Agglutination and hemolytic crossmatching to determine transfusion reaction differences between large and small breed goats

Abstract Background Blood transfusions are performed frequently in goats, but crossmatches are rarely performed. Hypothesis/Objectives Determine differences in the frequency of agglutination and hemolytic crossmatch reactions between large and small breed goats. Animals Healthy adult goats, 10 large and 10 small breed. Methods Two hundred eighty major and minor agglutination and hemolytic crossmatches: 90 large breed donor to large breed recipient (L‐L), 90 small breed donor to small breed recipient (S‐S), 100 large breed donor to small breed recipient (L‐S). A linear mixed model with treatment group (L‐L, S‐S, L‐S) as a fixed effect and individual crossmatch as a random effect was used to identify variations in reaction frequency among groups and individuals. Results Frequency of major agglutination reactions for L‐L, S‐S, and L‐S were 3/90 (3.3%), 7/90 (7.8%), and 10/100 (10.0%), respectively. Frequency of major hemolytic reactions for L‐L, S‐S, and L‐S were 27/84 (32.1%), 7/72 (9.7%), and 31/71 (43.7%). Individual pairings and groupings had no effect on agglutination reactions. Individual pairings had no effect on the frequency of hemolytic reactions. For major hemolytic crossmatches, pairwise comparisons identified higher frequencies of reactions when comparing L‐L to S‐S (P = .007) and L‐S to S‐S (P < .001). Conclusion and Clinical Importance Goats experience increased frequencies of hemolytic reactions compared to agglutination. Significant increases in hemolysis were seen between large breed donors and small breed recipients, compared to small breed pairings. Additional studies are required to determine correlations between crossmatches and transfusion reactions.

Whole blood transfusions are used to treat anemia attributed to various conditions in goats, including severe parasitism, trauma, and hemolytic anemia. [1][2][3] Crossmatches are performed rarely before the administration of whole blood because of a large number of blood groups in small ruminants combined with few choices for donor animals, expense, lack of expertise, and lack of established protocols.
Goats have at least 6 known blood group systems (A, B, C, E, F, and R), with the B system containing over 52 factors. 2 Consequently, blood typing is rarely performed, and the frequency of first-time transfusion reactions is considered relatively low. A recent retrospective study reported a transfusion reaction frequency of 15.6% for small ruminants, but risk related to size and breed was not evaluated. 3 The published maximum amount of blood that a small ruminant donor can provide is 10-15 mL/kg of body weight or 20% of total blood volume per month. 1 Many small breed goats (Nigerian Dwarf, Pygmy) can donate a maximum of 300-450 mL of blood, based on an average weight of 30 kg. 1 However, a large breed goat (Alpine, Boer, Nubian, Saanen, La Mancha) can donate 750-1125 mL of blood at a time based on an average weight of 75 kg. 1 Consequently, large breed goats have more utility as blood donors than do smaller breeds, especially if being kept as blood donors at a clinic.
Unlike small animals, ruminants are more prone to hemolytic reactions compared to agglutination reactions. 2 Therefore, performing both an agglutination and hemolytic crossmatch is recommended when screening donors for a given recipient. 2 No published protocols are available for agglutination or hemolytic crossmatches in small ruminants, and reports have not been published concerning the frequency of crossmatch incompatibility among goats of variable size and breed. Our objectives were to develop a crossmatch protocol for goats and determine the differences in frequency of agglutination and hemolytic crossmatch reactions among goats of similar and different sized breeds. We hypothesized that an increased frequency of crossmatch reactions would occur between large and small breed goats compared to crossmatches between goats of the same breed size, and that goats will have a higher frequency of hemolytic reactions as compared to agglutination reactions.

