Advertisement

Extracorporeal life support in cardiogenic shock: Impact of acute versus chronic etiology on outcome

Open ArchivePublished:February 28, 2015DOI:https://doi.org/10.1016/j.jtcvs.2015.02.043

      Abstract

      Background

      The role of extracorporeal life support (ECLS) in primary cardiogenic shock (PCS) is well established. In this study, we evaluated the impact of etiology on outcomes.

      Methods

      Between January 2009 and March 2013, we implanted a total of 249 patients with ECLS; we focused on 64 patients for whom peripheral ECLS was the treatment for PCS. Of these, 37 cases (58%) were “acute” (mostly acute myocardial infarction: 39%); 27 (42%) had an exacerbation of “chronic” heart failure (dilated cardiomyopathy: 30%; post-ischemic cardiomyopathy: 9%; and congenital: 3%).

      Results

      In the group with chronic etiology, 23 patients were bridged to a left ventricular assist device (52%) or heart transplantation (33%). In the group with acute etiology, ECLS was used as a bridge-to-transplantation in 3 patients (8%), a bridge-to-bridge in 9 (24%), and a bridge-to-recovery in 18 (49%). One patient in each group was bridged to conventional surgery. Recovery of cardiac function was achieved in only the group with acute primary cardiogenic shock (18 vs 0 patients, P = .0001). A mean flow during support of ≤60% of the theoretic flow (body surface area × 2.4) was a predictor of successful weaning (P = .02). Median duration of ECLS support was 7 days (range: 2-11.5 days). Nine patients (14%) died during support; 30-day overall survival was 80% (51 of 64 patients); and 59% of patients were discharged, in whom survival at 48 months was 90%. Thirty-day survival was correlated with duration of ECLS support.

      Conclusions

      In “chronic” heart failure, ECLS represents a bridge to a ventricular assist device or heart transplantation, whereas in “acute” settings, it offers a considerable chance of recovery, and is often the only required therapy.

      Key Words

      Abbreviations and Acronyms:

      AMI (acute myocardial infarction), A-PCS (acute primary cardiogenic shock), C-PCS (chronic primary cardiogenic shock), ECLS (extracorporeal life support), IABP (intra-aortic balloon pump)

      CTSNet classification

      Figure thumbnail fx1
      Overall survival of patients with primary cardiogenic shock with acute or chronic etiology.
      ECLS in C-PCS represents a bridge to VAD or heart-transplantation, whereas in A-PCS it offers chance of recovery. Better outcomes are observed with shorter and partial support.
      ECLS is a bridge-to-life and should be considered a first line treatment in PCS. Therapeutic strategy should be tailored considering the different outcomes with respect to etiology. Moreover, maintaining flows below 60% of the theoretical requirement and minimizing the duration of support appears to be the strategy which offers the best chance of survival.
      See Editorial Commentary page 341.
      Acute cardiogenic shock is a condition that continues to have very high mortality despite advances in medical therapy.
      • Goldstein D.
      • Neragi-Miandoab S.
      Mechanical bridge to decision: What are the options for the management of acute refractory cardiogenic shock?.
      • Koerner M.M.
      • Jahanyar J.
      Assist devices for circulatory support in therapy-refractory acute heart failure.
      • Russo C.F.
      • Cannata A.
      • Lanfranconi M.
      • Bruschi G.
      • Milazzo F.
      • Paino R.
      • et al.
      Veno-arterial extracorporeal membrane oxygenation using Levitronix centrifugal pump as bridge to decision for refractory cardiogenic shock.
      Conventional treatment typically comprises inotrope infusions, vasopressors, and an intra-aortic balloon pump (IABP).
      • Goldstein D.
      • Neragi-Miandoab S.
      Mechanical bridge to decision: What are the options for the management of acute refractory cardiogenic shock?.
      • McMurray J.J.
      • Adamopoulos S.
      • Anker S.D.
      • Auricchio A.
      • Böhm M.
      • Dickstein K.
      • et al.
      ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure 2012.
      When circulatory instability is refractory to these treatments, mechanical circulatory support represents the only hope for survival, as indicated by current guidelines.
      • Russo C.F.
      • Cannata A.
      • Lanfranconi M.
      • Bruschi G.
      • Milazzo F.
      • Paino R.
      • et al.
      Veno-arterial extracorporeal membrane oxygenation using Levitronix centrifugal pump as bridge to decision for refractory cardiogenic shock.
      • McMurray J.J.
      • Adamopoulos S.
      • Anker S.D.
      • Auricchio A.
      • Böhm M.
      • Dickstein K.
      • et al.
      ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure 2012.
      As most of these patients present with critical circulatory instability that requires urgent or emergent therapy, the chosen mechanical assistance should be able to be rapidly and easily implanted. For this reason, extracorporeal life support (ECLS) is the ideal “bridge to life,” and increasingly, it is used to keep patients alive while the optimal therapeutic management is determined (“bridge to decision”). Management may follow 1 of 3 courses: (1) a “bridge to recovery,” which involves patient recovery and weaning from ECLS; (2) a “bridge to transplant,” which involves direct orthotopic heart transplantation; and (3) a “bridge to bridge,” which involves placement of a ventricular assist device or total, artificial, longer-term support.
      • Marasco S.F.
      • Lukas G.
      • McDonald M.
      • McMillan J.
      • Ihle B.
      Review of ECMO (extra corporeal membrane oxygenation) support in critically ill adult patients.
      • Bermudez C.A.
      • Rocha R.V.
      • Toyoda Y.
      • Zaldonis D.
      • Sappington P.L.
      • Mulukutla S.
      • et al.
      Extracorporeal membrane oxygenation for advanced refractory shock in acute and chronic cardiomyopathy.
      Several large reports have been made on use of ECLS as a mechanical support in postcardiotomy patients
      • Russo C.F.
      • Cannata A.
      • Lanfranconi M.
      • Bruschi G.
      • Milazzo F.
      • Paino R.
      • et al.
      Veno-arterial extracorporeal membrane oxygenation using Levitronix centrifugal pump as bridge to decision for refractory cardiogenic shock.
      • Formica F.
      • Avalli L.
      • Martino A.
      • Maggioni E.
      • Muratore M.
      • Ferro O.
      • et al.
      Extracorporeal membrane oxygenation with a poly-methylpentene oxygenator (Quadrox D). The experience of a single Italian centre in adult patients with refractory cardiogenic shock.
      • Rastan A.J.
      • Dege A.
      • Mohr M.
      • Doll N.
      • Falk V.
      • Walther T.
      • et al.
      Early and late outcomes of 517 consecutive adult patients treated with extracorporeal membrane oxygenation for refractory postcardiotomy cardiogenic shock.
      • Hei F.
      • Lou S.
      • Li J.
      • Yu K.
      • Liu J.
      • Feng Z.
      • et al.
      Five-year results of 121 consecutive patients treated with extracorporeal membrane oxygenation at Fu Wai hospital.
      ; but relatively few, most of which are small case-series studies, have focused on the role of ECLS in primary, cardiogenic shock (PCS) outside of the postcardiotomy setting.
      • Arlt M.
      • Philipp A.
      • Voelkel S.
      • Camboni D.
      • Rupprecht L.
      • Graf B.M.
      • et al.
      Hand-held minimised extracorporeal membrane oxygenation: a new bridge to recovery in patients with out-of-centre cardiogenic shock.
      • Tang G.H.
      • Malekan R.
      • Kai M.
      • Lansman S.L.
      • Spielvogel D.
      Peripheral venoarterial extracorporeal membrane oxygenation improves survival in myocardial infarction with cardiogenic shock.
      • Moraca R.J.
      • Wanamaker K.M.
      • Bailey S.H.
      • McGregor W.E.
      • Murali S.
      • Benza R.
      • et al.
      Salvage peripheral extracorporeal membrane oxygenation using Cobe Revolution centrifugal pump as a bridge to decision for acute refractory cardiogenic shock.
      • Chen Y.S.
      • Yu H.Y.
      • Huang S.C.
      • Chiu K.M.
      • Lin T.Y.
      • Lai L.P.
      • et al.
      Experience and result of extracorporeal membrane oxygenation in treating fulminant myocarditis with shock: What mechanical support should be considered first?.
      We present the results of our center's experience in the treatment of PCS with ECLS as a bridge to decision. Furthermore, we evaluated the impact of etiology on patient outcomes by comparing refractory PCS secondary to acute myocardial infarction (AMI), myocarditis, pulmonary embolism, and postpartum cardiomyopathy with acute decompensation of a chronic cardiomyopathy, including dilated cardiomyopathy, ischemic cardiomyopathy, and adult congenital heart diseases. In addition, we analyzed whether duration and magnitude of support predict weaning and survival.

