Abstract
Background
In patients with cardiogenic shock (CS), acute myocardial infarction (AMI) is the most common cause, and a percutaneous microaxial ventricular assist device (Impella, Abiomed, Danvers, MA) is a choice for temporary mechanical circulatory support. However, data are limited on complications and outcomes of Impella treatment in patients with CS with and without AMI.
Methods and Results
Using nationwide prospective registry data in Japan, we included a total of 2047 patients with CS in whom the Impella devices were successfully placed between February 2020 and December 2021. Patients were divided into 2 groups according to the primary indication for the Impella use: AMI versus non‐AMI. The primary end point was a composite of in‐hospital all‐cause death and major complications. Of the 2047 patients, the Impella was indicated for AMI in 1337 (65.3%). In the group without AMI, myocarditis was the leading cause of CS. Patients with AMI‐CS were older and more likely to have cardiovascular risk factors than those with non‐AMI‐CS. The rates of in‐hospital mortality (46.0% versus 43.9%, P=0.38) and major complications (35.2% versus 34.7%, P=0.85) were similar between the 2 groups. Overall, multivariable analysis identified older age, higher body mass index, previous transient ischemic attack or stroke, out‐of‐hospital cardiac arrest, and the Impella 5.0 as factors significantly associated with the primary end point.
Conclusions
The use of Impella in patients with and without AMI was related to similar clinical outcomes with high mortality and complication rates. Further studies are needed to identify patients who may benefit from the Impella devices in CS.
Registration
URL: https://www.umin.ac.jp/english. Identifier: UMIN000033603.
Keywords: acute myocardial infarction, Impella, mechanical circulatory support, outcome
Subject Categories: Big Data and Data Standards, Cardiopulmonary Resuscitation and Emergency Cardiac Care, Myocardial Infarction
Nonstandard Abbreviations and Acronyms
- CS
cardiogenic shock
- J‐PVAD
Japan Registry for Percutaneous Ventricular Assist Device
Clinical Perspective.
What Is New?
Acute myocardial infarction accounted for approximately two‐thirds of causes of cardiogenic shock in a nationwide Impella registry in Japan.
Although considerable differences were found in baseline characteristics between patients with and without acute myocardial infarction, the rates of in‐hospital complications and mortality were high with no significant between‐group differences in the Impella treatment.
What Are the Clinical Implications?
Older age, higher body mass index, previous transient ischemic attack or stroke, out‐of‐hospital cardiac arrest, and the Impella 5.0 were identified as factors significantly associated with in‐hospital complications and mortality.
These factors may be considered when estimating patient risks in the Impella treatment in a setting of cardiogenic shock.
Cardiogenic shock (CS) is characterized by a low cardiac output state, which results in inadequate organ perfusion and tissue hypoxia. Despite recent advances in therapeutic strategies, the prognosis of CS remains poor with in‐hospital mortality ranging from 40% to 50%. 1 In the setting of CS, mechanical circulatory support devices allow immediate restoration of tissue perfusion, among which the Impella (Abiomed, Danvers, MA), a percutaneous ventricular assist device, has been increasingly used in clinical practice. 2 , 3 The Impella is a microaxial, continuous‐flow pump to draw blood from inside the ventricle and to eject it into the ascending aorta (or pulmonary artery), providing hemodynamic stability, 4 , 5 although the clinical evidence to support the use of Impella is insufficient, and the device may be associated with an increased risk of complications such as major bleeding, stroke, and all‐cause mortality. 3 , 6 , 7 Acute myocardial infarction (AMI) represents the most common cause of CS, and other forms of CS include fulminant myocarditis, acute or chronic heart failure, valvular heart disease, and ventricular arrythmia. 8 Patients with CS and no AMI differ from those with AMI‐CS in baseline characteristics, diagnostic process, timely therapeutic intervention (eg, coronary revascularization), and potentially prognosis. 9 However, whether the presence of AMI has an impact on adverse events and outcomes in the Impella treatment for CS is largely unknown. For instance, although a single‐center, retrospective study compared patient characteristics and outcomes and showed similar prognosis between Impella‐treated patients with and without AMI, 10 the findings have not been established yet. We sought to evaluate complications and clinical outcomes related to the Impella between patients with and without AMI in a setting of CS, using large‐scale contemporary nationwide registry data.
Methods
The data that support the findings of this study may be available upon the permission from the Council for Clinical Use of Ventricular Assist Device Related Academic Societies, Impella committee, through the corresponding author.
Data Source
The Japan Registry for Percutaneous Ventricular Assist Device (J‐PVAD) is a nationwide ongoing, prospective registry that records Impella treatment, including successful and unsuccessful placement, in Japan. 11 The J‐PVAD is led by the Council for Clinical Use of Ventricular Assist Device Related Academic Societies, Impella committee, consisting of 10 academic organizations in order to disseminate the device safely and effectively. The J‐PVAD registry study investigators are listed in Table S1. The J‐PVAD prospectively collects clinical variables and outcome data during hospitalization and is registered in the University Hospital Medical Information Network (UMIN000033603). Central monitoring was performed in the present J‐PAD data set. The present study complies with the Declaration of Helsinki and was approved by Osaka University Ethics Committee (approval number: 17232, approval date: November 21, 2017). Written informed consent was waived because of complete data anonymization and the observational nature of this study.