| Animals
Twenty goats divided equally into large and small breeds were included. A physical examination was performed on each goat before blood collection and enrollment in the study. Only goats considered healthy based on physical examination and medical history were included in the study. In addition, goats that were difficult to handle for blood collection and any goat that required >1 needle insertion or redirection were excluded to minimize the occurrence of sample hemolysis. The 10 large breed goats consisted of 5 Saanens, 2 Alpines, and 3 LaManchas. These goats were owned by the University of California, Davis, Animal Science Department. The 10 small breed goats consisted of 5 Nigerian Dwarf and 5 Pygmy goats that were privately owned. Of the 20, 17 were open females (14 currently lactating), 1 wether, and 2 bucks. All animals were routinely vaccinated for Clostridium tetani and C. perfringens type C and D and were seronegative for caprine arthritis and encephalitis virus, caseous lymphadenitis, and Mycobacterium avium paratuberculosis based on farm protocols. The study protocol was approved by the Animal Care and Use Committee at the University of California, Davis.

| Study design
Thirty milliliters of blood were drawn from the jugular vein of each goat, using 20-gauge vacutainer needles, and placed in 2 red top (additive free) tubes and 2 lavender top tubes containing ethylenediaminetetraacetic acid (EDTA). The EDTA-containing samples were inverted slowly to promote homogenization and prevent clot formation. To minimize hemolysis, samples were allowed to cool to room temperature for 30-60 minutes before being stored at 4 C for up to 72 hours.
One EDTA tube from each patient was submitted to the William

R. Pritchard Veterinary Medical Teaching Hospital (VMTH) Clinical
Diagnostic Laboratory for a CBC. Plasma protein concentrations were measured using a hand-held optic refractometer (ADE Advanced Optics inc., Oregon City, Oregon) for each sample after centrifugation (Unico, Dayton, New Jersey) of a microhematocrit tube at 3000 rpm for 3 minutes. Plasma fibrinogen concentration was determined using the heat precipitation method. 4 Finally, a Romanowski Wright'sstained blood smear from each animal was evaluated by a clinical pathologist (FA) to assess for erythrocytic, leukocytic, and thrombocytic morphological abnormalities. These diagnostic tests were performed in the event an individual animal was found to have increased frequencies of crossmatch reactions compared to others. These tests are not a part of the crossmatch protocol.
Each animal was used as both a donor and recipient in crossmatch combinations, as presented in Table 1. A total of 280 major and minor agglutination and hemolytic crossmatches were performed: 90 large breed donor to large breed recipient (L-L), 90 small breed donor to small breed recipient (S-S), and 100 large breed donor to small breed recipient (L-S). The crossmatch procedure was adapted from established protocols used in dogs and horses and is described below. 2,4,5 All agglutination and hemolytic crossmatches were graded by a single clinician (CK) throughout the study. Before the study, this individual was trained by the university's clinical pathology department.

| Complement solution
Before the crossmatch procedure, 10 mL of blood in a red top tube was collected from the jugular vein of a university-owned blood donor goat that was not included in the study. The donor goat was an adult castrated male Nubian. The blood was centrifuged for 5 minutes at

| Sample preparation
All donor and recipient red blood cells were washed before the crossmatch procedure. Briefly, 2 mL of whole blood from EDTA tubes and an equal volume of the 0.9% saline solution were placed in glass tubes using plastic transfer pipettes (Cole-Parmer, Vernin Hills, Illinois), and centrifuged at 3000 rpm for 2 minutes. The supernatant was discarded, and the RBC pellet was resuspended in saline. This washing procedure was repeated 3 times. Finally, a 5% solution of washed RBCs was created by mixing 0.1 mL of washed RBCs with 2.0 mL of 0.9% saline.
Blood from red top tubes was centrifuged at 3000 rpm for 5 minutes and serum was transferred to a clean glass tube and refrigerated at 4 C until needed.

