Chapter 109 

Inhibitors in Hemophilia

Donald I. Feinstein

Inhibitors in Hemophilia A

Factor VIII inhibitors are antibodies that develop in patients with
hemophilia A in response to factor VIII contained in various blood
products. Most of these antibodies neutralize factor VIII coagulant
activity. Although they do not increase the frequency of bleeding
episodes, they make treatment of bleeding much more difficult. The
incidence of factor VIII inhibitors complicating hemophilia A is
approximately 510% of all patients with hemophilia A and 1015% of
those with severe hemophilia A. 1,2 However, in more recent studies of
inhibitor formation in previously untreated patients receiving highly
purified plasma derived factor VIII, the incidence of inhibitors was
1835%, MLID91246858 MLID92031936 MLID92311847  35 and with
intermediate purity factor VIII concentrates the incidence was 2552%.
MLID92167720 MLID93133254  68 In prospective studies using recombinant
factor VIII, the incidence was approximately 20%. MLID93133254  7,8 In
most of these prospective studies, the inhibitors were detected early,
with a median number of exposure days before inhibitor formation of
911, MLID93133254  7,8 and about 50% were low titer and transient.
Thus, most patients with severe hemophilia who are destined to develop
an inhibitor do so early after exposure to factor VIII. MLID93133254
MLID88108142  79 The reason for the variable incidence with newer and
more purified concentrates is not explained.

Although the great majority of inhibitors develop in severe or moderate
forms of hemophilia A, well-documented cases include patients with mild
hemophilia A. MLID69278299 MLID71136796 MLID91082821  2,1014 In
general, these patients develop only low-titer inhibitors following
exposure to factor VIII. After abstinence from products containing
factor VIII, the inhibitors disappear after 412 weeks, and the
inhibitors do not necessarily reappear on re-exposure.

Patients with low-titer inhibitors (35 Bethesda units) that do not rise
after further exposure to factor VIII are known as low responders
(approximately 25% of hemophiliacs with inhibitors), whereas those that
rise markedly with further exposure to factor VIII (anamnestic response)
are known as high responders (approximately 75% of hemophiliacs with
inhibitors). MLID76206851  15 Inhibitor titers usually begin to rise 23
days after exposure to factor VIII, reach a maximum within 721 days,
and then decrease very slowly (Fig. 109-1). Once formed, high-titer
inhibitors tend to persist for long, although variable periods, and
detectable levels of inhibitor may be present 12 years later without
re-exposure to factor VIII. By contrast, low-titer inhibitors in low
responders occasionally disappear and may not reappear with exposure to
factor VIII.


The reasons for the development of inhibitors in a minority of patients
with hemophilia A remain unknown. Brother pairs have a
higher-than-expected incidence or absence of inhibitors. MLID88108142
MLID77224277  9,16 Another factor suggesting genetic susceptibility is
the finding of a lesser incidence of HLA-A1 and a higher incidence of
certain complement components located close to HLA-DR in hemophiliacs
with inhibitors. 17 By contrast, no correlation exists between factor
VIII gene deletions and the development of factor VIII inhibitors.
MLID85296150 MLID85213871  18,19 Shapiro and Hultin 10 suggested that
immune tolerance might be involved to explain the development of
inhibitors in patients with hemophilia A. Since factor VIII does not
cross the placenta, tolerant hemophiliacs could be exposed to factor
VIII in utero only as a result of maternal-fetal hemorrhage. Thus,
according to this hypothesis, only nontolerant hemophiliacs would
develop inhibitors following exposure to factor VIII. Induction of
tolerance in hemophiliacs with inhibitors results in the disappearance
of the inhibitor in a significant number of patients, which supports
this theory. MLID84272861 MLID86286530  2022

