- Email: [email protected]
Extracorporeal Membrane Oxygenation Is a Team Sport: Institutional Survival Benefits of a Formalized ECMO Team
Accepted Manuscript
ECMO Is a Team Sport: Institutional Survival Benefits of a formalized ECMO Team Adam A. Dalia MD,MBA , Jamel Ortoleva MD , Amy Fiedler MD , Mauricio Villavicencio MD , Ken Shelton MD , Gaston D. Cudemus MD PII: DOI: Reference:
S1053-0770(18)30390-2 10.1053/j.jvca.2018.06.003 YJCAN 4751
To appear in:
Journal of Cardiothoracic and Vascular Anesthesia
Received date:
27 March 2018
Please cite this article as: Adam A. Dalia MD,MBA , Jamel Ortoleva MD , Amy Fiedler MD , Mauricio Villavicencio MD , Ken Shelton MD , Gaston D. Cudemus MD , ECMO Is a Team Sport: Institutional Survival Benefits of a formalized ECMO Team, Journal of Cardiothoracic and Vascular Anesthesia (2018), doi: 10.1053/j.jvca.2018.06.003
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 Is a Team Sport: Institutional Survival Benefits of a formalized ECMO Team
(Both authors played an equal first author role)
Adam A. Dalia, MD,MBA1, Jamel Ortoleva, MD2, Amy Fiedler, MD3, Mauricio Villavicencio, MD3, Ken Shelton, MD1, Gaston D. Cudemus, MD1
AN US
CR IP T
1. Division of Cardiac Anesthesiology, Massachusetts General Hospital, Harvard Medical School 2. Division of Cardiac Anesthesiology, Beth-Israel Deaconess Medical Center, Harvard Medical School 3. Division of Cardiac Surgery, Massachusetts General Hospital, Harvard Medical School
Address for Correspondence:
M
Adam A. Dalia, MD, MBA 55 Fruit Street, Boston MA 02114 Tel: 617-726-3030 Email: [email protected]
CE
Abstract
PT
ED
Both authors played an equal first author role
AC
Objective: At our institution, prior to 2014, patients requiring care in the peri-ECMO period were treated by intensivists with specific training in ECMO but worked independently. This isolated form of care was addressed in late 2013 with the formal initiation of an ECMO team. We wanted to assess the difference in overall mortality for ECMO patients cared for prior to the initiation of a multidisciplinary team compared to after its establishment.
Design: This was a retrospective chart review based study.
ACCEPTED MANUSCRIPT Setting: This was a single center university based hospital setting.
Participants: The study included all adult patients at our institution who required ECMO support between the years of 2009 and 2017.
Intervention: The new multidisciplinary ECMO team established a set of protocols and
CR IP T
guidelines to care for ECMO patients. The formal ECMO team consisted of: cardiac surgery, cardiac anesthesia, intensivists, cardiology heart failure specialist, ICU nursing (NP/RN), perfusion services, respiratory therapy, nutrition, physical and occupational therapy, and an
AN US
ethics committee member.
Measurements and Main Results: Manual chart review was conducted and survival to discharge was collected and separated into two groups, 2009-2013 (pre-ECMO team) and 2014-2017 (post-ECMO team). There was a total of 279 patient charts reviewed who required
M
ECMO support. Survival to discharge for patients between 2009-2013 was 37.7% compared
ED
to a survival to discharge of 52.3% between 2014-2017 (P value = 0.02).
PT
Conclusions: Patients cared for after the initiation of an ECMO team showed improved
CE
survival compared to patients cared for prior to the creation of the ECMO team.
Keywords:
AC
Extracorporeal Membrane Oxygenation; ECMO; Multidisciplinary team; Extracorporeal life support; ECLS
Introduction
ACCEPTED MANUSCRIPT Extracorporeal membrane oxygenation (ECMO) has become an established treatment strategy for patients in cardiogenic shock or acute respiratory failure. Either veno-arterial (VA) or veno-venous (VV) cannulation strategies may be implemented depending on patient presentation. According to the January 2018 Extracorporeal Life Support Organization (ELSO) registry, the number of ECMO cases and centers have nearly doubled between 2010
CR IP T
and 20171. Research has demonstrated that ECMO patient outcomes, specifically
complications and in-hospital survival, relate closely with case load and volume at the patient center2.
AN US
The concept of the „Heart Team‟, used widely in cases of complex cardiovascular disease, specifically coronary revascularization and transcatheter aortic valve replacement, has demonstrated success in optimizing the management of complex patients using multidisciplinary teams. In these situations, the „Heart Team‟, composed of a cardiac
M
surgeon, cardiac anesthesiologists, interventional cardiologist, primary cardiologist, nursing
ED
staff, and a patient advocate allows the team to focus on patient specific factors and considerations. This approach has led to improved clinical outcomes and patient satisfaction.
PT
Data collection for the ECMO program at the Massachusetts General Hospital (MGH)
CE
began in 2009. When ECMO was initiated at our center patients were cared for in various intensive care units (ICU) by specialized intensivists, functioning in a siloed fashion. In an
AC
effort to improve outcomes, a formal „MGH ECMO Team‟ was instituted in 2014. The MGH Multidisciplinary ECMO team consists of cardiac surgeons, cardiac anesthesiologists, intensivists, cardiologists, ICU nursing staff, perfusionists, respiratory therapists, nutritionists, physical and occupational therapists, and an ethics committee member. In addition, VA ECMO patients were exclusively cared for in the cardiac surgical ICU and VV ECMO patients were cared for in the MICU. The creation of this team allowed for the establishment of specific protocols and guidelines to provide comprehensive care for ECMO
ACCEPTED MANUSCRIPT patients. In an effort to evaluate the performance of our ECMO team, we compared ECMO patients cared for prior to the institution of the ECMO team to those managed with the new collaborative ECMO team approach. We have chosen to focus on survival to discharge between these two groups.
