Acquired inhibitors of coagulation

INTRODUCTION — Acquired inhibitors of coagulation are antibodies that either inhibit the activity or increase the clearance of a clotting factor. A common clinical manifestation in affected patients is a hemorrhagic diathesis.

The natural history and management of these inhibitors are quite different from inhibition due to alloantibodies that occur in patients with various inherited bleeding disorders (eg, congenital deficiencies of factors VIII, IX, or XI) treated with factor replacement. Acquired coagulation inhibitors that lead to bleeding will be reviewed here. Inhibitors in patients with hemophilia and factor XI deficiency are discussed separately. (See "Inhibitors in hemophilia: Mechanisms, prevalence, diagnosis, and eradication" and "Factor XI (eleven) deficiency", section on 'Inhibitor development'.)

●Some disorders may be associated with antibodies to a variety of clotting factors. In systemic lupus erythematosus, for example, antiphospholipid antibodies and antibodies directed against factors II, VIII, IX, XI, XII, and XIII have been described [1]. (See "Clinical manifestations of antiphospholipid syndrome" and "Hematologic manifestations of systemic lupus erythematosus", section on 'Autoantibodies'.)

●Antiphospholipid antibodies represent a special problem; these antibodies prolong certain clotting assays, but typically result in thrombosis rather than bleeding. The pathogenesis, diagnosis, and treatment of conditions associated with antiphospholipid antibodies are discussed elsewhere in UpToDate. (See "Pathogenesis of antiphospholipid syndrome" and "Clinical manifestations of antiphospholipid syndrome" and "Management of antiphospholipid syndrome".)

FACTOR VIII INHIBITORS — The most common autoantibodies that affect clotting factor activity and lead to a bleeding disorder are directed against, and interfere with, the activity of factor VIII, a condition also called acquired hemophilia A [2-7]. Although there are no large series describing the immunoglobulin class of the factor VIII autoantibodies, most have been IgG antibodies that do not bind complement.

The reasons for the production of factor VIII autoantibodies in a particular individual are not clear, but may involve the presence of certain gene polymorphisms (eg, HLA, CTLA4) and/or autoreactive CD4+ T lymphocytes [8,9].

An individual patient, either nonhemophilic or hemophilic, can produce antibodies of different IgG subclasses directed against different epitopes on the factor VIII molecule [10-12]. However, the production of multiple antibodies is much more common in patients with hemophilia [13]. Most of these antibodies bind to the C2 or, less often, the A2 domain on factor VIII [13-15]. Loss of the C2 domain, which binds to the procoagulant phospholipid phosphatidylserine on activated platelets and endothelial cells and to von Willebrand factor [16], leads to reduced procoagulant activity [14,15]. (See "Biology and normal function of factor VIII and factor IX".)

A similar finding has been noted with factor V inhibitors that are associated with bleeding manifestations; the antibodies are usually directed against the C2 domain of the protein (see 'Factor V inhibitors' below) [17,18], which is also responsible for binding to phosphatidylserine on activated platelets and endothelial cells [19].

Epidemiology — Population-based unselected cohorts of patients seen in South and West Wales and the United Kingdom have calculated an incidence of acquired factor VIII inhibitor of 1.3 to 1.5 cases per million population per year [20,21]. In another survey of 118 physicians likely to see such patients, 215 cases were described over a 10-year period; the following characteristics were noted [2]:

●Most of the patients were over age 50, except for women who were pregnant or postpartum.

●The major identifiable causes were pregnancy or the postpartum period, rheumatoid arthritis, malignancy, systemic lupus erythematosus, and drug reactions (each responsible for 5 to 10 percent). However, in almost one-half of the patients no underlying disorder was present.

A meta-analysis of series and surveys of patients with acquired factor VIII inhibitors, including the author's own experience, and comprising 249 patients, was published in 2003 [22]. Median age was 64 (range: 8 to 93); 55 percent were female. Median inhibitor and factor VIII levels were 10 Bethesda units (range: 0.9 to 32,000) and 2 percent (range: 0 to 30), respectively. The most common associated conditions were malignancy, post-partum status, and autoimmune disorders. On multivariate analysis, factors predicting for improved overall survival were attainment of complete remission, age <65, and post-partum status. Although treatment data appeared to suffer from publication bias, the highest rates of complete remission and lowest rates of disease-specific mortality were noted in patients treated with cyclophosphamide.

In 2012, a large prospective study detailed the demographics and clinical characteristics of 501 European patients who presented between 2003 and 2008; findings were consistent with the earlier meta-analysis of smaller studies [23].

●Postpartum – Several retrospective studies and reviews of acquired factor VIII inhibitors associated with pregnancy have come to similar conclusions regarding the course and prognosis of these inhibitors [2,3,24-26]. The inhibitors are typically diagnosed within two to three months postpartum. Most cases were associated with the first pregnancy (ie, in primigravidas) [24]. The survival of patients with pregnancy associated factor VIII inhibitors is nearly 100 percent, substantially better than that of patients with these inhibitors not associated with pregnancy. This excellent survival might reflect the younger age, overall better health, and reversibility of the condition that provoked the development of the inhibitor [25]. (See 'Natural history' below.)

