1. Introduction
Cardiogenic shock (CS) is reported to be the most common cause of death following acute myocardial infarction (MI).[1] Fluid administration for the maintenance of volume status is the first-line therapy for CS. In addition, inotropes and vasopressors are used to improve the risk of cardiac mortality in over 90% of CS cases.[2] In the Should We Emergently Revascularize Occluded Coronaries for Cardiogenic Shock (SHOCK) trial, early revascularization with initial medical stabilization yielded favorable long-term clinical outcomes.[3] Regarding percutaneous coronary intervention (PCI) strategies for CS, current guidelines recommend culprit-only PCI followed by staged PCI after hemodynamic stabilization based on the results of the randomized, multicenter Culprit-Lesion-Only PCI versus Multi-vessel PCI in Cardiogenic Shock trial.[4]
There have been recent advances in the evidence-based management of patients in CS. However, CS’s 40 to 50% mortality rate has not changed in a decade.[5] Mechanical circulatory support (MCS) reduces the need for inotropes and vasopressors and improves hemodynamic instability which could lead to unfavorable clinical outcomes.[6] However, the intra-aortic balloon counterpulsation (IABP)-SHOCK II trial[7] showed that the IABP does not provide survival benefits. This finding may lead to a decrease in the use of IABP and an increase in the use of other MCS methods, such as the use of percutaneous ventricular assist devices or extracorporeal membrane oxygenation (ECMO).[8] However, despite the increase in the number of devices used for mechanical support in CS, data from randomized clinical trials regarding their effectiveness are lacking.
A small randomized controlled trial demonstrated no differences in cardiac function (50.0% vs 50.8%; P = .86) and 30-day mortality (19% vs 33%; P = .37) between using ECMO alone and medical management.[9] Studies regarding the comparisons between the use of ECMO and IABP separately in MI-CS revealed a significantly increased risk of 30-day mortality in the ECMO group.[10–13] One study compared the mortality between ECMO ± IABP and IABP alone (81% vs 74%; P = .5),[14] while another revealed a lower incidence of IABP ± ECMO compared with IABP alone (33% vs 68%; P = .01).[15] However, these observational registry-based studies only reported unadjusted incidences of short-term mortality, which could lead to false interpretations of the effect of ECMO support in MI-CS.
Sustaining adequate perfusion and unloading of the ventricle during CS is essential to improve clinical outcomes. Venoarterial ECMO (VA-ECMO) can provide >4.5 L of oxygen per minute without the need for surgery. In addition, oxygenation is generally preferred over other types of MCS.[16] Moreover, a recent systematic analysis of 17 observational studies (3997 patients) concluded that ECMO support in addition to left ventricular (LV) unloading is associated with decreased 30-day mortality risk (relative ratio, 0.79; 95% confidence interval [CI], 0.72–0.87; P < .00001).[17] Therefore, VA-ECMO with unloading, in addition to early revascularization, could effectively maintain adequate perfusion and result in a good clinical outcome of MI-CS. This study aimed to evaluate the adjusted incidences of 30-day and 1-year mortality of ECMO + IABP over medical management ± IABP for acute MI patients with CS who underwent early revascularization.
2. Method
2.1. Study design and population
The study was conducted using data extracted from the Korea Acute Myocardial Infarction-National Institute of Health (KAMIR-NIH) registry from October 2011 to December 2015. The KAMIR-NIH is a prospective, multicenter registry study designed to characterize the clinical features and prognoses of acute MI in Korean patients. Data coordinators met with the study investigators at least twice a year to discuss the research and examine the study protocol.[18]
Patients were included in the study in accordance with the following inclusion criteria: age > 18 years; initial presentation with MI, which was diagnosed based on clinical manifestations; elevated levels of myocardial biomarkers, including creatinine kinase (CK)-myoglobin and troponin-I or T; and changes in the 12 leads of an electrocardiogram. CS was defined as a systolic blood pressure < 90 mm Hg for at least 30 minutes that required the administration of catecholamines, such as adrenaline, to maintain it at ≥90 mm Hg.[19] In addition to this definition, at least one of the following pieces of evidence of end-organ damage was expected: altered mental status, cold skin and extremities, and decreased urine output (<30 mL/h). Furthermore, the included patients must have undergone early PCI. Conversely, patients who met at least one of the following criteria were excluded: coronary angiogram did not show a coronary artery disease; did not undergo PCI; underwent coronary artery bypass grafting; lost to follow-up; incomplete hemodynamic data; and did not initially present with CS. Only patients with MI complicated by CS were included. The included patients were categorized into VA-ECMO with IABP and no VA-ECMO support groups.
PCIs were performed according to the current practice guidelines. Any type of stent could be used without restriction. The decision to perform multi-vessel revascularization was left to the attending physician. The indications for VA-ECMO initiation were generally based on the Extracorporeal Life Support Organization guideline for Venoarterial Extracorporeal Membrane Oxygenation in Adult Cardiac Patients[19]; however, the final decision was at the discretion of the attending physician.
2.2. Clinical outcomes and statistical analysis
This study’s primary and secondary endpoints were 30-day and 1-year mortality, respectively. Cardiac death was defined as death resulting from arrhythmia, pump failure, or mechanical complications, including free-wall and ventricular septal ruptures. Death resulting from causes besides cardiac ones was considered a non-cardiac death.[20] Regarding comparative analysis, continuous variables were compared using Student t test or the Wilcoxon rank-sum test, whereas categorical variables were compared using the χ2 or Fisher exact test.
