Bringing doctors together

or

A free, private network for verified physicians

Leading voices

Dr. Brahmajee Nallamothu

Interventional Cardiology · UMHS

View

Dr. Mohamad Adnan Alkhouli

Interventional Cardiology · Mayo Clinic College of Medicine

View

Dr. Ajay Kirtane

Interventional Cardiology · CUIMC

View

Dr. Manesh Patel

Interventional Cardiology · Duke University

View

Dr. Deepak Bhatt

Interventional Cardiology · Mount Sinai

View

Dr. Sunil Rao

Interventional Cardiology · NYU Langone

View

Dr. Alexander Fanaroff

Interventional Cardiology · UPenn

View

Dr. James Tcheng

Interventional Cardiology · Duke

View

Dr. Pinak Shah

Interventional Cardiology · PHS

View

Dr. Dorairaj Prabhakaran

General Cardiology

View
  • Discover your community

  • Stay current

  • A profile that represents you

  • Find your people

About RoonPrivacyTerms

    Overview

    Effects of Perfusion, Coronary Artery Disease Burden, and Revascularization in Establishing Organized Cardiac Rhythm During Extracorporeal Cardiopulmonary Resuscitation for Shockable Refractory Out-of-Hospital Cardiac Arrest.

    Oct 18, 2024DOI: 10.1161/jaha.123.033907Journal ArticleAHA

    Written by: Ruben Crespo-DiazMarinos KosmopoulosGanesh RaveendranSergey GurevichDemetris YannopoulosJason A Bartos

    Find these authors on Roon

    GR
    Dr. Ganesh Raveendran
    Interventional Cardiology · University of Minnesota

