Welcome to Correspondence Sign in | Join | Help

Peri-operative extracorporeal cardiopulmonary resuscitation

Last post 27 Nov 2018, 2:48 AM by Jong-Hwan Lee. 1 replies.
To submit a new response, please Register or Sign in.
Sort Letters: Previous Next
  •  12 Nov 2018, 7:07 AM 2686

    Peri-operative extracorporeal cardiopulmonary resuscitation

    The retrospective observational study from Min et al. suggests peri-operative extracorporeal cardiopulmonary resuscitation (ECPR) is feasible, and may even be indicated, for cases of catastrophic intra-operative haemorrhage [1]. Given our positive experience of peri-operative ECPR, albeit limited to a single case [2], we welcome and support these results and conclusions. There are nevertheless several aspects of the study that we wish to comment on and we invite the authors response.

    Firstly, a priority should be to shorten the intra-operative CPR and low-flow times. Are the ECPR team called for all cases of intra-operative cardiac arrest at the study site or only after declaring the arrest as ‘refractory’? Was there an ECPR ‘crash team’ with a hospital-wide alert system (such as 2222)? Were members of this team immediately available and free from other clinical duties at all times? Was there a primed circuit ready for use at any time, or were circuits prepared as required? In setting up our ECPR programme, we have given attention to all these points along with establishing a cannulation and human factors simulation programme to reduce the low-flow time.

    Secondly, non-survivors appear to have been somewhat older than survivors. Were all patients suffering refractory cardiac arrest at the study site eligible for ECPR? If not, what were the minimum criteria for ECPR? For example, it is very unlikely that patients 1, 3, 4, 7, 8, 9, 11, 14, 15 and 18 would have met our criteria for ECPR. That said, patients 1 and 18 survived neurologically intact to hospital discharge. We believe that patient selection is key to a successful ECPR programme.

    Thirdly, three of the included patients received veno-venous extracorporeal membrane oxygenation (VV-ECMO), which is an unusual approach to intra-operative ‘cardiac arrest’. If this was for management of acute lung injury due to massive transfusion, can it still be termed ECPR? True ECPR is, arguably, the rapid institution of veno-arterial ECMO (VA-ECMO) for refractory cardiac arrest. Though we use urgent VA- or VV-ECMO in some circumstances, it is not correct to refer to this as ECPR. Grouping urgent VA- and VV-ECMO together with ECPR may, therefore, confuse the overall picture.

    Fourthly, did the authors find any difficulties instituting ECPR in the context of intra-operative haemorrhage? For example, the surgeons would need to step back from the operating table during cannulation and intravenous heparin would be required.

    Finally, there is little evidence about survival rates from intra-operative cardiac arrest treated with ECPR. If the VV-ECMO cases are removed from this analysis, we are left with a survival rate of 4/20 (20%), which is low, much lower than ECPR figures on the ELSO registry (29%) [3]. Would the authors not have expected their survival rates to be higher than, for example, out of hospital cardiac arrests treated with ECPR?

    Overall, we argue the key principles to improving outcomes for an ECPR service are reducing low flow times, patient selection, excellent CPR including the use of mechanical compression devices and well defined and implemented protocols.


    M. Charlesworth

    A. Ashworth

    Wythenshawe Hospital

    Manchester University NHS Foundation Trust

    Manchester, UK

    Email: mda05mc@gmail.com


    No external funding or conflict of interest. MC is Social Media Editor for Anaesthesia.



    1. Min JJ, Tay CK, Ryu DK et al. Extracorporeal cardiopulmonary resuscitation in refractory intra-operative cardiac arrest: an observational study of 12-year outcomes in a single tertiary hospital. Anaesthesia 2018; 73: 1515–23.

    2. Charlesworth M, Barker JM, Greenhalgh D, Ashworth AD. Peri-operative extracorporeal cardiopulmonary resuscitation: the defibrillator of the 21st century? A&A Practice 2018; 11: 87.

    3. ECLS registry report. July 2018 https://www.elso.org/Registry/Statistics.aspx (accessed 10/11/2018).

  •  27 Nov 2018, 2:48 AM 2697 in reply to 2686

    Peri-operative extracorporeal cardiopulmonary resuscitation - a reply

    We thank Charlesworth and Ashworth for their comments on our paper [1]. We had no predefined ECPR protocol before 2014; the decision to call the extracorporeal membrane oxygenation (ECMO) team to the operating room was made by the attending anaesthesiologist, the leader of intraoperative CPR, and final decision for ECPR was made in discussion with the ECMO team leader. In January 2014, we built a multidisciplinary ECMO team including critical care medicine, cardiology, and cardiovascular surgery, and developed shared protocols and institutional guidelines for ECMO activation and management [2]. The ECMO leader on-call is responsible for the decision to initiate ECMO, and the first available cardiovascular surgeon uses a pre-primed, ready-to-use ECMO circuit.

    We are involved in ongoing discussions with our ECMO team about the minimun criteria for ECPR, particularly around when not to initiate treatment. Generally, discussion occurs in the operating room once the ECMO team has been contacted. ECMO therapy is expensive and controversial in patients with haemorrhage, but can be life-saving in some patients who would have been very unlikely to survive without ECPR. We agree that appropriate patient selection is the key to a successful ECPR.

    We included veno-venous ECMO (VV-ECMO) in our analysis because 'ECPR' generally refers to the implantation of venous-arterial ECMO. In the absence of a clear and univocal definition of ECPR, we thought that CPR with the assistance of VV-ECMO for pulmonary resuscitation should be considered to be ECPR when the return of spontaneous circulation (ROSC) could not be sustained due to profound respiratory failure. In the three patients with severe acute respiratory failure in our study, immediate ROSC after conventional CPR was not sustained without VV-ECMO.

    We found that femoral ECMO procedures could be performed without interruping abdominal or haemostatic procedures, including compression or clamping. Any decision about ECMO was made only when the surgeon was confident of haemostasis. In some cases, the cardiovascular surgeon helped repar a vascular injury. The routine use of systemic anticoagulation could be dangerous in coagulopathic patient, necessiating point-of-care coagulation monitoring.

    We think the real survival rate may have been underestimated in our retrospective analysis. We did not review all cases of intraoperative CPR during the study periods, and so we could not assess the survival of patients who might have benefitted from ECPR correctly initiated during their intraoperative cardiac arrests.

    We agree with Charlesworth and Ashworth that improving patient outcomes using ECPR requires multidisciplinary teamwork, and involves well-defined and implemented protocols.

    J.-H. Lee

    Sungkyukwan University School of Medicine,

    Seoul, Korea.

    Email: jonghwanlee75@gmail.com


    No external funding and no conflicts of interest declared.



    1. Min JJ, Tay CK, Ryu DK, et al. Extracorporeal cardiopulmonary resuscitation in refractory intra-operative cardiac arrest: an observational study of 12-year outcomes in a single tertiary hospital. Anaesthesia 2018; 73: 1515-23.

    2. Na SJ, Chung CR, Choi HJ, et al. The effect of multidisciplinary extracorporeal membrane oxygenation team on clinical outcomes in patients with severe acute respiratory failure. Annals of Intensive Care 2018; 8: 31.

View as RSS news feed in XML

Our site uses cookies to improve your experience.You can find out more about our use of cookies in our standard cookie policy, including instructions on how to reject and delete cookies if you wish to do so.

By continuing to browse this site you agree to us using cookies as described in our standard cookie policy .