (267,345). They operations are scheduled over weekends and include increased plasma recovery (the additive holidays to avoid shortages. solution replaces the buffering function of plasma during storage), better red cell survival in the re- The development in 1981 of new plastics for cipient, and transfusion viscosity similar to that storing platelets increased shelf life from 3 to 5 of whole blood, days and has alleviated some of the production constraints. Initial studies show improved effi- Potential disadvantages include the additional ciency by an increase in product availability and cost of the collection container (approximately 13 decrease in outdating (414b). Longer shelf life has percent more), extra manipulation during com- also allowed a reduction in the number of shut- ponent preparation, and the potential need to tles and weekend operations, further improving remove excess solution in patients subject to cir- component production efficiency. Notable in- culatory overload (392). Nevertheless, the ad- creases in platelet demand and utilization were ex- ditive solution approach represents a technology perienced in 1983 in the Red Cross, where 75 per- designed to make possible specific preservative cent of the platelet production was in the 5-day solutions tailor-made for each blood component plastic containers (48,351). (315). The extended shelf life of platelets is possible Future component storage research is aimed at because of two different new plastics which allow further increases in platelet shelf life and the use better exchange of oxygen and carbon dioxide of ascorbate (vitamin C) to improve maintenance through the walls of the platelet storage container, of certain red blood cell functions. thereby preventing the buildup of acid (as meta- Recent advances and commercial successes in bolic processes continue in the stored product) extending the shelf life of platelets are likely to that is toxic to platelets. The new plastics have spawn further improvements in dating periods. an additional advantage over previous formula- Use of additive solution systems, if widely actions in that they lack DEHP, a common plasti- cepted in the blood banking community, will cizer used in construction, home furnishings, facilitate research into component-specific pre- clothing, etc., which is known to leach into blood servatives. There also is interest in automating during storage and therefore has caused concern component production by the adaptation of over potential toxicity (562). robotics systems used in the U.S. automobile in- The storage life of liquid-stored red blood cells dustry. Such a system has been designed, and its and whole blood was extended from 21 to 35 days feasibility and potential impact are under study by the addition of adenine to the basic preserva- (209). tive solution in 1978. The improved efficiencies from this change have resulted in 50 percent re- Testing and Labeling of Blood duction in outdating, now at approximately 5 per- From Donors cent (48). Further extension of red blood cell shelf life to 49 days is now possible. Red blood cells With every donation, blood samples are tested are first suspended in CPD anticoagulant (one of for selected red blood cell antigens and antibodies the standard preservatives), then an additive nu- and certain transmissible diseases (table 28). Test trient solution (AS-1, or ADSOLTM, from the methods, their scientific bases, and potential clin- Fenwal Division of Travenol Laboratories) is ical relevance vary greatly. 86 ● Blood Policy and Technology

Table 28.—Testing of Blood Donor Samples

Test Test detects Reason performed Clinical significance

SOURCE’ U.S. Department of Health and Human Services, FDA Panel, 1979

Red blood cell tests determine the donor’s ABO Table 29 demonstrates the serological basis for and Rh type. Mild incompatibility can result in determining the ABO group of an individual. Be- an ineffective transfusion due to reduced red blood cause A and B antibodies are naturally produced cell survival. Major ABO group incompatibility by individuals who lack one or both of the an- can result in fatal hemolytic transfusion reactions. tigens, it is possible to determine the ABO group Because of this danger, the donor ABO and Rh by testing for either antigens or antibodies. Com- type are confirmed by the transfusion service, and mon practice is to perform both tests and check ABO compatibility may be verified by cross- for agreement. hatching. Rh type is determined by testing for D antigen Standard tests for red cell antigens and anti- on red cells; there is no naturally occurring an- bodies are based on the principle of hemagglutination; i.e., the clumping of red blood cells that can occur as a result of specific antigen-antibody re- Table 29.—Determination of ABO Blood Group actions. When a red cell antibody comes into con- Antigens tact with its target antigen on the surface of red on donor’s Antibodies blood cells, it combines with the antigen, form- red blood cells in donor plasma ing antibody “bridges” between cells, which lead Donor blood group A B A B to visible clumps. Some antibodies (e.g., A and 0 ...... No No Yes Yes A ...... Yes No No Yes B) are very effective in bridge formation; others B No Yes Yes No need antiglobulin (Coombs) reagent to form vis- AB : : : : : : : : : : : : : : : : : Yes Yes No No ible antibody bridges. SOURCE E. Rossiter, “Technologies for Blood Collection, ” 1984. Ch. 4—Blood Technologies ● 87

