- Email: [email protected]
Cardiogenic shock complicating acute coronary syndromes: Insights from the Global Registry of Acute Coronary Events
Clinical Investigations
Acute Ischemic Heart Disease
Cardiogenic shock complicating acute coronary syndromes: Insights from the Global Registry of Acute Coronary Events Hamza H. Awad, MD, PhD, a Frederick A. Anderson, Jr, PhD, a Joel M. Gore, MD, a Shaun G. Goodman, MD, MSc, b and Robert J. Goldberg, PhD a Worcester, MA; and Ontario, Canada
Introduction Despite advances in the management of patients with an acute coronary syndrome (ACS), cardiogenic shock (CS) remains the leading cause of death in these patients. The objective of this observational study was to describe the characteristics, management, and hospital outcomes of patients with an ACS complicated by CS. Our secondary study objective was to describe trends in the incidence and hospital case-fatality rates (CFRs) of CS and predictors of increased hospital mortality in these high-risk patients. Methods The population consisted of patients enrolled in the GRACE study between 1999 and 2007 who were hospitalized with an ACS. Results During the years under study, 2,992 patients (4.6%) developed CS. Patients with CS were more likely to be older, have a history of diabetes or atrial fibrillation, and present with a higher pulse rate or cardiac arrest. Cardiac catheterization was performed on 1,706 (57%) and in-hospital revascularization on 1,408 patients (47%) with CS. Patients with CS were less likely to receive evidence-based cardiac medications compared with patients who did not develop CS. The in-hospital CFR of patients with CS was 59.4%, compared with 2.3% in those who did not develop CS. Factors associated with an increased risk of dying in patients with CS included advanced age, diabetes mellitus, angina, and stroke. Adjusted incidence rates and hospital CFRs of CS showed modest declines over time. Conclusion Continued efforts are needed to reduce the incidence and CFRs of CS complicating ACS. (Am Heart J 2012;163:963-71.) Cardiogenic shock (CS) remains the most serious clinical complication and the leading cause of death for patients hospitalized with an acute coronary syndrome (ACS). 1,2 Despite recent studies suggesting possible declines in the risk of dying during hospitalization for patients with CS, which has been linked to advances in medical treatment, coronary revascularization techniques, and mechanical support, in-hospital case-fatality rates (CFRs) associated with CS remain high, exceeding 50%. 3 Moreover, despite the clinical importance of this complication, the incidence rates of CS in patients with an ACS have remained relatively constant over the past 30 years, averaging approximately 7%. 3-5 Using data from
a large multinational coronary disease registry, we describe the demographic and clinical characteristics of patients presenting with an ACS complicated by CS, the management of these high-risk patients, and their shortterm outcomes in comparison with patients who did not develop CS. A secondary study objective was to describe nearly decade-long (1999-2007) trends in the incidence and CFRs of CS complicating ACS and factors associated with an increased risk of dying in patients with CS.
Methods The full details of the methods used in the GRACE study have been previously published. 6,7
Patient population a
b
From the University of Massachusetts Medical School, Worcester, MA, and Canadian Heart Research Centre, St. Michael's Hospital, University of Toronto, Toronto, Ontario, Canada. Submitted October 17, 2011; accepted March 12, 2012. Reprint requests: Robert J. Goldberg, PhD, Division of Epidemiology of Chronic Diseases and Vulnerable Populations, Department of Quantitative Health Sciences, University of Massachusetts Medical School, 55 Lake Avenue North, Worcester, MA 01655. E-mail: [email protected] 0002-8703/$ - see front matter © 2012, Mosby, Inc. All rights reserved. doi:10.1016/j.ahj.2012.03.003
This large multinational prospective study was designed to reflect an unselected population of patients with an ACS irrespective of geographic region. A total of 123 hospitals located in 14 countries in North and South America, Europe, Australia, and New Zealand have contributed data to this observational study. Adult patients (≥18 years old) admitted with a presumptive diagnosis of ACS to participating hospitals were potentially eligible. Patients with noncardiovascular causes for their clinical presentation were excluded.
