- Email: [email protected]
Association between percutaneous hemodynamic support device and survival from cardiac arrest in the state of Michigan
YAJEM-57039; No of Pages 4 American Journal of Emergency Medicine xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
American Journal of Emergency Medicine journal homepage: www.elsevier.com/locate/ajem
Association between percutaneous hemodynamic support device and survival from cardiac arrest in the state of Michigan☆ Andrew Pressman, MD a, Kelly N. Sawyer, MD, MS b, William Devlin, MD c, Robert Swor, DO d,⁎ a
Department of Emergency Medicine, Beaumont Hospital-Troy, Troy, MI, United States Department of Emergency Medicine, University of Pittsburgh, School of Medicine, Pittsburgh, PA, United States Division of Cardiology, Beaumont Hospital-Troy, Troy, MI, United States d Department of Emergency Medicine, Oakland University William Beaumont School of Medicine, Rochester, MI, United States b c
a r t i c l e
i n f o
Article history: Received 16 July 2017 Received in revised form 12 October 2017 Accepted 12 October 2017 Available online xxxx Keywords: Cardiac arrest Outcome Ventricular assist device
a b s t r a c t Introduction: The role of circulatory support in the post-cardiac arrest period remains controversial. Our objective was to investigate the association between treatment with a percutaneous hemodynamic support device and outcome after admission for cardiac arrest. Methods: We performed a retrospective study of adult patients with admission diagnosis of cardiac arrest or ventricular fibrillation (VF) from the Michigan Inpatient Database, treated between July 1, 2010, and June 30, 2013. Patient demographics, clinical characteristics, treatments, and disposition were electronically abstracted based on ICD-9 codes at the hospital level. Mixed-effects logistic regression models were fit to test the effect of percutaneous hemodynamic support device defined as either percutaneous left ventricular assist device (pLVAD) or intra-aortic balloon pump (IABP) on survival. These models controlled for age, sex, VF, myocardial infarction (MI), and cardiogenic shock with hospital modeled as a random effect. Results: A total of 103 hospitals contributed 4393 patients for analysis, predominately male (58.8%) with a mean age of 64.1 years (SD 15.5). On univariate analysis, younger age, male sex, VF as the initial rhythm, acute MI, percutaneous coronary intervention, percutaneous hemodynamic support device, and absence of cardiogenic shock were associated with survival to discharge (each p b 0.001). Mixed-effects logistic regressions revealed use of percutaneous hemodynamic support device was significantly associated with survival among all patients (OR 1.8 (1.28–2.54)), and especially in those with acute MI (OR 1.95 (1.31–2.93)) or cardiogenic shock (OR 1.96 (1.29–2.98)). Conclusion: Treatment with percutaneous hemodynamic support device in the post-arrest period may provide left ventricular support and improve outcome. © 2017 Elsevier Inc. All rights reserved.
1. Introduction The benefit of circulatory support in the post-cardiac arrest period remains controversial. Historically, intra-aortic balloon pumps (IABP) have been used as the primary circulatory support device for myocardial infarction (MI) complicated by cardiogenic shock [1]. However, a contemporary prospective, randomized controlled trial has revealed no mortality benefit with this therapy at one month or one year for patients in cardiogenic shock after MI [2]. While use of IABP for cardiogenic shock in the setting of acute MI was once a level 1b recommendation in America, it has recently been downgraded to a level 2b
☆ Sources of support: This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. ⁎ Corresponding author at: William Beaumont Hospital, 3601 W. 13 Mile Rd, Royal Oak, MI 48073, United States. E-mail address: [email protected] (R. Swor).
recommendation [1,3]. Advances in technology have allowed for mechanical support using percutaneous left ventricular assist devices (pLVAD). While numerous studies have shown improved hemodynamics with these devices, no randomized controlled trial has been performed to assess overall mortality benefit [4,5,6]. Furthermore, studies assessing both IABP and pLVAD use in patients resuscitated from outof-hospital cardiac arrest (OHCA) are limited given the difficulty in studying this patient population [7]. Data from IABP use to treat post-cardiac arrest cardiogenic shock are more readily available than data from other pLVADs. Two studies retrospectively studying the subject have found that IABP use was strongly associated with neurologically intact survival [8,9]. However, a further prospective study found IABP use associated with increased mortality in MI patients complicated by cardiac arrest [10]. There have been promising porcine studies evaluating the use of pLVAD during cardiac arrest resuscitation, doubling the rate of return of spontaneous circulation (ROSC) compared to standard care [11]. In human subjects, studies
https://doi.org/10.1016/j.ajem.2017.10.036 0735-6757/© 2017 Elsevier Inc. All rights reserved.
