Predictors of Clinical Outcome After Early Veno-Arterial Extracorporeal Membrane Oxygenation in Cardiogenic Shock Complicating ST-Elevation Myocardial Infarction

Abstract

Objectives. Despite increasing use of veno-arterial extracorporeal membrane oxygenation (VA-ECMO) in patients with cardiogenic shock (CS) secondary to ST-segment elevation myocardial infarction (STEMI), a paucity of adequate evidence for this therapy remains. The aim of this single-center clinical registry study was to identify predictors of survival and discern the possible optimal time to initiate VA-ECMO in this cohort. Methods and Results. Seventy-nine consecutive patients with CS complicating STEMI who received VA-ECMO support were included in this analysis. The primary endpoint was survival at 6 months after initiation of VA-ECMO. Mean age was 60 ± 11 years. Forty-six patients (58%) were successfully weaned from VA-ECMO and 30 patients (38%) could be discharged. Of these, 23 patients (29% of the overall population) survived up to 6-month follow-up. Multivariate analysis to identify determinants of survival showed no association between the time of CS onset to VA-ECMO start time and 6-month survival (P=.75). Glomerular filtration rate on admission (P<.001), white blood cell count on admission (P≤.01), age (P≤.01), and arterial lactate level 1 and 24 hours after VA-ECMO initiation (P=.01) were the strongest predictors of survival. Conclusions. The timing of VA-ECMO initiation in patients with CS complicating STEMI was not a prognostic factor of survival. Renal function, white blood cell count, age, and lactate level were the strongest predictors of death during 6-month follow-up.

J INVASIVE CARDIOL 2021;33(5):E329-E335. doi:10.25270/jic/20.00542

Key words: ECMO, extracorporeal life support organization, refractory cardiogenic shock, STEMI

The use of veno-arterial extracorporeal membrane oxygenation (VA-ECMO) in patients with cardiogenic shock (CS) complicating acute ST-segment elevation myocardial infarction (STEMI) has increased over recent years.¹ The significant economic burden and lack of prospective randomized trials investigating VA-ECMO use in patients with CS did not reverse this trend.² Different scores have been established to help predict the outcome of CS in patients undergoing VA-ECMO.^3,4 However, indications for VA-ECMO in this patient cohort are not clearly defined and the level of evidence remains low (class IIb, level of evidence C).⁵ Furthermore, optimal timing for initiating VA-ECMO in patients with acute myocardial infarction and CS remains undetermined.

Several studies previously focused on the outcome of VA-ECMO treatment in patients with STEMI-associated CS. While these studies focused exclusively on patients with the need for acute VA-ECMO implantation due to severe refractory CS despite percutaneous coronary intervention (PCI), intra-aortic balloon pump (IABP), and medical support, our study included all consecutive patients who developed profound CS after STEMI (on admission or during hospital stay). The purpose of the current study was to identify predictors of survival and to discern a possible optimal time to initiate VA-ECMO in patients with STEMI and CS.

Methods

Patients. Between July 2005 and September 2017, all adult patients who received VA-ECMO support due to CS complicating STEMI were prospectively entered into a general hospital database at a single high-volume tertiary-care center. Additional data were evaluated retrospectively.

Definitions. STEMI was defined according to previous and current definitions.⁶CS was clinically defined as hypotension in the presence of adequate intravascular volume or the necessity of inotropic or vasopressor support.⁷ The definition includes laboratory (elevated arterial lactate >2 mmol/L) and clinical signs of hypoperfusion (altered mental status, cold skin, oliguria). Clinically relevant bleeding considered Blood Academic Research Consortium bleeding criteria type 2 and 3a-c.⁸

