Predictors of Mortality and Discharge Location for Patients Supported by Venoarterial Extracorporeal Membrane Oxygenation
Venoarterial extracorporeal membrane oxygenation (VA-ECMO) is a form of advanced life support that is increasingly used to treat cardiac or pulmonary failure, however morbidity and mortality for patients supported by VA-ECMO remains high. Elucidation of predictors of mortality and discharge location for survivors may aid clinicians, patients, and families in prognostication for patients supported by VA-ECMO. Methods: We performed a retrospective cohort study using registry data from the Extracorporeal Life Support Organization. Inclusion criteria were adults supported by VA-ECMO from 2017-2022. We calculated the relative risk of inpatient mortality in relation to patient demographic and clinical characteristics. Among survivors, we calculated the relative risk of discharge to a facility in relation to patient demographic and clinical characteristics. Results: Data were extracted for 24,481 patients at 334 ESLO sites across North America. Patients were 33.2% female, 60.7% White, and had a mean age 55.6 years (SD 14.5). Inpatient mortality was 56%. The relative risk (RR) for inpatient mortality among patients in the oldest age category relative to youngest was 1.49 (CI 1.44 – 1.55). The age-adjusted RR for inpatient mortality for patients with a BMI 35 relative to 18.5 – 24.9 kg/m2 was 1.25 (CI 1.17 – 1.33), for patients treated as a bridge to transplant was 0.50 (CI 0.44 – 0.58), for patients requiring renal replacement therapy was 1.31 (CI 1.24 – 1.38), for patients with a pH <7.14 at initiation relative to normal was 1.42 (CI 1.32 – 1.53), and for patient with an ECMO flow >4.7 L/min 24 hours after initiation was 1.23 (CI 1.14 – 1.33). Among survivors, 63.5% were discharged to a facility rather than home. The relative risk for discharge to a facility among patients in the oldest age category relative to youngest was 1.53 (CI 1.45 – 1.61). The age-adjusted RR for discharge to a facility for patients with a BMI 35 kg/m2 was 1.21 (CI 1.12 – 1.32), for patients on ECMO for 21 days was 1.33 (CI 1.21 – 1.46), for patients who suffered major complications was 1.39 (CI 1.25 – 1.56), and for patients who ambulated relative to those who were immobile was 0.55 (CI 0.45 – 0.67). Conclusion: In this large cohort study of patients supported by VA-ECMO across North America, we found several patient demographic and clinical variables to be associated with inpatient mortality and discharge to a facility among survivors. These findings have potential to inform patients, family members, and clinicians regarding possible outcomes for patients started on VA-ECMO.