Please cite this article as: Pressman A, et al, Association between percutaneous hemodynamic support device and survival from cardiac arrest in the state of Michigan, American Journal of Emergency Medicine (2017), https://doi.org/10.1016/j.ajem.2017.10.036
2
A. Pressman et al. / American Journal of Emergency Medicine xxx (2017) xxx–xxx
assessing pLVAD use in cardiac arrest have been limited to case series. These series showed that pLVAD use was associated with a sicker patient populations but found no difference in neurologically intact survival [12,13]. It is clear that little literature exists studying pLVAD and IABP use in post cardiac arrest patients across larger populations. The objective of this study was to investigate whether, in a statewide cohort of post cardiac arrest patients, use of percutaneous hemodynamic support device was associated with survival. 2. Methods We performed a retrospective study of adult (age ≥ 18 years) patients with an admission diagnosis of cardiac arrest (ICD-CM 427.5) or ventricular fibrillation (ICD-9 CM 427.41) from the Michigan Inpatient Database (MIDB), treated between July 1, 2010, and June 30, 2013. This definition of cardiac arrest includes both OHCA patients that survived to hospital admission, as well as those who arrested in the ED and survived to admission. No patient with CPR in progress is included in this cohort of patients. The MIDB is run by the Michigan Health & Hospital Association and has been collecting data on patients requiring inpatient admission since the 1980s. It currently collects data on 1.3 million admitted patients annually from N 100 hospitals across Michigan [14,15]. This database collects information on all patients who survived to admission into an acute care hospital, including patient demographics, receiving facility, discharge diagnoses, procedures performed, and patient outcomes. Patient outcomes recorded included survival to hospital discharge and discharge destination (home, short term rehabilitation, skilled nursing facility, hospital). For this study we defined surviving to hospital discharge as any patient discharged from the hospital to home or another facility other than hospice. Data on patient demographics (age, sex), clinical characteristics (arrest rhythm of ventricular fibrillation (VF), cardiogenic shock (ICD-9 CM 98551), myocardial infarction (ICD-9 CM, 41000–41010)), procedures performed (e.g. percutaneous coronary intervention (PCI), percutaneous hemodynamic support device placement), and outcome (survived to discharge) were electronically abstracted based on ICD-9 codes at the hospital level. Percutaneous hemodynamic support device was defined as either IABP or pLVAD (Impella 2.5). Duplicate cases, patients b 18 years old, and cases with missing data were excluded. Univariate relationships with outcome were compared using a Welch's two-sample t-test for continuous data and a chi-squared test of independence for categorical data. Mixed-effects logistic regression models were fit to test the effect of percutaneous hemodynamic support device on outcome, with hospital modeled as a random effect. In the mixed-effects logistic regression, we controlled for those variables known to have an effect on survival from our univariate analysis (age, sex, VF, MI, and cardiogenic shock), but also attempted to control for specific hospital effect on survival. This study was approved by the William Beaumont Hospital Institutional Review Board. All data cleaning and analysis was performed using R (version 3.2.1) and RStudio (version 0.99.467). 3. Results A total of 103 hospitals in Michigan contributed 4393 patients for analysis with a mean age of 64.1 years (SD 15.5) and 41% being female. From the total, 1422 patients (32.4%) had a diagnosis of VF, 1174 (26.7%) with MI, and 779 (17.7%) cardiogenic shock. Only 197 (4.5%) patients received a percutaneous hemodynamic support device, of which most were IABP (182 (4.1%)). These were placed in 35 different hospitals, with only 20 hospitals placing them more than once a year. Table 1 and Fig. 1 highlight the results of the univariate analyses, revealing that younger age, male sex, VF initial rhythm, acute MI, PCI, percutaneous hemodynamic support device, and absence of cardiogenic shock were associated with survival to discharge. This effect was most notable
Table 1 Univariate analysis of factors associated with survival to hospital discharge. Patient variables
Survival (n = 1688)
Deaths (n = 2705)
p-Value
Age, mean (SD) Male sex, n(%) VF initial rhythm, n(%) Acute MI, n(%) Cardiogenic shock, n(%) Received PCI, n(%) Hemodynamic support device, n(%)
61.5 (15.3) 1083 (64.2) 821 (48.6) 510 (30.2) 220 (13.0) 278 (16.5) 103 (6.1)
65.7 (15.5) 1499 (55.4) 601 (22.2) 664 (24.5) 559 (20.7) 122 (4.5) 94 (3.5)
b0.001a b0.001b b0.001b b0.001b b0.001b b0.001b b0.001b
VF-ventricular fibrillation; MI-myocardial infarction; PCI-percutaneous intervention. a Compared using Welch's two-sample t-test. b Compared using chi-squared test of independence.
for an initial rhythm of VF (OR, 3.3, 95% CI, 2.9–3.8) and receiving PCI (OR, 4.2, 95% CI, 3.3–5.2) but was still significant for those receiving a percutaneous hemodynamic support device (OR, 1.8, 95% CI, 1.4–2.4). As would be expected, cardiogenic shock was negatively associated with survival (OR, 0.6, 95% CI, 0.5–0.7). Fig. 2 shows results of the mixed-effects logistic regression model assessing the association of percutaneous hemodynamic support device on survivability in all patients, acute MI patients, and patients complicated by cardiogenic shock. The mixed-effects logistic regression model attempted to control for variables found in our univariate analysis such as age, sex, VF initial rhythm, cardiogenic shock and presence of MI while also taking into account a theoretical hospital effect towards survivability. While controlling for these variables, use of a percutaneous hemodynamic support device was associated with survival in all patients (OR, 1.8, 95% CI, 1.3–2.5) and even more so when applied to acute MI patients (OR, 1.9, 95% CI, 1.3–2.9) and those in cardiogenic shock (OR, 2.0, 95% CI, 1.3–3.0).
4. Discussion These data identify that the use of percutaneous hemodynamic support device in post cardiac arrest patients admitted to Michigan hospitals was associated with an improved survival. This was true in all patients, patients who had a diagnosis of MI, and patients with a diagnosis of cardiogenic shock. While an improved outcome was observed for all patients, the association was even stronger with those who had suffered an acute MI or suffered from cardiogenic shock. To our knowledge this is the largest dataset that has evaluated use of percutaneous hemodynamic support device in the post arrest population. Our findings are important, as the literature specifically evaluating percutaneous hemodynamic support device use in cardiac arrest patients is sparse [7]. While existing literature evaluating IABP use in MI patients suffering from cardiogenic shock found no mortality benefit, there is some data to suggest that IABP use is positively associated with neurologically intact survival in post arrest patients [1,8,9]. In the largest prospective trial assessing IABP use in MI patients complicated by cardiac arrest, IABP was associated with higher mortality [10]. Because IABP was used in a sicker population, and the primary outcome of mortality did not assess for neurologic status, results were likely negatively biased. It is possible that these variables as well as our inclusion of pLVAD use could account for our positive association with survival. Furthermore, this study did not delineate between in hospital and OHCA while our study specifically looked at those admitted for cardiac arrest. In human subjects, very little data exists specifically addressing pLVAD use in OHCA. A single study compared IABP use to pLVAD (Impella 2.5) use in OHCA patients and found no difference in neurologically intact survival at 28 days. However, no comparison was made between patients that received either LVAD device to those that received neither [12,13]. Our findings expand the literature as our patient population represented patients who suffered either OHCA or Emergency
Please cite this article as: Pressman A, et al, Association between percutaneous hemodynamic support device and survival from cardiac arrest in the state of Michigan, American Journal of Emergency Medicine (2017), https://doi.org/10.1016/j.ajem.2017.10.036
A. Pressman et al. / American Journal of Emergency Medicine xxx (2017) xxx–xxx
3
Fig. 1. Unadjusted odds ratios of factors associated with survival to hospital discharge.