VA-ECMO indications. All included patients underwent VA-ECMO due to refractory CS unresponsive to conventional volume and pharmacological therapy, where the chance of survival was considered minimal without mechanical support. VA-ECMO was not considered for patients with terminal malignancy (if previously known), relevant and uncontrollable bleeding, irreversible brain damage after refractory cardiac arrest, and prolonged cardiopulmonary resuscitation (CPR). All patients with STEMI and CS undergoing VA-ECMO as bridge to rescue bypass surgery or permanent left ventricular assist device (LVAD) were also excluded. The decision for VA-ECMO treatment was made by cardiologists experienced in intensive care medicine and interventional cardiology. VA-ECMO cannulae were inserted percutaneously in the catheterization laboratory and start of treatment was performed under the assistance of a perfusionist. All patients who could be weaned from VA-ECMO underwent percutaneous decannulation to minimize the risk of vascular complications. PCI was performed according to current clinical practice and guideline recommendations prior to or at the time of VA-ECMO initiation.

Intensive care treatment. All patients were admitted to an intensive care unit (ICU), where clinical, functional, and laboratory parameters were repeatedly assessed according to standard of care. The use of antibiotics, additional medication, or therapeutic measures such as renal replacement therapy were performed according to standard clinical practice and guideline recommendations. Lung protective ventilation was favored. Inotropic and vasopressor agents were administered additionally if necessary, to achieve a mean arterial blood pressure (MAP) >60 mm Hg. Transfusions of blood components were restricted to the presence of clinically relevant bleeding and significant anemia. The short-term therapeutic goal was to reach a stable cardiopulmonary condition to allow weaning from VA-ECMO, which was performed according to an established in-hospital protocol.

Outcome definitions. The primary outcome was defined as all-cause mortality during a 6-month follow-up period. Main secondary endpoints were successful weaning from VA-ECMO and survival to hospital discharge as well as 30-day all-cause mortality.

Statistical analysis. All categorical variables are expressed as numbers with percentages. All continuous variables are presented as mean ± standard deviation or as median with interquartile range (IQR). Normality tests were used to determine the distribution of data. Normally distributed variables were tested by Student’s t-test or one-way analysis of variance. Mann-Whitney U-test or Kruskal-Wallis test were performed for non-normally distributed variables. Categorical variables were compared by Fisher’s exact test or Chi-square test, when appropriate. Overall 6-month time to survival was calculated and plotted using the Kaplan-Meier method. Cox proportional hazards model was applied to identify predictors of 6-month mortality. A probability value of <.05 was considered statistically significant. Statistical analysis was performed using SPSS statistics software, version 24 (IBM).

Results

Baseline characteristics. In total, 79 patients (65 males) were included in this observational study. The follow-up was 6 months. The baseline characteristics of the study population based on 6-month survival are presented in Table 1. Survivors were significantly younger (mean age, 55 ± 9 years vs 63 ± 12 years; P≤.01) and had lower body mass index (26 ± 3 kg/m² vs 29±5 kg/m²; P=.01). Other relevant differences between 6-month survivors and non-survivors included kidney function (creatinine and estimated glomerular filtration rate [eGFR]), white blood cell count, activated partial thromboplastin time, and lactic acid. MAP prior to VA-ECMO initiation was <70 mm Hg under continuous intravenous infusion of inotropic and/or vasopressor agents. There were no significant differences in established cardiovascular risk factors among the 2 patient groups. Table 2 presents essential laboratory tests of patients obtained on admission.

Procedure and complications. Table 3 shows VA-ECMO-relevant technical parameters, duration, and related complications. The duration of VA-ECMO support was 6 ± 3 days in 6-month survivors and 5 ± 4 days in non-survivors. The median duration from symptom onset to PCI was comparable in both groups (4.8 hours [IQR, 2.9-13.5 hours] in survivors vs 4.75 hours [IQR, 3.0-14.5 hours] in non-survivors). The median duration from diagnosis of CS to VA-ECMO initiation was 8.4 hours (IQR, 2.1-26.5 hours) in survivors vs 7.6 hours (IQR, 3.1-18.7 hours) in non-survivors (Figure 1). The median duration from PCI to VA-ECMO initiation was 11.9 hours (IQR, 0.9-25.8 hours) in survivors and 5.7 hours (IQR, 2.1-26.5 hours) in non-survivors. Three patients (4%) underwent ECMO insertion before revascularization and 76 patients (96%) underwent ECMO insertion after revascularization. Forty patients (51%) received VA-ECMO within the initial procedure. The remaining 39 patients (49%) have not undergone immediate VA-ECMO implantation. Twelve patients (31%) were not in CS on admission and 27 patients (69%) were stabilized after PCI and believed to recover under supportive treatment. Later, all of these patients developed profound refractory CS with the need for VA-ECMO treatment. There were no significant differences in the incidence of cerebral ischemia, clinically relevant bleeding, acute kidney injury requiring renal replacement therapy, acute limb ischemia, and compartment syndrome between the cohorts.