    Methods

    Data related to the University of Minnesota ECPR cohort are available from the corresponding author upon reasonable request.We performed a retrospective analysis of 289 consecutive patients transported for the University of Minnesota ECPR program between December 2015 and June 2022. All data were collected as part of clinical care with retrospective review for this study. The Institutional Review Board at the University of Minnesota approved this study and data management (No. 1703M11301) with waiver of informed consent.Details of the University of Minnesota ECPR program have been described previously (Figure 1).2, 9, 10, 11, 12 All patients met the following field transport criteria as assessed by paramedics at the scene: age 18 to 75 years, OHCA with an initial rhythm of VF or VT, refractory cardiac arrest as defined by ongoing VT/VF despite ≥3 defibrillation attempts and ≥300 mg of amiodarone or conversion into pulseless electrical activity or asystole with ongoing CPR. Upon arrival to the cannulation center, patients were assessed for venoarterial ECMO cannulation criteria including end‐tidal CO2 ≥ 10 mm Hg, arterial partial pressure of oxygen ≥50 mm Hg, and lactic acid ≤18 mmol/L. For patients meeting the cannulation criteria, CPR was continued while peripheral venoarterial cannulation and initiation were performed by interventional cardiologists. Coronary angiography was then immediately performed followed by PCI when indicated at the discretion of the interventional cardiologist. Routine rhythm assessment was performed during pauses in CPR for rhythm/pulse checks as described for ACLS. Once ECMO cannulation was completed and flow was initiated, CPR was halted and the cardiac rhythm was assessed continuously using routine rhythm monitoring equipment at the cannulation sites. The patient's rhythm was assessed through all interventions, with defibrillation performed when a shockable rhythm was present. Grouping patients into the resuscitation stages depicted in Figure 1 required maintenance of an organized rhythm throughout the subsequent stages. All patients were cannulated with venoarterial ECMO; pump flow and sweep were individualized to the patient's clinical intensive care unit (ICU) needs and tolerance.2, 10 Coronary angiography was then immediately performed, followed by PCI when indicated at the discretion of the interventional cardiologist. The patient's rhythm was assessed through all interventions, with defibrillation performed when shockable rhythm was present. Patients were then transferred to the cardiac intensive care unit where postarrest care was provided.Stage 1 involves out‐of‐hospital advanced cardiac life support while transferring to hospital. Stage 2, refractory cardiac arrest patient is taken directly to the catheterization laboratory for ECMO cannulation. Stage 3, immediately following cannulation coronary angiography is performed along with PCI when a culprit lesion or lesions are identified. Stage 4, patient is then transferred to the cardiac intensive care unit. ECMO indicates extracorporeal membrane oxygenation; ICU, intensive care unit; and PCI, percutaneous coronary intervention.
    Patients were grouped based on the stage during ECPR (Figure 1) in which they achieved a sustained organized rhythm. An organized rhythm was defined as organized cardiac electrical activity visible on ECG (eg, not VF, VT, or asystole) independent of cardiac mechanical activity. Patients who achieved an organized rhythm before arriving at the ECMO cannulation center are denoted as pre‐ECMO. Patients who converted to an organized rhythm with defibrillation after ECMO cannulation and initiation, but before any PCI, are denoted as post‐ECMO. Patients who required PCI before achieving an organized cardiac rhythm are labeled as post‐PCI. When an organized rhythm was not obtained despite venoarterial ECMO, PCI, and 90 minutes of additional medical therapy, then patients were labeled as ICU admit. Intermittent return of spontaneous circulation (ROSC) was defined as intermittent pulses for <20 minutes. Patients with sustained ROSC (>20 minutes) were excluded from this study. Patients with intermittent ROSC upon arrival to cannulation but in refractory cardiogenic shock (Society for Cardiovascular Angiography & Interventions Stage E) despite maximum doses of 2 or more inotrope or pressors were included. Neurologically favorable survival was determined by Cerebral Performance Category score (CPC) ≤2 at the time of patient discharge. Patients with CPC score 3 to 5 were considered to have poor outcome. Cardiac rhythm was continuously recorded throughout the resuscitation process. Arterial blood samples and lactic acid levels from each patient were collected at the time of arrival to the cannulation center. Standard coronary angiography views were taken and analyzed immediately following ECMO cannulation to determine the presence of CAD and indication for PCI. Obstructive CAD was defined as ≥70% diameter stenosis in the left anterior descending (LAD), left circumflex (LCx), or right coronary (RCA) arteries or ≥50% stenosis of the left main coronary artery. Multivessel disease was noted when 2 or more vessels met the obstructive CAD criteria. Angiographic images and reports were analyzed to construct a map highlighting the presence of obstructive CAD, distribution, and PCI profile in each coronary territory. The Gensini5, 13 and SYNTAX14 (Synergy Between PCI With Taxus and Cardiac Surgery) scores were calculated in all patients.
    Categorical variables are presented as numbers and their frequencies (percentage). Continuous variables were reported as a mean±SD if data were normally distributed or median with the interquartile range if without normal distribution. Comparison of categorical variables was performed using the chi‐square test. Comparison of continuous variables when differences were significant was performed with Tukey (parametric) or Wilcoxon (nonparametric) tests for pairwise comparisons and each corrected for multiple comparisons. All probability values were 2‐sided and a P value <0.05 was considered statistically significant and adjusted for multiple comparisons using parametric and nonparametric methods. Organized rhythm based on ECPR stage distribution outcomes were analyzed using logistic regression and reported as adjusted odds ratios (ORs) with 95% CIs. Age, pH, CPR duration, and lactic acid level were used for adjusted ORs. Data were stored in a Research Electronic Data Capture database at the University of Minnesota and exported to JMP (SAS Institute, Cary, NC) and SAS software (SAS Institute) for statistical studies.

