Extracorporeal membrane oxygenation as a rescue measure for postcardiotomy cardiogenic shock: A retrospective observational analysis

Abstract

Background

Nearly 500,000 open-heart operations are performed annually in the United States, with complications such as postcardiotomy cardiogenic shock (PCCS) occurring in 2–9% of cases. Extracorporeal membrane oxygenation (ECMO) is a critical salvage therapy for patients unable to wean from cardiopulmonary bypass (CPB). This study aims to review our institution's experience with ECMO in managing PCCS and to analyze patient outcomes.

Methods

Following IRB approval, a retrospective observational study was conducted on adult patients aged 32–84 years who underwent open-heart procedures with the use of CPB requiring ECMO support due to failed wean from CPB from 1 April 2014 to 31 December 2022. Data were analyzed from electronic medical records for demographics, procedural details, ECMO therapy duration, and outcomes.

Results

Of 45 patients identified from 6346 open-heart procedures, 33 were male, and 12 were female, with an average age of 59.9 years. The majority of patients were Caucasian (88.8%, n = 40). Extracorporeal membrane oxygenation was initiated using venoarterial configurations in 100% (n = 45) of cases, with three patients transitioning to venovenous configurations. The median ECMO duration for all patients in the study was 4 days. In-hospital mortality was 51.1% (n = 23), while 48.9% (n = 22) of patients survived to discharge. Survivors were discharged to rehabilitation facilities (54.5%, n = 12), home (31.8%, n = 7), long-term acute care hospitals (9.1%, n = 2), or detention centers (4.5%, n = 1).

Conclusions

Extracorporeal membrane oxygenation remains a valuable rescue therapy for PCCS, achieving a 48.9% survival rate. This study highlights the importance of timely intervention and underscores the need for future research into optimizing patient selection and perioperative management.

Keywords

Cardiogenic shock extracorporeal membrane oxygenation adult salvage therapy heart–lung machine retrospective

Introduction

Postcardiotomy cardiogenic shock (PCCS) is a rare but severe complication of cardiac surgery, with an incidence of 2–9% among patients undergoing open-heart procedures.^1–3 This condition, characterized by the inability of the heart to maintain adequate cardiac output following separation from cardiopulmonary bypass (CPB), is associated with exceedingly high mortality rates if untreated.⁴

Extracorporeal membrane oxygenation (ECMO) has emerged as a pivotal therapy for PCCS. While traditional mechanical circulatory support (MCS) devices, such as intra-aortic balloon pumps (IABPs), Impella, and TandemHeart, play important roles in stabilizing hemodynamics, their capabilities are often insufficient for cases of refractory shock.⁵ ECMO therapy provides full cardiopulmonary support, bridging patients to myocardial recovery or definitive interventions such as heart transplantation.⁶

Despite its potential, outcomes for the use of ECMO in treating PCCS remain variable. Previous studies have reported survival rates ranging from 30% to 50%, influenced by patient characteristics, timing of intervention, and institutional expertise.⁷ This study aims to address the knowledge gap by reviewing our institution's experience with ECMO in managing PCCS, focusing on patient outcomes and procedural variables.

Methods

This study received ethical approval from our institution's IRB (approval #92795) on 22 January 2024. All patient information was de-identified, and patient consent was not required. Patient data will not be shared with third parties. The study was designed to provide a comprehensive overview of ECMO's impact on patient outcomes within the context of PCCS. The study included adult patients aged 32–84 years who underwent open-heart procedures, defined as a sternotomy with implementation of CPB, requiring ECMO support due to failed wean from CPB between 1 April 2014 and 31 December 2022. Patients were identified through an electronic medical record (EMR) search using Current Procedural Terminology codes. Relevant data were extracted from patients’ EMRs, including demographic details, surgical procedures, ECMO initiation parameters, and clinical outcomes. Patients requiring ECMO initiation in the intensive care unit (ICU) after successful CPB wean were not included in this study, nor were those who used other forms of mechanical circulatory devices. The primary outcome was survival to discharge, while secondary outcomes included the duration of ECMO therapy, discharge disposition, and associated complications. The demographic and procedural characteristics of the cohort are presented in Tables 1 and 2. Extracorporeal membrane oxygenation configurations and associated outcomes are detailed in Tables 3 to 5.

