SCAI cardiogenic shock classification after out of hospital cardiac arrest and association with outcome

We aimed to validate the Society for Cardiovascular Angiography and Interventions (SCAI) classification to evaluate association with outcome in a real‐world population and effect of invasive therapies.


| INTRODUCTION
Out of Hospital Cardiac Arrest (OOHCA) occurs in over a quarter of a million patients a year and presents a major public health challenge. 1 Hemodynamic instability is present in approximately 50% of patients after OOHCA and several studies have shown a relationship between the development of shock and poor outcome after OOHCA, including neurological outcome. 2,3 Clinical trials of cardiogenic shock (CS) have been composed of significant proportions of patients with OOHCA, 4,5 but the pathophysiology of shock in these patients may represent a distinct clinical entity due to the presence of concomitant post-arrest myocardial stunning (PAMS) and systemic inflammatory response syndrome (SIRS). This is reflected in a heterogeneity of outcomes after OOHCA with mortality attributed to multi-organ dysfunction syndrome and cardiac etiology death in −40% of patients only. This indicates a requirement for improved early classification of CS in patients with OOHCA which might support emergency selection of invasive therapies on arrival to a heart attack center, particularly revascularization and use of mechanical circulatory support (MCS) devices. 6 CS has classically been defined as a systolic blood pressure < 90 mmHg for >30 min without vasopressor support or requirement of these agents. 7 The Society for Cardiovascular Angiography and Interventions (SCAI) have recently developed a novel CS classification to reflect a spectrum of shock which has been endorsed by multiple societies. 6 In a recent validation of this classification in patients admitted to cardiac intensive care units, there was a signal that OOHCA significantly increases mortality across the spectrum of shock grades. However, in this study, the classification was applied a later time-point which would not guide early use of invasive therapies and also did not evaluate the relationship of SCAI grade with mode of death nor the relative effect of these invasive therapies on outcome.
Accordingly, the purpose of this study was to validate the SCAI shock classification in a cohort of primary cardiac etiology OOHCA particularly on admission to a Heart Attack Centre-atime-point that enables initiation of emergency therapies-and to evaluate the association with 30-day and 12-month mortality. In addition, we sought to characterize requirement for renal replacement therapy and mode of death by shock grade.

| Data collection
Data was collected using a dedicated database based on the Utstein style recommendations. 8,9 Prehospital data was collected including initial rhythm, use of bystander cardiopulmonary resuscitation (CPR) and time of return of spontaneous circulation (ROSC). Zero-flow time was defined as time from the cardiac arrest to commencement of CPR and the low-flow time as duration of CPR until ROSC. Hospital data including arterial blood gas results such as pH and lactate and routine blood tests, which were all collected on admission were recorded. In addition, we recorded baseline cardiovascular investigations including electrocardiography, left ventricular ejection fraction and results of CAG where appropriate.
Significant coronary artery disease was defined as a lesion of 70% severity in a single angiographic plane. A culprit lesion was classified as an atherothrombotic occlusion with presence of thrombus and/or easy passage of the coronary guidewire and any lesion over 70% in a single angiographic plane that the clinician treated as a culprit or with evidence of less than TIMI III flow. These definitions are in accordance with previously published articles in this area. 10,11 MCS devices were implanted at clinicians' discretion.

| SCAI shock classification
The SCAI grade is a consensus-based classification based on hemodynamic, metabolic, clinical, and physical examination parameters. 6 Briefly, this classification is a pragmatic tool which classifies CS from A-E with "A" modifier signifying the occurrence of cardiac arrestnamely-At risk, Beginning Shock, Classic Shock, Deteriorating Shock and Extremis. 6 In accordance with the classification, we defined the SCAI shock grades using objective markers of hemodynamics including blood pressure and heart rate, vasopressor requirements and presence of renal failure (study definition shown in Table 1). Patients were ascribed the SCAI shock classification retrospectively by an independent clinician who was blinded to patient outcome and mode of death and this was defined as the worst grade of shock within the first 1 hr of arrival to the cardiac catheterization laboratory.

| Outcome
The prespecified primary end-point was mortality at 30 days. The principal secondary end-point was mortality at 12 months. Additional secondary end-points were poor neurological outcome (defined as CPC 3-5-severe disability to death) 12 and mode of death classified as neurological, 13 multiorgan dysfunction syndrome 14 or cardiac etiology (refractory arrhythmia or refractory CS) at 12 months in accordance with previous studies. 5 Follow-up was tracked using the medical records and by telephone contact and patients with incomplete data were excluded from the final analysis.