    Results

    The ECPR process was composed of 4 stages as shown in Figure 1. Stage 1 is composed of prehospital and pre‐ECMO ACLS care including CPR, defibrillations, vasoactive medications, and antiarrhythmics. Stage 2 consists of ECMO cannulation and initiation of ECMO flow. Stage 3 includes coronary angiography and PCI as indicated. Stage 4 involves admission to the ICU for postarrest care. Defibrillation and medical therapy including vasoactive therapies were provided during all stages as needed. The specific rhythm distribution across all stages is depicted in Table S1.
    Out of all 289 patients with refractory OHCA and a shockable presenting rhythm transferred for ECMO cannulation as part of the University of Minnesota ECPR program, 148 (51%) established an organized rhythm at the pre‐ECMO stage, 87 (30%) achieved an organized rhythm post‐ECMO initiation but before the coronary angiogram, and 37 (13%) had refractory VF/VT until after PCI was performed (Figure 2). This resulted in 94% of patients having an organized rhythm in the catheterization laboratory and only 6% without an organized rhythm until after ICU admission. Patients were grouped based on the stage in which an organized rhythm was achieved (Table 1). There were no statistically significant differences in age, sex, or race. Intermittent ROSC occurred most frequently (96/148 [65%]) in patients achieving an organized rhythm before arrival at the cannulation center, compared with 27 of 87 (31%) with successful defibrillation after ECMO initiation, 11 of 37 (29%) patients achieving an organized rhythm after PCI, and 1 of 17 (6%) patients failing to achieve an organized rhythm before ICU admission (P<0.0001). Initial lactic acid levels upon arrival to the cannulation center were significantly lower in the pre‐ECMO group (9.9±4.0 mmol/L) compared with the post‐ECMO (12±4.2 mmol/L; P<0.005) and ICU admit (14±6.6 mmol/L; P<0.001) groups. The pH was also higher in pre‐ECMO patients (7.1±0.18) compared with post‐ECMO (7.0±0.17; P<0.05) and ICU admit patients (6.9±0.19; P<0.02). Arterial PaCO2 and arterial partial pressure of oxygen were not significantly different between groups.
    CAD indicates coronary artery disease; ECMO, extracorporeal membrane oxygenation; ICU, intensive care unit; and PCI, percutaneous coronary intervention.
    Patient and Arrest Characteristics Separated by the Stage of Achieving an Organized Cardiac Rhythm
    Values are mean±SD, n (%), or median (interquartile range).
    ECMO indicates extracorporeal membrane oxygenation; ICU, intensive care unit; and PCI, percutaneous coronary intervention.
    Statistical significance with P≤0.05 when compared with pre‐ECMO group and adjusted for multiple comparisons.
    Categorical variables comparison with P≤0.05.
    CPR duration, defined as time from 911 call to ECMO start of flow initiation was significantly different only in the pre‐ECMO group (52±22 min) compared with patients failing to achieve an organized rhythm before ICU admission (68±14 min; P< 0.01). The number of shocks before cannulation center arrival was lower in the pre‐ECMO group (4.0±2.0) when compared with post‐ECMO (5±3.0; P<0.001) and post‐PCI patients (5.0±4.0; P<0.001), as this would be expected with earlier acquisition of an organized nonshockable rhythm. There was no difference in the dose of epinephrine administered among the groups.
    Obstructive CAD was present in 192 of 289 (66%) patients. Of these, 87 of 192 (45%) had an organized rhythm pre‐ECMO, 57 (30%) achieved an organized rhythm after perfusion with ECMO, 37 (19%) converted to an organized rhythm post‐PCI, and 11 (6%) did not achieve an organized rhythm before ICU admission (Figure 2). This distribution was significantly different compared with patients without obstructive CAD (P<0.0005), in which 61 of 97 (63%) converted to an organized rhythm before arrival at the cannulation center, 30 (31%) achieved an organized rhythm after ECMO initiation, and 6 (6%) did not achieve an organized rhythm before ICU admission (Figure 2). The majority of patients in both groups, 144 of 192 (75%) with obstructive CAD and 91 of 97 (94%) without obstructive CAD, achieved an organized rhythm without PCI. A similar small portion of both groups failed to achieve an organized rhythm despite all interventions before ICU admission. Furthermore, the metabolic and gas exchange profile of obstructive and non‐obstructive CAD patients at arrival to the catheterization laboratory was not statistically significantly different (Table S2).
