Extracorporeal Membrane Oxygenation Use in Acute Myocardial Infarction in the United States, 2000-2014

Abstract

Background:

Extracorporeal membrane oxygenation (ECMO) is increasingly used in acute myocardial infarction (AMI); however there are limited large-scale national data.

Methods:

Using the National Inpatient Sample database from 2000–2014, a retrospective cohort of AMI utilizing ECMO was identified. Use of percutaneous coronary intervention (PCI), intra-aortic balloon pump (IABP) and percutaneous left ventricular assist device (pLVAD) was also identified in this population. Outcomes of interest included temporal trends in utilization of ECMO alone and with concomitant procedures (PCI, IABP, and pLVAD), in-hospital mortality and resource utilization.

Results:

In ~9 million AMI admissions, ECMO was used in 2,962 (<0.01%) and implanted a median of one day after admission. ECMO was used in 0.5% and 0.3% AMI admissions complicated by cardiogenic shock and cardiac arrest respectively. ECMO was used more commonly in admissions that were younger, non-white, and with less comorbidity. ECMO use was 11-times higher in 2014 as compared to 2000 (odds ratio 11.37 [95% confidence interval 7.20–17.97]). Same-day PCI was performed in 23.1%; IABP/pLVAD was used in 57.9%, of which 30.3% were placed concomitantly. In-hospital mortality with ECMO was 59.2% overall, but decreased from 100% (2000) to 45.1% (2014). Durable LVAD and cardiac transplantation were performed in 11.7% as an exit strategy. Of the hospital survivors, 40.8% were discharged to skilled nursing facilities. Older age, male sex, non-white race, and lower socio-economic status, were independently associated with higher in-hospital mortality with ECMO use.

Conclusions:

In AMI admissions, a steady increase was noted in the utilization of ECMO alone and with concomitant procedures (PCI, IABP, pLVAD). In-hospital mortality remained high in AMI admissions treated with ECMO.

Concomitant cardiac arrest and/or cardiogenic shock (CS) are seen in about 3–10% of all patients with acute myocardial infarction (AMI), and is associated with nearly 30–50% in-hospital mortality.^{1, 2} These patients typically present with attendant respiratory failure, hemodynamic collapse and electrical instability, all of which may result in systemic hypoperfusion and multi-organ failure.^3–5 Furthermore, AMI patients with CS or cardiac arrest patients remain at high risk of clinical decompensation during coronary angiography or percutaneous coronary intervention (PCI) due to ongoing hemodynamic compromise. In contemporary practice, the advent of advanced percutaneous mechanical circulatory support (MCS) devices has made it possible to offer hemodynamic support for the management of AMI.^{6, 7} Though there are prior data on the intra-aortic balloon pump (IABP) and emerging data on the percutaneous left ventricular assist devices (pLVAD);^{8, 9} there are limited large-scale data on the use of extra-corporeal membrane oxygenation (ECMO) in AMI.¹⁰ Using a 15-year nationally-representative database we sought to assess the use, temporal trends, timing, and outcomes of ECMO in AMI. We also sought to evaluate the use of ECMO across demographic categories and the use of concomitant cardiac interventions with ECMO in AMI.

MATERIAL AND METHODS

Study Population, Variables and Outcomes

The data are publicly available with the Agency for Healthcare Research and Quality for other researchers to replicate the results of this study.¹¹ Institutional Review Board approval was not sought due to the publicly available nature of the de-identified data. The Healthcare Quality and Utilization Project-National (Nationwide) Inpatient Sample (HCUP-NIS) is the largest all-payer database of hospital inpatient stays in the United States and contains discharge data from a 20% stratified sample of community hospitals.¹¹ Information regarding each discharge includes patient demographics, primary payer, hospital characteristics, principal diagnosis, up to 24 secondary diagnoses, and procedural diagnoses. The HCUP-NIS does not capture individual patients, but captures all information for a given hospitalization.