American Heart Journal June 2012
964 Awad et al
Table I. Characteristics of patients with an ACS complicated by CS Patients with ACS complicated by CS
Characteristic Age (y) mean (SD) Age (y) mean (SD) b55 55-64 65-74 ≥75 Female, mean (SD) Medical history, mean (SD) Angina Atrial fibrillation Chronic heart failure Coronary artery disease Current smoking Diabetes mellitus Hypertension Myocardial infarction PCI CABG Renal insufficiency Stroke/TIA Clinical presentation, mean (SD) Pulse Systolic blood pressure Cardiac arrest STEMI Killip class on admission I II III IV
Total (n = 2992)
Presenting with CS on admission (n = 666)
Developing CS in-hospital after admission (n = 2326)
Patients with ACS not complicated by CS (n = 62 134)
n (%)
n (%)
n (%)
n (%)
71.01 (12.5)
69.8 (12.7)
71.4 (12.5)⁎
65.42 (13.2) †
368 (12.4) 529 (17.8) 817 (27.5) 1262 (42.4) 1143 (38.3)
93 (14.1) 140 (21.2) 177 (16.9) 249 (37.8) 249 (37.7)
275 (11.9) 388 (16.8) 640 (27.6) 1013 (43.7)⁎ 894 (38.6)
14 685 (23.8) 14 760 (23.8) 16 102 (26.0) 16 367 (26.5) † 20 100 (32.5) †
1346 (45.4) 263 (8.9) 516 (17.5) 665 (22.8) 1475 (50.1) 916 (31.0) 1768 (59.9) 837 (28.3) 309 (10.5) 272 (9.2) 325 (11.0) 324 (11.0)
232 (35.3) 56 (8.6) 115 (17.7) 138 (21.5) 324 (49.6) 166 (25.5) 364 (55.5) 193 (29.4) 73 (11.2) 57 (8.7) 49 (7.5) 48 (7.3)
1113 (48.2)⁎ 207 (9.0) 401 (17.4) 526 (23.1) 1150 (50.2) 750 (32.5)⁎ 1403 (61.1)⁎ 644 (27.8) 236 (10.2) 215 (9.3) 276 (12.0)⁎ 276 (12.0)⁎
32 608 (52.7) † 4690 (7.6) † 6090 (9.9) † 19 098 (31.3) † 35 762 (57.8) † 15 388 (24.9) † 38 386 (62.1) † 18 710 (30.3) † 11 045 (17.9) † 7746 (12.5) † 4600 (7.4) † 5070 (8.2) †
87.1 (28.3) 117.1 (37.9) 368 (12.5) 1909 (63.8)
86.0 (34.7) 94.9 (41.6) 193 (29.3) 414 (62.2)
87.4 (26.2) 123.1 (34.4)⁎ 175 (7.6)⁎ 1495 (64.3)
79.1 (20.8) † 142.5 (29.0) 933 (1.5) † 21 809 (35.1) †
1240 (42.5) 634 (21.7) 379 (13.0) 666 (22.8)
0 0 0 666 (100)
1240 (55.0) 634 (28.1) 379 (16.8) 0 (0)⁎
51 709 (85.1) 6982 (11.5) 2060 (3.4) None †
TIA, Transient ischemic attack. ⁎ P b .05 comparing patients presenting with CS with those developing it later during hospitalization. † P b .05 comparing patients with CS with patients without CS.
Prospective (warm pursuit) and retrospective (cold pursuit) surveillance approaches for identifying cases of ACS were used, as has been previously described. 6,7 Where required, study investigators received approval from their local hospital ethics or institutional review board. Standardized definitions of all patient-related variables, clinical diagnoses, and selected hospital complications and outcomes were based on the American College of Cardiology key data elements and definitions for measuring the clinical management and outcomes of patients hospitalized with ACS. 8 Patients were defined as having an ACS if they had symptoms typical of an ACS accompanied by at least 1 of the following: electrocardiographic changes consistent with ACS, serial increases in biochemical markers of cardiac necrosis (creatine kinase-MB fraction, creatine phosphokinase, or troponin), and documented coronary artery disease. Cardiogenic shock was defined as a systolic blood pressure of b80 mm Hg and congestive heart failure (Killip class IV) occurring at any time during the short-term hospitalization. 9 The specific timing of shock onset was not collected, but patients presenting
to the hospital with CS were differentiated from those developing shock in-hospital by Killip class on hospital presentation. Coronary revascularization was defined as the receipt of a percutaneous coronary intervention (PCI) or coronary artery bypass graft (CABG) at any time during the index hospitalization.
Data collection The hospital records of patients with a validated ACS were abstracted for demographic and clinical data, complications occurring during hospitalization, electrocardiographic findings, and use of diagnostic procedures and therapeutic approaches. Data were stored and analyzed at the Center for Outcomes Research at the University of Massachusetts Medical School.
