Abstract
Background
Early revascularization improves survival in patients with ST-elevation myocardial infarction–related cardiogenic shock (STEMI-CS). However, lower rates of invasive management are seen in states with public reporting of outcomes for percutaneous coronary intervention and coronary artery bypass grafting surgery. The reasons for this remain speculative. We aim to report contemporary treatment patterns and examine the reasons for deferral of invasive management in patients with STEMI-CS in a New York cohort.
Methods
All patients with STEMI-CS in the Northwell-Shock Registry, a retrospective study of cardiogenic shock across 11 hospitals in New York, treated between January 2016 and August 2022, were included. Clinical variables and outcomes were compared between patients managed invasively and conservatively. Reasons for deferral of invasive management were collected manually from clinical documentation. Multivariable logistic regression was performed to examine the factors associated with a conservative management strategy.
Results
Invasive management was performed in 87% of patients, revascularization with percutaneous coronary intervention in 63% and coronary artery bypass grafting surgery in 8%. Hospital mortality was 27% for patients managed invasively and 81% for those managed conservatively. The most cited reasons for deferring invasive management were unclear neurologic status (35%), patient or family preference (29%), and complicating medical conditions (25%). Meanwhile, the factors independently associated with conservative management were older age, higher creatinine, cardiac arrest, and higher Society for Cardiovascular Angiography & Interventions (SCAI) stages.
Conclusions
Although rates of invasive management were high in this contemporary cohort of STEMI-CS patients in New York, risk aversion seems to still play a role in the deferral of invasive management in many patients.
Keywords: cardiogenic shock, invasive management, left heart catheterization, percutaneous coronary intervention, revascularization, ST-elevation myocardial infarction
Introduction
Revascularization with percutaneous coronary intervention (PCI) leads to improved survival in patients with acute myocardial infarction (AMI) and cardiogenic shock (CS) and is currently a class I recommendation in the American and European guidelines for acute coronary syndrome.1,2 Although the uptake of PCI in AMI-CS has been increasing over the last several decades in the United States,3 these rates appear to be lower in states that have implemented public reporting of outcomes for PCI.4
Public reporting is used in some states, including New York, as a way to motivate institutions and physicians to improve patient outcomes after PCI and coronary artery bypass grafting (CABG) surgery.4,5 However, some studies have suggested that an unintended consequence of this policy is the aversion of physicians to perform invasive procedures in high-risk patients.
To improve consistency with guideline-directed care and to prevent risk aversion, “refractory” CS patients were excluded from public reporting in 2008 in New York State (NYS). According to a study of patients treated in this state between 2002-2011, this change led to an increase in the rates of invasive management (cardiac catheterization, PCI, or CABG) for AMI-CS, but failed to increase them to the rates seen in 3 other states where public reporting was not adopted.6
Important advances in CS care have occurred in the last decade,7,8 and more contemporary rates of invasive management for AMI-CS in NYS have not been reported. In addition, the reasons why physicians decide not to pursue invasive management in this context have not been previously studied.
We describe the rates of invasive management in a contemporary multicenter cohort of patients with ST-elevation myocardial infarction (STEMI)-related CS in NYS and explore the reasons behind the deferral of invasive strategies among patients managed conservatively.
Materials and methods
Study population
The Northwell-Shock Registry is a multicenter, retrospective study of patients with CS. It was approved by the Northwell Health Institutional Review Board with a waiver of informed consent. Patient information is deidentified and securely stored in a protected database. Eleven of the 23 hospitals within the Northwell Health system, sharing a common electronic medical record, 6 of which have primary PCI capabilities, are included in this registry.
All adult patients (≥18 years of age) discharged between January 2016 and August 2022 with a principal or secondary diagnosis of CS, identified by the International Classification of Diseases, 10th Revision (ICD-10) code R57.0, as well as STEMI, identified by ICD-10 codes, were included (Supplemental Table S1).
Patient-level data were obtained directly from the electronic health record by automated collection.
Clinical characteristics, management procedures, and outcomes at hospital discharge were collected for all patients. For patients undergoing PCI, basic procedural details, including the target vessel, were collected. For all other patients, reasons for not receiving invasive management were collected.
Definitions
Invasive management was defined as diagnostic left heart catheterization (LHC), PCI, or CABG.
Following review of all patient records, we were able to group the reasons for not performing invasive management or PCI into 1 of 8 categories: (1) family or patient declined based on goals of care; (2) severe neurologic dysfunction; (3) complicating condition, referring to other concomitant conditions which made the benefit of PCI unclear; (4) mechanical complication of STEMI; (5) late presentation (as quoted in documentation by the treating medical team with no predefined time cutoff); (6) referral for coronary artery bypass grafting surgery; (7) collateralized vessel on angiogram suspected to be chronic total occlusion; and (8) technical inability to perform PCI despite a documented attempt.