The Kaplan–Meier method was used to estimate the incidence of clinical outcomes between the groups during follow-up, and the results were compared using the log-rank test. The weighted Kaplan–Meier method is used to control the censoring of survival data and can outperform the crude Kaplan–Meier method.[21] Variables that could affect the outcomes, such as age, sex, Killip class, culprit-only PCI, thrombolysis in myocardial infarction (TIMI) major bleeding, acute kidney injury, multiorgan failure, and cardiopulmonary resuscitation, were used to calculate weights using a generalized linear regression model and compared using the weighted log-rank test. Finally, landmark analysis was performed at the 30-day mark, and the adjusted incidence was compared between the groups before and after 30 days.
A propensity score was calculated to assess the clinical outcomes through adjusted baseline clinical differences between the 2 groups.[22] Multiple generalized linear regressions were performed to calculate the propensity score with ECMO support as the dependent variable. The variables that were significantly different between the groups were as follows: age, sex, Killip class, culprit-only PCI, for 30-days mortality and included TIMI major bleeding, acute kidney injury, multiorgan failure, and cardiopulmonary resuscitation for 1-year mortality.
Owing to the limited number of events that occurred in the study population, inverse probability of treatment weighting, calculated from the same variables used for propensity scoring, was used to estimate the average treatment effect of ECMO support and control for confounding factors.[23]
Comparisons of 30-day and 1-year mortality between the groups were based on Cox regression models, including unadjusted, propensity score-adjusted, and weighted models. Proportional hazard assumptions were checked using the log-log survival plot and Schoenfield residuals.
Simple and multiple Cox regression analyses were performed to identify the risk factors for 30-day mortality. A stepwise Cox regression model was created with bidirectional variable selection using the estimation of a function of the posterior probability of a model being true, under a certain Bayesian setup. In addition, another stepwise logistic regression model, which was created with bidirectional selection using Bayesian information criterion, was used to identify the risk factors for increased probability of requiring ECMO support at initial presentation. The following variables were included in the regression analysis: baseline characteristics, such as age, sex, comorbidities, hemodynamic data, and PCI characteristics, which showed significant associations in the simple logistic regression analysis (P < .1) with >90% data availability. Two-sided P values < .05 were considered statistically significant. Analyses were performed using the R software, version 3.1.1 (R Foundation for Statistical Computing, Vienna, Austria).
3. Results
During the study period, 13,105 patients were enrolled in the KAMIR-NIH registry at various study sites. After excluding patients whose coronary angiograms did not show coronary artery disease (n = 209) and those who did not undergo PCI (n = 128), underwent coronary artery bypass grafting (n = 37), were lost to follow-up (n = 127), had incomplete hemodynamic data (n = 345), and did not present with CS at initial presentation (n = 12,304), 300 patients with MI complicated by CS were included in this study. The included patients were categorized into VA-ECMO + IABP support (n = 39) and no VA-ECMO support (medical management ± IABP) (n = 261) groups (Fig. 1). Only 39 (13%) patients who were initially diagnosed with CS underwent ECMO in addition to IABP at the time of PCI. The remaining patients (n = 261, 87%) with CS were treated with only inotropes and vasopressors, and/or IABP (Fig. 1).
;)
Figure 1.: Flowchart of the study. IABP = intra-aortic balloon counterpulsation, IPTW = inverse probability of treatment weighting, KAMIR-NIH = the Korean Acute Myocardial Infarction Registry-National Institute of Health, K-M = Kaplan–Meier curve, MI = myocardial infarction, PCI = percutaneous coronary intervention, VA-ECMO = venoarterial extracorporeal membrane oxygenation.
The baseline and procedural characteristics of the patients are shown in Table 1. There were some differences in the baseline characteristics of the 2 patient groups. Patients from the VA-ECMO + IABP support group were more likely to be young and men in contrast to the no VA-ECMO support group. Additionally, patients from the VA-ECMO + IABP support group had a higher Killip class than those from the no VA-ECMO support group (97.4% vs 70.9%; P = .006). The prevalence of cardiovascular risk factors in both groups was similar; however, history of angina or PCI was more common in the VA-ECMO + IABP support group than in the no VA-ECMO support group. The patients from the ECMO + IABP support group had higher initial serum hemoglobin, peak CK, and troponin levels than those from the no VA-ECMO support group. Hospital survivors from the VA-ECMO + IABP support group were prescribed aspirins, beta-blockers, angiotensin-converting enzyme inhibitors/angiotensin receptor blockers, and statins less frequently than those from the no VA-ECMO support group, which could have led to poor outcomes. Additionally, patients from the VA-ECMO + IABP support group showed multi-vessel involvement and left main artery disease, whereas patients from the no VA-ECMO support group showed single-vessel involvement with the right coronary artery as the culprit vessel. However, there were no differences in successful PCI, multi-vessel revascularization, and final post-procedural TIMI flow grade between the 2 groups. Low LV ejection fraction and high regional wall motion score were more common in the VA-ECMO + IABP support group than in the no VA-ECMO support group. However, there were no differences in volume and dimensional indexes between the 2 groups.
