Abstract

CASE REPORT

Case 1

A 40-year-old 69-kg man who had recently undergone a second peripheral blood stem cell transplant from the same group B D+ matched sibling donor for relapsed acute myeloid leukemia had persistent thrombocytopenia 1 month posttransplant. He was scheduled for 6 units of PLTs (one-half of a double-apheresis PLT collection). Fifteen minutes and 100 mL into the transfusion, he developed severe back and flank pain associated with hypertension (blood pressure 112/72 to 150/92) and chest pain. The transfusion was discontinued; a posttransfusion specimen was noted to be grossly hemolyzed and the urine specimen was dark brown. The hemoglobin (Hb) level had decreased from 9.7 g/dL pretransfusion to 5.5 g/dL. The pre- and posttransfusion laboratory results are shown in Fig. 1. The patient was treated supportively with intravenous hydration and transfusion of 3 units of red blood cells (RBCs). Cardiac and renal function remained stable, and the patient was discharged 4 days later. When the other half of the apheresis PLT component was traced, it was discovered to have been administered 2 hours earlier to another patient (Case 2 below).

Open in a separate window

Fig. 1

Serial Hb () and total bilirubin (■) values for Case 1.

Case 2

A 5-year-old 26-kg boy with severe aplastic anemia 3 months postimmunosuppressive therapy had received the second half of the implicated apheresis PLTs. The component had been volume reduced (to 37 mL) in accordance with institutional policy for out-of-group PLTs to pediatric patients. At the completion of the transfusion, the patient had a witnessed syncopal episode after defecating. He remained minimally responsive for 1 minute and very drowsy thereafter. Computed tomography of the head did not show any intracranial hemorrhage. By the time of the index transfusion reaction (Case 1 above), the second patient had improved clinically and was awake and alert without any complaints. The pretransfusion and posttransfusion laboratory results are shown in Fig. 2. The patient had a nadir Hb level of 7.1 g/dL 9 hours after PLT transfusion (a decrease of 2.3 g/dL), received one RBC transfusion, and was discharged home 3 days later.

Open in a separate window

Fig. 2

Serial Hb () and total bilirubin (■) values for Case 2.

MATERIALS AND METHODS

RESULTS

The pretransfusion specimen from Patient 1 typed as group B D+ and had a negative DAT; the serum showed no evidence of hemolysis. The patient’s posttransfusion sample showed severe hemolysis with brown plasma and a DAT positive for C3d. The RBCs now reacted with anti-A and anti-B, but a saline control was also positive. When the RBCs were treated with dithiothreitol to prevent RBC-bound IgM interfering, the RBCs no longer reacted with anti-A and the saline control was negative. The patient’s serum reacted with group A and B RBCs, but not group O RBCs. The patient’s untreated RBCs were nonreactive on a lectin panel (data not shown), ruling out polyagglutination. An eluate demonstrated anti-B. Subsequent postreaction specimens obtained the following day(s) did not demonstrate any ABO discrepancy. The urine was grossly dark brown, had 3+ Hb on dipstick analysis, and had no RBCs on microscopic examination.

The pretransfusion specimen from Patient 2 typed as group B D+ and had a negative DAT; the serum was not hemolyzed. The posttransfusion specimen was grossly hemolyzed and typed as group B D+; a DAT was positive for both IgG and C3d. An eluate demonstrated anti-B. Urinalysis was 1+ positive for Hb without RBCs on microscopic examination. Of note, for both patients, the degree of hemolysis interfered with routine chemistry testing, and the results for several analytes were only obtainable using a sodium lithium collection tube.

DISCUSSION

To our knowledge, these cases represent the first reported cases of hemolytic transfusion reactions (HTRs) associated with a group A apheresis PLT donor. Although the true incidence of PMHTR is not known, the range is 1:1000 to 1:25,000 for all PLT transfusions and 1:100 to 1:9000 for minor-incompatible PLT transfusions.¹ In all prior reports implicated donors were group O, and recipients were usually group A with the exception of a few cases involving group AB and group B recipients.^1,2 The clinical presentations and severity of these reactions vary widely. In one recent review¹ summarizing 28 cases of PMHTR, 82% of the PLTs implicated were apheresis PLTs, the mean volume of minor-incompatible plasma infused was 350 mL, the mean saline anti-A/B titer was 2403 (range, 128 to 16,384), and the mean decrease in Hb was 5.8 g/dL. Our second index case and lookback cases underscore the fact that PMHTR can be underrecognized especially in patients receiving outpatient transfusion.