●Systemic rheumatic disease – Among the systemic rheumatic diseases, factor VIII inhibitors have primarily been described in patients with rheumatoid arthritis and systemic lupus erythematosus [1,2,27-29]. In occasional patients, these antibodies coexist with antiphospholipid antibodies [30].

●Malignancy – Anti-factor VIII antibodies are a rare complication of solid tumors. In a review of 27 analyzable cases, there was a close temporal relationship between detection of the inhibitors and diagnosis of the tumor [31]. There was no association with a specific tumor; the possible causative role of treatment with immunomodulatory agents (eg, interferon, fludarabine, ipilimumab) is unclear [32-36].

In a second review of 41 patients identified through the available literature, there were 25 with solid organ tumors and 16 with hematologic malignancies [37]. Adenocarcinomas comprised 64 percent of the solid tumors, with the most common sites being prostate and lung. Chronic lymphocytic leukemia was the most common of the hematologic malignancies. The median age was 70 and 68 percent of the subjects were men. Treatment of the inhibitor led to a complete resolution of the inhibitor (CR) in 70 percent of patients. Patients achieving CR were more likely to have early stage tumors and a lower median inhibitor titer at the time of presentation, compared with nonresponders.

●Drug-induced – Autoantibodies against factor VIII have been associated with some medications (eg, penicillin, sulfonamides, phenytoin, clopidogrel) and immunomodulatory agents such as immune checkpoint inhibitors, alemtuzumab, omalizumab, interferon, and fludarabine [4,38,39]. These antibodies often resolve following cessation of the offending drug.

Clinical features — The hallmark of acquired factor VIII antibodies is bleeding that, in some cases, is first noted during or after a surgical procedure. However, most occur spontaneously without apparent provocation [23]. Symptomatic patients often present with large hematomas, extensive ecchymoses or severe mucosal bleeding, including epistaxis, gastrointestinal bleeding, and gross hematuria. Spontaneous hemarthroses, which are common in hereditary factor VIII deficiency, are unusual in those with acquired disease (table 1).

The range of bleeding manifestations can be illustrated by observations in women with postpartum factor VIII inhibitors. Soft tissue bleeding was most common (29 of 51 patients), followed by muscle, vaginal, joint bleeding, and hematuria (13 to 18 patients) [3]. There were no episodes of intracerebral bleeding.

Bleeding is often severe, constituting a medical emergency. In the above survey of 215 patients, 87 percent experienced major bleeding and 22 percent died from complications attributed directly or indirectly to the inhibitor [2]. In the United Kingdom cohort, bleeding was the cause of death in 9 percent and remained a risk until the inhibitor had been eradicated [21]. A significant number of patients have repeat bleeding after initial therapy.

Diagnosis — The sudden presence of large hematomas or extensive ecchymoses in an older individual without significant trauma or known bleeding disorder should always raise the clinical suspicion of an acquired factor VIII inhibitor.

Coagulation tests — In addition to the clinical features, factor VIII inhibitors are characterized by a prolonged activated partial thromboplastin time (aPTT) and a normal prothrombin time (PT). The differential diagnosis of this profile in a patient with a bleeding diathesis includes deficiencies of factors XI, IX or VIII, von Willebrand disease, inhibitors of these factors, and the use of heparin (table 2). A prolonged aPTT is also seen in the antiphospholipid antibody syndrome, but these patients typically present with thrombosis rather than bleeding episodes. (See "Clinical manifestations of antiphospholipid syndrome".)

A reasonable first step is to exclude the use of heparin as a cause of the patient's bleeding and/or prolongation of the aPTT. In addition to a review of the patient's medications, the presence of heparin in the blood sample can be suspected via one of the following (see "Clinical use of coagulation tests", section on 'Activated partial thromboplastin time (aPTT)'):

●The simplest approach is to redraw a blood sample using an uncontaminated peripheral vein. If the repeat aPTT is normal, the original sample was probably contaminated with heparin.

●Perform a thrombin time and a reptilase time. Heparin is present if the thrombin time is prolonged and the reptilase time is normal.

Inhibitor screen (mixing test) — The primary initial diagnostic test for a factor VIII inhibitor (after documented a prolonged aPTT) is the mixing test or inhibitor screen. In this assay, varying amounts of patient plasma and pooled normal plasma are mixed and the aPTT measured. Generally only a 1:1 mix is done in an inhibitor screen. After mixing, measurement of the aPTT should be performed not only immediately but also after incubation at 37ºC for one to two hours. The second measurement is necessary to detect factor VIII inhibitors with slow reaction kinetics [40]. (See "Clinical use of coagulation tests".)