Department (ED) arrest and compared those receiving either percutaneous hemodynamic support device to those that received neither. Overall, few patients from the MIDB were treated with percutaneous hemodynamic support device after being admitted to the hospital for cardiac arrest. Our findings suggest a positive association between survival to discharge and percutaneous hemodynamic support device use defined as either IABP or pLVAD. Although retrospective and hypothesis generating, this study is important given the absence of prospective data on this therapy. A body of literature has evolved calling for the development of post cardiac arrest centers, which may provide interventions (emergent PCI, targeted temperature management, percutaneous hemodynamic support device) to post arrest patients [16-22]. This literature is mixed with respect to showing improved outcomes in such centers. Our data indicates that percutaneous hemodynamic support devices were placed in a small subset of hospitals across the state, and an even smaller subset of these placed them more frequently than once a year. This may further support the utilization of post cardiac arrest centers that have more experience with critically ill patients that may benefit from these devices. Data supporting identification of subsets of patients and therapies such as percutaneous hemodynamic support device placement are needed to guide decisions regarding the triage of post arrest patients. There is also growing literature showing that other novel innovations such as mechanical CPR and extracorporeal membrane oxygenation (ECMO) may improve outcomes in OHCA [23,24]. Mechanical CPR and ECMO CPR may serve as a “bridge-to-therapy”, giving the patient uninterrupted mechanical support until the appropriate therapeutic intervention (e.g. PCI) may be performed. Our findings suggest that the percutaneous hemodynamic support devices we studied may also be considered as a bridge-to-therapy while awaiting the targeted intervention that may resolve the cause of arrest. Furthermore, while most studies consider a positive outcome as survival or a neurologically intact state, further research may show that these bridge-to-therapy devices may also increase survival to organ donation [25]. While survival of the patient remains the foremost priority, in cases where neurologically intact survival is not possible, percutaneous hemodynamic support
devices may provide the necessary support to keep organs perfused while the subject of organ donation may be broached. Prospective data regarding the use of percutaneous hemodynamic support devices post arrest is needed, however, the relative infrequency of cardiogenic shock in this population (17.7% in this study) and limited number of centers that perform emergent percutaneous hemodynamic support device placement (b33% in this study) make it unlikely that large prospective multi-center trials will be performed to definitively answer the question of whether pLVAD therapy improves outcome. Furthermore, it would be difficult to implement a randomized control trial in this population as it would be unethical to withhold or place percutaneous hemodynamic support devices if felt there could be a benefit or harm for specific patients. In the interim, data from registries and administrative datasets play an important role in guiding clinicians and policy makers. 4.1. Limitations Use of ICD-9 CM coding is a fundamental limitation of this study, as we do not have physiological data to assess clinical characteristics and comorbidities of this population. All diagnoses are identified upon hospital discharge and we are unable to identify which clinical entities were present upon hospital admission. More in-depth analysis of the database subjects in regards to cardiac arrest type, MI type, total ischemic time, time to intervention with the percutaneous hemodynamic support device, and specifics about post-arrest hemodynamics were not possible and would have been of great interest. The vast majority of our population with percutaneous hemodynamic support devices received IABPs. Although pLVADs are relatively new devices that generate greater flow than IABPs, we are unable to assess the independent association of either technology alone to improve outcome. This dataset represents patients who were admitted to the hospital after resuscitation and have a continued perfusing rhythm, and who have an admission diagnosis of cardiac arrest. As such, patients who may have arrested in an ED are included in our population along with OHCA patients. This population may be more similar to out of hospital patients than inpatient
Fig. 2. Mixed-effects regression model testing effect of pLVAD use on survival to hospital discharge.
Please cite this article as: Pressman A, et al, Association between percutaneous hemodynamic support device and survival from cardiac arrest in the state of Michigan, American Journal of Emergency Medicine (2017), https://doi.org/10.1016/j.ajem.2017.10.036
4
A. Pressman et al. / American Journal of Emergency Medicine xxx (2017) xxx–xxx
cardiac arrests, but the relative number of these patients in our study cohort is not known. Our inclusion population does not include patients who have CPR in progress who received some sort of immediate circulatory support, such as extracorporeal membrane oxygenation (ECMO). While this immediate type of support is being implemented with increasing frequency and is an intriguing evolution of resuscitative care, it does not address our population. Finally, patients who expired prior to admission were not included in the MIDB and therefore are also not in our analysis. Our findings could possibly be confounded by survival bias as patients who survived long enough to receive percutaneous hemodynamic support device may have been more likely to survive overall. However, we feel this is unlikely, as numerous other studies have found that cardiac arrest patients are more likely to receive IABP and pLVAD if they have worse left ventricular function and hemodynamics [8,9,10]. Furthermore, unlike PCI, which is often performed after the patient may have awakened and their outcome status is clearer, IABP would typically be performed earlier in a patient's course, before an outcome can be anticipated. Finally we used survival to hospital discharge as our outcome and did not attempt to use any measures of quality of survival (e.g. discharge home vs skilled nursing, etc) in our analysis. Other measures of positive outcome, such as survival to organ transplantation or neurologically intact survival, should also be considered in the future. 4.2. Conclusion In this statewide population, treatment with percutaneous hemodynamic support device in the post-arrest period was associated with an improved survival to hospital discharge. These data support the need for further prospective work to assess the benefit of percutaneous hemodynamic support devices in post cardiac arrest patients. Acknowledgments
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13] [14] [15]
[16]
This study adhered to the conditions approved by the William Beaumont Hospital Institutional Review Board. The authors would like to acknowledge the work of the providers and hospitals that contribute their data to the MIDB.