CR IP T
Methods: This study is a retrospective, single institution cohort study approved by the Partners Institutional Review Board (IRB). We identified all patients greater than 18 years of age who were placed on VA or VV ECMO support between 2009 and 2017 from our institutional
AN US
database. Manual chart review was performed on all patients to identify key demographic, clinical, and survival to discharge data. Following data collection, the data and patients were sorted into two groups: pre-MGH ECMO Team 2009-2013 and post-MGH ECMO Team
M
2014-2017. Additionally, subgroup analysis of patients on VA and VV ECMO was performed to evaluate utilization and survival changes.
ED
Statistics
PT
Data were collected and analyzed in Microsoft Excel spread sheets. Pearson‟s chi squared test was utilized to assess for differences in mortality between the pre- MGH ECMO team era
CE
(2009-2013) and the MGH ECMO team era (2014-2017). A p-value of less than 0.05 was
AC
selected as a cutoff to reject the null hypothesis (defined as no difference in survival between pre ECMO team and post ECMO team cohorts). Pearson‟s chi squared test was also used to assess for differences in the utilization of VV and VA ECMO between the two cohorts, again with a P value less than 0.05 selected as a cutoff to reject the null hypothesis (which was no difference in the utilization of VA and VV ECMO as a percent of total ECMO runs).
ACCEPTED MANUSCRIPT Two tailed T tests were used to assess for differences between ECMO duration. A P-value of less than 0.05 was used as the cutoff for a statistically significant difference between the groups. Depending on the results of the F-test, the T tests were either heteroscedastic (if the F-test showed a statistically significant difference in variance of the compared groups) or homoscedastic (if the F-test showed no statistically significant difference in variance between
CR IP T
the compared groups). Results:
From 2009-2017 a total of 279 patients were placed on ECMO at our institution;130 patients
AN US
between 2009-2013 and 149 patients between 2014-2017. Survival to discharge for patients between 2009-2013 was 37.7% (49/130) compared to survival to discharge of 52.3% (78/149) between 2014-2017 (p=0.02). When examining solely VA ECMO patients, survival
M
to discharge was 37.4% (37/99) from 2009-2013 and 48.6% (36/74) from 2014-2017. Furthermore, VV ECMO patients had a 40% (10/25) survival to discharge from 2009-2013,
ED
and a 63% (29/46) survival to discharge from 2014-2017 (P = 0.06 by chi-squared test). The mixed ECMO group only had 33 patients between 2009 and 2017 and thus was not analyzed
PT
for comparison. Duration on VA ECMO was 4.74 4.84 days in the pre-ECMO team era and
CE
6.5 10.2 days in the post-ECMO era. Of note, the variance is significantly different between the two groups of the patients (F-test: P < 0.001) but the duration of ECMO is not (P = 0.164
AC
by heteroscedastic T-test). For the V-V ECMO cohort, duration of cannulation between the pre-ECMO team era in this subgroup was 14.52 21.5 days and 14.54 28.6 days in the post-ECMO team era with neither the variance or duration achieving statistical significance (F-test: P = 0.136, homoscedastic T-test: P = 0.997). Prior to the creation of the MGH ECMO team, ICU placement of any patient cannulated for ECMO was dependent on the patients‟ original location prior to cannulation as
ACCEPTED MANUSCRIPT well as hospital bed resources. As a result, between 2009-2013, the 130 patients on ECMO were distributed between various units: 49 in the Cardiac Surgical Intensive Care Unit (CSICU), 4 in the Cardiac Care Unit (CCU), 9 in the Medical Intensive Care Unit (MICU) and 63 in the Surgical Intensive Care Unit (SICU). After implementation of the multidisciplinary MGH ECMO team, a protocolized approach was created for the placement
CR IP T
of these patients. Between 2014-2017, the distribution of ICU location for the 149 patients changed markedly: 94 in the CSICU, 0 in the CCU, 34 in the MICU, 1 in the SICU and 1 offunit in the cardiac catheterization laboratory (Table 1). Selected VV ECMO patients started receiving care in the medical intensive care unit as a result of an initiative to create more bed
AN US
space in the CSICU for patients requiring specialized surgical care. Prior to putting these patients in the MICU, multiple intensivists with VV ECMO training were recruited and trained the existing intensivist group. Additionally, bedside staff including nurses and
M
respiratory therapists were trained for the care of these patients.
ED
Discussion:
As the utilization of ECMO increases for patients with cardiogenic and respiratory failure,
PT
institutions are striving to improve outcomes, specifically survival to discharge in this
CE
complex patient population. Volume and outcomes at individual centers with respect to ECMO utilization have been reported in both VA and VV ECMO populations; it has been
AC
suggested that high-volume centers have lower mortality2,3. Teerapuncharoen et al examined the ECMO experience at The University of Alabama at Birmingham as their institution moved from a low to a high volume ECMO center4. Their work suggests that higher annual ECMO volume, and therefore more experience, improves survival to hospital discharge4. Through multivariable logistic regression, they identified that low annual volume, advanced patient age, and prolonged hospital length of stay are associated with reduced survival to hospital discharge.