Characterization and Properties of Inhibitors

Factor VIII inhibitors in hemophilia A are IgG. Although light chain and
heavy chain subtyping has demonstrated restricted heterogeneity,
MLID73127873 MLID68199215 MLID69211684 MLID75206132 MLID77158812
MLID81226753 MLID81193362  2330 heavy chain subtyping of these
antibodies has shown a significant predominance of the IgG4 subtype.
MLID77158812 MLID77066042  25,28,31 IgG4 does not fix complement, which
might explain why hemophiliacs do not develop immune complex disease.
Factor VIII inhibitors show species specificity, both in vitro and in
vivo, in that human factor VIII is usually neutralized to a greater
extent than is bovine or porcine factor VIII, and infusion of porcine
factor VIII into a patient with an inhibitor often raises the factor
VIII level.

The reaction between factor VIII and inhibitors is time and temperature
dependent. Two different patterns of antigen/antibody reaction have been
described. In the type I pattern, characteristic of most alloantibodies,
as seen in hemophiliacs, factor VIII is completely inactivated in the
presence of excess inhibitor, MLID67093871 MLID73003152  3234 whereas
in the type II pattern characteristic of many autoantibodies, the
inhibitor frequently does not completely inactivate factor VIII in vitro
in the presence of excess inhibitor. MLID73127873 MLID73003152  23,3537
In the latter cases, despite the demonstration of a significant amount
of residual factor VIII activity in vitro, the patient bleeds as if
there were no coagulant function in vivo. MLID83023663 MLID82044272 

The antigenic regions on the factor VIII molecule to which neutralizing
factor VIII inhibitors bind have been identified. MLID89001198
MLID87185841 MLID92376757  4043 Interestingly, the epitopes to which
allo- or autoantibodies are directed are limited to certain areas of the
factor VIII light or heavy chains, or both MLID89001198 MLID87185841
MLID92376757  4043 (Fig. 109-2). Immunoblotting and binding studies
with fragments of factor VIII have shown that plasma from approximately
50% of inhibitor patients contain at least two different neutralizing
antibodies directed at both the light and heavy chain, whereas the other
50% bind to only the light chain. MLID92376757  43 In addition, it has
been noted that the inhibitor produced by a given patient may change
over the course of time and a few non-neutralizing antibodies may occur
as well. MLID89001198 MLID87185841 MLID92376757 MLID90105722  4044

Laboratory Evaluation and Quantification of Inhibitor Titer

The presence of an inhibitor to factor VIII should be suspected in a
hemophiliac if transfused factor VIII appears either to have a short
half-life or is not efficacious in achieving hemostasis, or both. This
can be suspected in the laboratory whenever the partial thromboplastin
time of a mixture of patient's plasma and normal plasma, after
incubation for 2 hours at 37C, is longer than that of a mixture of
patient plasma and hemophilic plasma known not to contain an inhibitor.
The sensitivity of this assay can be markedly increased by using a 4:1
patient/normal plasma mixture and incubating with kaolin/cephalin
suspension for 2 hours at 37C. MLID77158810 MLID82227542  45,46 As
these tests lack specificity, in order to confirm that an inhibitor acts
specifically with factor VIII, a dilution of the patient's plasma must
be incubated with an equal volume of normal plasma and factor VIII
levels measured in subsamples removed immediately and after 60 and 120
minutes. If the inhibitor is specific for factor VIII, factor VIII will
decrease over time in the incubation mixture. Other methods, using
nephelometry, MLID81084762  47 inhibition of coagulation in agarose gel,
MLID76062134 MLID85280367  48,49 and immunoradiometry, MLID82000379
MLID82278057 MLID83123080  5052 have been described but have not been
adopted for widespread use.

Most centers in the United States use the Bethesda assay for
quantification of factor VIII inhibitors. 53 Factor VIII assays are done
on 2-hour incubation mixtures of various dilutions of patient's plasma
with normal plasma. A test sample producing a residual factor VIII
activity of 50% of normal is considered to contain 1 Bethesda unit of
inhibitor per milliliter, and the inhibitor titer equals the reciprocal
of the dilution of inhibitor plasma that neutralizes 50% of normal
factor VIII. In England, the New Oxford Method is used to quantitate
factor VIII inhibitors. MLID82176818  54 One Bethesda unit equals 1.21
times 1 Oxford unit. MLID82176818  54 An inhibitor unit does not imply
that any specific number of factor VIII units infused into the patient
will neutralize any specific number of inhibitor units.