Correction of the prolonged aPTT suggests a factor deficiency or VWD, while persistent prolongation of the aPTT indicates the presence of an inhibitor. The mixing test will establish whether an inhibitor is present but will not identify the inhibitor's specificity.

The next step is adding a source of phospholipid to the mixed plasma. Correction of the aPTT suggests the presence of antiphospholipid antibodies. If the aPTT does not correct, the Bethesda assay is performed. The Bethesda assay both establishes the diagnosis of a factor VIII inhibitor and quantifies the antibody titer [41]. By standardizing the strength of factor VIII inhibitors, this assay also permits comparison between different treatment regimens.

Bethesda assay — In the Bethesda assay, the factor VIII activity level is measured as would be done in a patient with hemophilia A. Then, serial dilutions of patient plasma are incubated with pooled normal plasma at 37°C for two hours, and factor VIII activity is again measured, and the dilution at which factor VIII activity is restored. The reciprocal dilution of patient plasma that results in 50 percent factor VIII activity is defined as one Bethesda unit (BU). The stronger the inhibitor, the greater the dilution required to allow for factor VIII activity and the higher the BU titer.

Other testing — An enzyme-linked immunosorbent assay (ELISA)-based immunoassay for anti-factor VIII antibodies has also been investigated [42].

TREATMENT OF FACTOR VIII INHIBITORS — Initial treatment of acquired factor VIII inhibitors consists of two separate steps [43-45]:

●Control of bleeding and

●Elimination of the inhibitor

As with any other bleeding disorder, actions that might provoke bleeding (eg, intramuscular injections, invasive procedures, use of antiplatelet agents) should be avoided or minimized [46].

Control of active bleeding — Treatment strategies to control active bleeding include the use of desmopressin (DDAVP), factor VIII concentrates, activated prothrombin complex concentrates (aPCCs; eg, factor eight inhibitor bypassing activity [FEIBA]), recombinant human factor VIIa, and recombinant porcine sequence factor VIII concentrate [4,43,44,47-55].

The decision regarding initial treatment should be made based upon the severity of bleeding and the titer of the inhibitor [56].

●Non-life-threatening bleeding – Some patients with non-life-threatening bleeding and low inhibitor titers have received DDAVP at a dose of 0.3 mcg/kg subcutaneously per day for three to five days to control non-life-threatening bleeding or for hemostatic coverage of invasive procedures [46,57,58]. However, due to the overall superior efficacy of the factor bypassing agents, DDAVP is rarely recommended [54].

●Severe bleeding, low titer inhibitor – Human factor VIII products (recombinant factor VIII or a factor VIII concentrate) at high doses have been used for bleeding associated with a low titer inhibitor (eg, <5 Bethesda units). The typical dose of human factor VIII is 20 international units/kg intravenously for each Bethesda unit of the inhibitor, plus an additional 40 international units/kg, with monitoring of factor VIII activity 10 minutes following bolus injection, and repeat intravenous bolus dosing if the incremental recovery is not adequate [46]. However, due to the overall superior efficacy of the factor bypassing agents, high-dose factor VIII is not recommended [54,59].

●Severe bleeding, high titer inhibitor – For patients with higher titer factor VIII inhibitors (ie, ≥5 Bethesda units) and/or severe bleeding, we suggest the use of an activated prothrombin complex concentrate (FEIBA) or recombinant human factor VIIa (rFVIIa) [50,51]. Human factor VIII products generally cannot be given in high enough amounts to overcome the inhibitor. Recombinant porcine (pig) factor VIII is also available [55].

There are no comparative trials to determine whether an aPCC or rFVIIa is more effective in the setting of a high titer inhibitor and active bleeding. Results from an uncontrolled European registry in 501 patients with acquired factor VIII deficiency suggest that the two agents have similar efficacy [54]. Thus, the choice between FEIBA and rFVIIa should be guided by local experience and cost considerations.

Recommended doses for FEIBA and rFVIIa are similar to those employed in hemophilia patients with inhibitors (eg, typical FEIBA dose 75 units/kg; rFVIIa median starting dose 90.4 mcg/kg, range 45 to 181 mcg/kg). (See "Treatment of bleeding and perioperative management in hemophilia A and B", section on 'Bypassing products (rFVIIa products or FEIBA)'.)

Porcine (pig) factor VIII has protein sequence differences from human factor VIII, and may thus be less inactivated by factor VIII inhibitors. In a study using a previously available porcine factor VIII product, the inhibitor titer to porcine factor VIII was only 5 to 10 percent of the titer against human factor VIII [60]. A recombinant porcine factor VIII product engineered to lack the B domain (OBI-1, Obizur) is also available [55]. In a prospective study in 28 patients with an acquired factor VIII inhibitor and severe bleeding, this product controlled bleeding in 24 individuals (86 percent) [55]. Efficacy was greater in those who received OBI-1 as primary therapy versus those who received another hemostatic agent first (94 versus 73 percent). Patient comorbidities included a variety of underlying conditions such as malignancy or rheumatologic disease; 16 had no underlying etiology for acquired factor VIII inhibitor identified. Dosing was initiated at 200 units/kg and titrated according to clinical bleeding and factor VIII activity level. All individuals also received immunosuppressive therapy with a glucocorticoid and/or another agent (eg, rituximab, cyclophosphamide). Five patients developed new inhibitors to the porcine product. There were no serious adverse events.