ACCEPTED MANUSCRIPT Separately, Barbaro et. al utilized the Extracorporeal Life Support (ELSO) Registry to evaluate ECMO outcomes for the neonatal, pediatric, and adult populations3. This registry encompassed data from more than 56,000 patients across 290 centers. With respect to adults, from 2008 onwards, hospital volume played a significant role in adult mortality rates. Despite identifying and describing a relationship between center volume and patient outcomes, the
CR IP T
study did not attempt to elucidate the reason mortality was improved at these centers. What remains to be resolved is whether experience in it of itself is necessary to foster good outcomes or if a protocolized approach is what is needed for improved outcomes.
AN US
The use of multi-disciplinary teams in the inpatient hospital setting has been shown to limit adverse events, improve outcomes, and adds to patient and employee satisfaction5. Epstein has published widely on this subject, providing the conclusion that the best medical and surgical care is a “team sport”6. Epstein‟s group has demonstrated that cohesive
M
teamwork has improved communication between different levels of healthcare workers,
ED
allowing for improved patient outcomes and decreased length of stay. Despite the persuasive argument from Epstein and others with respect to the
PT
importance of multi-disciplinary care teams, very few studies have been published regarding
CE
institutional experience with a high volume ECMO center and a protocolized, team based approach4,7,8. There is a recent report by Cotza et. al describing their centers ECMO team
AC
approach at a tertiary care hospital in Italy9. Created in 2012, the team consists of cardiac surgeons, anesthesiologists, intensivists, nurses, and perfusionists. After one year of a multidisciplinary approach, retrospective analysis suggested improved survival to discharge, fewer bleeding complications, and fewer blood product transfusions9. Prior to the creation of the MGH ECMO Team, our institutional policy was to have an intensivist with ECMO experience manage the patients in a siloed approach. After initiation
ACCEPTED MANUSCRIPT of the MGH ECMO Team, both a multi-disciplinary care team and a set of protocols and guidelines for the institution of ECMO support were implemented (Figure 1, Appendix A-D). Initiation and implementation of the MGH ECMO team fostered a more collaborative approach to caring for these patients with improved outcomes. An impressive improvement in survival from 37.7% to 52.3% was noted; a 14.6% absolute improvement (NNT≈7).
CR IP T
Further analysis is needed to determine what other factors contributed to this substantial survival improvement.
Importantly, the number of VV ECMO patients drastically increased during the
AN US
analyzed period (25/130 from 2009-2013 vs 46/149 from 2014-2017, p=0.01). Additionally, the number of VA ECMO patients significantly decreased (99/130 from 2009-2013 vs 76/149 from 2014-2017, p <0.001). This trend may be multifactorial; however, it suggests a more prudent selection of therapeutic modalities and possibly better patient selection. This survival
M
improvement is also consistent with the ELSO registry findings in which patients on VV
ED
ECMO have a higher survival compared to patients on VA ECMO. Provider experience is also important, but throughout this analyzed period (2009-2017) there was significant staff
CE
experience.
PT
turnover reducing the likelihood that the improved survival is solely a result of provider
Based on our experience, there are several key components to our MGH ECMO Team
AC
and associated protocols which have contributed significantly to the improvement in our patient outcomes (Appendix A-D). One important component is the ICU location where patients on ECMO are cared for. Prior to the institution the MGH ECMO Team, patients supported with ECMO were cared for in multiple intensive care units throughout the hospital, including the Surgical Intensive Care Unit (SICU), Medical Intensive Care Unit (MICU), Cardiac Surgical Intensive Care Unit (CSICU), and the Cardiac Care Unit (CCU). We, as an institution, recognized that patients may benefit by care being performed in intensive care
ACCEPTED MANUSCRIPT units experienced in ECMO care. Thus, VA-ECMO patients are now exclusively cared for within the CSICU, and VV-ECMO patients are cared for within the MICU. Having dedicated intensive care units to care for ECMO patients has allowed us to focus on ECMO specific training for the nursing staff and care teams. Focusing these patients into the CSICU or MICU has streamlined care and improved experience and comfort in bedside providers caring
CR IP T
for these patients.
This study does have some notable limitations, including its single center
retrospective design. Particularly, given the fact the study was done over an 8-year period
AN US
advances in medicine that may be relevant to ECMO patients may have skewed the mortality in favor of the more recent cohort. Additionally, the statistical analysis performed in this study did not control for potential confounders that may have affected our results. Furthermore, data on blood transfusions, vasopressor use, or infection rates where not
M
analyzed. Despite these limitations, this is currently the largest single center review of ECMO
AC
CE
PT
ED
outcomes with respect to the effect of instituting a team based approach on mortality.
ACCEPTED MANUSCRIPT Conclusion: In summary, the implementation of a multi-disciplinary team-based approach to ECMO care has played an integral role in the improvement in survival to discharge of ECMO patients at our institution. This manuscript may serve as a framework to assist other
CR IP T
institutions interested in starting or searching for possible means to improve an ECMO
AC
CE
PT
ED
M
AN US
program.
ACCEPTED MANUSCRIPT References: 1. Extracorporeal Life Support Organization. Registry. https://www.elso.org/Registry/Statistics.aspx. Accessed on February 28,2018 2. Mccarthy FH, Mcdermott KM, Spragan D, et al. Unconventional Volume-Outcome Associations in Adult Extracorporeal Membrane Oxygenation in the United States. Ann
CR IP T
Thorac Surg. 2016;102(2):489-95.
3. Barbaro RP, Odetola FO, Kidwell KM, et al. Association of hospital-level volume of
extracorporeal membrane oxygenation cases and mortality. Analysis of the extracorporeal life support organization registry. Am J Respir Crit Care Med. 2015;191(8):894-901.
