Management of Blood Donors and Blood Donations From Individuals

Found to Have a Positive Direct Antiglobulin Test

Judith L. Hannon

The medical literature is replete with articles addressing evidence from Israel suggests that the finding of a positive
the diagnosis and management of patients with a positive DAT in a blood donor may not be as benign as previously
direct antiglobulin test (DAT). However, there is scant thought. Therefore, it may be prudent for blood collection
information addressing the management of blood donors agencies to periodically reexamine their approach to the
and blood donations found to have a positive DAT. management of blood donors with a positive DAT and
Practices vary considerably between countries and blood their donations. This article reviews the available litera-
suppliers within countries, and there is no standardized ture and explores options for the management of DAT-
approach to the management of these blood donors or the positive blood donors and their blood donations.
blood products prepared from their donations. Recent © 2012 Elsevier Inc. All rights reserved.

ONORS WITH A positive direct antiglobulin these authors used the antiglobulin test to detect Rh
D test (DAT or direct Coombs test) may be
identified at the time of blood donor screening or
antibodies on the RBCs of infants having hemolytic
disease of the fetus/newborn. [3] The same year,
when a red blood cell (RBC) unit is returned to Boorman et al [4] demonstrated the presence of
the blood supplier because of cross-matching autoantibodies on the RBCs of patients with
difficulties. The DAT detects the presence of acquired hemolytic anemia. The test used to detect
immunoglobulin or complement on the surface of binding of antibodies or complement to the RBCs
RBCs. Antihuman globulin (AHG or Coombs in vivo became known as the DAT (DAT or direct
serum) is a commercial reagent that contains Coombs test).
antibody to human immunoglobulin and/or com-
plement. These reagents are derived either from THE INCIDENCE OF POSITIVE DATS IN
animal (rabbit) or monoclonal sources. Monoclo- HEALTHY INDIVIDUALS
nal reagents are epitope specific as compared with In 1965, Weiner [5] first described the occur-
polyspecific reagents that have broad reactivity. In rence of a positive DAT in hematologically healthy
performing the DAT, AHG is added to a washed individuals in the United Kingdom. Of approxi-
suspension of test RBCs, and if immunoglobulin mately 60 000 healthy donors screened, 21 were
and/or complement is present on the cells, AHG found to have a positive DAT for an overall
binds to the molecules on the cell surface, incidence of 1:5000. Strength of reactivity varied: 3
bridging the gap between the RBCs causing to 4+ (12), 2+ (5), 1+ (1), and less than 1+ (3). The
visible agglutination. Monospecific AHG can be incidence of strongly positive DATs in healthy
used to determine the type of immunoglobulin or individuals was 1:3300 in this study. The human
complement component coating the RBCs. antiglobulin reagent used for testing contained
The antiglobulin test was first described by mainly anti-immunoglobin G (IgG) specificity,
Coombs et al [1,2] in 1945 as a test to detect and, as expected, all of these individuals had IgG
nonagglutinating antibodies in serum. A year later, coating their RBCs.
In 1980, Habibi et al [6] in France studied
From the Canadian Blood Services, Edmonton, Alberta, 892 000 blood donors between the ages of 20 and
Canada. 60 years, of which 69 (0.008%) were found to have
Address reprint requests to Judith L. Hannon, MD, FRCPC, a positive DAT for an incidence of 1:13 000. Of
Canadian Blood Services, 8249-114 St, Edmonton, Alberta, these individuals, 97% had IgG coating their RBCs,
Canada T6G 2R8. and 25% were receiving the drug alpha methyl
E-mail: judy.hannon@blood.ca
0887-7963/$ - see front matter dopa, which is known to cause a positive DAT and
© 2012 Elsevier Inc. All rights reserved. a clinical picture mimicking an idiopathic warm
doi:10.1016/j.tmrv.2011.08.004 autoimmune hemolytic anemia (AIHA). Forty-one