tibody to D as with A and B antigens. Some blood predominant manual test technique is the tube banks perform two different tests for D antigen method, although use of microplates (actually a to provide an internal check. Blood banks also miniaturization of the tube technique) has been perform additional, more sensitive antiglobulin increasing since the late 1970s. In 1984, within the (Coombs) testing on samples giving negative Red Cross alone, 14 centers used manual micro- results for D antigen. This technique will detect plate techniques as their primary method, and weak or suppressed D antigen, but its clinical sig- another 7 used microplates as a backup to auto- nificance is unclear. mated methods. ABO and Rh types are genetically determined The Technicon AutoAnalyzer, introduced in and do not change. For this reason, some Euro- 1967, was the first attempt to automate routine pean blood centers do not repeat ABO and Rh donor testing. It used continuous flow technol- determinations on donors who have been previ- ogy already proven in clinical chemistry analy- ously tested (sometimes multiple testing will be sis, It automatically samples and performs ABO, conducted before subsequent testing is discon- Rh, and STS (serological test for syphilis) tests, tinued). This practice would significantly reduce but relies on the operator to identify samples, read the testing workload in the United States, because reactions, and interpret and record results. Never- approximately 80 percent of blood donors have theless, the AutoAnalyzer was widely used (still donated before (48). However, Federal regulations about 200,000 worldwide), easy to repair and currently require ABO and Rh testing on every maintain, and, at about $30,000, within the price donation. On the other hand, some blood banks range of a number of medium to large blood in Canada and Europe do more elaborate Rh typ- centers; the largest of them often had two. The ing than is done in the United States, testing for AutoAnalyzer is no longer produced but is still C and/or E antigens (of the same family as D). used in a number of centers. This practice has been discontinued in most U.S. Table 31 lists the fully automated testing tech- blood banks with no increased risk to recipients nologies used in the United States in 1984. All (562). identify samples via bar-coded label, sample auto- ABO and Rh testing can be performed by man- matically, perform tests, read results by a photo- ual or automated methods. Table 30 summarizes meter, and interpret results via dedicated com- the different manual techniques used for detect- puter. All systems also have some capability to ing red cell antigens and antibodies. All manual perform antibody screening and STS, but are not techniques involve mixing of sera-containing an- generally used for these purposes in the United tibodies and red blood cell suspensions, and visual States. Costs per donor sample range from $0.83 examination for clumping. Tube and microplate to $1.23 v. $0.53 for the AutoAnalyzer (208). techniques allow centrifugation of the test mix- These systems are much more difficult to maintain ture for easier reading of the test reaction. The and repair. Overall costs of the fully automated

Table 30.—Comparison of Manual Technologies for Red Cell Serological Testing of Blood Donors

Type Advantages Disadvantages Slide/tile testing ...... Simple, fast, inexpensive. Less sensitive. Not adaptable to automation. Difficult to use in batch testing. Coombs testing impossible without sup- plemental use of tubes. Tube testing ...... Allows the use of more sensitive Requires labeling and manipulation of techniques than slide. Adaptable to batch tubes. Uses larger amount of reagent and testing up to approximately 200 samples. sample than slide or microplate. Microplate...... More sensitive than tube or slide. Uses More dexterity required for sampling and less sample, reagent and space. Easily plate handling. May require new reaction adaptable to large batch testing. interpretation techniques. SOURCE E Rossiter, “Technologies for Blood Collect ion,” 1984 88 ● Blood Policy and Technology