Table 1.

Patient demographics and characteristics.

Characteristic	Number in total population (N = 45)
Sex
Males	33
Females	12
Race
Caucasian	40
Black or African American	4
Asian American	1
Average age (years) ± SD	59.9 (32–84) ± 11.8
Obesity (BMI >30 kg/m²)	38
Chronic kidney disease	11
Heart failure	14
Diabetes mellitus	13
Redo sternotomy	15
Endocarditis	12

Table 2.

Initial operation performed.

Operative procedure	Frequency
Heart transplant	17
CABG	11
Valve replacement	10
CABG with valve replacement	1
Aortic root replacement	5
HVAD placement	1

CABG = coronary artery bypass graft; HVAD = HeartWare ventricular assist device.

Table 3.

Extracorporeal membrane oxygenation therapy placement and outcomes.

ECMO configuration	Frequency
Venoarterial (V-A)
V_f–A_AO	37
V_f–A_f	8
Transition from V_f–A_AO to V_f–A_f	2
Venovenous (V-V)
Transition from V_f–A_AO to V_f–V_f	3
Population	Time (days) on ECMO as median (IQR Q₁–Q₃)
Total study participants (45)	4 (3–6)
Survivors (22)	4.5 (3.25–5.75)
Nonsurvivors (23)	4 (2–7.5)

V_f – A_AO = femoro-aortic V-A ECMO; V_f – A_f = femoro-femoral V-A ECMO; V_f – V_f = femoro-femoral V-V ECMO.

Table 4.

Outcomes on mortality and discharge.

Characteristic	Frequency
Total deaths	23
Alive at discharge	22
Discharge deposition
Rehabilitation facility	12
Home	7
Long-term acute care hospital	2
Detention center	1

Table 5.

Postoperative complications.

Characteristic	Frequency
Bleeding	18
Acute kidney injury/failure	15
Respiratory failure	7
DVT or PE	5
Ischemic stroke	4
Cardiac arrest	3

Definition of PCCS

Postcardiotomy cardiogenic shock was defined as persistent hypotension and low cardiac output syndrome refractory to volume resuscitation and inotropic support following separation from CPB, leading to the inability to wean from bypass requiring emergent mechanical support.

Criteria for ECMO initiation

Patients were considered for ECMO if they met one or more of the following: (1) failure to wean from CPB despite maximal medical and mechanical support; (2) hemodynamic instability with systolic blood pressure <80 mmHg and/or cardiac index <2.0 L/min/m²; or (3) evidence of intraoperative ventricular dysfunction guided by real-time transesophageal echocardiogram and Swan-Ganz catheter measurements. Ultimately, the decision to initiate ECMO was made by the attending cardiac surgeon in conjunction with the perfusion team.

Timing and location of ECMO initiation

All patients (100%) selected for the study were placed on ECMO intraoperatively due to the inability to separate from CPB.

Criteria for ECMO weaning

At our institution, the protocol starts with ensuring adequate hemodynamic stability and pulmonary function. Hemodynamic stability is characterized as a CVP <16 mmHg, MAP >60 mmHg, pulse pressure >20 mmHg while on no or low-dose vasopressor/inotropic support. Pulmonary function is deemed appropriate if the pCO2 < 45 on a sweep of 2 liters per minute or less with a tidal volume of at least 5 cc/kg. Once these physiologic parameters are met, the ECMO flows are slowly weaned to 2 liters per minute. If the MAP > 60, ejection fraction (EF) > 20%, and no evidence of worsening LV distension on ECHO, then a recirculation bridge is built in and circulated. If the MAP > 60, EF > 20%, and no evidence of worsening LV distension on ECHO while using the recirculation bridge, then the patient is decannulated in the operating room.

Results

Cohort demographics

The study cohort included 45 patients meeting the inclusion criteria out of a total of 6346 open-heart procedures performed during the study period. Of these, 33 patients (73.3%) were male, and 12 (26.7%) were female, with a mean age of 60 years (32–84). Most patients (88.9%, n = 40) were Caucasian, with smaller proportions identifying as Black or African American (8.9%, n = 4) or Asian American (2.2%, n = 1) (Table 1).

