Extracorporeal cardiopulmonary resuscitation for ...

IJCA-24181; No of Pages 6 International Journal of Cardiology xxx (2016) xxx–xxx

Contents lists available at ScienceDirect

International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard

Extracorporeal cardiopulmonary resuscitation for refractory cardiac arrest: A multicentre experience☆ Mark Dennis a,b,⁎, Peter McCanny c, Mario D’Souza d, Paul Forrest a,f, Brian Burns a,e, David A Lowe c, David Gattas a,g, Sean Scott h, Paul Bannon a,i, Emily Granger j, Roger Pye c, Richard Totaro a,g, on behalf of the Sydney ECMO Research Interest Group a

Sydney Medical School, University of Sydney, Sydney, Australia Department of Cardiology, Royal Prince Alfred Hospital, Sydney, Australia c Department of Intensive Care, St Vincent's Hospital, Sydney, Australia d Sydney Local Health District, Clinical Research Centre, Australia e Greater Sydney Area Helicopter Emergency Medical Service, New South Wales Ambulance Service, Australia f Department of Cardiothoracic Anaesthesia, Royal Prince Alfred Hospital, Sydney, Australia g Department of Intensive Care, Royal Prince Alfred Hospital, Sydney, Australia h Department of Emergency Medicine, St Vincent's Hospital, Sydney, Australia i Institute of Academic Surgery, Royal Prince Alfred Hospital, Sydney, Australia j Department of Cardiothoracic Surgery, St Vincent's Hospital, Sydney, Australia. b

a r t i c l e

i n f o

Article history: Received 19 October 2016 Accepted 1 December 2016 Available online xxxx Keywords: Cardiopulmonary resuscitation CPR Extracorporeal membrane oxygenation ECMO ECPR Cardiac arrest ELS ECLS

a b s t r a c t Aim: To describe the ECPR experience of two Australian ECMO centres, with regards to survival and neurological outcome, their predictors and complications. Methods: Retrospective observational study of prospectively collected data on all patients who underwent extracorporeal cardiopulmonary resuscitation (ECPR) at two academic ECMO referral centres in Sydney, Australia. Measurements and main results: Thirty-seven patients underwent ECPR, 25 (68%) were for in-hospital cardiac arrests. Median age was 54 (IQR 47–58), 27 (73%) were male. Initial rhythm was ventricular fibrillation or pulseless ventricular tachycardia in 20 patients (54%), pulseless electrical activity (n = 14, 38%), and asystole (n = 3, 8%). 27 (73%) arrests were witnessed and 30 (81%) patients received bystander CPR. Median time from arrest to initiation of ECMO flow was 45 min (IQR 30–70), and the median time on ECMO was 3 days (IQR 1–6). Angiography was performed in 54% of patients, and 27% required subsequent coronary intervention (stenting or balloon angioplasty 24%). A total of 13 patients (35%) survived to hospital discharge (IHCA 33% vs. OHCA 37%). All survivors were discharged with favourable neurological outcome (Cerebral Performance Category 1 or 2). Pre-ECMO lactate level was predictive of mortality OR 1.35 (1.06–1.73, p = 0.016). Conclusions: In selected patients with refractory cardiac arrest, ECPR may provide temporary support as a bridge to intervention or recovery. We report favourable survival and neurological outcomes in one third of patients and pre-ECMO lactate levels predictive of mortality. Further studies are required to determine optimum selection criteria for ECPR. © 2016 Elsevier Ireland Ltd. All rights reserved.

1. Introduction Cardiac arrest remains the leading cause of sudden death in otherwise healthy adults [1] and accounts for almost half of all cardiac deaths [2]. Despite much research overall outcomes remain poor with out of hospital

☆ Work was performed at Royal Prince Alfred Hospital and St Vincent's Hospital, New South Wales, Australia. ⁎ Corresponding author at: Cardiology Department, Royal Prince Alfred Hospital, Missenden Road, Camperdown, NSW 2050, Australia. E-mail address: [email protected] (M. Dennis).

cardiac arrest (OHCA) survival ranging from 2 to 11% [1,3,4]. In-hospital arrest (IHCA) survival rates, although higher, are also unsatisfactory at 15–22% [5,6]. Moreover, significant morbidity occurs in patients who have survived prolonged cardiac arrest. Severe neurological deficits occur in up to 30–60% [3,7] of OHCA and 10–20% [5,6] of IHCA survivors. Extracorporeal membrane oxygenation (ECMO) can provide adequate perfusion to the brain and other vital organs, both during refractory cardiac arrest (RCA), while the underlying cause is treated, and following return of spontaneous circulation (ROSC). Several recent studies have suggested ECMO CPR (ECPR) strategies may improve outcomes in both out of OHCA [8–11] and IHCA [12–14] when compared to conventional cardiopulmonary resuscitation (CCPR) survival rates.

http://dx.doi.org/10.1016/j.ijcard.2016.12.003 0167-5273/© 2016 Elsevier Ireland Ltd. All rights reserved.