Table 31.—A utomated ABO/Rh Test Technologies in the United States, 1983

Number Price System name Manufacturer Principle of centers (x1,000) AutoGrouper ...... Technicon a Continuous flow analysis 11 $184 MiniGroupamatic ...... Kontron b Discrete analysis 5 $115 G-2000 ...... Kontron Discrete analysis 8 $260 G-360 ...... Kontron Discrete analysis 8 $460 aTarrytown, NY bEverett, MA SOURCE:L Friedman, ’’Status ot Automated ABO/Rh Testing i n the US” ARC internal Report, 1983

systems are beyond the grasp of most community have been approved for blood bank products, but blood centers (only 32 of 213 U.S. community many are under development. blood centers were using them in 1984). Recent increase.. ss in the use of microplates for The testing of blood donor samples for unex- manual ABO/Rh testing have led to renewed in- pected antibodies involves the use of reagent red terest in automating that technique. Initial de- bloodcells selected on the basis of antigenic con- velopmental work from at least three companies tent. The test was originally designed to detect indicates that “ELISA” readers, photometers cur- antibodies in recipient plasma that might cause rently used for reading enzyme-linked immuno- difficulty in crosshatching, but it has become a sorbent assays (an immunologic test) can be standard test for every sample entering the blood adapted for reading hemagglutination reactions. bank. Tube or microplate testing is necessary to Coupled with dedicated computers for data anal- detect clinically significant antibodies that react ysis and interpretation, market projections in- only with Coombs reagent. Microplate techniques dicate that these devices could be a more economi- are more efficient and more sensitive for this pur- cal alternative for automated ABO/Rh testing pose (487). (table 32). The automated microplate system pro- totypes require more manipulation of test samples Although traditional screening techniques have than the current automated technologies, but the included multi-phase testing with various enhancers techniques for manual and automated microplate and potentiators, there is little justification for testing are the same, allowing easy conversion elaborate procedures on donor samples. Regular from one technology to the other. blood donors who have not been transfused or pregnant since the last negative antibody screen Still another technology under development for could be excluded from subsequent testing, but several years by Gamma Biological, Inc. (Hous- the effort required to selectively exclude these ton, TX) involves modified microplate technol- donors may negate the benefits of decreased ogy (MicrotearTM), which offers visual or auto- testing. mated interpretation of hemagglutination by the “streaming” of red blood cells after centrifugation. A new and abundant source of antibody production was developed in the 1970s and is being All of the testing technologies described thus applied commercially in the 1980s. Monoclinal far are liquid-phase tests; i.e., they involve the antibody technology (see ch. 6, pt. 2) allows large- mixing of sera-containing antibodies with red scale production of antibodies with selected spec- blood cells in suspension, and the qualitative ificity and potency (575). Production of antibodies reading of the presence or absence of agglutina- in this manner should eventually become less ex- tion. A solid-phase immunoadsorbance technique pensive and have the added advantage of reduc- has been patented by Rosenfield and Kochwa ing variability in test results that is inherent in (U.S. patent # 4,275,053, June 23, 1981), and simi- using naturally produced human or animal serum. lar developmental work has been presented by Monoclinal antibodies also can be easily labeled Plapp and associates (1983), which offer quan- with indicator substances (radioactive, fluores- titative evaluation of antigen-antibody interac- cent, etc. ) (436). Only a few monoclinal reagents tions. Greatly increased sensitivity results from Ch. 4—Blood Technologies ● 89

Table 32.—Traditional Automated Technologies v. Automated Microplate Systems (AMS)

b Tradition ala AMS — Machine Costs ...... $115K-$420K $25K-$60K Cost per sample ABO/Rh...... $0.83-$1.23 $0.68 Operator intervention ...... Minimal Centrifugation, resuspen- sion, and loading of microplates required. Backup technology , ...... , . Second machine, manual Standard manual microplate tube, or microplate. technology is inherent alternative.