| Agglutination crossmatch
Diagrams of the agglutination and hemolytic crossmatch setups and procedures are depicted in Figure 1

| Hemolytic crossmatch
Major, minor, and auto control hemolytic crossmatches were performed. Each major, minor, and auto control tube contained the same mixture as described above, but 0.05 mL (1 drop) of 1:2 complement was added to each reaction mixture. The samples were vortexed, incubated, and centrifuged as described above. Hemolysis was graded on a scale of 0 to 4 ( Figure 2). This scoring system was adopted from the university's crossmatch protocol used in horses. A score of 0 was considered negative for hemolysis with a clear background of similar color to the control. A grade of 4 was assigned to samples with a dark red background and considered the most severe.
T A B L E 1 Blood crossmatching analysis design for 20 goats based on large (L) and small (S) breed size.
Each goat was assigned a letter-number combination (L1-L10, S1-S10) based on breed size. Groupings include 90 large breed combinations, 90 small breed combinations, and 100 large breed donor to small breed recipient combinations. Large breeds = Saanen, Alpine, La Mancha. Small breed = Nigerian Dwarf and Pygmy.

| Delayed reactions
Incubation and grading were repeated, immediately after initial grading, for each crossmatch, using the same samples to detect any delayed reactions in both the agglutination and hemolytic crossmatches. This second incubation was performed for 30 minutes at 38 C. After incubation, the samples were vortexed for 5 seconds and centrifuged for 2 minutes at 3000 rpm to create an RBC pellet. Grading for both agglutination and hemolytic crossmatches was assigned as described above. Delayed reactions were defined as an increase from grade 0 to ≥1 between the first and second incubations. All results include total reactions from both the first and the second incubation, unless stated otherwise.

| Statistical analysis
Data analysis was performed using statistical software (GraphPad Crossmatches were graded on a scale of 0-4, and grades ≥1 were Tukey's test were performed to identify specific differences between groups or pairs. A P value <.05 was considered significant. Note: For parametric data, mean ± SD were reported, whereas median (95% CI) were reported for nonparametric data. Lab-specific reference ranges (RR) are provided from the VMTH. Large breeds = Saanen, Alpine, La Mancha. Small breeds = Nigerian Dwarf and Pygmy.
Linear mixed model analysis indicated that treatment group (L-L, S-S, L-S) had no effect on the frequency of agglutination reactions for both major (P = .2) and minor (P = .4) crossmatches. Individual crossmatch pairs similarly had no effect for both major and minor crossmatches (P > .99). Further pairwise comparisons were not performed.

| Hemolysis
Sample hemolysis can hinder the ability to accurately grade hemolytic reactions. 2 Consequently, any hemolytic autologous control tube with lysis present during the first incubation and its corresponding crossmatches were excluded from the study.    Table 3.  Table 3B does not include the autologous control tubes that exhibited delayed hemolysis described here.
Linear mixed model analysis indicated that the random variable of individual crossmatch pairs had no effect on the frequency of hemolytic reactions for both major and minor crossmatches (P > .99). However, treatment groups were identified to have an effect in both major and minor crossmatches (P < .001). In the major hemolytic crossmatches, pairwise comparisons identified significant differences when comparing the L-L to the S-S groupings (P = .01) and the L-S to the S-S grouping (P < .001). In the minor hemolytic crossmatches, differences when comparing the L-L to the L-S groupings (P < 0.001), as well as the L-L to the S-S groupings (P < 0.001), were identified. In summary, when compared to the S-S groupings, a significantly increased frequency of hemolytic reactions was identified within the L-L and the L-S groupings.

| DISCUSSION
We developed a simple crossmatch protocol for goats. As hypothesized, our study results reinforced that goats, similar to other large animals, have a higher frequency of hemolytic reactions compared to agglutination. Similarly, differences in the frequency of hemolytic crossmatch reactions among breed groups were identified. Most pertinent to the objectives, a higher frequency of major hemolytic reactions occurred when blood from large breed donors was paired with small breed recipients compared to small breed donors paired with small breed recipients. Unexpectedly, the large breed pairings also had increased frequencies of hemolytic reactions compared to small breed pairings.

