Absence of rapid deployment extracorporeal membrane oxygenation (ECMO) team does not preclude resuscitation ecmo in pediatric cardiac patients with good results.
ABSTRACT We evaluated the results of using extracorporeal membrane oxygenation (ECMO) as resuscitation for cardiac patients undergoing cardiopulmonary resuscitation (CPR) in our setting where neither perfusionists nor surgeons are always on site, and no circuit may be ready. Between 2003 and 2006, we used ECMO for all cardiac patients who underwent cardiac arrest in the pediatric intensive care unit (PICU) or Cath Laboratory. We reviewed retrospectively 14 consecutive files (15 episodes). Mean CPR time before ECMO institution was 44 minutes (10-110 minutes). The surgeons, perfusionist, and scrub nurse, not on site for three of these patients, had to be called in simultaneously with institution of CPR. Two died on ECMO, the third one was successfully transplanted after 5 days. Globally, 10 patients could be weaned (66%). Eight patients (57%) survived to hospital discharge, seven without obvious neurological damage. One patient was bridged to a left ventricular assist device (LVAD) and was eventually successfully transplanted. He had an ischemic brain lesion with good recuperation and no sequel. We obtained good results with resuscitation ECMO in our setting where a permanently on-site rapid deployment ECMO team is not present at all times.
- SourceAvailable from: Venkatesan Ben Sivarajan[Show abstract] [Hide abstract]
ABSTRACT: Survival outcomes in children with heart disease after use of either non-emergent extracorporeal membrane oxygenation (ECMO) or cardiopulmonary resuscitation (CPR) onto ECMO (ECPR) are comparable. Concerns remain regarding the impact of CPR duration on survival and neurological outcome. Children with cardiac disease requiring ECMO were identified from our database. Demographic, operative and ECMO details were evaluated with respect to survival. In addition, resuscitation details were extracted for the recent subgroup requiring ECPR; these details were evaluated with respect to survival and neurological outcomes at midterm follow-up by univariate analysis and multivariable logistic regression. There were 126 ECMO runs in 116 children; 61 (53%) received ECPR. Forty-eight (41%) children survived to discharge; survival in the most recent era was 48%. Thirty-seven children underwent ECPR in the most recent era with 14 (38%) surviving to discharge. Duration of cardiopulmonary resuscitation differed significantly between survivors and nonsurvivors (15 vs. 40 min, p = 0.009); children requiring ≥30 min of CPR had 79% reduced odds of hospital survival (OR = 0.21, 95% CI = 0.05-0.96, p = 0.04). Two children died after hospital discharge; with 33% having paediatric cerebral performance category scores ≤2. Poor neurological outcome was associated with longer duration of CPR (32 vs. 17.5 min, p = 0.03). Despite comparable survival outcomes between ECPR and non-emergent ECMO in children with cardiac disease a significant association between CPR duration and outcome (survival and neurological) was noted in this population.European Journal of Intensive Care Medicine 03/2011; 37(5):853-60. · 5.17 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: Cost estimation for engineering projects has advanced beyond the bottom-up counting of costs and the top-down use of analogies. Parametric cost estimation is now employed to estimate volatile software costs, as well as the cost of Systems Engineering (SE) effort in engineering projects. The Constructive Systems Engineering Cost Model (COSYSMO) estimates the number of Person-Months (PM) necessary to complete systems engineering projects by using project size and cost parameters. On re-examining the nature of systems engineering and the structures of complex projects that must be built in an orderly fashion on a variety of scales, it is apparent that a parametric formula that employs fractal dimensionality principles should be well suited for their cost estimation. This thesis therefore develops the connections between fractal dimensionality and cost estimation with COSYSMO. The result is Fractal-COSYSMO, a novel cost estimating formulation that can be used to determine the cost of developing systems that show complexity on a broad scale.Procedia CS. 01/2011; 6:88-93.