Factors to be Considered in Selecting a  Blood Product for a Bleeding

1.  Patients known to be high responders should not  receive blood
products containing factor VIII to treat  minor hemorrhages, so as to
avoid an anamnestic  response (unless they are undergoing induction of 
immune tolerance). Conservative measures combined  with the
administration of factor IX complex  concentrate is frequently adequate,
since only a few inhibitor  titers rise on exposure to the small amount
of factor  VIII coagulant antigen that may contaminate such 

2.  If a hemorrhage is critical, an attempt should be made  to raise the
plasma factor VIII level into the hemostatic  range of 30 to 50 U/ml.

3.  In those patients with a serious hemorrhage who are  either low
responders or high responders with a low  inhibitor level (<5 Bethesda
units), high-purity human  factor VIII can be given initially in an
initial large bolus  of 5,00010,000 U, followed by a continuous
infusion  of 1,000 U/hr. Alternatively porcine factor VIII can be  used.
By contrast, the same type of patient (low  inhibitor level of <5
Bethesda units) with a minor  hemorrhage can be most easily managed with
factor IX  complex concentrate in doses of 75100 U/kg repeated  once or
twice at 8 - 12-hour intervals as necessary.  Factor IX complex
concentrate has only trace amounts of  factor VIII antigen and only
rarely causes an  anamnestic response. MLID80066000 55

4.  If the level of inhibitor against human factor VIII is  >10 Bethesda
units but the level of inhibitor against  porcine factor VIII is much
<10 Bethesda units, a large  dose of porcine factor VIII should be

5.  If a patient has a moderately high inhibitor level of  1030
Bethesda units to both porcine and human  factor VIII, the inhibitor
level may be lowered by about  5066% by a 1.5-vol plasmapheresis56 or 
extracorporeal immunoadsorption, MLID81208665 MLID91165179 57,58 after
which a large  infusion of factor VIII concentrate may achieve
hemostatic  levels. Alternatively, patients with moderately high 
inhibitor levels can be treated with factor IX complex or  with the
porcine factor VIII concentrate (particularly if  the level of inhibitor
against porcine factor VIII is <10  Bethesda units).

6.  If a patient has a very high inhibitor level (>30  Bethesda units),
the only alternatives are factor IX  complex concentrates or the porcine

General Considerations

The presence of an inhibitor in a hemophiliac, the titer of that
inhibitor, and whether the patient is a low responder or high responder
are important determinants of immediate and future therapy. Minor
hemorrhage may respond to conservative measures, such as immobilization
and compression. For a hemorrhage requiring blood product therapy, two
possible choices are available: (1) raise the factor VIII level by
infusion of human or porcine factor VIII and, if the inhibitor titer is
moderately high, precede this therapy by maneuvers to lower the titer;
or (2) infuse an agent that bypasses the need for factor VIII, such as
factor IX complex concentrates or factor VIIa.

High-Purity Human Factor VIII Concentrate

High-purity human factor VIII may be used successfully in the treatment
of critical hemorrhages in either low or high responders with low
inhibitor levels (<5 Bethesda units) and in patients with moderate
inhibitor levels (1030 Bethesda units) after reduction of the inhibitor
level by plasmapheresis or immunoadsorption. In order to neutralize the
inhibitor and achieve a hemostatic level of 3050 U/ml, an adult patient
is given an initial bolus of factor VIII of 5,00010,000 U, followed by
a continuous infusion of 3001,000 U/hr. Alternatively, a large amount
of factor VIII can be given every 14 hours (20 U of factor VIII for
each Bethesda unit plus an additional 40 U/kg). MLID91221500  59 Since
the dose of factor VIII concentrate needed to neutralize inhibitors and
provide a hemostatic level of factor VIII cannot be predicted by the
inhibitor level, factor VIII levels need to be assayed frequently in
order to monitor the plasma factor VIII level achieved in vivo. However,
although blood drawn for assay even a few minutes later may not have a
measurable factor VIII level by the time the blood is processed and the
assay completed, the infused factor VIII may achieve in vivo hemostasis
before it is inactivated.