Table 1 -
Baseline characteristics of the study population, including angiographic and echocardiographic assessment data.
|
No VA-ECMO N = 261 (87%) |
VA-ECMO + IABP N = 39 (13%) |
P value |
| Male, n (%) |
177 (67.8) |
34 (87.2) |
.023 |
| Age, median (SD) |
67.31 (12.99) |
62.10 (12.59) |
.020 |
| History of angina |
34 (13.0) |
3 (7.7) |
.494 |
| Systolic BP, mm Hg, median (SD) |
54.08 (32.53) |
47.82 (33.96) |
.286 |
| Diastolic BP, mm Hg, median (SD) |
30.65 (22.61) |
26.90 (24.42) |
.371 |
| Heat rate, per min, median (SD) |
60.87 (38.36) |
61.28 (41.87) |
.954 |
| Killip class, n (%) |
|
|
|
| I |
52 (19.9) |
1 (2.6) |
.006 |
| II |
13 (5.0) |
0 (0.0) |
|
| III |
11 (4.2) |
0 (0.0) |
|
| IV |
185 (70.9) |
38 (97.4) |
|
| Body mass index, kg/m2, median (SD) |
23.29 (3.12) |
24.14 (2.90) |
.161 |
| Hypertension, n (%) |
149 (57.1) |
21 (53.8) |
.835 |
| Diabetes, n (%) |
99 (37.9) |
13 (33.3) |
.707 |
| Dyslipidemia, n (%) |
21 (8.0) |
3 (7.7) |
1.000 |
| History of MI, n (%) |
22 (8.4) |
5 (12.8) |
.369 |
| History of angina, n (%) |
19 (7.3) |
7 (17.9) |
.059 |
| History of PCI, n (%) |
26 (10.0) |
9 (23.1) |
.035 |
| History of heart failure, n (%) |
10 (3.8) |
1 (2.6) |
1.000 |
| History of CVD, n (%) |
22 (8.4) |
3 (7.7) |
1.000 |
| Current smoker, n (%) |
140 (53.6) |
26 (66.7) |
.176 |
| Family history of CAD, n (%) |
7 (2.7) |
3 (8.1) |
.119 |
| WBC (×1000), median (SD) |
13.42 (5.15) |
14.22 (5.38) |
.390 |
| Serum Hb, median (SD) |
12.67 (2.39) |
13.73 (1.74) |
.001 |
| Serum glucose level, median (SD) |
250.99 (123.01) |
282.79 (128.39) |
.159 |
| Serum creatinine level, median (25th–75th quartile) |
1.20 (0.90–1.50) |
1.30 (0.95–1.70) |
.144 |
| Serum peak CK, median (25th–75th quartile) |
670 (137–2515) |
6632 (1383–12,938) |
<.001 |
| Serum peak troponin, median (25th–75th quartile) |
31.51 (14.69–89.08) |
153.48 (54.50–270.62) |
<.001 |
| High-density lipoprotein, median (SD) |
38.31 (10.64) |
34.35 (12.39) |
.192 |
| Low-density lipoprotein, median (SD) |
96.81 (40.97) |
81.68 (40.64) |
.135 |
| hsCRP, median (25th–75th quartile) |
0.30 (0.10–1.76) |
0.11 (0.04–0.96) |
.152 |
| BNP, median (25th–75th quartile) |
105.00 (33.00–411.48) |
535.00 (274.75–1058.00) |
.656 |
| HbA1c, median (25th–75th quartile) |
6.10 (5.60–6.90) |
6.05 (5.50–6.20) |
.493 |
| Discharge medications |
|
|
|
| Aspirin, n (%) |
256 (98.1) |
36 (92.3) |
.072 |
| P2Y12inhibitor, n (%) |
|
|
|
| Clopidogrel |
169 (66.8) |
25 (69.4) |
.750 |
| Ticagrelor |
55 (21.7) |
6 (16.7) |
|
| Prasugrel |
29 (11.5) |
5 (13.9) |
|
| Calcium channel blocker, n (%) |
6 (2.3) |
0 (0.0) |
1.000 |
| Beta-blocker, n (%) |
157 (60.2) |
5 (12.8) |
<.001 |
| ACEi and/or ARB, n (%) |
147 (56.3) |
7 (17.9) |
<.001 |
| Statin, n (%) |
173 (66.3) |
10 (25.6) |
<.001 |
| Hypoglycemic agent, n (%) |
44 (100.0) |
3 (75.0) |
.083 |
| PCI characteristics |
|
|
|
| Access site, n (%) |
|
|
|
| Radial |
21 (8.0) |
1 (2.6) |
.330 |
| Femoral |
240 (92.0) |
38 (97.4) |
|
| GP IIb/IIIa inhibitor, n (%) |
64 (24.5) |
12 (30.8) |
.523 |
| Use of closure device, n (%) |
44 (16.9) |
2 (5.1) |
.097 |
| Target vessel, n (%) |
|
|
|
| Left anterior descending |
91 (34.9) |
12 (30.8) |
<.001 |
| Left circumflex |
21 (8.0) |
5 (12.8) |
|
| Right coronary |
133 (51.0) |
7 (17.9) |
|
| Left main involve |
16 (6.1) |
15 (38.5) |
|
| Lesion type of target vessel, n (%) |
|
|
|
| Type A |
1 (0.4) |
0 (0.0) |
.308 |
| Type B1 |
19 (7.3) |
2 (5.1) |
|
| Type B2 |
99 (38.1) |
21 (53.8) |
|
| Type C |
141 (54.2) |
16 (41.0) |
|
| Number of involved vessels, n (%) |
|
|
|
| 1-vessel disease |
109 (41.8) |
12 (30.8) |
<.001 |
| 2-cessel disease |
82 (31.4) |
7 (17.9) |
|
| 3-vessel disease |
