Methyldopa-induced Autoantibodies

Against Red Blood Cells

W. G. Murphy, J. G. Kelton

SUMMA R Y. Methyldopa therapy results in the formation of red cell autoantibodies in lO-20%
of patients taking the drug for longer than 4 months. These red cell antibodies are true
autoantibodies, that is they are directed against an autoantigen on the red blood cell membrane and
not against the drug or against a drug-altered antigen. The target membrane antigen is usually
within the Rhesus system, although often the antibody specificity cannot be defined. Red cell
antibody is usually detectable in the patient’s sera as well as on the red cells. The autoantibody is
usually a warm reacting IgG antibody. Most patients who develop these autoantibodies do not go
on to develop hemolytic anemia in spite of high titres of antihodies on their red cells. In addition,
these patients do not tend to develop hemolysis if methyldopa therapy is continued. Rarely patients
develop hemolytic anemia which can be severe. Differences in antibody characteristics, including
subclass restriction, complement-binding ability, or titre do not explain why some patients with
autoantibody hemolyze while most do not. One group of investigators found that hemolyzing
patients had IgM on their red cells while those who did not had IgG only. But while this
observation could explain why some patients (IgM-sensitized red cells) hemolyze, it does not
explain why most patients with IgG-sensitized red cells do not hemolyze.
Why the autoantibody forms is not known but some investigators have proposed that the drug
may directly affect B or T cells with resulting impairment of immune tolerance. It is also possible
that an impairment of reticuloendothelial ceU function described in some methyldopa patients could
explain the unique combination of immunological disorders induced by this drug.

Three major problems remain to be solved in the
story of methyldopa-induced red cell autoantibodies.
(1) How does the drug induce red cell autoantibody in
l&20% of those who take the drug for longer than 4
months?; (2) why do only a small fraction of these
patients have hemolysis while most patients have
normal red cell lifespans in spite of abnormally high
concentrations of IgG on the red cell surface?;
W. G. Murphy, Ml3 MRCP, (UK), J. G. Kelton, MD, Depart-
ments of Medicine and Pathology, McMaster University Medical
Centre and the Hamilton Centre of the Canadian Red Cross Blood
Transfusion Service, Hamilton, Ontario, Canada.
(3) why does it take so long (usually 4 months) for the
autoantibody to develop after the start of drug
therapy?
In this review we will present an overview of what is
known about methyldopa-induced autoimmune he-
molytic anemia, what theories have been proposed to
answer the questions posed previously, and what
experimental evidence has been advanced in support
of these theories.
Alpha-methyldopa (L-cl-methyl-3,4, dihydroxyphe-
nylalanine) (Fig. 1) was introduced as an antihyper-
tensive agent in 1960. It exerts its antihypertensive
effect through a central mechanism.’ With the intro-
duction of other hypotensive agents its use has de-
clined; however, it continues to be widely used. Many

Blood Reviews (1988) 2, 3-2

Q 1988 Longman Group UK Ltd

unwanted effects have been reported. These include
fever, skin lesions and diarrhea,2 as well as more
serious side effects such as autoimmune hemolytic
anemia,3 hepatitis,4 myocarditis,5 thrombocyto-
penia,6 platelet function defects,‘,’ lupus-like syn-
drome,‘*” central nervous system abnormalities,’
abnormalities of lipids and glucose” and autoanti-
bodies against factor VIII.”
Methyldopa gives rise to two distinct immunologi-
cal abnormalities of red cells. These are (1) methyl-
dopa-induced autoimmune hemolytic anemia and (2)
a positive direct antiglobulin test without hemolysis.
Although these two conditions share many similari-
ties, there is little clinical overlap--patients who do
not hemolyze at presentation usually do not develop
hemolysis later. In the following sections, these two
disorders will first be considered separately; we will
then discuss the pathophysiological mechanisms that
distinguish between them and the implications of this
distinction for our understanding of the reticuloendo-
thelial system.