      Methods

       Patients

      Between January 2009 and March 2013, a total of 249 patients were treated with ECLS in our center. We prospectively analyzed 64 patients implanted with the PLS (MAQUET, Cardiopulmonary AG, Hirrlingen, Germany) ECLS system for treatment of cardiogenic shock refractory to maximal inotropic and vasopressor therapy and treatment with IABP therapy, excluding postcardiotomy, pediatric, and respiratory supports. We divided patients into 2 groups, according to etiology. A total of 37 were assigned to the group with acute primary cardiogenic shock (A-PCS), for whom the primary cause was an acute event in a previously healthy heart, including 26 cases of AMI, 4 of myocarditis, 6 of pulmonary embolism, and 1 of postpartum cardiomyopathy. In the second group were 27 patients with chronic primary cardiogenic shock (C-PCS), in whom the etiology was an acute deterioration of a chronic cardiomyopathy, including 20 who had dilated cardiomyopathy, 5 who had ischemic cardiomyopathy (C-PCS), and 2 who had grownup congenital heart diseases (Table 1). Written informed consent was obtained from all patients, and the study was approved by our institutional review board.
      Table 1Etiologies leading to ECLS implantation, in the overall population, the group who had A-PCS, and the group who had C-PCS
      EtiologyOverall (n = 64)A-PCS (n = 37)C-PCS (n = 27)
      AMI26 (41)26 (70)0
      Myocarditis4 (6)4 (11)0
      Pulmonary embolism6 (9)6 (16)0
      Postpartum CM1 (2)1 (3)0
      Dilated CM20 (31)020 (74)
      Ischemic CM5 (8)05 (19)
      Congenital2 (3)02 (7)
      Values are n (%). A-PCS, Acute primary cardiogenic shock; C-PCS, chronic primary cardiogenic shock; AMI, acute myocardial infarction; CM, cardiomyopathy.