Please cite this article as: M. Dennis, et al., Extracorporeal cardiopulmonary resuscitation for refractory cardiac arrest: A multicentre experience, Int J Cardiol (2016), http://dx.doi.org/10.1016/j.ijcard.2016.12.003

2

M. Dennis et al. / International Journal of Cardiology xxx (2016) xxx–xxx

These promising results have led to recent guideline updates recognizing the potential benefit of ECPR in limited circumstances as well as the need for further research [15]. The use of ECPR for RCA in Australia is increasing. We report the first multi-centre Australian experience with ECPR in terms of survival, neurological outcome and associated complications. 2. Materials and methods 2.1. Study design We performed a multi-centre retrospective analysis of prospectively collected data, on all patients who underwent ECPR for refractory cardiac arrest up to 2016 at Royal Prince Alfred Hospital and St Vincent's Hospital, Sydney, Australia. Patient data is routinely collected on all cardiac arrest patients as part of ongoing arrest audit practices at both hospitals. Additional required information was obtained from patient medical paper records, NSW Ambulance logs and hospital electronic medical record systems. Ethics approval was obtained for the study protocol. 2.2. Study setting and population Royal Prince Alfred and St Vincent's Hospitals are academic, ECMO referral centres that provide ECMO retrieval support for New South Wales (NSW) and internationally to New Caledonia servicing in excess of 7 million people. St Vincent's Hospital is the designated heart and lung transplant centre for NSW. Both hospitals have developed ECPR response teams, consisting of cardiothoracic surgery, anaesthetic, perfusion and intensive care personnel. ECPR teams support Medical Emergency Teams (MET) for in hospital cardiac arrests and emergency physicians for out of hospital cardiac arrests transported to the emergency department. Patients with refractory cardiac arrest without return of spontaneous circulation (ROSC) were eligible for ECPR at the discretion of the treating arrest team. Patients who received ECMO support for cardiogenic shock, following ROSC were excluded from this analysis. 2.3. Procedures and equipment For refractory cardiac arrest, conventional CPR was continued with no or minimal interruption either by the medical team or mechanical chest compression device (LUCAS™2, Physio-control Inc.). Following an unfractionated heparin bolus of 5000 international units, cardiothoracic surgical or emergency physician/intensive care team members performed percutaneous venoarterial femoral cannulation using the Seldinger technique with ultrasound guidance. 17Fr or 19Fr cannulae were used for femoral (return) arterial cannulation and 23Fr or 25Fr cannulae were used for venous (access) cannulation depending on vessel size determined on ultrasonography. Cannulae positioning was confirmed by transoesophageal echocardiography or fluoroscopy. Rotaflow or Cardiohelp pumps were used with Quadrox-D oxygenators and heparin bonded circuits (Maquet, Getinge Group, Rastatt, Germany). In the absence of any contraindication to anticoagulation, an infusion of unfractionated heparin was subsequently commenced to achieve a target activated partial thromboplastin time (APTT) of 50–70 s. ECMO flows were maintained between 2.5 and 5 l per minute, titrated to a target mean arterial pressure of p65 mmHg and supplemented with additional inotropic and/or vasopressor support as required. Fresh gas flow was titrated to the patient's partial pressure of carbon dioxide (paCO2). Right radial artery access was obtained for continuous invasive blood pressure monitoring and to monitor oxygenation of the upper body. After ECMO support was established, the patient was transferred to the intensive care unit, cardiac catheterization laboratory or surgical theatre depending on the suspected arrest aetiology and agreed team plan. A 7–9 French femoral arterial backflow cannula was inserted either at time of cannulation or upon completion of appropriate diagnostic or therapeutic interventions to provide distal limb perfusion. Regular transthoracic and transoesophageal echocardiography was performed to assess ventricular function, aortic valve opening, pericardial effusions, thrombus and cannula positioning. Prior 2014 the standard of care for at both institutions was to therapeutically cool arrest patients to 33 °C. Post March 2014 target temperature management (core temperature of 36 °C) was employed for 24 h post cardiac arrest as per Therapeutic Temperature Management (TTM) protocol [16]. In the absence of pulsatility or aortic valve opening, additional inotropic support or reduction of flows, where possible, is attempted to assist in prevention and management of significant left ventricular distension. Left ventricular “venting” is not performed prophylactically. Patients were weaned from ECMO, based on clinical and echocardiographic assessment of haemodynamic and cardiac function. Following successful weaning, patients were decannulated and underwent open repair of the femoral artery. Successfully weaning was defined as successful disconnection from extracorporeal support without mortality in a 48-hour period. 2.4. Data collection Both hospitals maintain databases of prospectively collected data on all cardiac arrest patients and patients requiring ECMO support as part of existing quality assurance protocols. Patient records, once identified, were interrogated to ensure study inclusion criteria were met. Patient data was collated and cross referenced from Electronic medical records