Methyldopa-induced Autoimmune Hemolytic
Anemia
Incidence
Methyldopa is the drug most commonly implicated in
drug-induced immune hemolytic anemia. l3 The
chance of developing hemolytic anemia for patients
taking methyldopa has been estimated to range from
0.02% to o.8%.3.‘4,‘5
Mechanism of Hemolysis
Methyldopa induces immune hemolytic anemia via
autoimmune red cell antibodies. This is in direct
contrast to other more ‘typical’ drug-induced immune
hemolytic anemias. In these disorders an antidrug
antibody forms and the drug-antibody complex binds
directly to the red cell membrane or deposits comple-
ment on the red cell surface. Quinidine and quinine
result in drug-induced hemolysis via complement
activation. In contrast, penicillin is an example of
drug-induced immune hemolysis in which the drug
binds to the red cell membrane and antidrug IgG in
turn binds to the cell surface. Subsequently, the red
cell is cleared through Fc receptors on reticuloendo-
thelial cells. These various mechanisms of immune
hemolysis have been recently reviewed.13 In contrast
y2
CH,-y-COOH
CH3
Fig. 1 The structure of a-methyldopa.
BLOOD REVIEWS 37
to these mechanisms, the antibody of methyldopa-
induced hemolysis is not directed against the drug but
against an antigen on the red cell membrane. Other
drugs have been reported to give rise to a similar
autoimmune hemolytic anemia: L-dopa, mefenamic
acid, phenacetin, chlorpromazine, ibuprofen,13 and
procainamide.’ 3~16 The cell antibodies in methyl-
dopa-induced autoimmune hemolytic anemia do not
activate complement (described in the next section):
the IgG-coated cells are removed by the Fc receptors
of the reticuloendothelial cells, especially those in the
spleen.
Characteristics of the Red Cell Autoantibody
Methyldopa-induced hemolytic anemia is usually of
the ‘warm’ type, and almost always of the IgG class.
Rarely, exceptions occur: IgM warm-reacting comple-
ment-fixing antibodies have been reported in autoim-
mune hemolytic anemia associated with methyldopa
in two reports. “J * IgA antibodies were described in
one patient.’ Et
Several studies suggested that the subclass of IgG
involved in methyldopa-induced autoimmune hemo-
lytic anemia was limited to IgG 1 or IgG 1 and
3. 19*20,21However, such studies are dependent upon
the quality of the anti-IgG preparations used to detect
the IgG subclasses. In a more recent report, all four
subclasses of IgG were found on the red cells of
patients with the disease.”
The IgG antibodies in this disease do not deposit
complement on the red cell membrane in high concen-
trations.‘4*22 The reasons for this are uncertain. In a
majority of cases, the autoantibody is directed against
an Rh antigen on the red cell membrane (described
subsequently). Characteristically, Rh antibodies do
not activate complement. It is possible that this is due
to the low density of antigenic sites on the red cell
membraneZ3 reducing the likelihood of achieving a
sufficiently high concentration of IgG necessary for
complement deposition.
The Red Cell Membrane Target Antigen in
Methyldopa-induced Autoimmune Hemolytic Anemia
(the SpeciJicity of the Autoantibody)
In many cases the specificity of the antibody cannot
be determined. Since antibody specificity is studied by
observing the reactions of test sera with cells lacking
known antigens (e.g. Rh,,ii cells), determination of
antibody specificity depends upon having cells that
lack the target antigen. When the antigen is ‘public’,
that is, present on the cells of almost all donors in a
population, it may be virtually impossible to identify
the target antigen by these ‘negative’ methods.