Figures are based on estimates from developers SOURCE Friedman, personal communication, 1983 affixing the reaction components to a solid sur- immunoassay (EIA) are of equal sensitivity as face (usually a microplate well), but the utility of radioimmunoassay (RIA) techniques, but are in- this technique in routine donor processing has not creasingly preferred over RIA due to cumbersome been proven. monitoring, licensing and waste disposal require- ments for the radioisotopes used in RIA testing. Except for the semiautomated AutoAnalyzer Both EIA and RIA have automatic reading of re- techniques developed in the late 1960s, screening actions and interpretation of results. Abbott Lab- technologies for syphilis have been unchanged for oratories, the leading company in the field of the past 15 to 20 years. The tests are not specific; HBsAg testing, is developing a more complete approximately 90 percent of positive reactions are automated system which identifies samples by bar due to causes other than syphilis. Questions have code and can interface with other blood center been raised about the health status of false positive computers. reactors; some scientists believe the test may be indicative of pre-disease states, the infectivity of Once all testing is completed on donor samples, which is unknown (562). However, the over- blood components with unsatisfactory test results whelming opinion of FDA expert panels, AABB, must be disposed of and satisfactory units labeled and ARC is that the STS test need not be required according to ABO, Rh and a statement of satisfac- for donor samples. Federal regulations and most tory test results (e.g., for HBsAg), and released State health departments currently require STS for distribution. This is a most crucial procedure, testing, in spite of the lack of scientific evidence since the mistaken release of a unit reactive to to support its continued use. HBsAg would go undetected, and blood labeled Hepatitis testing technologies have undergone incorrectly for ABO group could be transfused rapid evolution since the discovery of Australia into an incompatible recipient. antigen (80). The antigen so discovered was a Labeling and release of blood components are marker of hepatitis B virus and led to develop- tedious and laborious tasks. They are usually per- ment of a succession of increasingly more sensitive formed in batches by a team of two or more per- tests for hepatitis B surface antigen (HBsAg). Like sons, with one individual applying adhesive labels other tests on donor blood, it is repeated on every and the other checking laboratory and donor donation, regardless of the donation frequency records. To minimize human error and facilitate and history, and once a donor has had a con- the process, computers storing laboratory and firmed positive test, he or she is permanently dis- donor suitability data can be used to print appro- qualified as a donor. No confirmation of the (neg- priate labels or to verify that a correct bar-coded ative) test is performed by the transfusion service. label has been applied, and that all testing on a Current “third generation” technologies for blood component has been satisfactory. Such sys- HBsAg testing are compared in table 33. Enzyme tems, fully integrated with automated testing sys-

38-647 0 - 85 - 7 90• Blood Policy and Technology

Table 33.—Current Technologies for HBsAGa Testing

Agglutination test type Immunoassay test type Latex Red blood cell Radio-plastic Enzyme-plastic Test surface Slide/tile Micro plate Bead or tube Bead or tube Sensitivity ...... Good Good Excellent Excellent Specificity ...... Fair Good Excellent Excellent Interpretation of results ...... Subjective Subjective Objective Objective Positive results indicated by ...... Clumping of Clumping of “Radioactive”b Colored particles red blood cells bead solution Automated reading ...... No Under Yes Yes development

terns, are operational in probably not more than Figure 11 .—The Crosshatch 10 community blood centers, and no onsite printers Donor Recipient approved by the American Blood Commission are red blood red blood cells cells being used, but they are the ultimate goal of any facility introducing computers into the laboratory.

