Managing Traps and Pitfalls during Initial Steps of ECMO Retrieval Program Using Miniaturized Portable ECMO System: What Have We Learned from the First Year?

Objectives: Due to increasing evidence on improved outcomes using ECMO therapy for patients with heart or lung failure, several experienced ECMO centers launched ECMO retrieval services in recent years. However, compared with ECMO treatment in the own institution, ECMO implantation in foreign hospitals and transport of patients on ECMO support requires additional organizational, logistic and clinical performance. This study was designed to present initial mid-term results, traps and pitfalls of the newly launched ECMO retrieval program in our center.

Methods: From 01/2015 till 09/2016, a total of 22 patients received ECMO support as a part of our newly launched ECMO retrieval service. Each implantation was performed peripherally in percutaneous technique in foreign departments on ICU using miniaturized portable Cardiohelp System (Maquet, Rastatt Germany) followed by patient transport by road to our institution. Detailed analysis of logistic, organizational and clinical parameters as well as early and mid-term outcomes are presented.

Results: The mean age was 56 ± 15 years and 6 patients (27%) were female. Implantations were performed in 14 distant departments with the mean distance of 23 ± 18 km (maximum 60 km). 13 patients (59%) were cannulated under cardio-pulmonary resuscitation (CPR) with the mean duration of 59 ± 56 minute (maximum 180 minute), 4 (18%) required dialysis and 9 (41%) had multiorgan failure preoperatively. The mean lactate level was 12 ± 6 mmol/L. The median ECMO support duration was 56(19;93) hours, whereas 8 patients (36.4%) were weaned off support and discharged from hospital. Overall mid-term survival in patients without the need for CPR was 66.7% at one week and 44% at 1 month, 6 month and 1 year, whereas patients requiring CPR survived in 38.5% at one week, and 28.8% at one month, 6 month and 1 year (Log-Rank p = 0.294). Higher age (p = 0.022), lower platelet count before implantation (p = 0.018) and higher lactate after 24 hours of support (p = 0.011) were the only factors associated with higher mortality.

Conclusion: Our initial experience shows that launching an ECMO retrieval program in centers with sufficient ICU capacities and local ECMO experience is feasible and associated with acceptable “real world” results despite initial learning curve. Rapid logistical organization and team flexibility are the key points to ensure comparable survival of patients requiring prolonged CPR.

No conflict of interest has been declared by the author(s).