Where it has been characterized, the most common
specificity of autoantibody is against the antigens in
the Rhesus system.3*20,24*25 Much less commonly
reported target antigens are Wrb,25 Jk”,26 and U.27
The proportion of cases in which the antibody is
38 METHYLDOPA-INDUCED AUTOANTIBODIES AGAINST RED BLOOD CELLS
directed against an antigen within the Rhesus system
varies from 6 out of 6” to 10 out of 33.25 In some
cases the antibody has specificity for more than one
antigen.22’28
Clinical Course of Methyldopa-induced
Autoimmune Hemolytic Anemia
Hemolysis due to methyldopa is always autoantibody
mediated, and is always extravascular in type. This
condition is relatively uncommon and it arises in less
than 1% of patients taking the drug.3 It can occur in
those taking less than 1 gm/day, but there appears to
be an increasing frequency of autoantibody formation
with increasing drug dosage.22p2g The onset is slow. It
can occur by 3 months after starting the drug2’ and
has arisen as late as 4 years after therapy began.14
The hemolysis can be severe and deaths have occur-
red.30
Treatment of Methyldopa-induced Hemolytic Anemia
Stopping of methyldopa therapy usually results in a
rapid resolution of the hemolysis;‘3 however, it can
take many months before the direct antiglobulin test
(DAT) becomes negative. Corticosteroids may speed
the recovery,3*10*14 but controlled trials are lacking
and it is not possible to know if this therapy is
beneficial. On rare occasions blood transfusions may
be required. Although no evidence is available at
present, high dose intravenous IgG should be con-
sidered in patients with severe hemolysis since the
hemolysis is mediated via the Fc receptors on macro-
phages. Consequently, blocking these receptors with
intravenous IgG could diminish red cell destruc-
tion.31 The re-introduction of methyldopa is con-
traindicated, since the autoantibody may redevelop.2g
Methyldopa-induced Positive Direct
Antiglobulin Test Without Hemolysis
Incidence
Although less than 1% of patients receiving methyl-
dopa therapy develop hemolysis, lO-20% of those
taking the drug for longer than 3-6 months will
develop a positive direct antiglobulin test, indicating
abnormally high concentrations of IgG bound to the
surface of their red cells. Of these, 60% will also have
a positive indirect antiglobulin test,32 indicating that
red cell autoantibodies are present in their serum.
Characteristics of the Red Cell Autoantibody
The IgG red cell autoantibody in those patients
without hemolysis is the same as the autoantibody in
those patients with hemolysis. In particular (a) all
four subclasses of IgG are represented; (b) the specifi-
city is the same; (c) complement binding does not
occur. Possible explanations for why some patients
hemolyze and others do not will be discussed later in
this report.
Clinical Course
Those patients who have a positive antiglobulin test
without hemolysis usually do not develop hemolysis if
the drug is continued. In fact, the autoantibody can
spontaneously disappear. 11*33
Treatment
For most patients who have methyldopa-induced
autoantibodies, but who are not hemolyzing, no
treatment is required. Usually it is acceptable to
continue the drug. However, deciding to continue
methyldopa therapy in a patient with positive direct
antiglobulin test who is not hemolyzing requires that
the following be considered: the positive indirect
antiglobulin test will interfere with crossmatching,
should the patient require blood transfusions. This
could add to the risk of the transfusion. The autoanti-
body will not shorten the life of the transfused red
cells (provided it does not hemolyze the patient’s own
cellist), but it could obscure clinically relevant allo-
antibodies in the crossmatch. However, the use of
additional techniques such as autoabsorption of the
autoantibody allows accurate crossmatching to be
performed in almost all cases.
Pathophysiology of the Immunological
Abnormalities Associated with Methyldopa
The clearance from the circulation of a particle or cell
foreign to the body can be effected by three mechan-
isms. (1) The particle is coated with immunoglobulin,
especially IgG, with or without C3b. These red cell-
bound proteins bind to specific receptors on the
reticuloendothelial cells, particularly those in the
spleen and liver; (2) the complement cascade is com-
pleted through either the classical or the alternate
pathways, resulting in the intravascular destruction of
the cells; (3) circulating reticuloendothelial cells with
receptors for the Fc portion of IgG, or cells with
natural killer activity, destroy the sensitized cells
intravascularly. Only the first of these three pathways
causes the red cell destruction in patients with methyl-
dopa-induced autoimmune hemolytic anemia.