142 Transfusion Medicine Reviews, Vol 26, No 2 (April), 2012: pp 142-152

BLOOD DONORS AND DONATIONS FROM DAT-POSITIVE INDIVIDUALS 143

percent had various other antitissue antibodies increasing incidence of positive DATs with age
identified on laboratory testing. Habibi et al was also noted in this study.
followed up 63 of these donors for 5 years. Ten Therefore, in various studies, the incidence of a
percent of these individuals had a confirmed mild positive DAT in the blood donor population ranges
anemia. Seventy-two percent had biochemical from 1 in 1000 [8] to 1 in 14 000 [7]. The incidence
evidence of hemolysis by laboratory investigations varies between reports likely because of differences
alone. Follow-up was considered to have been in the type of test performed and the type of data
insufficient to determine any long-term sequelae reported rather than real differences in the popula-
related to these laboratory and clinical findings. tions studied. Issitt and Anstee [11] reported that, if
Gorst et al [7], in 1980, reported 65 healthy all positive reactions of ± or greater with C3, C4, or
individuals with a positive DAT of whom 32 were IgG present on the RBCs were included, the
recalled for further investigation. The incidence of a incidence of a positive DAT may be as high as 1
positive DAT in this donor group was 1:14 000 per 1000 blood donors. In contrast, if only strongly
donations (not corrected for no. of donors). In 81% positive DATs of 3 to 4+ with IgG were considered,
of these individuals, the positive DAT persisted at 6 the incidence would be 1 in 10 000 to 1 in 25 000.
months to 18 years of follow-up. Of 65, 31 (48%) It is difficult to compare reports on the incidence
were associated with IgG sensitization, 28 (43%) of or strength of the DAT reaction or the presence of
65 with complement only, and 6 (9%) of 65 were IgG, complement, or both on the RBCs because of
not determined. The incidence of a positive DAT differences in the techniques used in various
was shown to correlate with increasing age. There laboratories. For example, gel techniques have
was no clinical or laboratory evidence of hemolysis been reported to be more sensitive than tube
in any of these individuals. At 2 years postdiscov- techniques [12], whereas flow cytometry and
ery of a positive DAT, 1 individual with strongly radioimmunoassay are more sensitive than the
positive IgG sensitization of her RBCs developed techniques used for routine testing. The use of
the sudden onset of warm AIHA. These findings polyspecific vs monospecific AHG, the specific-
indicated that the incidence of a positive DAT in ities of the AHG, and the grading practices in the
healthy individuals increases with age and that a laboratory can also impact the ability to perform
positive DAT tends to persist over time in any data comparison.
particular individual.
In 1980, Allan and Garratty [8,9,14] reported the THE SIGNIFICANCE OF A POSITIVE DAT IN A
results of screening of 1.3 million blood donors in HEALTHY INDIVIDUAL
the United States over a 5-year period. Incidence of A significant proportion of otherwise hematolo-
a positive DAT in this donor population was gically healthy individuals has been reported to
1:1000. Eighty percent of the positive DAT have a positive DAT [8,11,13-15]. The positive
reactions were of a strength less than or equal to DAT in these cases has been of a benign nature and
1+. Only 4% were of a strength 3+ or greater, for an not associated with any apparent clinically signif-
incidence of 1 in 25 000. Two thirds of the positive icant consequences. A positive DAT may occur
DATs were associated with IgG sensitization, and because of immunoglobulin or complement binding
one third demonstrated complement only on the to the RBCs in vivo or in vitro. Clinically
RBCs. This analysis was performed using a spin significant in vivo causes of a positive DAT include
tube technique, and negative results were confirmed AIHAs due either to warm- or cold-reactive
microscopically. antibodies, drug-induced positive DAT with or
In 1985, Bareford et al [10] described 26 positive without hemolytic anemia, hemolytic transfusion
DATs in UK blood donors for an incidence of reactions, hemolytic disease of the fetus/newborn,
1:7500. All of these individuals had IgG coating autoimmune disorders such as systemic lupus
their RBCs. Twelve percent had bound complement erythematosis (SLE), and certain malignant disor-
in addition to IgG. Nine of these donors were ders [16].
followed up for up to 18 years. One individual As early as 1953, Boursnell et al [17] reported
developed AIHA, 1 donor developed ulcerative uptake of radiolabeled anti-IgG onto RBCs in vitro,
colitis, 1 had mild arthritis, 5 remained well, and the and in 1967, Grob et al [18] demonstrated that this
outcome for 1 donor was not reported. An antibody was binding specifically to RBC-bound
144 JUDITH L. HANNON

IgG. In 1983, Stratton described 22 healthy donors is a low level of complement activation in healthy
with less than 1000 molecules of IgG1 or IgG4 on individuals. Complement activation may play a
the RBCs. The finding of IgG on the RBCs of role in the normal removal of immune complexes
healthy individuals has subsequently been con- and foreign proteins [28]. The level of comple-
firmed by numerous investigators. All healthy ment activation may be increased in infection,
individuals have some IgG on the surface of their liver disease, or autoimmune disease.
RBCs. Whether this IgG is confined to a small The incidence of AIHA in the population at large
population of RBCs or is evenly distributed over is variously reported to be in the range of 1 in a
the RBC population is not clear. It does appear that million blood donations (Gorst et al [7]), 1 in 5
the amount of bound IgG is greater on older RBCs million blood donors (Bareford et al [10]), and 1 in
than on their younger counterparts [19]. There is 80 000 of the general population in the United
evidence that IgG on the surface of RBCs may be States [8]. These estimates pertain to individuals
involved in the normal process of RBC senescence with clinical evidence of a hemolytic process.
[20-23]. Recognition of IgG on the surface of RBCs Habibi et al [6], on the other hand, aggressively
may be a mechanism for macrophage recognition pursued laboratory evidence of hemolysis in his
and removal of RBCs from the circulation at the cohort of healthy individuals with positive DATs
end of its life span. Whether cell bound IgG and reported an incidence of hemolytic anemia in
represents a physiologic autoantibody or cytophilic this group as 1:24 000. Issitt and Anstee [11]
binding of immunoglobulin to the surface of RBCs reported that, of blood donors with a positive DAT
is uncertain. and IgG coating the RBCs, 5% to 10% will develop
The amount of IgG and complement detectable AIHA, 20% to 25% will become DAT negative
on the surface of the RBC will vary depending on over time, and 60% to 70% will remain DAT
the sensitivity of the technique used for detection positive but hematologically normal. Of individuals
[24]. Using a sensitive quantitative radioimmuno- with positive DATs, Garratty [8] found that about
assay with a radiolabeled monoclonal antibody two thirds of individuals had IgG coating the
against a C3d determinant found on all forms of RBCs, of which about half had IgG only and the
cell-bound C3, Merry et al [25] showed that RBCs other half IgG plus complement. The remaining
from healthy individuals carry approximately 420 ± one third of individuals had complement only coating
140 C3 molecules per RBC. Garratty [8] states that the RBCs.
all RBCs have IgG present on their surface, Positive DATs may also occur as an in vitro
generally less than 50 molecules per RBC, a level phenomenon reflecting nonspecific uptake of
undetectable with routine antiglobulin testing plasma IgG during RBC storage. This can be
(lower limit of detection, 100 molecules per reversed by washing and suspending the donor
RBC). Other investigators have shown that RBCs RBCs in pH 6.8 saline or Alsever's solution, and
from healthy donors carry approximately 31 IgG the RBCs can then be used successfully for
molecules per cell. JeJe et al [26] showed that RBCs cross-matching. The incidence of false-positive
from healthy individuals had 3.7 to 16 fg IgG per DATs in vitro is reported to be as high as 91%
10 3 RBCs with a mean of 14 fg per 10 3 RBCs. after 24 days of storage. Some studies have
Most healthy individuals with a positive DAT suggested an increased frequency of positive
do not show clinical or laboratory evidence of DATs because of detection of weak, low affinity
hemolysis, and the strength of the DAT is not autoantibodies using more sensitive gel and
necessarily an indicator of the presence or severity microcolumn techniques [15,29,30].
of hemolysis. Garratty and Nance [27], using flow
cytometry techniques, demonstrated considerable CLINICAL SIGNIFICANCE OF A POSITIVE DAT IN
overlap in the range of fluorescence of IgG-coated A HOSPITALIZED PATIENT
RBCs from patients with and without immune Approximately 8% of hospitalized patients are
hemolysis and concluded that one cannot differ- reported to have a positive DAT usually (80%)
entiate between these groups based on the levels because of complement only coating their RBCs
of IgG binding on the RBCs. Increased comple- [11,13-15,28,29,31,32]. Some authors report a
ment activation may result in a positive DAT higher incidence of positive DAT in hospitalized
because of complement binding on RBCs. There individuals with up to 15% of in-hospital patients
BLOOD DONORS AND DONATIONS FROM DAT-POSITIVE INDIVIDUALS 145