(eMR), Intensive Care Department's IntelliVue Clinical Information Portfolio (ICIP) database, cardiac arrest notes and ambulance department records. 2.5. Measurements The primary outcome was survival to hospital discharge. Secondary outcomes included neurological status at hospital discharge, bleeding and thrombotic events, need for renal replacement and vascular complications. Neurological status was assessed according to the cerebral performance score (CPC). A favourable neurological outcome was classified as CPC scores 1 and 2 [17]. Bleeding was classified according to the Bleeding Academic Research Consortium (BARC) consensus report [18]. Bleeding events were then subdivided further into minor and major bleeding; minor bleeding being BARC bleeding groups b3 and major being bleeding events BARC group ≥3. Patient related thrombotic events including deep vein thrombosis (DVT) or Pulmonary Embolism (PE) confirmed on ultrasound or computed tomography (CT) and circuit thrombotic events, as defined by oxygenator or circuit change or haemolysis were noted. All patients had vascular ultrasonography of cannulated vessels routinely after decannulation. On coronary angiography, culprit coronary occlusion was determined by treating interventional cardiologist at time of procedure. Clinical characteristics, cardiovascular risk factors and morbidities were sought and reported. Initial laboratory test results, were taken at the time of cannulation. 2.6. Statistical analysis Baseline and follow-up categorical variables were summarized using number and percentages and a p-value for the chi-square test for the general association between the two groups and the corresponding levels of the variables. Baseline and follow-up numeric variables were summarized using the summary statistics n, median and interquartile range (IQR). A p-value for the parametric t-test for the difference in the means between two groups was also determined and presented. A value of b0.05 was considered significant. Logistic regression analyses were performed to determine the predictors of mortality. Selection of the variables for logistic regression is based on known clinical relevance to mortality prior to development logistic modelling. The statistics produced were the odds ratio, 95% confidence interval of the odds ratio and the overall p-value. All tests were completed with SAS v9.3 software, Cary NC.

3. Results 3.1. Baseline characteristics Baseline patient characteristics are shown in Table 1. From 2009 to 2016, 37 patients were treated with ECPR at a median age of 54 years (IQR 47–58). All patients with attempted ECPR were successfully cannulated and commenced on ECMO support. All patients received therapeutic hypothermia as per protocol. Significantly more patients who survived to hospital discharge had type 2 diabetes, hypertension or hypercholesterolemia (p = 0.04, p = 0.04, p = 0.02 respectively). Initial blood tests at time of ECMO insertion revealed a significant difference in lactate levels between survivors (5.15, IQR 4.1–10.1) and nonsurvivors 11.2 (IQR 8.6–15.0), p = 0.01. Two patients had Intra-aortic balloon pumps (IABP) in situ at commencement of extracorporeal support, no patients received an IABP once on ECMO support. 3.2. Cardiac arrest details Arrest aetiology and coronary intervention findings are summarized in Table 2. Twenty-five (76%) were IHCA and 12 (24%) OHCA. 27 (73%) arrests were witnessed with no difference in survivors or non-survivors, 9 (69%) vs. 18 (75%), p = 0.67. A majority of patients received bystander CPR 30 (81%) and most CPR commenced within 5 min of the arrest. Initial rhythm was ventricular tachycardia (VT) or ventricular fibrillation (VF) in 20 (51%) patients. The median time from arrest to commencement of ECMO flow was 45 min (IQR 37.5–50) in survivors compared to 55 (IQR 30–79) min in non-survivors (p = 0.42). When comparing IHCA and OHCA arrest to ECMO times there was no statistically significant difference for all patients (p = 0.08), survivors (p = 0.37) or nonsurvivors (p = 0.16). Acute coronary syndrome accounted for 11 (30%) of cardiac arrest causes, pulmonary embolism occurred in 5 (14%) of patients. After commencement with ECMO support 20 (54%) patients underwent coronary



3

Table 1 Demographic and baseline data. Variable Demographics Median age (year), (IQR) Male, n (%) Cardiovascular risk factors n (%) Current smoker Ex-smoker Type 2 diabetes Hypertension History of ischaemic heart disease Peripheral vascular disease Hypercholesterolemia History of congestive cardiac failure Pre arrest treatment n (%) Anticoagulated Thrombolysed Intra-aortic balloon pump Left ventricular assist device Vasopressor or inotrope Laboratory data Worst pre-ECMO pH, median (IQR) Worst pre-ECMO HCO3− (mmol/L), median (IQR) Worst pre-ECMO lactate (mmol/L), median (IQR) Haemoglobin (g/L), median (IQR) Creatinine (mmol/L), median (IQR) Platelets (109/L), median (IQR)

Total (n = 37)

Survivors (n = 13)

Non-survivors (n = 23)

p-Value

54 (47–58) 27 (73)

52 (48–58) 9 (69)

54 (47–58) 18 (75)

0.37 0.71

7 (19) 3 (8) 12 (32) 18 (49) 9 (24) 2 (5) 14 (38) 5 (14)

4 (31)

3 (13) 3 (13) 5 (21) 9 (38) 6 (25) 1 (4) 8 (33) 2 (8)

0.36

0.25 0.65 0.65 0.46 0.69 0.96 0.07 0.01 0.76 0.29 0.50

7 (54) 9 (69) 3 (23) 1 (8) 6 (46) 3 (23)

10 (27) 2 (5) 2 (5) 1 (3) 27 (73)

5 (38) 1 (8) 1 (8) 10 (77)

5 (21) 1 (4) 1 (4) 1 (4) 17 (71)

7.17 (6.94–7.24) 15.0 (12–19) 9.4 (5.5–13) 109 (92–132) 137 (92–149) 162 (117–252)

7.14 (6.92–7.27) 19.0 (14–21) 5.2 (4.1–10.1) 109 (100–125) 129 (103–151) 173 (122–271)

7.19 (6.94–7.24) 13.0 (9–16) 11.2 (8.6–15) 108 (85–134) 119 (88–135) 161 (104–241)

angiography, 9 of which had percutaneous coronary intervention and one patient underwent emergent coronary artery bypass grafting. There was no statistically significant difference between survivors and non-survivors in terms of percutaneous coronary intervention.