Recombinant ovine (sheep) factor VIII is also being investigated [61].

Eliminating the inhibitor — Elimination of the factor VIII inhibitor requires the use of immunosuppressive modalities which are described below. Of importance, some inhibitors will regress spontaneously. (See 'Natural history' below.)

There are no convincing data from randomized trials that one immunosuppressive regimen is better than another or that the choice of regimen should be based upon the inhibitor titer or factor VIII level [45]. The lack of definitive data from non-randomized studies was shown in a retrospective analysis of the outcomes of various immunosuppression regimens in 331 subjects enrolled in the European Acquired Hemophilia Registry [62].

●The most commonly employed initial immunosuppressive regimens were glucocorticoids alone (n = 142), glucocorticoids plus cyclophosphamide (n = 83), and glucocorticoids plus rituximab (n = 28).

●Complete remissions (ie, undetectable inhibitor levels, factor VIII levels >70 International units/mL, and immunosuppression stopped) were noted in 48, 70, and 59 percent of those treated with glucocorticoids alone, glucocorticoids plus cyclophosphamide, and glucocorticoids plus rituximab, respectively.

●The median times to complete remission were approximately five weeks for the use of glucocorticoids with or without cyclophosphamide; rituximab-based regimens required approximately twice as long. Immunoglobulin administration did not improve outcome.

●Adverse events were common in all three treatment groups, and were reported in 25, 41, and 37 percent, respectively.

●Second-line therapy was successful in approximately 60 percent of cases that failed first-line therapy; outcome was not affected by the choice of first-line therapy.

Although patients were more likely to respond to glucocorticoids plus cyclophosphamide than with other regimens, outcome at final follow-up was not affected by the choice of first-line therapy. Data were insufficient to adjust the immunosuppression regimen based upon factor VIII levels and/or the inhibitor titer.

A prospective study suggested that baseline factor VIII activity could be used to individually choose immunosuppression regimens [63].

Prednisone and cyclophosphamide — A prospective randomized trial evaluated the efficacy of prednisone, cyclophosphamide, and the combination of both drugs in 31 nonhemophilic patients with factor VIII antibodies [64]. All patients initially received prednisone (1 mg/kg per day PO) for three weeks; if the antibody persisted, the patients were randomly assigned to continuation of prednisone for another six weeks, tapering of prednisone and initiation of cyclophosphamide (2 mg/kg per day orally), or continuation of prednisone with the addition of cyclophosphamide. The antibody disappeared in 10 of the 31 subjects (32 percent) during the initial course of prednisone. Twenty of the 21 nonresponding patients were then randomly assigned to further treatment, with the following results [64]:

●The antibody disappeared in three of the four subjects randomly assigned to receive continued prednisone alone, making the overall response to prednisone alone 42 percent.

●The antibody disappeared in three of six randomly assigned to receive cyclophosphamide alone.

●The antibody disappeared in 5 of 10 randomly assigned to receive prednisone plus cyclophosphamide.

The antibody titer was significantly lower in responders than nonresponders, but seven patients with titers >5 Bethesda units had complete remissions. The authors concluded that all patients should receive initial prednisone therapy and that cyclophosphamide was effective second-line therapy in many patients who were steroid-resistant.

This conclusion was supported by results of a non-randomized retrospective analysis of 88 patients from a UK cohort study [21]. No significant difference was found in the proportion of patients achieving complete remission or in the time to complete remission following treatment with steroids alone versus those treated with steroids plus cytotoxic therapy (usually cyclophosphamide). Of interest, all four patients with rheumatoid arthritis responded to treatment with steroids alone, in contrast to reports that suggested such patients are resistant to treatment with steroids alone and do better when treated with cyclophosphamide [27,28].

Similar findings have been noted in patients with postpartum disease. In the above retrospective series of 51 patients, prednisone did not appear to be more effective than no therapy, but a shorter time to complete response was noted in patients treated with cyclophosphamide, azathioprine, or 6-mercaptopurine [3]. The overall outcome was quite favorable, with 97 percent survival at two years and almost 100 percent probability of complete remission by 30 months.

Other options — Another therapy of acquired factor VIII inhibitors that has been reported is the administration of intravenous immune globulin (IVIG) [29,46,65]. In a prospective study, 19 patients were treated with IVIG (1 g/kg IV for two days or 400 mg/kg IV for five days) [65]. Two patients had a decline in inhibitor titer within several days, and four patients responded over weeks to months. Because only some patients respond to IVIG and multiple courses are often required, and given the excellent results obtained with prednisone and cyclophosphamide, we suggest that IVIG not be used as initial therapy.

