INTRODUCTION

There has been an increasing number of extracorporeal membrane oxygenation (ECMO) centers in the past few decades. Bleeding and thrombosis are common complications of extracorporeal membrane oxygenation. In adult ECMO patients, pre-existing conditions like coronary artery disease and innate bleeding disorders may predispose and make adult-ECMO patients even more vulnerable to these complications. Critical illness, factors related to the ECMO circuit itself (anticoagulation, prothrombotic surfaces), and prophylactic use of heparin also increases the risk of bleeding and thrombosis complications including cannula or surgical bleeding, organ bleeding or intracranial hemorrhage. Despite this, the incidence and contributing factors for these complications in adult populations remain unclear. The increasing availability and use of ECMO in the adult population mandates a need for research in this population.

METHODS

We conducted a retrospective study from 2015-2019 at Atrium Health in Charlotte, NC. A total of 47 patients were extrapolated all of whom required ECMO secondary to cardiogenic or pulmonary causes. Data collected included demographics, ECMO indication, mode of ECMO, anticoagulation regimen, thrombotic/bleeding events. Patients who died or were decannulated within 24 hours were excluded, as were post-cardiac surgery patients due to innate altered hemostatic profiles.

RESULTS

A total of 47 patients were included in the analysis with a median age of 39 years. The cohort was made up of 17 females and 30 males, and racially, there were 24 White, 21 Black, and 2 patients from other backgrounds. Approximately 40.4% of the patients experienced thrombotic events, while 61.7% had bleeding events. Laboratory results showed a mean hemoglobin level of 12.28 g/dL, mean prothrombin time (PT) of 17.8 seconds, mean platelet count of 189,000/µL, and mean serum creatinine of 1.86 mg/dL. After starting ECMO, 43 out of 47 patients were given heparin for anticoagulation. There was no significant effect of gender, race, comorbidities, or laboratory results that could be predictive of thrombotic or bleeding complications.

CONCLUSION

In this retrospective study, no statistically significant associations were identified for thrombotic or bleeding complications. This indicates that other, possibly unmeasured elements related to ECMO, such as specific components of the circuit or inflammatory reactions might play a role in increasing risks of bleeding and thrombosis. It is also possible that conventional comorbidities or initial characteristics might not be good predictors. However, this study is limited by its small sample size (n = 47), which could compromise our ability to identify significant associations and may limit how broadly we can apply these findings. There is a need for larger, multicenter study to assess for risk factors.

This content is only available as a PDF.
2025
Sign in via your Institution