    The percentage of patients with CAD in the pre‐ECMO group was 59% (87/148), post‐ECMO 66% (57/87), post‐PCI 100% (37/37), and ICU admit group 65% (11/17; P<0.002; Table 2). Left main disease was significantly more prevalent in the severely refractory ICU admit group (5/11 [45%]) compared with 7 of 87 (8%) in pre‐ECMO, 4 of 57 (7%) post‐ECMO, and 7 of 37 (18%) post‐PCI (P<0.01). Yet, the presence of LAD, LCx, RCA, chronic total occlusion, multivessel, or proximal disease was not statistically different among the resuscitation stages.
    Coronary Artery Disease Profile by Resuscitation Stage With an Organized Cardiac Rhythm
    CAD indicates obstructive coronary artery disease; CTO, chronic total occlusion; ECMO, extracorporeal membrane oxygenation; ICU, intensive care unit; LAD, left anterior descending artery; LCx, left circumflex artery; LM, left main artery; Multivessel, multivessel obstructive CAD; and RCA, right coronary artery.
    Comparison with P≤0.05.
    Gensini and SYNTAX scores were calculated to provide an objective assessment of CAD burden in all patients. The median Gensini score for the post‐PCI group (65; interquartile range [IQR = 44]) was significantly higher than scores of the pre‐ECMO (24 [IQR = 80]) and post‐ECMO (32 [IQR = 82]) groups (P<0.01 and P<0.05, respectively, Figure 3A, left panel). The median SYNTAX score for the post‐PCI group (22 [IQR = 27]) was also significantly higher than both the pre‐ECMO and post‐ECMO groups, which were both 9.0 (IQR = 23, P<0.002 and P<0.02, respectively; Figure 3B, left panel). No further differences in Gensini and SYNTAX scores were observed between the remaining group comparisons. Gensini and SYNTAX scores were then calculated only in patients with obstructive CAD (Figure 3A through 3B, right panels). Interestingly, there were no statistically significant differences in either score between the resuscitation stages in patients with obstructive CAD. Indeed, the median Gensini score in the pre‐ECMO group was 72 (IQR = 66), post‐ECMO 61 (IQR = 67), post‐PCI 66 (IQR = 43), and ICU admit 84 (IQR = 106); and the median SYNTAX score in the pre‐ECMO group was 22 (IQR = 15), post‐ECMO 20 (IQR = 18), post‐PCI 22 (IQR = 16), and ICU admit 27 (IQR = 14).
    Gensini (A) and SYNTAX (B) scores in all patients and those with obstructive coronary artery disease. CAD indicates coronary artery disease; ECMO, extracorporeal membrane oxygenation; ICU, intensive care unit; PCI, percutaneous coronary intervention; and SYNTAX, Synergy Between PCI With Taxus and Cardiac Surgery.
    Of the 192 patients with obstructive CAD, 165 (86%) had angiographically successful PCI. Overall, PCI was successfully performed in 83%, 86%, 100%, and 82% of patients with obstructive CAD in the pre‐ECMO, post‐ECMO, post‐PCI, and ICU admit groups, respectively. The coronary distribution for revascularization was not statistically different among the groups (left main coronary artery P=0.22; LAD P=0.59; LCx P=0.3; RCA P=0.93; multivessel disease P=0.25; respectively) (Table 3).
    Revascularization Profile by Resuscitation Stage With an Organized Cardiac Rhythm
    CTO indicates chronic total occlusion; ECMO, extracorporeal membrane oxygenation; LAD, left anterior descending artery; LCx, left circumflex artery; LM, left main artery; Multivessel, multivessel obstructive coronary artery disease; PCI, percutaneous coronary interventions; and RCA, right coronary artery.
    Overall, 117 of all 289 (40%) patients had neurological favorable survival with a CPC score 1 or 2. Of the patients with CPC score 1 or 2, 74 of 117 (63%) converted to an organized rhythm pre‐ECMO, 34 (29%) achieved an organized rhythm post‐ECMO, 7 (6%) converted to an organized rhythm post‐PCI, and 2 (2%) failed to achieve an organized rhythm before admission to the ICU (Figure 4). For patients with poor neurologic outcomes (CPC score 3–5), 74 of 172 (43%) had an organized rhythm before arriving at the cannulation center, 53 (31%) were successfully defibrillated after ECMO initiation, 30 (17%) required PCI to achieve an organized rhythm, and 15 (9%) failed to achieve an organized rhythm before ICU admission. When comparing the groups based on neurologic outcomes, significantly more patients with CPC score 1 or 2 achieved an organized rhythm before ECMO cannulation, and fewer patients required PCI to achieve an organized rhythm (P<0.0002). There were also fewer patients failing to achieve an organized rhythm before ICU admission.