Data analysis Continuous variables were summarized by their means and SDs or medians, as appropriate. Categorical variables were summarized by counts and percentages. Univariate comparisons of patient characteristics, clinical presentation, treatments
American Heart Journal Volume 163, Number 6
Awad et al 965
Table II. Treatment of patients with an ACS complicated by CS Patients with ACS complicated by CS
Treatment Procedures Cardiac catheterization PCI Patients undergoing PCI (% stent) CABG IABP Pulmonary artery catheter Medications ACE inhibitors Angiotensin receptor blockers Aspirin β-Blockers Calcium-channel blockers Fibrinolysis GP IIb/IIIa inhibitor Nitrates Pressors/inotropes
Total (n = 2993)
Presenting with CS on admission (n = 666)
Developing CS in-hospital after admission (n = 2326)
Patients with ACS not complicated by CS (n = 62 134)
n (%)
n (%)
n (%)
n (%)
1706 (57.3) 1190 (40.1) 977 (88.0) 272 (9.2) 852 (29.0) 945 (32.0)
375 (56.6) 294 (44.3) 250 (90.6) 28 (4.2) 155 (23.9) 181 (27.7)
1331 (57.5) 896 (38.9)⁎ 727 (87.2) 244 (10.6)⁎ 697 (30.5)⁎ 764 (33.3)⁎
37 823 (61.4) † 23 622 (38.3) 21 309 (93.2) † 3107 (5.1) † 892 (1.5) † 2285 (3.7) †
1595 (53.9) 97 (3.3) 2512 (84.4) 1656 (56.1) 368 (12.6) 627 (21.2) 894 (30.4) 1916 (64.7) 2368 (79.8)
309 (46.8) 16 (2.4) 519 (78.5) 294 (44.8) 52 (8.0) 127 (19.4) 178 (37.1) 295 (45.2) 489 (74.0)
1286 (56.0)⁎ 81 (3.6) 1993 (86.1)⁎ 1362 (59.4)⁎ 316 (15.0)⁎
40 497 (65.7) † 3532 (5.8) † 58 130 (93.7) † 52 625 (85.2) † 14 439 (23.6) † 8003 (13.0) † 15 798 (25.7) † 49 875 (80.8) † 7030 (11.5) †
499 (21.7) 716 (31.3)⁎ 1620 (70.3)⁎ 1878 (81.5)⁎
ACE, Angiotensin-converting enzyme; GP, glycoprotein. ⁎ P b .5 comparing patients presenting with CS with those developing it later during hospitalization. † P b .5 comparing patients with CS with patients without CS.
prescribed, and hospital outcomes between patients with and without CS were carried out using the Wilcoxon rank sum or χ 2 test, as appropriate. Short-term survival rates were estimated using the KaplanMeier method, and log-rank tests were used for between group (CS present vs absent) comparisons. The short-term prognosis in each study year was examined by calculating in-hospital and 30-day postadmission CFRs. A logistic multivariable regression analysis was used to examine changes over time in the incidence rates of CS and in-hospital CFRs while controlling for several potentially confounding demographic (eg, age or sex) and clinical (eg, prior comorbidities or ACS type) factors of prognostic importance. These variables were considered as potential confounders based on our univariate results and the findings from previous research. The HosmerLemeshow goodness-of-fit test was used to examine the adequacy of our logistic regression models. Unadjusted and multivariable-adjusted Cox proportional hazards models were used to determine factors associated with an increased risk of dying in the hospital among patients who developed CS. Given the nonrandomized nature of the present study and the caveats and difficulties involved in the interpretation of any multivariable-adjusted estimates of association, we did not control for the hospital use of cardiac treatments in our regression analyses in which hospital or 30-day postadmission survival status was the key outcome. All analyses were performed with STATA version 11.0 (Stata Corporation, College Station, TX). No extramural funding was used to support this work. The authors are solely responsible for the design and conduct of this study, all study analyses, and drafting and editing of the paper.