Utility of the anti-CD20 antibody rituximab, alone or with cyclophosphamide and/or prednisone, has been demonstrated in a number of patients with acquired factor VIII inhibitors [66-75]. Highly anecdotal information suggests that subjects with high inhibitor titers (eg, >100 BU) may require the use of all three agents [46,68,69,71,75].

Anecdotal information indicates that some patients resistant to all of the above therapies have responded to cyclosporine [76-79] or to cladribine [80]. In the rare patient not responding to the above measures, the use of extracorporeal plasmapheresis with an immunoadsorption column to absorb the autoantibody can be tried [56,81-83].

Pregnancy — Management of a factor VIII inhibitor during pregnancy is similar to non-pregnant individuals. The most relevant aspect of management is prevention and/or control of bleeding. Immunosuppression with prednisone may be used. Other agents are generally avoided unless maternal health is critically dependent on their use. (See "Safety of rheumatic disease medication use during pregnancy and lactation".)

The efficacy of therapy is monitored similarly to non-pregnant individuals. (See 'Monitoring response to treatment' below.)

Monitoring response to treatment — The primary goal of treatment for an acquired factor VIII inhibitor is cessation of bleeding, followed ultimately by a decrease in the titer of the inhibitor. The former is monitored via the usual clinical and laboratory observations (eg, observable blood loss, blood in urine or stool, repeated blood counts).

An adequate response is complete absence of the inhibitor, which is the goal of treatment but not always achieved. We monitor inhibitor titers during immunosuppressive therapy. However, since inhibitor titers drop very slowly following successful treatment, it is neither necessary nor advisable to check the patient's aPTT or inhibitor titer more often than every two to four weeks once immunosuppressive therapy has been started. We also check a factor VIII activity level after approximately four weeks of immunosuppressive therapy to determine if factor activity is detectable.

Natural history — In terms of eliminating the inhibitor, most patients have been treated with immunosuppressive drugs. Although these drugs hasten the time to complete remission, there is an appreciable incidence of spontaneous recovery. In the survey of 215 nonhemophilic patients noted above, 31 received no therapy other than supportive transfusions or factor concentrates; the inhibitor disappeared spontaneously in 11 (36 percent) at an average duration of 14 months [2]. Similar findings were noted in a report of 16 patients who received little or no therapy for an average of 31 months [84]. There were no spontaneous hemarthroses, two died from retroperitoneal bleeding, and five (31 percent) had spontaneous disappearance of the inhibitor. Given the continued risk of bleeding, we do not advise withholding immunosuppressive therapy; if this approach is selected, it should be an informed decision made by the patient with the understanding that immunosuppressive therapies are usually successful.

The relapse rate after a first complete remission has been estimated at approximately 20 percent; 70 percent of such relapsing patients achieve a second complete remission [21,43].

The relapse rate of pregnancy-associated factor VIII inhibitors appears to be lower [85]. Relapse of pregnancy-associated acquired factor VIII inhibitors may rarely occur in subsequent pregnancies; the antibody may affect fetal factor VIII levels, resulting in life-threatening hemorrhage in a subsequent fetus because of transplacental transfer of IgG antibodies [24,43,86,87]. Patients with low antibody titers (ie, <5 Bethesda units) tend to have remissions within months, whereas those with higher titers may have antibody persistence for years [24].

In at least one case, treatment of the underlying condition (eg, malignancy) has resulted in disappearance of the inhibitor [88].

OTHER COAGULATION FACTOR INHIBITORS — Autoantibodies can be directed against a variety of clotting factors other than factor VIII, with variable clinical manifestations [89]. These will be described below, with the exception of antibodies to von Willebrand factor, which are discussed separately. (See "Acquired von Willebrand syndrome" and "Pathophysiology of von Willebrand disease", section on 'Causes of reduced VWF in acquired VWS'.)

Prothrombin (factor II) inhibitors — Antibodies to prothrombin are most often detected in patients with antiphospholipid antibodies (aPL) [90,91]. The aPL antibody that correlates with antiprothrombin is the antiphosphatidyl serine antibody, which correlates with the lupus anticoagulant phenomenon and likely causes functional interference in the prothrombin assay. (See "Clinical use of coagulation tests", section on 'Causes of prolonged aPTT' and "Pathogenesis of antiphospholipid syndrome", section on 'Lupus anticoagulant'.)

Antiprothrombin antibodies can also occur. Antiprothrombin antibodies, which can cause significant clinical bleeding, usually bind to a nonactive portion of the molecule, resulting in accelerated clearance of prothrombin [90]. Technically, since these antibodies bind prothrombin and deplete it from the circulation rather than interfering with function, they are not inhibitors in the strict sense of the word. The inhibitor screen will be most consistent with a factor deficiency rather than an inhibitor, since the functional activity of prothrombin is not impaired in a clotting assay when additional prothrombin is added. (See "Clinical manifestations of antiphospholipid syndrome".)