In patients who are high responders and who sustain a critical
hemorrhage, hemostasis should be achieved early, before the anamnestic
response occurs. As the factor VIII inhibitor level rises after several
days of factor VIII therapy, factor VIII should be continued at a
frequent interval or continuouslyhemostasis can be maintained because
of the slow neutralization rate of factor VIII. MLID67093871  32

Porcine Factor VIII Concentrate

The suitability of a patient for porcine factor VIII depends partially
on the degree of cross-reactivity of the patient's inhibitor with
porcine factor VIII. The level of inhibitor to porcine factor VIII can
be determined in the Bethesda test or the Oxford test by substituting
porcine factor VIII for human factor VIII. Although the degree of
cross-reactivity can vary widely from 0% to 75%, MLID84080838
MLID85041086 MLID89272459  6063 on average it is usually about 25%
(i.e., about 1 U of antiporcine factor VIII equals 4 U of antihuman
factor VIII MLID84080838  60 (Kasper CK, personal communication). After
treatment with porcine factor VIII, the degree of cross-reactivity
increases. Thus, measurements of cross-reactivity are useful in
predicting efficacy before the use of the porcine concentrate.

Although dosage is relatively arbitrary, in patients with inhibitor
titers of <5 Bethesda units, 50 U of porcine factor VIII/kg should be
given initially, and in patients with >5 Bethesda units, 50150 U/kg
MLID84080838 MLID89272459  60,63 (Kasper CK, personal communication). It
is very important to monitor the level of factor VIII achieved in vivo
following the infusion to determine the adequacy of the dose
administered; if the initial dose is inadequate, rapid upward titration
of the dose is important.

Several patients have had multiple courses of therapy with the porcine
concentrate without loss of efficacy. In only the occasional patient
have heterologous antibodies to porcine factor VIII developed. Classic
anamnestic responses to porcine factor VIII occasionally occur,
particularly in high responders who are infrequently treated. 61 In
addition, a few cases of thrombocytopenia in response to the porcine
concentrate have been reported, MLID89272459 MLID89116256  63,64 as has
the occasional anaphylactic reaction. Mild febrile reactions are common
and are easily modulated with prophylactic antihistamine and

Management of Critical, Life-Threatening  Hemorrhages or Emergency
Surgery In   Patients With Inhibitors 

In an emergent situation with a critical hemorrhage or  with impending
lifesaving surgery, both a high responder  and low responder should be
initially treated with a bolus  infusion of the porcine factor VIII
concentrate 100 U/kg.  Ten minutes after the infusion, blood should be
drawn  for factor VIII assay in order to determine the adequacy  of the
dose; if inadequate, the dose should be increased  by 50 U/kg, until the
patient has a postinfusion factor VIII  level of 3050%. Simultaneously,
the preinfusion  antibody titer to both porcine and human factor VIII
can be  determined. If the titer to porcine factor VIII is >10  Bethesda
units and clinical efficacy has not been  established, plasmapheresis
can be performed, followed by  a massive dose of porcine factor VIII.
Once efficacy is  established, the patient should be treated at frequent
 intervals or by continuous infusion at a dose of 1,000 U/ hr. Factor
VIII levels should be monitored frequently. As  long as hemostatic
levels are achieved, the porcine  concentrate should be continued for
several days. If the  inhibitor titer begins to rise and hemostatic
levels can no  longer be achieved, anti-inhibitor coagulant complex 
concentrate or factor IX complex concentrate should be  used for several

If the patient does not respond initially to porcine  factor VIII,
anti-inhibitor complex concentrate may be  administered, starting at a
dose of 100 U/kg. Factor VIIa is  currently available on a compassionate
basis on research  protocol by the manufacturer.