AN US
4. Tchantchaleishvili V, Hallinan W, Massey HT. Call for Organized Statewide Networks for Management of Acute Myocardial Infarction-Related Cardiogenic Shock. JAMA Surg. 2015;150(11):1025-6.
5. Moll V, Teo EY, Grenda DS, et al. Rapid Development and Implementation of an ECMO
M
Program. ASAIO J. 2016;62(3):354-8.
6. Epstein NE. Multidisciplinary in-hospital teams improve patient outcomes: A review. Surg
ED
Neurol Int. 2014;5(Suppl 7):S295-303.
PT
7. Na SJ, Chung CR, Choi HJ, et al. The effect of multidisciplinary extracorporeal membrane oxygenation team on clinical outcomes in patients with severe acute respiratory failure. Ann
CE
Intensive Care. 2018;8(1):31. 8. Fiedler AG, Song TH, D'alessandro DA. Redesigning Care for Patients With Acute
AC
Myocardial Infarction Complicated by Cardiogenic Shock: The "Shock Team". JAMA Surg. 2016;151(7):685.
9. Cotza M, Carboni G, Ballotta A, et al. Modern ECMO: why an ECMO programme in a tertiary care hospital. Eur Heart J Suppl. 2016;18(Suppl E):E79-E85.
ACCEPTED MANUSCRIPT Figure Legend:
AC
CE
PT
ED
M
AN US
CR IP T
Figure 1. Institutional ECMO initiation algorithm
CR IP T
ACCEPTED MANUSCRIPT
AN US
Figure Appendix A. Nursing ECMO protocols Appendix A: Nursing ECMO Protocols Education:
ICU Nurse teach annual ECMO class for all nursing staff caring for ECMO patients.
-
Any new CCRN/NP/PA to the CSICU must take the introductory ECMO class.
-
All nursing staff must take a retraining ECMO class each year.
-
ECMO training classes are offered on a bi-monthly basis.
ED
M
-
Patient Care:
-
Daily sedation “holidays”
AC
CE
PT
-
All new ECMO patients have a cooler of 4 PRBCs and 4 FFP at their bedside for the first 24 hours‟ post cannulation. No PIVs are to be placed or removed by nursing staff while on ECMO due to high risk for compartment syndrome and need for fasciotomy. Blood cultures are to be done by mid-level staff or higher. Assess and document pupils q2hrs
-
CR IP T
ACCEPTED MANUSCRIPT
-
Sterile dressing changes daily
-
Emergency‟s require MD notification: bleeding, air in the circuit, accidental decannulation, lethal Arrhythmias/CPR
-
Transfusion Thresholds:
AN US
o Hgb > 8.5 o If SVO2 is low, keep Hgb > 10.0 o Plts > 20,000 o If bleeding, keep > 80, 000 o INR < 2.5
M
o If bleeding, keep < 1.6 o Fibrinogen > 150 Anticoagulation goals:
ED
-
o Heparin Bolus (typically 5,000-10,000 units IVP) on initiation of ECMO or weaning/clamping ECMO Circuit
CE
PT
o Heparin Infusion starting at 10units/kg/hr titrating to goal PTT 60-80 (if bleeding, 40-60)
AC
Figure Appendix B-D. Electronic medical record ECMO order sets
AC
CE
PT
ED
M
AN US
CR IP T
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
M
AN US
CR IP T
ACCEPTED MANUSCRIPT
Table 1. ECMO Patient Data Analysis
CR IP T
ACCEPTED MANUSCRIPT
2009 – 2013 (n=130)
2014-2017 (n=149)
p-value
Patient age (years)
50.96 ± 14.9
52.8 ± 14.5
48.8 ± 15.1
0.03
ECMO duration (days)
9.3 ± 17.8
6.9 ± 11.6
11.43 ± 21.15
0.025
VA
175 (62.7)
99 (76.2)
76 (51)
<0.001
VV
71(25.4)
25 (19.2)
46 (30.8)
0.01
Mixed
33 (12.45)
AN US
Total (n=279)
ED
Unit
M
Mode of ECMO
6 (4.6)
27 (18.1)
154 (55.2)
49 (37.7)
105 (70.5)
CCU
4 (1.4)
4 (3.1)
0 (0.0)
MICU
46 (16.5)
9 (6.9)
37 (24.8)
64 (22.9)
63 (48.5)
1 (0.7)
4 (1.4)
3 (2.3)
1 (0.7)
PT
CSICU
AC
Cath lab
CE
SICU
<0.001
CR IP T
ACCEPTED MANUSCRIPT
7 (2.5)
2 (1.5)
5 (3.4)
Survived to discharge
127 (45.5)
49 (37.7)
78 (52.3)
AC
CE
PT
ED
M
AN US
Multiple units
0.021
Table 2: ECMO Indications for Cannulation
IPF: 7
Flu/PNA: 3
Pulmonary Embolism: 3
VA-ECMO ‘09-13:
MI: 25
Post-cardiotomy: 12
Pulmonary Embolism: 13
ARDS: 12
Flu/PNA: 9
IPF: 4
MI: 13
Postcardiotomy: 7
Cardiac arrest: 6
Unspecified: 21 Pulmonary Embolism: 5
AC
CE
PT
ED
M
VV-ECMO ‘14-17: VA-ECMO ‘14-17:
Other/unspecified: 12 Myocarditis: 5 Other/unspecified: 44
AN US
VV-ECMO ‘09-13:
CR IP T
ACCEPTED MANUSCRIPT
Cardiomyopathy: Myocarditis: VT storm: 4 4 3
Other/unspecified: 32
ECMO Is a Team Sport: Institutional Survival Benefits of a formalized ECMO Team Adam A. Dalia MD,MBA , Jamel Ortoleva MD , Amy Fiedler MD , Mauricio Villavicencio MD , Ken Shelton MD , Gaston D. Cudemus MD PII: DOI: Reference:
S1053-0770(18)30390-2 10.1053/j.jvca.2018.06.003 YJCAN 4751
To appear in:
Journal of Cardiothoracic and Vascular Anesthesia
Received date:
27 March 2018
Please cite this article as: Adam A. Dalia MD,MBA , Jamel Ortoleva MD , Amy Fiedler MD , Mauricio Villavicencio MD , Ken Shelton MD , Gaston D. Cudemus MD , ECMO Is a Team Sport: Institutional Survival Benefits of a formalized ECMO Team, Journal of Cardiothoracic and Vascular Anesthesia (2018), doi: 10.1053/j.jvca.2018.06.003
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 Is a Team Sport: Institutional Survival Benefits of a formalized ECMO Team
(Both authors played an equal first author role)
Adam A. Dalia, MD,MBA1, Jamel Ortoleva, MD2, Amy Fiedler, MD3, Mauricio Villavicencio, MD3, Ken Shelton, MD1, Gaston D. Cudemus, MD1