[17]
Author disclosure statement
[19]
No competing financial interests exist. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
[18]
[20]
[21]
References [1] Antman EM, Anbe DT, Armstrong PW, et al. ACC/AHA guidelines for the management of patients with ST-elevation myocardial infarction–executive summary. A report of the American College of Cardiology/American Heart Association task force on practice guidelines (writing committee to revise the 1999 guidelines for the management of patients with acute myocardial infarction). J Am Coll Cardiol 2004;44: 671–719. [2] Thiele H, Zeymer U, Neumann FJ, et al. Intraaortic balloon pump in cardiogenic shock II (IABP-SHOCK II) trial investigators. Intra-aortic balloon counterpulsation in acute myocardial infarction complicated by cardiogenic shock (IABP-SHOCK II): final 12 month results of a randomized, open-label trial. Lancet 2013;382(9905): 1638–45. [3] O'Gara PT, Kushner FG, Ascheim DD, et al. ACCF/AHA guideline for the management of ST-elevation myocardial infarction: a report of the American College of Cardiology
[22] [23]
[24]
[25]
Foundation/American Heart Association task force on practice guidelines. J Am Coll Cardiol 2013;61:78–140. Lauten A, Engström AE, Jung C, et al. Percutaneous left-ventricular support with the Impella-2.5-assist device in acute cardiogenic shock: results of the ImpellaEUROSHOCK-registry. Circ Heart Fail 2013;6(1):23–30. Casassus F, Corre J, Leroux L, et al. The use of Impella 2.5 in severe refractory cardiogenic shock complicating an acute myocardial infarction. J Interv Cardiol 2015;28: 41–50. Rihal CS, Naidu SS, Givertz MM, et al. 2015 SCAI/ACC/HFSA/STS clinical expert consensus statement on the use of percutaneous mechanical circulatory support devices in cardiovascular care: endorsed by the American Heart Association, the Cardiological Society of India, and Sociedad Latino Americana de Cardiologia Intervencion; affirmation of value by the Canadian Association of Interventional CardiologyAssociation Canadienne de Cardiologie d'intervention. J Am Coll Cardiol 2015;65: 7–26. Mukku VK, Cai Q, Gilani S, et al. Use of impella ventricular assist device in patients with severe coronary artery disease presenting with cardiac arrest. Int J Angiol 2012 Sep;21(3):163–6. Iqbal MB, Al-Hussaini A, Rosser G, et al. Predictors of survival and favorable functional outcomes after out-of-hospital cardiac arrest in patients systematically brought to a dedicated heart attack center (from the Harefield cardiac arrest study). Am J Cardiol 2015;115(6):730–7. Iqbal MB, Al-Hussaini A, Rosser G, et al. Intra-aortic balloon pump counterpulsation in the post-resuscitation period is associated with improved functional outcomes in patients surviving an out-of-hospital cardiac arrest: insights from a dedicated heart attack centre. Heart Lung Circ 2016;25(12):1210–7. Kim HK, Jeong MH, Ahn Y, et al. Clinical outcomes of the intra-aortic balloon pump for resuscitated patients with acute myocardial infarction complicated by cardiac arrest. J Cardiol 2016;67(1):57–63. Derwall M, Brucken A, Bleilevens C, et al. Doubling survival and improving clinical outcomes using a left ventricular assist device instead of chest compressions for resuscitation after prolonged cardiac arrest: a large animal study. Crit Care 2015;19: 123. Manzo-Silberman S, Fichet J, Mathonnet A, et al. Percutaneous left ventricular assistance in post cardiac arrest shock: comparison of intra aortic blood pump and IMPELLA recover LP2.5. Resuscitation 2013;84:609–15. Vase H, Christensen S, Christiansen A, et al. The impella CP device for acute mechanical circulatory support in refractory cardiac arrest. Resuscitation 2017;112:70–4. Michigan Health & Hospital Association. Data services. Available at: http://www. datakoala.com/about.htm, Accessed date: 14 September 2017. Reeves MJ, Nickles AV, Roberts S, et al. Assessment of the completeness and accuracy of case ascertainment in the Michigan Stroke Registry. Circ Cardiovasc Qual Outcomes 2014;7:757–63. Sunde K. SOPs and the right hospitals to improve outcome after cardiac arrest. Best Pract Res Clin Anaesthesiol 2013;27(3):373–81. https://doi.org/10.1016/j.bpa.2013. 07.009. Hasegawa K, Tsugawa Y, Camargo CA, Hiraide A, Brown DFM. Regional variability in survival outcomes of out-of-hospital cardiac arrest: the All-Japan Utstein Registry. Resuscitation 2013;84(8):1099–107. https://doi.org/10.1016/j.resuscitation.2013. 03.007. Carr BG, Kahn JM, Merchant RM, Kramer AA, Neumar RW. Inter-hospital variability in post-cardiac arrest mortality. Resuscitation 2009;80(1):30–4. https://doi.org/10. 1016/j.resuscitation.2008.09.001. Cudnik MT, Sasson C, Rea TD, et al. Increasing hospital volume is not associated with improved survival in out of hospital cardiac arrest of cardiac etiology. Resuscitation 2012;83(7):862–8. https://doi.org/10.1016/j.resuscitation.2012.02.006. Martin-Gill C, Dilger CP, Guyette FX, Rittenberger JC, Callaway CW. Regional impact of cardiac arrest center criteria on out-of-hospital transportation practices. Prehosp Emerg Care 2011;15(3):381–7. https://doi.org/10.3109/10903127.2011.561409. Donnion MW, Rittenberger JC, Gaeski D, et al. The development and implementation of cardiac arrest centers. Resuscitation 2011;82(8):974–8. Lurie K, Idris A, Holcomb R. Level 1 cardiac arrest centers: learning from the trauma surgeons. Acad Emerg Med 2005;12(1):79–80. Siao FY, Chiu CC, Chiu CW, et al. Managing cardiac arrest with refractory ventricular fibrillation in the emergency department: conventional cardiopulmonary resuscitation versus extracorporeal cardiopulmonary resuscitation. Resuscitation 2015;92: 70–6. Stub D, Bernard S, Pellegrino V, et al. Refractory cardiac arrest treated with mechanical CPR, hypothermia, ECMO and early reperfusion (the CHEER trial). Resuscitation 2015;86:88–94. Alarhayem AQ, Cohn SM, Muir MT, et al. Organ donation, an unexpected benefit of aggressive resuscitation of trauma patients presenting dead on arrival. J Am Coll Surg 2017;224(5):926–32.