Results Baseline characteristics The study population consisted of 65,119 patients hospitalized with an ACS who were enrolled in GRACE between 1999 and 2007. Of these, 4.6% (n = 2,992) developed CS during their index hospitalization for an ACS. Of all patients with CS, 22.3% presented with CS on admission to the hospital, whereas the remainder of patients with CS developed it later during their index hospitalization. These proportions remained relatively unchanged during the years under study. Compared with patients without CS, patients who developed shock were more likely to be older (6 years on average), women, and to have a prior history of diabetes, atrial fibrillation, chronic heart failure, renal insufficiency, and stroke. Patients with CS were more likely to have a higher pulse rate, an ST-segment elevation myocardial infarction (STEMI), and to be in cardiac arrest compared with patients without CS at the time of hospital admission (Table I). Multivariableadjusted predictors of developing CS were older age (adjusted odds ratio [OR] per year 1.03, 95% CI 1.011.04), history of diabetes mellitus (OR 1.20, 95% CI 1.101.31) or atrial fibrillation (OR 1.34, 95% CI 1.16-1.54), and presenting with a higher pulse rate (OR 1.01, 95% CI 1.01-1.02) or cardiac arrest (OR 9.04, 95% CI 7.84-10.43). Multivariable predictors among patients with STEMI were similar to the predictors in the overall population, whereas additional predictors included a history of
American Heart Journal June 2012
966 Awad et al
Table III. Univariate and multivariate predictors of dying during hospitalization for patients with an ACS complicated by CS
Characteristic Age (y), mean (SD) Age (y) b55 55-64 65-74 ≥75 Female Medical history Angina Atrial fibrillation CAD PCI CABG Current smoking Diabetes mellitus Heart failure Hypertension Renal insufficiency Stroke/TIA Clinical presentation Presented with shock Cardiac arrest ST-segment elevation Non–ST-segment elevation
Died (n = 1775)
Survived (n = 1213)
n (%)
n (%)
P
73.9 (11.6)
66.8 (12.7)
b.01
139 (7.9) 239 (13.5) 460 (26) 929 (52.6) 733 (41.5)
229 (19) 289 (24) 355 (29.5) 331 (47.9) 409 (33.9)
b.01 b.01
872 (49.7) 188 (10.8) 419 (24.2) 177 (11.0) 170 (9.7) 763 (43.9) 591 (33.8) 356 (20.4) 1090 (62.5) 225 (12.8) 230 (13.2)
472 (39.0) 74 (6.2) 244 (20.6) 132 (10.2) 102 (8.5) 709 (58.9) 324 (26.9) 160 (13.3) 675 (56.0) 100 (8.3) 94 (7.8)
b.01 b.01 .02 .48 .24 b.01 b.01 b.01 b.01 b.01 b.01
322 (18.7) 228 (13.0) 1094 (61.7) 412 (23.2)
344 140 814 288
b.01 .28 b.01 .17
(29.0) (11.7) (67.1) (23.7)
Multivariable-adjusted hazards ratio (95% CI)
1 1.22 (09-1.6) 1.79 (1.4-2.4) 3.83 (3.0-5.0)
1.62 (1.3-1.8)
0.70 (0.6-0.8) 1.41 (1.2-1.6)
1.54 (1.1-1.9)
CAD, Coronary artery disease.
hypertension or myocardial infarction in patients with non–ST-segment elevation ACS.
cated by shock compared with patients without shock (Table II).
Treatment for patients with CS Although patients with ACS who developed CS were less likely to undergo cardiac catheterization (57.3% vs 61.4%), once they underwent angiography, they were more likely to undergo PCI (40.1% vs 38.3%) or CABG (9.2% vs 5.1%) compared with patients without shock. Stent use was significantly lower in patients with shock who underwent PCI compared with patients who did not develop CS (88.0% vs 93.2%) (Table II). Pulmonary artery catheterization and intraaortic balloon pumps [IABPs] as well as cardiac supportive drugs, including inotropes/pressors, were used significantly more often in patients with CS compared with patients without shock (Table II). Fibrinolytic therapy was administered more often to patients with shock compared with patients without CS (21.2% vs 13.0%). Patients with CS were less likely to have received angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, aspirin, β-blockers, calcium-channel blockers, low-molecular-weight heparin, and nitrates than patients without CS during their index hospitalization (Table II). Of note, only glycoprotein IIb/IIIa inhibitors and unfractionated heparin were used more frequently among ACS cases compli-
Short-term CFRs The crude in-hospital CFR for patients with CS was 59.4% compared with 2.3% in patients without shock (P b .001). Hospital mortality was markedly lower for patients who underwent coronary revascularization compared with those in whom a less aggressive approach was adopted (45.3% vs 72.0%, P b .001). Patients developing CS during their short-term hospitalization were at significantly higher risk for dying in-hospital compared with patients presenting to the hospital in CS (62.6% vs 48.4%, P b .001). In examining differences in possible prognostic factors in patients who died, as compared with those who survived CS, older patients and those with a history of diabetes, heart failure, hypertension, and renal insufficiency were more likely to die after developing CS than respective comparison groups (Table III). Multivariable-adjusted survival regression models showed that older age and a history of either angina, diabetes mellitus, or stroke were associated with an increased risk of dying among patients who developed CS. There were no differences in the predictors of death for patients with CS according to the presence or absence of ST-elevation in their baseline electrocardiogram.