45 (17.2) |
4 (10.3) |
|
| Left main involvement |
25 (9.6) |
16 (41.0) |
|
| Pre-TIMI flow of target vessel, n (%) |
|
|
|
| 0 |
172 (66.2) |
25 (64.1) |
.969 |
| 1 |
22 (8.5) |
4 (10.3) |
|
| 2 |
30 (11.5) |
4 (10.3) |
|
| 3 |
36 (13.8) |
6 (15.4) |
|
| Treatment of target vessel, n (%) |
|
|
|
| Balloon only |
22 (8.4) |
5 (12.8) |
.369 |
| Stent implantation |
239 (91.6) |
34 (87.2) |
|
| Post-TIMI flow, n (%) |
|
|
|
| 1 |
6 (2.3) |
2 (5.1) |
.280 |
| 2 |
12 (4.6) |
3 (7.7) |
|
| 3 |
243 (93.1) |
34 (87.2) |
|
| Type of index procedure, n (%) |
|
|
|
| Culprit-only revascularization |
214 (82.0) |
32 (82.1) |
1.000 |
| Multi-vessel revascularization |
47 (18.0) |
7 (17.9) |
|
| Staged PCI, n (%) |
27 (10.3) |
2 (5.1) |
.396 |
| Final diagnosis, n (%) |
|
|
|
| NSTEMI |
45 (17.3) |
8 (20.5) |
.792 |
| STEMI |
215 (82.7) |
31 (79.5) |
|
| Echocardiographic characteristics, median (SD) |
| Left ventricle EF |
48.47 (12.79) |
32.85 (13.48) |
<.001 |
| Regional wall motion score index |
1.52 (0.42) |
1.81 (0.46) |
.026 |
| Left ventricle ESD |
34.48 (8.88) |
38.36 (8.69) |
.160 |
| Left ventricle EDD |
49.15 (6.39) |
48.36 (8.66) |
.763 |
| Left ventricle ESV |
53.47 (36.63) |
68.15 (31.13) |
.211 |
| Left ventricle EDV |
100.33 (45.09) |
104.79 (38.45) |
.749 |
ACE = angiotensin-converting enzyme, ARB = angiotensin receptor blocker, BNP = B-type natriuretic peptide, BP = blood pressure, CAD = coronary artery disease, CK = creatinine kinase, CVD = cerebrovascular disease, EDD = end-diastolic dimension, EDV = end-diastolic volume, EF = ejection fraction, ESD = end-systolic dimension, ESV = end-systolic volume, Hb = hemoglobin, HbA1c = glycated hemoglobin, hsCRP = high- sensitivity C-reactive protein, IABP = intra-aortic balloon counterpulsation, LV = left ventricular, NSTEMI = non-ST segment elevation myocardial infarction, PCI = percutaneous coronary intervention, SD = standard deviation, STEMI = ST-segment elevation myocardial infarction, TIMI = thrombolysis in myocardial infarction, VA-ECMO = venoarterial-extracorporeal membrane oxygenation, WBC = white blood cell.
The crude incidence of major complications within 30 days in both groups is shown in Table 2. The incidence of TIMI major bleeding, ventricular fibrillation, acute kidney injury, multiorgan failure, and cardiopulmonary resuscitation was significantly higher in the VA-ECMO + IABP support group than in the no VA-ECMO support group. Severe bleeding[24] (no data about the cause of bleeding) and multiorgan failure[25] are well-known indicators of poor outcomes of CS.
Table 2 -
Comparison of major complications in both groups within 30 days.
|
No VA-ECMO |
VA-ECMO + IABP |
P value |
| N = 261 (87%) |
N = 39 (13%) |
| CVA, n (%) |
0 (0.0) |
1 (2.6) |
.130 |
| Brain hemorrhage, n (%) |
1 (0.0) |
2 (5.2) |
.130 |
| TIMI major bleeding, n (%) |
18 (6.9) |
15 (38.5) |
<.001 |
| TIMI minor bleeding, n (%) |
19 (7.3) |
7 (17.9) |
.059 |
| AV block, n (%) |
43 (16.5) |
4 (10.3) |
.447 |
| VT-controlled by antiarrhythmics, n (%) |
14 (5.4) |
4 (10.3) |
.269 |
| VT- controlled by shock, n (%) |
57 (21.8) |
14 (35.9) |
.085 |
| Ventricular fibrillation, n (%) |
46 (17.6) |
18 (46.2) |
<.001 |
| Atrial fibrillation, n (%) |
43 (16.5) |
6 (15.4) |
1.000 |
| Acute kidney injury, n (%) |
8 (3.1) |
6 (15.4) |
.005 |
| Multi-organ failure, n (%) |
11 (4.2) |
7 (17.9) |
.004 |
| Cardiopulmonary resuscitation, n (%) |
140 (53.6) |
37 (94.9) |
<.001 |
AV = atrioventricular, CVA = cerebrovascular disease, IABP = intra-aortic balloon counterpulsation, TIMI = thrombolysis in myocardial infarction, VA-ECMO = venoarterial-extracorporeal membrane oxygenation, VT = ventricular tachycardia.