Using the HCUP-NIS data from 2000–2014, a retrospective cohort study of admissions with AMI in the primary diagnosis field was identified (ST-elevation myocardial infarction – ICD-9CM 410.1x-410.6x, 410.8x, 410.9x, non ST-elevation myocardial infarction – ICD-9CM 410.70–410.79).^{2–5, 12, 13} CS and cardiac arrest were identified using ICD-9CM codes of 785.51, 427.5 and 427.41. CS was defined as shock resulting from diminution of cardiac output in cardiovascular disease, shock resulting from primary pump failure of the heart, as in myocardial infarction, severe cardiomyopathy, or mechanical obstruction or compression of the heart or shock resulting from the failure of the heart to maintain adequate output.¹⁴ Validation studies have shown a specificity of 99.3%, a sensitivity of 59.8%, a positive predictive value of 78.8%, and a negative predictive value of 98.1% for the ICD-9-CM code 785.51 to identify CS.¹⁴ Use and timing of ECMO (ICD-9CM 39.65), IABP (ICD-9CM 37.61) and pLVAD (ICD-9CM 37.68) were identified using procedure codes and procedure day for timing.^{4, 6, 7, 12} Durable LVAD (ICD-9CM 37.66) and cardiac transplant (ICD-9CM 37.5, 37.51, 33.6) were identified for all admissions.^{6, 7} Consistent with prior literature from the HCUP-NIS, we used the listed procedure day to define concomitant ECMO placement with PCI, IABP and pLVAD placement.¹⁵ The Deyo’s modification of Charlson Comorbidity Index was used to identify the burden of co-morbid diseases (Supplementary Table 1).¹⁶ Patient and hospital demographic characteristics, cardiac and non-cardiac diagnostic and therapeutic procedures and outcomes were identified similar to prior literature.^{2–7, 12, 13, 17, 18} The primary outcome was the incidence and temporal trends for ECMO use in AMI. Secondary outcomes included in-hospital mortality, resource utilization, the use of a second temporary MCS device, timing of ECMO with relation to PCI and other MCS devices, concomitant placement of ECMO and IABP/pLVAD and the use of durable LVAD or cardiac transplant as an exit strategy. Though the Agency for Healthcare Research and Quality has released the HCUP-NIS data till 2016, due to the change in coding practices from ICD-9CM to ICD-10CM in October 2015 we sought to restrict the data to 2014. The HCUP-NIS from 2015 and 2016 databases lack the Clinical Classification System for ICD-9CM codes used in the study. Furthermore, the ICD-10-CM codes lack extensive validation studies unlike the ICD-9CM codes and therefore need further evaluation prior to incorporation into temporal analyses.¹⁹

Statistical Analysis

As recommended by HCUP-NIS, survey procedures using discharge weights provided with the HCUP-NIS database were used to generate national estimates. Using the trend weights provided by the HCUP-NIS, samples from 2000–2011 were re-weighted to adjust for the 2012 HCUP-NIS re-design.²⁰ Chi-square and t-tests were used to compare categorical and continuous variables respectively. The inherent restrictions of the HCUP-NIS database related to research design, data interpretation, and data analysis were reviewed and addressed.²⁰ Odds ratio (OR) with 95% confidence interval (CI) were used to represent univariable and multivariable analyses. Adjusted temporal trends were assessed using multivariable logistic regression analysis incorporating age, gender, race, admission year, primary payer status, socio-economic stratum, hospital characteristics, comorbidities, acute organ failure, coronary angiography, PCI, second temporary MCS use, invasive hemodynamic monitoring, invasive ventilation and hemodialysis use for ECMO use (referent year 2000) and in-hospital mortality (referent year 2014). A multivariable logistic regression incorporating age, sex, race, primary payer status, socio-economic stratum, comorbidity, hospital location, AMI type, acute organ failure, cardiac and non-cardiac procedures was performed for predictors of in-hospital mortality in AMI admissions with the use of ECMO. For the multivariable modeling, regression analysis with purposeful selection of statistically (p<0.20) and clinically relevant variables was conducted. Two-tailed p<0.05 was considered statistically significant. All statistical analyses were performed using SPSS version 25.0 (IBM Corp, Armonk NY).