Factor IX Complex Concentrate

If human factor VIII cannot be used because the inhibitor level is too
high (>30 Bethesda units), an attempt may be made to bypass the need for
factor VIII, by giving a factor IX complex concentrate. MLID79042000
MLID81270622 MLID80232757 MLID81270615 MLID84025044 MLID83075888  6572
These concentrates contain phospholipids, prothrombin, and factors VII,
IX, and X, but the substance(s) responsible for factor VIII bypassing
activity remain(s) unknown. MLID79166675 MLID83283809  73,74 These
concentrates are of two types: unactivated intermediate-purity factor IX
complex concentrates; and activated concentrates, so-called
anti-inhibitor coagulant complex. The activated concentrates (Autoplex
[Baxter]; FEIBA [Immuno]), contain a greater amount of the factor VIII
bypassing material and are specifically designed for patients with
factor VIII inhibitors. Both types of concentrates have proved
efficacious in controlled clinical trials, MLID81270615 MLID84025044 
70,71 and no significant differences have been demonstrated. However,
some patients have described successful use of anti-inhibitor coagulant
complex in life-threatening situations in which the unactivated
concentrates were ineffective. These concentrates should be
distinguished from the highly purified factor IX concentrates used for
the treatment of hemophilia B.

In addition to their unpredictable efficacy, these concentrates have
other limitations. Dosage is relatively arbitrary, and no reliable in
vitro method is available that reflects in vivo efficacy. These
concentrates also induce a hypercoagulable state and may be associated
with thromboembolic complications. These concentrates contain a small
amount of factor VIII antigen and can induce an anamnestic response
after their use. If selected for use in critical situations, these
concentrates should be used early in the course of the bleeding episode.
MLID80066000  55

Human Factor VIIa Concentrate

Purified components of the prothrombin complex have recently been used
to treat hemorrhages in patients with inhibitors. A highly purified
concentrate containing factor VIIa achieved good hemostasis MLID83239008
 75 and more recently, a recombinant preparation of human factor VIIa
has shown significant efficacy in patients with inhibitors. MLID89272459
 63,76 Recombinant factor VIIa is currently an experimental drug.

Suppression of Inhibitors

Long-term goals in the management of these patients should be aimed at
the prevention of anamnesis and suppression of further antibody
production. Immunosuppressive therapy has been attempted in many
patients, with variable and unpredictable results. MLID69211684
MLID74047581 MLID73226034 MLID75089231 MLID76232703  10,26,7782 In
general, with few exceptions, corticosteroid therapy alone is not
efficacious in patients with hemophilia A and inhibitors. MLID76232703 
82 By contrast, Dormandy and Sultan MLID76175273  83 collected the
experience of many investigators using cyclophosphamide and factor VIII.
Some degree of success was achieved on 40 occasions in 18 of 45
patients. The data reported by Hultin et al. MLID76232703  82 are in
agreement with these observations. Although total elimination of
antibody occurs very rarely, immunosuppressive therapy may be helpful in
patients requiring factor VIII infusion, particularly low responders
with low inhibitor titers. MLID76232703  82 Rarely, treatment of
inhibitors soon after their appearance and before re-exposure to factor
VIII may be an important determinant of successful outcome. MLID76232703
 82 Despite these encouraging reports, the use of immunosuppressive
drugs has not gained wide acceptance.