For patients referred to CABG, we further examined what proportion went on to receive it, and we collected the time from presentation to surgery.
Vital signs, laboratory parameters, and hemodynamics were collected at the time of admission (initial) and at the point of the worst value for each specific parameter (worst). The vasoactive-inotropic score (VIS) was calculated using the aggregate type and dosage of inotropes and vasopressors administered concomitantly, measured at admission and at the point of maximum dosing.9
The presence of pulmonary artery catheter (PAC) and different mechanical circulatory support (MCS) devices was determined using ICD-10 procedural codes (Supplemental Table S1).
Shock severity was defined at admission and the point of most severe illness using the Cardiogenic Shock Working Group (CSWG) modification of the Society for Cardiovascular Angiography & Interventions (SCAI) SHOCK stages classification.10 The Acute Physiology and Chronic Health Evaluation version IV (APACHE IV) was calculated for all patients at the time of ICU admission.11
Statistical analysis
Data analysis and graphic creation were performed using Prism (GraphPad Software) and Office (Microsoft Corp) applications. Continuous variables are presented as mean ± SD, whereas categorical variables are reported as percentages and absolute numbers.
Comparisons of continuous variables were conducted using an unpaired t test, whereas categorical variables were compared using χ2 and Fisher exact tests.
Logistic multivariate regression was conducted to assess both categorical and continuous variables. Laboratory parameters were analyzed by comparing normal values vs quartiles of the abnormal values, whereas other continuous variables were assessed as quartiles of their respective distribution. Statistical significance was determined using a threshold of P < .05.
Results
A total of 857 STEMI-related CS (STEMI-CS) admissions were identified during the study period. Following exclusions for incomplete data (n = 40) and duplicate records due to interhospital transfers (n = 24), the final cohort was composed of 793 unique patients.
Invasive management
Invasive management was defined as LHC, PCI, or CABG. A total of 690 (87%) patients underwent LHC, 499 (63%) underwent PCI, and 62 (8%) underwent CABG (Figure 1). Hence, the total number of patients receiving invasive management was 691 (87%), whereas 13% were managed conservatively. The main findings of our study are presented in the Central Illustration.
Figure 1.
Flowchart of study participants. CABG, coronary artery bypass grafting; LHC, left heart catheterization; PCI, percutaneous coronary intervention; STEMI, ST-elevation myocardial infarction.
Central Illustration.
Contemporary management of ST-elevation myocardial infarction (STEMI)-related cardiogenic shock in the New York metropolitan area. CTO, chronic total occlusion; LHC, left heart catheterization; PCI, percutaneous coronary intervention; SCAI, Society for Cardiovascular Angiography & Interventions.
Compared to patients managed conservatively, those selected for invasive management were younger in age (66.1 ± 12.5 years vs 74.7 ± 14.2 years; P < .01) and had lower comorbidity burden as reflected by a lower Charlson comorbidity index score. They also had lower rates of cardiac arrest (5% vs 18%; P < .01) and less severe hypotension at the time of admission. Initial and maximal SCAI-CSWG CS stages, as well as initial Acute Physiology And Chronic Health Evaluation (APACHE) score at the time of ICU admission, were all lower among patients managed invasively (Table 1). Moreover, the rate of invasive management decreased progressively with increasing presenting SCAI-CSWG stage. The progression of CS stages during hospital admission among patients with invasive and conservative management is depicted in Supplemental Figure S1.
Table 1.
Baseline characteristics, management strategies, and outcomes of STEMI-CS patients managed invasively and conservatively.
