Intra-cardiac thrombosis in the setting of heparin-induced t
Key clinical message is to know that HIT is a massive thrombotic storm. Atrial fibrillation cardioversion might not be safe in HIT using the standard 48-hour cutoff from arrhythmia onset. Also, the case serves as a reminder of how to suspect, diagnose, and treat HIT.

A 56-year-old man had been admitted to the hospital with heart failure exacerbation. He had been on telemetry since admission and had been in sinus rhythm. He had a loop recorder for the last three months with no evidence of atrial fibrillation. He had acute onset atrial fibrillation in the hospital, off-note he had been on enoxaparin for thromboprophylaxis since admission. He complained of calf pain on day 8, and a duplex lower extremities showed right common femoral deep venous thrombosis. Laboratory work is significant for a platelet count drop from 400 on day 4 to 130 on day 8 of hospital stay. The rest of the laboratory work was unremarkable.

Despite the concise duration of atrial fibrillation, we decided against cardioversion without a transoesophageal echo(TEE) as we were suspecting heparin-induced thrombocytopenia(HIT), and we were afraid of thrombotic complications. TEE showed a massive left atrial thrombus.

His 4T score was eight, which is a high probability, and platelet factor 4 ELISA IgG optical density came back positive at 2.2, which confirms a diagnosis of HIT. He was started on bivalirudin and was later shifted to apixaban for chronic atrial fibrillation anticoagulation.

The key clinical message is to know that HIT is a massive thrombotic storm.1 Atrial fibrillation cardioversion might not be safe in HIT using the standard 48-hour cutoff from arrhythmia onset. Also, the case serves as a reminder of how to suspect, diagnose, and treat HIT.

Source: https://onlinelibrary.wiley.com/doi/10.1002/ccr3.4677?af=R
Like
Comment
Share