Study Population
Between February 2020 and December 2021, the Impella devices were attempted in 2143 patients with CS at 179 hospitals across Japan. The Impella devices were successfully placed in 2121 (99.0%) patients. Major exclusion criteria included unsuccessful Impella insertion, age <20 years, missing data on in‐hospital survival, and postcardiotomy CS (Figure 1). Thus, a total of 2047 patients with CS receiving Impella devices were included in the present study. CS was defined as systolic blood pressure ≤90 mm Hg for ≥30 minutes and cardiac index <2.2 L/min per m2 if measured, or the need for intravenous inotropes and vasopressors to maintain the blood pressure and cardiac index, accompanied by end‐organ hypoperfusion. When a patient had received other mechanical circulatory support devices (eg, extracorporeal membrane oxygenation and intra‐aortic balloon pumping) before the Impella placement, the patient was considered to have CS, irrespective of the indications of other mechanical circulatory support devices. During the study period, the Impella 2.5, CP, and 5.0 were available for left ventricular support in patients with drug‐resistant acute heart failure including CS. In Japan, the Impella 2.5 and 5.0 were approved in September 2016 and commercially available with reimbursement in September 2017. The Impella CP became commercially available in July 2019. 11 Only 1 primary indication for the Impella use at the time of placement was selected from the following diseases by treating physicians at each institution in the J‐PVAD: AMI (with or without mechanical complications), myocarditis, nonischemic cardiomyopathy, chronic coronary syndrome (ie, ischemic cardiomyopathy), ventricular arrhythmia, valvular heart disease, and others. The choice of primary indication was left at the discretion of local physicians. In the present study, patients were divided into 2 groups according to the presence or absence of AMI.
Figure 1. Study flow chart.
Impella (Abiomed, Danvers, MA) is an intravascular microaxial left ventricular assist device. AMI indicates acute myocardial infarction; and CS, cardiogenic shock.
Definitions and End Points
The primary interest of this study included death and major complications during the hospitalization according to the indication of Impella (ie, AMI versus non‐AMI). Impella‐related major complications were defined as a composite of major bleeding requiring blood transfusion, access site‐related infection, vascular injury, lower limb ischemia, ischemic and bleeding stroke, and myocardial infarction. Access site‐related infection was defined as clinical evidence of infection such as pain, fever, purulent discharge, blood test results (eg, elevated white blood cell count), and positive culture and the requirement of (nonprophylactic) antimicrobial drugs. Vascular injury included false aneurysm; arteriovenous fistula; vascular thrombus; dissection, perforation, and rupture of the vessel; and vascular narrowing related to the Impella devices. Lower limb ischemia was defined as clinical evidence of lower limb hypoperfusion requiring interventional treatment. Stroke was defined as clinically significant neurological deficit or apparent infarction or intracranial hemorrhage. Myocardial infarction was defined as a clinically significant increase in cardiac biomarkers (eg, troponin) and significant change on electrocardiogram. 11 Even in patients with AMI, a new increase in cardiac biomarkers and electrocardiographic changes that were different from those that would be normally expected as a part of trend for patients suffering AMI was considered as a complication of MI. In the present study, major complications related to the first and subsequent (another Impella re‐placed after the first Impella had been removed) Impella devices were acknowledged as end points, although the association of each (the first and subsequent) Impella with individual complications was not specified. The timing of Impella treatment (ie, days from hospital admission to the Impella placement) was also evaluated. The outcomes were further assessed among causes of non‐AMI‐CS.
Statistical Analysis
Statistical analysis was performed using SAS software version 9.3 (SAS Institute, Cary, NC). Data are expressed as mean±SD, median (interquartile range), or frequency (percentage). Continuous variables were compared using Student's t test, Mann–Whitney U test, ANOVA, or Kruskal–Wallis test, as appropriate. Categorical data were assessed with the Fisher's exact test. In‐hospital mortality and major complications between patients with and without AMI were evaluated. Using logistic regression analysis, baseline characteristics (Table 1) were included into the multivariable model to identify factors associated with the primary end point of composite of in‐hospital mortality and complications. Multivariable analyses to identify factors associated with the primary end point were also performed in the subgroups of patients with and without AMI and those receiving the Impella CP. A value of P<0.05 was considered statistically significant.
Table 1.