| Parameter | Invasive management (n = 691) | Conservative management (n = 102) | P value |
|---|---|---|---|
| Male sex | 475 (68.7) | 64 (62.7) | .23 |
| Age, y | 66.1 ± 12.5 | 74.7 ± 14.2 | <.001 |
| Length of stay, d | 12.5 ± 17.3 | 7 ± 14.9 | .001 |
| Body mass index, kg/m2 | 27.8 ± 5.5 | 25.6 ± 6.4 | .004 |
| Race | .13 | ||
| Asian | 64 (9.3) | 3 (2.9) | |
| Black | 51 (7.4) | 10 (9.8) | |
| White | 382 (55.3) | 66 (64.7) | |
| Other | 140 (20.3) | 12 (11.8) | |
| Unknown | 54 (7.8) | 11 (10.8) | |
| Admission | .75 | ||
| 2015-2019 | 364 (52.7) | 52 (51.0) | |
| 2020-2022 | 327 (47.3) | 50 (49.0) | |
| Weekend | 210 (30.4) | 26 (25.5) | .31 |
| Nighttime | 358 (51.8) | 50 (49.0) | .59 |
| Transfers | 300 (43.4) | 33 (32.4) | .035 |
| Cardiac arrest on admission | 36 (5.2) | 18 (17.6) | <.001 |
| Intensive care unit stay | 589 (85.2) | 80 (78.4) | .08 |
| Initial APACHE | 80.7 (38) | 118.9 (35.8) | <.001 |
| Past medical history | |||
| Charlson comorbidity index | 4.8 ± 2.9 | 5.9 ± 3.3 | .003 |
| Hypertension | 396 (57.3) | 64 (62.7) | .29 |
| Chronic kidney disease | 93 (13.5) | 9 (8.8) | .19 |
| Coronary artery disease | 256 (37.0) | 30 (29.4) | .13 |
| Type 2 diabetes mellitus | 244 (35.3) | 23 (22.5) | .011 |
| Atrial fibrillation | 132 (19.1) | 20 (19.6) | .90 |
| Anemia | 151 (21.9) | 19 (18.6) | .46 |
| Hemodynamics | |||
| Heart rate, bpm | 82.2 ± 24.1 | 76.3 ± 33 | .09 |
| SBP, mm Hg | 101.9 ± 27.1 | 77.6 ± 27.1 | <.001 |
| DBP, mm Hg | 58 ± 18.6 | 47.3 ± 49 | .046 |
| MAP, mm Hg | 72.7 ± 18.8 | 57.4 ± 35.5 | <.001 |
| Lowest MAP, mm Hg | 55.3 ± 13.2 | 46.7 ± 13.2 | <.001 |
| CVP, mm Hg | 12.3 ± 7.1 | 12.9 ± 7.7 | .76 |
| PAS, mm Hg | 40 ± 15.7 | 35.9 ± 13.2 | .28 |
| PAD, mm Hg | 21.4 ± 12.2 | 20 ± 9.2 | .59 |
| Mean PAP, mm Hg | 27.6 ± 12.7 | 25.3 ± 10.3 | .43 |
| CI, L/min/m2 | 2.2 ± 0.9 | 2.3 ± 0.8 | .50 |
| LVEF, % | 33.2 ± 15.3 | 30.8 ± 12.2 | .26 |
| Laboratory parameters | |||
| Lactate, mmol/L | 5.9 ± 5.2 | 7.6 ± 6.4 | .037 |
| Creatinine, mg/dL | 1.5 ± 1.0 | 2.1 ± 1.5 | <.001 |
| Sodium, mEq/L | 136.4 ± 4.8 | 136.5 ± 4.8 | .80 |
| Bilirubin, mg/dL | 0.9 ± 0.7 | 0.9 ± 0.6 | .94 |
| ALT, IU/L | 242.2 ± 607.2 | 274.2 ± 749.1 | .69 |
| WBC, k/μL | 17 ± 7.1 | 17.4 ± 8.4 | .64 |
| Hemoglobin, g/dL | 12.1 ± 2.6 | 11.7 ± 2.7 | .1 |
| Initial VIS score | 18 (6-36) | 20 (12.5-50.5) | .24 |
| Highest VIS score | 25 (10-50) | 41.5 (17.3-92.5) | .05 |
| Initial SCAI SHOCK stage | n = 685 | n = 101 | <.001 |
| A | 85 (12.4) | 4 (4.0) | |
| B | 103 (15.0) | 9 (8.9) | |
| C | 139 (20.3) | 10 (9.9) | |
| D | 141 (20.6) | 16 (15.8) | |
| E | 217 (31.7) | 62 (61.4) | |
| Worst SCAI SHOCK stage | n = 689 | n = 102 | <.001 |
| A | 0 (0.0) | 0 (0.0) | |
| B | 9 (1.3) | 0 (0.0) | |
| C | 86 (12.5) | 5 (4.9) | |
| D | 166 (24.1) | 17 (16.7) | |
| E | 428 (62.1) | 80 (78.4) | |
| Management strategies | |||
| LHC | 691 (100) | 0 (0.0) | – |
| PCI | 499 (72.2) | 0 (0.0) | – |
| CABG | 62 (9.0) | 0 (0.0) | – |
| PAC | 396 (57.3) | 59 (57.8) | .92 |
| IABP | 402 (58.2) | 9 (8.8) | <.001 |
| Impella | 166 (24.0) | 3 (2.9) | <.001 |
| VA-ECMO | 60 (8.7) | 5 (4.9) | .19 |
| Mechanical ventilation | 393 (56.9) | 69 (67.6) | .039 |
| Transfusion | 237 (34.3) | 20 (19.6) | .003 |
| Hemodialysis | 93 (13.5) | 8 (7.8) | .11 |
| Heart transplant | 3 (0.4) | 0 (0.0) | .51 |
| Durable LVAD | 2 (0.3) | 0 (0.0) | .59 |
| Outcomes | |||
| In-hospital mortality | 183 (26.5) | 83 (81.4) | <.001 |
| Transferred to other hospital | 28 (4.2) | 1 (1) | .121 |
| Discharge to a skilled facility | 275 (39.8) | 7 (6.9) | <.001 |
| Discharge to hospice | 9 (1.3) | 4 (3.9) | .05 |
| Routine discharge | 196 (28.4) | 7 (6.9) | <.001 |
| Postdischarge outcomes (total n – patients lost to follow-up) | n = 634 | n = 91 | |
| 90-Day mortality | 201 (31.7) | 83 (91.2) | <.001 |
| 90-Day hospitalization rate | 142 (20.5) | 3 (2.9) | <.001 |
Data presented as n (%), mean ± SD, or median (IQR).