- Circulation 10/2010; 122(16 Suppl 2):S466-515. · 15.20 Impact Factor
Absence of Rapid Deployment Extracorporeal Membrane
Oxygenation (ECMO) Team Does Not Preclude Resuscitation
ECMO in Pediatric Cardiac Patients With Good Results
OLIVIER GHEZ,* VIRGINIE FOUILLOUX,* ARNAUD CHARPENTIER,† PATRICK FESQUET,* FREDERIC LION,* LIONEL LEBRUN,*
MAGALI COMMANDEUR,* ALAIN FRAISSE,‡ DOMINIQUE METRAS,* AND BERNARD KREITMANN*
We evaluated the results of using extracorporeal membrane
oxygenation (ECMO) as resuscitation for cardiac patients un-
dergoing cardiopulmonary resuscitation (CPR) in our setting
where neither perfusionists nor surgeons are always on site,
and no circuit may be ready. Between 2003 and 2006, we
used ECMO for all cardiac patients who underwent cardiac
arrest in the pediatric intensive care unit (PICU) or Cath
Laboratory. We reviewed retrospectively 14 consecutive files
(15 episodes). Mean CPR time before ECMO institution was
44 minutes (10–110 minutes). The surgeons, perfusionist, and
scrub nurse, not on site for three of these patients, had to be
called in simultaneously with institution of CPR. Two died on
ECMO, the third one was successfully transplanted after 5
days. Globally, 10 patients could be weaned (66%). Eight
patients (57%) survived to hospital discharge, seven without
obvious neurological damage. One patient was bridged to a
left ventricular assist device (LVAD) and was eventually suc-
cessfully transplanted. He had an ischemic brain lesion with
good recuperation and no sequel. We obtained good results
with resuscitation ECMO in our setting where a permanently
on-site rapid deployment ECMO team is not present at all
times. ASAIO Journal 2007; 53:692–695.
Extracorporeal membrane oxygenation (ECMO) use for re-
suscitation is increasing. Ideally, resuscitation ECMO im-
plies having permanently on site a surgical team ready to
cannulate in emergency (surgeon, nurses, perfusionist), as
well as a circuit primed ready; however, this is not achiev-
able in all centers. We report our results with resuscitation
ECMO in our center where no circuit is kept primed ready
and where the surgical team is not on site after hours, and
we propose a minimal setting to achieve resuscitation
Patients and Methods
Our pediatric cardiac surgery unit is located in a children’s
hospital. The cardiac surgical team includes three surgeons,
three perfusionists, five scrub nurses, and two surgical resi-
dents. After hours, the on-call cardiac surgery team is com-
posed of one surgeon, one perfusionist, one resident, and one
scrub nurse, who are not on site. Patients undergoing cardiac
surgery are cared for postoperatively in a general pediatric
intensive care unit (PICU).
Between 2003 and 2006, we performed ECMO a total of 49
times in 43 patients with a 60% survival to hospital discharge
(26 of 43 patients). We used ECMO for all cardiac patients who
underwent a cardiac arrest not responding rapidly to usual
resuscitation in the PICU or Catheterization Laboratory.
We reviewed retrospectively the files of 14 consecutive
patients (15 episodes) who underwent resuscitation ECMO
(installed during active chest compression). Eleven patients
were in PICU after cardiac surgery and two for severe cardio-
myopathy. Two patients were cannulated in the Cath Labora-
tory after an interventional procedure. One of them already
had a postoperative resuscitation ECMO during a previous
hospitalization. Pathologies and procedures leading to ECMO
are detailed in Table 1.
Method for ECMO Alert
The intensive care and catheterization teams called the
surgical team for implantation as needed. When the problem
occurred after hours, the intensive care team had a very low
threshold to call for ECMO. The surgical team, composed of
one surgeon, one resident, one scrub nurse, and one perfu-
sionist, gathered on site while blood was obtained for priming.
All ECMO materials, i.e., ECMO circuit, oxygenator (con-
nected just before priming), were stored along with the pump
and a cart was ready with the rest of the material (connectors,
cannulae, clamps). The circuit was simplified (Figure 1) to be
connected and deaired rapidly as we did not keep any primed
ready circuit. We never used crystalloid priming. Blood could
be obtained within approximately 20 minutes.
The circuit included a nonocclusive pump and pretreated
tubing and cannulas. Oxygenators were adapted to patients’
body surface area (Minimax or Affinity, Medtronic Inc., Min-
neapolis, MN). More recently, we used long-term oxygenators
specifically designed for ECMO (Jostra Quadrox, Maquet cardio-
From the Departments of *Thoracic and Cardiovascular Surgery,
†Pediatric Intensive Care, and ‡Pediatric Cardiology, Children’s Hos-
pital La Timone, Marseille, France.
Submitted for consideration May 2007; accepted for publication in
revised form June 2007.
Presented in part at the Third International Conference on Pediatric
Mechanical Circulatory Support Systems and Pediatric Cardiopulmo-
nary Perfusion, May 17–19, 2007, Hershey, PA.