Baseline Characteristics
| Variable | All | AMI‐CS | Non‐AMI‐CS | P value |
|---|---|---|---|---|
| (n=2047) | (n=1337) | (n=710) | ||
| Age, y | 66.6±14.1 | 69.1±12.2 | 61.8±16.0 | <0.001 |
| Men | 1582/2047 (77.3%) | 1072/1337 (80.2%) | 510/710 (71.8%) | <0.001 |
| Body mass index | 23.7±4.3 | 23.8±4.1 | 23.4±4.6 | 0.07 |
| Hypertension | 1166/1933 (60.3%) | 829/1252 (66.2%) | 337/681 (49.5%) | <0.001 |
| Diabetes | 770/1959 (39.3%) | 541/1273 (42.5%) | 229/686 (33.4%) | <0.001 |
| Dyslipidemia | 869/1933 (45.0%) | 618/1252 (49.4%) | 251/681 (36.9%) | <0.001 |
| Hemodialysis | 96/2047 (4.7%) | 35/1337 (2.6%) | 61/710 (8.6%) | <0.001 |
| Previous transient ischemic attack or stroke | 151/1960 (7.7%) | 104/1271 (8.2%) | 47/689 (6.8%) | 0.29 |
| Previous heart failure | 484/1944 (24.9%) | 190/1263 (15.0%) | 294/681 (43.2%) | <0.001 |
| Previous atrial fibrillation | 184/2047 (9.0%) | 69/1337 (5.2%) | 115/710 (16.2%) | <0.001 |
| Previous ventricular arrhythmia | 96/2047 (4.7%) | 34/1337 (2.5%) | 62/710 (8.7%) | <0.001 |
| Clinical presentation | ||||
| ST‐segment–elevation myocardial infarction | 1070/2047 (52.3%) | 1070/1337 (80.0%) | 0 (0%) | <0.001 |
| Out‐of‐hospital cardiac arrest | 458/2038 (22.5%) | 308/1329 (23.2%) | 150/709 (21.2%) | 0.32 |
| Vasopressor/inotrope use | 1596/2047 (78.0%) | 1013/1337 (75.8%) | 583/710 (82.1%) | <0.001 |
Continuous variables are expressed as mean±SD. AMI indicates acute myocardial infarction; and CS, cardiogenic shock.
Results
Of the 2047 patients with CS, the Impella was indicated for AMI in 1337 (65.3%), whereas the remaining 710 (34.7%) patients had a different indication of Impella use (Figure 1). A median Impella volume per hospital was 9 [5, 16] during the 2‐year study period, ranging from 1 to 55. In the present study population, the Impella 2.5, CP, and 5.0 were used in 5.8%, 90.7%, and 3.6%, respectively. Table 1 lists baseline characteristics. Patients with AMI‐CS, in whom ST‐segment–elevation MI accounted for 80.0%, were older and more likely to have cardiovascular risk factors than those with non‐AMI‐CS (Table 1). On the other hand, previous heart failure, atrial fibrillation, and ventricular arrhythmia were more frequently observed in the group with non‐AMI‐CS than in the group with AMI‐CS (Table 1). Of note, the rate of out‐of‐hospital cardiac arrest did not differ significantly between the 2 groups. Myocarditis was the most frequent indication of the Impella use in the group with non‐AMI‐CS, followed by nonischemic and ischemic cardiomyopathy (Table 2). Table 3 lists baseline characteristics among patients with non‐AMI‐CS, indicating that patients with myocarditis were younger and had fewer cardiovascular risk factors.
Table 2.
Indications and Mode of MCS Device Use
| Variable | All | AMI‐CS | Non‐AMI‐CS | P value |
|---|---|---|---|---|
| (n=2047) | (n=1337) | (n=710) | ||
| Primary indications for Impella | ||||
| AMI | 1337 (65.3%) | 1337 (100%) | 0 (0%) | |
| Mechanical complications of MI | 68 (3.3%) | 68 (5.1%) | 0 (0%) | |
| Myocarditis | 194 (9.5%) | 0 (0%) | 194 (27.3%) | |
| Nonischemic cardiomyopathy | 130 (6.4%) | 0 (0%) | 130 (18.3%) | |
| Ischemic cardiomyopathy | 128 (6.3%) | 0 (0%) | 128 (18.0%) | |
| Ventricular arrhythmia | 126 (6.2%) | 0 (0%) | 126 (17.8%) | |
| Valvular heart disease | 53 (2.6%) | 0 (0%) | 53 (7.5%) | |
| Others | 79 (3.9%) | 0 (0%) | 79 (11.1%) | |
| Type of Impella | 0.002 | |||
| Impella 2.5 | 116 (5.8%) | 76 (5.7%) | 40 (5.6%) | |
| Impella CP | 1857 (90.7%) | 1227 (91.8%) | 630 (88.7%) | |
| Impella 5.0 | 74 (3.6%) | 34 (2.5%) | 40 (5.6%) | |
| Access site of Impella | <0.001 | |||
| Femoral artery | 1951 (95.3%) | 1291 (96.6%) | 660 (93.0%) | |
| Subclavian artery | 90 (4.4%) | 42 (3.1%) | 48 (6.8%) | |
| Others | 6 (0.3%) | 4 (0.3%) | 2 (0.3%) | |
| Other MCS devices before Impella | ||||
| Intra‐aortic balloon pumping | 205 (10.0%) | 141 (10.6%) | 64 (9.0%) | 0.28 |
| Extracorporeal membrane oxygenation | 644 (31.5%) | 378 (28.3%) | 266 (37.5%) | <0.001 |
| Additional Impella | 130 (6.4%) | 81 (6.1%) | 49 (6.9%) | 0.45 |
AMI indicates acute myocardial infarction; CS, cardiogenic shock; MCS, mechanical circulatory support; and MI, myocardial infarction.
Table 3.