ALT, alanine aminotransferase; APACHE, Acute Physiology And Chronic Health Evaluation; CABG, coronary artery bypass grafting; CI, cardiac index; CS, cardiogenic shock; CVP, central venous pressure; DBP, diastolic blood pressure; IABP, intraaortic balloon pump; LHC, left heart catheterization; LVAD, left ventricular assist device; LVEF, left ventricular ejection fraction; MAP, mean arterial pressure; PAC, pulmonary artery catheter; PAD, pulmonary artery diastolic pressure; PAP, pulmonary artery pressure; PAS, pulmonary artery systolic pressure; PCI, percutaneous coronary intervention; SBP, systolic blood pressure; SCAI, Society for Cardiovascular Angiography & Interventions; STEMI, ST-elevation myocardial infarction; VA-ECMO, veno-arterial extracorporeal membrane oxygenation; VIS, vasoactive-inotropic score; WBC, white blood cell count.
Patients managed invasively were also more frequently admitted via interhospital transfer (43% vs 32%; P = .03), but no difference was seen in the rate of invasive management for patients presenting after hours or during weekends compared to those presenting during working hours (defined as 8:00 AM to 5:00 PM).
In multivariable regression analysis, increasing age (OR, 1.8; CI, 1.4-2.4; P < .01), admission with cardiac arrest (OR, 0.5; CI, 2.2-9.1; P < .01), higher CSWG-SCAI Shock stages (OR, 0.49; CI, 0.2-0.9; P = .04), and higher baseline creatinine (OR, 1.6; CI, 1.3-1.8; P < .01), were all independently associated with a lower likelihood of receiving invasive management. Meanwhile, a history of diabetes mellitus (OR, 0.5; CI, 0.3-0.8; P = .01) was associated with a higher chance of receiving invasive management (Figure 2).
Figure 2.
Forest plot for factors associated with invasive management. Factors independently associated with invasive management were increasing age, admission with cardiac arrest, higher CSWG-SCAI stage on admission, higher baseline creatinine, and diabetes mellitus. ALT, alanine aminotransferase; CCI, Charlson comorbidity index; CSWG, Cardiogenic Shock Working Group; SCAI, Society for Cardiovascular Angiography & Interventions; WBC, white blood cell count.
The reasons cited for not performing LHC among patients managed conservatively (n = 102) were severe neurologic dysfunction (35%), family or patient preference (29%), concomitant complicating condition (25%), late presentation of STEMI (4%), and mechanical complications associated with AMI (4%) (Table 2).
Table 2.
Reasons for deferral of left heart catheterization or PCI among patients with STEMI-CS.
| Reason | Deferral of left heart catheterization (n = 102) | Deferral of PCI (n = 167) |
|---|---|---|
| Family/patient preference | 30 (29.4) | 6 (3.6) |
| Neurologic dysfunction | 36 (35.3) | 9 (5.4) |
| Complicating condition | 28 (27.5) | 5 (3.0) |
| Mechanical complication | 4 (3.9) | 17 (10.2) |
| Late presentation | 4 (3.9) | 16 (9.6) |
| Required CABG | – | 83 (49.7) |
| Collateralized; suspected CTO | – | 7 (4.2) |
| Technical inability to perform PCI | – | 25 (15.0) |
Data presented as n (%)
CABG, coronary artery bypass grafting; CS, cardiogenic shock; CTO, chronic total occlusion; PCI, percutaneous coronary intervention; STEMI, ST-elevation myocardial infarction.