Extracorporeal membrane oxygenation configuration, duration, and outcomes

The surgical procedures leading to ECMO support varied. Heart transplant was the most common procedure (37.7%, n = 17), followed by coronary artery bypass graft (CABG) (24.4%, n = 11), aortic or mitral valve replacement (22.2%, n = 10), aortic root replacement (11.1%, n = 5), CABG with valve replacement (2.2%, n = 1), and HeartWare ventricular assist device placement (2.2%, n = 1) (Table 2).

Extracorporeal membrane oxygenation configurations were collected and reported according to the Extracorporeal Life Support Organization Maastricht Treaty for nomenclature in extracorporeal life support to ensure accurate and concise communication. All patients (100%, n = 45) were initiated on venoarterial ECMO, with the primary route of cannulation being V_f – A_AO (82.2%, n = 37) followed by V_f – A_f (17.7%, n = 8) Two patients transitioned from V_f – A_AO to V_f – A_f during therapy, while three others transitioned from V_f – A_AO to V_f – V_f (Table 3). Stratified analysis revealed that survivors had a slightly longer median ECMO duration 4.5 (3.25–5.75) days, compared to the overall median, while nonsurvivors had a median ECMO duration of 4 (2–7.5) days (Table 3). To monitor for signs of LV distention in this cohort, all patients left the operating room with evidence of an opening aortic valve and then underwent echocardiograms for worsening pulse pressure. No patients required LV venting. Furthermore, pulmonary edema related to the VA ECMO circuit was not measured; however, all of the patients had evidence of vascular congestion on daily CXR.

Survival outcomes

A total of 22 patients (48.9%) survived to discharge, while 23 (51.1%) died during their hospital stay. Survivors were discharged to rehabilitation facilities (54.5%, n = 12), home (31.8%, n = 7), long-term acute care hospitals (9.1%, n = 2), or detention centers (4.5%, n = 1) (Table 4). Discharge destinations were influenced by functional status at decannulation, duration of mechanical ventilation, and social support availability.

Complications and causes of mortality

Postoperative complications were common and included bleeding (n = 18), renal failure or acute kidney injury (n = 15), respiratory failure (n = 7), thromboembolic events (e.g., deep venous thrombosis or pulmonary embolus, n = 5), stroke (n = 4), and cardiac arrest (n = 3) (Table 5). Several patients experienced multiple complications concurrently, including multi-organ failure and infectious sequelae such as sepsis or pneumonia. The most common cause of death in the 23 patients who died during hospital stay was cardiogenic shock (n = 13) followed by multi-organ failure (n = 5), respiratory failure (n = 1), hemorrhagic shock (n = 1), primary graft dysfunction (PGD) (n = 1), brain death (n = 1), and vasoplegic shock (n = 1) (Table 6).

Table 6.

Causes of death.

Characteristic	Frequency
Cardiogenic shock	13
Multi-organ failure	5
Respiratory failure	1
Hemorrhagic shock	1
Primary graft dysfunction	1
Brain death	1
Vasoplegic shock	1

Discussion

Postcardiotomy cardiogenic shock remains one of the most challenging complications in adult cardiac surgery, often resulting in rapid hemodynamic deterioration. Postcardiotomy cardiogenic shock is rare with a reported incidence of 2–9%.^1–3 This condition requires prompt intervention to prevent rapid hemodynamic decline and multi-organ failure.⁴ While traditional MCS devices such as IABPs, Impella, and TandemHeart are commonly employed, they often fall short in cases of severe circulatory collapse, necessitating the use of ECMO.⁵ Extracorporeal membrane oxygenation therapy provides full cardiopulmonary support, facilitating myocardial recovery and stabilizing end-organ perfusion.⁶