Clinical outcomes. Outcome data were available for all patients. The median duration of hospitalization was 12 days (IQR, 4-30 days) for all patients. Thirty eight percent of all patients (n = 30) survived to discharge and 46 patients (58% of the total cohort) could be primarily weaned from VA-ECMO. The 30-day mortality was 62% (n = 49), the 6-month mortality was 71% (n = 56). Almost half of the 6-month non-survivors died within 2 weeks after CS onset (median, 13 days; IQR, 3.5-180 days). Cumulative time to survival is shown in Figure 2.

Predictors of clinical outcome. The parameters significantly associated with 6-month mortality on univariate analysis were MAP (P=.03), creatinine level (P=.02), eGFR (P<.001), arterial lactate at 1 and 24 hours after initiation of VA-ECMO (P=.01 for both), age (P≤.01), and white blood cell count on admission (P≤.01). Multiple stepwise logistic regression analysis revealed that age >60 years (hazard ratio [HR], 1.03; 95% confidence interval [CI], 1.00-1.06; P=.03) along with eGFR <52 mL/min/1.73m² (HR, 0.98; 95% CI, 0.97-1.00; P=.03) and arterial lactate level >7.5 mmol/L prior to VA-ECMO initiation (HR, 1.08; 95% CI, 1.02-1.15; P=.01) were significant predictors of 6-month mortality. Lactate level as a continuous variable before VA-ECMO initiation did not reach statistical significance as a predictor (P=.051). Additionally, the analysis showed no association between the time of CS onset to VA-ECMO start time as well as time between PCI and VA-ECMO implantation and 6-month survival (P=.75 and P=.91, respectively).

Discussion

Mechanical circulatory support is considered a possible life-saving therapy for patients with refractory CS. However, data from large prospective and randomized studies are lacking. The present analysis is one of the few specifically focusing on a homogeneous cohort of STEMI patients in profound CS receiving VA-ECMO.^9,10 All previous studies included only patients with severe refractory CS on admission resulting in prompt VA-ECMO support. In comparison, this is an all-comers study of patients with profound CS after STEMI. We included both patients in refractory CS at initial presentation with the need for immediate VA-ECMO support and patients with later onset of refractory CS after PCI at the ICU. In addition, the current analysis focused on the time of CS onset to VA-ECMO initiation and did not find an influence of symptom and CS onset and VA-ECMO initiation on survival.

Current guidelines recommend that VA-ECMO may be considered in patients with STEMI and CS (recommendation class IIb, level of evidence C) but the optimal timing for ECMO therapy is not well defined.^11,12 A recent trial showed an improved outcome in patients with STEMI and CS who received VA-ECMO before PCI.¹³ The majority of patients in our study received mechanical circulatory support after revascularization (96%). In addition, nearly half of our patients (49%) received delayed VA-ECMO support after transfer to the ICU and not within the initial procedure. We could not prove a clear benefit of prompt initiation of VA-ECMO after PCI. On the basis of our results, the precise VA-ECMO insertion time remains undetermined.