    Organized rhythm distribution based on neurologically favorable survival. CPC indicates cerebral performance category; ECMO, extracorporeal membrane oxygenation; ICU, intensive care unit; and PCI, percutaneous coronary intervention.
    The presence of obstructive CAD was not associated with neurologic outcomes, with nearly identical rates of obstructive CAD in both CPC score 1 or 2 and CPC score 3 to 5 groups (66% and 67%, respectively, Table 4). When looking at the distribution of the obstructive CAD, patients with poor neurologic outcome (CPC score 3–5) more commonly had severe left main coronary artery disease compared with CPC score ≤2 (20/115 [17%] versus 3/77 [4%], respectively, P<0.005). There were no differences between the neurologic outcome groups in the presence of disease in the LAD, LCx, RCA, chronic total occlusion, multivessel, or proximal disease. The frequency of PCI was similar in CPC score ≤2 (90%) and CPC score 3 to 5 (83%) patients who had obstructive CAD. Also, revascularization of the left main, LAD, LCx, RCA, and multivessel territories were not significantly different between CPC groups, P=0.83, 0.24, 0.34, 0.98 and 0.30, respectively (Table 5).
    Coronary Artery Disease Profile by Neurologic Outcome as Demonstrated by Cerebral Performance Category
    CAD indicates obstructive coronary artery disease; CPC, Cerebral Performance Category; CTO, chronic total occlusion; LAD, left anterior descending artery; LCx, left circumflex artery; LM, left main artery; Multivessel, multivessel obstructive CAD; and RCA, right coronary artery.
    Comparison with P≤0.05.
    Revascularization Profile Separated by Neurologic Outcome as Demonstrated by Cerebral Performance Category
    CPC indicates Cerebral Performance Category; CTO, chronic total occlusion; LAD, left anterior descending artery; LCx, left circumflex artery; LM, left main artery; Multivessel, multivessel obstructive coronary artery disease; PCI, percutaneous coronary interventions; and RCA, right coronary artery.
    Comparison with P≤0.05.
    When grouping by CPC, the median Gensini and SYNTAX scores in all patients (Figure 5A through 5B left panels) or in only those with obstructive CAD (Figure 5A through 5B, right panels) failed to demonstrate significant differences. In patients with CPC score 1 or 2 and CPC score 3 to 5, the median Gensini scores were 32 (IQR = 79) and 32 (IQR = 70), respectively; and the median SYNTAX scores were 10 (IQR = 23) and 11 (IQR = 15) in CPC score ≤2 and CPC score 3 to 5, respectively. Median Gensini scores in patients with obstructive CAD with CPC score ≤2 and CPC score 3 to 5 were 68 (IQR = 53) and 73 (IQR = 66), respectively. Median SYNTAX scores in patients with obstructive CAD with CPC score ≤2 and CPC score 3 to 5 were 20 (IQR = 16) and 23 (IQR = 13), respectively.
    Gensini (A) and SYNTAX (B) scores in all patients and those with obstructive coronary artery disease. CAD indicates coronary artery disease; CPC, cerebral performance category; and SYNTAX, Synergy Between PCI With Taxus and Cardiac Surgery.
    Highlighting survival as a function of each resuscitation stage, we found that 50% of the pre‐ECMO patients had a CPC ≤2 versus 39% post‐ECMO, 19% post‐PCI, and 12% in the ICU admit group (P<0.001; Figure 6, left panel). Additionally, in patients with obstructive CAD, 45% had a CPC score ≤2 at the pre‐ECMO stage, 50% post‐ECMO, 19% post‐PCI, and 9% in the ICU admit group (P<0.001; Figure 6, middle panel). In patients with obstructive CAD who received PCI, 47% had a CPC score ≤2 at the pre‐ECMO stage, 55% at post‐ECMO, 20% at post‐PCI, and 11% in the ICU admit group (P<0.001; Figure 6A, right panel).
    A, Neurologically favorable distribution through each resuscitation stage. B, Adjusted odds ratio of neurologically favorable survival when achieving an organized rhythm before PCI compared with those with an organized rhythm after PCI. CAD indicates coronary artery disease; ECMO, extracorporeal membrane oxygenation; ICU, intensive care unit; and PCI, percutaneous coronary intervention.
    The adjusted OR for neurologically favorable survival (Figure 6B) when factoring age, pH, CPR duration, and lactic acid levels, comparing patients achieving an organized rhythm before PCI and those with an organized rhythm after PCI was 3.1 (95% CI, 1.1–9.7; P=0.049) in all patients, 3.4 (95% CI, 1.1–11.0; P=0.039) in those with obstructive CAD, and 3.9 (95% CI, 1.2–12.0; P=0.024) for those who underwent PCI. Thus, the occurrence of neurologically favorable survival was more likely when achieving an organized rhythm before coronary revascularization.