In general, complement is not activated by the
methyldopa-induced autoantibody, and the red cells
are destroyed by the tissue reticuloendothelial cells
following the binding of the opsonized red cells to the
Fc receptors of the reticuloendothelial cells. Thus
there are two major steps in this system of immune
destruction. (1) The production of specific antibody
and, (2) the destruction of the antibody-sensitized red
cells by the reticuloendothelial cells. Abnormalities of
both of these components have been described in
patients with methyldopa-induced autoantibodies. In

the next sections we first consider the question: why
do some patients taking methyldopa develop red
blood cell antibodies? Next, we discuss theoretical
reasons why some patients hemolyze and others do
not. Last, the function of the reticuloendothelial
system in patients treated with methyldopa will be
discussed.
Why Do Some Patients Taking Methyldopa Develop
Red Cell Autoantibodies?
Several theories have been proposed to explain the
formation of methyldopa-induced red blood cell
autoantibodies. These include the following:
Altered Red Cell Membrane. It has been suggested
that methyldopa alters the red cell membrane, result-
ing in the formation of a neoantigen on the cell
surface. These antigens are the targets of the autoan-
tibody. 3,20 Such a mechanism would imply that me-
thyldopa initially binds to and subsequently alters
the red cell membrane. Although one group of inves-
tigators did report binding of methyldopa to red
cells,35 others were unable to demonstrate this phe-
nomenon.20 Another group of investigators reported
that the drugs glafenine and latamoxef induced auto-
antibody formation which developed in conjunction
with drug antibodies and that the autoantibodies
were directed against the drug-binding sites on the
red cell membrane.36 These observations were inter-
preted to support a hypothesis that drug-membrane
binding was the initial event for some drug-induced
immune hemolytic disorders. This theory does not
explain why the autoantibody persists long after
the drug has been stopped and presumably cleared
from the body. It also fails to explain why sera or
the IgG fraction alone reacts with normal red cells
without the presence of the drug. Finally, this theory
does not explain the more general autoimmune phe-
nomena associated with methyldopa where many
other cells or proteins become the target of autoanti-
bodies.
Spontaneous Formation of Autoantibodies. Methyl-
dopa could directly affect B-lymphocytes allowing the
proliferation of autoantibody-producing cell lines.37
Kirtland and coworkers demonstrated that methyl-
dopa caused a sustained elevation in cyclic-AMP
concentrations in lymphocytes3’ This effect induced
an inhibition of T-cell proliferation with diminished
suppressor cell maturation and activation. They pro-
posed that the resultant loss of suppressor function
allowed unregulated autoantibody production by
B-cells. These theories could explain many of the
serological observations. Unfortunately, other groups
of investigators were unable to confirm these
studies.39*40 Thus this most important issue remains
unresolved.
As yet no satisfactory explanation has been pro-
posed to explain why it takes over 3 months for the
autoantibody to appear.
BLOOD REVIEWS 39
Why Do Most Patients Who Develop Methyldopa-
induced Red Cell Autoantibodies not Hemolyze?
It is far more likely for a patient taking methyldopa to
form red cell autoantibodies and not to hemolyze,
than to form the autoantibodies and to hemolyze.
Approximately 1@20% of patients who receive me-
thyldopa for longer than 3-6 months will develop a
positive direct antiglobulin test. But, as noted pre-
viously, less than 1% of these patients will have
clinical or laboratory evidence of increased red cell
destruction. This issue is not just of biological in-
terest: the understanding of why some patients have
antibody sensitizing their red cells but do not destroy
these red cells could have therapeutic implications in
the management of patients with autoimmune dis-
orders. We will briefly list possible reasons why
patients with autoantibodies on their red cells might
not have hemolysis and summarize some of the
studies addressing these various issues.
Characteristics Of The Autoantibody
DiJerences in antibody class. The direct antiglobulin
test gives an approximate estimate of the amount of
IgG or complement on the surface of red cells.