| Agglutination
In our study, 12.9% of the total crossmatches performed were positive for agglutination. Within the L-L, S-S, and L-S groups, 6.7%, 15.6%, and 16.0% of crossmatches showed agglutination, respectively. This finding is similar to a previous study that assessed transfusions in sheep that cited 17.1% incompatible agglutination crossmatches. 6

| Hemolysis
Overall is unclear. These findings could indicate a higher concentration or diversity of naturally occurring alloantibodies in goats than previously has been thought or an increase in R-positive blood types in large breed goats. However, the reason is difficult to determine without additional studies to assess specific breed differences, blood types, and alloantibodies, as has been done in other species. 5,7 The importance of these alloantibodies and their clinical correlation with an acute hemolytic transfusion reaction is unknown. Animals with low concentrations of alloantibodies may not always mount a massive acute hemolytic response but rather may slowly upregulate alloantibody production and gradually destroy transfused RBCs over a few days. 5 In dogs and cats, the antigenicity of certain RBC antigens and their association with acute hemolytic reactions versus delayed destruction of RBCs (if alloantibodies are present) is known. 5,7

| Limitations
The main limitations of our study were sample storage time and sample hemolysis. All samples were used within 72 hours, but some samples were utilized at the start of this period and others at the end. All serum samples were separated and stored in individual glass tubes before the study. Whole blood was stored in EDTA tubes, and aliquots were removed to prepare washed RBCs as needed. In the clinical setting, crossmatches are performed immediately after blood collection, and little information is available on the effect of storage, especially for small ruminants. One study in horses cited poor reproducibility in hemolytic and agglutination crossmatch results with blood that had been stored for 1-4 weeks. 8 However, this study did not evaluate storage times <7 days. 8  Finally, all scores were assigned by a single individual who was not blinded to the crossmatch pairings throughout the study. Without blinding, an inherent risk of bias exists because of awareness of crossmatch groupings (L-L, S-S, and L-S) and individual pairings.
The applicability of our findings in a clinical setting when performing blood transfusions is unknown and requires further investigation.
Although our study did not identify any individual effect on crossmatch reaction frequencies, it is important to note that our results represent a small population of healthy animals. Compromised patients may be at an increased risk of transfusion reactions. Similarly, prior dystocia and the use of blood-contaminated needles could lead to development of alloantibodies, potentially increasing the risk for crossmatch and transfusion reactions.
Although we reported the frequency of agglutination and hemolytic crossmatch reactions, it is unknown whether crossmatch scores in goats equate to acute or delayed transfusion reactions, shortened RBC survival, or outcome. Future studies correlating in vitro crossmatch results to in vivo effects are needed. A study assessing transfusion reactions in cats found a larger increase in mean PCV 24 hours post-transfusion when major crossmatches were performed. 13 Similarly, RBC survivability based on crossmatch results has been studied in horses, and cross-match incompatibility was significantly associated with decreased RBC survival leading to an RBC half-life of 4.7 days versus 33.5 days with compatible crossmatches. 14 Interestingly, horses transfused with crossmatch incompatible donors developed acute febrile episodes 30 days after transfusion. 14 Acute intravascular hemolysis with concurrent hemoglobinemia and hemoglobinuria can be seen with incompatible combinations in small animals. 2 Similarly, delayed reactions can be seen up to 14 days after transfusion, and extravascular hemolysis is more prominent. 2 Many current publications in small ruminants focus on acute changes in vital parameters during the actual transfusion, with hyperthermia being the most commonly cited acute adverse effect. 3 However, this reaction is most likely a response to leukocyte or platelet antigens or less commonly bacterial contamination of blood products rather than an erythrocyte antigen-antibody response. 2

| CONCLUSIONS
Based on our results, smaller breeds are potentially at increased risk for hemolytic reactions when receiving blood transfused from larger breed goats. However, additional studies involving the correlations among crossmatch score, transfusion reactions, and RBC survival time are required to support this hypothesis. A second incubation of donor's and recipient's washed RBCs and serum seems to be important to increase the sensitivity of a crossmatch reaction in this species.

ACKNOWLEDGMENT
No funding was received for this study.