       Criteria for Extracorporeal Life Support Installation

      Patients who met the criteria of profound cardiogenic shock due to pump failure were candidates for ECLS. Profound shock was defined as: systolic blood pressure of <75 mm Hg; a cardiac index of ≤1.8 L/minute/m2, with left ventricular end-diastolic pressure of >20 mm Hg, despite multiple, high-dose, intravenous inotropic agents (dopamine ≥10 μg/kg/minute, dobutamine ≥10 μg/kg/minute, epinephrine ≥0.1 μg/kg/minute, and norepinephrine ≥0.1 μg/kg/minute) and/or IABP treatment, in association with clinical signs of pulmonary congestion; and impaired end-organ perfusion (renal, respiratory, and hepatic failure and altered mental status).
      • Tsao N.W.
      • Shih C.M.
      • Yeh J.S.
      • Kao Y.T.
      • Hsieh M.H.
      • Ou K.L.
      • et al.
      Extracorporeal membrane oxygenation-assisted primary percutaneous coronary intervention may improve survival of patients with acute myocardial infarction complicated by profound cardiogenic shock.
      The diagnosis of impaired end-organ function required ≥1 of the following: altered mental status; cold, clammy skin and extremities; oliguria with urine output of <30 ml per hour; respiratory failure, defined as partial pressure of oxygen in arterial blood <60 mm Hg and/or partial pressure of carbon dioxide in arterial blood >45 mm Hg (type I-II), or need for mechanical ventilation in the setting of cardiogenic shock (type IV); serum lactate level >2.0 mmol per liter; bilirubin or transaminases >3 times the upper normal limit set by the local laboratory; and multiorgan failure, defined as failure of ≥2 organs, in addition to cardiac dysfunction.
      • McMurray J.J.
      • Adamopoulos S.
      • Anker S.D.
      • Auricchio A.
      • Böhm M.
      • Dickstein K.
      • et al.
      ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure 2012.

      Extracorporeal Life Support Organization Guidelines for cardiopulmonary extracorporeal life support, version 1.3. Available at: www.elso.org/Resources/Guidelines.aspx. Accessed March 1, 2014.

      The only absolute contraindication to ECLS was presence of severe neurologic involvement after arrest (significant anisocoria and signs of decerebration or focality); age >75 years and severe peripheral vascular disease were considered relative contraindications.
      • Bermudez C.A.
      • Rocha R.V.
      • Toyoda Y.
      • Zaldonis D.
      • Sappington P.L.
      • Mulukutla S.
      • et al.
      Extracorporeal membrane oxygenation for advanced refractory shock in acute and chronic cardiomyopathy.

       Extracorporeal Life Support System

      The system used in our institute was the PLS.
      • Formica F.
      • Avalli L.
      • Martino A.
      • Maggioni E.
      • Muratore M.
      • Ferro O.
      • et al.
      Extracorporeal membrane oxygenation with a poly-methylpentene oxygenator (Quadrox D). The experience of a single Italian centre in adult patients with refractory cardiogenic shock.
      • Arlt M.
      • Philipp A.
      • Voelkel S.
      • Rupprecht L.
      • Mueller T.
      • Hilker M.
      • et al.
      Extracorporeal membrane oxygenation in severe trauma patients with bleeding shock.
      This system has a portable, “all-in-one” design, including an oxygenator (Quadrox D), a centrifugal pump (Rotaflow), and heparin-coated tubes, as well as an optional heat exchanger, with specific features to minimize thrombotic risk.
      • Formica F.
      • Avalli L.
      • Martino A.
      • Maggioni E.
      • Muratore M.
      • Ferro O.
      • et al.
      Extracorporeal membrane oxygenation with a poly-methylpentene oxygenator (Quadrox D). The experience of a single Italian centre in adult patients with refractory cardiogenic shock.
      • Arlt M.
      • Philipp A.
      • Voelkel S.
      • Rupprecht L.
      • Mueller T.
      • Hilker M.
      • et al.
      Extracorporeal membrane oxygenation in severe trauma patients with bleeding shock.
      Furthermore, it has obtained CE (European conformity) approval for 14 days of support.

       Extracorporeal Life Support Placement

      In our institute, when feasible, we opted to implant ECLS with the patient awake and breathing independently, with local anesthetic, at the bedside. Cannulation was performed using a percutaneous veno-arterial Seldinger technique, with a venous drainage cannula (18F to 28F) placed in the femoral vein, and an arterial return cannula (18F to 22F) placed in the femoral artery. When this strategy was unsuccessful or contraindicated (eg, by significant peripheral vascular disease, with small femoral vessels), the patient was taken to the operating theater, sedated, and intubated.
      At the surgeon's discretion, 1 of the following cannulation techniques was chosen: subclavian or inguinal dissection and anastomosis of a 6- to 10-mm Dacron vascular graft (FlowWeave, Jotec GmbH, Hechingen, Germany) onto the subclavian or femoral artery for cannulation. In a single case, a sternotomy and central cannulation (a venous cannula in the right atrium and an arterial cannula in the ascending aorta) were used. Limb ischemia subsequent to percutaneous femoral arterial cannulation was handled via distal cannulation using a smaller cannula (either percutaneous or surgical), or by shifting the cannulation site.

       Anticoagulant Management

      Before placement of cannulae, a heparin bolus of 70 U/Kg (usually 5000 units) was administered to obtain an activated clotting time of 180 seconds. After this step, we performed activated partial thromboplastin time, international normalized ratio, and antithrombin assays 4 times per day; platelet counts, fibrinogen, and d-dimer assays were conducted once daily. Patients were kept anticoagulated with heparin maintenance, and with activated partial thromboplastin time in the range of 50 to 60 seconds. When d-dimer was elevated, or in cases of hemorrhagic or thrombotic complications, a thromboelastometry was carried out, and targeted therapy, based on its results, was initiated.