The possible presence of prothrombin antibodies should be suspected if a patient with antiphospholipid antibodies develops bleeding rather than the expected thrombotic events [91,92]. An autoantibody directed against prothrombin has been well characterized in one patient who did not have a lupus anticoagulant [93]. It bound to a different epitope on the prothrombin molecule than described above, but still led to depletion of prothrombin without affecting prothrombin function.

Specific immunochemical measurement of the prothrombin concentration is required to establish the diagnosis.

Treatment of active bleeding consists of the use of fresh frozen plasma (FFP) to increase prothrombin levels. The recommended dose of FFP is 15 to 20 ml/kg, with a target prothrombin level >30 percent.

The modalities used to treat factor VIII inhibitors also may apply to patients with prothrombin inhibitors [91-94]. (See 'Treatment of factor VIII inhibitors' above.)

Thrombin (factor IIa) inhibitors — Inhibitors specific for human thrombin are quite rare, as one might expect given the multiple important functions of this enzyme. (See "Overview of hemostasis".)

The majority of antibodies directed against thrombin are detected incidentally in the absence of clinical bleeding because of prolongation of the thrombin clotting time (thrombin time, TT) and other clot-based anticoagulation assays performed for some other reason.

Thrombin antibodies typically arise after exposure to bovine thrombin found in topical thrombin or fibrin-glue preparations used in surgery [95-98]. These antibodies do not cause clinical bleeding unless they crossreact with human thrombin or, more often, when antibodies are also present against factor V, a common contaminant in bovine thrombin preparations [96-99]. (See "Fibrin sealants".)

In these patients with antibodies to bovine thrombin, the thrombin time is prolonged when using bovine reagent thrombin and is normal when the thrombin reagent is of human origin. The aPTT is prolonged and the prothrombin time is increased if factor V antibodies are also present [96,98]. The availability of human, rather than bovine, thrombin for fibrin sealant and topical thrombin preparations should decrease the incidence of thrombin antibody formation. (See "Clinical use of coagulation tests", section on 'Thrombin time (TT)'.)

Thrombin antibodies have also been described in several patients without exposure to exogenous thrombin. Most of these patients had a comorbid condition such as systemic lupus erythematosus, rheumatoid arthritis, or monoclonal gammopathy [100-102]; however, one patient had no obvious underlying disease [103]. The finding of autoantibodies to prothrombin in the setting of a lupus anticoagulant has been termed "lupus anticoagulant hypoprothrombinemia syndrome" (LAHPS) [104].

These inhibitors (autoantibodies to prothrombin or thrombin in individuals with rheumatologic or lymphoproliferative disorders) may cause severe bleeding [101,103]. Anecdotal treatment approaches include immunosuppression with prednisone, cyclophosphamide, other immunosuppressive agents, and/or plasma exchange [103,104]. Given the severity of bleeding that can occur in this setting, aggressive initial management including plasma exchange is probably warranted.

Factor V inhibitors — As with thrombin inhibitors, the majority of factor V inhibitors are autoantibodies that arise from exposure to topical fibrin glues or bovine thrombin preparations that are contaminated with bovine factor V [96-98,105]. One review evaluated 105 published cases of acquired factor V inhibitors [99]. The following results were noted:

●Forty-six (44 percent) were associated with bovine thrombin exposure. In the only prospective study that has evaluated this relationship, such exposure led to the generation of factor V inhibitors in 11 of 24 patients after cardiac surgery and 2 of 10 patients after neurosurgery, for a combined incidence of 38 percent [98]. Repeated exposure appeared to lead to more potent inhibitors.

●Twenty-four patients (23 percent) developed inhibitors following surgery without known exposure to bovine proteins. These antibodies showed specificity for human factor V, but can crossreact with bovine factor V. They do not prolong the thrombin time, in contrast to patients exposed to bovine thrombin.

●Other associated conditions included malignancy, autoimmune disease, the postpartum state, and congenital factor V deficiency. Nineteen had apparently idiopathic disease.

In one series of 12 patients (with and without bovine thrombin exposure), only the antibodies from patients with hemorrhagic manifestations inhibited factor V activity [17]. These antibodies usually recognize the C2 domain of the protein. Loss of this domain, which binds to the procoagulant phospholipid phosphatidylserine on activated platelets and endothelial cells [19], results in loss of procoagulant activity [18]. As noted above, some factor VIII antibodies are directed against the C2 domain on that protein [14,15].

A case of acquired factor V inhibitor after exposure to topical human thrombin has also been reported [106].

Clinical manifestations — The hemorrhagic manifestations associated with factor V inhibitors are variable, ranging from asymptomatic to life-threatening bleeding [17,96,97,99]. In the large review cited above, bleeding occurred in one-third of patients with inhibitors following exposure to bovine proteins, compared with two-thirds of other cases [99].