| Statistical analysis
Normally distributed data are represented as mean ± standard deviation and non-normally distributed data are presented as medians with ranges or interquartile range (IQR) as appropriate. Comparisons were performed with Student's t test or ANOVA for continuous, normally distributed variables; Chi-square for proportions and Mann-Whitney for non-normally distributed variables. After stratification by baseline SCAI class, survival curves were constructed using the Kaplan-Meier method, and compared using a log rank test. Event times were measured from time of index cardiac arrest until the end-point. We constructed a multivariate logistic regression model with forced inclusion of variables independently associated with mortality in this population from existing literature (age, initial cardiac arrest rhythm, and lactate) as well as early coronary angiography based on SCAI grade. Results are presented as odds ratios (ORs) with 95% confidence intervals (CIs). A p < .05 was considered statistically significant. All analyses were undertaken using SPSS version 25.   Figure 2).

| Association with mode of death and requirement for renal replacement therapy
While absolute mortality rate increased across the shock grades, we also observed a significant difference in relative mode of death between the different classes ( Figure 3 (Table 6). We did find a significant interaction between performance of early CAG and initial rhythm in Grades A-C but not in Grades D-E (Table S1).

| DISCUSSION
In this study, we validated the SCAI shock classification in OOHCA patients on arrival to a specialist center and our principal findings are that, when applied at this early time-point, this grade is associated with mortality at 30 days and at 12 months. Higher SCAI shock grades were also associated with mortality attributed to multi-organ dysfunction syndrome and cardiac etiology death and requirement for renal replacement therapy. Finally, in a selected group of patients with CS and OOHCA, we found that early CAG might potentially be associated with improved outcome in SCAI grade D.
CS is observed in 40-70% of OOHCA patients and is associated with increased mortality. 4  or left ventricular predominance. [16][17][18][19] These classifications individually confer specific advantages but, to date, no classification has been developed specifically for patients with CS after OOHCA and the majority have significant limitations for applicability in an acute setting. While it is known that hemodynamic instability after OOHCA classically deteriorates over a 72 hr period after the event, we were most interested in application of the grade on arrival since, it is at this point, that a decision on an early invasive approach with MCS can be made.
The SCAI classification of shock is a consensus-based classification which was designed in order to be a pragmatic and practical tool that could be easily and specifically applied on arrival to center, partly to support selection of therapies. While developed by experts in the field and with endorsement of several societies, the classification is empirical and was not previously validated in a clinical cohort in an acute setting. A recent study by Jentzer et al retrospectively validated the SCAI classification in a large cohort of patients in the first 24 hr after admission to cardiac intensive care units and found a similar association with hospital mortality as our study. 20 There was a signal from this study that the prognosis after OOHCA was particularly poor but the proportion of these patients in this registry was low, there was significant heterogeneity of etiologies and with no data on mode of death, which is an important consideration in this group of patients.
In our study, we found that the SCAI classification is easily applied on arrival to a specialist center with accurate reflection of the spectrum of CS after OOHCA, with association with early and late mortality as well as requirement for renal replacement therapy.

| Limitations
This was a single center, retrospective study and while this has enabled collection of a rich dataset and a standardized protocol of care, it also introduces the possibility of bias. For example, the decision, or otherwise, to perform CAG and PCI were ultimately based on clinical discretion. Therefore, it cannot be excluded that any association of improved outcome with early CAG might be attributed to an element of patient selection and requires further validation. Second, the SCAI shock classification is practical and consensus-based and an individual patient might at one time possess specific characteristics of several grades. Hence, we sought to objectively define the shock grade using hemodynamic and vasopressor requirements only and the use of these variables did appear to correlate well with shock severity.
Third, the rates of usage of advanced MCS (Impella and ECMO) were relatively low, which reflects absence of guideline support and current practice for OOHCA across many centers. These findings might not therefore reflect outcome where VA-ECMO and E-CPR for OOHCA is performed routinely.
Fourth, the numbers in this study were relatively small which reduced the power, so the findings should be viewed as hypothesis generating and require robust validation in larger prospective studies.
Finally, the patients were selected on arrival to specialist heart attack center with access to 24/7 emergency angiography and MCS so the observed outcomes may not be applicable outside of this setting.

| Conclusions
The SCAI shock classification can be applied on admission to patients with primary cardiac etiology OOHCA. Higher SCAI shock grades were associated with mortality, requirement for renal replacement therapy and mortality attributed to multi-organ dysfunction syndrome