    Discussion

    In this analysis of a patient population with refractory OHCA undergoing ECPR for refractory VF, several hypothesis‐generating conclusions arise. First, half of all patients achieved an organized rhythm before ECMO initiation with conversion to an organized rhythm despite the presence of significant CAD. The addition of perfusion with ECMO was sufficient to attain an organized rhythm in an additional 30% of patients, suggesting a role for ECMO in supplementing coronary perfusion. An organized rhythm was not obtained in 13% of patients until PCI was performed; thus, the role of PCI during ECPR in achieving a stable organized rhythm might be limited but critical for this subset of patients. Second, neurologically favorable survival decreased as patients failed to acquire an organized rhythm throughout the resuscitation stages. Although the mechanism of this decreased survival is unclear, differences in CPR duration, presence of ROSC, lactic acid levels, and pH support a partial link with cardiac arrest severity and ease of conversion into an organized rhythm. In our study, an absolute difference of 31% for neurologically favorable survival between patients with rhythm conversion pre‐ECMO compared with those who require PCI to convert to an organized rhythm is significant. This differential magnitude was independent of obstructive CAD presence. Third, there were no significant differences in coronary disease anatomical distribution, complexity, or revascularization profile when evaluating for CPC or resuscitation stages. Of note, patients without obstructive CAD were more likely to achieve an organized cardiac rhythm pre‐ECMO, but these differences are not sufficient for prediction of CAD status based on this rhythm conversion.
    More than half of patients with OHCA are refractory to standard ACLS therapies, resulting in a mortality rate of >90%.1 In patients for whom CAD may have a role in triggering and sustaining the shockable rhythm, ACLS does not address the underlying pathophysiological process. ECPR has significantly improved neurologically favorable survival in patients with refractory OHCA by bypassing the heart to provide systemic perfusion.2 The impact of this perfusion is multifold. ECMO‐mediated systemic perfusion halts ischemic injury and metabolic derangement in critical tissues such as the brain and heart. This supports overall survival while potentially assisting with recovery of cardiac rhythm as shown in this study. The role of ECMO‐mediated cardiac perfusion is unclear with respect to return of cardiac function and was not assessed in this study.
    ECMO‐based stabilization also allows continued treatment of the underlying cause for cardiac arrest, including coronary revascularization if needed. Importantly, most patients in our study achieved an organized rhythm before coronary angiography and PCI, suggesting that CAD is irresponsible for maintaining refractory shockable OHCA in a limited number of cases. However, the additive effect of reduced coronary flow due to stenosis in combination with reduced global perfusion during CPR may play a more prominent role as 30% of patients in our study achieved an organized rhythm after systemic perfusion with ECMO. The ability to achieve an organized rhythm with CPR leading to conversion to pulseless electrical activity suggests that the metabolic demands of cardiac electrical signaling can be maintained with considerably lower coronary perfusion compared with pumping function. One may hypothesize that VF is triggered by a stochastic mechanism in which CAD increases its risk. However, this study suggests that refractory VF requires a more persistent mechanism, including systemic hypoperfusion, due to the limitation of conventional CPR in some patients.
    Obstructive coronary artery disease has been reported to account for 70% to 80% of cardiac arrest due to shockable rhythms.1, 3, 4, 5, 6 Early revascularization during cardiac arrest attempts to restore myocardial perfusion, stabilize the metabolic derangements within the myocardium, reduce cardiac necrosis, and quell the proarrhythmogenic substrate.15 Although this approach has improved outcomes in cardiac arrest patients with ST‐segment elevations,16, 17, 18 it has been called into question when ST‐elevation is not present. Four randomized trials, PEARL (Early Coronary Angiography Versus Delayed Coronary Angiography),19, 20 EMERGE (Emergency vs Delayed Coronary Angiogram in Survivors of Out‐of‐Hospital Cardiac Arrest),16 TOMAHAWK (Immediate Unselected Coronary Angiography Versus Delayed Triage in Survivors of