However, this measure is very inexact and it is
possible that the precise amount of autoantibody on
red cells is not strictly comparable in patients with
methyldopa-induced IgG autoantibodies that are as-
sociated with hemolysis and those that are not associ-
ated with hemolysis. Furthermore, IgM is far more
efficient than IgG at complement activation. Conse-
quently, if patients had IgG plus small amounts of
IgM on their red cells, one would anticipate a much
more rapid red cell destruction than if the patient had
IgG alone on the red cells. To address this issue,
Lalezari and associates used an auto-analyzer to
estimate semi-quantitatively the amount of IgG and
IgM on red cells from patients with positive DAT
tests associated with methyldopa therapy.’ 7 These
investigators found that 8 of 11 patients with positive
DAT tests had both IgM and IgG autoantibodies on
their red cells. All of these patients had evidence of
hemolysis. The three patients with IgG only on their
red cells did not have hemolysis. These investigators
postulated that these findings explained why only this
subset of patients with methyldopa-induced autoanti-
bodies had hemolysis, i.e. these patients with both
IgG and IgM autoantibodies had two mechanisms of
cell destruction making hemolysis more likely to
occur. However, this study does not answer why those
patients with ZgG only on their red cells did not
hemolyze. The results of this study await confirma-
tion by other investigators.
Differences in the subclass of the autoantibody. The
direct antiglobulin test provides a relative estimate of
the amount of IgG on red cells. However, it does not
give information about the spatial orientation of the
antibody: whether it is clustered on certain membrane
40 METHYLDOPA-INDUCED AUTOANTIBODIES
proteins, or whether the antigen-antibody complexes
are mobile within the red cell membrane. Neither does
it provide information on the subclass of the anti-
body. All of these factors could affect the rate of
clearance of IgG sensitized red cells. Indeed, there is
evidence that IgG subclasses 2 and 4 interact weakly
with the Fc receptors on the reticuloendothelial cells;
therefore, if certain methyldopa-induced autoanti-
bodies were primarily subclasses 2 and 4, then
hemolysis would not occur. In contrast, those autoan-
tibodies mostly comprised of IgG subclasses 1 and 3
would cause hemolysis. This issue has been investi-
gated by a number of investigators and although
some have proposed that patients who did not have
hemolysis were more likely to have IgG subclasses 2
and 4 1g*20*21recent studies have shown that the
autoantibody consists of all subclasses of 1gG.l’
The amount of autoantibody. Because the direct
antiglobulin test is semi-quantitative at best, investi-
gators have suggested that those patients who have
methyldopa-induced autoimmune hemolysis have
more IgG on their red cells compared to those
patients who have autoantibody but no hemo-
lysis. 3~1g*33,41However, there do not appear to be
major quantitative differences in the amount of auto-
antibody on the red cells from these two groups of
patients, making this an unlikely explanation.
Reticuloendothelial Function in Patients with
Methyldopa-induced Red Cell Immunological
Abnormalities
It is important to remember that the type and amount
of autoantibody, its subclass and its membrane mo-
bility represent only one-half of the cell clearance
equation: the other being the functional integrity of
the reticuloendothelial system. For example, patients
with a very active reticuloendothelial system are more
tXTERMlNANK OF CELL CLEARANCE
Sasitimtii
RE Cdl Fuxtim
Easily Mcasvsd DMkllltteMaarc
Fig. 2 A schematic representation of some of the determinants
of red cell clearance. The amount of immunoglobulin (usually
IgG) or complement on a cell is easily measured and
consequently most biological models of cell clearance focus on
these cell-specific factors. Reticuloend~helial function probably
is just as important in cell clearance, but because it usually has
not been assessed, its contribution tends to be overlooked.