RESULTS

There were an estimated 9,747,034 (95% CI 9,736,108–9,757,960) admissions for AMI between January 1, 2000 and December 31, 2014, of which ECMO was used in 2,962 (<0.01%). In these ~9 million AMI admissions, CS and cardiac arrest were noted in 4.6% and 4.5%, respectively. ECMO was used in 0.5% and 0.3% of AMI admissions complicated by CS and cardiac arrest respectively. The cohort receiving ECMO was on average younger, more frequently non-White, covered by private insurance, presenting with ST-elevation, on average with lower comorbidity and higher organ failure (Table 1). AMI admissions with ECMO use more often had cardiac arrest, CS, and less use of coronary angiography and PCI (Table 1). In the admissions receiving ECMO support, 86.7% and 92.2% had been admitted to large and urban teaching hospitals, respectively. The 15-year unadjusted and adjusted trends (referent year 2000) in the use of ECMO in AMI are presented in Figure 1A and 1B, respectively. There was an 11.37-times higher odds of ECMO use in 2014 as compared to 2000 after adjusting for patient and hospital characteristics (95% CI 7.20–17.97). ECMO use stratified by patient characteristics is presented in Supplementary Figure 1. Per 100,000 AMI admissions, the Northeast region had highest use of ECMO (60.3), followed by the Midwest (28.4), the West (21.7) and the Southern (18.7) regions of the United States.

Table 1.

Baseline characteristics of cohorts with and without ECMO use in acute myocardial infarction

Characteristic		ECMO (N = 2,962)	No ECMO (N = 9,744,072)	P
Age (years)		59.0 ± 10.9	67.7 ± 14.3	<0.001
Female sex		26.1	40.1	<0.001
Race	White	70.6	78.0	<0.001
	Black	8.8	9.2
	Other	20.6	12.8
Primary payer	Medicare	33.4	57.8	<0.001
	Medicaid	10.6	5.6
	Private	45.2	28.3
	Uninsured	6.9	5.2
	No charge	1.6	0.5
Quartile of median household income for zip code	0–25th	23.9	23.1	0.24
	26th–50th	28.3	27.2
	51st–75th	23.6	24.7
	75th–100th	24.1	25.0
Charlson Comorbidity Index	0–3	43.1	35.9	<0.001
	4–6	46.8	46.3
	≥ 7	10.2	17.8
AMI type	ST-segment elevation	67.3	39.0	<0.001
	Non-ST-segment elevation	32.7	61.0
Acute organ failure	Respiratory	56.5	7.6	<0.001
	Renal	62.1	1.0	<0.001
	Hepatic	30.2	0.8	<0.001
	Neurologic	27.3	2.8	<0.001
Out-of-hospital cardiac arrest		47.5	4.5	<0.001
Cardiogenic shock		80.6	4.5	<0.001
Coronary angiography		50.9	62.0	<0.001
Percutaneous coronary intervention		39.3	40.2	0.31
Invasive hemodynamic monitoring		22.2	4.9	<0.001
Coronary artery bypass grafting		37.1	9.3	<0.001

Represented as percentage or mean ± standard deviation

Abbreviations: ECMO: extracorporeal membrane oxygenation

Figure 1. Temporal trends of ECMO use in AMI.

A: Unadjusted 15-year temporal trends of ECMO use in all AMI, AMI with CS and AMI with CA. B: adjusted odds ratio for temporal trends of ECMO use in AMI (referent year 2000)*, the Y-axis is presented in logarithmic scale; all p<0.001 for trend

*Adjusted for: age, sex, race, primary payer, socio-economic status, hospital location/teaching status, hospital bedsize, hospital region, comorbidity, acute organ dysfunction, cardiogenic shock, cardiac arrest, use of coronary angiography, percutaneous coronary intervention, invasive hemodynamic assessment, invasive mechanical ventilation and hemodialysis

Abbreviations: AMI: acute myocardial infarction; CA: cardiac arrest; CS: cardiogenic shock; ECMO: extra-corporeal membrane oxygenation

ECMO was implanted at a median of one day (interquartile range 0–3 days) after admission. Same-day PCI was performed in 686 (23.1%) admissions and a second temporary MCS device (IABP/pLVAD) was used in 1,716 (57.9%) admissions. Temporal trends in the use of same-day PCI, placement of concomitant IABP/pLVAD with ECMO, and use of a second temporary MCS device during the index admission are presented in Figure 2A. The timing of the second temporary MCS device relative to ECMO was available in 1,579/1,716 (92.0%) admissions (Figure 2B). Of these admissions, 30.3% received concomitant IABP/pLVAD with ECMO with an increasing trend during the study period (Figure 2A and 2B). Vascular complications, lower limb amputations, and use of blood transfusions were noted in 4.7%, 1.1% and 25.9%, respectively, of the AMI admissions with ECMO use.

Figure 2. Concomitant cardiac procedures in AMI admissions on ECMO.