Please cite this article as: Pressman A, et al, Association between percutaneous hemodynamic support device and survival from cardiac arrest in the state of Michigan, American Journal of Emergency Medicine (2017), https://doi.org/10.1016/j.ajem.2017.10.036
Contents lists available at ScienceDirect
American Journal of Emergency Medicine journal homepage: www.elsevier.com/locate/ajem
Association between percutaneous hemodynamic support device and survival from cardiac arrest in the state of Michigan☆ Andrew Pressman, MD a, Kelly N. Sawyer, MD, MS b, William Devlin, MD c, Robert Swor, DO d,⁎ a
Department of Emergency Medicine, Beaumont Hospital-Troy, Troy, MI, United States Department of Emergency Medicine, University of Pittsburgh, School of Medicine, Pittsburgh, PA, United States Division of Cardiology, Beaumont Hospital-Troy, Troy, MI, United States d Department of Emergency Medicine, Oakland University William Beaumont School of Medicine, Rochester, MI, United States b c
a r t i c l e
i n f o
Article history: Received 16 July 2017 Received in revised form 12 October 2017 Accepted 12 October 2017 Available online xxxx Keywords: Cardiac arrest Outcome Ventricular assist device
a b s t r a c t Introduction: The role of circulatory support in the post-cardiac arrest period remains controversial. Our objective was to investigate the association between treatment with a percutaneous hemodynamic support device and outcome after admission for cardiac arrest. Methods: We performed a retrospective study of adult patients with admission diagnosis of cardiac arrest or ventricular fibrillation (VF) from the Michigan Inpatient Database, treated between July 1, 2010, and June 30, 2013. Patient demographics, clinical characteristics, treatments, and disposition were electronically abstracted based on ICD-9 codes at the hospital level. Mixed-effects logistic regression models were fit to test the effect of percutaneous hemodynamic support device defined as either percutaneous left ventricular assist device (pLVAD) or intra-aortic balloon pump (IABP) on survival. These models controlled for age, sex, VF, myocardial infarction (MI), and cardiogenic shock with hospital modeled as a random effect. Results: A total of 103 hospitals contributed 4393 patients for analysis, predominately male (58.8%) with a mean age of 64.1 years (SD 15.5). On univariate analysis, younger age, male sex, VF as the initial rhythm, acute MI, percutaneous coronary intervention, percutaneous hemodynamic support device, and absence of cardiogenic shock were associated with survival to discharge (each p b 0.001). Mixed-effects logistic regressions revealed use of percutaneous hemodynamic support device was significantly associated with survival among all patients (OR 1.8 (1.28–2.54)), and especially in those with acute MI (OR 1.95 (1.31–2.93)) or cardiogenic shock (OR 1.96 (1.29–2.98)). Conclusion: Treatment with percutaneous hemodynamic support device in the post-arrest period may provide left ventricular support and improve outcome. © 2017 Elsevier Inc. All rights reserved.
1. Introduction The benefit of circulatory support in the post-cardiac arrest period remains controversial. Historically, intra-aortic balloon pumps (IABP) have been used as the primary circulatory support device for myocardial infarction (MI) complicated by cardiogenic shock [1]. However, a contemporary prospective, randomized controlled trial has revealed no mortality benefit with this therapy at one month or one year for patients in cardiogenic shock after MI [2]. While use of IABP for cardiogenic shock in the setting of acute MI was once a level 1b recommendation in America, it has recently been downgraded to a level 2b
☆ Sources of support: This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. ⁎ Corresponding author at: William Beaumont Hospital, 3601 W. 13 Mile Rd, Royal Oak, MI 48073, United States. E-mail address: [email protected] (R. Swor).
recommendation [1,3]. Advances in technology have allowed for mechanical support using percutaneous left ventricular assist devices (pLVAD). While numerous studies have shown improved hemodynamics with these devices, no randomized controlled trial has been performed to assess overall mortality benefit [4,5,6]. Furthermore, studies assessing both IABP and pLVAD use in patients resuscitated from outof-hospital cardiac arrest (OHCA) are limited given the difficulty in studying this patient population [7]. Data from IABP use to treat post-cardiac arrest cardiogenic shock are more readily available than data from other pLVADs. Two studies retrospectively studying the subject have found that IABP use was strongly associated with neurologically intact survival [8,9]. However, a further prospective study found IABP use associated with increased mortality in MI patients complicated by cardiac arrest [10]. There have been promising porcine studies evaluating the use of pLVAD during cardiac arrest resuscitation, doubling the rate of return of spontaneous circulation (ROSC) compared to standard care [11]. In human subjects, studies
https://doi.org/10.1016/j.ajem.2017.10.036 0735-6757/© 2017 Elsevier Inc. All rights reserved.