AGAINST RED BLOOD CELLS
likely to clear IgG-sensitized cells than are patients
with a less active reticuloendothelial system. This
concept is illustrated schematically in Figure 2. The
functional activity of the reticuloendothelial cell sys-
tem can be studied either in vitro or in vivo. In vitro
studies are performed by separating reticuloendothe-
lial cells (usually monocytes) from peripheral blood
and then studying their function. Such studies have
been criticized because the cells that are collected, the
circulating reticuloendothelial cells, are probably less
important than the tissue-bound cells of the reticulo-
endothelial system for the clearance of anti-
body-sensitized cells and particles. Furthermore, the
isolation and processing of these cells may alter their
function. Nonetheless, some in vitro reticuloendothe-
lial studies have provided new and useful informa-
tion. Branch and associates42 studied monocytes and
macrophages collected from the peripheral blood of
patients receiving methyldopa and found that the
circulating monocytes of the hemolyzing patients, but
not of the non-hemolyzing patients, phagocytosed the
DAT positive red cells in vitro. Other investigators
have reported similar findings.‘g,43
Reticuloendothelial function can also be studied in
vivo. This is accomplished by measuring the rate of
clearance of radiolabelled autologous red cells which
have been heat damaged (to measure the sieving
capacity of the reticuloendothelial system) or sensi-
tized by alloantibodies. This latter manoeuvre allows
one to study the Fc component of the reticuloendothe-
lial system, i.e., the ability of monocytes and macro-
phages to remove IgG sensitized cells. One group of
investigators performed this type of study in patients
treated with methyldopa. 44 The reticuloendothelial
clearance studies from these studies demonstrated the
following: (a) in one patient with methyldopa-induced
autoantibodies who had hemolysis, the reticuloendo-
thelial system was normal; (b) in several patients who
had methyldopa-induced red cell autoantibodies, but
who did not have hemolysis, the activity of the
reticuloendothelial system was markedly impaired; (c)
in several patients treated with methyldopa, who did
not have red cell autoantibodies, the reticuloendothe-
lial system also was impaired. These studies suggest
that in patients with methyldopa-induced red cell
autoantibodies, the absence of hemolysis is caused by
drug-induced impaired Fc-dependent reticuloendo-
thelial function. The demonstration of impaired reti-
culoendothelial function in several patients who did
not have autoantibodies suggests that the drug itself is
responsible for the impaired reticuloendothelial func-
tion. Indirectly, this supports the concept that drug-
induced autoantibody production is independent of
drug-induced impaired reticuloendothelial function.
Consequently, one can propose a number of different
scenarios. In most patients, methyldopa does not
produce autoantibodies nor does it impair reticuloen-
dothelial function: such patients would not have
hematological consequences. In other patients, the
drug impairs reticuloendothelial function but does

not initiate autoantibody formation, again a clinically
undetectable event. In some patients, autoantibodies
form plus the reticuloendothelial function is impaired
and in these patients there are detectable red cell
autoantibodies, but no hemolysis. Finally, the smallest
subset of patients, autoantibodies form without im-
paired reticuloendothelial function and these patients
have clinical hemolysis.
Further elucidation of the unique effects of methyl-
dopa on the immune system may yield valuable
insights into the pathophysiology of autoimmune
disease and reticuloendothelial function; it may also
allow physicians to manipulate the immune system
in cases of harmful destruction of circulating blood
cells.
Acknowledgement
Studies described in this report were funded entirely by a grant
from the Medical Research Council of Canada.
References
Frohlich E D 1980 Methyldopa. Mechanisms and treatment
25 years later. Archives of Internal Medicine 140: 954-959
Furhoff A-K 1978 Adverse reactions with methyldopa-
a decade’s reports. Acta Medica Scandinavica 203: 425-428
Worlledge S M, Carstairs K C, Dacie J V 1966 Autoimmune
haemolytic anemia associated with alpha-methyldopa
therapy. Lancet 2: 135-139
4. Neuberger J, Kenna J G, Aria K N, Williams R 1985
Antibody mediated hepatocyte injury in methyldopa induced
hepatotoxicity. GUT 26: 1233-1239
5. Seeverens H, Bruin C D de, Jordans J G M 1982
Myocarditis and methyldopa. Acta Medica Scandinavica
Zli: 233-235
6. Manohitharajah S M, Jenkins W J, Roberts P D, Clarke
R C 1971 Methvldopa and associated thrombocvtooenia.