The cumulative incidence of 30-day mortality was 84.8% in the VA-ECMO + IABP support group and 49% in the no VA-ECMO support group (Fig. 2A). The difference in mortality between the groups was maintained at the 1-year follow-up (93.9% vs 62.7%; P = .001) (Fig. 2B). However, the adjusted cumulative incidence derived from the weighted Kaplan–Meier method by adjusting for confounding risk factors was different. The difference in 30-day mortality between the 2 groups decreased and lost its statistical significance (P value increased from 0.001 to 0.083) (Fig. 2C). After the extension of the follow-up period up to 1 year, there was no difference in mortality between the 2 groups (Fig. 2D). Landmark analysis at the 30-day timepoint showed that the adjusted incidence of mortality in the ECMO support group significantly decreased 30 days after the index hospitalization (Fig. 2D, picture-in-picture).
Figure 2.: Crude cumulative incidence of (A) 30-day mortality and (B) 1-year mortality, and VA-ECMO + IABP versus no VA-ECMO support and adjusted incidence curve with weighted Kaplan–Meier methods for (C) 30-day and (D) 1-year mortality. Landmark analysis at the 30-day timepoint. IABP = intra-aortic balloon counterpulsation, Tx = treatment, VA-ECMO = venoarterial extracorporeal membrane oxygenation.
The propensity score-adjusted and weighted Cox regression models, which were constructed to adjust for possible confounding factors, showed similar results (Table 3). The crude risks of 30-day and 1-year mortality in the VA-ECMO + IABP support group were significantly higher than those in the no VA-ECMO support group (unadjusted hazard ratio [HR], 1.97 and 2.05; 95% CI, 1.26–3.09 and 1.34–3.13; P value, .003 and <.001; respectively). However, the propensity-adjusted and inverse probability of treatment weighting models showed no differences in 30-day (HR, 1.57; 95% CI, 0.92–2.77 P = .346 and HR, 1.44; 95% CI, 0.42–3.17; P = .452, respectively) and 1-year mortality (HR, 1.56; 95% CI, 0.95–2.56; P = .076 and HR, 1.33; 95% CI, 0.57–3.06; P = .51, respectively) between the groups.
Table 3 -
Outcomes of extracorporeal membrane oxygenation support.
|
Unadjusted HR (95% CI) |
P value |
*PS-adjusted HR (95% CI) |
P value |
†IPTW HR (95% CI) |
P value |
| 30-d mortality |
1.97
(1.26–3.09) |
.003 |
1.57
(0.92–2.77) |
.346 |
1.44
(0.42–3.17) |
.452 |
1-yr
mortality |
2.05
(1.34–3.13) |
<.001 |
1.56
(0.95–2.56) |
.076 |
1.33
(0.57–3.06) |
.510 |
CI = confidence interval, HR = hazard ratio, IPTW = inverse probability of treatment weighting, PS = propensity score.
*Adjusted PS model: propensity score-adjusted Cox regression model constructed using the following variables: age, sex, Killip class, culprit-only PCI, for 30-days mortality and included TIMI major bleeding, acute kidney injury, multiorgan failure, and cardiopulmonary resuscitation for 1-year mortality.
†IPTW model: weighted Cox model with calculated average weighted treatment effect constructed using the same variables used for the adjusted PS model.
A stepwise Cox regression analysis showed that old age (HR, 1.018; 95% CI, 1.002–1.035; P = .032), higher Killip class at initial presentation (HR, 1.501; 95% CI, 1.138–1.980; P = .067), low post-TIMI flow grade after PCI (HR, 0.423; 95% CI, 0.266–0.668; P ≤ 0.001), and cardiopulmonary resuscitation (HR, 9.494; 95% CI, 3.795–23.749; P < .001) were significantly associated with increased risk of 30-day mortality (Fig. 3). Another stepwise logistic regression analysis showed that patients with peak serum CK level at initial presentation were likely to receive ECMO support (odds ratio, 1.000093; 95% CI, 1.000183–1.000273; P < .001)
Figure 3.: Independent predictors of 30-day mortality. CI = confidence interval, HR = hazard ratio, TIMI = thrombolysis in myocardial infarction.
4. Discussion
The main finding of this study was that in the study population of patients with MI in CS, adjusted 30-day and 1-year mortality did not differ between the VA-ECMO + IABP support and no VA-ECMO support groups. However, the better survival benefit of VA-ECMO + IABP support 30 days after index hospitalization could be supposed. More patients from the VA-ECMO + IABP support group died between days 4 and 30 than from the no VA-ECMO support group. However, this finding was not statistically significant. Most patients with peak CK levels at initial presentation received early VA-ECMO + IABP support. Old age, high Killip class, non-recovery of TIMI flow grade after PCI, and cardiopulmonary resuscitation were independent predictors of 30-day mortality in patients with MI in CS.