       Management of Extracorporeal Life Support

      Extracorporeal life support is intended to be a bridge-to-life, and at first, all patients need full-flow support (patient's estimated required cardiac output, calculated as: body surface area × 2.4 l/minute). In this phase, stabilization is obtained, with cardiac function totally replaced by ECLS, which guarantees circulatory support and organ perfusion.
      After this phase, some pulsatility may be observed, and this is a sign of initial recovery of cardiac function. In this second phase, we maintain inotropic support and the IABP when it is already present before ECLS implantation; we wake the patient to guarantee sympathetic tone; we perform weaning from mechanical ventilation, achieving extubation as soon as possible, to reduce pulmonary resistance; we start to reduce support, monitoring pulsatility and organ perfusion (lactate, urine output, central venous pressure, and pulmonary capillary wedge pressure). We maintain inotrope therapy, to help the native heart open the aortic valve and empty the left ventricle, thus eliminating the need for a vent. If the patient remains stable, the third phase is initiated, which is a weaning protocol.

       Weaning Trial

      After obtaining hemodynamic stabilization and improvement of organ function (either neurologic, respiratory, renal, or hepatic), ECLS support was progressively decreased to 1 L per minute. Standard management involved serial echocardiograms during this phase, but we relied mostly on persistence of satisfactory hemodynamics and organ function, on low-to-medium dose inotropic support, to assess the feasibility of weaning from ECLS. Specifically, we performed echocardiographic evaluation of ventricular function (left ventricular ejection fraction >35%; good right ventricular contractility) and volume (absence of excessive ventricular distension, or severe tricuspid regurgitation), as well as clinical parameters, such as normal systemic pressure (systolic blood pressure >85 mm Hg) and central venous pressure, normal blood lactate level, and urine output.

       Data Analysis

      Continuous variables are expressed as mean ± SD, or median (range). Categoric variables are summarized by reporting absolute frequency distribution and percentage; they were compared using χ2 analysis or the Fisher exact test, as appropriate. The Student t test (for unpaired data) or the Mann-Whitney U test was used to compare continuous variables, as appropriate (normal distribution was assessed by the Shapiro-Wilk normality test). Overall survival and survival after discharge were estimated by the Kaplan-Meier method.
      To estimate effect of length of extracorporeal life support on 30-day survival, a multistate Cox Markov model
      • Meira-Machado L.
      • Cadarso-Suarez C.
      • De Una-Alvarez J.
      tdc.msm: an R library for the analysis of multi-state survival data.
      was used, with the 3 states represented by death, recovery, and surgery procedure (ventricular assist device, heart transplantation, or conventional surgery). Thirty-day survival was computed using the cessation of extracorporeal life support as landmark point.
      • Van Houwelingen H.C.
      Dynamic prediction by landmarking in event history analysis.
      Statistical analysis was performed using STATA software, release 10.0 (Stata Corporation, College Station, Tex), the R system,

      R Foundation for Statistical Computing/ R Development Core Team. The R Project for Statistical Computing. Available at: http://www.R-project.org. Accessed March 1, 2014.

      and the Time Dependent Cox libraries.
      • Meira-Machado L.
      • De Una-Alvarez J.
      • Cadarso-Suarez C.
      • Andersen P.K.
      Multi-state models for the analysis of time-to-event data.