0.04 0.04 0.90 0.80 0.02 0.26

replacement therapy was required in 12 (32%) patients with no patients requiring ongoing dialysis on discharge. Thrombotic complications of pulmonary embolism, deep vein thrombosis, backflow clot, circuit change or haemolysis occurred in a total of 8 patients. No patients required venting of the left ventricle.

3.3. Primary outcomes – survival to discharge and neurological outcome 4. Discussion Thirteen patients (35%) survived to hospital discharge (Table 3). Median ECMO duration was 3 days (IQR 1–6) days with a median intensive care unit (ICU) stay of 9.5 (IQR 2–18) and total hospital length stay of 11 days (IQR 2–34). Survivors had a longer ICU and hospital stay than non-survivors (p = 0.03). Fifteen (41%) patients survived to ICU discharge and two of these patients died before hospital discharge. There was no significant difference in survival between patients suffering an out of hospital cardiac arrest versus in hospital cardiac arrest (31% vs. 69% p = 0.87). Survival outcomes were similar between the two institutions (36% Royal Prince Alfred Hospital vs. 35% at St Vincent's Hospital). All survivors had a favourable neurological outcome (CPC 1 or 2) at hospital discharge. There was no significant difference in survivors or favourable neurological outcome between those cooled using preMarch 2014 protocol vs. those cooled using the post-March 2014 protocol, p = 0.30. On logistic regression analysis only initial lactate was found to be predictive of mortality OR 1.35 (1.06–1.73, p = 0.01) Table 4. 3.4. Organ donation A total of 10 patients were referred for organ donation. Seven patients were deemed suitable and of these consent was gained in two. 1 patient did not die within the required timeframe and 1 patient went on to successful organ donation. 3.5. Complications Major bleeding events occurred in 14 (38%) patients with no significant difference between survivors and non-survivors 6 (46%) vs. 8 (33%), p = 0.44, Table 5 Supplementary material. 28 (76%) patients required red blood cell transfusion. Seven (19%) patients experienced ischaemic limb complications with one requiring amputation. Renal

Our study is the first multicenter study of ECPR outcomes in Australia. Thirty-seven patients between 2009 and 2016 underwent ECPR across two ECMO referral centres, with an overall survival rate of 35%, and excellent neurological outcomes in survivors. Recent prospective and retrospective studies have shown superior survival rates in RCA (both IHCA and OHCA) when managed with ECPR compared to conventional CPR (CCPR) [9,11,12,14,19,20]. The reported survival to discharge rates for ECPR in IHCA range from 15 to 60% [12,14,19,21–24] and for OHCA, 4–55% [8–11,19,23–27]. In a recent meta-analysis of propensity matched observational data the overall survival to discharge rate of ECPR was twice that of CCPR [13]. Our data are similar to these previously documented ECPR survival rates. Our cases were not managed according to a predefined management pathway. Bundled treatment programs including mechanical CPR, ECMO support and immediate coronary angiography have produced promising results with survival to discharge rates above 50% [19,26]. In our study acute coronary syndromes precipitated 30% of arrests and only 25% of patients went on to receive revascularisation. Both of these figures are substantially lower than other cohorts [8,12,20] and are likely due to the large proportion of IHCA patients that have a more heterogeneous spread of cardiac arrest etiologies than OHCA. In contrast to previous reports [23,24] we found no difference in survival rates between refractory OHCA and IHCA. Further, in our study there was no significant difference in median arrest to ECMO flow times between OHCA and IHCA, nor between those who survived and those who did not. This is significant from three perspectives: 1) it may represent selection bias for OHCA patients; with longer times from arrest to arrival at the emergency department, the less likely the patients may have been to be selected for ECMO support 2) there was still a significant time delay in placing a patient on ECMO support in hospital and 3) It supports the concept that any survival difference that may


4


Table 2 Arrest and intervention data Variable Arrest data Bystander CPR Arrest to CPR b5 min 5–10 min Unknown Time from arrest to ECMO (min), median (IQR) IHCA arrest to ECMO (min), median (IQR) OHCA arrest to ECMO (min), median (IQR) IHCA vs. OHCA arrest to ECMO time Witnessed n (%) Unclear if witnessed n (%) In hospital cardiac arrest (IHCA) n (%) Out of hospital cardiac arrest (OHCA) n (%) Pre-March 2014 cooling protocol Post-March 2014 cooling protocol Initial rhythm VT/VF Pulseless electrical activity Asystole Aetiology n (%) Acute coronary syndrome Pulmonary embolism Arrhythmia Other Angiography data n (%) Angiogram No disease Triple vessel disease Significant left main disease Single vessel Multi vessel Interventional data Balloon angioplasty Received stent First stent location Left main Left anterior descending Left circumflex Right coronary artery Second stent location Left anterior descending Posterior descending artery/posterolateral SVG to OM graft