Our institution's experience demonstrates the utility of ECMO in managing PCCS, with a survival-to-discharge rate of 48.9%. This finding aligns with prior studies reporting survival rates between 30% and 50%. Rastan et al. documented a 31% survival rate in a similar cohort, while Rao et al. emphasized the importance of timely ECMO initiation in improving outcomes.^6,7 Notably, our results also reflect those of Pérez et al., who highlighted the role of multidisciplinary ECMO teams in achieving comparable survival rates.⁵ Conversely, our outcomes surpass the 30% survival rates reported by Whitman, potentially reflecting differences in institutional protocols and patient management strategies.⁸

Despite these encouraging results, the in-hospital mortality rate of 51.1% underscores the severity of PCCS and the limitations of current therapeutic approaches. The observed mortality likely reflects the critical condition of patients requiring ECMO, many of whom present with refractory shock and significant comorbidities. Moreover, the resource-intensive nature of ECMO therapy necessitates careful patient selection to optimize outcomes.

Interestingly, our data reveal a diverse range of discharge destinations among survivors, with 54.5% transitioning to rehabilitation facilities and 31.8% returning home. This highlights the significant recovery potential among ECMO-supported patients, though it also stresses the need for robust postdischarge care strategies to facilitate long-term recovery.

We also observed that survivors tended to have longer ECMO duration than nonsurvivors, though this was not statistically analyzed due to the sample size. The median ECMO duration for survivors was 4.5 (3.25–5.75) days, compared to 4 (2–7.5) days for nonsurvivors. While this may suggest that prolonged support allows for myocardial recovery in selected patients, it may also reflect survivorship bias or more aggressive supportive strategies in select patients.

Of the patients included in this study, the most common procedure was heart transplantation. With an incidence of 20–40%, PGD has a significant impact on overall mortality when compared to patients without PGD. Although various factors such as donor/recipient age, ischemic time, and recipient preoperative MCS contribute to increased PGD risk, management is focused on optimizing hemodynamic support via inotropes, IABP, or VA ECMO while the PGD is treated. With these management strategies, mortality for those with PGD has improved.^9,10

Another consideration from this cohort is that this study population did not include those requiring ECMO initiation for shock after transfer to the ICU from the operating room. Furthermore, patients requiring IABP or Impella support postoperatively were also not included in this study. As a result, our 0.7% incidence is lower than that observed in other publications. We believe if those patients were added, our institution's PCCS incidence would align with previously published rates.

Our study raises several areas for future research, including earlier initiation of ECMO in high-risk patients. Mariani et al. identified a statistically significant survival difference favoring postcardiotomy patients treated with intraoperative VA ECMO versus postoperative VA ECMO in the ICU.¹¹ Additionally, recent studies utilizing Impella 5.5 implantation during low EF CABG procedures to assist with postoperative recovery have shown promising results, further highlighting the benefit of planned intraoperative MCS support in high-risk patient populations.^12,13 Moreover, developing risk stratification tools and specific biomarker testing to identify these high-risk patient cohorts would help guide decision-making for MCS utilization.

Our findings prompt consideration of combined MCS strategies. Though not employed in this cohort, devices such as IABPs or Impella in conjunction with ECMO may help unload the left ventricle, reduce afterload, and mitigate complications such as pulmonary edema or ventricular distension. Further studies exploring the integration of these therapies in PCCS are warranted.

This study adds to the growing body of evidence supporting ECMO as a lifesaving therapy for PCCS. However, it also highlights the need for continued innovation and multidisciplinary collaboration to refine patient selection, optimize timing, and enhance overall outcomes. By addressing these gaps, future investigations can pave the way for more effective management strategies and improved survival for this critically ill population.

Limitations

This study has several limitations. First, its retrospective design introduces inherent biases related to data collection and interpretation. Second, the single-center nature of the study limits the generalizability of the findings to other institutions. Third, the absence of a control group prevents direct comparison of ECMO outcomes with alternative therapies. Finally, variability in patient characteristics, comorbidities, and ECMO initiation may influence outcomes, warranting cautious interpretation of the results.

Conclusion

Extracorporeal membrane oxygenation remains a valuable rescue therapy for PCCS, achieving a 48.9% survival rate in our cohort. This study reinforces ECMO's role in high-risk cardiac surgical populations while emphasizing the need for further research into optimizing patient selection and perioperative management. By addressing the challenges and unanswered questions in PCCS management, future investigations can pave the way for improved outcomes in this vulnerable population.

