Ecmo as a bridge to decision: Recovery, VAD, or heart transplantation? N. Rousse, F. Juthier, C. Pinc¸on, I. Hysi, C. Banfi, E. Robin, G. Fayad, B. Jegou, A. Prat, A. Vincentelli PII: DOI: Reference:
S0167-5273(15)00523-9 doi: 10.1016/j.ijcard.2015.03.283 IJCA 20048
To appear in:
International Journal of Cardiology
Received date: Revised date: Accepted date:
5 October 2014 26 February 2015 2 March 2015
Please cite this article as: Rousse N, Juthier F, Pin¸con C, Hysi I, Banfi C, Robin E, Fayad G, Jegou B, Prat A, Vincentelli A, Ecmo as a bridge to decision: Recovery, VAD, or heart transplantation?, International Journal of Cardiology (2015), doi: 10.1016/j.ijcard.2015.03.283
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT ECMO AS A BRIDGE TO DECISION: RECOVERY, VAD, OR HEART
T
TRANSPLANTATION?
IP
N. Rousse, MD1*, F. Juthier, MD1*, PhD, C. Pinçon, PhD2, I. Hysi, MD1,C. Banfi1, MD1, PhD, E.
SC R
Robin, MD3, PhD, G. Fayad, MD1, B. Jegou, MD1, A. Prat, MD1, A. Vincentelli, MD1, PhD
NU
*authors have equally contributed
Centre Hospitalier Régional et Universitaire de Lille, Lille 59035, France. Department of
MA
Cardiovascular surgery1
Université Lille Nord de France, Lille 59000. UDSL, Inserm UMR1011 & UDSL, Lille 59035, France Université Lille Nord de France, Lille 59000. UDSL, EA2694, Department of Biostatistics2
D
Centre Hospitalier Régional et Universitaire de Lille, Lille 59035, France. Department of
AC
CE P
TE
Anestesiology3
Corresponding author:
Pr. A. Vincentelli, MD, PhD Department of Cardiovascular surgery, CHU Lille Phone: +333 204 453 56 Fax: +333 204 456 61 Mail: [email protected]
1
ACCEPTED MANUSCRIPT Abstract: Background: Our 8-year experience with ECMO support as a bridge to decision was reviewed.
T
Methods: A cohort of 124 consecutive patients received ECMO for refractory cardiogenic shock in our
IP
institution. Twenty-six of these were out of hospital cardiac arrests and were excluded from this
SC R
analysis. The median age was 43 years, in the range of 11 to 73 years.
Results: The median duration of ECMO support was 4.5 days. Mortality while supported by ECMO was 50% with a median support time of 2 days. Weaning from ECMO was achieved for 49 patients
NU
with outcome as follow: cardiac recovery (60%), heart transplantation (26%), VAD implantation (14%). Median duration of support before weaning was 8 days. Hospital survival was respectively of
MA
83%, 61.5% and 71% for cardiac recovery, heart transplantation or VAD implantation. ECMO weaning was significantly improved respectively, in all patients who had normalized their renal
D
function, and when duration of support > 6 days ( HR: 4.255 [1.255 – 14.493], p=0.02 and HR: 2.164
TE
[1.152 – 4.082], p=0.02). A creatinine level > 14mg/l the day of weaning was a significant predictor of death (HR: 5.807 [1.089– 30.953]; p=0.04). Median follow up was 2.4 years; one-year survival rate
CE P
was 78%, 51% and 75% for cardiac recovery, heart transplantation and VAD implantation respectively. Conclusion: With at least 6 days of support, ECMO allowed a better patient selection for myocardial
AC
recovery, VAD implantation or heart transplantation. Whether VAD implantation or heart transplant in those patients is a better indication remains to be evaluated.
Keywords : Heart Failure, Shock, Extra Corporeal Life Support, Heart transplantation, Ventricle Assist Device
2
ACCEPTED MANUSCRIPT
INTRODUCTION Treatment of INTERMACS level 1 and 2 population remains the most challenging issue in the field of
IP
T
refractory end-stage heart failure. The SHOCK trial reported a 30-day mortality rate in ischemic cardiogenic shock of 50 to 80% even after revascularization (1; 2). According to current North
SC R
American and European heart failure and transplantation guidelines heart transplantation in this population is no longer an indication (3; 4).
