Page 100 - JSOM Summer 2019

Minor Crossmatch antibody response is present in addition to activation of com-
plement. Interestingly, the B– samples from the commercial
All samples in the minor crossmatch showed hemolysis or test kit hemolyzed in both the heated and nonheated samples.
marked agglutination. Of the 80 samples tested, 51 samples The reason for this reaction is unclear but may have to do
showed hemolysis and 29 samples showed marked agglutina- with the interaction of the heated plasma constituents and
tion. (Figures 2 and 3). There was no difference between DEA stabilizing factors contained in the sample from the test kit.
1–positive and DEA 1–negative samples (p = .48). All seven of Regardless, the results of all of the crossmatches in the heat
the controls were considered negative.
denaturation study were consistent insofar as they suggest that
human RBCs should not be administered to canines.
Heat Denaturation Experiment
There are numerous limitations in this study. First and fore-
All samples using canine plasma with human RBCs showed most, this was an in vitro study and results of this study do
marked hemolysis (to include the B– RBCs from the test kit). not necessarily reflect the immune response in a living, severely
All controls were considered negative. In the samples where injured dog. In fact, numerous studies have shown that hu-
the plasma was heated, the samples mixed with A– and O– man serum albumin is antigenic in healthy dogs, 24,25 but some
RBCs taken from donors showed agglutination rather than dogs with severe illness can tolerate an infusion of human
hemolysis. The samples that contained B– RBCs from the com- albumin. 26,27 However, given the absence of any other data,
mercial kit continued to show hemolysis.
the results of this pilot study should give significant pause to
clinicians considering administration of human blood to a
Discussion canine as the data suggest that a potentially life-threatening
hemolytic reaction may occur. Second, this was a pilot study
The results of this pilot study suggest that transfusion of hu- and not powered to detect small differences. While the results
man blood into canines has the potential to cause severe trans- of this study appear fairly consistent, a larger study would be
fusion reactions, in most cases an acute hemolytic reaction. needed to draw definitive conclusions. Finally, a limited num-
Every sample tested (except for controls) showed cross-reac- ber of human blood donors providing RBCs were used in this
tivity, which would preclude transfusion in a normal clinical study, increasing the possibility of both type 1 and type 2 er-
scenario. Our results are consistent with early in vitro attempts rors. Careful consideration should be given to further testing
to transfuse various animal blood into humans, often leading of xenotransfusions from human to canines while weighing
to lethal transfusion reactions. 21
the potential benefits with the likely significant risks.
In the major crossmatch, the only samples that showed agglu-
tination rather than hemolysis were the samples that used A– Conclusion
and B– RBCs from the commercial test kit. While it is possible Strong cross-reactivity in the form of both hemolysis and ag-
that this result is due to differing antigens on the human RBC glutination was seen in every sample in both the major and mi-
surface, it is more likely that the difference is due to stabilizing nor crossmatches between human and canine blood. Based on
agents contained within the washed RBC preparations. The the results of this study, the authors must strongly recommend
commercial test kits contained citric acid, and lower pH has against transfusion of human blood to injured canines as the
been shown to inhibit complement response while preserving risk of a life-threatening transfusion reaction is high.
22
cells’ ability to aggregate. However, we included these data
for two reasons. First, the authors had no other samples of B–
blood readily available and, despite the apparent complement Funding
This study was funded by a grant from the Department of
inactivation, robust aggregation still occurred, suggesting that Defense Special Operations Command.
B– human blood is not compatible in a canine.
The minor crossmatch showed more variability between ag- Disclaimer
glutination and hemolysis, but there was no difference in the The views expressed in this manuscript are those of the au-
proportion of hemolysis between DEA 1–positive and DEA thors and do not reflect the official policy or position of the
1–negative blood. This suggests that human plasma is incom- Departments of the Army, the Department of Defense, or the
patible with canine RBCs and that the transfusion of human US Government.
whole blood or plasma carries a high risk of transfusion re-
action if administered to a canine. The variable response in Disclosure
the crossmatch shown here suggests that the complement sys- The authors declare no conflicts of interest.
tem may not cause hemolysis in all cases and that, in some in-
stances, agglutination from antibodies plays a more important Author Contributions
role. TE, RB, EM, and AC conceived of the study concept. RB and
TE obtained funding. TE, NW, and SG collected and analyzed
The heat denaturation experiment was undertaken to try to the data. TE, EM, NW, and SG wrote the first draft. All au-
gain more understanding about the role of complement in thors read, edited, and approved the final manuscript.
the reaction. It has long been known that heat will denature
complement, rendering it inactive. We therefore heated the References
23
canine plasma to eliminate the contribution of complement in 1. Eastridge BJ, Mabry RL, Seguin P, et al. Death on the battlefield
the reaction. In the A– and O– samples, the resultant cross- (2001–2011): implications for the future of combat casualty care. J
Trauma Acute Care Surg. 2012;73(6 Suppl 5):S431–S437.
match showed agglutination in the same samples that showed 2. Anonymous. Tactical Combat Casualty Care Guidelines for Medi-
hemolysis in the nonheated samples, suggesting that a robust cal Personnel. 03 JUN 2015. JSOM 2015;2016(01/15):14.

98 | JSOM Volume 19, Edition 2 / Summer 2019