      Results

      Preoperative characteristics, mechanical support data, and outcomes are reported in Table 2, Table 3, respectively. Mean age was 50 ± 16 years, with no differences between the groups who had A-PCS versus C-PCS. The group who had A-PCS, versus C-PCS, had a higher incidence of: malignant arrhythmias (49% vs 41%); cardiopulmonary resuscitation, either in the preceding 72 hours (70% vs 22%) or during ECLS implantation (38% vs 11%); IABP treatment (51% vs 22%); and mechanical ventilation (78% vs 63%).
      Table 2Preoperative characteristics and implantation technique
      Characteristics and techniquesOverall (n = 64)A-PCS (n = 37)C-PCS (n = 27)P value
      Preoperative characteristics
       Age (y)50 ± 1652 ± 1548 ± 17.38
       Gender, female12 (19)10 (27)2 (7).004
       BSA (m2)1.8 ± 0.21.9 ± 0.21.8 ± 0.1.001
       Malignant arrythmia29 (45)18 (49)11 (41).04
       CPR within 72 h32 (50)26 (70)6 (22)<.0001
       No. of inotropes1.9 ± 11.7 ± 12.1 ± 1.1.009
       IABP25 (39)19 (51)6 (22).001
       Respiratory failure55 (86)32 (86)23 (85).04
       Mechanical ventilation46 (72)29 (78)17 (63).01
       Renal failure32 (50)14 (38)18 (67).001
       CVVH8 (13)4 (11)4 (15).71
       Hepatic failure20 (31)7 (19)13 (48).0007
       Multi-organ failure24 (38)12 (32)12 (44).32
       APACHE IV score for predicted mortality (%)61 ± 1658 ± 1665 ± 17.38
      Implantation technique
       Bedside31 (48)21 (57)10 (37).11
       Salvage CPR17 (27)14 (38)3 (11).02
       Peripheral F-F51 (80)32 (86)19 (70).11
       Peripheral S-F12 (19)4 (11)8 (30).1
       Upgrade from F-F to S-F9 (14)4 (11)5 (19).48
       Central1 (1)1 (3)0.04
      Values are n (%) or mean ± SD, unless otherwise indicated. P values are for A-PCS versus C-PCS. A-PCS, Acute primary cardiogenic shock; C-PCS, chronic primary cardiogenic shock; BSA, body surface area; CPR, cardiopulmonary resuscitation; IABP, intra-aortic balloon pump; CVVH, continuous veno-venous hemofiltration; APACHE, Acute Physiology and Chronic Health Evaluation; F-F, femoro-femoral; S-F, subclavian-femoral.
      Table 3Parameters, outcomes, and complications of ECLS support
      Parameter, outcome, complicationOverall (n = 64)A-PCS (n = 37)C-PCS (n = 27)P value
      Parameter
       Duration (d)7 (2-11.5)7 (2.75-10.25)7 (2-12.75).86
       Flow (% of theoretic CO)61 ± 1557 ± 1367 ± 15.004
       No. of inotropes2.6 ± 1.12.2 ± 1.13.1 ± 0.8<.001
       Serum lactates (mmol/L)2.55 (1.7-3.6)2.3 (1.55-3.25)2.85 (1.9-4.3).14
       TnI peak (μg/L)3.75 (0.33-100)53 (2.2-231.9)0.47 (0.15-1.98).0007
      Outcome
       BTD9 (14)6 (16)3 (11).72
       BTR18 (28)18 (49)0<.0001
       BTS2 (3)1 (3)1 (4).04
       BTB23 (36)9 (24)14 (52).001
       BTT12 (19)3 (8)9 (33).001
       Mortality in ECLS9 (14)6 (16)3 (11).72
       Mortality, 30-d13 (20)6 (16)7 (26).02
       Discharged from hospital37 (58)22 (59)15 (56).75
      Complications during ECLS
       Neurologic12 (19)8 (22)4 (15).49
       Limb ischemia9 (14)6 (16)3 (11).72
       Leg amputation0001.00
       Bleeding13 (20)8 (22)5 (19).76
       Hemolysis1 (2)1 (3)01.00
       Renal failure25 (39)12 (32)13 (48).20
       CVVH22 (34)10 (27)12 (44).14
       Oxygenator change14 (22)5 (14)9 (33).06
       Malfunction4 (6)2 (5)2 (7)1.00
       ARDS/pulmonary congestion5 (8)1 (3)4 (15).15
       Sepsis9 (14)3 (8)6 (22).15
       MOF, post14 (22)8 (22)6 (22).95
      Values are n (%), mean ± SD, or median (interquartile range), unless otherwise indicated. P values are for A-PCS versus C-PCS. ECLS, Extracorporeal life support; A-PCS, acute primary cardiogenic shock; C-PCS, chronic primary cardiogenic shock; CO, cardiac output; TnI, troponin I; BTD, bridge-to-decision; BTR, bridge-to-recovery; BTS, bridge-to-conventional surgery; BTB, bridge-to-bridge; BTT, bridge-to-transplant; CVVH, continuous veno-venous hemofiltration; ARDS, acute respiratory distress syndrome; MOF, multi-organ failure.
      In the group who had C-PCS, versus A-PCS, we observed a greater rate of end-organ dysfunction, demonstrated by the higher number of inotropes (2.1 vs 1.7), and renal (67% vs 38%) and hepatic failure (48% vs 19%). The Acute Physiology and Chronic Health Evaluation (APACHE) IV score indicated a 61% ± 16% predicted mortality, with no differences between the 2 groups. Veno-arterial ECLS was implanted peripherally in all but 1 case. A total of 51 patients were treated with femoro-femoral cannulation, and 12 with subclavian-femoral ECLS. Nine patients required shifting of the cannulation site from the femoral to the subclavian artery, because of limb ischemia or issues in femoral cannulation (Table 2).
      The median duration of ECLS support was 7 days (interquartile range: 2-11.5 days), with no differences between the 2 groups. Mean percentage of theoretic flow was 61% ± 15%. Patients in the C-PCS group required greater support, compared with those who had A-PCS (67% vs 57%), to achieve adequate organ perfusion, as confirmed by the absence of difference in serum lactate levels (expressed as the median of the peak lactate on a daily basis). As expected, given that the frequency of AMI was higher in the A-PCS group, peak troponin was significantly greater in this group.
      The outcomes of ECLS support differed significantly between the 2 groups. In the group who had C-PCS, 23 patients were bridged to either a left ventricular assist device (52%) or heart transplantation (33%). Conversely, in the group who had A-PCS, ECLS was used as a bridge-to-transplantation in 3 patients (8%), a bridge-to-bridge in 9 patients (24%), and a bridge-to-recovery in 18 patients (49%). One patient in each group was bridged to conventional surgery.
      Recovery of cardiac function was achieved only in the group who had A-PCS (18 vs 0 patients, P = .0001). The group with A-PCS, versus C-PCS, were comparable in: mortality during ECLS support (16% in A-PCS vs 11% in C-PCS), hospital discharge (59% in A-PCS vs 56% in C-PCS) and survival at 1-year follow-up (49% in group A-PCS vs 44% in C-PCS), However, 30-day mortality was significantly greater in the group who had C-PCS, versus A-PCS (26% vs 16%, P = .02). Complications during support were similar in the 2 groups (Table 3).
      Evaluation of the impact of ECLS duration on survival indicated that risk of dying at 30 days increased at a constant rate throughout the duration of ECLS (for every 4 days on ECLS, the hazard ratio is 6.11, 95% confidence interval: 1.15-32.3) in patients undergoing a surgical procedure. In patients who recovered, the risk of dying at 30 days increased steadily only after 9 days of ECLS (for every 2 days on ECLS, the hazard ratio is 14.53, 95% confidence interval: 1.12-189.0). The confidence intervals are huge because of the low sample size of patients on ECLS for >16 days (Figure 1). Analysis of flow rates has demonstrated that in the A-PCS group, the required percentage of support is a significant predictor of recovery (Figure 2, A [P = .03]). In fact, we have shown that flow rates <60% correlate with the probability of recovery (P = .02). A subgroup of particular interest is patients who have AMI, in whom this phenomenon is particularly strong (Figure 2, B [P = .01]).
      Figure thumbnail gr1
      Figure 1Impact of duration of ECLS on 30-day survival for the 2 distinct states: “recovery” and “surgical procedure.” Risk of dying at 30 days increases at a constant rate throughout the duration of ECLS for patients in the surgical procedure stage. In patients who recovered, the risk of dying at 30 days increases steadily only after 9 days on ECLS. Surg, Surgery; ECMO, extracorporeal membrane oxygenation.
      Figure thumbnail gr2
      Figure 2Analysis of mean flow versus (A) recovery in the group who had A-PCS (P = .03); and (B) recovery in the group who had AMI (P = .01). In the box-and-whisker plot, the central box represents the values from the lower to upper quartile (25th-75th percentile). The middle line represents the median. The horizontal line extends from the minimum to the maximum value, excluding one outlier value, which is displayed as separate circle. A-PCS, Acute primary cardiogenic shock; AMI, acute myocardial infarction.
      Overall survival did not differ between the 2 groups; for A-PCS and C-PCS, respectively, survival was 73% versus 70% at 1 month; 57% versus 56% at 6 months; and 51% versus 55% at 1 year (Figure 3, A). The overall actuarial survival of the cohort after discharge was 89% at 1 year (Figure 3, B). A result of particular interest was the survival of those patients who recovered cardiac function, which was 83% at 1-year follow up (Figure 3, C).
      Figure thumbnail gr3
      Figure 3Kaplan-Meier estimates of actuarial survival in 3 populations: (A) all patients; (B) discharged patients; and (C) bridge-to-recovery subgroup after discharge. A-PCS, Acute primary cardiogenic shock; C-PCS, chronic primary cardiogenic shock; CI, confidence interval.