Baseline Characteristics in Patients With Non‐AMI‐CS
| Variable | Myocarditis | Non‐ICM | ICM | VA | VHD | P value |
|---|---|---|---|---|---|---|
| (n=194) | (n=130) | (n=128) | (n=126) | (n=53) | ||
| Age, y | 54.9±17.4 | 57.7±14.6 | 70.1±10.8 | 61.1±14.7 | 73.9±12.3 | <0.001 |
| Men | 119/194 (61.3%) | 93/130 (71.5%) | 108/128 (84.4%) | 102/126 (81.0%) | 33/53 (62.3%) | <0.001 |
| Body mass index | 23.1±4.1 | 23.5±4.7 | 23.2±4.7 | 24.4±5.3 | 21.7±3.6 | 0.009 |
| Hypertension | 62/189 (32.8%) | 45/126 (35.7%) | 93/125 (74.4%) | 62/118 (52.5%) | 31/50 (62.0%) | <0.001 |
| Diabetes | 19/189 (10.1%) | 33/126 (26.2%) | 79/126 (62.7%) | 49/120 (40.8%) | 17/51 (33.3%) | <0.001 |
| Dyslipidemia | 33/187 (17.7%) | 37/127 (29.1%) | 80/125 (64.0%) | 55/118 (46.6%) | 22/51 (43.1%) | <0.001 |
| Hemodialysis | 2/194 (1.0%) | 9/130 (6.9%) | 26/128 (20.3%) | 8/126 (6.4%) | 9/53 (17.0%) | <0.001 |
| Previous transient ischemic attack or stroke | 4/188 (2.1%) | 10/129 (7.8%) | 13/126 (10.3%) | 6/122 (4.9%) | 5/51 (9.8%) | 0.01 |
| Previous heart failure | 29/185 (15.7%) | 83/129 (64.3%) | 67/124 (54.0%) | 48/119 (40.3%) | 36/50 (72.0%) | <0.001 |
| Previous atrial fibrillation | 11/194 (5.7%) | 34/130 (26.2%) | 19/128 (14.8%) | 20/126 (15.9%) | 17/53 (32.1%) | <0.001 |
| Previous ventricular arrhythmia | 10/194 (5.2%) | 12/130 (9.2%) | 12/128 (9.4%) | 19/126 (15.1%) | 2/53 (3.8%) | 0.03 |
| Out‐of‐hospital cardiac arrest | 26/193 (13.5%) | 22/130 (16.9%) | 18/128 (14.1%) | 69/126 (54.8%) | 3/53 (5.7%) | <0.001 |
AMI indicates acute myocardial infarction; CS, cardiogenic shock; ICM, ischemic cardiomyopathy; VA, ventricular arrhythmia; and VHD, valvular heart disease.
Table 4 shows in‐hospital complications and outcomes. Patients received the Impella within 24 hours after hospital arrival in 78.6% and 51.9% in the groups with AMI‐CS and non‐AMI‐CS (P<0.001). Days from hospital admission to the first Impella and length of hospital stay from the Impella to discharge were significantly longer in patients with non‐AMI‐CS than those with AMI‐CS (Table 4). Overall, the rates of major complications and in‐hospital mortality was high (35.0% and 45.3%), with no significant between‐group differences (Table 4). Among patients with non‐AMI‐CS, the incidence of major bleeding events was highest and in‐hospital mortality was lowest in patients with myocarditis (Table 5). In‐hospital complications and outcomes in patients with unsuccessful Impella placement are also shown in Table S2. Multivariable analysis identified older age, higher body mass index, previous transient ischemic attack or stroke, out‐of‐hospital cardiac arrest, and the use of Impella 5.0 (as compared with the Impella 2.5 or CP) as factors significantly associated with the primary end point of composite of in‐hospital mortality and complications, whereas the presence of AMI did not have a significant effect on the end point (Table 6). When dividing patients into the 2 groups based on the presence or absence of AMI as a cause of CS, overall results of the multivariable analysis were unchanged, although the significant effect of previous transient ischemic attack or stroke on the primary end point may be more evident in patients with AMI‐CS (Table S3) than those with non‐AMI‐CS (Table S4). In addition, when focusing only on patients receiving the Impella CP, the results were consistent (Table S5). Clinical characteristics and outcomes according to the type of first Impella are listed in Table S6. Overall results are summarized in Figure 2.
Table 4.
In‐Hospital Complications and Outcomes
| Variable | All | AMI‐CS | Non‐AMI‐CS | P value |
|---|---|---|---|---|
| (n=2047) | (n=1337) | (n=710) | ||
| Days from admission to Impella | 0 [0–1] | 0 [0–0] | 0 [0–3] | <0.001 |
| Within 24 h* | 1412/2037 (69.3%) | 1044/1328 (78.6%) | 368/709 (51.9%) | <0.001 |
| Days from Impella to discharge† | 34 [21–60] | 32 [21–56] | 39 [24–62] | <0.001 |
| Death and major complications | 1287 (62.9%) | 847 (63.4%) | 440 (62.0%) | 0.56 |
| Major complications | 716 (35.0%) | 470 (35.2%) | 246 (34.7%) | 0.85 |
| Major bleedings‡ | 548 (26.8%) | 365 (27.3%) | 183 (25.8%) | 0.46 |
| Access site‐related infection | 49 (2.4%) | 36 (2.7%) | 13 (1.8%) | 0.29 |
| Vascular injury | 37 (1.8%) | 25 (1.9%) | 12 (1.7%) | 0.86 |
| Lower limb ischemia | 102 (5.0%) | 65 (4.9%) | 37 (5.2%) | 0.75 |
| Stroke | 132 (6.4%) | 87 (6.5%) | 45 (6.3%) | 0.92 |
| Ischemic stroke | 88 (4.3%) | 59 (4.4%) | 29 (4.1%) | 0.82 |
| Bleeding stroke | 50 (2.4%) | 32 (2.4%) | 18 (2.5%) | 0.88 |
| Myocardial infarction | 4 (0.2%) | 2 (0.2%) | 2 (0.3%) | 0.61 |
| Death | 927 (45.3%) | 615 (46.0%) | 312 (43.9%) | 0.38 |
AMI indicates acute myocardial infarction; and CS, cardiogenic shock.