Total (n = 37)

Survivors (n = 13)

Non-survivors (n = 23)

p-Value

30 (81)

10 (77)

20 (83)

0.56 0.49

28 (76) 1 (3) 8 (22) 45 (30–70) 40 (30–55) 63 (45–77) p = 0.08 27 (73) 6 (16) 25 (68) 12 (32) 15 (41) 22 (59)

9 (69) 4 (31) 4 (31) 45 (37.5–51) 42.5 (32.5–45) 51 (45–63.5) p = 0.37 9 (69) 3 (23) 9 (69) 4 (31) 7 (46) 6 (27)

19 (79) 1 (4) 4 (17) 55 (30–79) 40 (30–60) 70 (45.5–89) p = 0.16 18 (75) 3 (13) 16 (67) 8 (33) 8 (54) 16 (73)

19 (51) 14 (38) 3 (8)

9 (69) 4 (31)

11 (46) 10 (42) 3 (13)

0.25

11 (30) 5 (14) 3 (8) 18 (49)

5 (38) 1 (8) 1 (8) 6 (46)

6 (25) 4 (17) 2 (8) 12 (50)

0.79

20 (54) 4 (11) 7 (19) 4 (11) 4 (11) 8 (22)

8 (62) 2 (15) 3 (23) 2 (15) 1 (8) 4 (31)

12 (50) 2 (8) 4 (17) 2 (8) 3 (13) 4 (17)

0.50 0.60 0.66 0.62 0.42 0.55

1 (3) 8 (22)

1 (8) 3 (23)

5 (21)

2 (5) 4 (11) 1 (3) 1 (3)

1 (8) 1 (8) 1 (8)

2 (5) 1 (3) 1 (3)

1 (8)

0.42 0.75 0.41 0.67 0.87 0.30

0.57

2 (8) 3 (13)

1 (4) 1 (4) 1 (4)

CPR – cardiopulmonary resuscitation, ECMO – extracorporeal membrane oxygenation VT – ventricular tachycardia, VF – ventricular fibrillation, SVG – saphenous venous graft, OM – obtuse marginal artery.

be seen between IHCA and OHCA is driven by the difference in length of arrest to effective flow times [23]. Arrest to ECMO flow time has been shown to be predictive of outcomes [12,19,24,28,29] and very long arrest to ECMO times have had very poor outcomes [27]. We undertook additional analysis looking at survival with patients with an arrest to ECMO time of ≤ 60 min vs. those N 60 min. Despite having a trend towards increased survival with times ≤ 60 min it did not reach significance. With effective conventional CPR, only 25–30% of cardiac output is delivered [30], in contrast the initiation of ECMO provides effective coronary and organ perfusion. Shortening time to ECMO flow should become akin to the key performance indicator used in primary angioplasty of “door to balloon” time. Ambulance initiated ECPR calls; a dedicated ECMO team paging system and simulation training are all critical in mobilizing resources in the shortest possible time frames and have already been implemented at both institutions. Equally as important as survival in cardiac arrest patients is neurological outcome. Nagao et al. [31] and Reynolds et al. [32] have previously demonstrated the importance of return of spontaneous circulation on neurological outcomes. ECPR has been shown to increase the return of spontaneous circulation in animal [33] and clinical trials [12]. Despite this, concerns still exist over the possibility of survival being increased at the expense of significant neurological morbidity. Our study adds to

the increasing evidence that this is not the case. Although likely affected by selection bias, 100% of survivors having a CPC of 1 or 2 is surprising and reinforces the excellent neurological results seen in the prospective CHEER Study [19] and other trials [22]. Further the improved ECPR neurological outcomes seen in IHCA patients are maintained at 2 years of follow up [34] and OHCA ECPR neurological outcomes may also be better than matched CCPR populations [11,13]. Whilst not all prospective or retrospective ECPR trials have shown such dramatic neurological outcomes as in our cohort, they do not support an increase in survival at the expense of survivors with severe neurological deficit. Traditional prognostic factors of age, witnessed arrest, IHCA, bystander CPR, and initial rhythm type were not predictive of positive outcomes in our study. These factors are well described in CCPR however their contribution is more variable in the ECPR literature [20], but still important. Unexpectedly, risk factors for chronic coronary artery disease and cardiovascular events in our study were found to be more prevalent in survivors. Whilst this likely a Type 2 error, an alternate explanation may relate to the development of collateral blood flow prior to the arrest. In our study, pre-ECMO lactate level was found to be predictive of mortality. Lactate levels have been reported as a survival and neurological prognostic marker in CCPR [35]. Wang et al. [36]. retrospectively


M. Dennis et al. / International Journal of Cardiology xxx (2016) xxx–xxx Table 3 Outcome data. Variable Days on ECMO, median (IQR) Days in ICU, median (IQR)

Total Survivors Non-survivors p-Value (n = 37) (n = 13) (n = 23)