AN US
CR IP T
1. Division of Cardiac Anesthesiology, Massachusetts General Hospital, Harvard Medical School 2. Division of Cardiac Anesthesiology, Beth-Israel Deaconess Medical Center, Harvard Medical School 3. Division of Cardiac Surgery, Massachusetts General Hospital, Harvard Medical School
Address for Correspondence:
M
Adam A. Dalia, MD, MBA 55 Fruit Street, Boston MA 02114 Tel: 617-726-3030 Email: [email protected]
CE
Abstract
PT
ED
Both authors played an equal first author role
AC
Objective: At our institution, prior to 2014, patients requiring care in the peri-ECMO period were treated by intensivists with specific training in ECMO but worked independently. This isolated form of care was addressed in late 2013 with the formal initiation of an ECMO team. We wanted to assess the difference in overall mortality for ECMO patients cared for prior to the initiation of a multidisciplinary team compared to after its establishment.
Design: This was a retrospective chart review based study.
ACCEPTED MANUSCRIPT Setting: This was a single center university based hospital setting.
Participants: The study included all adult patients at our institution who required ECMO support between the years of 2009 and 2017.
Intervention: The new multidisciplinary ECMO team established a set of protocols and
CR IP T
guidelines to care for ECMO patients. The formal ECMO team consisted of: cardiac surgery, cardiac anesthesia, intensivists, cardiology heart failure specialist, ICU nursing (NP/RN), perfusion services, respiratory therapy, nutrition, physical and occupational therapy, and an
AN US
ethics committee member.
Measurements and Main Results: Manual chart review was conducted and survival to discharge was collected and separated into two groups, 2009-2013 (pre-ECMO team) and 2014-2017 (post-ECMO team). There was a total of 279 patient charts reviewed who required
M
ECMO support. Survival to discharge for patients between 2009-2013 was 37.7% compared
ED
to a survival to discharge of 52.3% between 2014-2017 (P value = 0.02).
PT
Conclusions: Patients cared for after the initiation of an ECMO team showed improved
CE
survival compared to patients cared for prior to the creation of the ECMO team.
Keywords:
AC
Extracorporeal Membrane Oxygenation; ECMO; Multidisciplinary team; Extracorporeal life support; ECLS
Introduction
ACCEPTED MANUSCRIPT Extracorporeal membrane oxygenation (ECMO) has become an established treatment strategy for patients in cardiogenic shock or acute respiratory failure. Either veno-arterial (VA) or veno-venous (VV) cannulation strategies may be implemented depending on patient presentation. According to the January 2018 Extracorporeal Life Support Organization (ELSO) registry, the number of ECMO cases and centers have nearly doubled between 2010
CR IP T
and 20171. Research has demonstrated that ECMO patient outcomes, specifically
complications and in-hospital survival, relate closely with case load and volume at the patient center2.
AN US
The concept of the „Heart Team‟, used widely in cases of complex cardiovascular disease, specifically coronary revascularization and transcatheter aortic valve replacement, has demonstrated success in optimizing the management of complex patients using multidisciplinary teams. In these situations, the „Heart Team‟, composed of a cardiac
M
surgeon, cardiac anesthesiologists, interventional cardiologist, primary cardiologist, nursing
ED
staff, and a patient advocate allows the team to focus on patient specific factors and considerations. This approach has led to improved clinical outcomes and patient satisfaction.
PT
Data collection for the ECMO program at the Massachusetts General Hospital (MGH)
CE
began in 2009. When ECMO was initiated at our center patients were cared for in various intensive care units (ICU) by specialized intensivists, functioning in a siloed fashion. In an
AC
effort to improve outcomes, a formal „MGH ECMO Team‟ was instituted in 2014. The MGH Multidisciplinary ECMO team consists of cardiac surgeons, cardiac anesthesiologists, intensivists, cardiologists, ICU nursing staff, perfusionists, respiratory therapists, nutritionists, physical and occupational therapists, and an ethics committee member. In addition, VA ECMO patients were exclusively cared for in the cardiac surgical ICU and VV ECMO patients were cared for in the MICU. The creation of this team allowed for the establishment of specific protocols and guidelines to provide comprehensive care for ECMO
ACCEPTED MANUSCRIPT patients. In an effort to evaluate the performance of our ECMO team, we compared ECMO patients cared for prior to the institution of the ECMO team to those managed with the new collaborative ECMO team approach. We have chosen to focus on survival to discharge between these two groups.