Data are missing in 10 patients.
Days from the first Impella to discharge only in patients who survived to discharge.
Major bleeding events requiring transfusion. Continuous variables are expressed as median [interquartile range].
Table 5.
In‐Hospital Complications and Outcomes in Patients With Non‐AMI‐CS
| Variable | Myocarditis | Non‐ICM | ICM | VA | VHD | P value |
|---|---|---|---|---|---|---|
| (n=194) | (n=130) | (n=128) | (n=126) | (n=53) | ||
| Days from admission to Impella | 0 (0–1) | 1 (0–3) | 1 (0–6) | 0 (0–2) | 2 (0–9) | <0.001 |
| Within 24 h* | 127/194 (65.5%) | 62/129 (48.1%) | 50/128 (39.1%) | 75/126 (59.5%) | 21/53 (39.6%) | <0.001 |
| Days from Impella to discharge† | 32 (24–60) | 50 (29–81) | 38 (18–61) | 37 (25–57) | 41 (25–68) | 0.06 |
| Death and major complications | 115 (59.3%) | 81 (62.3%) | 76 (59.4%) | 83 (65.9%) | 33 (62.3%) | 0.79 |
| Major complications | 78 (40.2%) | 48 (36.9%) | 38 (29.7%) | 43 (34.1%) | 13 (24.5%) | 0.15 |
| Major bleedings‡ | 65 (33.5%) | 33 (25.4%) | 25 (19.5%) | 33 (26.2%) | 8 (15.1%) | 0.02 |
| Access site‐related infection | 7 (3.6%) | 3 (2.3%) | 3 (2.3%) | 0 (0%) | 0 (0%) | 0.17 |
| Vascular injury | 6 (3.1%) | 3 (2.3%) | 2 (1.6%) | 1 (0.8%) | 0 (0%) | 0.59 |
| Lower limb ischemia | 9 (4.6%) | 7 (5.4%) | 9 (7.0%) | 5 (4.0%) | 3 (5.7%) | 0.83 |
| Stroke | 13 (6.7%) | 13 (10.0%) | 6 (4.7%) | 5 (4.0%) | 3 (5.7%) | 0.35 |
| Ischemic stroke | 8 (4.1%) | 6 (4.6%) | 4 (3.1%) | 4 (3.2%) | 2 (3.8%) | 0.97 |
| Bleeding stroke | 6 (3.1%) | 7 (5.4%) | 2 (1.6%) | 1 (0.8%) | 1 (1.9%) | 0.21 |
| Myocardial infarction | 0 (0%) | 0 (0%) | 0 (0%) | 1 (0.8%) | 0 (0%) | 0.28 |
| Death | 64 (33.0%) | 59 (45.4%) | 55 (43.0%) | 65 (51.6%) | 29 (54.7%) | 0.004 |
AMI indicates acute myocardial infarction; CS, cardiogenic shock; ICM, ischemic cardiomyopathy; VA, ventricular arrhythmia; and VHD, valvular heart disease.
Data are missing in 1 patient.
Days from the first Impella to discharge only in patients who survived to discharge.
Major bleeding events requiring transfusion. Continuous variables are expressed as median (interquartile range).
Table 6.
Factors Associated With the Primary End Point
| Variable | Univariable | Multivariable | ||
|---|---|---|---|---|
| OR (95% CI) | P value | OR (95% CI) | P value | |
| Age, y | 1.01 (1.01–1.02) | <0.001 | 1.02 (1.01–1.03) | <0.001 |
| Men | 0.85 (0.68–1.05) | 0.14 | 0.84 (0.66–1.08) | 0.17 |
| Body mass index | 1.05 (1.03–1.08) | <0.001 | 1.06 (1.03–1.09) | <0.001 |
| Hypertension | 1.31 (1.08–1.57) | 0.005 | 1.04 (0.83–1.31) | 0.74 |
| Diabetes | 1.06 (0.88–1.28) | 0.52 | 0.92 (0.74–1.14) | 0.42 |
| Dyslipidemia | 0.93 (0.78–1.12) | 0.46 | 0.82 (0.66–1.01) | 0.06 |
| Hemodialysis | 1.72 (1.08–2.73) | 0.02 | 1.88 (1.13–3.13) | 0.02 |
| Previous transient ischemic attack or stroke | 1.91 (1.31–2.80) | <0.001 | 1.81 (1.19–2.75) | 0.006 |
| Previous heart failure | 1.21 (0.97–1.49) | 0.09 | 1.10 (0.85–1.43) | 0.46 |
| Previous atrial fibrillation | 0.98 (0.72–1.34) | 0.91 | 0.92 (0.65–1.30) | 0.63 |
| Previous ventricular arrhythmia | 1.25 (0.81–1.94) | 0.32 | 1.26 (0.77–2.08) | 0.36 |
| Acute myocardial infarction‐cardiogenic shock | 1.06 (0.88–1.28) | 0.54 | 1.00 (0.79–1.26) | 0.99 |
| Out‐of‐hospital cardiac arrestpresentation | 1.57 (1.25–1.97) | <0.001 | 1.65 (1.27–2.14) | <0.001 |
| Impella 5.0 (vs 2.5 or CP) | 3.16 (1.69–5.89) | <0.001 | 3.23 (1.65–6.31) | <0.001 |
| Vasopressor/inotrope use at baseline | 1.36 (1.10–1.68) | 0.005 | 1.21 (0.96–1.52) | 0.11 |
OR indicates odds ratio.