      Discussion

      The aim of this study was to evaluate the impact of etiology—in particular, acute versus chronic—on the outcome of patients in primary cardiogenic shock who are treated with ECLS. The outcomes assessed were death during ECLS support, 30-day mortality, hospital discharge, and survival at follow up, both overall and in the 2 etiology groups. The 2 groups were comparable in terms of their preimplantation condition and postimplantation complications.
      Differences between the groups could be attributed to the differing etiology of cardiogenic shock. In the group with C-PCS, the shock represents the end stage of a progressive decline in cardiac function with chronic structural damage. This scenario was not seen in the group with A-PCS, in which the most common cause of cardiogenic shock was AMI. In the analysis of primary endpoints, the only difference between the groups was that 30-day mortality was higher with C-PCS, compared with A-PCS.
      These data are consistent with the recent series published by Bermudez and colleagues,
      • Bermudez C.A.
      • Rocha R.V.
      • Toyoda Y.
      • Zaldonis D.
      • Sappington P.L.
      • Mulukutla S.
      • et al.
      Extracorporeal membrane oxygenation for advanced refractory shock in acute and chronic cardiomyopathy.
      but differ regarding long-term survival. In our series, the 2 groups were comparable in mortality during support, hospital discharge, and even survival at follow up. Overall, the outcomes are in line with the literature, as is overall survival.
      • Chen Y.S.
      • Yu H.Y.
      • Huang S.C.
      • Chiu K.M.
      • Lin T.Y.
      • Lai L.P.
      • et al.
      Experience and result of extracorporeal membrane oxygenation in treating fulminant myocarditis with shock: What mechanical support should be considered first?.
      • Wu M.Y.
      • Lee M.Y.
      • Lin C.C.
      • Chang Y.S.
      • Tsai F.C.
      • Lin P.J.
      Resuscitation of non-postcardiotomy cardiogenic shock or cardiac arrest with extracorporeal life support: the role of bridging to intervention.
      In light of these results, a reasonable conclusion is that use of ECLS allows survival to be maximized in patients who are otherwise untreatable and face an extremely severe prognosis, independent of etiology of the shock, and of whether it is acute or chronic. In our experience, unlike that of Bermudez et al,
      • Bermudez C.A.
      • Rocha R.V.
      • Toyoda Y.
      • Zaldonis D.
      • Sappington P.L.
      • Mulukutla S.
      • et al.
      Extracorporeal membrane oxygenation for advanced refractory shock in acute and chronic cardiomyopathy.
      patients in a chronic setting can additionally benefit from ECLS as a bridge to candidacy for a ventricular assist device or a transplant.
      As secondary endpoints, we looked at the proportion of patients who had recovery of cardiac function (bridge-to-recovery), to determine if the duration and magnitude of support are predictors of recovery and survival. In keeping with the markedly different profiles of the 2 groups, recovery occurred in only the group with A-PCS. This difference was highly significant and reinforces the idea that cardiogenic shock, in the context of chronic progressive cardiac failure, represents the terminal stage of the disease, which cannot be remedied without replacing the heart's pumping function. As a result, in the group who had C-PCS, ECLS acted essentially either as a bridge to a ventricular assist device or as a bridge-to-transplant. Conversely, in the group who had A-PCS, ECLS support temporarily rests the damaged myocardium, allowing it to recover functionality.

      R Foundation for Statistical Computing/ R Development Core Team. The R Project for Statistical Computing. Available at: http://www.R-project.org. Accessed March 1, 2014.