CR IP T
Methods: This study is a retrospective, single institution cohort study approved by the Partners Institutional Review Board (IRB). We identified all patients greater than 18 years of age who were placed on VA or VV ECMO support between 2009 and 2017 from our institutional
AN US
database. Manual chart review was performed on all patients to identify key demographic, clinical, and survival to discharge data. Following data collection, the data and patients were sorted into two groups: pre-MGH ECMO Team 2009-2013 and post-MGH ECMO Team
M
2014-2017. Additionally, subgroup analysis of patients on VA and VV ECMO was performed to evaluate utilization and survival changes.
ED
Statistics
PT
Data were collected and analyzed in Microsoft Excel spread sheets. Pearson‟s chi squared test was utilized to assess for differences in mortality between the pre- MGH ECMO team era
CE
(2009-2013) and the MGH ECMO team era (2014-2017). A p-value of less than 0.05 was
AC
selected as a cutoff to reject the null hypothesis (defined as no difference in survival between pre ECMO team and post ECMO team cohorts). Pearson‟s chi squared test was also used to assess for differences in the utilization of VV and VA ECMO between the two cohorts, again with a P value less than 0.05 selected as a cutoff to reject the null hypothesis (which was no difference in the utilization of VA and VV ECMO as a percent of total ECMO runs).
ACCEPTED MANUSCRIPT Two tailed T tests were used to assess for differences between ECMO duration. A P-value of less than 0.05 was used as the cutoff for a statistically significant difference between the groups. Depending on the results of the F-test, the T tests were either heteroscedastic (if the F-test showed a statistically significant difference in variance of the compared groups) or homoscedastic (if the F-test showed no statistically significant difference in variance between
CR IP T
the compared groups). Results:
From 2009-2017 a total of 279 patients were placed on ECMO at our institution;130 patients
AN US
between 2009-2013 and 149 patients between 2014-2017. Survival to discharge for patients between 2009-2013 was 37.7% (49/130) compared to survival to discharge of 52.3% (78/149) between 2014-2017 (p=0.02). When examining solely VA ECMO patients, survival
M
to discharge was 37.4% (37/99) from 2009-2013 and 48.6% (36/74) from 2014-2017. Furthermore, VV ECMO patients had a 40% (10/25) survival to discharge from 2009-2013,
ED
and a 63% (29/46) survival to discharge from 2014-2017 (P = 0.06 by chi-squared test). The mixed ECMO group only had 33 patients between 2009 and 2017 and thus was not analyzed
PT
for comparison. Duration on VA ECMO was 4.74 4.84 days in the pre-ECMO team era and
CE
6.5 10.2 days in the post-ECMO era. Of note, the variance is significantly different between the two groups of the patients (F-test: P < 0.001) but the duration of ECMO is not (P = 0.164
AC
by heteroscedastic T-test). For the V-V ECMO cohort, duration of cannulation between the pre-ECMO team era in this subgroup was 14.52 21.5 days and 14.54 28.6 days in the post-ECMO team era with neither the variance or duration achieving statistical significance (F-test: P = 0.136, homoscedastic T-test: P = 0.997). Prior to the creation of the MGH ECMO team, ICU placement of any patient cannulated for ECMO was dependent on the patients‟ original location prior to cannulation as
ACCEPTED MANUSCRIPT well as hospital bed resources. As a result, between 2009-2013, the 130 patients on ECMO were distributed between various units: 49 in the Cardiac Surgical Intensive Care Unit (CSICU), 4 in the Cardiac Care Unit (CCU), 9 in the Medical Intensive Care Unit (MICU) and 63 in the Surgical Intensive Care Unit (SICU). After implementation of the multidisciplinary MGH ECMO team, a protocolized approach was created for the placement
CR IP T
of these patients. Between 2014-2017, the distribution of ICU location for the 149 patients changed markedly: 94 in the CSICU, 0 in the CCU, 34 in the MICU, 1 in the SICU and 1 offunit in the cardiac catheterization laboratory (Table 1). Selected VV ECMO patients started receiving care in the medical intensive care unit as a result of an initiative to create more bed
AN US
space in the CSICU for patients requiring specialized surgical care. Prior to putting these patients in the MICU, multiple intensivists with VV ECMO training were recruited and trained the existing intensivist group. Additionally, bedside staff including nurses and
M
respiratory therapists were trained for the care of these patients.
ED
Discussion:
As the utilization of ECMO increases for patients with cardiogenic and respiratory failure,
PT
institutions are striving to improve outcomes, specifically survival to discharge in this
CE
complex patient population. Volume and outcomes at individual centers with respect to ECMO utilization have been reported in both VA and VV ECMO populations; it has been
AC
suggested that high-volume centers have lower mortality2,3. Teerapuncharoen et al examined the ECMO experience at The University of Alabama at Birmingham as their institution moved from a low to a high volume ECMO center4. Their work suggests that higher annual ECMO volume, and therefore more experience, improves survival to hospital discharge4. Through multivariable logistic regression, they identified that low annual volume, advanced patient age, and prolonged hospital length of stay are associated with reduced survival to hospital discharge.