The induction of immune tolerance by the frequent regular infusions of
factor VIII has become increasingly accepted, but the financial
implications of this therapy are great. MLID84272861  20,61 A patient
who is a high responder with a high-titer inhibitor is given daily or
alternate-day infusions of various amounts of human factor VIII.
MLID84272861  20,61 Following chronic administration of factor VIII, the
inhibitor titer first increases and then progressively decreases, with
most patients achieving either a very low titer (<2.5 U), or the
inhibitor is undetectable. Moreover, the patient no longer has anamnesis
with continued exposure to factor VIII. Similar results have been
achieved at some centers, either with less amounts of factor VIII
MLID83276392 MLID87049531  21,61,8486 or by giving factor VIII
repeatedly but at irregular intervals. MLID83000167  87 Immune tolerance
usually must be maintained by giving low doses of factor VIII every few

Successful outcomes of the induction of tolerance occur in about 75% of
patients and appear to be more common in patients <20 years of age, in
patients who are low responders, and in those receiving higher doses of
factor VIII (E50 U/kg/day). Human immunodeficiency virus status does not
appear to affect outcome (Kasper CK, personal communication, 1993).

Successful regimens using factor VIII infusions combined with
immunosuppressive drugs appear to shorten the induction of immune
tolerance significantly, attaining remission at a much lower cost.
MLID86123621  88

In recent studies, MLID88174925 MLID90175503  89,90 high-dose
intravenous immunoglobulin given together with high-dose factor VIII and
cyclophosphamide appeared to have a synergistic effect in achieving a
state of immune tolerance. Intravenous IgG may have both a short- and
long-term immunosuppressive effect. MLID85110011 MLID89114721  9193
Anti-idiotypic antibodies are present in the IgG preparations
MLID85011599 MLID90053531  9496; the emergence of anti-idiotypic
antibodies may be the explanation for the occasional occurrence of
spontaneous remission and some instances of remission induced by
immunosuppressive agents. MLID90053531 MLID85059880 MLID89134724 

Inhibitors in Hemophilia B

Inhibitors to factor IX occur in hemophilia B secondary to
alloimmunization following transfusion similar to hemophilia A. Although
the overall incidence of inhibitors complicating hemophilia B is only
2.42.8%, MLID74303213  10,99 the incidence in patients with severe
hemophilia B is approximately 12%. 10 Patients with major gene deletions
who do not have demonstrable factor IX protein in their plasma may be
especially susceptible to the development of factor IX antibodies,
MLID83192505  100 but more recent observations have not supported this

Like factor VIII inhibitors, factor IX inhibitors are IgG. MLID77202199 
10,101103 Some have restricted light chain and heavy chain
heterogeneity, MLID83286434  102,104 whereas others are polyclonal.
MLID77202199 MLID81232699  103,105 The kinetic behavior of factor IX
inhibitors differs from that of factor VIII inhibitors in that factor IX
inhibitors produce an immediate loss of factor IX activity, with no
progressive loss on further incubation. 10,101

Laboratory Evaluation and Quantification

Factor IX inhibitors should be suspected in a hemophiliac if transfused
factor IX either appears to have a short half-life or is not efficacious
in achieving hemostasis, or both. This can be suspected in the
laboratory whenever a partial thromboplastin time of a mixture of the
patient's plasma and normal plasma is longer than that of a mixture of
patient plasma and factor IX-deficient plasma known not to contain an
inhibitor. Specificity can be determined by doing factor IX assays of an
incubation mixture of patient plasma and normal plasma over time. A
modification of the Bethesda assay is used to quantitate the inhibitor
level. 53


In patients with hemophilia B who have inhibitors, the clinical course
is not different from that seen in those with hemophilia A with
inhibitors. Most of these cases can be managed with factor IX complex
concentrates or the anti-inhibitor/coagulant complex. Whether the
efficacy of these concentrates is due to neutralization of the inhibitor
or to bypassing activity similar to that observed in hemophilia A, or
both, is unknown. In addition, the induction of immunotolerance has been
achieved in hemophilia B patients with the infusion of high doses of
intravenous IgG in combination with cyclophosphamide and factor IX
concentrates. MLID87067471  106