Reprint Requests: Olivier Ghez, MD, Department of Thoracic and
Cardiovascular Surgery, Children’s Hospital La Timone, 13005 Mar-
ASAIO Journal 2007
pulmonary AG, Germany, or Lilliput2 D902 ECMO, Dideco,
Sorin Group, Italy). The pump used was nonocclusive in all cases
(Biomedicus, Medtronic Inc., Minneapolis, MN or more recently
Revolution, Cobe, Sorin Group, Mirandola, Italy).
With this organization, the perfusionist could prepare in a
few minutes the necessary material, adapted to the weight of
the child, which had to be communicated at the time of the
call. The scrub nurse came with an emergency cart, a front
light, and a diathermy helped by the surgical resident. Mean-
while, the surgeon installed the patient with the intensivist or
anesthetist, as cardiopulmonary resuscitation (CPR) was per-
formed. In all cases, an arterial line was available to check the
Table 1. Patients’ Characteristics
PatientDiagnosis OperationWeaned DischargedMode of Death
Double outlet right ventricle
Double inlet single ventricle
Pulmonary atresia VSD
Post heart transplantation
Small left ventricle
Coarctation repair and PA banding
Pulmonary artery catheter dilation
Norwood type operation
Catheter biopsy for acute rejection
Bridge to transplantation
Aortic coarctation repair
PA-VSD-MAPCAs, pulmonary atresia with ventricular septal defect and major aortopulmonary arteries; VSD, ventricular septal defect; Y,
yes; N, No; MOF, multiorgan failure.
Figure 1. Extracorporeal membrane oxygen-
ation (ECMO) circuit.
RESUSCITATION ECMO FOR CARDIAC PATIENTS
efficacy of the CPR and to perform blood gas analysis. More
recently, near infrared spectroscopy (NIRS) was also available
(Invos, Somanetics, Troy, MI). The operation began while the
circuit was prepared and primed (which took altogether 15–20
minutes). Central cannulation between the right atrium and
aorta was performed in postoperative patients to gain time.
Peripheral venoarterial cannulation through the neck or fem-
oral vessels was performed in other cases. Cardiopulmonary
resuscitation was suspended for several 10 second intervals to
allow for dissection and cannulation. Heparin (100 U/kg) was
administered and cannulation was performed to start ECMO as
rapidly as possible. Cannulas were then secured to the skin,
after checking that a good, stable flow was obtained, or mo-
bilized if necessary. Dressing was then addressed and the child
transported to intensive care if the procedure took place in the
Catheterization Laboratory. Peritoneal dialysis or hemodiafiltra-
tion connected on the ECMO circuit (Prisma, Gambro/Hospal AG,
Germany) was instituted if needed.
This study is descriptive. Data are presented as mean ?
standard deviation. Median and extreme values are mentioned
Fifteen episodes occurred in 14 patients (Table 1). Mean age
was 43 ? 73 months (6 days to 21 years, median 14.5 months),
weight 14 ? 13 kg (3.1–47 kg). The surgeons, perfusionist, and
scrub nurse were called in simultaneously with institution of
CPR in most cases. Mean CPR time before ECMO institution was
44 ? 27 minutes (10–110 minutes). Mean ECMO support was
144 ? 82 hours. On three occasions, the surgical team was not
on site but could be called before the arrest occurred. Two of
these patients (patients 8 and 13) died on ECMO, the third one
(patient 7) was successfully transplanted after 5 days. Central
cannulation was used in seven cases, peripheral cannulation
in eight (six through the neck and two through the femoral
Ten of 14 patients could be weaned (66%). Eight patients
(57%) survived to hospital discharge. Causes of death are
detailed in Table 1. Of note, one 21-year-old patient (4) died of
a massive cerebral ischemic lesion. One 2-year-old patient
had to be placed on resuscitation ECMO in the postoperative
period of a pulmonary atresia VSD MAPCAs repair (6). He
could be weaned and eventually discharged. However, 5
months later, he suffered a cardiac arrest during an interven-
tional catheterization procedure to dilate a pulmonary artery.
He did not survive despite ECMO implantation in the Cathe-
terization Laboratory, dying of multiorgan failure. Patient 11
presented with severe acute rejection after heart transplanta-
tion. This 6-year-old patient underwent a cardiac catheteriza-
tion and suffered a cardiac arrest when a myocardial biopsy
was done. She underwent CPR for 110 minutes before ECMO
could be instituted. She recovered after treatment of the rejec-
tion. She could be weaned and discharged with no sequel.