Notably, only 42% of the patients who were deferred due to severe neurologic dysfunction or an unresponsive state had suffered a cardiac arrest.
Within the 28% of patients deferred due to a complicating condition, the most common reasons were: advanced age or poor functional status (14%), respiratory failure (14%), refractory arrhythmias (14%), multiorgan failure (14%), active malignancy (11%), and renal failure (7%). A full list is provided in the supplemental material (Supplemental Table S2).
Revascularization
Among the 690 patients who underwent LHC, a culprit vessel was identified in 667 (97%). Of those, 72% received PCI. The most frequent PCI targets were the left anterior descending artery in 46%, followed by the right coronary artery in 27% and the left circumflex in 11% of patients. Multivessel PCI at the index procedure was performed in 14% of the patients (Table 3).
Table 3.
Coronary vessel disease according to revascularization status and PCI target vessels.
| Underwent revascularization (n = 560) | Did not undergo revascularization (n = 131) | P value | |
|---|---|---|---|
| No. of diseased vessels | <.001 | ||
| 0 | 0 | 16 (12.2) | |
| 1 | 240 (42.9) | 44 (33.6) | |
| 2 | 167 (29.8) | 24 (18.3) | |
| ≥3 | 146 (26.1) | 40 (30.5) | |
| Unknown | 7 (1.3) | 7 (5.3) | |
| LM disease | 58 (10.4) | 22 (16.8) | .055 |
| Target vessel | n = 499 | ||
| Left main | 6 (1.2) | ||
| Left anterior descending | 231 (46.3) | ||
| Left circumflex | 57 (11.4) | ||
| Right coronary artery | 133 (26.7) | ||
| Ramus | 3 (0.6) | ||
| Bypass graft | 1 (0.2) | ||
| 2-vessel PCI | 59 (11.8) | ||
| 3-vessel PCI | 9 (1.8) |
Data presented as n (%).
PCI, percutaneous coronary intervention.
The most common reasons for not performing PCI after LHC (n = 167) were referral for CABG (49%), technical inability to perform PCI despite an attempt (15%), appearance of late presentation based on the presence of collaterals on coronary angiogram (14%), and mechanical complications of AMI (10%) (Table 2).
Overall, 11% of STEMI-CS patients were referred for CABG. Most of these (76%) ultimately underwent surgery, but only 16% received it within the first 24 hours of admission. Instead, the median time to CABG was 3 days (IQR, 1, 5; range, 0-21).
In total, 209 (26%) patients with STEMI-CS did not undergo revascularization. These patients were older, more likely to be women, had higher comorbidity burden based on the Charlson comorbidity index, and had higher presenting serum lactate and creatinine levels. They also had higher initial and maximal SCAI-CSWG CS stages, with declining rates of revascularization as presenting stages increased (Supplemental Table S3). The regression demonstrating the predictors of revascularization is presented in Supplemental Table S4.
ICU therapies and MCS use
Patients managed invasively and conservatively both had similar rates of ICU admissions as well as use of critical care therapies, including invasive mechanical ventilation, hemodynamic monitoring with PAC, and acute hemodialysis. Meanwhile, the use of vasoactive medications (VIS) was higher among patients managed conservatively, and the need for blood transfusions was higher among those managed invasively.
Mechanical circulatory support use was higher among patients who were managed invasively, driven by higher use of IABP and Impella CP, whereas the utilization of veno-arterial extracorporeal membrane oxygenation was low and similar in both groups.
The most cited reason for deferring PCI among patients who received MCS was referral for CABG (52%).
Outcomes
Hospital mortality (27% vs 81%; P < .01) was lower among patients receiving invasive management. Mortality rates were comparable among patients undergoing invasive management independent of revascularization status: diagnostic LHC only 34% vs LHC and PCI or CABG 27%; P = .07. Hospital mortality was higher in the conservative management arm of patients in SCAI stages B, D, and E, and was only numerically higher in stages A and C (Figure 3).
Figure 3.
Mortality according to treatment strategy by presenting Society for Cardiovascular Angiography & Interventions stage. Mortality was significantly higher in patients receiving conservative management after presenting with Society for Cardiovascular Angiography & Interventions stages of B, D, and E, and was numerically higher for those with stages A and C.
At 90 days, 57 patients (11%) in the invasive management group and 11 (58%) in the conservative group were lost to follow-up. Among those with available follow-up information, the all-cause mortality rate reached 32% and 91%, respectively. However, the all-cause rehospitalization rate was 21% in the invasive group and 3% in the conservative group (P < .001 for all).