3 (1–6) 9.5 (2–18) Days in hospital, median (IQR) 11 (2–34) Survived to ICU Discharge, n (%) 15 (41) Survived to hospital discharge, 13 (35) n (%) Cerebral performance category (CPC) n (%) CPC 1 11 (30) CPC 2 2 (5) Cause of death Cardiac failure and multi 15 (41) organ failure Ischaemic gut and multi 1 (3) organ failure Hypoxic brain injury 5 (14) Subarachnoid haemorrhage 1 (3) Embolic cerebrovascular event 1 (3) Sepsis cardiac failure 1 (3)

3 (1–4) 20.0 (8.5–20) 34 (30–44) 13 (100) 13 (100)

2.5 (1–7.5) 4.5 (2–18)

0.57 0.03

4.5 (1–12)

b0.0001

2 (8)

11 (85) 2 (15)

5

numbers it is underpowered for outcome predictors, however this is an inherent limitation in many ECPR studies given the relative infrequency of ECPR cases 5. Conclusion ECPR survival rates across two major referral hospitals may provide improved survival when compared to previous conventional CPR survival rates, with improved neurological outcomes. Serum lactate may be a useful variable for prognostication of ECPR candidates and patients. Randomised control or well-designed multi-centre cohort studies are required to confirm the benefit and provide firmer guidelines on patient selection for ECPR. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.ijcard.2016.12.003.

15 (63) 1 (4)

Conflicts of interest

5 (21) 1 (4) 1 (4) 1 (4)

No conflicts of interest declared. No funding was applied for in this study. References

ECMO – extracorporeal membrane oxygenation, ICU – intensive care unit.

reviewed 340 patients with conventional CPR and found levels b9 mmol/l within 10 min of CPR was associated with survival and a non-linear inverse relationship with survival occurred as lactate increased. Anaerobic metabolism from tissue hypoxia and hypoperfusion contributes to the development of lactate [36] and therefore it may represent a surrogate marker of “no or low flow time” with inadequate tissue perfusion [37]. In the ECPR population, the prognostic value of lactate has had limited review. A paediatric ECPR study [38] showed higher pre-ECMO lactate was associated with increased mortality and a retrospective review of adult extracorporeal support found baseline serum lactate as the strongest predictor of outcome [39]. In contrast Leick et al. [28] did not find it significant. Our data supports the inclusion of lactate in a decision making paradigm although further research to clarify its weight and role is required. Complication rates are high with the invasive nature of ECMO and may be higher in ECPR given the time-pressured nature of implementation and concomitant administration of anticoagulants and antiplatelets. Major bleeding complications are seen in up to one third of all ECMO patients and are associated with worse outcomes [40]. Our bleeding rates are similar or less than other cohorts [19]. Distal limb ischaemia occurred in 19% of patients and 1 patient required an above knee amputation. Backflow cannulae implementation can be challenging and the timing of insertion varies between centres ensuring appropriate volume of cases by operators is key to minimizing complication rates. Our study has several limitations. Firstly, our study is retrospective and subject to patient selection bias. Secondly, despite having information on bystander CPR on most of our patients, the quality of this for OHCA patients is difficult to qualify. Finally, with relatively small patient

Table 4 Logistic regression analysis. Variable

Odds ratio

p-Value

Sex Location IHCA Witnessed arrest Age Arrest to ECMO flow Worst pre ECMO lactate Arrest to ECMO b60 min

0.76 (0.07, 8.26)

0.82

8.0 (0.63, 102.19) 0.35 (0.01, 10.05) 0.98 (0.91, 1.05) 0.99 (0.95, 1.03) 1.35 (1.06, 1.73) 0.33 (0.06, 1.88)

0.16 0.74 0.58 0.64 0.02 0.17

IHCA – in hospital cardiac arrest, ECMO – extracorporeal membrane oxygenation.