ACCEPTED MANUSCRIPT Separately, Barbaro et. al utilized the Extracorporeal Life Support (ELSO) Registry to evaluate ECMO outcomes for the neonatal, pediatric, and adult populations3. This registry encompassed data from more than 56,000 patients across 290 centers. With respect to adults, from 2008 onwards, hospital volume played a significant role in adult mortality rates. Despite identifying and describing a relationship between center volume and patient outcomes, the
CR IP T
study did not attempt to elucidate the reason mortality was improved at these centers. What remains to be resolved is whether experience in it of itself is necessary to foster good outcomes or if a protocolized approach is what is needed for improved outcomes.
AN US
The use of multi-disciplinary teams in the inpatient hospital setting has been shown to limit adverse events, improve outcomes, and adds to patient and employee satisfaction5. Epstein has published widely on this subject, providing the conclusion that the best medical and surgical care is a “team sport”6. Epstein‟s group has demonstrated that cohesive
M
teamwork has improved communication between different levels of healthcare workers,
ED
allowing for improved patient outcomes and decreased length of stay. Despite the persuasive argument from Epstein and others with respect to the
PT
importance of multi-disciplinary care teams, very few studies have been published regarding
CE
institutional experience with a high volume ECMO center and a protocolized, team based approach4,7,8. There is a recent report by Cotza et. al describing their centers ECMO team
AC
approach at a tertiary care hospital in Italy9. Created in 2012, the team consists of cardiac surgeons, anesthesiologists, intensivists, nurses, and perfusionists. After one year of a multidisciplinary approach, retrospective analysis suggested improved survival to discharge, fewer bleeding complications, and fewer blood product transfusions9. Prior to the creation of the MGH ECMO Team, our institutional policy was to have an intensivist with ECMO experience manage the patients in a siloed approach. After initiation
ACCEPTED MANUSCRIPT of the MGH ECMO Team, both a multi-disciplinary care team and a set of protocols and guidelines for the institution of ECMO support were implemented (Figure 1, Appendix A-D). Initiation and implementation of the MGH ECMO team fostered a more collaborative approach to caring for these patients with improved outcomes. An impressive improvement in survival from 37.7% to 52.3% was noted; a 14.6% absolute improvement (NNT≈7).
CR IP T
Further analysis is needed to determine what other factors contributed to this substantial survival improvement.
Importantly, the number of VV ECMO patients drastically increased during the
AN US
analyzed period (25/130 from 2009-2013 vs 46/149 from 2014-2017, p=0.01). Additionally, the number of VA ECMO patients significantly decreased (99/130 from 2009-2013 vs 76/149 from 2014-2017, p <0.001). This trend may be multifactorial; however, it suggests a more prudent selection of therapeutic modalities and possibly better patient selection. This survival
M
improvement is also consistent with the ELSO registry findings in which patients on VV
ED
ECMO have a higher survival compared to patients on VA ECMO. Provider experience is also important, but throughout this analyzed period (2009-2017) there was significant staff
CE
experience.
PT
turnover reducing the likelihood that the improved survival is solely a result of provider
Based on our experience, there are several key components to our MGH ECMO Team
AC
and associated protocols which have contributed significantly to the improvement in our patient outcomes (Appendix A-D). One important component is the ICU location where patients on ECMO are cared for. Prior to the institution the MGH ECMO Team, patients supported with ECMO were cared for in multiple intensive care units throughout the hospital, including the Surgical Intensive Care Unit (SICU), Medical Intensive Care Unit (MICU), Cardiac Surgical Intensive Care Unit (CSICU), and the Cardiac Care Unit (CCU). We, as an institution, recognized that patients may benefit by care being performed in intensive care
ACCEPTED MANUSCRIPT units experienced in ECMO care. Thus, VA-ECMO patients are now exclusively cared for within the CSICU, and VV-ECMO patients are cared for within the MICU. Having dedicated intensive care units to care for ECMO patients has allowed us to focus on ECMO specific training for the nursing staff and care teams. Focusing these patients into the CSICU or MICU has streamlined care and improved experience and comfort in bedside providers caring
CR IP T
for these patients.
This study does have some notable limitations, including its single center
retrospective design. Particularly, given the fact the study was done over an 8-year period
AN US
advances in medicine that may be relevant to ECMO patients may have skewed the mortality in favor of the more recent cohort. Additionally, the statistical analysis performed in this study did not control for potential confounders that may have affected our results. Furthermore, data on blood transfusions, vasopressor use, or infection rates where not
M
analyzed. Despite these limitations, this is currently the largest single center review of ECMO
AC
CE
PT
ED
outcomes with respect to the effect of instituting a team based approach on mortality.
ACCEPTED MANUSCRIPT Conclusion: In summary, the implementation of a multi-disciplinary team-based approach to ECMO care has played an integral role in the improvement in survival to discharge of ECMO patients at our institution. This manuscript may serve as a framework to assist other
CR IP T
institutions interested in starting or searching for possible means to improve an ECMO
AC
CE
PT
ED
M
AN US
program.
ACCEPTED MANUSCRIPT References: 1. Extracorporeal Life Support Organization. Registry. https://www.elso.org/Registry/Statistics.aspx. Accessed on February 28,2018 2. Mccarthy FH, Mcdermott KM, Spragan D, et al. Unconventional Volume-Outcome Associations in Adult Extracorporeal Membrane Oxygenation in the United States. Ann
CR IP T
Thorac Surg. 2016;102(2):489-95.
3. Barbaro RP, Odetola FO, Kidwell KM, et al. Association of hospital-level volume of
extracorporeal membrane oxygenation cases and mortality. Analysis of the extracorporeal life support organization registry. Am J Respir Crit Care Med. 2015;191(8):894-901.