      In patients with refractory shock secondary to an acute etiology, ECLS offers a substantial chance of recovery, often representing, after failed percutaneous coronary intervention, the only therapy required.
      In patients with the best 30-day survival, the duration of ECLS was significantly shorter. Specifically, for patients recovering their cardiac function, risk of death does not increase until the 9th day; our data suggest that recovery beyond this limit, although possible, is burdened by higher mortality. Conversely, patients in need of additional surgical procedures, owing to the constant increase in mortality over time, would likely benefit from early planning of a therapeutic strategy. Potentially, the shorter duration of support reflects, at least partly, a less substantial systemic compromise.
      Another part of this study is related to the management of the flow rates. After the first phase of “resuscitation” at full flow, we aimed to reduce the support, whenever possible. Specific strategies make this possible. First, the use of ECLS reduces venous return and hence the preload of the left ventricle, thus reducing the work it is required to do. Second, ECLS provides most of the splanchnic circulation, but does not increase the afterload on the left ventricle as much as full support would. Finally, with the help of inotropes, the native heart is often able to open the aortic valve and empty the left ventricle, thus eliminating the need for a vent and providing most of the coronary and cerebral perfusion.
      Within the group with A-PCS, analyses of flow rates demonstrate that the maintenance of support at <60% of the theoretic requirement was significantly associated with recovery of cardiac function, which likely reflects a less-severe initial myocardial condition and a greater degree of reversibility. Of particular interest in the group of patients who had A-PCS is the subgroup who had AMI, because of both the epidemiologic significance of this condition and its high frequency within this group (n = 26). In this subgroup, the proportion of patients who recover cardiac function is very promising, totaling 42%.
      These results are comparable to those in the literature, particularly in studies reporting outcomes for patients with cardiogenic shock secondary to AMI.
      • Wu M.Y.
      • Lee M.Y.
      • Lin C.C.
      • Chang Y.S.
      • Tsai F.C.
      • Lin P.J.
      Resuscitation of non-postcardiotomy cardiogenic shock or cardiac arrest with extracorporeal life support: the role of bridging to intervention.
      A reasonable possibility is that, in this group, ECLS helps to contain the damage caused by ischemia-reperfusion injury and evolution of the infarcted area, by directly reducing the afterload against which the heart has to work, and hence reducing the oxygen demand of the myocardium.
      • Schopka S.
      • Philipp A.
      • Lunz D.
      • Camboni D.
      • Zacher R.
      • Rupprecht L.
      • et al.
      Single-center experience with extracorporeal life support in 103 nonpostcardiotomy patients.
      In this subgroup, the degree of ECLS support was a predictor of cardiac function recovery: Patients who experienced recovery had mean flows of 55% of the theoretic predicted requirement.
      The analysis of all patients as a single cohort showed that 30-day survival was approximately 80%, and discharge from the hospital was approximately 60%. A particularly encouraging result was 89% survival at 2-year follow up in patients discharged from the hospital. The results for the subgroup with shock due to AMI was consistent with these findings, but peculiar in that ECLS was the only necessary treatment after failed percutaneous coronary intervention, for 42% of the patients, and provided long-term survival on the order of 80%. These results support the use of ECLS in PCS, whereas treatment with IABP was recently shown to confer no benefit in terms of mortality.
      • Thiele H.
      • Zeymer U.
      • Neumann F.J.
      • Ferenc M.
      • Olbrich H.G.
      • Hausleiter J.
      • et al.
      Intraaortic balloon support for myocardial infarction with cardiogenic shock.

      Conclusions

      Extracorporeal life support is intended as a bridge-to-life and should be considered as a first line of support in patients with refractory cardiogenic shock. In the setting of a chronic decline in cardiac function, ECLS represents a bridge to implantation of a ventricular assist device or to an orthotopic heart transplant. In patients with refractory shock due to acute etiologies, ECLS offers a substantial chance of recovery, often representing, after failed percutaneous coronary intervention, the only necessary treatment. In our experience, the magnitude and duration of the ECLS support significantly affects the chance of weaning, and of survival, reflecting a less severe initial myocardial condition and a greater degree of reversibility. Maintaining flows at approximately 60% of the theoretic requirement and minimizing the duration of support seems to be the strategy that offers the best chance of survival.

       Conflict of Interest Statement

      Authors have nothing to disclose with regard to commercial support.