Out‐of‐Hospital Cardiac Arrest Without ST‐Segment Elevation),21 and COACT 22, 23 showed no survival difference between emergent versus delayed coronary angiography and PCI. However, these trials significantly differ from our study as they included patients with cardiac arrest who had already achieved ROSC, thus not in a refractory state. They also excluded patients with cardiogenic shock, signs of significant myocardial damage, electrical instability, and ongoing ischemia in clear contrast to this current study. In addition, <40% of all patients in the aforementioned studies had successful PCI compared with our population in which 66% of all patients had obstructive CAD and 86% of those had successful PCI. Although our study revealed that revascularization was necessary to achieve an organized rhythm in only 13% of patients, the overall effect of PCI on cardiac recovery and prevention of subsequent pathology remains unknown. Thus, the optimal timing of coronary angiography and PCI in the ECPR population might differ greatly from the population studied in PEARL,19, 20 EMERGE,16 TOMAHAWK,21 and COACT.22, 23
    Our study population is also distinct compared with the recently published ECLS‐SHOCK (Extracorporeal Life Support in Cardiogenic Shock) trial.24 Specifically, we used ECMO for patients with refractory cardiac arrest and ongoing CPR. Furthermore, the median CPR duration in our study was 56 minutes. In the ECLS‐SHOCK study, 77.7% of patients underwent CPR, but exclusion criteria were met if CPR duration was >45 minutes or ROSC was not present at the time of randomization. All patients in our study would be considered Society for Cardiovascular Angiography & Interventions class E shock. Almost half the patients in the ECLS‐SHOCK trial were Society for Cardiovascular Angiography & Interventions shock stage C and almost two‐thirds were either stage C or D. In agreement with the lower acuity seen in the ECLS‐SHOCK trial where patients had a median lactate of 6.8 mmol/L, our patients median lactate was 11 mmol/L. Therefore, although our study and ECLS‐SHOCK both explore the use of ECMO for patients with cardiogenic shock, we intentionally selected patients at very different severities in the cardiogenic shock spectrum.
    This study has several limitations. Patients analyzed in our study had VT or VF as the presenting cardiac arrest rhythm and they were in a refractory state, thus they represent a subset of OHCA that may not be generalizable to all patients with OHCA. Our study was performed in a single center with extensive expertise and use of ECMO; therefore, some of these observations and outcomes may not be similar in centers with less experience. The number of patients in this study may prevent discernment of small differences between groups or substantial differences between small subgroups. Given the inability to measure rhythm continuously during CPR in the prehospital setting, it is impossible to know if transient organized rhythms were achieved between pulse checks in this prehospital setting. This could cause underestimation of these short conversions to organized rhythm before venoarterial ECMO. Also, the cause of pulseless electrical activity in those patients who convert to an organized rhythm is unclear in many cases and may have significant implications for prognosis.25

    Conclusions

    In this study including patients with refractory OHCA due to shockable presenting rhythms, most patients achieved an organized rhythm before coronary angiography and PCI. Half of all patients achieved an organized rhythm before ECMO whereas nearly one‐third of patients only achieved an organized rhythm after perfusion with ECMO was initiated. Coronary angiography and PCI were necessary before achieving an organized rhythm in 13% of patients. This observation was independent of CAD, coronary disease complexity, and revascularization profile. Importantly, neurologically favorable survival was more prevalent in patients who achieved an organized rhythm before PCI; however, neurologically favorable survival was still observed in 19% of patients after revascularization. Thus, the role of PCI in long‐term cardiac recovery and neurologically favorable survival remains unknown.

    Sources of Funding

    This work was supported by R61 HL142696/HL/National Heart, Lung, and Blood Institute, National Institutes of Health, Department of Health and Human Services.

    Disclosures

    None.

    Supporting information

    Tables S1–S2
    GR

    Dr. Ganesh Raveendran

    Cardiology · Interventional Cardiology
    University of Minnesota
    ganeshan