Among patients managed conservatively, mortality varied according to the reason cited for invasive management deferral. The highest mortality was seen in patients with severe neurologic dysfunction (94%), followed by those in whom the family or patient declined (80%), a concomitant complicating condition (76%), those with a mechanical complication of AMI (75%), followed by those with late presenting myocardial infarction (25%).
Among patients who did not receive PCI after LHC, the highest mortality was also seen in patients with neurologic dysfunction (89%), followed by deferral based on family or patient wishes (67%), technical inability to perform PCI (52%), late presentation STEMI (44%), mechanical complication of AMI (35%), and those with a concomitant complicating condition (20%) (Figure 4).
Figure 4.
Mortality according to the reason for deferral of invasive management. Mortality rates varied based on the reasons for deferring left heart catheterization (LHC) and percutaneous coronary intervention (PCI). The highest mortality rates were observed in patients deferred due to neurologic dysfunction and patient or family preferences, followed by patients conservatively managed because of complicating concomitant conditions. CABG, coronary artery bypass grafting; CTO, chronic total occlusion.
Mortality among patients who underwent CABG was 6%. However, for those who were referred for CABG but did not ultimately receive surgery, hospital mortality was 48%. The reasons cited for deferral of CABG were not collected.
Discussion
In this multicenter registry of 793 patients with STEMI-CS in NYS, we report on the contemporary rates of invasive management and describe the reasons cited by treating physicians for deferring it among patients managed conservatively. The main findings of our study are as follows: (1) the majority of patients (87%) underwent invasive management, and 71% received revascularization via PCI or CABG; (2) patients managed conservatively faced a nearly 3-fold higher hospital mortality rate (81% vs 27%) compared to those treated invasively; (3) severe neurologic dysfunction (35%), family/patient preference (29%), and complicating medical conditions (27%) were the primary reasons cited for deferring diagnostic LHC, whereas referral for CABG (49%) and technical inability to perform PCI (15%) were the main reasons for deferral of primary PCI; (4) rates of invasive management declined with higher SCAI stages at presentation, but invasive treatment was associated with much lower mortality in patients with the most severe shock stages; and (5) cardiac arrest (OR = 0.23) was the factor most strongly associated with conservative management on adjusted analysis.
Mortality rates in this study were very high among patients with the most common reasons for deferral of invasive management. Together with the low and comparable rates of mortality among patients treated invasively regardless of revascularization status, this finding supports the notion that patient selection was adequately informed by a grim clinical outlook. However, to what extent mortality could be modified in these patients as a result of invasive management is unclear.
For instance, although poor neurologic status was the reason for deferral of invasive treatment in 35% of the patients, less than half of them had been resuscitated from cardiac arrest. Neurologic dysfunction can be a manifestation of severe shock, and although it is associated with poorer outcomes,12 most unresponsive patients with AMI-CS survived to hospital discharge after revascularization with CABG in a report from the Society of Thoracic Surgeons Database.13 In addition, even among patients with out-of-hospital cardiac arrest, neurologic injury has been reported to be the cause of death in only 4% of CS patients without out-of-hospital cardiac arrest.14 Hence, this group could potentially benefit from revascularization, especially because early prediction of neurologic recovery can be highly inaccurate.15,16 Moreover, patients with altered mental status have not been excluded from contemporary revascularization trials in AMI-CS, which underscores this potential benefit.7
Patients cited as having concomitant complicating conditions such as renal or other extracardiac organ failure, as well as arrhythmias, are another group who could benefit from invasive management but whose deferral from invasive procedures could be the result of physician risk aversion. Risk aversion also seemed to play a role among patients referred for surgery because a high proportion of patients referred for CABG after LHC did not undergo surgery (24%), and there were significant delays in the time to CABG for those who ultimately did undergo surgery.
Our study suggests that despite the exclusions of refractory CS from public reporting, its influence on treatment selection persists among patients with STEMI and CS.
This issue appears to particularly impact patients in whom the assessment of futility may be fraught with subjectivity, particularly early in the hospital course. These patients tend to be those with the most complex clinical pictures. In fact, patients with the worst severity of shock at presentation were more frequently selected for conservative management, even though their outcomes may have still been modifiable with timely intervention.
These findings are not unique to this study. By 2016, it was reported that lower rates of invasive management persisted in NYS compared to states without public reporting, despite the exclusion of refractory CS from public reporting.6 Moreover, patients managed conservatively in our cohort were treated aggressively in other ways, with similar rates of ICU admission, PAC placement, hemodialysis, and even relatively high rates of MCS. The apparent ongoing risk aversion in performing LHC and revascularization in STEMI-CS apparent in our findings, may be in part explained by the nuanced definition of the conditions that allow exclusion from reporting systems.