In most cases of PMHTR the cause of the high-titer anti-A or anti-B in the donor is unknown.³ Known causes include pregnancy, recent vaccination, and ingestion of high-dose live bacteria. Immunization with pneumococcal vaccine⁴ and certain toxoids⁵ including tetanus and diphtheria was shown to cause a marked rise in isoagglutinins titers, especially anti-A, attributable to A-like substance in the polysaccharide capsule of Streptococcus pneumoniae, and the presence of A-like substance in hog stomach pepsin used in toxoid production. It is not known how long these high titers persist after vaccination.

The implicated donor’s high-titer anti-B was associated with probiotic ingestion. Probiotics are dietary supplements that contain live bacteria and exert their putative benefits by preventing bowel colonization, adhesion, and invasion by pathogenic organisms and by modulating host immune response. Among healthy individuals, they are promoted to alleviate the symptoms of lactose intolerance, antibiotic-associated gastrointestinal side effects, allergy manifestations, and the common cold.^6,7 Our inhibition studies demonstrate that the bacteria in some probiotic formulations contain B-like antigens that may stimulate increased production of anti-B when ingested by subjects lacking the corresponding antigen. We postulate that our donor’s exposure to high doses of probiotic stimulated the formation of high-titer anti-B, although we do not have stored plasma that dates back more than 3 years to test our hypothesis.

The formation of anti-A and anti-B is known to result from exposure to bacterial antigens that are similar in structure to the A and B blood group antigens as the gut becomes colonized in early infancy. These antibodies develop after the first 4 to 6 months of life. Most of the bacteria expressing A, B, and H substance are ubiquitous Gram-negative bacteria that eventually colonize the normal gut. For example, Escherichia coli serotype 086:B7 has a B-like antigen on its lipopolysaccharide chain that can stimulate anti-B production. Springer and Horton⁸ exposed subjects to E. coli 086:B7, either by ingestion or by inhalation, and measured pre- and postexposure anti-B titers. One-third of healthy adults had a significant increase in the titer of anti-B, and the anti-B response was strongest in those who produced high-titer antibodies to E. coli.

In addition to directly stimulating anti-B production, probiotics may nonspecifically enhance humoral immunity. In one study, lactobacillus administered to children with acute gastroenteritis increased IgG, IgA, and IgM levels;⁹ in another study, probiotics given to infants to prevent atopy improved antibody response to Haemophilus influenzae Type b vaccination.¹⁰

Various strategies have been proposed to reduce the risk of a PMHTR (Table 1). Screening for high-titer anti-A, anti-B, or anti-A,B has been in place in the United Kingdom since 2002, although donor screening using a titer cutoff of 100 on an automated platform has not eliminated PMHTR;¹¹ a recent study using a standardized method of titration still showed variable titer results highlighting the difficulty in setting a reproducible cutoff titer.¹² Other strategies include volume reducing or washing minor-incompatible PLTs, limiting the amount of incompatible plasma transfused over 24 hours, the use of PLT storage media that replaces much of the donor plasma, and transfusion of whole blood–derived PLT concentrates in lieu of apheresis PLTs.^1,13 Many of these methods will not prevent a PMHTR if very-high-titer anti-A or anti-B is present. A volume as small as 30 to 50 mL can cause a PMHTR whether derived from 1 unit of pooled PLTs, the residual volume of plasma in a volume reduced apheresis component, or even PLTs stored in additive storage media. In our second case, 37 mL of residual plasma in a volume-reduced product was sufficient to cause hemolysis. In another report two group A patients had a PMHTR when transfused with “dry PLTs” resuspended in additive storage media.¹⁴ A recent report also found anti-A and anti-B titers in pooled PLTs to be comparable to those of apheresis PLTs.¹⁵

The incidental finding of an ABO discrepancy on the first postreaction specimen in Case 1 was unexpected. We speculate that the patient’s RBCs were so heavily coated with passive anti-B that they spontaneously agglutinated on initial testing. Such spontaneous in vitro agglutination has been described in the setting of passively acquired high-titer anti-B.¹⁶ The initial reactivity with reverse grouping B cells was similarly attributable to passively acquired anti-B from donor plasma.

In conclusion, we have demonstrated that non–group O PLT donors can have high-titer anti-B that can mediate HTRs, that PMHTR may be underrecognized, and that probiotic ingestion represents a novel mechanism of stimulating high-titer anti-B. The effects of probiotic ingestion on anti-B production in otherwise healthy adults and their clinical significance merit further investigation in a prospective study. We are currently investigating this to determine if donors should be questioned about probiotic consumption and to guide management of donors who admit to probiotic consumption.

ABBREVIATIONS

Footnotes

References