Assessing a bolus-free heparin strategy for peripheral veno-arterial ECMO cannulation in patients with cardiogenic shock

It is difficult to strike the right balance between bleeding and thrombosis in patients with cardiogenic shock (CS) supported with a venoarterial extracorporeal membrane oxygenator (VA ECMO). Systemic anticoagulation with unfractioned heparin (UFH), aimed at reducing thrombosis of the circuit, continues to be the standard of care. However, there is currently no evidence supporting the recommendation for administering an UFH bolus during ECMO cannulation.

To explore whether an UFH-bolus-free strategy during peripheral VA ECMO cannulation might decrease the incidence of bleeding events for the first 24 hours without increasing thrombotic events.

It is a retrospective observational study of adult patients who underwent successful emergency femoral VA ECMO implantation between January 2020 and September 2024 in a tertiary hospital in Spain. Heparin-coated membranes and circuits were used. Patients were divided into two groups based on whether UFH bolus was administered during cannulation or not, besides cannulae flush with a 10 IU/mL UFH solution. Bleeding events were categorized based on the BARC scale.

Of the 59 VA ECMO patients, 35 (59.3%) received an UFH bolus during cannulation and 24 (40.7%) did not. The mean dose of the UFH bolus was 74 IU/kg. Patients in the UFH-bolus group had a higher BMI and they more frequently had a percutaneous coronary angioplasty performed and a coronary stent implanted in the same procedure (Table 1). There were no statistically significant differences in the number of patients that reached activated partial thromboplastin time (aPTT) ≥ 50 ms on admission (above the lower limit of the therapeutic anticoagulation range). A trend towards higher levels of aPTT in the first hours was observed in patients who received an UFH bolus (Figure 1).

A major bleeding event (BARC≥3) occurred within the first 24 hours in 15 patients (42.9%) of UFH-bolus patients compared to 6 (25.0%) in the no-UFH-bolus (adjusted RR 1.54; 95% CI: 0.64 - 3.70). Bleeding from vascular accesses were the most common in both groups. Two fatal bleeding events were recorded, both in the UFH-bolus group: intracranial haemorrhage and alveolar bleeding. No significant differences were observed between groups regarding the occurrence of thrombotic events during admission: 5 patients (14.3%) in the UFH-bolus group compared to 4 patients (16.7%) in the no-UFH-bolus (p = 0.803). There was no clinically significant thrombosis of the cannulae or ECMO circuit.