A Canadian Blood Banker reports that her laboratory routinely screens patients for unexpected red cell alloantiabodies by testing SERUM (Becton Dickinson, red top, no additive) by a tube antiglobulin method using PEG (polyethylene gycol; supplied by Dominion Biologicals Ltd [DBL]). They use 2 drops of serum, 1 drop of 5% red cells and 2 drops of PEG, incubated at 37C for 10 minutes, washed in DACII cell washers and read with DBL Murine Monoclonal Blend IgG per the manufacturer's instructions. Recently they discovered a serum antibody that defies identification by reacting 'microscopically' with all cells tested, including autologous cells. They repeated the PEG screen using a different antiglobulin reagent, Gamma Biological Anti-IgG, but still got the same microscopic pan agglutination results. Their screening cells and panel cells are obtained from DBL. They used a Gamma twenty panel. In this case they also ran the same tests with PLASMA (K3 EDTA from Becton Dickinson) and the results were negative. The Canadian adds that they sent the positive specimen out to a reference lab for further investigation. While specific details were not given, the reference lab report indicates that all clinically significant antibodies were excluded. High Titer Low Avidity antibodies (HTLA) such as Anti-Ch, -Rg, -Csa, -Yka, -Kna, -McCa, -JMH etc and (HLA) Human Leukocyte Antibodies also were ruled out. It is the inquiring blood banker's understanding that reference lab ran a panel of rare cells to rule out or confirm HTLA and HLA, and that chloroquine was used to treat reactive red cells to remove Bga antigens. The inquiring blood banker's lab has also tested the reactive serum by other methods. They used a LISS-indirect Antiglobulin Test using PS Antibody Potentiating Solution supplied by DBL (2 drops serum, 2 drops PS, 1 drop 5% red cells; 10 minutes incubation at 37C, washed in DACII cell washers and read with DBL IgG, Gamma's IgG and Anti-Human Globulin Polyspecific DBL. They got the same results as with PEG. In addition, they tested the serum by a saline indirect antiglobulin test which demonstrated the same results. Their saline-IAT consists of 2 drops serum, 1 drop 5% red cells, 30-minute incubation, washed in DACII cell washer and read with both DBL's IgG and Gamma's IgG. The inquiring Canadian would like input to explain their observed serologic reactions when testing serum (getting positive results) versus testing plasma (getting negative results).

In response to the above case report and question, here are several helpful replies.

1. A colleague wonders if when the Canadian blood banker says that no additive was used in the red top tubes, if they meant that the red top tubes were not serum-separator-tubes, because serum-separator tubes are known to cause some problems.

2. The Canadian blood banker confirmed that the red top tubes used were NOT serum separator tubes, but merely plain red top tubes.

3. A blood banker from Portland wrote that an "antibody" reacting in serum but not in plasma might be related to a medical condition known as ulcerative colitis. According to the responding blood banker, an abstract from TRANSFUSION, Vol. 20, No. 5, Sept.-Oct., 1980, pg. 646, by Garratty et al, entitled 'IgG Red Cell Sensitization Associated with in vitro Clotting in Patients with Ulcerative Colitis' might explain this situation. In this abstract it states "We have encountered eight ulcerative colitis patients with a positive direct antiglobulin test (DAT) due to IgG sensitization of clotted red cells, but negative or very much weaker positive when anticoagulated blood was tested. The weakly positive DAT on anticoagulated blood only occurred occasionally in some patients. Eluates from clotted red cells sensitized normal red cells with IgG. The patients' sera, but not the plasmas, from a variety of anticoagualted samples reacted by indirect antiglobulin test (IAT); individual patients varied from 1+ to 4+ IAT. No specificity could be demonstrated. This appeared to be an in vitro phenomenon. When fresh blood from the patient was tested immediately, the DAT was negative. IgG sensitization of the patients' red cells developed during in vitro clotting. No obvious signs of hemolytic anemia were present. Some of the patients were transfused with blood that was incompatible with the serum, but compatible with the plasma.; no ill effects were noted. A chromium-51 red cell survivaL study in one patient was normal. A prospective study has, so far, not demonstrated the described phenomenon in 30 patients with ulecrative colitis (four of these had positive DAT). It is hypothesized that the eight patients described may have an antibody to an activated coagulation factor present in their plasma; during clotting, immune complexes may form in the serum and attach to red cells." The Portland blood banker reports that he has encountered a case similar to the ones described in the above abstract sometime in the mid 1980's, at which time he consulted Dr. Garratty about the problem.

4. John Judd, FIBMS, MIBiol, Professor of Immunohematology at the University of Michigan comments that "the one thing that comes immediately to mind is a phenomenon that Dr. George Garratty can expound on in great detail. In 1980, at an AABB meeting, he described 8 cases of patients with ulcerative colitis who had autoantibodies detectable in serum but not in plasma (Transfusion 1980;20646). This topic is discussed in the 3rd edition of Applied Blood Group Serology, but we have been unable to find it in the 4th edition! The inquiring Canadian may want to do an IgG DAT on RBCs from the clot and the EDTA tube. There was also a report at the previous year's meeting by Tregellas who described a single case of the phenomenon; his case had autoanti-Sc1 specificity. My word of advice to the inquiring individual is not to read tests microscopically, especially PEG tests! He or she should also be aware of the fact that PEG precipitation of proteins, especially fibrinogen in plasma samples, can lead to the deposition of immunoglobulins onto RBCs that results in false-negative IATs (Coombs control cells do not react). This probably doesn't apply to the case in hand, but one PEG manufacturer cautions against using plasma for PEG tests."

5. Dr. George Garratty, Scientific Director at the American Red Cross, Southern California Region, responded that he would refer the Canadian Blood Banker to two abstracts that were published by his group in TRANSFUSION, the first being published in 1980, volume 20 p646; and the second in 1993 (Suppl) 3322S, as well as a review by him in Immunohematology entitled 'In vitro reactions with RBCs that are not due to blood group antibodies a review', Immunohematolgy 1998;14:1-11 (PDF). These publications discuss a strange phenomenon where positive DATs and IATs (IgG) are found with RBCs/serum from clotted blood but not anticoagulated blood (RBCs/plasma). There is an association with ulcerative colitis, but the cause is still unknown. Plasma can be made to react by adding any serine proteases to the plasma. Garratty et al have found no blood group specificity associated with any of a large number of cases, but Tregellis et al (Transfusion 1979;19:650) found one serum to show anti-Sc1 specificity. There appears to be no in vivo significance regarding blood transfusion or HDN (i.e. we have transfused patients where the plasma is giving non-reactive compatibility testing and the serum is reactive; there were no signs of HTRs; we have also seen this phenomenon in pregnancy, with no clinical effects on the baby).

ADDENDA Oct. 14, 2002

6. A transfusion medicine physician from the midwest comments that with regard to the serum-reactive, but plasma non-reactive serology, there are some proteins, more usually enzymes, which are dependent on ionized calcium or magnesium for either conformation or activity. This would be unlikely for an antibody, but the issue could be easily checked. Just take a small volume of the serum, and add EDTA in an amount proportionate to the original EDTA plasma collection tube. Now repeat the PEG test using the EDTA serum.

7. A reference laboratory technologist in California wrote that the different serological results with serum versus plasma may be due to an antibody to 'activated clotting factor(s)'. She is aware of the publications referenced by other responding blood bankers above, and reports having seen this phenomenon in a 'healthy' donor, the antibody reacting 3+ when serum was used and negative with plasma, tested by LISS IgG tube test.

Please submit comments to the e-Network Forum.

Ira A. Shulman, MD
CBBS e-Network Forum Editor & Moderator