Firstly, the definition “refractory cardiogenic shock” within the NYS Percutaneous Coronary Intervention Reporting System differs from the one used in the National Cardiovascular Data Registry and requires a systolic blood pressure <80 mm Hg and/or a cardiac index <2.0 L/min/m2 immediately prior to PCI, despite the use of pharmacological or mechanical support. Patients who meet CS criteria but are not on this level of support are instead categorized as hemodynamically unstable and remain included in reporting, which may contribute to hesitancy in pursuing invasive management.17 Our study demonstrates that most patients present without refractory shock, which is stages D or E within SCAI classification, whereas the mortality in stages A to C is very high, particularly in the conservative arm, making the application of this criterion even more limited.
Secondly, “anoxic brain injury,” which can also be used for exclusion, has a nuanced definition that requires the presence of cardiac arrest before arrival to the catheterization laboratory and the proper documentation of an inadequate response to external stimuli. As observed in our cohort, half of the patients with altered mental status did not have prior cardiac arrest, and how many would meet this definition of a comatose state is unknown. This again limits the applicability of this exclusion criterion.
Future directions
These findings highlight the need for refined policies regarding reporting of STEMI outcomes after PCI to ensure that patients who may benefit from invasive management are not treated conservatively due to risk aversion. The exclusion from reporting of patients with any type of CS and/or neurologic dysfunction, regardless of cardiac arrest, should be strongly considered to simplify physician decision-making and lower the threshold for intervention, particularly because this decision is made within a constrained timeframe.16
A more radical and perhaps more impactful approach to improve the quality of care in this population could be to move beyond a focus on PCI reporting and adopt a system that captures the outcomes for STEMI-CS patients by hospital, regardless of the treatment received. A large body of emerging research could provide adequate definitions of CS as well as risk adjustment tools for this purpose. This approach would incentivize rather than disincentivize the use of potentially life-saving interventions.
Limitations
Our study is based on retrospective observational data and is subject to the biases inherent to this study design. In addition, both CS and STEMI were identified using ICD-10 diagnostic codes, and this may limit the generalizability of our findings. Also, although our registry is multicenter, it is limited to a single health system, and practice patterns specific to this system may influence our results. We also lack information on the vascular access site used in patients treated invasively, which can impact complication rates and outcomes directly. Lastly, because we do not have contemporary data from states without public reporting, we cannot assess whether the practice patterns reported here are exclusive to NYS or directly related to public reporting of outcomes.
Conclusion
In conclusion, this study demonstrates that although proper assessments of futility appear central in the decision to defer invasive management in STEMI-CS in NYS, there is evidence that physician risk aversion and subjective judgment also play a significant role. The exclusion from public reporting systems of STEMI patients with “non-refractory” CS, neurologic dysfunction, or unresponsive state, regardless of cardiac arrest, has the potential to enhance the adoption of evidence-based care and improve outcomes for this high-risk population.
Acknowledgments
Declaration of competing interest
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding sources
This work was not supported by funding agencies in the public, commercial, or not-for-profit sectors.
Ethics statement and patient consent
This study was approved by the institutional review board of Northwell Health and was conducted in accordance with the Declaration of Helsinki. A waiver of informed consent was obtained due to the retrospective nature of the study and minimal to no risk to participants.
Footnotes
To access the supplementary material accompanying this article, visit the online version of the Journal of the Society for Cardiovascular Angiography & Interventions at 10.1016/j.jscai.2025.103926.
Contributor Information
Abduljabar Adi, Email: aadi@northwell.edu.
Miguel Alvarez Villela, Email: malvarezvill@northwell.edu.