Pre-Transfusion Testing Techniques The purpose of test procedures in the transfusion service is to minimize risk of major serologic incompatibility. The ABO/Rh type of the donor unit is rechecked, the ABO/Rh type of the recipient determined, and a crosshatch performed between the recipient sample and selected donor samples. Though not required by Federal regulations, AABB accredited transfusion services must Donor Recipient also perform a recipient antibody screen. Recipi- plasma plasma ent antibody screening yields much the same in- SOURCE: Rossiter, E., “’Technologies for Blood Collection,” 1984. formation as the crosshatch, and there has been increasing questioning of the need for both tests In spite of the diversity of approaches taken in since the antibody screen was required for accred- testing, decisions about when to crosshatch and itation by AABB in 1970. The next revision of the how many units to hold available have become AABB standards is likely to allow some options increasingly standardized. Eighty-three percent of in choosing between the tests. responders to a recent survey (588) used a type- The first crosshatch was a simple mixing of two and-screen (T&S) procedure in which preliminary blood samples and observing for hemagglutina- testing (ABO, Rh, and antibody screen) is per- tion. The test eventually evolved into two distinct formed on low-risk surgical patients, and blood procedures —mixing of recipient plasma with is made available (though not crosshatched). The donor red blood cells, and mixing of donor plasma T&S policy is implemented after each blood bank with recipient red blood cells (fig. 11). The former reviews its own transfusion statistics across “sur- is now regarded as the “major” crosshatch, be- gery-related” groups and determines which elec- cause it is of prime importance in predicting the tive surgical procedures seldom result in trans- likelihood of a hemolytic transfusion reaction (re- fusion. In the rare instances in which blood is cipient’s antibodies would destroy the infused red needed during or after these surgeries, blood will cells). The latter, “minor” crosshatch, is no longer be released for transfusion with an abbreviated required by Federal or voluntary standards. crosshatch. Ch. 4—Blood Technologies . 91

Studies of the safety of T&S practices indicate more effective, and have increased their safety. that they are both safe (408b) and economical The biggest improvements have been in efficiency, (203). T&S and similar policies which establish with increased availability and better utilization a standard number of crosshatched units for of products. surgeries that usually require transfusion, improve Trends for the next 5 to 10 years in the use of blood resources management through lower in- technologies will be continued increases in auto- ventory needs, improved utilization, and de- mation and computerization, and further im- creased outdating. provements in blood preservation. These changes are a natural result of technological changes made Conclusions since the late 1970s and reflect the dynamic climate Technologies developed and put into use dur- in which U.S. blood banks operate. ing the last 10 years have made blood products