Mortality among patients in CS remains high. The 30-day and 6-month survival rates in the present study of 38% and 29%, respectively, are below those reported in some other studies.^9,10,14 These data of lower survival rates compared with other studies could be attributed to different factors. First, in the present study, roughly half of the patients received ECMO support after transferring to the ICU and not within the initial procedure of revascularization. In comparison, ECMO support in other studies with CS after STEMI was started during the index procedure.^9,10,15 Patients in these studies were in profound CS without hemodynamic stabilization despite IABP and high-dose inotropic and vasopressor support. Patients who developed late CS were stable under mechanical and medical support and therefore not included. Hence, selection bias may play a role in affecting survival rates. In our group, nearly half of the patients were believed to recover after PCI and supportive treatment with fluids, vasopressor, and inotropic agents. These patients were initially not in profound CS or stabilized after PCI and conventional therapy. Deterioration or even late development of refractory CS in the ICU may have caused delay of VA-ECMO insertion in these patients. Our study provides more accurate representation of a clinical practice wherein not every patient with STEMI and CS is in profound CS immediately. Nevertheless, optimal timing of mechanical unloading of the heart in STEMI patients remains unclear. Further research in this field is mandatory.

Second, in comparison with other studies, none of the patients in our group received assistance of additional percutaneous ventricular assist device (PVAD) for venting after implementation of VA-ECMO. This may have affected survival rates because VA-ECMO support may be associated with increased left ventricular pressures due to an increase in afterload and observational studies showed that left ventricular unloading was associated with a reduced mortality.¹⁶ Nevertheless, the possible benefit of left ventricular unloading is currently a hypothesis because prospective randomized studies are lacking.

It is known that acute kidney injury as a consequence of CS is associated with poorer outcome with higher risk in patients with pre-existing chronic kidney disease.⁷ Increased creatinine level and lower eGFR are independent predictors of mortality among patients in CS undergoing VA-ECMO.¹⁵ In the ENCOURAGE mortality score, which was developed to predict mortality of severe CS in acute myocardial infarction patients receiving an ECMO, creatinine >150 µmol/L was independently associated with increased mortality.⁴ Confirming these previous findings, in the present study, non-survivors had significantly higher mean creatinine levels compared with survivors.

Renal failure requiring temporary renal replacement therapy was the most frequently observed event in our study population (28%). However, it is markedly below the numbers reported in recent meta-analyses.^17,18 Surprisingly, the need for renal replacement therapy during VA-ECMO support did not correlate with outcome.

Some investigators showed that older age is a significant predictor of outcome in patients undergoing VA-ECMO.^19,20 Although age alone should not exclude VA-ECMO support for CS, these patients have a higher incidence of multiorgan dysfunction and lower in-hospital survival.^1,20 Our findings confirm a poorer 6-month survival for elderly patients, even though the study population was relatively young.

A significant role of arterial lactate level and its association with mortality among patients in CS has also been described previously.⁷ Our results showed that the level of arterial lactate at 1 hour and 24 hours after insertion of VA-ECMO has a strong correlation with mortality. In our study, 67% of survivors had arterial lactate levels ≤2 mmol/L after 24 hours as opposed to 36% in non-survivors. This indicates that lactate level after VA-ECMO initiation might be a better prognosticator than lactate prior to VA-ECMO.

Both groups were characterized by leukocytosis, with significantly higher values in the deceased group. Furthermore, it was found to be independently associated with poorer outcome. Recent studies showed leukocytosis as marker of inflammation in atherosclerosis and as independent predictor of cardiovascular events and mortality in patients with coronary artery disease.^21,22 The role of systemic inflammatory response syndrome and its association with impaired prognosis in patients with CS has been reported previously.^11,12

Some other biomarkers were identified as independent predictors of short-term mortality in infarct-related CS. However, their routine use is currently still limited.^23,24 Our findings suggest that an elevated leukocyte count should be considered an independent, generally available, and important prognostic factor in patients with STEMI complicated by CS. Nevertheless, a remaining unresolved issue is the role and potential treatment options of concomitant inflammation by patients in CS and STEMI.