At least two factors contribute to this variability. First, bleeding is primarily associated with antibodies directed against the C2 domain of the protein [17]. Second, functionally important factor V is concentrated on platelets. One patient who did not have a severe bleeding disorder had an inhibitor that neutralized plasma factor V but not the less accessible platelet factor V [107]. Platelet factor V may be more important for assembly of the prothrombinase complex than circulating factor V [108]. (See "Overview of hemostasis".)

Diagnosis — The presence of a factor V inhibitor should be suspected in a patient with prolongation of the aPTT and PT but a normal thrombin time and a decreased factor V activity level [97]. However, the thrombin time may be prolonged in patients with bovine thrombin exposure and the production of antibodies to thrombin as well as to factor V. As described for factor VIII inhibitors, mixing studies should confirm the diagnosis and the inhibitor titer can be determined using the Bethesda assay.

Treatment — Acute severe bleeding due to a factor V inhibitor may be life-threatening. For those patients who survive the acute bleeding episode, the prognosis is generally excellent [99]. Most inhibitors disappear within a period of months; it is not clear if immunosuppressive therapy accelerates this response.

Treatment modalities for acutely bleeding patients include platelet transfusions, plasma exchange [98,99], and intravenous immune globulin (IVIG) [109,110]. For serious bleeding, we suggest the use of platelet transfusions and plasma exchange; such patients should also receive immunosuppressive therapy with prednisone and cyclophosphamide, as described above in patients with factor VIII inhibitors; rituximab has also been used successfully in a few patients with acquired factor V inhibitors refractory to other therapy [111]. (See 'Treatment of factor VIII inhibitors' above.)

Factor VII inhibitors — Antibodies directed against factor VII are rare and only a small number of cases have been reported [112-115]. As expected, the PT is prolonged but the aPTT is normal, since factor VII is not involved in either the intrinsic coagulation pathway or the final common pathway (table 2). Mixing studies will confirm the presence of an inhibitor; a low level or absence of residual factor VII activity, especially after prolonged incubation (one hour) is necessary to confirm the diagnosis.

The clinical manifestations in the few reported cases have ranged from minor bleeding to life-threatening events such as intracranial hemorrhage [114,115]. One such patient did not improve after IVIG but was successfully treated with cyclophosphamide, corticosteroids, and plasma exchange [114]. This response is similar to that described for severe bleeding with other acquired inhibitors.

Factor IX inhibitors — Acquired antibodies directed against factor IX in the absence of hemophilia B are also very rare; two cases have been reported in the postpartum period [116-119]. The natural history is similar to factor V inhibitors. The role of immunosuppressive therapy is uncertain, but one child seemed to respond to IGIV and dexamethasone [117].

Factor X inhibitors — An extremely limited number of patients have been described with documented inhibitors against factor X [120-122]. These patients have presented with the sudden onset of bleeding, prolonged PT and aPTT measurements, and transient, severe deficiency of factor X. There was often an antecedent acute respiratory infection, and inhibitors were not conclusively demonstrated.

Amyloidosis can lead to deficiency of factor X due to binding of the protein to amyloid fibrils [123-125]. In this disorder, other hemostatic defects probably also contribute to the bleeding [126]. (See "Clinical presentation, laboratory manifestations, and diagnosis of immunoglobulin light chain (AL) amyloidosis", section on 'Clinical presentation'.)

One well-characterized case of an acquired inhibitor involved the spontaneous appearance of an autoantibody in an older male who presented with hemoptysis, hematuria, and bleeding into the thigh after an intramuscular injection [122]. The patient was successfully treated with steroids with disappearance of the inhibitor within two weeks. It is unclear if the inhibitor would have resolved spontaneously had such therapy not been given.

Factor XI inhibitors — Factor XI autoantibodies not occurring in patients with congenital factor XI deficiency are often associated with systemic lupus erythematosus [127,128]. Affected patients usually have little or no bleeding despite a prolonged aPTT that does not fully correct in a mixing study. The relative lack of bleeding, which is also noted in congenital factor XI deficiency, probably reflects the tertiary role for factor XI in factor X activation, although bleeding can occur under certain circumstances. (See "Overview of hemostasis".)

Factor XIII inhibitors — Activated factor XIII stabilizes and crosslinks overlapping fibrin strands. Factor XIII inhibitors, which are exceedingly rare, can act by one of several mechanisms: inhibition of activation of factor XIII; interference with enzymatic function; or prevention of binding to fibrin [129-131]. (See "Overview of hemostasis", section on 'Continuation of the coagulation cascade'.)

The clinical hallmark of a factor XIII inhibitor is delayed bleeding after surgery or an invasive procedure, since the initial clot which does form is mechanically weak and unstable. Large spontaneous hematomas or intracranial hemorrhage can occur. Most patients are over age 50, and death from bleeding is not uncommon [129,132]. As an example, in a review of 33 cases, nine died of hemorrhage [132]. In a review of 39 patients with factor XIII activity <70 percent and bleeding, approximately one-third died in the hospital [133].