Please cite this article as: Pressman A, et al, Association between percutaneous hemodynamic support device and survival from cardiac arrest in the state of Michigan, American Journal of Emergency Medicine (2017), https://doi.org/10.1016/j.ajem.2017.10.036
2
A. Pressman et al. / American Journal of Emergency Medicine xxx (2017) xxx–xxx
assessing pLVAD use in cardiac arrest have been limited to case series. These series showed that pLVAD use was associated with a sicker patient populations but found no difference in neurologically intact survival [12,13]. It is clear that little literature exists studying pLVAD and IABP use in post cardiac arrest patients across larger populations. The objective of this study was to investigate whether, in a statewide cohort of post cardiac arrest patients, use of percutaneous hemodynamic support device was associated with survival. 2. Methods We performed a retrospective study of adult (age ≥ 18 years) patients with an admission diagnosis of cardiac arrest (ICD-CM 427.5) or ventricular fibrillation (ICD-9 CM 427.41) from the Michigan Inpatient Database (MIDB), treated between July 1, 2010, and June 30, 2013. This definition of cardiac arrest includes both OHCA patients that survived to hospital admission, as well as those who arrested in the ED and survived to admission. No patient with CPR in progress is included in this cohort of patients. The MIDB is run by the Michigan Health & Hospital Association and has been collecting data on patients requiring inpatient admission since the 1980s. It currently collects data on 1.3 million admitted patients annually from N 100 hospitals across Michigan [14,15]. This database collects information on all patients who survived to admission into an acute care hospital, including patient demographics, receiving facility, discharge diagnoses, procedures performed, and patient outcomes. Patient outcomes recorded included survival to hospital discharge and discharge destination (home, short term rehabilitation, skilled nursing facility, hospital). For this study we defined surviving to hospital discharge as any patient discharged from the hospital to home or another facility other than hospice. Data on patient demographics (age, sex), clinical characteristics (arrest rhythm of ventricular fibrillation (VF), cardiogenic shock (ICD-9 CM 98551), myocardial infarction (ICD-9 CM, 41000–41010)), procedures performed (e.g. percutaneous coronary intervention (PCI), percutaneous hemodynamic support device placement), and outcome (survived to discharge) were electronically abstracted based on ICD-9 codes at the hospital level. Percutaneous hemodynamic support device was defined as either IABP or pLVAD (Impella 2.5). Duplicate cases, patients b 18 years old, and cases with missing data were excluded. Univariate relationships with outcome were compared using a Welch's two-sample t-test for continuous data and a chi-squared test of independence for categorical data. Mixed-effects logistic regression models were fit to test the effect of percutaneous hemodynamic support device on outcome, with hospital modeled as a random effect. In the mixed-effects logistic regression, we controlled for those variables known to have an effect on survival from our univariate analysis (age, sex, VF, MI, and cardiogenic shock), but also attempted to control for specific hospital effect on survival. This study was approved by the William Beaumont Hospital Institutional Review Board. All data cleaning and analysis was performed using R (version 3.2.1) and RStudio (version 0.99.467). 3. Results A total of 103 hospitals in Michigan contributed 4393 patients for analysis with a mean age of 64.1 years (SD 15.5) and 41% being female. From the total, 1422 patients (32.4%) had a diagnosis of VF, 1174 (26.7%) with MI, and 779 (17.7%) cardiogenic shock. Only 197 (4.5%) patients received a percutaneous hemodynamic support device, of which most were IABP (182 (4.1%)). These were placed in 35 different hospitals, with only 20 hospitals placing them more than once a year. Table 1 and Fig. 1 highlight the results of the univariate analyses, revealing that younger age, male sex, VF initial rhythm, acute MI, PCI, percutaneous hemodynamic support device, and absence of cardiogenic shock were associated with survival to discharge. This effect was most notable
Table 1 Univariate analysis of factors associated with survival to hospital discharge. Patient variables
Survival (n = 1688)
Deaths (n = 2705)
p-Value
Age, mean (SD) Male sex, n(%) VF initial rhythm, n(%) Acute MI, n(%) Cardiogenic shock, n(%) Received PCI, n(%) Hemodynamic support device, n(%)
61.5 (15.3) 1083 (64.2) 821 (48.6) 510 (30.2) 220 (13.0) 278 (16.5) 103 (6.1)
65.7 (15.5) 1499 (55.4) 601 (22.2) 664 (24.5) 559 (20.7) 122 (4.5) 94 (3.5)
b0.001a b0.001b b0.001b b0.001b b0.001b b0.001b b0.001b
VF-ventricular fibrillation; MI-myocardial infarction; PCI-percutaneous intervention. a Compared using Welch's two-sample t-test. b Compared using chi-squared test of independence.
for an initial rhythm of VF (OR, 3.3, 95% CI, 2.9–3.8) and receiving PCI (OR, 4.2, 95% CI, 3.3–5.2) but was still significant for those receiving a percutaneous hemodynamic support device (OR, 1.8, 95% CI, 1.4–2.4). As would be expected, cardiogenic shock was negatively associated with survival (OR, 0.6, 95% CI, 0.5–0.7). Fig. 2 shows results of the mixed-effects logistic regression model assessing the association of percutaneous hemodynamic support device on survivability in all patients, acute MI patients, and patients complicated by cardiogenic shock. The mixed-effects logistic regression model attempted to control for variables found in our univariate analysis such as age, sex, VF initial rhythm, cardiogenic shock and presence of MI while also taking into account a theoretical hospital effect towards survivability. While controlling for these variables, use of a percutaneous hemodynamic support device was associated with survival in all patients (OR, 1.8, 95% CI, 1.3–2.5) and even more so when applied to acute MI patients (OR, 1.9, 95% CI, 1.3–2.9) and those in cardiogenic shock (OR, 2.0, 95% CI, 1.3–3.0).
4. Discussion These data identify that the use of percutaneous hemodynamic support device in post cardiac arrest patients admitted to Michigan hospitals was associated with an improved survival. This was true in all patients, patients who had a diagnosis of MI, and patients with a diagnosis of cardiogenic shock. While an improved outcome was observed for all patients, the association was even stronger with those who had suffered an acute MI or suffered from cardiogenic shock. To our knowledge this is the largest dataset that has evaluated use of percutaneous hemodynamic support device in the post arrest population. Our findings are important, as the literature specifically evaluating percutaneous hemodynamic support device use in cardiac arrest patients is sparse [7]. While existing literature evaluating IABP use in MI patients suffering from cardiogenic shock found no mortality benefit, there is some data to suggest that IABP use is positively associated with neurologically intact survival in post arrest patients [1,8,9]. In the largest prospective trial assessing IABP use in MI patients complicated by cardiac arrest, IABP was associated with higher mortality [10]. Because IABP was used in a sicker population, and the primary outcome of mortality did not assess for neurologic status, results were likely negatively biased. It is possible that these variables as well as our inclusion of pLVAD use could account for our positive association with survival. Furthermore, this study did not delineate between in hospital and OHCA while our study specifically looked at those admitted for cardiac arrest. In human subjects, very little data exists specifically addressing pLVAD use in OHCA. A single study compared IABP use to pLVAD (Impella 2.5) use in OHCA patients and found no difference in neurologically intact survival at 28 days. However, no comparison was made between patients that received either LVAD device to those that received neither [12,13]. Our findings expand the literature as our patient population represented patients who suffered either OHCA or Emergency
Please cite this article as: Pressman A, et al, Association between percutaneous hemodynamic support device and survival from cardiac arrest in the state of Michigan, American Journal of Emergency Medicine (2017), https://doi.org/10.1016/j.ajem.2017.10.036
A. Pressman et al. / American Journal of Emergency Medicine xxx (2017) xxx–xxx
3
Fig. 1. Unadjusted odds ratios of factors associated with survival to hospital discharge.