Sternal closure was delayed in nine cases for a duration of
8 ? 5 days (up to 21 days), mostly at the beginning of the
experience. Four of these patients underwent one to four re-
entries for bleeding. Two patients developed a mediastinitis.
One patient had his sternal closure delayed for 21 days and
necessitated application of a vacuum assisted drainage to pro-
vide healing of the sternotomy site, as primary closure of the
skin was not possible on the sternum. A skin graft was then
Ten patients underwent dialysis (four survivors to hospital dis-
charge). Peritoneal dialysis was used in five cases. Hemodialysis
was used in seven cases (in two cases after failure of peritoneal
dialysis). The hemodialysis was connected to the ECMO circuit
(from the arterial line, before the oxygenator, to the venous line).
One 8-month-old patient was bridged to a left ventricular assist
device (LVAD) and was eventually successfully transplanted. He
had an ischemic brain lesion with good recuperation and no
sequel. This lesion, detected only after transplantation, could be
attributed either to the ECMO period, the LVAD period, or the
transplantation intervention with no certainty.
We could apply ECMO resuscitation in this setting of pa-
tients with a survival to discharge of 57% (8 of 15 episodes),
comparable to the literature. A rapid deployment ECMO team
as described by Duncan et al. seems the way to achieve the
best results.1This report from the Boston children’s group
evaluated the results with a permanently ready, on-site, rapid
deployment ECMO team with a modified circuit allowing
institution of support within 15 minutes of notification. In the
rapid resuscitation group, compared with a historical control
group, CPR time was ?30 minutes less with a mean duration
of 55 minutes. The survival improved to 64% of 11 patients.
Our setting does not allow having a team permanently on
site, ready for cannulation, with a primed circuit. Proposing
the possibility of resuscitation ECMO was possible with the
collaboration of the intensivists, cardiologists, surgeons, per-
fusionists, and nurses. Extracorporeal membrane oxygenation
was set as a systematic therapy for cardiac patients undergoing
untreatable low cardiac output or cardiac arrest so that the
surgical team could be called as early as possible. This attitude
was developed after the good results we obtained with ECMO
in the postoperative period after cardiac surgery.2
The material was stored for rapid access. After hours, an
ECMO alert had to be called when a patient was degrading
rapidly despite medical therapy, so the surgical team could
gather before the onset of a cardiac arrest. The threshold for
ECMO alert was therefore decided to be low. The transport of
the material from the operative theater could be rapidly done
so that cannulation in the PICU or Catheterization Laboratory
could be realized as has been advocated by others.3
Duration of CPR before the start of ECMO circulation remains
a concern and should be kept as low as possible before ECMO
institution. However, several reports showed that good outcome
is possible even with very long CPR before ECMO.1,4–8It is
difficult to establish a relationship between duration of CPR and
outcome in the ECMO group, as shown by the Children’s Hos-
pital of Philadelphia group.6This group reported their experience
GHEZ ET AL.
with extracorporeal resuscitation in 64 patients, with a survival to
discharge of 33%. They showed that cardiac patients had a better
survival than noncardiac patient (44% vs. 9.5%), and that CPR
duration before ECMO institution was not significantly different
between survivors and nonsurvivors (50 vs. 46 minutes). The
good flow and oxygenation provided by ECMO may explain the
advantage in terms of outcome we can obtain as compared with
conventional means after which a patient will most likely be in
low cardiac output. These good results are also favored by the
optimal CPR, which can be provided in the PICU or catheteriza-
tion laboratory with continuous arterial pressure monitoring, near
infra red spectroscopy cerebral monitoring, and invasive ventila-
tion. Viewing these results, it seemed reasonable to propose
ECMO resuscitation even without an on-site dedicated ECMO
team. Total time from call to ECMO in our center can be esti-
mated at 40 minutes (20 minutes enroute to the hospital plus 20
minutes preparation and cannulation) at best before ECMO. Our
mean duration of CPR before ECMO was 44 minutes. One of our
patients had a very long CPR of 110 minutes before institution of
ECMO in the Cath Laboratory. She however survived without
neurological sequel. The surgical team was not on site only on
three occasions; two of these patients died. These patients were
both in the postoperative period of a complex coarctation repair
in the setting of a relative left ventricular hypoplasia. Whether the
biventricular repair intended was a poor indication or whether
cases the call for ECMO was prompted by a low cardiac output
state. The cardiac arrest occurred in both cases as the team was
already there, preparing for cannulation, in the intensive care
unit. The third patient with a severe cardiomyopathy had a car-
diac arrest prompting the call for ECMO. He was resuscitated and
recovered a cardiac rhythm as the team gathered. He underwent
a second arrest as the team was preparing for cannulation. This
patient was successfully bridged to transplantation and survived
without neurological damage.