PART 2: PLASMA FRACTIONATION TECHNOLOGIES

Introduction Forces had authorized regular use of normal serum albumin (NSA) and issued production contracts Plasma fractionation began in this country dur- to several pharmaceutical firms. Training of tech- ing World War II. At the onset of the war, liquid nical personnel from these firms was conducted human plasma was in use in the battlefields of Eur- by Cohn and his staff. Cutter began to supply the ope, but there was a need for a blood substitute product in late 1942, and by 1943 seven U.S. com- that was easy to transport and which could be panies were involved in fractionation and supply administered on the battlefield under widely vary- of human plasma products. Following World War ing climatic conditions. II, these companies continued to process plasma At the request of the Medical Advisory Com- to meet the demand for various plasma proteins mittee of the Red Cross and the National Research for civilian use. Council’s Committee on Blood Transfusions, Ed- During the beginning of plasma fractionation win J. Cohn, head of the Department of Physical in the 1940s, albumin production and sale pro- Chemistry at Harvard University, agreed to vided the economic support for the industry. Later undertake an investigation to determine whether in the 1950s, the need for gamma globulin for pro- or not the plasma of animals could be made safe phylaxis against poliomyelitis, infectious hepatitis, for transfusion. By mid-1940), Cohn notified the and childhood viral diseases provided financial Red Cross of the development of a system which security for the industy. Subsequently, prevalence would yield not only albumin but other protein of these diseases was markedly decreased, and at components of plasma as well, but said he felt that about the same time methods for recovering Fac- dependence should be placed on the use of human tor VIII were greatly improved, leading to in- plasma rather than that of animals. The Red Cross creased treatment of hemophiliacs in the United then opened a blood donor facility at Peter Bent States and abroad. As a result, sales of Factor VIII Brigham Hospital in Boston to provide plasma to have been a significant source of revenues from support Cohn’s work. In April 1941, the first the mid-1960s to the present. human serum albumin produced by Cohn was released for clinical trials, and by July 1941, pilot production facilities had been expanded. Plasma Collection In December 1941, clinical studies had not yet Three types of plasma are used to produce been concluded, but human derivatives from Har- plasma derivatives; liquid single-donor plasma, vard were rushed to Pearl Harbor for use in treat- fresh-frozen, single-donor plasma, and “source” ment of injured servicemen. By 1942, the Armed plasma, which is collected by plasmapheresis. Liq- 92 . Blood Policy and Technology uid and fresh-frozen, single-donor plasma are pri- be used to prepare platelet concentrates, buffy marily obtained through volunteer agencies as coats (leukocytes), and washed red cells. “recovered” plasma from whole blood donations. “Recovered” plasma is either fresh frozen within On a theoretical basis, filtration through mem- 6 to 24 hours of collection or is salvaged from out- branes of appropriate pore size should be the dated whole blood at the time the red cells are simplest way to separate liquid plasma from the discarded. Only fresh-frozen plasma is suitable cells in whole blood. Earlier development of this for Factor VIII fractionation. Increased use of technology has been hampered by the fact that components has resulted in an increasing propor- filtration does not simultaneously yield cellular tion of recovered plasma being frozen within a components such as platelets to help amortize the few hours of collection to protect the labile coag- development costs. However, its potential value ulation factors. Single units of fresh-frozen plasma in therapeutic plasmapheresis may catalyze its are then shipped directly to the fractionators. evaluation in the collection of normal (source) plasma. Membrane filtration at low temperatures In plasmapheresis, 500 ml of whole blood is has also been attempted experimentally to isolate drawn into a sterile container in which there is Factor VIII-rich cryoprecipitate. Large-scale ap- anticoagulant solution (either ACD Formula A, plication of this new technology has not been at- CPD, or Trisodium Citrate). The container is then tempted. detached and the blood centrifuged in order to separate the red cells from the plasma. During this The estimated time required for a centrifugal time the phlebotomy needle is kept open with plasmapheresis separation is 30 to 40 minutes, and an infusion of isotonic salt (NaCl) solution. The for continuous flow membrane filtration, 50 to red cells are then reinfused into the donor, and 60 minutes. When donor preparation time is the process is repeated a second time. The 0.5 to added in, the overall plasmapheresis time is slightly 0.6 liters of collected plasma are then frozen. The longer than 1 hour. This compares with approx- average elapsed time for such plasmapheresis is imately 90 minutes for the traditional manual sys- 90 minutes. Each donor is permitted to undergo tem (2). The interrupted flow centrifugal system a double-bleeding of this type twice per week. also requires only a single venipuncture, as op- posed to the two needles or large, double lumen At least three systems for the automation of needle necessary for continuous flow. The prin- plasmapheresis are currently undergoing research cipal advantages of automated plasmapheresis and development: membrane filtration, continu- are: 1) the donor is always attached to his/her ous flow centrifugation, and intermittent flow own red cells or whole blood, thus preventing ac- (batch) centrifugation. As yet, none is able to cidental administration of mismatched cells; and compete economically with the traditional man- 2) the potential population of plasma donors may ual technique. Although the manual system is be able to be expanded significantly throughout labor-intensive, the total cost of a double pack the world. Automation also raises the possibility of plasma is in the range of $22 to $25 (277), in- of decentralized plasmapheresis, and, by shorten- cluding $7 to $10 for the donor fee, $8 for the ing the overall procedure time, also enhances the blood bag set and saline solution, $2 for donor possibility of inclusion of volunteer donors for screening and testing, and $5 for labor. In the case source plasma. The remaining unanswered ques- of automated centrifugal systems, the disposable tion is whether or not increased volumes of plastic plastic bowls alone cost more than this; Haemonetics disposable will lower costs sufficiently to com- $55, Fenwal $60, and IBM $65. However, these pete with a manual system in which labor costs costs could plummet if simplified to be used for are only 20 percent of final production costs. source plasma alone. The state of the art of the automated centrifugal system has been developed There are two types of plasmapheresis donors: far beyond the needs for plasma collection because standard donors and specialized donors. High- the same basic hardware (with costs of approx- titer antibody preparations (hyperimmune glob- imately $20,000) and disposable plastic bowls can ulins) are made from the plasma of specialized Ch. 4—Blood Technologies ● 93