A: Fifteen-year temporal trends of same-day PCI, concomitant use of ECMO and IABP/pLVAD and use of a second MCS device (IABP/pLVAD) in AMI admissions receiving ECMO (represented as total admissions); all p<0.001 for trend. B: Timing of IABP/pLVAD in AMI admissions receiving ECMO; negative values denote IABP/pLVAD use before ECMO and positive values denote IABP/pLVAD use after ECMO (represented as percentage of total AMI admissions); all p<0.001 for trend

Abbreviations: AMI: acute myocardial infarction; ECMO: extra-corporeal membrane oxygenation; IABP: intra-aortic balloon pump; MCS: mechanical circulatory support; PCI: percutaneous coronary intervention; pLVAD: percutaneous left ventricular assist device

In-hospital mortality occurred in 1,752 (59.2%) admissions treated with ECMO during the study period. The 15-year temporal trends for mortality are presented in Figure 3A. In an adjusted analysis, with 2014 as the referent year, there was a 15-times lower odds of in-hospital mortality in this population during the study period (Figure 3B). The use of durable LVAD, cardiac transplant and discharge disposition of the AMI admissions receiving ECMO is presented in Figure 4. In a multivariable logistic regression analysis for in-hospital mortality, older age, male sex, non-white race, lower socio-economic status, acute hepatic failure and use of organ support systems were independently predictive of higher in-hospital mortality in AMI admissions receiving ECMO support (Figure 5). Age >80 years was the strongest predictor of in-hospital mortality (OR 4.14 [95% CI 1.89–9.07]; p<0.001) in the population receiving ECMO in AMI. There was a statistically significant interaction between PCI and concomitant CS (p=0.02 for interaction). Use of PCI was associated with lower in-hospital mortality in the overall cohort (OR 0.72 [95% CI 0.59–0.87); p<0.001) and when stratified by presence (OR 0.74 [95% CI 0.59–0.91]; p=0.005) or absence (OR 0.37 [95% CI 0.20–0.70]; p=0.002) of CS.

Figure 3. Temporal trends of in-hospital mortality in AMI admissions receiving ECMO therapy.

A: Unadjusted 15-year temporal trends of in-hospital mortality in AMI admissions receiving ECMO therapy represented as percentage. B: Adjusted odds ratio for temporal trends of in-hospital mortality of ECMO use in AMI (referent year 2014)*, the Y-axis is presented in logarithmic scale; all p<0.001 for trend

*Adjusted for: age, sex, race, primary payer, socio-economic status, hospital location/teaching status, hospital bedsize, hospital region, comorbidity, acute organ dysfunction, cardiogenic shock, cardiac arrest, use of a second temporary mechanical circulatory support device, coronary angiography, percutaneous coronary intervention, invasive hemodynamic assessment, invasive mechanical ventilation and hemodialysis

The odds ratio for year 2000 in Figure 3B could not be calculated due to 100% mortality (infinite odds ratio)

Abbreviations: AMI: acute myocardial infarction; ECMO: extra-corporeal membrane oxygenation

Figure 4. In-hospital mortality, disposition and use of permanent exit-strategies in AMI admissions receiving ECMO therapy.

Abbreviations: AMI: acute myocardial infarction; ECMO: extra-corporeal membrane oxygenation; HHC: home health care; HTx: heart transplantation; LVAD: left ventricular assist device; SNF: skilled nursing facilities

Figure 5. Adjusted odds ratio for in-hospital mortality in AMI admissions receiving ECMO therapy.

Odds ratio with 95% confidence interval using multivariable hierarchical regression analysis for prediction of in-hospital mortality; for cohorts with multiple categories (i.e. age, sex, race, primary payer, CCI, SES, hospital region, AMI type) the first category was used as reference category for calculating odds ratios. The X-axis is presented in logarithmic scale.