Mode of death was either multiorgan failure or catastrophic
brain lesions leading to withdrawal of support. Specific mor-
bidity in our series concerned patients undergoing central
cannulation and prolonged period on ECMO. Mediastinitis
occurred in these cases, as well as sternotomy healing prob-
lems. Currently, we still undergo central cannulation if the
patient is in the early postoperative period as it is the fastest
vascular access. We convert this cannulation to a peripheral
cannulation after 3–5 days if the duration of ECMO is expected
to exceed 5 days.
our series. However, a precise evaluation is under investigation,
and we know from other series that the prevalence of neurolog-
ical injury is a concern in the ECMO population.9,10Outcome is,
however, demonstrated to be better after ECMO resuscitation
than after conventional CPR alone.6,11Cost-effectiveness of this
approach has also been studied favorably.12Long-term neurolog-
ical evaluation of our patients is being undertaken and will be
We achieved good results in our series in terms of survival
with resuscitation ECMO, in a setting not allowing a perma-
nently on-site rapid deployment ECMO team. Close coopera-
tion with the intensive care, cardiology, perfusion, nursing,
and surgical team is necessary.
1. Duncan BW, Ibrahim AE, Hraska V, et al: Use of rapid-deployment
extracorporeal membrane oxygenation for the resuscitation of pe-
diatric patients with heart disease after cardiac arrest. J Thorac
Cardiovasc Surg 116: 305–311, 1998.
2. Ghez O, Feier H, Ughetto F, et al: Postoperative extracorporeal
life support in pediatric cardiac surgery: Recent results. ASAIO
J 51: 513–516, 2005.
3. Allan CK, Thiagarajan RR, Armsby LR, et al: Emergent use of
extracorporeal membrane oxygenation during pediatric cardiac
catheterization. Pediatr Crit Care Med 7: 212–219, 2006.
4. del Nido PJ, Dalton HJ, Thompson AE, Siewers RD: Extracorporeal
membrane oxygenator rescue in children during cardiac arrest
after cardiac surgery. Circulation 86: II300–II304, 1992.
5. Morris MC, Ittenbach RF, Godinez RI, et al: Risk factors for mor-
tality in 137 pediatric cardiac intensive care unit patients man-
aged with extracorporeal membrane oxygenation. Crit Care
Med 32: 1061–1069, 2004.
6. Morris MC, Wernovsky G, Nadkarni VM: Survival outcomes after
extracorporeal cardiopulmonary resuscitation instituted during
active chest compressions following refractory in-hospital pe-
diatric cardiac arrest. Pediatr Crit Care Med 5: 440–446, 2004.
7. Kelly RB, Porter PA, Meier AH, et al: Duration of cardiopulmonary
resuscitation before extracorporeal rescue: How long is not
long enough? ASAIO J 51: 665–667, 2005.
8. Yu HY, Yeh HL, Wang SS, et al: Ultra long cardiopulmonary
resuscitation with intact cerebral performance for an asystolic
patient with acute myocarditis. Resuscitation 73: 307–308,
9. Petrou S, Bischof M, Bennett C, et al: Cost-effectiveness of neo-
natal extracorporeal membrane oxygenation based on 7-year
results from the United Kingdom Collaborative ECMO Trial.
Pediatrics 117: 1640–1649, 2006.
10. Hanekamp MN, Mazer P, van der Cammen-van Zijp MH, et al:
Follow-up of newborns treated with extracorporeal membrane
oxygenation: A nationwide evaluation at 5 years of age. Crit
Care 10: R127, 2006.
11. de Mos N, van Litsenburg RR, McCrindle B, et al: Pediatric
in-intensive-care-unit cardiac arrest: Incidence, survival, and
predictive factors. Crit Care Med 34: 1209–1215, 2006.
12. Mahle WT, Forbess JM, Kirshbom PM, et al: Cost-utility analysis of
salvage cardiac extracorporeal membrane oxygenation in chil-
dren. J Thorac Cardiovasc Surg 129: 1084–1090, 2005.
RESUSCITATION ECMO FOR CARDIAC PATIENTS