Figure 2. Schematic figure of the present study results.
AMI indicates acute myocardial infarction; CS, cardiogenic shock; ICM, ischemic cardiomyopathy; VA, ventricular arrhythmia; and VHD, valvular heart disease.
Discussion
The present analysis from the nationwide registry data in Japan demonstrated that AMI was the most common cause of CS in approximately two‐thirds of patients, whereas myocarditis and nonischemic and ischemic cardiomyopathy were major causes in the group with non‐AMI‐CS. This study confirmed the high in‐hospital mortality of 45.3% in patients with CS treated with the Impella. Although baseline characteristics were largely different between patients with and without AMI, in‐hospital mortality and major complication risks were high with no significant between‐group differences. Among patients with non‐AMI‐CS, the rate of complications and mortality differed significantly according to the underlying causes. Although multivariable analysis identified several clinical factors related to in‐hospital events, the presence of AMI was not a significant factor. To the best of our knowledge, this is the first study to evaluate the impact of indications of Impella, the presence or absence of AMI, on complications and mortality during hospitalization in patients with CS.
Impella in Cardiogenic Shock
CS is the most severe manifestation of acute heart failure, in which AMI is the most common cause. 4 , 8 The Impella devices theoretically and potentially result in a better hemodynamic support and clinical outcomes in patients with CS. 4 , 8 To date, however, the evidence supporting the use of Impella in a setting of CS derives predominantly from observational studies and industry‐sponsored registries, 12 most of which have focused on patients with AMI. Several large‐scale observational studies have suggested that the Impella use may be associated with an increase in major bleeding events, stroke, mortality, and medical cost. 3 , 6 Thus, more data are needed to define and characterize the appropriate role of Impella in specific patient populations.
In the Italian Impella registry, older age, higher body mass index, and previous transient ischemic attack or stroke were identified as factors associated with all‐cause mortality as were shown in the present study. 12 Another registry in the United States also found that age, body mass index, renal insufficiency, anoxic brain injury, and cardiac arrest were associated with in‐hospital mortality. 13 A recent meta‐analysis focusing on clinical outcomes in patients with CS treated with the Impella demonstrated that in‐hospital or 30‐day mortality was 46.5% (95% CI, 42.7%–50.3%) in 63 observational studies (n=3896). 14 In this meta‐analysis, the incidence of access site bleeding, limb ischemia, and stroke were reported to be 25.8%, 6.1%, and 5.5%, respectively, with a large heterogeneity. 14 The present nationwide study with a large sample (n=2047) provides robust data on the incidence of in‐hospital complications and mortality, in which the risks may be higher in the elderly and obese and in those with previous cerebrovascular events and a critically ill presentation. In addition to the high in‐hospital mortality, our results reinforce the high complication rate related to the Impella devices. Interestingly, the Impella 5.0 was associated with an increased risk of death and complications in the present study, possibly due to the large profile, but further investigations are warranted given the small sample size (n=74). Because >90% of patients in the present analysis received the Impella CP, our results may be particularly applicable to this patient population.
Presence of Acute Myocardial Infarction in Cardiogenic Shock
AMI is the leading cause of CS and thus, data are limited in patients with CS and non‐AMI causes. In the setting of CS, patients with and without AMI differ in baseline characteristics, diagnostic process, and timely therapeutic intervention. For instance, patients with AMI are likely to be older and have more comorbidities as shown in the present study, although they have an immediate therapeutic option such as primary percutaneous coronary intervention. 15 , 16 , 17 Differences in a diagnostic and therapeutic time frame may lead to different outcomes in patients with and without AMI. In this study, indeed, patients with AMI received the Impella earlier after hospital arrival than those without, suggesting that the devices were placed in an emergency situation in AMI than in non‐AMI. However, the risks of major complications are similar between the 2 groups. In a single‐center, retrospective study (n=106), baseline characteristics and outcomes were compared between patients with CS with and without AMI, in which age was higher, the prevalence of atrial fibrillation was lower, and renal function was impaired in the group with AMI. 10 In the single‐center investigation, mortality at 30 days was nonsignificantly higher in the group with AMI than in the group with non‐AMI (54.3% versus 44.4%, P=0.34). 10 In the aforementioned meta‐analysis, short‐term mortality was also numerically higher in patients with AMI‐CS, without statistical significance (52.1% versus 42.0%, P=0.09). 14 In the present study, in‐hospital mortality was similar between the 2 groups (46.0% versus 43.9%, P=0.38). Therefore, it is likely that once a patient develops severe CS that needs the Impella treatment, complications and outcomes may be similar regardless of the indications (ie, AMI versus non‐AMI). Of note, as seen in the comparison between AMI and non‐AMI, baseline characteristics and outcomes varied widely among non‐AMI‐CS causes in this study, with lowest in‐hospital mortality in patients with myocarditis (33.0%) and highest in those with valvular heart disease (54.7%). These findings suggest that future studies investigating the Impella treatment across a spectrum of CS should delve into the causes of non‐AMI‐CS.