. .
British Medical Journal 1: 494
7. Ziegler Z, Hasiba U, Lewis J, Vagnucci A, West V, Bezek E
1984 Exaggeration of defective platelet function by
methyldopa in hypothyroidism. American Journal of
Hematology 17: 209215
8. Polk 0 D, Kletter G G, Smith J, Castro 0 1982 Impaired
clot retraction in thrombocytopenia due to methyldopa.
Southern Medical Journal 75: 374-375
9. DuPont A, Six R 1982 Lupus-like syndrome induced by
methyldopa. British Medical Journal 285: 693-694
10. Sherman J D, Love D E, Haninton J F 1967 Anemia,
positive lupus and rheumatoid factors with methyldopa.
Archives of Internal Medicine lu): 321-326
11. Velasco M, Silva H, Feldstein E, Pellicer R, Morillo J,
Urbina-Quintana A, Hemandez-Pieretti 0 1985 Effects of
prazosin and alphamethyldopa on blood lipids and
lipoproteins in hypertensive patients. European Journal of
Clinical Pharmacology 28: 513-516
12. Devereux S, Fisher D M, Roter B L T, Hedge U M 1983
Factor VIII inhibitor and raised platelet IgG levels
associated with methyldopa therapy. British Journal of
Haernatology 54: 485-488
13. Petz L D, Branch D R 1985 Drug-induced hemolytic anemia.
In: Chaplin H ted) Methods in Hematoloev. 12. Immune
Hemolyiic Anemias. New York: ChurchilrLivingstone,
47-94
14. Worlledge S M 1969 Immune drug-induced haemolytic
anaemias. Seminars in Hematology 6: 18l-200
15. Bottiger L E, Westerholm B 1973 Acquired haemolytic
anaemia. Acta Medica Scandinavica 193: 227-231
16. Kleinman S, Nelson R, Smith L, Goldfinger D 1984 Positive
direct antiglobulin test and immune hemolytic anemia in
patients receiving procainamide. New England Journal of
Medicine 311: 809812
17. Lalezari P, Louie J E, Fadlallah N 1982 Serologic profile of
BLOOD REVIEWS 41
alphamethyldopa-induced hemolytic anemia: correlation
between cell-bound IgM and hemolysis. Blood 59: 61-68
18. Hsu T C S, Rosenfield R E, Brukart P, Wong K W,
Kochwa S 1974 Instrumented PVP-augmented antiglobulin
tests II: evaluation of acquired hemolytic anemia. VOX
Sanguinis 26: 305-325
19. Yust I, Frisch B, Goldsher N 1986 Antibody-dependent cell-
mediated cytotoxicity and phagocytosis of autologous red
blood cells in alphamethyldopa-induced haemolysis.
Scandinavian Journal of Haematology 36 211-216
20. LoBuglio A F, Jandl J H 1967 The nature of the alpha-
methyldopa red cell antibody. New England Journal of
Medicine 276: 658665
21. Bakemeier R F, Leddy J P 1968 Erythrocytes autoantibody
associated with alphamethyldopa: Heterogeneity of structure
and specificity. Blood 32: l-14
22. Petz L D, Garratty G 1980 Acquired immune hemolytic
anemias. New York: Churchill Livingstone, 267-304
23. Mollison P L 1983 Blood transfusion in clinical medicine
(7th ed). Oxford: Blackwell Scientific Publications, 351-352
24. Carstairs K C, Breckenbridge A, Dollery C T, Worlledge
S M 1966 Incidence of a positive Coombs tests in patients on
a-methyldopa. Lancet ii: 133-135
25. Issitt P D, Pavone B G, Goldfinger D, Zwicker H, Issitt C,
Tessel J A, Kroovand S W, Bell C A 1976 Anti-Wrb and
other autoantibodies responsible for positive direct
antiglobulin test in 150 individuals. British Journal of
Haematology 34: 518
26. Patten E, Beck C E, Scholl C, Stroope R A, Wukasch C 1977
Autoimmune hemolytic anemia with anti-Jk’ specificity in a
patient taking aldomet. Transfusion 17: 517-520
27. Kessey E C, Pierce S, Beck M L, Bayer W L 1973 Alpha-
methyldopa-induced hemolytic anemia involving
autoantibody with U specificity. Transfusion 13: 360
28. Worlledge S M 1973 Immune drug induced haemolytic
anaemias. In: Girdwood R H (ed) Blood Disorders Due to
Drugs and Other Agents. Amsterdam, London, Princeton:
Excerpta Medica
29. Breckenbridge A, Dollery C T, Worlledge S M, Holborow
E J, Johnson G D 1967 Positive direct Coombs tests and
antinuclear factor in patients treated with methyldopa.