Abbreviations: AMI: acute myocardial infarction; CCI: Charlson comorbidity index; ECMO: extracorporeal membrane oxygenation; IHDM: invasive hemodynamic monitoring; IMV: invasive mechanical ventilation; MCS: mechanical circulatory support; NSTEMI: non-ST-elevation myocardial infarction; OHCA: out-of-hospital cardiac arrest; PCI: percutaneous coronary intervention; SES: socio-economic status; STEMI: ST-elevation myocardial infarction

DISCUSSION

In this nationally-representative study of ~9 million AMI admissions, ECMO use was noted in 2,962 (<0.01%). There was an 11-times higher odds of use of ECMO for AMI in 2014 compared to 2000. There was a temporal increase in same-day PCI, use of concomitant IABP/pLVAD with ECMO and the use of multiple MCS devices during a single admission. During this 15-year period, the overall in-hospital mortality for AMI admissions receiving ECMO remained high at 59.2%, although there was a significant decrease from 100% mortality in 2000 to 45.1% mortality in 2014. Older age, male sex, non-white race, lower socio-economic status, and acute non-cardiac organ failure were independently associated with in-hospital mortality. Of the hospital survivors, 41.4% of those receiving ECMO were discharged to skilled nursing facilities suggestive of high post-hospitalization resource utilization.

In patients with AMI complicated by CS and/or cardiac arrest, ECMO offers significant advantages over other MCS devices by virtue of respiratory and biventricular support, higher flow rates, ease of deployment and relatively low costs.^21–23 Prior work from large databases has either focused on unselected ECMO use or use in unselected CS admissions.^{10, 24} Our data are consistent with prior studies that demonstrate an increased adoption of ECMO for AMI-CS.^{12, 22, 25, 26} In addition to CS, in AMI ECMO is often used to facilitate cardiopulmonary resuscitation (eCPR).²⁷ Consistent with the Extracorporeal Life Support Organization registry that reported a 10-times increase in the adoption of eCPR between 2003 and 2014, we note a steady increase in ECMO adoption in AMI with cardiac arrest.²⁸ In AMI with refractory cardiac arrest, ECMO can be used to facilitate PCI despite not attaining return of spontaneous circulation.^{27, 29} However, given the limitations of this administrative database, we are unable to discern the timing of ECMO placement in cardiac arrest and if it was used per an eCPR protocol.

The improvements of ECMO technology have resulted in its increasing adoption to facilitate PCI in AMI with CS or cardiac arrest.^25–27 Consistent with these data, we demonstrate a rapid increase in ECMO-facilitated PCI in this nationally representative population of AMI. The lower rates of angiography and PCI in this study are consistent with prior real-world literature that reflects reluctance to perform angiography in higher risk cohorts despite robust guideline recommendations.³⁰ Furthermore, use of MCS in AMI needs careful assessment of clinical and hemodynamic profiles prior to selection of MCS device. Despite strong guideline recommendations, angiography and PCI may frequently be deferred in patients with non-ST-elevation AMI placed on ECMO to evaluate for organ recovery or neurological function.³¹ The higher rates of coronary artery bypass grafting in this population may have resulted in consequent lower rates of PCI. The heterogeneity of this population and associated complex decision-making often requires multi-disciplinary care teams to achieve optimal outcomes.^{5, 32}

In this study we noted that nearly 57.9% of admissions had a second temporary MCS device besides ECMO which is increasingly being used in modern practice.^{21, 22} In a prior meta-analysis, we noted the ECMO+IABP strategy to have lower mortality compared to ECMO alone only in AMI-CS.²² Preliminary data with unloading the left ventricle with a pLVAD has shown greater myocardial recovery and faster weaning of ECMO and is worthy of further study.²¹ Despite well-established randomized trial data on the use of IABP in cardiogenic shock, there are limited adequately powered studies evaluating the VA-ECMO with equal rigor.³³ Furthermore, all studies on combination therapy (ECMO+IABP/pLVAD) are observational in nature, and therefore are prone to confounding by indication and heterogeneity.^{21, 22} Though hemodynamically, the combination therapy of ECMO with pLVAD has a strong physiological rationale, further studies are needed prior to adoption in routine clinical practice. Our study reflects the contemporary practice in the United States. Despite the lack of strong evidence from randomized trials, 57.9% of our population received combination therapy.

