Introduction: In recent years, the use of extracorporeal membrane oxygenation (ECMO) has expanded beyond university centers and is now being performed at community hospitals with limited resources. There is currently no consensus approach to weaning from venoarterial ECMO (VA-ECMO) after cardiogenic shock, with multiple different methods supported in the literature. Many of these approaches are complicated and resource-intensive, requiring equipment or personnel that are not available in the community setting. In our community hospital, we developed a simplified ECMO weaning protocol, and this study aimed to assess the feasibility and efficacy of our protocol. Methods: In this retrospective single-center study, we evaluated the outcomes of patients who underwent our ECMO weaning protocol between May 2021 and December 2023. To qualify for our ECMO weaning protocol, the patient needed to be well perfused with stable settings on the VA-ECMO. Our ECMO weaning protocol was as follows: Maintain anticoagulation with a partial thrombin time (PTT) goal of ~50 seconds. The ECMO flow was weaned at 0.5 L/min every 30–60 min. Once the ECMO flow reached ~2 L/min, a bolus of fluid and inotrope was started. A transthoracic echocardiogram would be obtained to evaluate the left and right ventricular functions. Then, the ECMO flow rate was further decreased to 1.5 L/min. Based on the echocardiogram findings, a discussion would be held to evaluate whether the patient was a candidate for decannulation, if the patient required an additional mechanical-assisted device, or if they should be transferred for a higher level of care. Results: A total of 64 patients who underwent VA-ECMO at our institution were evaluated for decannulation from ECMO using our weaning protocol. In this group, 25 eventually passed our weaning trial, while 39 failed to pass our weaning trial at any point. Of the 25 patients who passed our trial, 19 were weaned off mechanical support, four were weaned to a left ventricular assist device (LVAD), and two were weaned to a right ventricular assist device (RVAD). Of the patients who underwent successful weaning, four patients experienced post-ECMO cardiogenic shock, leading to death or reinitiation of ECMO. Our protocol had a sensitivity of 95.5% (95% confidence interval (CI): 77.2% to 99.9%) with a specificity of 87.1% (95% CI: 70.2% to 96.4%). Conclusion: In a community hospital with limited resources, the existence of a clear ECMO weaning protocol that optimizes the use of available resources is essential. In our experience, this protocol was predictable and produced acceptable results.
Mangan et al. (Tue,) studied this question.