Study Limitations
Some limitations to the present study should be acknowledged. Despite the large sample size, this was an observational study with the relatively small number of patients included per hospital and thus, our findings should be considered hypothesis‐generating. Because only patients receiving the Impella devices were included, data on potential candidates with CS for the Impella devices were lacking in the present data set. The presence of AMI was likely to be associated with the Impella use and outcomes, and the conditioning on the Impella use can lead to misspecification in the association between the presence of AMI and outcomes. However, the results may be informative for future studies in this field. Some laboratory data potentially related to the outcomes including a lactate level and left ventricular ejection fraction were not available. 18 In the group with non‐AMI‐CS, detailed information in patients with other causes than myocarditis, nonischemic cardiomyopathy, ischemic cardiomyopathy, ventricular arrhythmia, and valvular heart disease was lacking. Because the Impella CP was predominantly used in this study population, analysis on other types of Impella may be underpowered. In Japan, the Impella use has been limited only in a setting of drug‐resistant acute heart failure including CS, with no indications of right ventricular support and high‐risk percutaneous coronary intervention. In the J‐PVAD registry, the preceding use of other mechanical circulatory support devices (eg, extracorporeal membrane oxygenation and intra‐aortic balloon pumping) is a part of definitions of CS, although this may include some patients without traditional CS criteria of end‐organ hypoperfusion. Furthermore, the timing of extracorporeal membrane oxygenation use relative to the Impella placement was not clear. Additionally, although only 1 primary indication for the Impella use was selected by local physicians in the J‐PVAD, it is likely that some patients had multiple indications. The previous investigation from the J‐PVAD included data between October 2017 and January 2020 under a postmarketing surveillance, 11 which was a different data set from that in the present study (ie, an academia‐led prospective registry investigation). Thus, the combined data set was not available. Even though the central monitoring was done in the J‐PVAD, the accuracy of coding the definition of CS was not validated by audits. All complications were locally adjudicated by treating physicians.
Conclusions
The present Japanese nationwide Impella registry showed that AMI accounted for approximately two‐thirds of CS causes, and considerable differences were found in baseline characteristics between patients with and without AMI. The rates of in‐hospital complications and mortality were high with no significant between‐group differences. Adequately powered observational and randomized studies are warranted to identify patients with and without AMI who may benefit from the Impella devices in a setting of CS.
Source of Funding
This work was supported by grants from Takeda Science Foundation.
Disclosures
Dr Koichi Toda reports speaker fees from Abiomed Japan. Dr Yoshio Kobayashi reports honoraria from Abiomed Japan. The remaining authors have no disclosures to report.
Supporting information
Tables S1–S6.
This article was sent to Yen‐Hung Lin, MD, PhD, Associate Editor, for review by expert referees, editorial decision, and final disposition.
Supplemental Material is available at https://www.ahajournals.org/doi/suppl/10.1161/JAHA.123.030819
For Sources of Funding and Disclosures, see page 9.
References
- 1. Naidu SS, Baran DA, Jentzer JC, Hollenberg SM, van Diepen S, Basir MB, Grines CL, Diercks DB, Hall S, Kapur NK, et al. SCAI SHOCK stage classification expert consensus update: a review and incorporation of validation studies: this statement was endorsed by the American College of Cardiology (ACC), American College of Emergency Physicians (ACEP), American Heart Association (AHA), European Society of Cardiology (ESC) Association for Acute Cardiovascular Care (ACVC), International Society for Heart and Lung Transplantation (ISHLT), Society of Critical Care Medicine (SCCM), and Society of Thoracic Surgeons (STS) in December 2021. J Am Coll Cardiol. 2022;79:933–946. doi: 10.1016/j.jacc.2022.01.018 [DOI] [PubMed] [Google Scholar]
- 2. Khan MZ, Munir MB, Khan MU, Osman M, Agrawal P, Syed M, Ghaffar YA, Alharbi A, Khan SU, Balla S. Trends, outcomes, and predictors of revascularization in cardiogenic shock. Am J Cardiol. 2020;125:328–335. doi: 10.1016/j.amjcard.2019.10.040 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Amin AP, Spertus JA, Curtis JP, Desai N, Masoudi FA, Bach RG, McNeely C, Al‐Badarin F, House JA, Kulkarni H, et al. The evolving landscape of Impella use in the United States among patients undergoing percutaneous coronary intervention with mechanical circulatory support. Circulation. 2020;141:273–284. doi: 10.1161/CIRCULATIONAHA.119.044007 [DOI] [PubMed] [Google Scholar]