having a positive DAT in the absence of incidence of positive DATs in AIDS patients of
hemolysis [16]. The DAT in these cases is usually 18%, again mostly in association with IgG sensiti-
weak or 1+. About 1% to 2% of hospitalized zation. Therefore, there appears to be an association
patients have a weak IgG coating on their RBCs between the finding of a positive DAT and
in the absence of hemolysis. Using sensitive autoimmune disease, hemoglobinopathy, and ma-
methods for detection, virtually all normal RBCs lignancy, both nonhematologic and hematologic, in
carry some complement components on their surface, patient populations that have been examined.
usually below the level of detectability of a routine
DAT [11,13-15,28]. Many clinical disorders result in CLINICAL SIGNIFICANCE OF A POSITIVE DAT IN
in vivo complement activation, so it is not surprising A BLOOD DONOR
that hospitalized patients have an increased incidence Over 25 years of follow-up of blood donors with
of positive DATs. Certain drugs may cause comple- a positive DAT in London, Ontario, only 2
ment binding to RBCs. Recent studies suggest that examples were found where notification of a
modifications to commercial AHG, with a lower positive DAT led to a diagnosis. In both cases,
amount of anti-C3d activity, have reduced the the donors had IgG myeloma previously undetected
incidence of positive DATs in hospital populations (personal communication, Dr R. Barr, 2002). In
to closer to 1% to 3.5%. In a patient with hemolysis, a over 20 years of donor follow-up in Sudbury,
positive DAT has a high positive predictive value for Ontario, there were no known cases where a donor
an immune etiology, but the positive predictive value with a positive DAT developed a serious medical
of the DAT is poor (1.3%) in an unselected group condition such as AIHA (personal communication,
such as the blood donor population [8,33]. Dr T. Ciszewski, 2003). In Hamilton, Ontario, a
Individuals with hypergammaglobulinemia have DAT (IgG)–positive incidence of approximately 1
significantly increased nonspecific uptake of immu- in 3000 was found, and there were no known cases
noglobulin molecules onto the RBC surface and of anemia related to autoimmune hemolysis
often have a positive DAT [34,35]. The finding of a reported over a 10-year period of observation
positive DAT is not uncommon in individuals (personal communication, Dr M. Blajchman,
receiving intravenous immune globulin (IVIG), for 2008). However, a caveat in the interpretation of
example. Hypergammaglobulinemia is reported to be this information is the lack of systematic follow-up
associated with an increased incidence of positive of the donors with a positive DAT.
DAT in the absence of hemolysis. Up to 50% of such Published data and anecdotal reports suggest that
individuals may have a positive DAT, and eluates the risk of a healthy blood donor with a positive
prepared from the RBCs are usually nonreactive. DAT in the absence of underlying clinical symp-
Positive DAT can be associated with antipho- toms progressing to clinically significant disease is
spholipid (APL) antibodies in healthy individuals. very small. Other than the study by Habibi et al [6],
Of 474 545 blood donors in the west of Scotland the literature suggests that hemolysis, either overt or
from 1991 through 1993, 42 (0.009%) donors had a subclinical, is unusual in an otherwise healthy
positive DAT of which 3 (7%) had a positive DAT blood donor despite the finding of a positive DAT.
in association with a false-positive VDRL (Venereal However, the robustness and length of follow-up in
Disease Research Laboratory) test. All of these most studies has been insufficient to detect subtle
individuals had circulating APL antibodies [36]. subclinical and longer term adverse outcomes.
There is a known association of APL antibodies In December 2007, Rottenberg et al [40] from
with SLE. Hadassah University Hospital, Israel, presented an
In 1968, Bohnen et al [37] reported the incidence abstract at the American Society of Hematology
of selected diseases in a population of patients with Annual Meeting in which they provided evidence to
positive DATs. Seventeen percent had SLE; 11%, a support the hypothesis that a positive DAT may be a
malignant tumor; 13%, leukemia or lymphoma; 8%, risk marker (as opposed to a risk factor) for a
a hemoglobinopathy; 7%, AIHA; and 5%, a benign malignant disease. In May 2009, an article was
tumor. Lau et al [38] in 1976 reported that 22% of published by the same group of investigators
their study population with positive DATs and IgG providing detailed data that expanded upon the
on their RBCs had an immunologic disease; 17%, a findings of the preliminary report [41]. Donors were
malignancy; and 5%, SLE. Toy et al [39] reported an universally screened, including an autocontrol (AC)
146 JUDITH L. HANNON