Department (ED) arrest and compared those receiving either percutaneous hemodynamic support device to those that received neither. Overall, few patients from the MIDB were treated with percutaneous hemodynamic support device after being admitted to the hospital for cardiac arrest. Our findings suggest a positive association between survival to discharge and percutaneous hemodynamic support device use defined as either IABP or pLVAD. Although retrospective and hypothesis generating, this study is important given the absence of prospective data on this therapy. A body of literature has evolved calling for the development of post cardiac arrest centers, which may provide interventions (emergent PCI, targeted temperature management, percutaneous hemodynamic support device) to post arrest patients [16-22]. This literature is mixed with respect to showing improved outcomes in such centers. Our data indicates that percutaneous hemodynamic support devices were placed in a small subset of hospitals across the state, and an even smaller subset of these placed them more frequently than once a year. This may further support the utilization of post cardiac arrest centers that have more experience with critically ill patients that may benefit from these devices. Data supporting identification of subsets of patients and therapies such as percutaneous hemodynamic support device placement are needed to guide decisions regarding the triage of post arrest patients. There is also growing literature showing that other novel innovations such as mechanical CPR and extracorporeal membrane oxygenation (ECMO) may improve outcomes in OHCA [23,24]. Mechanical CPR and ECMO CPR may serve as a “bridge-to-therapy”, giving the patient uninterrupted mechanical support until the appropriate therapeutic intervention (e.g. PCI) may be performed. Our findings suggest that the percutaneous hemodynamic support devices we studied may also be considered as a bridge-to-therapy while awaiting the targeted intervention that may resolve the cause of arrest. Furthermore, while most studies consider a positive outcome as survival or a neurologically intact state, further research may show that these bridge-to-therapy devices may also increase survival to organ donation [25]. While survival of the patient remains the foremost priority, in cases where neurologically intact survival is not possible, percutaneous hemodynamic support
devices may provide the necessary support to keep organs perfused while the subject of organ donation may be broached. Prospective data regarding the use of percutaneous hemodynamic support devices post arrest is needed, however, the relative infrequency of cardiogenic shock in this population (17.7% in this study) and limited number of centers that perform emergent percutaneous hemodynamic support device placement (b33% in this study) make it unlikely that large prospective multi-center trials will be performed to definitively answer the question of whether pLVAD therapy improves outcome. Furthermore, it would be difficult to implement a randomized control trial in this population as it would be unethical to withhold or place percutaneous hemodynamic support devices if felt there could be a benefit or harm for specific patients. In the interim, data from registries and administrative datasets play an important role in guiding clinicians and policy makers. 4.1. Limitations Use of ICD-9 CM coding is a fundamental limitation of this study, as we do not have physiological data to assess clinical characteristics and comorbidities of this population. All diagnoses are identified upon hospital discharge and we are unable to identify which clinical entities were present upon hospital admission. More in-depth analysis of the database subjects in regards to cardiac arrest type, MI type, total ischemic time, time to intervention with the percutaneous hemodynamic support device, and specifics about post-arrest hemodynamics were not possible and would have been of great interest. The vast majority of our population with percutaneous hemodynamic support devices received IABPs. Although pLVADs are relatively new devices that generate greater flow than IABPs, we are unable to assess the independent association of either technology alone to improve outcome. This dataset represents patients who were admitted to the hospital after resuscitation and have a continued perfusing rhythm, and who have an admission diagnosis of cardiac arrest. As such, patients who may have arrested in an ED are included in our population along with OHCA patients. This population may be more similar to out of hospital patients than inpatient
Fig. 2. Mixed-effects regression model testing effect of pLVAD use on survival to hospital discharge.
Please cite this article as: Pressman A, et al, Association between percutaneous hemodynamic support device and survival from cardiac arrest in the state of Michigan, American Journal of Emergency Medicine (2017), https://doi.org/10.1016/j.ajem.2017.10.036
4
A. Pressman et al. / American Journal of Emergency Medicine xxx (2017) xxx–xxx
cardiac arrests, but the relative number of these patients in our study cohort is not known. Our inclusion population does not include patients who have CPR in progress who received some sort of immediate circulatory support, such as extracorporeal membrane oxygenation (ECMO). While this immediate type of support is being implemented with increasing frequency and is an intriguing evolution of resuscitative care, it does not address our population. Finally, patients who expired prior to admission were not included in the MIDB and therefore are also not in our analysis. Our findings could possibly be confounded by survival bias as patients who survived long enough to receive percutaneous hemodynamic support device may have been more likely to survive overall. However, we feel this is unlikely, as numerous other studies have found that cardiac arrest patients are more likely to receive IABP and pLVAD if they have worse left ventricular function and hemodynamics [8,9,10]. Furthermore, unlike PCI, which is often performed after the patient may have awakened and their outcome status is clearer, IABP would typically be performed earlier in a patient's course, before an outcome can be anticipated. Finally we used survival to hospital discharge as our outcome and did not attempt to use any measures of quality of survival (e.g. discharge home vs skilled nursing, etc) in our analysis. Other measures of positive outcome, such as survival to organ transplantation or neurologically intact survival, should also be considered in the future. 4.2. Conclusion In this statewide population, treatment with percutaneous hemodynamic support device in the post-arrest period was associated with an improved survival to hospital discharge. These data support the need for further prospective work to assess the benefit of percutaneous hemodynamic support devices in post cardiac arrest patients. Acknowledgments
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13] [14] [15]
[16]
This study adhered to the conditions approved by the William Beaumont Hospital Institutional Review Board. The authors would like to acknowledge the work of the providers and hospitals that contribute their data to the MIDB.