Lancet ii:1265-1268
30. De Gruchy G C 1975 Drug-induced blood disorders. Oxford:
Blackwell Scientific Publications, 158
31. Kelton J G I985 The interaction of IgG with
reticuloendothelial cells: biological and therapeutic
implications. In: Garratty G (ed) Current Concepts in
Transfusion Therapy. Arlington, Va: American Association
of Blood Banks
32. Dacle J V, Worlledge S M 1969 Auto-immune hemolytic
anemias. Progress in Hematology 6: 82-120
33. Habibi B 1983 Disappearance of alpha-methyldopa induced
red cell autoantibodies despite continuation of the drug.
British Journal of Haematology 54: 493-495
34. Petz L D 1980 Drug-induced immune haemolytic anaemia.
Clinical Haematology 9: 455482
35. Green F A, Jung C Y, Rampal A, Lorusso D J 1980 Alpha-
methyldopa and the erythrocyte membrane. Clinical
Experimental Immunology 40: 554-560
36 Habibi B 1985 Drug induced red blood cell autoantibodies
co-developed with drug specific antibodies causing
haemolytic anaemias. British Journal of Haematology 61:
139-143
37. Dameshek W 1967 Alpha-methyldopa red-cell antibody:
cross reactivity or forbidden clones? New England Journal
of Medicine 276: 1382
38. Kirtland H H, Mohler D N, Horwitz D A 1980 Methyldopa
inhibition of suppressor-lymphocyte function. New England
Journal of Medicine 302: 825-832
39. Phan-Dinh-Tuy F, Habibi R, Bach M A, Chatenoud L,
Salmon C, Bach J F 1983 T cell subpopulations defined by
monoclonal antibodies in autoimmune hemolytic anemia.
Biomedicine and Pharmacotherapy 37: 75-80
40. Garratty G, Arndt P, Prince H, Shuhnan I 1986 In vitro
IgG production following preincubation of lymphocytes with
methyldopa and procainamide. Blood 68: (suppl 1) 108a
41. Meulen F W van der, Bruin H G de, Goosen P C M.
Bruynes E C E, Joustra-Maas C J, Telkamp H G, Borne
A E G Kr von dean, Engelfriet C P 1980 Quantitative aspects
of the destruction of red cells sensitized with IgGl
42 METHYLDOPA-INDUCED AUTOANTIBODIES AGAINST RED BLOOD CELLS

autoantibodies: an application of flow cytometry. British Evaluation of reticuloendothelial function in autoimmune

Journal of Haematology 46: 47-56
42. Branch D R, Gallagher M T, Shulman I A, Mison A P, Sy
Siok Hain A L, Petz L D 1983 Reticuloendothelial ceil
function in a-methyldopa-induced hemolytic anemia. VOX
Sanguinis 45: 278-287
43. Gallagher M T, Branch D R, Mison A, Petz L D 1983
hemolytic anemia using an in vitro assay of monocyte-
macrophage interaction with erythrocytes. Experimental
Hematology 11: 82-89
44. Kelton J G 1985 Impaired reticuloendothelial function in
patients treated with methyldopa. New England Journal of
Medicine 313: 596600