using an automated analyzer (Autogrouper 1986- effort was made to exclude malignancies that might
2000 [Siemens Medical Solutions Diagnostics, have been present but undiagnosed at the time of
Tarrytown, NY]; Olympus PK7200 [Olympus, blood donation by excluding cases presenting
Melville, NY] after 2000). Donors with a positive within 12 months of blood donation, and the
AC were tested by DAT using a tube and/or Diamed increased cancer incidence remained statistically
gel card technique with anti-IgG, anti-C3, and anti- significant. The authors further noted that a positive
IgG/C3. The DAT-positive donors were matched to DAT can predate clinical detection of a cancer by
a control group of DAT-negative donors for sex, months or years, and this is particularly true of
year of donation, and age (by 5-year categories). certain lymphomas and plasma cell dyscrasias
Cancer incidence was ascertained through the Israel where an abnormal protein may be present for an
Cancer Registry. Cancer rates in DAT-positive extended period before the onset of the overt
donors were compared with expected rates in the illness. It was also interesting to note in this study
general Israeli population. The study population that there was a borderline but significantly
included 586 DAT+ donors and 2344 DAT− increased risk for solid tumors, particularly breast
donors, of which 62% were male in both groups. cancer, in the DAT-negative donor population as
The mean age was 34.5 years (DAT+) and 32.0 compared with the general population studied. The
years (DAT−). Donors were matched for year of reason for this is not clear, but the authors speculate
donation. Malignancies occurred in 17 (2.9%) of the that this may reflect better health screening and
DAT-positive donors and in 27 (1.2%) of the DAT- lifestyle choices by the blood donor population in
negative donors during the follow-up period. general. The above data raise questions as to
Excluding 3 individuals in the DAT-positive whether blood donors with a positive DAT should
group who were diagnosed with hematopoietic be allowed to continue donating blood. In addition,
malignancies within 12 months of the finding of a one must also consider the ethical obligation of a
positive DAT (Hodgkin disease within 2 months, health care agency to provide individuals with
non-Hodgkin lymphoma within 4 months, and information that may have relevance for their
multiple myeloma within 8 months) and with a underlying health.
mean follow-up period of 66 months (range, 14-91
months), the relative risk of developing a malignancy RISK TO A TRANSFUSION RECIPIENT OF
was 2.14 (95% confidence interval [CI], 1.13-4.10) RECEIVING BLOOD COMPONENTS FROM A
in blood donors with a positive DAT as compared DONOR WITH A POSITIVE DAT
with the DAT-negative group. The relative risk for a Evidence indicates that there is no immediate
hematological malignancy was 8.3 (95% CI, 1.5- harm to a transfusion recipient in receiving RBCs
43.2). Comparing DAT-positive donors with the from a donor with a positive DAT if cross-matching
general population, standardized incidence rates can be done successfully. In fact, facilities using an
(observed/expected cases) were elevated at 2.11 immediate spin or a computerized cross-match
(95% CI, 1.15-3.54; P = .16) for all malignancies and would not detect this abnormality before transfu-
8.03 (95% CI, 2.2-20.6; P = .003) for hematologic sion, and so these units (including some with a
malignancies. Malignancies included 2 cases each of strongly positive DAT) are unknowingly transfused
lymphoma, myeloma, thyroid, prostate, and gastro- from time to time in most facilities [24]. Based on
intestinal cancer. The incidence of malignancy was published data and clinical experience, there is little
not increased in the DAT-negative donor group as reason to suspect that RBCs weakly coated with IgG
compared with the general population [41]. will have a decreased posttransfusion survival [29].
The authors concluded that even with the Previous studies have suggested that the risk of a
relatively short follow-up period of 5.5 years in donor with a positive DAT progressing to a serious
this study, the finding of a positive DAT in an clinical illness is small, probably less than 5% of the
apparently healthy blood donor may be a risk donors who are found to have this abnormal
marker for a malignancy, although this represents a laboratory result. However, most studies have not
small absolute risk as compared with DAT-negative followed up these donors for long enough or closely
individuals. In particular, the findings supported an enough to identify long-term adverse effects related
increased incidence of hematologic malignancies in to this laboratory finding. As discussed above, new
the follow-up of DAT-positive blood donors. An evidence, albeit preliminary, suggests that the
BLOOD DONORS AND DONATIONS FROM DAT-POSITIVE INDIVIDUALS 147

finding of a positive DAT in a blood donor may be sample if they receive a customer complaint
of more significance than formerly recognized. regarding difficulty in the cross-match, a positive
This raises the question as to whether a blood weak D test occurs on donor Rh testing, a positive
donor who may be in the early stages of a malignant manual Rh control occurs on manual repeat RhD
process can actually transmit the illness to a testing in the donor testing facility, a previous
transfusion recipient. Early reports suggested an history of a positive DAT is present in the donor
increased risk of cancer, particularly non-Hodgkin record, or a positive AC is encountered during an
lymphoma, among transfused patients. Immune antibody investigation.
modulation, transmission of factors that encourage AABB Standards for Blood Banks and Transfu-
cancer development, and engraftment of donor sion Services, 26th Ed, 2009, Section 5.2.4 [44]
RBCs were postulated as contributing to this states that “the medical director shall establish the
outcome. The question of transmission of malig- means to notify all donors (including autologous
nant disease through transfusion was recently donors) of any medically significant abnormality
reviewed by Yang et al [42]. These authors reported detected during the pre-donation evaluation or as a
that “transmission of malignancy via blood trans- result of laboratory testing or recipient follow-up. In
fusion has never been established.” They performed the case of autologous donors, the referring
a comprehensive search of the literature in June physician shall also be notified. Appropriate
2009 and were unable to identify a single education, counselling and referral shall be offered.”
convincing report of cancer transmission through Guidelines for the Blood Transfusion Services in the
allogeneic blood transfusion. This is in contrast to UK (2005), Sections 13.12 to 13.14 [45] state that
reports in the literature of transmission of malig- “non-red cell components may be prepared and
nancy and autoimmune illness through transplan- issued from DAT positive donations” and “red cell
tation of bone marrow and peripheral hematopoetic units may be prepared and issued from DAT
stem cells as well as solid organ transplantation. positive RBCs provided that the ABO groups are
Edgren et al [43], in a retrospective cohort study of confirmed and RBC antibodies have been excluded
12 012 Scandinavian blood recipients whose as per the mandatory antibody screen. Donors who
donors developed cancer within 5 years of blood have been found incidentally to have a positive DAT
donation, found no difference in cancer incidence in at donation testing may remain as blood donors
recipients of blood from these donors as compared provided they continue to pass the health screening
with recipients receiving blood from “noncancer- questionnaire and have a normal haemoglobin.”
ous” donors. Even with tumors known to dissem- This regulation does not address donor notification
inate through a hematogenous route, there was no or follow-up. As for Canadian Standards, neither
identifiable increased risk. These authors concluded CAN/CSA Z902-2010 Canadian Standards Associ-
that “their data provided no support that blood ation: Blood and Blood Components [46] nor the
transfusions from precancerous blood donors are Canadian Society for Transfusion Medicine, Stan-
associated with an increased risk of cancer among dards for Hospitals and Transfusion Services v.2
recipients compared with transfusions from ‘non- 2007 [47] requires that a DAT be performed on
cancerous’ donors.” These studies are reassuring in donor units nor do they provide guidance on donor
that even if a blood donor presenting with a positive follow-up in the event of a positive DAT result.
DAT does go on to develop a malignant disease, the Internationally, there is considerable variability
risk of transmission to a blood transfusion recipient in the manner in which donors with a positive DAT
is exceedingly small and, in fact, must be and their donations are managed. In most countries,
considered only theoretical at this time. there is no established national policy for deferral of
DAT-positive donors or referral to a physician for
REGULATIONS AND INTERNATIONAL follow-up. Because of a lack of clear direction,
STANDARDS PERTAINING TO THE most donor collection facilities establish local in-
PERFORMANCE OF A DAT ON BLOOD DONORS house policies. In some cases, facilities choose to
Review of regulatory requirements indicates that defer on the first positive DAT to avoid the logistics
performance of a DAT is not required as a test of of obtaining follow-up samples. In other situations,
record for blood donation in most jurisdictions. donors are allowed to donate on 1 or more
Most collection facilities perform a DAT on a donor subsequent occasions and are deferred if the DAT
148 JUDITH L. HANNON