CS is not a single disease entity caused by decreased cardiac output. The IABP-SHOCK study showed that the survival of patients with MI in CS after the initial 24 hours is not associated with the cardiac index.[26] After initial shock occurred, clinical factors such as age, body mass index, lactate, anterior wall infarction, TIMI-3 flow after PCI, cardiopulmonary resuscitation time, and time from arrest to extracoporeal cardiopulmonary resiscitation may affect the clinical outcomes in MI-CS[27]; however, the main cause of mortality during shock is multiorgan dysfunction, which is triggered by severe systemic inflammation cascade,[28] microvascular dysfunctions, and decreased perfusion to peripheral organs.[25]
To prevent multiorgan dysfunction, device-related complications should be minimized during shock.[16] ECMO has several issues associated with an increased risk of access site bleeding and an increase in afterload, which may lead to increased end-diastolic pressure and LV volume. Severe bleeding complications during shock occur in 20 to 90% of CS cases and are associated with increased use of MCS. In addition, severe bleeding complications are not just related to the access site but to consumptive coagulopathy and acquired platelet dysfunction as well.[24] A meta-analysis of 20 studies (1866 patients) indicated that the risk of major bleeding during ECMO support is 40.8% (26.8–56.6%). In addition, the study showed that the more severe the bleeding events, the higher the risk of 30-day mortality (bleeding: HR, 2.11; severe bleeding: HR, 2.80).[29] This could explain the increased adjusted incidence of short-term mortality after 4 days among patients from the VA-ECMO + IABP support group, who were enrolled in a more severe shock state and had more 30-day complications, including bleeding, than those from the no VA-ECMO support group.
Increased afterload is another problem encountered while maintaining ECMO support. Various unloading techniques, such as concomitant use of IABP, impella, and left atrium or left pulmonary artery venting, have been described for the prevention of volume overloading to the left ventricle.[30] IABP can also be a good option for an unloading technique during ECMO support. VA-ECMO could increase afterload, which would result in pulmonary edema.[31] But ECMO support, in addition to IABP, have several beneficial effects: decreased myocardial oxygen demand and reduced pre-capillary wedge pressure in the pulmonary artery, which results in decreased hydrostatic pulmonary edema[32]; creates pulsatile perfusion flow, which could deactivate the coagulation pathway,[33] protect from kidney injury,[34] and improve lung perfusion[35]; increased hemodynamic profile in animal models, such as mean coronary artery flow, diastolic pressure time index, LV pressure-volume area, and tension time index.[36] Recent study showed that ECMO, in addition to IABP support with proper management of device-related complications, may lead to improved clinical outcomes of CS in patients with MI (HR 0.30, 95% CI, 0.25–0.37), similar results as shown in the present study.[37]
The strength of this study was that long-term mortality included short-term mortality with VA-ECMO + IABP support, which we provided with adjusted incidence rates using various statistical methods. A recent systematic review showed that most studies regarding the clinical outcomes of ECMO support only reported crude incidence and indicated the harmful effects of ECMO during the early stage of MI-CS.[38] Moreover, ECMO support, in addition to LV unloading, is crucial for improving the clinical outcomes of MI-CS,[39] which our study reflected in a real-world clinical setting.
Several randomized controlled trials conducted to determine the efficacy of ECMO support for treating patients with MI-CS are ongoing. The ECLS-SHOCK (NCT03637205),[40] EURO-SHOCK (NCT03813134),[41] and ANCHOR (NCT04184635) trials are adequately powered to evaluate 30-day mortality as the primary endpoint. In addition, high-risk patients in CS are enrolled in these trials. Guidelines on the efficacy, proper management, and appropriate selection of ECMO support will be produced from these trials.
The present study had some limitations. First, this was a registry-based study and thus prone to selection bias. Although we used weighted Cox regression and adjusted Kaplan–Meier curves to control for such confounding factors, unidentified or residual confounders may have affected the study results. Second, the management of patients with MI in CS may vary depending on the volume of CS cases in an institution. Compared to hospitals with low volumes of CS cases, which may adopt a more defensive approach, hospitals with a large volume of CS cases have low mortality rates, even after adjusting for recent advances in the management of CS, including early revascularization.[42] Third, experience in the initiation and management of ECMO support may have influenced the results. This includes different treatment durations of ECMO support among the study participants; therefore, we could not assess the correlation between treatment duration and clinical outcomes. Fourth, the lack of detailed information regarding ECMO support in special cases of cardiac arrest with ongoing cardiopulmonary resuscitation is another limitation. This is because ECMO cardiopulmonary resuscitation may be correlated with an increase in 30-day mortality by up to 13%.[17] Additionally, the lack of details regarding the dosages of vasopressors and inotropes could have influenced the study results. There were also not enough data about the causes of the complications observed during the 30-day period. Finally, recent guidelines for the use of IABP in CS do not recommend it anymore; however, IABP is still used as a last-resort treatment for patients in severe CS to increase myocardial perfusion.[43]
In conclusion, this study demonstrated that patients with MI in CS treated with ECMO, in addition to IABP support, have comparable overall 30-day and 1-year risks of mortality compared to those not managed with ECMO. However, it appears that survivors after short-term ECMO support may have a better clinical outcome which should be considered hypothesis-generating.
Acknowledgment
We would like to thank Editage (www.editage.com) for English language editing.
Author contributions
Conceptualization: Jin Sup Park.