[17]
Author disclosure statement
[19]
No competing financial interests exist. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
[18]
[20]
[21]
References [1] Antman EM, Anbe DT, Armstrong PW, et al. ACC/AHA guidelines for the management of patients with ST-elevation myocardial infarction–executive summary. A report of the American College of Cardiology/American Heart Association task force on practice guidelines (writing committee to revise the 1999 guidelines for the management of patients with acute myocardial infarction). J Am Coll Cardiol 2004;44: 671–719. [2] Thiele H, Zeymer U, Neumann FJ, et al. Intraaortic balloon pump in cardiogenic shock II (IABP-SHOCK II) trial investigators. Intra-aortic balloon counterpulsation in acute myocardial infarction complicated by cardiogenic shock (IABP-SHOCK II): final 12 month results of a randomized, open-label trial. Lancet 2013;382(9905): 1638–45. [3] O'Gara PT, Kushner FG, Ascheim DD, et al. ACCF/AHA guideline for the management of ST-elevation myocardial infarction: a report of the American College of Cardiology
[22] [23]
[24]
[25]
Foundation/American Heart Association task force on practice guidelines. J Am Coll Cardiol 2013;61:78–140. Lauten A, Engström AE, Jung C, et al. Percutaneous left-ventricular support with the Impella-2.5-assist device in acute cardiogenic shock: results of the ImpellaEUROSHOCK-registry. Circ Heart Fail 2013;6(1):23–30. Casassus F, Corre J, Leroux L, et al. The use of Impella 2.5 in severe refractory cardiogenic shock complicating an acute myocardial infarction. J Interv Cardiol 2015;28: 41–50. Rihal CS, Naidu SS, Givertz MM, et al. 2015 SCAI/ACC/HFSA/STS clinical expert consensus statement on the use of percutaneous mechanical circulatory support devices in cardiovascular care: endorsed by the American Heart Association, the Cardiological Society of India, and Sociedad Latino Americana de Cardiologia Intervencion; affirmation of value by the Canadian Association of Interventional CardiologyAssociation Canadienne de Cardiologie d'intervention. J Am Coll Cardiol 2015;65: 7–26. Mukku VK, Cai Q, Gilani S, et al. Use of impella ventricular assist device in patients with severe coronary artery disease presenting with cardiac arrest. Int J Angiol 2012 Sep;21(3):163–6. Iqbal MB, Al-Hussaini A, Rosser G, et al. Predictors of survival and favorable functional outcomes after out-of-hospital cardiac arrest in patients systematically brought to a dedicated heart attack center (from the Harefield cardiac arrest study). Am J Cardiol 2015;115(6):730–7. Iqbal MB, Al-Hussaini A, Rosser G, et al. Intra-aortic balloon pump counterpulsation in the post-resuscitation period is associated with improved functional outcomes in patients surviving an out-of-hospital cardiac arrest: insights from a dedicated heart attack centre. Heart Lung Circ 2016;25(12):1210–7. Kim HK, Jeong MH, Ahn Y, et al. Clinical outcomes of the intra-aortic balloon pump for resuscitated patients with acute myocardial infarction complicated by cardiac arrest. J Cardiol 2016;67(1):57–63. Derwall M, Brucken A, Bleilevens C, et al. Doubling survival and improving clinical outcomes using a left ventricular assist device instead of chest compressions for resuscitation after prolonged cardiac arrest: a large animal study. Crit Care 2015;19: 123. Manzo-Silberman S, Fichet J, Mathonnet A, et al. Percutaneous left ventricular assistance in post cardiac arrest shock: comparison of intra aortic blood pump and IMPELLA recover LP2.5. Resuscitation 2013;84:609–15. Vase H, Christensen S, Christiansen A, et al. The impella CP device for acute mechanical circulatory support in refractory cardiac arrest. Resuscitation 2017;112:70–4. Michigan Health & Hospital Association. Data services. Available at: http://www. datakoala.com/about.htm, Accessed date: 14 September 2017. Reeves MJ, Nickles AV, Roberts S, et al. Assessment of the completeness and accuracy of case ascertainment in the Michigan Stroke Registry. Circ Cardiovasc Qual Outcomes 2014;7:757–63. Sunde K. SOPs and the right hospitals to improve outcome after cardiac arrest. Best Pract Res Clin Anaesthesiol 2013;27(3):373–81. https://doi.org/10.1016/j.bpa.2013. 07.009. Hasegawa K, Tsugawa Y, Camargo CA, Hiraide A, Brown DFM. Regional variability in survival outcomes of out-of-hospital cardiac arrest: the All-Japan Utstein Registry. Resuscitation 2013;84(8):1099–107. https://doi.org/10.1016/j.resuscitation.2013. 03.007. Carr BG, Kahn JM, Merchant RM, Kramer AA, Neumar RW. Inter-hospital variability in post-cardiac arrest mortality. Resuscitation 2009;80(1):30–4. https://doi.org/10. 1016/j.resuscitation.2008.09.001. Cudnik MT, Sasson C, Rea TD, et al. Increasing hospital volume is not associated with improved survival in out of hospital cardiac arrest of cardiac etiology. Resuscitation 2012;83(7):862–8. https://doi.org/10.1016/j.resuscitation.2012.02.006. Martin-Gill C, Dilger CP, Guyette FX, Rittenberger JC, Callaway CW. Regional impact of cardiac arrest center criteria on out-of-hospital transportation practices. Prehosp Emerg Care 2011;15(3):381–7. https://doi.org/10.3109/10903127.2011.561409. Donnion MW, Rittenberger JC, Gaeski D, et al. The development and implementation of cardiac arrest centers. Resuscitation 2011;82(8):974–8. Lurie K, Idris A, Holcomb R. Level 1 cardiac arrest centers: learning from the trauma surgeons. Acad Emerg Med 2005;12(1):79–80. Siao FY, Chiu CC, Chiu CW, et al. Managing cardiac arrest with refractory ventricular fibrillation in the emergency department: conventional cardiopulmonary resuscitation versus extracorporeal cardiopulmonary resuscitation. Resuscitation 2015;92: 70–6. Stub D, Bernard S, Pellegrino V, et al. Refractory cardiac arrest treated with mechanical CPR, hypothermia, ECMO and early reperfusion (the CHEER trial). Resuscitation 2015;86:88–94. Alarhayem AQ, Cohn SM, Muir MT, et al. Organ donation, an unexpected benefit of aggressive resuscitation of trauma patients presenting dead on arrival. J Am Coll Surg 2017;224(5):926–32.
Please cite this article as: Pressman A, et al, Association between percutaneous hemodynamic support device and survival from cardiac arrest in the state of Michigan, American Journal of Emergency Medicine (2017), https://doi.org/10.1016/j.ajem.2017.10.036