Study limitations. This study has several limitations. First, it is an observational, single-center study. Second, nearly half of the patients received VA-ECMO support outside of the initial procedure. These patients developed refractory CS later in the ICU, which may have led to delayed VA-ECMO support. Third, left ventricular venting techniques were not used in any patient, which may have affected survival rates. Fourth, the reported sample size was relatively small (n = 79). Therefore, the number of risk factors included in the multivariate model was limited and other in-hospital factors, such as severe bacterial, fungal, or viral infections, or use of antimicrobial agents for presumed sepsis, were not taken into account. Finally, the study included predominantly male patients (n = 65). Hence, a possible interaction between sex and its influence on the primary outcome could not be investigated.

Conclusion

Impaired kidney function, an increased white blood cell count on admission, age, and elevated arterial lactate levels at 1 hour and 24 hours after VA-ECMO initiation were negative predictors for 6-month clinical outcome in patients with STEMI complicated by CS. Additionally, the current study could not identify an association between CS onset to VA-ECMO start time and 6-month survival.

Impact on daily practice. The timing of initiation of VA-ECMO in patients with CS complicating STEMI was not a prognostic factor of survival in our study and did not significantly influence outcome in this population. Renal function, white blood cell count, age, and lactate level were the strongest predictors of death during 6-month follow-up. Due to lack of evidence, the use of VA-ECMO should always be considered based on individual patient assessments and ethical aspects.

^aJoint first authors. ^bJoint senior authors.

From the ¹Heart Center Leipzig at University of Leipzig, Department of Internal Medicine/Cardiology and Leipzig Heart Institute, Leipzig, Germany; and ²University Clinic of Cardiac Surgery, Leipzig Heart Center, Leipzig, Germany.

Disclosure: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. The authors report no conflicts of interest regarding the content herein.

Manuscript accepted September 15, 2020.

Address for correspondence: Lukasz Szczanowicz, MD, Heart Center Leipzig at University of Leipzig, Department of Internal Medicine/Cardiology, Strümpellstr. 39, D-04289 Leipzig, Germany. Email: lukasz.szczanowicz@onet.eu