[1] Z.J. Zheng, J.B. Croft, W.H. Giles, G.A. Mensah, Sudden cardiac death in the United States, 1989 to 1998, Circulation 104 (2001) 2158–2163. [2] C.S. Fox, J.C. Evans, M.G. Larson, W.B. Kannel, D. Levy, Temporal trends in coronary heart disease mortality and sudden cardiac death from 1950 to 1999: the Framingham Heart Study, Circulation 110 (2004) 522–527. [3] M.J. Moore, B.M. Glover, C.J. McCann, N.A. Cromie, P. Ferguson, D.C. Catney, et al., Demographic and temporal trends in out of hospital sudden cardiac death in Belfast, Heart 92 (2006) 311–315. [4] S. Girotra, S. van Diepen, B.K. Nallamothu, M. Carrel, K. Vellano, M.L. Anderson, et al., Regional variation in out-of-hospital cardiac arrest survival in the United States, Circulation 133 (2016) 2159–2168. [5] S. Girotra, B.K. Nallamothu, J.A. Spertus, Y. Li, H.M. Krumholz, P.S. Chan, et al., Trends in survival after in-hospital cardiac arrest, N. Engl. J. Med. 367 (2012) 1912–1920. [6] Z.D. Goldberger, P.S. Chan, R.A. Berg, S.L. Kronick, C.R. Cooke, M. Lu, et al., Duration of resuscitation efforts and survival after in-hospital cardiac arrest: an observational study, Lancet 380 (2012) 1473–1481. [7] Y.J. Kim, S. Ahn, C.H. Sohn, D.W. Seo, Y.S. Lee, J.H. Lee, et al., Long-term neurological outcomes in patients after out-of-hospital cardiac arrest, Resuscitation 101 (2016) 1–5. [8] T.S. Ha, J.H. Yang, Y.H. Cho, C.R. Chung, C.M. Park, K. Jeon, et al., Clinical outcomes after rescue extracorporeal cardiopulmonary resuscitation for out-of-hospital cardiac arrest, Emerg. Med. J. (2016). [9] K. Maekawa, K. Tanno, M. Hase, K. Mori, Y. Asai, Extracorporeal cardiopulmonary resuscitation for patients with out-of-hospital cardiac arrest of cardiac origin: a propensity-matched study and predictor analysis, Crit. Care Med. 41 (2013) 1186–1196. [10] K. Nagao, N. Hayashi, K. Kanmatsuse, K. Arima, J. Ohtsuki, K. Kikushima, et al., Cardiopulmonary cerebral resuscitation using emergency cardiopulmonary bypass, coronary reperfusion therapy and mild hypothermia in patients with cardiac arrest outside the hospital, J. Am. Coll. Cardiol. 36 (2000) 776–783. [11] T. Sakamoto, N. Morimura, K. Nagao, Y. Asai, H. Yokota, S. Nara, et al., Extracorporeal cardiopulmonary resuscitation versus conventional cardiopulmonary resuscitation in adults with out-of-hospital cardiac arrest: a prospective observational study, Resuscitation 85 (2014) 762–768. [12] Y.S. Chen, J.W. Lin, H.Y. Yu, W.J. Ko, J.S. Jerng, W.T. Chang, et al., Cardiopulmonary resuscitation with assisted extracorporeal life-support versus conventional cardiopulmonary resuscitation in adults with in-hospital cardiac arrest: an observational study and propensity analysis, Lancet 372 (2008) 554–561. [13] S.J. Kim, H.J. Kim, H.Y. Lee, H.S. Ahn, S.W. Lee, Comparing extracorporeal cardiopulmonary resuscitation with conventional cardiopulmonary resuscitation: a metaanalysis, Resuscitation 103 (2016) 106–116. [14] J. Blumenstein, J. Leick, C. Liebetrau, J. Kempfert, L. Gaede, S. Gross, et al., Extracorporeal life support in cardiovascular patients with observed refractory in-hospital cardiac arrest is associated with favourable short and long-term outcomes: a propensity-matched analysis, Eur. Heart J. Acute Cardiovasc. Care (2015). [15] R.W. Neumar, M. Shuster, C.W. Callaway, L.M. Gent, D.L. Atkins, F. Bhanji, et al., Part 1: executive summary: 2015 American Heart Association guidelines update for cardiopulmonary resuscitation and emergency cardiovascular care, Circulation 132 (2015) S315–S367. [16] N. Nielsen, J. Wetterslev, T. Cronberg, D. Erlinge, Y. Gasche, C. Hassager, et al., Targeted temperature management at 33 °C versus 36 °C after cardiac arrest, N. Engl. J. Med. 369 (2013) 2197–2206. [17] R.O. Cummins, D.A. Chamberlain, N.S. Abramson, M. Allen, P.J. Baskett, L. Becker, et al., Recommended guidelines for uniform reporting of data from out-of-hospital cardiac arrest: the Utstein style. A statement for health professionals from a task force of the American Heart Association, the European Resuscitation Council, the


6


[18]

[19]

[20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

[28]

Heart and Stroke Foundation of Canada, and the Australian Resuscitation Council, Circulation 84 (1991) 960–975. R. Mehran, S.V. Rao, D.L. Bhatt, C.M. Gibson, A. Caixeta, J. Eikelboom, et al., Standardized bleeding definitions for cardiovascular clinical trials: a consensus report from the Bleeding Academic Research Consortium, Circulation 123 (2011) 2736–2747. D. Stub, S. Bernard, V. Pellegrino, K. Smith, T. Walker, J. Sheldrake, et al., Refractory cardiac arrest treated with mechanical CPR, hypothermia, ECMO and early reperfusion (the CHEER trial), Resuscitation 86 (2015) 88–94. C.H. Wang, N.K. Chou, L.B. Becker, J.W. Lin, H.Y. Yu, N.H. Chi, et al., Improved outcome of extracorporeal cardiopulmonary resuscitation for out-of-hospital cardiac arrest–a comparison with that for extracorporeal rescue for in-hospital cardiac arrest, Resuscitation 85 (2014) 1219–1224. L. Avalli, E. Maggioni, F. Formica, G. Redaelli, M. Migliari, M. Scanziani, et al., Favourable survival of in-hospital compared to out-of-hospital refractory cardiac arrest patients treated with extracorporeal membrane oxygenation: an Italian tertiary care centre experience, Resuscitation 83 (2012) 579–583. D. Fagnoul, F.S. Taccone, A. Belhaj, B. Rondelet, J.F. Argacha, J.L. Vincent, et al., Extracorporeal life support associated with hypothermia and normoxemia in refractory cardiac arrest, Resuscitation 84 (2013) 1519–1524. A. Haneya, A. Philipp, C. Diez, S. Schopka, T. Bein, M. Zimmermann, et al., A 5-year experience with cardiopulmonary resuscitation using extracorporeal life support in non-postcardiotomy patients with cardiac arrest, Resuscitation 83 (2012) 1331–1337. E. Kagawa, I. Inoue, T. Kawagoe, M. Ishihara, Y. Shimatani, S. Kurisu, et al., Assessment of outcomes and differences between in- and out-of-hospital cardiac arrest patients treated with cardiopulmonary resuscitation using extracorporeal life support, Resuscitation 81 (2010) 968–973. N. Morimura, T. Sakamoto, K. Nagao, Y. Asai, H. Yokota, Y. Tahara, et al., Extracorporeal cardiopulmonary resuscitation for out-of-hospital cardiac arrest: a review of the Japanese literature, Resuscitation 82 (2011) 10–14. D. Yannopoulos, J.A. Bartos, C. Martin, G. Raveendran, E. Missov, M. Conterato, et al., Minnesota Resuscitation Consortium's advanced perfusion and reperfusion cardiac life support strategy for out-of-hospital refractory ventricular fibrillation, J. Am. Heart Assoc. 5 (2016). M. Le Guen, A. Nicolas-Robin, S. Carreira, M. Raux, P. Leprince, B. Riou, et al., Extracorporeal life support following out-of-hospital refractory cardiac arrest, Crit. Care 15 (2011) R29. J. Leick, C. Liebetrau, S. Szardien, U. Fischer-Rasokat, M. Willmer, A. van Linden, et al., Door-to-implantation time of extracorporeal life support systems predicts mortality