NU
In 2012, J. Kirklin’s report from the INTERMACS registry which included 4366 patients implanted with a Ventricle Assist Device (VAD) from 2006 to 2011, indicated that INTERMACS level 1 & 2
MA
were significant risk factors for death after VAD implantation (5). Moreover, as a consequence, the rate of implantation of VAD in this population significantly dropped from 40.8% in 2006 to 14.0% in 2011. Thus VAD implantation in INTERMACS level 1 or 2 patients remains highly debatable. Rao et
TE
D
al. revised screening scale to predict survival after Left Ventricle Assist Device (LVAD) insertion demonstrated that post cardiotomy shock remained a significant risk factor for death (6).
CE P
Midterm circulatory support to stabilize patients with (Extra Corporal Membrane Oxygenation) ECMO as a bridge to decision in order to improve outcomes in this high-risk population has been advocated (7- 9), but little is known about predicting factors of success, with weaning modes and
AC
survival after ECMO support in terms of duration of support and organ function improvement. We report here our experience with the use of ECMO in INTERMACS level 1 patient over an 8-year period.
PATIENTS AND METHODS: Patients Patient with a refractory cardiogenic shock was defined as a patient with an unstable hemodynamic status despite a maximal inotropes treatment. All the patients with refractory cardiogenic shock in our institution were referred for ECMO. In case of postcardiotomy patient, ECMO was implanted when patients could not be weaned from CBP despite high doses of inotropes or after postoperative refractory cardiac arrest.
3
ACCEPTED MANUSCRIPT The ECMO Circuit used was a centrifugal blood pump (Rotaflow ®, Maquet, Hirrlingen, Germany or Stockert Instrumente GmbH, Munchen, Germany) with a hollow fiber oxygenator (Quadrox ®, Maquet, Hirrlingen, Germany) or a polymethylpenthen membrane (Dideco, Sorin Group, Mirandola,
IP
T
Italy).
The majority of patients (78 patients or 80%) were supported by a peripheral ECMO, where arterial
SC R
and venous cannulation was performed through a Seldinger technique in the femoral artery and vein mostly after a limited surgical cut-down of the groin. We used mainly17Fr arterial and 21Fr venous
NU
cannula, and a 6 Fr distal leg perfusion catheter was systematically implanted. Central ECMO was used in 20 patients (20.4%). Through a median sternotomy, ECMO was established between the
MA
ascending aorta, the right atrium and the LV which was discharged with a cannula inserted into the right pulmonary vein and through the mitral valve. The ECMO system including tubes, membrane oxygenator and pump were coated either with heparin (Maquet, Hirrlingen, Germany) or were heparin
TE
D
free using a phosphorylcholine coating (Phisio, SorinGroup, Mirandola, Italy). Post ECMO insertion anticoagulation was started as soon as coagulation disorders were corrected.
CE P
Management of anticoagulation was achieved through a target activated clotting time of 160 to 250 seconds, with a heparin level < 0.2UI/ml and a rate of Prothrombin > 75%. Before ECMO set up, 13 patients were already supported with intra- aortic counter pulsation balloon
AC
pump (13.3%). ECMO was set up during cardiac arrest and cardiopulmonary resuscitation in 37 patients (37.8%).