1. Lorusso R, Gelsomino S, Parise O, et al. Venoarterial extracorporeal membrane oxygenation for refractory cardiogenic shock in elderly patients. Trends in application and outcome from the extracorporeal life support organization (ELSO) registry. Ann Thorac Surg. 2017;104:62-69.
2. Chiu R, Pillado E, Sareh S, De La Cruz K, Shemin RJ, Benharash P. Financial and clinical outcomes of extracorporeal mechanical support. J Card Surg. 2017;32:215-221.
3. Schmidt M, Burrell A, Roberts L, et al. Predicting survival after ECMO for refractory cardiogenic shock. The survival after veno-arterial-ECMO (SAVE)-score. Eur Heart J. 2015;36:2246-2256.
4. Muller G, Flecher E, Lebreton G, et al. The ENCOURAGE mortality risk score and analysis of long-term outcomes after VA-ECMO for acute myocardial infarction with cardiogenic shock. Intensive Care Med. 2016;42:370-378.
5. Ibanez B, James S, Agewall S, et al. 2017 ESC guidelines for the management of acute myocardial infarction in patients presenting with ST-segment elevation. The task force for the management of acute myocardial infarction in patients presenting with ST-segment elevation of the European Society of Cardiology (ESC). Eur Heart J. 2018;39:119-177.
6. Thygesen K, Alpert JS, Jaffe AS, et al. Third universal definition of myocardial infarction. J Am Coll Cardiol. 2012;60:1581-1598.
7. Van Diepen S, Katz JN, Albert NM, et al. Contemporary management of cardiogenic shock: a scientific statement from the American Heart Association. Circulation. 2017;136:e232-e268.
8. Mehran R, Rao SV, Bhatt DL, et al. Standardized bleeding definitions for cardiovascular clinical trials: a consensus report from the Bleeding Academic Research Consortium. Circulation. 2011;123:2736-2747.
9. Sheu J-J, Tsai T-H, Lee F-Y, et al. Early extracorporeal membrane oxygenator-assisted primary percutaneous coronary intervention improved 30-day clinical outcomes in patients with ST-segment elevation myocardial infarction complicated with profound cardiogenic shock. Crit Care Med. 2010;38:1810-1817.
10. Lee W-C, Fang C-Y, Chen H-C, et al. Associations with 30-day survival following extracorporeal membrane oxygenation in patients with acute ST segment elevation myocardial infarction and profound cardiogenic shock. Heart Lung. 2016;45:532-537.
11. Mebazaa A, Combes A, van Diepen S, et al. Management of cardiogenic shock complicating myocardial infarction. Intensive Care Med. 2018;44:760-773.
12. Thiele H, Ohman EM, de Waha-Thiele S, Zeymer U, Desch S. Management of cardiogenic shock complicating myocardial infarction: an update 2019. Eur Heart J. 2019;40:2671-2683.
13. Huang CC, Hsu JC, Wu YW, et al. Implementation of extracorporeal membrane oxygenation before primary percutaneous coronary intervention may improve the survival of patients with ST-segment elevation myocardial infarction and refractory cardiogenic shock. Int J Cardiol. 2018;269:45-50.
14. Negi SI, Sokolovic M, Koifman E, et al. Contemporary use of veno-arterial extracorporeal membrane oxygenation for refractory cardiogenic shock in acute coronary syndrome. J Invasive Cardiol. 2016;28:52-57.
15. Chung S-Y, Tong M-S, Sheu J-J, et al. Short-term and long-term prognostic outcomes of patients with ST-segment elevation myocardial infarction complicated by profound cardiogenic shock undergoing early extracorporeal membrane oxygenator-assisted primary percutaneous coronary intervention. Int J Cardiol. 2016;223:412-417.
16. Al-Fares AA, Randhawa VK, Englesakis M, et al. Optimal strategy and timing of left ventricular venting during veno-arterial extracorporeal life support for adults in cardiogenic shock: a systematic review and meta-analysis. Circ Heart Fail. 2019;12:e006486.
17. Cheng R, Hachamovitch R, Kittleson M, et al. Complications of extracorporeal membrane oxygenation for treatment of cardiogenic shock and cardiac arrest. A meta-analysis of 1,866 adult patients. Ann Thorac Surg. 2014;97:610-616.
18. Zangrillo A, Landoni G, Biondi-Zoccai G, et al. A meta-analysis of complications and mortality of extracorporeal membrane oxygenation. Crit Care Resusc. 2013;15:172-178.
19. Waha S de, Graf T, Desch S, et al. Outcome of elderly undergoing extracorporeal life support in refractory cardiogenic shock. Clin Res Cardiol. 2017;106:379-385.
20. Lee W, Kim Y, Choi H, et al. Advanced age as a predictor of survival and weaning in venoarterial extracorporeal oxygenation. A retrospective observational study. Biomed Res Int. 2017;2017:3505784.
21. Madjid M, Awan I, Willerson JT, Ward Casscells S. Leukocyte count and coronary heart disease. Implications for risk assessment. J Am Coll Cardiol. 2004;44:1945-1956.
22. Ghaffari S, Nadiri M, Pourafkari L, et al. The predictive value of total neutrophil count and neutrophil/lymphocyte ratio in predicting in-hospital mortality and complications after STEMI. J Cardiovasc Thorac Res. 2014;6:35-41.
23. Fuernau G, Poenisch C, Eitel I, et al. Growth-differentiation factor 15 and osteoprotegerin in acute myocardial infarction complicated by cardiogenic shock – a biomarker substudy of the IABP-SHOCK II-trial. Eur J Heart Fail. 2014;16:880-887.
24. Iborra-Egea O, Rueda F, García-García C, Borràs E, Sabidó E, Bayes-Genis A. Molecular signature of cardiogenic shock. Eur Heart J. 2020;41:3839-3848.