American Heart Journal Volume 163, Number 6
Awad et al 967
Figure 1
Trends in hospital incidence rates of cardiogenic shock among patients with acute coronary syndromes.
Models that further included time of development of shock showed that patients with CS who survived their short-term hospitalization were significantly younger and were more likely to have presented to the hospital in CS. The overall crude 30-day CFR for patients with ACS complicated by CS was 59% compared with 2.3% for patients who did not develop CS during their index hospitalization (P b .001) (Table III). Significant univariate as well as multivariable-adjusted predictors of an increased risk of dying during the 30 days after development of ACS for patients with CS were similar to the predictors of in-hospital mortality.
In examining the association between in-hospital death rates and the shock index (SI) (SI = heart rate/systolic blood pressure), patients with SI b0.8 experienced an inhospital CFR of 54.9%, whereas patients with an SI ≥0.8 had a CFR of 64.5%. In addition, in categorizing patients according to SI quartiles, the corresponding in-hospital CFRs were 52.9%, 53.8%, 60.4%, and 67.4%, respectively.
Time trends in the incidence rates of CS The crude hospital incidence rates of CS decreased from 5.1% in 1999 to 3.6% in 2007. Trends in incidence
968 Awad et al
American Heart Journal June 2012
Figure 2
Trends in hospital case-fatality rates for acute coronary syndrome patients with and without cardiogenic shock.
rates by type of ACS and by time of development of shock are shown in Figure 1. The multivariable-adjusted risk of developing CS, adjusted for age, sex, medical history, clinical presentation, and length of hospital stay, showed significant declines over the years under study (annual OR 0.97, 95% CI 0.95-0.98).
Time trends in hospital CFRs The crude in-hospital CFRs among patients with CS declined from 62% in 1999 to 54% in 2005, with an increase in these death rates in 2007 (Figure 2). Trends in hospital CFRs by type of ACS and by time of development of shock are shown in Figure 3. In-hospital CFRs of patients with CS adjusted for age, sex, medical history, clinical presentation, length of hospital stay, and study site showed significant declines during the years under study (annual OR 0.94, 95% CI 0.90-0.99).
Discussion The results of our large multinational observational study provide insights into the magnitude, management, short-term mortality rates, and the characteristics of patients with an ACS likely to develop and survive CS. Encouragingly, our study showed declining short-term mortality and incidence rates related to CS during the years under study.
Baseline characteristics Identifying patients at increased risk for developing CS and providing these high-risk patients with urgent medical care is essential to decreasing the risk of developing this serious complication among patients
hospitalized with an ACS. Proper monitoring, risk stratification, and aggressive intervention have been associated with improved survival among patients developing CS. 10-13 Our study demonstrated that patients with ACS at high risk for developing CS were older, more likely to be female, diabetic, having chronic heart failure, and more likely to present to the hospital with STEMI and cardiac arrest. Similar to previously reported findings from studies in patients with an ACS, less than onequarter of all ACS patients with CS presented to the hospital with CS on admission. 14 Our results are similar to previously published findings carried out in different population settings including the observational Worcester Heart Attack Study and the GUSTO and SHOCK randomized trials. 10,15,16 These findings create a profile of individuals with an ACS who are at particularly increased risk for developing CS. Several of these high-risk individuals can be identified in advance and should be placed under increased hospital surveillance as well as considered for the receipt of effective aggressive treatment modalities, including coronary revascularization.
Treatment for patients with CS Previously published results from clinical trials and observational studies have demonstrated that early mechanical revascularization in patients with CS has been associated with lower mortality compared with initial medical stabilization (including IABP counterpulsation and fibrinolytic therapy) followed by late or no revascularization. 11,17 Based on these findings, the American College of Cardiology and the American Heart Association have elevated early mechanical revascularization for CS to a class I
American Heart Journal Volume 163, Number 6
Awad et al 969
Figure 3
Trends in hospital case-fatality rates of cardiogenic shock among patients with acute coronary syndromes.