Footnotes

Acknowledgments

The Professional Student Mentored Research Fellowship (PSMRF) Project is supported by the National Center for Advancing Translational Sciences through Grant UL1TR001998, UK HealthCare, and the University of Kentucky College of Medicine. During the preparation of this work, the authors used OpenAI in order to improve language, readability, and to ensure proper formatting of tables and individual sections for the article. After using this tool/service, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.

ORCID iDs

Keenan Conley

Daniel Mohammadi

Sibu Saha

Ethical considerations

This study received ethical approval from the University of Kentucky Office of Research Integrity IRB (approval #92795) on 22 January 2024. All patient information was de-identified, and patient consent was not required. Patient data will not be shared with third parties.

Author contributions

KC contributed to study design, led data collection and analysis, and drafted the manuscript. SS supervised the project and contributed to the study design and critical manuscript revision. DM contributed additional analysis and revisions to the manuscript during the peer-review process. All authors contributed to the interpretation of the data, critical revision of the manuscript for important intellectual content, and approved the final version.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Keenan Conley received a stipend from the Professional Student Mentored Research Fellowship (PSMRF) Project supported by the National Center for Advancing Translational Sciences through Grant UL1TR001998, UK HealthCare and the University of Kentucky College of Medicine. This stipend is for medical students and was distributed over a 10-month period. There were no other sources of funding for this project.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

References

Masud

Gheewala

Giesecke

, et al. Cardiogenic shock in perioperative and intraoperative settings: a team approach. Methodist Debakey Cardiovasc J 2020; 16: e1–e7.

Cheng

Wang

Zhang

, et al. Risk factors for postoperative myocardial injury-related cardiogenic shock in patients undergoing cardiac surgery. J Cardiothorac Surg 2023; 18: 220.

Stein

Silvestry

. Algorithmic management of postcardiotomy shock with mechanical support: bring a map, a plan, and your parachute—and know how to use all three. JTCVS Open 2021; 8: 55–65.

Khorsandi

Shaikhrezai

Prasad

, et al. Advanced mechanical circulatory support for post-cardiotomy cardiogenic shock: a 20-year outcome analysis in a non-transplant unit. J Cardiothorac Surg 2016; 11: 29.

Pérez Vela

Llanos Jorge

Duerto Álvarez

, et al. Clinical management of postcardiotomy shock in adults. Med Intensiva 2022; 46: 312–325.

Rastan

Dege

Mohr

, 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: 302–11, 311.e1.

Rao

Khalpey

Smith

, et al. Venoarterial extracorporeal membrane oxygenation for cardiogenic shock and cardiac arrest. Circ Heart Fail 2018; 11: e004905.

Whitman

GJR

. Extracorporeal membrane oxygenation for the treatment of postcardiotomy shock. J Thorac TiCardiovasc Surg 2017; 153: 95–101.

Singh

SSA

Dalzell

Berry

, et al. Primary graft dysfunction after heart transplantation: a thorn amongst the roses. Heart Fail Rev 2019; 24: 805–820.

10.

Nicoara

Ruffin

Cooter

, et al. Primary graft dysfunction after heart transplantation: incidence, trends, and associated risk factors. Am J Transplant. 2018; 18: 1461–1470.

11.

PELS-1 (PELS-1, Post-Cardiotomy Extracorporeal Life Support Study) Investigators, et al. The importance of timing in postcardiotomy venoarterial extracorporeal membrane oxygenation: a descriptive multicenter observational study. J Thorac Cardiovasc Surg 2023; 166: 1670–1682.e33.

12.

Cabrucci

Baudo

Yamashita

, et al. Beating heart coronary artery bypass grafting with preemptive Impella^™ 5.5 assist device in ischemic cardiomyopathy. Biomedicines 2025; 13: 1259.

13.

Benke

Korça

Boltjes

, et al. Preventive Impella® support in high-risk patients undergoing cardiac surgery. J Clin Med 2022; 11: 5404.