Supplementary material
References
- 1.Lawton J.S., Tamis-Holland J.E., Bangalore S., et al. 2021 ACC/AHA/SCAI guideline for coronary artery revascularization: a report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation. 2022;145(3):e18–e114. doi: 10.1161/CIR.0000000000001038. [DOI] [PubMed] [Google Scholar]
- 2.Byrne R.A., Rossello X., Coughlan J.J., et al. 2023 ESC guidelines for the management of acute coronary syndromes. Eur Heart J. 2023;44(38):3720–3826. doi: 10.1093/eurheartj/ehad191. [DOI] [PubMed] [Google Scholar]
- 3.Vallabhajosyula S., Prasad A., Gulati R., Barsness G.W. Contemporary prevalence, trends, and outcomes of coronary chronic total occlusions in acute myocardial infarction with cardiogenic shock. Int J Cardiol Heart Vasc. 2019;24 doi: 10.1016/j.ijcha.2019.100414. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Apolito R.A., Greenberg M.A., Menegus M.A., et al. Impact of the New York State Cardiac Surgery and Percutaneous Coronary Intervention Reporting System on the management of patients with acute myocardial infarction complicated by cardiogenic shock. Am Heart J. 2008;155(2):267–273. doi: 10.1016/j.ahj.2007.10.013. [DOI] [PubMed] [Google Scholar]
- 5.Feldman D.N., Yeh R.W. Public reporting of percutaneous coronary intervention mortality in New York State: are we helping our patients? Circ Cardiovasc Qual Outcomes. 2017;10(9) doi: 10.1161/CIRCOUTCOMES.117.004027. [DOI] [PubMed] [Google Scholar]
- 6.Bangalore S., Guo Y., Xu J., et al. Rates of invasive management of cardiogenic shock in New York before and after exclusion from public reporting. JAMA Cardiol. 2016;1(6):640–647. doi: 10.1001/jamacardio.2016.0785. [DOI] [PubMed] [Google Scholar]
- 7.Thiele H., Akin I., Sandri M., et al. PCI strategies in patients with acute myocardial infarction and cardiogenic shock. N Engl J Med. 2017;377(25):2419–2432. doi: 10.1056/NEJMoa1710261. [DOI] [PubMed] [Google Scholar]
- 8.Møller J.E., Engstrøm T., Jensen L.O., et al. Microaxial flow pump or standard care in infarct-related cardiogenic shock. N Engl J Med. 2024;390(15):1382–1393. doi: 10.1056/NEJMoa2312572. [DOI] [PubMed] [Google Scholar]
- 9.Gaies M.G., Gurney J.G., Yen A.H., et al. Vasoactive-inotropic score as a predictor of morbidity and mortality in infants after cardiopulmonary bypass. Pediatr Crit Care Med. 2010;11(2):234–238. doi: 10.1097/PCC.0b013e3181b806fc. [DOI] [PubMed] [Google Scholar]
- 10.Kapur N.K., Kanwar M., Sinha S.S., et al. Criteria for defining stages of cardiogenic shock severity. J Am Coll Cardiol. 2022;80(3):185–198. doi: 10.1016/j.jacc.2022.04.049. [DOI] [PubMed] [Google Scholar]
- 11.Zimmerman J.E., Kramer A.A., McNair D.S., Malila F.M. Acute Physiology and Chronic Health Evaluation (APACHE) IV: hospital mortality assessment for today’s critically ill patients. Crit Care Med. 2006;34(5):1297–1310. doi: 10.1097/01.CCM.0000215112.84523.F0. [DOI] [PubMed] [Google Scholar]
- 12.Kataja A., Tarvasmäki T., Lassus J., et al. Altered mental status predicts mortality in cardiogenic shock – results from the CardShock study. Eur Heart J Acute Cardiovasc Care. 2018;7(1):38–44. doi: 10.1177/2048872617702505. [DOI] [PubMed] [Google Scholar]
- 13.Cox M.L., Gulack B.C., Thibault D.P., et al. Outcomes after coronary artery bypass grafting in patients with myocardial infarction, cardiogenic shock and unresponsive neurological state: analysis of the Society of Thoracic Surgeons Database. Eur J Cardiothorac Surg. 2018;54(4):710–716. doi: 10.1093/ejcts/ezy114. [DOI] [PubMed] [Google Scholar]
- 14.Davodian L.W., Larsen J.K.P., Povlsen A.L., et al. Timing and causes of death in acute myocardial infarction complicated by cardiogenic shock (from the RETROSHOCK cohort) Am J Cardiol. 2022;171:15–22. doi: 10.1016/j.amjcard.2022.01.050. [DOI] [PubMed] [Google Scholar]
- 15.Tamis-Holland J.E., Menon V., Johnson N.J., et al. Cardiac catheterization laboratory management of the comatose adult patient with an out-of-hospital cardiac arrest: a scientific statement from the American Heart Association. Circulation. 2024;149(5):e274–e295. doi: 10.1161/CIR.0000000000001199. [DOI] [PubMed] [Google Scholar]
- 16.Sandroni C., Cariou A., Cavallaro F., et al. Prognostication in comatose survivors of cardiac arrest: an advisory statement from the European Resuscitation Council and the European Society of Intensive Care Medicine. Intensive Care Med. 2014;40(12):1816–1831. doi: 10.1007/s00134-014-3470-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Hannan E.L., Zhong Y., Cozzens K., et al. The impact of excluding shock patients on hospital and physician risk-adjusted mortality rates for percutaneous coronary interventions: the implications for public reporting. JACC Cardiovasc Interv. 2017;10(3):224–231. doi: 10.1016/j.jcin.2016.10.040. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.