remains consistently positive. In most programs, if abnormal donations, including those with a positive
the donor is deferred, donor notification is under- DAT, to be either discarded or issued for research,
taken, the most common approach being for the medical teaching, or clinical laboratories for
blood collection facility is to send a letter to the procedure development and optimization. Further-
donor advising them to seek follow-up with their more, the standard operating procedure required that
personal physician. In some jurisdictions, the donor the names of donors demonstrating a serological
is notified and deferred if the DAT is positive with abnormality be referred to the blood center physi-
IgG coating the RBCs on a first donation, and this cian for follow-up. A survey of CBS centers carried
may depend on the strength of reactivity (≥2+). out in 2002 showed that each CBS center had its
Reinstatement is an option in some countries if a own policies and donor notification procedures for
donor is DAT negative on follow-up testing by their follow-up and management of these donors.
own physician. In 2009, in an effort to introduce some
consistency to the management of these donors/
CANADIAN BLOOD SERVICES REVIEW OF donations, CBS undertook a detailed review of the
THE MANAGEMENT OF DONORS WITH A processes for managing donors with a positive
POSITIVE DAT DAT. Between January 1 and December 31, 2007,
Routine donor testing is performed in 3 region- 1678 DATs were performed in the national CBS
alized Canadian Blood Services (CBS) Donor DTLs on 973 007 donations. This represented
Testing Laboratories (DTLs) across the country 0.17% of all donations and included repeat testing
that use standardized testing methodology. Before of donors previously coded by CBS for a positive
2009, monoclonal blend anti-D and a polyclonal DAT. Three hundred seventy-nine DAT-positive
(slide and tube) reagent were used for RhD typing. blood donor samples were identified, of which
Direct antiglobulin test was performed if a positive approximately 8% (33/379) had been reported to
Rh control occurred during donor testing or if a CBS as customer complaints relating to difficulty in
discrepancy was found on automated RhD typing cross-matching the donor RBC units (Table 1).
of a donor. Direct antiglobulin test was also Data relating to monospecific DAT testing of
performed in the event of a positive antibody these donor samples were not available because
screen, in which case performance of a panel these test results were not entered into the donor
routinely included an AC. If the AC was positive, a laboratory test result database at that time. Based on
DAT was performed using a polyspecific AHG the literature, however, it was estimated that one
(anti-IgG and anti-C3d specificity). Monospecific half to two thirds of these individuals (or ∼190-253
DAT was only performed if the DAT was positive cases in Canada in 2007) would have had a positive
using the polyspecific AHG reagent. A DAT was DAT associated with IgG with or without
also performed on a donor previously found to be
DAT positive based on a test pending code applied
to the donor record. Finally, donors were investi- Table 1. Positive DATs Identified by the 13 CBS Centers in 2007

gated if an RBC product was returned from a Total Collections from January 1 to December 31, 2007, n = 973 007

hospital because of cross-matching difficulties. Collection DAT Positive DAT Positive Reported
Site at CBS by Hospital Site Total
The components prepared from a blood donation
by a donor with a positive DAT could not be A 22 8 30
B 28 0 28
released through the CBS Laboratory Information
C 35 1 36
System and were therefore discarded. Autologous D 44 4 48
donations were quarantined and could be issued E 32 1 33
with approval of the blood center physician and F 23 1 24
the use of a deviation procedure. A code was G 15 0 15
H 13 2 15
appended to the donor record to trigger retesting
I 22 1 23
on a subsequent visit, which was not considered J 26 1 27
optimal because the donor frequently did not return K 31 12 43
to donate. L 28 1 29
The CBS standard operating procedures included M 27 1 28
Subtotal 346 33 Total DAT, 379
a requirement for all components of serologically
BLOOD DONORS AND DONATIONS FROM DAT-POSITIVE INDIVIDUALS 149