[29]

[30] [31]

[32]

[33]

[34]

[35]

[36]

[37]

[38]

[39]

[40]

in patients with out-of-hospital cardiac arrest, Clin. Res. Cardiol. 102 (2013) 661–669. S.J. Han, H.S. Kim, H.H. Choi, G.S. Hong, W.K. Lee, S.H. Lee, et al., Predictors of survival following extracorporeal cardiopulmonary resuscitation in patients with acute myocardial infarction-complicated refractory cardiac arrest in the emergency department: a retrospective study, J. Cardiothorac. Surg. 10 (2015) 23. W.G. Barsan, R.C. Levy, Experimental design for study of cardiopulmonary resuscitation in dogs, Ann. Emerg. Med. 10 (1981) 135–137. K. Nagao, H. Nonogi, N. Yonemoto, D.F. Gaieski, N. Ito, M. Takayama, et al., Duration of prehospital resuscitation efforts after out-of-hospital cardiac arrest, Circulation 133 (2016) 1386–1396. J.C. Reynolds, B.E. Grunau, J.C. Rittenberger, K.N. Sawyer, M.C. Kurz, C.W. Callaway, The association between duration of resuscitation and favorable outcome after out-of-hospital cardiac arrest: implications for prolonging or terminating resuscitation, Circulation (2016). D. Stub, M. Byrne, V. Pellegrino, D.M. Kaye, Extracorporeal membrane oxygenation to support cardiopulmonary resuscitation in a sheep model of refractory ischaemic cardiac arrest, Heart Lung Circ. 22 (2013) 421–427. T.G. Shin, I.J. Jo, M.S. Sim, Y.B. Song, J.H. Yang, J.Y. Hahn, et al., Two-year survival and neurological outcome of in-hospital cardiac arrest patients rescued by extracorporeal cardiopulmonary resuscitation, Int. J. Cardiol. 168 (2013) 3424–3430. M.W. Donnino, L.W. Andersen, T. Giberson, D.F. Gaieski, B.S. Abella, M.A. Peberdy, et al., Initial lactate and lactate change in post-cardiac arrest: a multicenter validation study, Crit. Care Med. 42 (2014) 1804–1811. C.H. Wang, C.H. Huang, W.T. Chang, M.S. Tsai, P.H. Yu, Y.W. Wu, et al., Monitoring of serum lactate level during cardiopulmonary resuscitation in adult in-hospital cardiac arrest, Crit. Care 19 (2015) 344. D.L. Carden, G.B. Martin, R.M. Nowak, C.C. Foreback, M.C. Tomlanovich, Lactic acidosis as a predictor of downtime during cardiopulmonary arrest in dogs, Am. J. Emerg. Med. 3 (1985) 120–124. S.C. Huang, E.T. Wu, C.C. Wang, Y.S. Chen, C.I. Chang, I.S. Chiu, et al., Eleven years of experience with extracorporeal cardiopulmonary resuscitation for paediatric patients with in-hospital cardiac arrest, Resuscitation 83 (2012) 710–714. C. Jung, K. Janssen, M. Kaluza, G. Fuernau, T.C. Poerner, M. Fritzenwanger, et al., Outcome predictors in cardiopulmonary resuscitation facilitated by extracorporeal membrane oxygenation, Clin. Res. Cardiol. 105 (2016) 196–205. C. Aubron, A.C. Cheng, D. Pilcher, T. Leong, G. Magrin, D.J. Cooper, et al., Factors associated with outcomes of patients on extracorporeal membrane oxygenation support: a 5-year cohort study, Crit. Care 17 (2013) R73.
