Case of acquired haemophilia a in Southeast Asia following COVID-19 vaccine

Background

Acquired haemophilia A (AHA) occurs due to the development of autoantibodies directed against clotting factor VIII (FVIII). This disease is commonly observed among two groups—pregnant women and the elderly.1 The risk is higher among those above 65 years old and those with malignancies, with a mortality rate approaching 20%.2 Accurate diagnosis and prompt treatment of AHA has been shown to reduce its bleeding mortality risk.3

AHA is known to be associated with autoimmune diseases, pregnancy, malignancies and medications, such as antibiotics and anticonvulsants. However, the majority of cases are idiopathic.4 Studies have reported an association between vaccinations and AHA, namely the seasonal influenza vaccine and H1N1 vaccination.5 6 Nonetheless, a definitive link between the vaccine and AHA remains to be proven.

Case presentation

A man in his 80s was seen in a district hospital clinic with a 4-day history of bruising over the upper and lower limbs. There was no other bleeding tendency. There was no recent history of falls or trauma or a family history of bleeding disorders. He had multiple comorbidities, which included type 2 diabetes mellitus, hypertension, dyslipidaemia, chronic kidney disease stage 3a, benign prostatic hyperplasia (BPH) and glaucoma in both eyes. He had a history of an ischaemic stroke 7 years previously with subsequent cognitive impairment. There was no history of COVID-19 infection. Medications included calcium carbonate, alfuzosin, cardiprin, pantoprazole, atorvastatin, bisoprolol and metformin. The patient had received his first dose of the Pfizer COVID-19 vaccine 2 weeks before the onset of symptoms. He had no other recent vaccination. On examination, his temperature was 36.8°C, blood pressure 128/68 mm Hg and pulse 73 beats per minute. There were bruises on the right posterior thigh and left upper limb. Blood results demonstrated an isolated prolonged activated partial thromboplastin time (aPTT) of 90 s (25.4–38.4). The platelet count, international normalised ratio and prothrombin time were normal. His creatinine level was 136 umol/L and liver function was normal. A compression bandage was applied to the haematoma and he was discharged and asked to return for review a week later. He was advised to defer the second dose of the COVID-19 vaccine.

Unfortunately, 5 days later, he presented to the emergency department with tiredness and a painful swollen left lower limb. He appeared frail. He had no constitutional symptoms. No recent falls or trauma were recorded. On examination, his temperature was 36.5°C, blood pressure 127/64 mm Hg, pulse 98 beats per minute, respiratory rate 20 breaths per minute and oxygen saturation 99% while breathing ambient air. There were multiple ecchymoses over his left cubital fossa, left wrist and right arm extending to the forearm (figure 1). There was a painful swelling on the anteromedial aspect of his left thigh with associated oedema (figure 2). The rest of the physical examination was normal. He had a mild cognitive impairment with a Mini-Mental State Examination score of 18/30.

Figure 1

Ecchymosis over the right medial aspect of the arm and forearm.

Figure 2

Ecchymosis over the left thigh with intramuscular haematoma.

Differential diagnosis

Senile purpura can occur in the elderly following minor trauma due to thin sun-damaged skin, especially on the forearms and dorsum of the hands. This patient, however, had an unprovoked left thigh haematoma with an isolated prolonged aPTT effectively excluding senile purpura.

Drug-induced bleeding is unlikely. Although on aspirin, he was not on any anticoagulants, including unfractionated heparin or direct thrombin inhibitors. An antiphospholipid antibody syndrome was unlikely due to his age, and this typically presents with thrombosis. There was no personal or family history of haemostatic disease. There were no features of connective tissue disease. Furthermore, he had a normal platelet count and liver function test. Disseminated intravascular coagulation was excluded in the absence of any infection or sepsis.

A mixing study was performed to differentiate a coagulation factor deficiency from the presence of factor inhibitors. In this case, the test showed a prolonged aPTT uncorrected even after 2 hours of incubation. This implies the presence of a clotting factor inhibitor. He had a low FVIII level, which was only 6.7% (62.6–165.3). FVIII inhibitor titre was detected at 7.5BU, confirming the diagnosis of AHA. His von Willebrand activity was normal.

Further investigations were performed to look for the aetiology of his AHA. Apart from the recent vaccination history, his initial COVID-19 rapid antigen test was negative and the chest X-ray was normal. Tumour markers and screening for autoantibodies were unremarkable (table 1). CT of thorax-abdomen-pelvis did not reveal any malignancy. His esophagogastroduodenoscopy and sigmoidoscopy showed no sign of malignancy.

Treatment

Oral tranexamic acid 500 mg was served immediately at the district hospital with no availability of the bypassing agent. He was transfused with two units of packed red blood cells and four units of fresh frozen plasma. He was then transferred to a specialist centre where aggressive treatment with intravenous methylprednisolone 500 mg daily for 3 days and a single dose of recombinant activated FVII (rFVIIa) 90 µg/kg were given to eliminate inhibitors and achieve haemostasis. The patient responded to this initial treatment and was started on azathioprine 100 mg daily subsequently. He was also commenced on high dose steroids (oral prednisolone 60mg daily) in divided doses to be tapered down over six weeks.

Concurrently, the patient had folate and vitamin B₁₂ deficiency as shown in table 1. He was given oral mecobalamin 500 µg three times a day, as intramuscular cyanocobalamin was contraindicated in this case.

Outcome and follow-up

Given the above, his second dose of the COVID-19 vaccine was deferred. Unfortunately, he developed Covid-19 pneumonia during the third week of tapering dose prednisolone. This resulted in readmission to hospital. Despite this, the AHA remained under control. There was a reduction in the haematoma size and dissipation of the ecchymosis. There was no bleeding episode and his aPTT remained within the normal range. The oral prednisolone was changed to intravenous dexamethasone as advised by the haematologist. This was continued for six days before being converted to oral dexamethasone. He was asked to continue on oral dexamethasone till the end of planned therapy for the AHA (see online supplemental table 1 for details of corticosteroid therapy). The azathioprine was continued at a dose of 100mg daily. He was subsequently discharged well. After 6 weeks of treatment with corticosteroids and azathioprine, his FVIII titres had normalised and FVIII inhibitors were undetectable. Corticosteroids were stopped, whereas oral azathioprine 100 mg daily was continued.

His clinical course was complicated by an episode of urosepsis and acute urinary retention 2 months after diagnosis. This responded well to intravenous antibiotics and he was scheduled for a renal ultrasound. There was no further bleeding episode and oral azathioprine was reduced to 50 mg daily. At the sixth month of follow-up, he remained in remission while on oral azathioprine 50 mg daily. The haematologist advised further continuation of the azathioprine for a total duration of nine months.