complement sensitization. The remainder would Table 3. Results of DATs in Relationship to the Reason for
Performing the DAT in DTL 1, December 1–31, 2007
have had complement only coating their RBCs. The
DAT DAT
literature suggests that complement only coating Reason for DAT Testing Positive Negative
the RBCs is of little clinical significance and is
Testing because of Rh control positive 1 0
often transitory in nature and may be related to an
Testing because of positive manual weak D 0 18
underlying clinical condition. test (PK “DU” results)
The CBS operates 3 DTLs across Canada: DTLs Testing because of positive weak D test 0 43
1, 2, and 3. Between December 1 and December 31, (PK “RH???” results)
2007, DTL 1 performed 77 DATs on 34 832 Testing because of DAT POS history 10 0
Testing because of current antibody screen 3 0
samples (0.22% of samples tested) of which 14
positive and positive DAT history
(18%) were found to be positive and 63 (82%) were Testing because of current antibody screen 0 1
negative. During the same timeframe, DTL 2 positive and autocontrol positive
performed 54 DATs on 31 284 donations of Previous unidentified antibody: no previous 0 1
which 13 (24%) were positive and 41 (76%) were DAT history
Total 14 63
negative. Donor testing laboratory 3 performed 10
DATs on 9102 donations of which 7 (70%) proved
to be positive and 3 (30%) tested negative (Table 2).
The reason for the higher percentage of positive unidentified serum antibody with no report of a
DATs in DTL 3 was not clear, but one may positive DAT. Both of these individuals tested
speculate that more of these individuals were tested DAT negative on follow-up testing.
because of the historical finding of a positive DAT, Based on the above, the incidence of a positive
a situation that seems to correlate with a high rate of DAT in the Canadian blood donor population was
continued positivity (Table 3). It is unlikely that this considered to be very low, and it was further noted
finding related to methodological differences as all that a donor found to have a positive DAT was likely
CBS donor laboratories were using standardized to remain DAT positive on subsequent testing.
procedures for performing the DAT at the time. After this survey, new reagents were imple-
The DAT was most commonly performed for mented in the DTLs that markedly reduced the
resolution of discrepant or unclear results on number of donations producing unclear results on
automated testing in DTLs 1 and 2. Most of the the automated instruments in use in these
samples tested on this basis were negative on laboratories. Donor testing procedures were mod-
follow-up DAT testing. Ten donors in DTL 1 were ified in June 2010 to replace polyspecific AHG
tested because of a history of a positive DAT and with monospecific anti-IgG AHG for routine
continued to be DAT positive. Three individuals DAT. This procedure allowed the detection of
with a currently positive antibody screen and a donors with IgG alone or IgG and complement
history of a positive DAT were also found to be coating their RBCs. Donors with a positive DAT
DAT positive (Table 3). Therefore, 13 of 13 or were deferred without the opportunity for rein-
100% of donors retested based on a previously statement after 3 DAT positive donations. Donors
positive DAT remained DAT positive on repeat were counseled to see their personal physician for
testing. One individual was tested because of a follow-up. Components from DAT-positive dona-
currently positive antibody screen with a positive tions were discarded. A decision was also made
AC, and another was tested because of a previously not to recall components from previous donations.
Directed donations were to be managed as for
allogeneic donations, and autologous units could
Table 2. Number of DATs Performed on CBS Blood Donors CBS be issued for transfusion at the discretion of the
in its 3 DTLs, December 1–31, 2007
center medical director in consultation with the
No. of DATs treating physician.
No. of Units Performed No. of No. of
Collected/ (% of collections), Positive DATs, Negative After these measures, the number of donations
Tested n (%) n (%) DATs (%) requiring a DAT decreased substantially from an
DTL 1 34 832 77 (0.22) 14 (18) 63 (82) average of 310 per month in 2006 to an average of
DTL 2 31 284 54 (0.17) 13 (24) 41 (76) 46 per month in 2010, an average reduction in test
DTL 3 9102 10 (0.11) 7 (70) 3 (30)
volume of approximately 85%.
150 JUDITH L. HANNON

Because only donors with IgG coating the Varied practices between blood collection agen-
RBCs were deferred, donor loss was minimized. cies also create potential risk management issues
This approach was considered justifiable based on if the policies in one facility are more stringent
observations in the literature that the finding of than in others. If donors attend at more than one
complement alone on the RBCs is common and is collection site, the lack of a consistent policy may
not known to be associated with clinically lead to confusion for blood donors. Therefore, a
significant sequelae. It was, however, considered standardized approach to the management of
prudent to permanently defer donors with persis- these donors has the potential to avert risk-
tence of IgG on their RBCs, with or without management concerns for individual blood centers
complement, as up to 10% of such donors may be while improving the donation experience for
at risk for progression to clinically significant donors by introducing consistency in donor
disease. This approach ensures that donors most at screening and ensuring that donors are not
risk for significant autoimmune disease or malig- deferred unnecessarily. Various approaches can
nancy down the road and who are excluded from be taken.
blood donation are deferred and referred to their A moderate approach to the management of these
physician for assessment and follow-up. It must be donors/donations is to defer only those donors with
recognized that some donors not at risk for clinical a positive DAT and IgG (with or without
illness are also deferred unnecessarily. However, complement) on the surface of the RBCs. These
the number of these donors who could be returned individuals should be referred to their personal
to the donor pool was considered low based on the physician for follow-up. Reinstatement of such
fact that most individuals with a persistent positive donors could occur according to a reinstatement
DAT because of IgG coating the RBCs remained protocol, if the DAT becomes negative at a
DAT positive on subsequent testing and therefore predetermined interval after the initial DAT posi-
ineligible to donate. tive result, and the donor meets other donor
It is recognized that allowing individuals with eligibility criteria. Components from the index
complement only on the RBCs to donate does donation should be discarded. Individuals could be
present a small future risk in that new evidence allowed to donate subsequently, but if the donation
may find that such donors are at risk for the remains DAT positive, deferral and referral to a
development of a clinically significant illness physician would be prudent. The benefit of this
given an extended period of follow-up (years). approach is that donor loss is minimized and that
However, this must be considered only a theoret- donor referral to a physician for follow-up would be
ical risk at present, as this outcome has not been a consistent recommendation.
documented in numerous studies reported in the However, as a small proportion of individuals
literature to date. with a positive DAT because of IgG on the RBCs
have been reported to evolve to a clinically
CONSIDERATIONS IN DEVELOPING A POLICY significant illness over several years of follow-up
FOR THE MANAGEMENT OF BLOOD DONORS (5%-10% depending on the study), reinstatement
AND BLOOD DONATIONS FOUND TO HAVE A may not be a desirable option. The literature
POSITIVE DAT suggests that approximately 60% to 70% of
Currently, most blood suppliers assess the individuals with a positive DAT will remain
suitability of DAT-positive donations and defer positive on follow-up testing. Therefore, very few
and/or refer blood donors with a positive DAT of these donors would be eligible for reinstatement.
based on site-specific policies. This approach has Canadian data, discussed previously, suggest that
been convenient because it allows collecting the number of donors who remain positive may be
facilities to determine their own procedures and higher than 70% and closer to 100%. A donor
avoids costly modifications to laboratory infor- reinstatement protocol requires resources, and there
mation systems that may be required to accom- may be implications for donor center computer
modate changes in deferral policies for these records. With these considerations, the resources
individuals. However, this approach may result in required for follow-up and reinstatement of blood
the unnecessary deferral of suitable donors donors in the context of a positive DAT may be
depending on the procedures in use in the facility. more effectively used.
BLOOD DONORS AND DONATIONS FROM DAT-POSITIVE INDIVIDUALS 151