Factor XIII inhibitors can be associated with immune dysregulation. In one review, development of an anti-factor XIII antibody was associated with systemic lupus erythematosus in a third of cases [134]. Other associated conditions include other autoimmune disorders, lymphoid malignancies, and some drugs [132,134].

Factor XIII inhibitors produce a unique pattern on coagulation testing. The PT and aPTT, which measure fibrin generation, are normal, but the clot stability assay is abnormal, since the absence of fibrin crosslinking causes the clot to lyse much more rapidly in 5 molar urea, 2 percent acetic acid, or 1 percent monochloroacetic acid. In one review, clot solubility testing was used to make the diagnosis in half of the patients, with the remainder diagnosed by a variety of other tests (eg, Berichrom FXIII, which is a photometric assay for quantitative measurement of FXIII level) [132]. (See "Approach to the adult with a suspected bleeding disorder", section on 'Tests for defects in fibrin crosslinking or fibrinolysis'.)

Another distinguishing laboratory feature of a factor XIII inhibitor is that despite extensive clinical bleeding, the D-dimer is normal rather than elevated (as would be expected in severe bleeding). This is because D-dimer is formed by the action of XIIIa in crosslinking of the two D domains, which does not occur with factor XIII deficiency. (See "Clinical use of coagulation tests", section on 'Fibrin D-dimer'.)

Management typically involves immunosuppression; most patients have received immunosuppressive therapies, including prednisone, cyclophosphamide, plasma exchange, IVIG, rituximab, and/or infusion of factor XIII concentrates [135-139]. Reviews of patients with acquired factor XIII antibodies have reported eventual disappearance of the antibodies in up to half of cases [129,132].

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Acquired bleeding disorders".)

SUMMARY AND RECOMMENDATIONS

●Acquired hemophilia results from the development of autoantibodies to the coagulation factors. Antibodies to factor VIII are the most common, and their treatment has been more extensively studied than that of other acquired inhibitors. (See 'Factor VIII inhibitors' above.)

●Acquired antibodies to the clotting factors are diagnosed by mixing studies, in which addition of patient plasma to normal plasma causes the relevant clotting test to become abnormal due to the antibody. The clotting tests affected by each type of inhibitor are discussed above.

●The most common conditions associated with the development of acquired factor VIII inhibitors are rheumatic diseases, the postpartum period, malignancy, and some drugs. (See 'Epidemiology' above.)

●For patients with factor VIII inhibitors who have active bleeding, we suggest initial control of active bleeding using an activated prothrombin complex concentrate (aPCC; eg, factor VIII inhibitor bypassing activity [FEIBA]) or recombinant human factor VIIa (rFVIIa) rather than a human factor VIII product (Grade 2C). This applies to inhibitors of any titer. Human factor VIII generally cannot be given in high enough amounts to overcome the inhibitor. Recombinant porcine factor VIIIa is also an option. The choice of product depends on availability, initial or previous responses, and clinician preference. (See 'Control of active bleeding' above.)

●To eliminate factor VIII inhibitors we recommend the use of prednisone at an initial oral dose of 1 mg/kg per day in all patients (Grade 1B). Based upon weak evidence from the literature and our own experience, we suggest that oral cyclophosphamide (2 mg/kg per day) be added to the initial prednisone treatment regimen (Grade 2C). (See 'Eliminating the inhibitor' above.)

●Inhibitors of prothrombin are frequently associated with antiphospholipid antibodies. Treatment of active bleeding is done with fresh frozen plasma (FFP). Modalities associated with factor VIII antibodies may also be useful. (See 'Prothrombin (factor II) inhibitors' above and 'Treatment of factor VIII inhibitors' above.)

●Inhibitors of thrombin can develop in patients exposed to bovine thrombin, or in patients with autoimmune diseases. Antibodies provoked by exposure to exogenous thrombin do not always cause bleeding, while those associated with autoimmune disease can cause severe bleeding. Although only anecdotal information is available for treatment, we suggest immunosuppression and aggressive initial management including plasma exchange (Grade 2C). (See 'Thrombin (factor IIa) inhibitors' above.)

●Inhibitors of factor V tend to arise after exposure to exogenous clotting products such as thrombin glue. The clinical phenotype can be quite variable. For serious bleeding, we suggest the use of platelet transfusions and plasma exchange; such patients should also receive immunosuppressive therapy with prednisone and cyclophosphamide (Grade 2C). (See 'Factor V inhibitors' above.)

●Inhibitors of factors VII, IX, X, XI, and XIII are exceedingly rare. Suggested treatment for these conditions includes cyclophosphamide, corticosteroids, and plasma exchange (Grade 2C). Infusion of coagulation factor concentrates, where available (eg, factor XI and factor XIII) may also be helpful. (See 'Other coagulation factor inhibitors' above.)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Steven Coutre, MD (deceased), who contributed to an earlier version of this topic review.