- 4. Balthazar T, Vandenbriele C, Verbrugge FH, Den Uil C, Engström A, Janssens S, Rex S, Meyns B, Van Mieghem N, Price S, et al. Managing patients with short‐term mechanical circulatory support: JACC review topic of the week. J Am Coll Cardiol. 2021;77:1243–1256. doi: 10.1016/j.jacc.2020.12.054 [DOI] [PubMed] [Google Scholar]
- 5. Nix C, Ishikawa K, Meyns B, Yasuda S, Adriaenssens T, Barth S, Zayat R, Leprince P, Lebreton G. Comparison of hemodynamic support by Impella vs. peripheral extra‐corporeal membrane oxygenation: a porcine model of acute myocardial infarction. Front Cardiovasc Med. 2020;7:99. doi: 10.3389/fcvm.2020.00099 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Miller PE, Bromfield SG, Ma Q, Crawford G, Whitney J, DeVries A, Desai NR. Clinical outcomes and cost associated with an intravascular microaxial left ventricular assist device vs intra‐aortic balloon pump in patients presenting with acute myocardial infarction complicated by cardiogenic shock. JAMA Intern Med. 2022;182:926–933. doi: 10.1001/jamainternmed.2022.2735 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Ito F, Kimura M, Hyogo M, Yamashita E, Shuntoh K, Takahashi A, Sawada T. Aortic saddle embolism just after Impella CP removal in a patient with alcoholic cardiomyopathy. Cardiovasc Interv Ther. 2023;38:139–140. doi: 10.1007/s12928-022-00876-w [DOI] [PubMed] [Google Scholar]
- 8. Attinger‐Toller A, Bossard M, Cioffi GM, Tersalvi G, Madanchi M, Bloch A, Kobza R, Cuculi F. Ventricular unloading using the ImpellaTM device in cardiogenic shock. Front Cardiovasc Med. 2022;9:856870. doi: 10.3389/fcvm.2022.856870 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Ouweneel DM, Henriques JP. Percutaneous cardiac support devices for cardiogenic shock: current indications and recommendations. Heart. 2012;98:1246–1254. doi: 10.1136/heartjnl-2012-301963 [DOI] [PubMed] [Google Scholar]
- 10. Haurand JM, Haberkorn S, Haschemi J, Oehler D, Aubin H, Akhyari P, Boeken U, Kelm M, Westenfeld R, Horn P. Outcome of patients with non‐ischaemic cardiogenic shock supported by percutaneous left ventricular assist device. ESC Heart Fail. 2021;8:3594–3602. doi: 10.1002/ehf2.13546 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Toda K, Ako J, Hirayama A, Kinugawa K, Kobayashi Y, Ono M, Nishimura T, Sato N, Shindo T, Takayama M, et al. Three‐year experience of catheter‐based micro‐axial left ventricular assist device, Impella, in Japanese patients: the first interim analysis of Japan registry for percutaneous ventricular assist device (J‐PVAD). J Artif Organs. 2023;26:17–23. doi: 10.1007/s10047-022-01328-1 [DOI] [PubMed] [Google Scholar]
- 12. Chieffo A, Ancona MB, Burzotta F, Pazzanese V, Briguori C, Trani C, Piva T, De Marco F, Di Biasi M, Pagnotta P, et al. Observational multicentre registry of patients treated with IMPella mechanical circulatory support device in ITaly: the IMP‐IT registry. EuroIntervention. 2020;15:e1343–e1350. doi: 10.4244/EIJ-D-19-00428 [DOI] [PubMed] [Google Scholar]
- 13. Singh H, Mehta RH, O'Neill W, Kapur NK, Lalonde T, Ohman M, Ghiu I, Chen‐Hsing Y, Dutcheshen K, Schreiber T, et al. Clinical features and outcomes in patients with cardiogenic shock complicating acute myocardial infarction: early vs recent experience with Impella. Am Heart J. 2021;238:66–74. doi: 10.1016/j.ahj.2021.04.002 [DOI] [PubMed] [Google Scholar]
- 14. Tan SR, Low CJW, Ng WL, Ling RR, Tan CS, Lim SL, Cherian R, Lin W, Shekar K, Mitra S, et al. Microaxial left ventricular assist device in cardiogenic shock: a systematic review and meta‐analysis. Life (Basel). 2022;12:1629. doi: 10.3390/life12101629 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Hochman JS, Sleeper LA, Webb JG, Dzavik V, Buller CE, Aylward P, Col J, White HD; SHOCK Investigators . Early revascularization and long‐term survival in cardiogenic shock complicating acute myocardial infarction. JAMA. 2006;295:2511–2515. doi: 10.1001/jama.295.21.2511 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Ozaki Y, Hara H, Onuma Y, Katagiri Y, Amano T, Kobayashi Y, Muramatsu T, Ishii H, Kozuma K, Tanaka N, et al. CVIT expert consensus document on primary percutaneous coronary intervention (PCI) for acute myocardial infarction (AMI) update 2022. Cardiovasc Interv Ther. 2022;37:1–34. doi: 10.1007/s12928-021-00829-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Saito Y, Kobayashi Y. Complete revascularization in acute myocardial infarction: a clinical review. Cardiovasc Interv Ther. 2023;38:177–186. doi: 10.1007/s12928-022-00907-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Sato T, Saito Y, Suzuki S, Matsumoto T, Yamashita D, Saito K, Wakabayashi S, Kitahara H, Sano K, Kobayashi Y. Prognostic factors of in‐hospital mortality in patients with acute myocardial infarction complicated by cardiogenic shock. Life (Basel). 2022;12:1672. doi: 10.3390/life12101672 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Tables S1–S6.