Ethical and legal implications must also be CONCLUSIONS

considered, as once the facility becomes aware of Recently published data suggest that blood sup-
a test result that could impact the donor's or the pliers should continue to revisit their protocols for
recipient's health, it may create a responsibility to managing DAT-positive donors and should closely
act on that information. One must be certain that the monitor the literature to determine in a timely manner
finding of a positive DAT is not significant in terms whether the finding of a positive DAT on a donor unit
of the donor's or the recipient's health, either may indicate a risk to either donor or recipient.
present or future, if one is to allow the donor to Regular review of procedures used in the DTL should
continue donating. The decision not to allow be undertaken, with a view to minimizing any
reinstatement of a donor with a positive DAT potential transfusion risk as well as minimizing
may offer a higher degree of safety, as the donor loss. Donors who have passed the health
appearance of a clinically significant illness in screening procedures are presumably healthy and not
association with a positive DAT may take several anemic. Donors with a positive DAT rarely have
months or years to become evident. This is clinical evidence of hemolysis. However, a small
illustrated by the recently published data from number of donors, in particular, those with persistence
Israel [40,41] that suggests that a positive DAT in a of IgG on the RBCs, may be in the early stages of an
blood donor may be a risk marker (as opposed to a autoimmune illness and may be developing a serum
risk factor) for malignancy, in particular, a protein abnormality that could presage a more serious
hematological malignancy in that individual. How- illness such as a malignancy or a hematologic
ever, this observation has not yet been confirmed by disorder. As the opportunity exists for early diagnosis
other investigators. Clearly, donor deferral becomes of a clinically significant and potentially treatable
a difficult issue under these circumstances because illness, the follow-up of blood donors with a positive
there is no valid evidence to inform the donor DAT and IgG on the RBCs is recommended by most
notification or counseling process. authors. If a donor is deferred based on a positive
The most aggressive approach would be to defer DAT, it is appropriate that they be referred to a
all DAT positive donors regardless of whether there physician for periodic assessment and follow-up.
is IgG and/or complement coating the RBCs Legal and ethical standards recognize that a
without reinstatement. All donors would be referred blood donor has the right to be provided with any
to their personal physician for follow-up. This information that may be significant for his or her
would result in maximum donor loss in a situation health and that these are also important consider-
where the literature indicates that donors with ations in determining policies for management of
complement only on the RBCs are unlikely to pose blood donors in this situation.
a risk to a transfusion recipient. This overly
cautious approach is probably not appropriate
given the current state of the knowledge, as the ACKNOWLEDGMENT
presence of complement alone on the RBCs is not The author wishes to acknowledge Dr. Dana
known to be associated with any clinically Devine for her review of the manuscript and to
significant sequelae and is a common finding in thank the staff of the CBS DTLs for providing data
the general population. to support this publication.

REFERENCES
[1] Coombs RRA, Mourant AE, Race RR. Detection of weak [5] Weiner W. “Coombs positive” “normal” people. Bibl
and “incomplete” Rh agglutinins: a new test. Lancet Haematol 1965;9:35.
1945;246:15-6. [6] Habibi B, Muller A, Lelong F, et al. Auto-immunisation
[2] Coombs RRA, Mourant AE, Race RR. A new test for the erythrocytaire dans la population “normale”. Nouv Presse
detection of weak and “incomplete” Rh agglutinins. Br J Méd 1980;9:3253-7.
Exp Pathol 1945;26:255-66. [7] Gorst DW, Rawlinson VI, Merry AH, et al. Positive direct
[3] Coombs RRA, Mourant AE, Race RR. In-vivo isosensiti- antiglobulin test in normal individuals. Vox Sang 1980;38:
zation of red cells in babies with haemolytic disease. Lancet 99-105.
1946;247:264-6. [8] Garratty G. The significance of IgG on the red cell surface.
[4] Boorman KE, Dodd BE, Loutit JF. Haemolytic icterus Transfus Med Rev 1987;1:47-57.
(acholuric jaundice) congenital and acquired. Lancet [9] Allan J, Garratty G. Positive direct antiglobulin tests in
1946;247:812-4. normal blood donors. In: Abstracts, 16th Congress of the
152 JUDITH L. HANNON