Weaning from the pump was considered in patients who no longer required mechanical ventilation and were in stable hemodynamic and metabolic condition, without any signs of malperfusion on only a low dose of inotropes. Pump flow was gradually reduced to low flow (1.5 l/min or 0.5 l/min/m2) support. Echocardiography was performed to evaluate myocardial recovery. Myocardial recovery was defined as a left ventricle outflow tract velocity time index measured between 13 and 15 cm, with a left ventricle ejection fraction above 40%. The right ventricle function was also carefully evaluated. The most relevant echocardiographic predictors of good right ventricular function after ECMO implantation were: fractional area change (FAC) >40%, Tricuspid annular plane systolic excursion
4
ACCEPTED MANUSCRIPT (TAPSE) > 17mm, tissue Doppler-derived lateral annular systolic velocity (S') > 10cm/s, Tei index 14 mg/l at the time of weaning was also a predictive factor of death and significantly impaired survival. Current North American and European guidelines for acute and chronic heart failure (3; 4) do not
9
ACCEPTED MANUSCRIPT either recommend heart transplantation in stable patient with significant impaired renal function. In France, recent recommendations from the “Agence de la Biomédecine”, the French organ transplantation regulatory office, has decided that significant renal failure (e.g. creatinine clearance
14 mg/l
4.344 [1.575 – 11.982]
0.005
T
Variable
IP
Table 3: Predictive risk factors of death (univariate and multivariate logistic regression) ; HR Hazard
AC
CE P
TE
D
MA
NU
SC R
Ratio, CI Confidence Intervals
20
ACCEPTED MANUSCRIPT
HR [CI95%]
p-value
Post-cardiotomy 6 days of follow-up and after
4.255 [0.001- 14.493]
0.02
Log creatinine (per log unit decrease)a
2.164 [1.152- 4.082]
0.02
SC R
IP
T
Variable
Table 4: Predictive risks factors for unsuccessful weaning of ECMO (univariate and multivariate
AC
CE P
TE
D
MA
NU
logistic regression); aTime dependent covariate
21
ACCEPTED MANUSCRIPT
TE
D
MA
NU
SC R
IP
T
1
AC
CE P
Figure 1
22
TE
D
MA
NU
SC R
IP
T
ACCEPTED MANUSCRIPT
AC
CE P
Figure 2
23
TE
D
MA
NU
SC R
IP
T
ACCEPTED MANUSCRIPT
AC
CE P
Figure 3
24
TE
D
MA
NU
SC R
IP
T
ACCEPTED MANUSCRIPT
AC
CE P
Figure 4
25
TE
D
MA
NU
SC R
IP
T
ACCEPTED MANUSCRIPT
AC
CE P
Figure 5
26
TE
D
MA
NU
SC R
IP
T
ACCEPTED MANUSCRIPT
AC
CE P
Figure 6
27
S0167-5273(15)00523-9 doi: 10.1016/j.ijcard.2015.03.283 IJCA 20048
To appear in:
International Journal of Cardiology
Received date: Revised date: Accepted date:
5 October 2014 26 February 2015 2 March 2015
Please cite this article as: Rousse N, Juthier F, Pin¸con C, Hysi I, Banfi C, Robin E, Fayad G, Jegou B, Prat A, Vincentelli A, Ecmo as a bridge to decision: Recovery, VAD, or heart transplantation?, International Journal of Cardiology (2015), doi: 10.1016/j.ijcard.2015.03.283
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT ECMO AS A BRIDGE TO DECISION: RECOVERY, VAD, OR HEART
T
TRANSPLANTATION?
IP
N. Rousse, MD1*, F. Juthier, MD1*, PhD, C. Pinçon, PhD2, I. Hysi, MD1,C. Banfi1, MD1, PhD, E.
SC R
Robin, MD3, PhD, G. Fayad, MD1, B. Jegou, MD1, A. Prat, MD1, A. Vincentelli, MD1, PhD
NU
*authors have equally contributed
Centre Hospitalier Régional et Universitaire de Lille, Lille 59035, France. Department of
MA
Cardiovascular surgery1
Université Lille Nord de France, Lille 59000. UDSL, Inserm UMR1011 & UDSL, Lille 59035, France Université Lille Nord de France, Lille 59000. UDSL, EA2694, Department of Biostatistics2
D
Centre Hospitalier Régional et Universitaire de Lille, Lille 59035, France. Department of
AC
CE P
TE
Anestesiology3
Corresponding author:
Pr. A. Vincentelli, MD, PhD Department of Cardiovascular surgery, CHU Lille Phone: +333 204 453 56 Fax: +333 204 456 61 Mail: [email protected]
1
ACCEPTED MANUSCRIPT Abstract: Background: Our 8-year experience with ECMO support as a bridge to decision was reviewed.