Data curation: Jeong Cheon Choe, Sun-Hack Lee, Jin Hee Ahn, Hye Won Lee, Jun-Hyok Oh, Jung Hyun Choi, Han Cheol Lee, Kwang Soo Cha, Myung Ho Jeong.
Formal analysis: Jin Sup Park.
Funding acquisition: Jin Sup Park.
Investigation: Jeong Cheon Choe, Sun-Hack Lee, Jin Hee Ahn, Hye Won Lee, Jun-Hyok Oh, Jung Hyun Choi, Han Cheol Lee, Kwang Soo Cha, Myung Ho Jeong.
Methodology: Jin Sup Park.
Writing – review & editing: Dominick J Angiolillo, Jin Sup Park.
References
[1]. Hochman JS, Buller CE, Sleeper LA, et al. Cardiogenic shock complicating acute myocardial infarction – etiologies, management and outcome: a report from the SHOCK trial registry. SHOULD we emergently revascularize occluded coronaries for cardiogenic shock? J Am Coll Cardiol. 2000;36(suppl A):1063–70.
[2]. Thiele H, Zeymer U, Neumann F, et al. Intraaortic balloon support for myocardial infarction with cardiogenic shock. N Engl J Med. 2012;367:1287–96.
[3]. Hochman JS, Sleeper LA, Webb JG, et al. Early revascularization in acute myocardial infarction complicated by cardiogenic shock. SHOCK investigators. Should we emergently revascularize occluded coronaries for cardiogenic shock. N Engl J Med. 1999;341:625–34.
[4]. Collet J-P, Thiele H, Barbato E, et al. 2020 ESC guidelines for the management of acute coronary syndromes in patients presenting without persistent ST-segment elevation. Eur Heart J. 2021;42:1289–367.
[5]. Thiele H, Akin I, Sandri M, et al. PCI strategies in patients with acute myocardial infarction and cardiogenic shock. N Engl J Med. 2017;377:2419–32.
[6]. Thiele H, Ohman EM, de Waha-Thiele S, et al. Management of cardiogenic shock complicating myocardial infarction: an update 2019. Eur Heart J. 2019;40:2671–83.
[7]. Thiele H, Zeymer U, Thelemann N, et al. Intraaortic balloon pump in cardiogenic shock complicating acute myocardial infarction: long-term 6-year outcome of the randomized IABP-SHOCK II trial. Circulation. 2019;139:395–403.
[8]. Shah M, Patnaik S, Patel B, et al. Trends in mechanical circulatory support use and hospital mortality among patients with acute myocardial infarction and non-infarction related cardiogenic shock in the United States. Clin Res Cardiol. 2018;107:287–303.
[9]. Brunner S, Guenther SPW, Lackermair K, et al. Extracorporeal life support in cardiogenic shock complicating acute myocardial infarction. J Am Coll Cardiol. 2019;73:2355–7.
[10]. Choi KH, Yang JH, Park TK, et al. Differential prognostic implications of vasoactive inotropic score for patients with acute myocardial infarction complicated by cardiogenic shock according to use of mechanical circulatory support. Crit Care Med. 2021;49:770–80.
[11]. Vojjini R, Patlolla SH, Cheungpasitporn W, et al. Racial disparities in the utilization and outcomes of temporary mechanical circulatory support for acute myocardial infarction-cardiogenic shock. J Clin Med. 2021;10:1459.
[12]. Lauridsen MD, Josiassen J, Schmidt M, et al. Prognosis of myocardial infarction-related cardiogenic shock according to preadmission out-of-hospital cardiac arrest. Resuscitation. 2021;162:135–42.
[13]. Vallabhajosyula S, Subramaniam AV, Murphree DH, et al. Complications from percutaneous-left ventricular assist devices versus intra-aortic balloon pump in acute myocardial infarction-cardiogenic shock. PLoS One. 2020;15:e0238046.
[14]. Aiba T, Nonogi H, Itoh T, et al. Appropriate indications for the use of a percutaneous cardiopulmonary support system in cases with cardiogenic shock complicating acute myocardial infarction. Jpn Circ J. 2001;65:145–9.
[15]. Tsao NW, Shih CM, Yeh JS, et al. Extracorporeal membrane oxygenation-assisted primary percutaneous coronary intervention may improve
survival of patients with acute myocardial infarction complicated by profound cardiogenic shock. J Crit Care. 2012;27:530.e1–11.
[16]. Werdan K, Gielen S, Ebelt H, et al. Mechanical circulatory support in cardiogenic shock. Eur Heart J. 2014;35:156–67.
[17]. Russo JJ, Aleksova N, Pitcher I, et al. Left ventricular unloading during extracorporeal membrane oxygenation in patients with cardiogenic shock. J Am Coll Cardiol. 2019;73:654–62.
[18]. Lee JM, Rhee TM, Hahn JY, et al. Multivessel percutaneous coronary intervention in patients with ST-segment elevation myocardial infarction with cardiogenic shock. J Am Coll Cardiol. 2018;71:844–56.
[19]. Lorusso R, Shekar K, MacLaren G, et al. ELSO interim guidelines for venoarterial extracorporeal membrane oxygenation in adult cardiac patients. ASAIO J. 2021;67:827–44.