AN US
4. Tchantchaleishvili V, Hallinan W, Massey HT. Call for Organized Statewide Networks for Management of Acute Myocardial Infarction-Related Cardiogenic Shock. JAMA Surg. 2015;150(11):1025-6.
5. Moll V, Teo EY, Grenda DS, et al. Rapid Development and Implementation of an ECMO
M
Program. ASAIO J. 2016;62(3):354-8.
6. Epstein NE. Multidisciplinary in-hospital teams improve patient outcomes: A review. Surg
ED
Neurol Int. 2014;5(Suppl 7):S295-303.
PT
7. Na SJ, Chung CR, Choi HJ, et al. The effect of multidisciplinary extracorporeal membrane oxygenation team on clinical outcomes in patients with severe acute respiratory failure. Ann
CE
Intensive Care. 2018;8(1):31. 8. Fiedler AG, Song TH, D'alessandro DA. Redesigning Care for Patients With Acute
AC
Myocardial Infarction Complicated by Cardiogenic Shock: The "Shock Team". JAMA Surg. 2016;151(7):685.
9. Cotza M, Carboni G, Ballotta A, et al. Modern ECMO: why an ECMO programme in a tertiary care hospital. Eur Heart J Suppl. 2016;18(Suppl E):E79-E85.
ACCEPTED MANUSCRIPT Figure Legend:
AC
CE
PT
ED
M
AN US
CR IP T
Figure 1. Institutional ECMO initiation algorithm
CR IP T
ACCEPTED MANUSCRIPT
AN US
Figure Appendix A. Nursing ECMO protocols Appendix A: Nursing ECMO Protocols Education:
ICU Nurse teach annual ECMO class for all nursing staff caring for ECMO patients.
-
Any new CCRN/NP/PA to the CSICU must take the introductory ECMO class.
-
All nursing staff must take a retraining ECMO class each year.
-
ECMO training classes are offered on a bi-monthly basis.
ED
M
-
Patient Care:
-
Daily sedation “holidays”
AC
CE
PT
-
All new ECMO patients have a cooler of 4 PRBCs and 4 FFP at their bedside for the first 24 hours‟ post cannulation. No PIVs are to be placed or removed by nursing staff while on ECMO due to high risk for compartment syndrome and need for fasciotomy. Blood cultures are to be done by mid-level staff or higher. Assess and document pupils q2hrs
-
CR IP T
ACCEPTED MANUSCRIPT
-
Sterile dressing changes daily
-
Emergency‟s require MD notification: bleeding, air in the circuit, accidental decannulation, lethal Arrhythmias/CPR
-
Transfusion Thresholds:
AN US
o Hgb > 8.5 o If SVO2 is low, keep Hgb > 10.0 o Plts > 20,000 o If bleeding, keep > 80, 000 o INR < 2.5
M
o If bleeding, keep < 1.6 o Fibrinogen > 150 Anticoagulation goals:
ED
-
o Heparin Bolus (typically 5,000-10,000 units IVP) on initiation of ECMO or weaning/clamping ECMO Circuit
CE
PT
o Heparin Infusion starting at 10units/kg/hr titrating to goal PTT 60-80 (if bleeding, 40-60)
AC
Figure Appendix B-D. Electronic medical record ECMO order sets
AC
CE
PT
ED
M
AN US
CR IP T
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
M
AN US
CR IP T
ACCEPTED MANUSCRIPT
Table 1. ECMO Patient Data Analysis
CR IP T
ACCEPTED MANUSCRIPT
2009 – 2013 (n=130)
2014-2017 (n=149)
p-value
Patient age (years)
50.96 ± 14.9
52.8 ± 14.5
48.8 ± 15.1
0.03
ECMO duration (days)
9.3 ± 17.8
6.9 ± 11.6
11.43 ± 21.15
0.025
VA
175 (62.7)
99 (76.2)
76 (51)
<0.001
VV
71(25.4)
25 (19.2)
46 (30.8)
0.01
Mixed
33 (12.45)
AN US
Total (n=279)
ED
Unit
M
Mode of ECMO
6 (4.6)
27 (18.1)
154 (55.2)
49 (37.7)
105 (70.5)
CCU
4 (1.4)
4 (3.1)
0 (0.0)
MICU
46 (16.5)
9 (6.9)
37 (24.8)
64 (22.9)
63 (48.5)
1 (0.7)
4 (1.4)
3 (2.3)
1 (0.7)
PT
CSICU
AC
Cath lab
CE
SICU
<0.001
CR IP T
ACCEPTED MANUSCRIPT
7 (2.5)
2 (1.5)
5 (3.4)
Survived to discharge
127 (45.5)
49 (37.7)
78 (52.3)
AC
CE
PT
ED
M
AN US
Multiple units
0.021
Table 2: ECMO Indications for Cannulation
IPF: 7
Flu/PNA: 3
Pulmonary Embolism: 3
VA-ECMO ‘09-13:
MI: 25
Post-cardiotomy: 12
Pulmonary Embolism: 13
ARDS: 12
Flu/PNA: 9
IPF: 4
MI: 13
Postcardiotomy: 7
Cardiac arrest: 6
Unspecified: 21 Pulmonary Embolism: 5
AC
CE
PT
ED
M
VV-ECMO ‘14-17: VA-ECMO ‘14-17:
Other/unspecified: 12 Myocarditis: 5 Other/unspecified: 44
AN US
VV-ECMO ‘09-13:
CR IP T
ACCEPTED MANUSCRIPT
Cardiomyopathy: Myocarditis: VT storm: 4 4 3
Other/unspecified: 32