      References

        • Goldstein D.
        • Neragi-Miandoab S.
        Mechanical bridge to decision: What are the options for the management of acute refractory cardiogenic shock?.
        Curr Heart Fail Rep. 2011; 8: 51-58
        • Koerner M.M.
        • Jahanyar J.
        Assist devices for circulatory support in therapy-refractory acute heart failure.
        Curr Opin Cardiol. 2008; 23: 399-406
        • Russo C.F.
        • Cannata A.
        • Lanfranconi M.
        • Bruschi G.
        • Milazzo F.
        • Paino R.
        • et al.
        Veno-arterial extracorporeal membrane oxygenation using Levitronix centrifugal pump as bridge to decision for refractory cardiogenic shock.
        J Thorac Cardiovasc Surg. 2010; 140: 1416-1421
        • McMurray J.J.
        • Adamopoulos S.
        • Anker S.D.
        • Auricchio A.
        • Böhm M.
        • Dickstein K.
        • et al.
        ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure 2012.
        Eur Heart J. 2012; 33: 1787-1847
        • Marasco S.F.
        • Lukas G.
        • McDonald M.
        • McMillan J.
        • Ihle B.
        Review of ECMO (extra corporeal membrane oxygenation) support in critically ill adult patients.
        Heart Lung Circ. 2008; 17: S41-S47
        • Bermudez C.A.
        • Rocha R.V.
        • Toyoda Y.
        • Zaldonis D.
        • Sappington P.L.
        • Mulukutla S.
        • et al.
        Extracorporeal membrane oxygenation for advanced refractory shock in acute and chronic cardiomyopathy.
        Ann Thorac Surg. 2011; 92: 2125-2131
        • Formica F.
        • Avalli L.
        • Martino A.
        • Maggioni E.
        • Muratore M.
        • Ferro O.
        • et al.
        Extracorporeal membrane oxygenation with a poly-methylpentene oxygenator (Quadrox D). The experience of a single Italian centre in adult patients with refractory cardiogenic shock.
        ASAIO J. 2008; 54: 89-94
        • Rastan A.J.
        • Dege A.
        • Mohr M.
        • Doll N.
        • Falk V.
        • Walther T.
        • et al.
        Early and late outcomes of 517 consecutive adult patients treated with extracorporeal membrane oxygenation for refractory postcardiotomy cardiogenic shock.
        J Thorac Cardiovasc Surg. 2010; 139: 302-311
        • Hei F.
        • Lou S.
        • Li J.
        • Yu K.
        • Liu J.
        • Feng Z.
        • et al.
        Five-year results of 121 consecutive patients treated with extracorporeal membrane oxygenation at Fu Wai hospital.
        Artif Organs. 2011; 35: 572-578
        • Arlt M.
        • Philipp A.
        • Voelkel S.
        • Camboni D.
        • Rupprecht L.
        • Graf B.M.
        • et al.
        Hand-held minimised extracorporeal membrane oxygenation: a new bridge to recovery in patients with out-of-centre cardiogenic shock.
        Eur J Cardiothorac Surg. 2011; 40: 689-694
        • Tang G.H.
        • Malekan R.
        • Kai M.
        • Lansman S.L.
        • Spielvogel D.
        Peripheral venoarterial extracorporeal membrane oxygenation improves survival in myocardial infarction with cardiogenic shock.
        J Thorac Cardiovasc Surg. 2013; 145: e32-e33
        • Moraca R.J.
        • Wanamaker K.M.
        • Bailey S.H.
        • McGregor W.E.
        • Murali S.
        • Benza R.
        • et al.
        Salvage peripheral extracorporeal membrane oxygenation using Cobe Revolution centrifugal pump as a bridge to decision for acute refractory cardiogenic shock.
        J Card Surg. 2012; 27: 521-527
        • Chen Y.S.
        • Yu H.Y.
        • Huang S.C.
        • Chiu K.M.
        • Lin T.Y.
        • Lai L.P.
        • et al.
        Experience and result of extracorporeal membrane oxygenation in treating fulminant myocarditis with shock: What mechanical support should be considered first?.
        J Heart Lung Transplant. 2005; 24: 81-87
        • Tsao N.W.
        • Shih C.M.
        • Yeh J.S.
        • Kao Y.T.
        • Hsieh M.H.
        • Ou K.L.
        • et al.
        Extracorporeal membrane oxygenation-assisted primary percutaneous coronary intervention may improve survival of patients with acute myocardial infarction complicated by profound cardiogenic shock.
        J Crit Care. 2012; 27: 530.e1-530.e11
      1. Extracorporeal Life Support Organization Guidelines for cardiopulmonary extracorporeal life support, version 1.3. Available at: www.elso.org/Resources/Guidelines.aspx. Accessed March 1, 2014.

        • Arlt M.
        • Philipp A.
        • Voelkel S.
        • Rupprecht L.
        • Mueller T.
        • Hilker M.
        • et al.
        Extracorporeal membrane oxygenation in severe trauma patients with bleeding shock.
        Resuscitation. 2010; 81: 804-809
        • Meira-Machado L.
        • Cadarso-Suarez C.
        • De Una-Alvarez J.
        tdc.msm: an R library for the analysis of multi-state survival data.
        Comput Methods Programs Biomed. 2007; 86: 131-140
        • Van Houwelingen H.C.
        Dynamic prediction by landmarking in event history analysis.
        Scand J Stat. 2007; 34: 70-85
      2. R Foundation for Statistical Computing/ R Development Core Team. The R Project for Statistical Computing. Available at: http://www.R-project.org. Accessed March 1, 2014.

        • Meira-Machado L.
        • De Una-Alvarez J.
        • Cadarso-Suarez C.
        • Andersen P.K.
        Multi-state models for the analysis of time-to-event data.
        Stat Methods Med Res. 2009; 18: 195-222
        • Wu M.Y.
        • Lee M.Y.
        • Lin C.C.
        • Chang Y.S.
        • Tsai F.C.
        • Lin P.J.
        Resuscitation of non-postcardiotomy cardiogenic shock or cardiac arrest with extracorporeal life support: the role of bridging to intervention.
        Resuscitation. 2012; 83: 976-981
        • Schopka S.
        • Philipp A.
        • Lunz D.
        • Camboni D.
        • Zacher R.
        • Rupprecht L.
        • et al.
        Single-center experience with extracorporeal life support in 103 nonpostcardiotomy patients.
        Artif Organs. 2013; 37: 150-156
        • Thiele H.
        • Zeymer U.
        • Neumann F.J.
        • Ferenc M.
        • Olbrich H.G.
        • Hausleiter J.
        • et al.
        Intraaortic balloon support for myocardial infarction with cardiogenic shock.
        N Engl J Med. 2012; 4: 1287-1296