T
Methods: A cohort of 124 consecutive patients received ECMO for refractory cardiogenic shock in our
IP
institution. Twenty-six of these were out of hospital cardiac arrests and were excluded from this
SC R
analysis. The median age was 43 years, in the range of 11 to 73 years.
Results: The median duration of ECMO support was 4.5 days. Mortality while supported by ECMO was 50% with a median support time of 2 days. Weaning from ECMO was achieved for 49 patients
NU
with outcome as follow: cardiac recovery (60%), heart transplantation (26%), VAD implantation (14%). Median duration of support before weaning was 8 days. Hospital survival was respectively of
MA
83%, 61.5% and 71% for cardiac recovery, heart transplantation or VAD implantation. ECMO weaning was significantly improved respectively, in all patients who had normalized their renal
D
function, and when duration of support > 6 days ( HR: 4.255 [1.255 – 14.493], p=0.02 and HR: 2.164
TE
[1.152 – 4.082], p=0.02). A creatinine level > 14mg/l the day of weaning was a significant predictor of death (HR: 5.807 [1.089– 30.953]; p=0.04). Median follow up was 2.4 years; one-year survival rate
CE P
was 78%, 51% and 75% for cardiac recovery, heart transplantation and VAD implantation respectively. Conclusion: With at least 6 days of support, ECMO allowed a better patient selection for myocardial
AC
recovery, VAD implantation or heart transplantation. Whether VAD implantation or heart transplant in those patients is a better indication remains to be evaluated.
Keywords : Heart Failure, Shock, Extra Corporeal Life Support, Heart transplantation, Ventricle Assist Device
2
ACCEPTED MANUSCRIPT
INTRODUCTION Treatment of INTERMACS level 1 and 2 population remains the most challenging issue in the field of
IP
T
refractory end-stage heart failure. The SHOCK trial reported a 30-day mortality rate in ischemic cardiogenic shock of 50 to 80% even after revascularization (1; 2). According to current North
SC R
American and European heart failure and transplantation guidelines heart transplantation in this population is no longer an indication (3; 4).
NU
In 2012, J. Kirklin’s report from the INTERMACS registry which included 4366 patients implanted with a Ventricle Assist Device (VAD) from 2006 to 2011, indicated that INTERMACS level 1 & 2
MA
were significant risk factors for death after VAD implantation (5). Moreover, as a consequence, the rate of implantation of VAD in this population significantly dropped from 40.8% in 2006 to 14.0% in 2011. Thus VAD implantation in INTERMACS level 1 or 2 patients remains highly debatable. Rao et
TE
D
al. revised screening scale to predict survival after Left Ventricle Assist Device (LVAD) insertion demonstrated that post cardiotomy shock remained a significant risk factor for death (6).
CE P
Midterm circulatory support to stabilize patients with (Extra Corporal Membrane Oxygenation) ECMO as a bridge to decision in order to improve outcomes in this high-risk population has been advocated (7- 9), but little is known about predicting factors of success, with weaning modes and
AC
survival after ECMO support in terms of duration of support and organ function improvement. We report here our experience with the use of ECMO in INTERMACS level 1 patient over an 8-year period.
PATIENTS AND METHODS: Patients Patient with a refractory cardiogenic shock was defined as a patient with an unstable hemodynamic status despite a maximal inotropes treatment. All the patients with refractory cardiogenic shock in our institution were referred for ECMO. In case of postcardiotomy patient, ECMO was implanted when patients could not be weaned from CBP despite high doses of inotropes or after postoperative refractory cardiac arrest.
3
ACCEPTED MANUSCRIPT The ECMO Circuit used was a centrifugal blood pump (Rotaflow ®, Maquet, Hirrlingen, Germany or Stockert Instrumente GmbH, Munchen, Germany) with a hollow fiber oxygenator (Quadrox ®, Maquet, Hirrlingen, Germany) or a polymethylpenthen membrane (Dideco, Sorin Group, Mirandola,
IP
T
Italy).