International Society of Blood Transfusion. Montreal: [29] Judd WJ. The clinical insignificance of a positive direct
Canada; 1980. antiglobulin test. In: Direct Antiglobulin Testing in the
[10] Bareford D, Lonster G, Gilks L, et al. Follow-up of normal New Millennium. Bethesda: American Association of
individuals with a positive antiglobulin test. Scand J Blood Banks; 1999.
Haematol 1985;35:348-53. [30] Petz LD, Garratty G. Immune hemolytic anemias. 2nd ed.
[11] Issitt PD, Anstee DJ. Applied blood group serology. 4th ed. Burlington (Mass): Elsevier Inc.; 2004. p. 204-8.
Canada: Montgomery Scientific Publications; 1998. [31] Freedman J. False-positive antiglobulin tests in healthy
p. 1015-20. subjects and hospital patients. J Clin Pathol 1979;32:
[12] Nathalang O, Chuansumrit A, Prayoonwiwat W, et al. 1014-8.
Comparison between the conventional tube technique and [32] Cid J, Ortin V, Excoda L, et al. The direct antiglobulin test
the gel technique in direct antiglobulin tests. Vox Sang in a hospital setting. Immunohematology 2003;19:16-8.
1997;72:169-71. [33] Kaplan HS, Garratty G. Predictive value of direct
[13] Mollison PL, Englefreit CP, Contreras M. Blood transfu- antiglobulin test results. Diagn Med 1985;8:29-33.
sion in clinical medicine. 9th ed. Oxford: Blackwell [34] Heddle NM, Kelton JG, Turchyn KL, et al. Hypergamma-
Scientific Publications; 1993. p. 295-8. globulinemia can be associated with a positive direct
[14] Simon TL, Dzik WN, Snyder EL, et al. Rossi's principles antiglobulin test, a nonreactive eluate, and no evidence of
of transfusion medicine. 3rd ed. Philadelphia: Lippincott hemolysis. Transfusion 1988;28:29-33.
Williams & Wilkins; 2002. p. 350-1. [35] Rushton L, et al. Increasing frequency of positive direct
[15] Garratty G. The clinical significance of a positive direct anti-human globulin tests in blood donors. [Abstract]
antiglobulin test. Direct antiglobulin testing in the new Transfus Med 1997;7(Suppl):22.
millennium. Bethesda: American Association of Blood [36] Win N, Islam SI, Peterkin MA, Walker ID. Positive
Banks; 1999. direct antiglobulin test due to antiphospholipid antibodies
[16] Zarandona JM, Yazer MH. The role of the Coombs test in normal healthy blood donors. Vox Sang 1997;72:
in evaluating hemolysis in adults. CMAJ 2006;174: 182-4.
305-7. [37] Bohnen RF, Ultmann JE, Gorman JG, et al. The direct
[17] Boursnell JC, Coombs RRA, Rizk V. Studies with marked Coomb's test: Its clinical significance. Ann Intern Med
antisera: Quantitative studies with antisera marked with 1968;68:19-32.
iodine-131 isotope and their corresponding antigens. [38] Lau P, Haesler WE, Wurzel HA. Positive direct antiglob-
J Biochem 1953;55:745. ulin reaction in a patient population. Am J Clin Pathol
[18] Grob PJ, Frommel D, Isliker HC, et al. Interaction of IgG 1975;65:368-75.
and its fragments with red cells. Immunology 1967;13: [39] Toy PTCY, Reid ME, Burns M. Positive direct antiglobulin
489-99. test associated with hyperglobulinemia in acquired immu-
[19] Szymanski IO, Odgren PR Fortier NL, Snyder LM. Red nodeficiency syndrome (AIDS). Am J Hematol 1985;19:
blood cell associated IgG in normal and pathologic states. 145-50.
Blood 1980;55:48-54. [40] Rottenberg Y, Yahalom V, Shinar E, Adler B, Paltiel O.
[20] Kay MMB. Mechanism for removal of senescent cells by Blood donors with a positive direct antiglobulin test are at
human macrophages in situ. Proc Natl Acad Sci U S A increased risk for hematologic malignancies and cancer in
1975;72:3521-5. general. [Abstract 2903]. ASH Annual Meeting Abstracts,
[21] Kay MMB. Role of physiologic autoantibody in the Part 1, Vol. 118; 2007. p. 854a.
removal of senescent human red cells. J Supramol Struct [41] Rottenburg Y, Yahalom V, Shinar E, et al. Blood donors
1978;9:555-67. with positive direct antiglobulin tests are at increased risk
[22] Kay MMB, Sorensen K, Wong P, et al. Antigenicity, for cancer. Transfusion 2009;49:838-42.
storage and aging: Physiologic autoantibodies to cell [42] Yang H, Lee J, Seed CR, et al. Can blood transfusion
membrane and cell membrane proteins and the senescent transmit cancer: A literature review. Transfus Med Rev
cell antigen. Mol Cell Biochem 1982;49:65-85. 2010;24:235-43.
[23] Kay MMB. Localization of senescent cell antigen on band [43] Edgren G, Hjalgrim H, Reilly M, et al. Risk of cancer
3. Proc Natl Acad Sci U S A 1984;81:5753-7. after blood transfusion from donors with subclinical
[24] Bellia M, Georgopoulos J, Twevrenis V. The investigation cancer: A retrospective cohort study. Lancet 2007;369:
of the significance of a positive direct antiglobulin test in 1724-30.
blood donors. Immunohematol 2002;18:78-81. [44] AABB standards for blood banks and transfusion services.
[25] Merry AH, Thomson EE, Rawlinson VI, et al. A 25th ed. Bethesda: America Association of Blood Banks;
quantitative antiglobulin test for IgG use in blood 2009. p. 16.
transfusion serology. Clin Lab Haematol 1982;4:393-402. [45] UK Blood Transfusion and Tissue Transplantation Ser-
[26] JeJe MO, Blajchman MA, Steeves K, et al. Quantitation of vices. Guidelines for the blood transfusion services in the
red cell–associated IgG using an immunoradiometric assay. United Kingdom. 7th ed. United Kingdom: The Stationary
Transfusion 1984;24:473-6 . Office; 2005. p. 181-2. www.transfusionguidelines.org.uk.
[27] Garratty G, Nance SJ. Correlation between in vivo hemolysis [46] Canadian Standards Association. CAN/CSA-Z902-10,
and the amount of red cell–bound IgG measured by flow standard for blood and blood components. Standards
cytometry. Transfusion 1990;30:617-21. Council of Canada; 2010. p. 40.
[28] Freedman J. The significance of complement on the red cell [47] Standards for hospital transfusion services. 2nd ed. Ottawa:
surface. Transfus Med Rev 1987;1:58-70. Canadian Society for Transfusion Medicine; 2007.