donors. Successful passive immunization is greatly Fraction VI, the residue remaining, is currently improved by use of gamma globulin fractions high discarded due to lack of known therapeutic use- in specific antibody content against the individ- fulness. ual diseases. The hyperimmunization of selected Method XII, the final technique evolved by the (specialized) donors through repeated stimulation Cohn group (128), gave promise of considerable of their immune systems with specific antigens simplification by using metallic precipitation of (noninfectious products made from the inactivated the globulins with zinc diglycinate, but this method protein of the underlying virus or bacteria) can was never commercially adopted due to prelimi- provide a pool of high-titer antisera. This tech- nary evidence of contamination of the albumin nique has provided specific globulins for the man- fraction with hepatitis virus. It also was imprac- agement of such diverse disorders as rabies, her- tical in its requirement for decalcified plasma pes zoster, and hepatitis B. (through ion-exchange treatment) rather than the In a similar manner, individuals who inad- usual citrated plasma. vertently have been immunized through prior Due to a shortage of industrial sources of al- blood transfusions or pregnancy, or who elect to cohol in England following World War II, the become donors of blood group antibody sera, can principal fractionation facilities which were con- have their titers restimulated over a period of structed at the Lister Institute were modified to years and maintain their role as specialized plas- permit use of ether rather than ethanol. The basic mapheresis donors for the production of anti-A, principle of the four-variable Cohn system was anti-B, or anti-Rh immune globulin for blood nonetheless used, and this method was continued group typing, or, in the case of anti-Rh, for the for a period of approximately 20 years, follow- prevention of Rh hemolytic disease of the new- ing which the ethanol system was again employed. born (erythroblastosis fetalis). Nevertheless, many minor modifications of the Some overlap occurs between standard and spe- basic alcohol precipitation technique have been cialized plasmapheresis donors. Immunization introduced successfully with improvement in against tetanus, for example, is so commonly yield, cost effectiveness, and stability of individ- practiced and easily accomplished that it is com- ual products. mon for standard donors to receive occasional booster doses, making possible the recovery of Factor VIII a high titer of tetanus antibodies from routine In the original alcohol fractionation method, gamma globulin fractionation. Factor VIII precipitated with fibrinogen in Frac- tion 1-.. The yield of Factor VIII by this method Fractionation Methods and Products was similar to that developed by cryoprecipita- tion, but contamination of the final product with The principal method used for the isolation of other proteins was considerably higher with the the different plasma proteins continues to be the original alcohol fractionation method. These con- cold-ethanol precipitation technique of Cohn and centrates were used to treat hemophilia during the collaborators (127). This method employs a four- 1940s and 1950s, and were marketed primarily for variable system of temperature, ionic strength, their fibrinogen content. After the development ethanol concentration, and pH to precipitate: of the cryoprecipitate technique by Pool, et al. ● Fraction I—chiefly Factor VIII and fibrinogen, (442), Fraction I-o was abandoned as too crude ● Fraction II—the gamma globulins, a preparation to use for its Factor VIII content. ● Fractions III and IV—other coagulation pro- The fibrinogen, which the recipient received as teins and trace components such as cerulo- a side effect of the Factor VIII replacement, led plasmin and iron-binding globulin, and to problems in red cell crosshatching. Contamina- ● Fraction V—the albumins. tion with hepatitis virus was also so prevalent as 94 ● Blood Policy and Technology

