We report the case of a young woman with no significant medical history who was diagnosed with a postpartum acquired factor VIII inhibitor. Activated factor VIII serves as a cofactor for activated factor IX, and together they activate factor X, ultimately leading to generation of thrombin. Factor VIII is deficient in hemophilia A, an X-linked disorder. It can be corrected in the case of a simple deficiency by replacement of the missing factor. The presence of an inhibitor is more difficult to diagnose and to treat. Delayed recognition of the source of continued bleeding can lead to multiple transfusions and procedures.

Our patient’s problems began with an abortion induced by medication in late August 2021. When she presented approximately five weeks afterward, after much on-going blood loss at home, it was thought that she might have retained products of conception, and she underwent dilation and curettage. She had multiple admissions over the following weeks, with continued bleeding. She underwent uterine artery embolization and aberrant cervical vessel ligation. Continued bleeding necessitated a hysterectomy. She had difficulties with vaginal cuff dehiscence in the weeks following her hysterectomy, ultimately requiring laparotomy for an intra-abdominal hematoma. She required intensive care unit admission and was intubated with an open abdomen for a brief period of time. During this final admission, in early November 2021, the hematology consultant diagnosed the acquired factor VIII inhibitor.

Acquired factor VIII inhibitor (also known as acquired hemophilia) is the most lethal of factor inhibitors, with a mortality rate of 8-22%. Early recognition of the clues of its presence is of paramount importance.1 An acquired factor VIII inhibitor typically causes prolongation of the activated partial thromboplastin time (aPTT), as factor VIII is part of the intrinsic clotting pathway. However, prolongation of the prothrombin time (PT) can also be seen because of consumption of clotting factors in the face of on-going bleeding.2 Our patient’s aPTT was often > 150 seconds. Ultimately, PT prolonged to 22.2 seconds (INR 1.85).

A mixing study is needed to distinguish inhibitor from deficiency. When the affected patient’s plasma is mixed 1:1 with normal plasma, the abnormality in aPTT (or PT) should fully correct if the problem is a deficiency. Normal aPTT (or PT) only requires 50% factor activity (i.e., 100% from the normal, mixed plasma and ~0% from a patient with a severe deficiency). However, full correction will not occur in the setting of an inhibitor, which also impacts the normal plasma it is mixed with.2 Our patient’s mixing study aPTT partially corrected from 91.6 seconds to 56.5 seconds.

The mixing study should be checked immediately and after two hours of incubation at 37 degrees Celsius. Typically, because of the time it takes for the inhibitor to bind the factor VIII, the initial mixing study will correct, sometimes close to normal. However, the aPTT will prolong significantly at the two-hour mark.1,2 Two-hour incubation was not performed on our patient. It was not required, given that the initial mixing study did not correct to normal or near normal.

A mixing study can be performed even if a patient has recently received blood products (e.g., packed red blood cells [pRBCs] or fresh frozen plasma [FFP]). In fact, a patient who has received a large amount of FFP (e.g., 10-12 units = 3000-3600 mL) in a short period of time without correction of the aPTT/PT has essentially had an in vivo mixing study with failure to correct. Our patient’s aPTT after roughly ten units of FFP was 91.6 seconds. Further testing was not recommended by hematology when initially consulted during an October admission because of concern that the blood products already received would make results of much testing uninterpretable. However, the presence of an inhibitor could have been suspected based on the in vivo mixing study with failure to correct.

A factor VIII activity level and inhibitor titer should be checked to confirm the diagnosis. The inhibitor titer will be reported in Bethesda units (BUs). The value is derived from taking the reciprocal of the dilution at which 50% of the factor activity is restored. As an example, one part patient plasma in eight parts normal plasma yielding 50% factor VIII activity would mean an inhibitor titer of 8 BUs.2 Our patient’s factor VIII activity level was 0%, and her inhibitor titer was 128 BUs (i.e., dilution of one part patient plasma in 128 parts normal plasma required to yield 50% factor VIII activity).

Acquired factor VIII inhibitors occur in specific settings (associated conditions found in 50% of cases): autoimmune disorders, malignancies (solid tumors and blood cancers), infections (e.g., HIV and hepatitis B and C), and postsurgical and postpartum states. Some medications reported to be associated include penicillin and its derivatives, sulfa antibiotics, phenytoin, methyldopa, interferon-α, fludarabine, levodopa, and clopidogrel. Acquired factor VIII inhibitors occur most commonly in the elderly (higher prevalence of associated malignancies and medications), but there is a biphasic age distribution, with females contributing predominantly to the peak occurring between ages 20 and 30.1 Postpartum state is a common setting, and the inhibitor typically develops after parturition. However, it can occur during labor and cause maternal hemorrhage. There is a risk of neonatal hemorrhage if the inhibitor develops during pregnancy. Favorable outcomes are reported in 97% of postpartum cases, likely due to the good health of the patients and the transient nature of the “disease” state. The high overall mortality (8-22%) largely reflects the underlying disease states in which it is found and the advanced age of most of the patients.2

After diagnosis, treatment is focused on bypassing factor VIII in the clotting cascade and on eradicating the inhibitor. Desmopressin (to increase factor VIII release from platelets and endothelial cells) and factor VIII concentrates are not advisable strategies because inhibitors weak enough to be overcome by increasing the factor VIII level are rare.2,3 The most commonly used bypassing agents include activated prothrombin complex concentrates (e.g., FEIBA) and activated factor VII (e.g., NovoSeven). The efficacy of these agents in stopping bleeding should be judged clinically (i.e., by whether the patient continues to bleed), as aPTT is not affected because the agents act on the extrinsic pathway (FEIBA and NovoSeven) +/- the common pathway (FEIBA).4

Steroids (e.g., prednisone 1 mg/kg/day) and cyclophosphamide (1-2 mg/kg/day) are commonly used to eradicate the inhibitor, and the combination yields better results than steroids alone. Some patients with high titers may also require weekly rituximab. Intravenous immunoglobulin (IVIG) is a second-line treatment that can also help.1,2 Our patient initially received prednisone and cyclophosphamide, each at 1 mg/kg/day. She also received IVIG 400 mg/kg/day for five days and rituximab weekly for four weeks (begun in the hospital and completed in the clinic). While hospitalized, she received FEIBA and NovoSeven on an alternating basis until no longer bleeding. (Both of these medications are quite expensive and require on-going conversation with hospital pharmacy to ensure availability.) The aPTT dropped as the inhibitor slowly receded (not because of the bypassing agent). She was discharged from the hospital four weeks after immunosuppression begun, at which time she was without clinically evident bleeding.

In conclusion, the majority of cases of acquired factor VIII inhibitor will resolve, though some will not despite use of immunosuppressants. Currently, our patient’s prednisone has been weaned off (from starting dose of 100 mg/day), while she remains on cyclophosphamide 100 mg/day. Her aPTT now (five months after start of therapy) is down to 39.3 seconds, and her factor VIII assay is up to 44%, a substantial improvement from initial levels of > 150 seconds and 0%, respectively.

Consent was obtained from the patient prior to writing this case report.

Conflicts of interest

The authors have nothing to disclose regarding conflicts of interest.