It is important to note that the overall survival in these observational studies was low at <50%. This is consistent with the cumulative in-hospital mortality in this study was 59.2%, with 45.1% mortality in 2014. The decrease in mortality could potentially be due to improvements in the delivery of AMI care in patients with concomitant cardiac arrest and CS.² Our data are consistent with unselected ECMO patients from a population database in Taiwan that demonstrated 30% hospital survival.³⁴ Chang et al. noted only 24% of the total population to be alive at one-year emphasizing the high post-hospitalization mortality and morbidity as noted indirectly in this study.³⁴ This study identifies significant health care disparities in in-hospital mortality in AMI admissions with ECMO use. Admissions of non-white race, without insurance, and lower socio-economic status were independently associated with higher in-hospital mortality, similar to disparities in patients with other acute cardiovascular conditions.³⁵ Sex-based disparities were also noted in this study given the high male preponderance in ECMO use that is consistent with prior literature.³⁶ We noted a lower mean age in ECMO recipients compared to the general AMI population. This could potentially be explained by the higher comorbidity burden and the lack of a well-defined exit strategy in the older population. Importantly, older patients are often not candidates for cardiac transplantation or durable LVAD which was used in 11.7% of this population.⁶ The reasons for these patient-specific disparities are incompletely understood and are likely due to differences in cultural and religious beliefs, treatment preferences and incomplete knowledge.⁵ Lastly, it is important to note that the use of ECMO is associated with significant financial incentives.³⁷ The insertion and maintenance of ECMO is has high hospitalization costs and resource utilization.^{38, 39} Recent policy changes from the Center for Medicare and Medicaid Services that treat peripheral and central ECMO differently were met with strong resistance needing reversal of the Diagnosis Relation Groups policy. In contemporary practice, peripheral ECMO is treated on par with central ECMO for reimbursement, which is significantly higher in costs than other percutaneous MCS devices.⁴⁰ In light of these issues, careful assessment of the indications, continued need and management of ECMO is needed in AMI patients.

Limitations

This study has several limitations, some of which are inherent to the analysis of a large administrative database. Given the administrative nature of this database, we cannot accurately distinguish type-1 from type-2 AMIs. However, this study included admissions with a primary diagnosis of AMI (i.e. the reason most likely for the admission), and therefore is less likely to include type-2 AMIs which often have an alternate primary diagnosis. Nonetheless, this could be an important issue where additional research is necessary. Concomitant use of other MCS or PCI with ECMO was defined as those performed on the same procedure day. However since further granularity in timing is unavailable and AMI often evolves dynamically, it is possible this study included admissions that received ECMO for cardiopulmonary resuscitation, worsening CS or post-PCI complications. However, since the HCUP-NIS does not record duration of support or explantation of organ support, this study did not evaluate duration of ECMO support alone or in-combination with another MCS device. The lack of angiographic data, such PCI location, lesion classification, presence of multi-vessel disease, and revascularization failure, that may significantly influence outcomes, were not available in this database. Though in patients with AMI with or without CS, veno-arterial ECMO is the most commonly used configuration of ECMO support, it is possible that the ICD-9CM code used in this study may refer to the veno-venous ECMO, used for respiratory support. Due to the limitations of the administrative coding, this database is unable to distinguish veno-arterial and veno-venous configurations. Despite these limitations, this study addresses an important knowledge gap highlighting the contemporary use of ECMO in AMI.

CONCLUSIONS

In this study, we observed a rapid increase in the use of ECMO in AMI over a 15-year study period. There was a temporal increase in same day PCI, use of concomitant IABP and pLVAD with ECMO and the use of multiple MCS devices in a single admission. Fifty nine percent of the admissions that received ECMO for AMI died during the hospitalization and those discharged had high post-hospitalization resource utilization. The use of ECMO in AMI needs further careful study in randomized studies prior to widespread adoption given the high costs and mortality in this sick population.

SHORT COMMENTARY.

What is new?

In the largest study of extracorporeal membrane oxygenation use in acute myocardial infarction, we note an 11-times higher odds increase in extracorporeal membrane oxygenation use between 2000 and 2014.

The in-hospital mortality (59.2%) continues to be high in this population.

What are the clinical implications?

The use of extracorporeal membrane oxygenation in acute myocardial infarction needs further careful study in randomized studies prior to widespread adoption given the high costs and mortality in this sick population.

ABBREVIATIONS

AMI

acute myocardial infarction

CS

cardiogenic shock

ECMO

extracorporeal membrane oxygenation

eCPR

extracorporeal membrane oxygenation-facilitated cardiopulmonary resuscitation

HCUP

Healthcare Cost and Utilization Project

IABP

intra-aortic balloon pump

ICD-9CM

International Classification of Diseases-9 Clinical Modification

MCS

mechanical circulatory support

NIS

National/Nationwide Inpatient Sample

PCI

percutaneous coronary intervention

pLVAD

percutaneous left ventricular assist device