to no longer justify its commercial preparation, Gamma Globulins and approval of Fraction I-0 (fibrinogen) as a Essentially all gamma globulins are recovered plasma product was withdrawn by FDA in the from Cohn Fraction II using the classic cold- late 1970s. ethanol technique. Immune serum globulin (ISG) “Cryoprecipitate” was discovered by Pool and is dispensed as a 16 percent protein solution for associates (442,443) to trap a significant portion intramuscular injection in amounts up to 10 ml, of Factor VIII, and thus provided a simple physi- or as a solution for intravenous injection (Intra- cal step to harvest it. Currently, cryoprecipitate venous gamma globulin, or IVGG). is either prepared as a single unit in a plastic bag during the routine collection and processing of a single unit of plasma in a blood bank, or the Albumin plasma is frozen and shipped to the fractionation industry. The production of albumin from fraction V-o utilizes pH, ionic strength, and ethanol concen- The concentration of Factor VIII in cryoprecip- tration to precipitate a highly purified concentrate itates varies widely, depending on such factors as of this principal oncotic agent of plasma. The the size of the ice crystals which form and the local powdered albumin is then resolubilized as either salt concentrations left behind when water crys- a 5 percent solution, with the same oncotic pres- tallizes into ice (419). Federal regulations require sure as normal plasma, or as a 25 percent con- that the average Factor VIII potency of a single centrate of salt-poor albumin. The 25 percent con- unit of cryoprecipitate be greater than 80 units centrate is marketed as a hyperoncotic concentrate as determined on four representative containers which draws water from the extravascular tissues each month (21 CFR pt. 640.56). (The Factor VIII to restore blood volume and circulatory com- content of any single unit that is being used ther- petence in a recipient suffering from hypovolemic apeutically cannot, of course, be measured. ) For shock. example, the average content of a bag of cryoprecipitate as determined by one Red Cross blood An alternative method of albumin production bank was assayed as containing: 1) 120 units of is the elimination of the subfractionation steps Factor VIII (range of 105 to 130), 2) 46.6 mgm which precipitate Fractions IV and V and recon- of fibronectin, 3) 326 mgm of fibrinogen, and 4) stitution of these pooled fractions as an albumin present but not quantified amounts of von Wille- solution. This material, plasma protein fraction brand factor (420). (PPF), contains 83 to 90 percent albumin, no more With commercial preparations of Factor VIII than 17 percent of the total protein as alpha or concentrates, a higher concentration of Factor VIII beta globulin, and no more than 1 percent as can be achieved with less volume of infusion gamma globulin. Currently, after the preceding needed to administer it, and an exact Factor VIII fractions are removed, 30 percent of all frac- content of each preparation can be assayed rather tionated plasma goes into PPF production and 70 than assumed. Moreover, the stability and pack- percent into albumin production. aging of the concentrates is more suitable for Albumin (and gamma globulin) can also be pre- home therapy. pared from human placentas. Such processing is When large pools of frozen material are proc- different from plasma because of the large amount essed, thawing is carefully controlled to improve of connective tissue, free supernatant hemoglo- yield and stability. The best yields of Factor VIII bin, and other extraneous material which must from cryoprecipitate are approximately 25 per- be removed by filtration, adsorption, and cen- cent of the starting material. (As mentioned in ch. trifugation, but the final products are of high 3, heat-treated Factor VIII concentrates are now purity. The average yield of albumin by this tech- available. ) nique is 5 gin/kg of placenta.