The majority of patients (78 patients or 80%) were supported by a peripheral ECMO, where arterial
SC R
and venous cannulation was performed through a Seldinger technique in the femoral artery and vein mostly after a limited surgical cut-down of the groin. We used mainly17Fr arterial and 21Fr venous
NU
cannula, and a 6 Fr distal leg perfusion catheter was systematically implanted. Central ECMO was used in 20 patients (20.4%). Through a median sternotomy, ECMO was established between the
MA
ascending aorta, the right atrium and the LV which was discharged with a cannula inserted into the right pulmonary vein and through the mitral valve. The ECMO system including tubes, membrane oxygenator and pump were coated either with heparin (Maquet, Hirrlingen, Germany) or were heparin
TE
D
free using a phosphorylcholine coating (Phisio, SorinGroup, Mirandola, Italy). Post ECMO insertion anticoagulation was started as soon as coagulation disorders were corrected.
CE P
Management of anticoagulation was achieved through a target activated clotting time of 160 to 250 seconds, with a heparin level < 0.2UI/ml and a rate of Prothrombin > 75%. Before ECMO set up, 13 patients were already supported with intra- aortic counter pulsation balloon
AC
pump (13.3%). ECMO was set up during cardiac arrest and cardiopulmonary resuscitation in 37 patients (37.8%).
Weaning from the pump was considered in patients who no longer required mechanical ventilation and were in stable hemodynamic and metabolic condition, without any signs of malperfusion on only a low dose of inotropes. Pump flow was gradually reduced to low flow (1.5 l/min or 0.5 l/min/m2) support. Echocardiography was performed to evaluate myocardial recovery. Myocardial recovery was defined as a left ventricle outflow tract velocity time index measured between 13 and 15 cm, with a left ventricle ejection fraction above 40%. The right ventricle function was also carefully evaluated. The most relevant echocardiographic predictors of good right ventricular function after ECMO implantation were: fractional area change (FAC) >40%, Tricuspid annular plane systolic excursion
4
ACCEPTED MANUSCRIPT (TAPSE) > 17mm, tissue Doppler-derived lateral annular systolic velocity (S') > 10cm/s, Tei index 14 mg/l at the time of weaning was also a predictive factor of death and significantly impaired survival. Current North American and European guidelines for acute and chronic heart failure (3; 4) do not
9
ACCEPTED MANUSCRIPT either recommend heart transplantation in stable patient with significant impaired renal function. In France, recent recommendations from the “Agence de la Biomédecine”, the French organ transplantation regulatory office, has decided that significant renal failure (e.g. creatinine clearance
14 mg/l
4.344 [1.575 – 11.982]
0.005
T
Variable
IP
Table 3: Predictive risk factors of death (univariate and multivariate logistic regression) ; HR Hazard
AC
CE P
TE
D
MA
NU
SC R
Ratio, CI Confidence Intervals
20
ACCEPTED MANUSCRIPT
HR [CI95%]
p-value
Post-cardiotomy 6 days of follow-up and after
4.255 [0.001- 14.493]
0.02
Log creatinine (per log unit decrease)a
2.164 [1.152- 4.082]
0.02
SC R
IP
T
Variable
Table 4: Predictive risks factors for unsuccessful weaning of ECMO (univariate and multivariate
AC
CE P
TE
D
MA
NU
logistic regression); aTime dependent covariate
21
ACCEPTED MANUSCRIPT
TE
D
MA
NU
SC R
IP
T
1
AC
CE P
Figure 1
22
TE
D
MA
NU
SC R
IP
T
ACCEPTED MANUSCRIPT
AC
CE P
Figure 2
23
TE
D
MA
NU
SC R
IP
T
ACCEPTED MANUSCRIPT
AC
CE P
Figure 3
24
TE
D
MA
NU
SC R
IP
T
ACCEPTED MANUSCRIPT
AC
CE P
Figure 4
25
TE
D
MA
NU
SC R
IP
T
ACCEPTED MANUSCRIPT
AC
CE P
Figure 5
26
TE
D
MA
NU
SC R
IP
T
ACCEPTED MANUSCRIPT
AC
CE P
Figure 6
27