[20]. Ahn KT, Oh JK, Seong SW, et al. One-year clinical outcomes between single- versus multi-staged PCI for ST elevation myocardial infarction with multi-vessel coronary artery disease: from Korea Acute Myocardial Infarction Registry-National Institute of Health (KAMIR-NIH). Korean Circ J. 2020;50:220–33.
[21]. Lee S-H, Lee E-J. Modifying weighted Kaplan-Meier test for two-sample
survival comparison. EJMATH. 2022;3:51–8.
[22]. Heinze G, Jüni P. An overview of the objectives of and the approaches to propensity score analyses. Eur Heart J. 2011;32:1704–8.
[23]. Mansournia MA, Altman DG. Inverse probability weighting. BMJ. 2016;352:i189.
[24]. Thiele H, Jobs A, Ouweneel DM, et al. Percutaneous short-term active mechanical support devices in cardiogenic shock: a systematic review and collaborative meta-analysis of randomized trials. Eur Heart J. 2017;38:3523–31.
[25]. De Backer D, Creteur J, Dubois MJ, et al. Microvascular alterations in patients with acute severe heart failure and cardiogenic shock. Am Heart J. 2004;147:91–9.
[26]. Prondzinsky R, Lemm H, Swyter M, et al. Intra-aortic balloon counterpulsation in patients with acute myocardial infarction complicated by cardiogenic shock: the prospective, randomized IABP SHOCK trial for attenuation of multiorgan dysfunction syndrome*. Crit Care Med. 2010;38:152–60.
[27]. Pang S, Miao G, Zhao X. Effects and safety of extracorporeal membrane oxygenation in the treatment of patients with ST-segment elevation myocardial infarction and cardiogenic shock: a systematic review and meta-analysis. Front Cardiovasc Med. 2022;9:963002.
[28]. Kohsaka S, Menon V, Lowe AM, et al. Systemic inflammatory response syndrome after acute myocardial infarction complicated by cardiogenic shock. Arch Intern Med. 2005;165:1643–50.
[29]. 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;97:610–6.
[30]. Lüsebrink E, Binzenhöfer L, Kellnar A, et al. Venting during venoarterial extracorporeal membrane oxygenation. Clin Res Cardiol. 2022. Online ahead of print.
[31]. Reynolds HR, Hochman JS. Cardiogenic shock: current concepts and improving outcomes. Circulation. 2008;117:686–97.
[32]. Bréchot N, Demondion P, Santi F, et al. Intra-aortic balloon pump protects against hydrostatic pulmonary oedema during peripheral venoarterial-extracorporeal membrane oxygenation. Eur Heart J Acute Cardiovasc Care. 2018;7:62–9.
[33]. Onorati F, Esposito A, Comi MC, et al. Intra-aortic balloon pump-induced pulsatile flow reduces coagulative and fibrinolytic response to cardiopulmonary bypass. Artif Organs. 2008;32:433–41.
[34]. Onorati F, Presta P, Fuiano G, et al. A randomized trial of pulsatile perfusion using an intra-aortic balloon pump versus nonpulsatile perfusion on short-term changes in kidney function during cardiopulmonary bypass during myocardial reperfusion. Am J Kidney Dis. 2007;50:229–38.
[35]. Onorati F, Cristodoro L, Bilotta M, et al. Intraaortic balloon pumping during cardioplegic arrest preserves lung function in patients with chronic obstructive pulmonary disease. Ann Thorac Surg. 2006;82:35–43.
[36]. Sauren LDC, Reesink KD, Selder JL, et al. The acute effect of intra-aortic balloon counterpulsation during extracorporeal life support: an experimental study. Artif Organs. 2007;31:31–8.
[37]. Nishi T, Ishii M, Tsujita K, et al. Outcomes of venoarterial extracorporeal membrane oxygenation plus intra-aortic balloon pumping for treatment of acute myocardial infarction complicated by cardiogenic shock. J Am Heart Assoc. 2022;11:e023713.
[38]. Ouazani Chahdi H, Berbach L, Boivin-Proulx LA, et al. Percutaneous mechanical circulatory support in post–myocardial infarction cardiogenic shock: a systematic review and meta-analysis. Can J Cardiol. 2022;38:1525–38.
[39]. Schrage B, Becher PM, Bernhardt A, et al. Left ventricular unloading is associated with lower mortality in patients with cardiogenic shock treated with venoarterial extracorporeal membrane oxygenation: results from an international, multicenter cohort study. Circulation. 2020;142:2095–106.
[40]. Thiele H, Freund A, Gimenez MR, et al. Extracorporeal life support in patients with acute myocardial infarction complicated by cardiogenic shock – design and rationale of the ECLS-SHOCK trial. Am Heart J. 2021;234:1–11.
[41]. Banning AS, Adriaenssens T, Berry C, et al. Veno-arterial extracorporeal membrane oxygenation (ECMO) in patients with cardiogenic shock: rationale and design of the randomised, multicentre, open-label euro SHOCK trial. EuroIntervention. 2021;16:e1227–36.
[42]. Shaefi S, O’Gara B, Kociol RD, et al. Effect of cardiogenic shock hospital volume on mortality in patients with cardiogenic shock. J Am Heart Assoc. 2015;4:e001462.
[43]. Tie EN, Dinh D, Chan W, et al. Trends in intra-aortic balloon pump use in cardiogenic shock in the post-SHOCK II trial era. Am J Cardiol. 2023;191:125–32.