recommendation for patients b75 years old 18 and a class IIA recommendation for those N75 years old who are considered to be suitable candidates. 19 However, the availability of clinical practice guidelines does not necessarily translate to changes in “real-world” practice. 20 The results of our study showed that N40% of patients with CS were N75 years old, a proportion that is consistent with the previously published literature. 4,10 Treatment for this high-risk group and appropriately selecting patients eligible for invasive management remain an important clinical challenge. Although patients with CS were less likely to undergo cardiac catheterization than patients who did not develop CS, once they were catheterized, they were more likely to
undergo PCI or CABG. This can likely be explained by the presence of more extensive myocardial damage, a greater frequency of multivessel underlying disease, and because of their clinically unstable status. Patients with shock were also more likely to receive fibrinolysis compared with patients who did not develop CS. Although the benefits of fibrinolysis in patients with CS are less established compared with patients with acute myocardial infarction, this reperfusion approach has been convincingly shown to reduce the risk of shock, 21 which is of considerable importance because most patients develop CS after hospital presentation (N6 hours). 9,10 The DANAMI-2 trial, which randomly assigned patients with
970 Awad et al
STEMI to either PCI or fibrinolysis, found no difference in the proportion of patients developing CS between the 2 comparison groups. 22 Patients with CS were also less likely to receive evidence-based medications compared with patients without CS. Although these medications are associated with better outcomes in patients with ACS, several of these therapies might aggravate hypotension in patients with CS. Inasmuch, these medications are often withheld until the patient is stabilized. 23 Because of the lack of information on the exact timing of development of CS in our study, we were unable to determine the relationship between the time of onset of CS and the prescribing of in-hospital medications or procedures. As expected, patients developing CS were more likely to receive supportive cardiac care. Future studies should continue to monitor the use of various treatment approaches in this high-risk patient population and identify patient groups less likely to be treated with evidence-based treatments.
In-hospital and 30-day CFRs Cardiogenic shock remains the most frequent cause of in-hospital death among patients with an ACS. Historically, CFRs associated with the development of CS in previous studies have ranged from 50% to 80%. However, and despite the persistently high death rates associated with CS, the prognosis for patients hospitalized with CS has improved considerably over the past several decades. 4,11,17,24 Declines in hospital mortality rates have been associated with advances in supportive care and a more aggressive approach to coronary revascularization. 11,17 As expected, and consistent with the previous literature, patients with CS were significantly more likely to die in-hospital than patients who did not develop this complex hemodynamic complication. 14 Several randomized trials 11,17 and reports from observational studies suggest that PCI often results in improved short-term survival in patients with CS with survival contingent on the successful establishment of coronary reperfusion. 9 Uncontrolled observational studies have suggested an association between the receipt of CABG surgery and an improvement in short-term survival among patients with CS when they are treated soon after shock has developed. 15 However, we were unable to assess the role of these interventional procedures because we could not determine whether shock preceded or followed the use of these treatment strategies or determine the reasons why certain patients received these therapeutic regimens and others did not (confounding by indication). Consistent with the findings from previously published studies, older age, diabetes mellitus, and a history of stroke were associated with an increased risk of inhospital mortality among patients developing CS. How-
American Heart Journal June 2012
ever, developing CS during hospitalization was associated with the highest risk of dying during hospitalization among the possible predictors of in-hospital mortality among patients with CS. A recently published study from the APEX-AMI trial on patients with STEMI developing CS reported similar predictors of death among this highrisk population. 14 Future studies should explore factors associated with patient's changing hemodynamic status during hospitalization for an ACS and the risk of developing CS in this high-risk population.
Trends in the incidence and in-hospital CFRs of CS It is difficult to compare the actual incidence rates of CS in published studies because of varying population characteristics, definitions of CS, and whether patients developing shock in the prehospital setting were included. 4,25 The incidence of CS reported from previous studies has ranged from 3% to 15% 26,27 depending on the study population characteristics. The overall incidence rate of CS reported in our study (4.6%) falls within this range. Moreover, we demonstrated that the incidence rates of CS declined over the period under study, consistent with findings from previous studies examining trends in CS incidence rates during the 2000s. 15,28 Potential contributors to encouraging declines in the incidence rates of CS include the increased adoption of early revascularization of patients with ACS, increased use of evidence-based cardiac medications over time, and enhanced patient monitoring. The results of our study show a declining trend in the short-term mortality rates associated with CS between 1999 and 2006 with an increase in these death rates in 2007. Although the reasons for this increase in crude short-term CFRs during our most recent study year remain unexplained, declines in the multivariable-adjusted hospital death rates of patients with CS occurred during the years under study when other covariates of prognostic importance were adjusted for. It is likely that the increased and early use of aggressive treatment regimens and supportive cardiac care contributed to these observed declines in adjusted hospital CFRs. Study strengths and limitations Our study strengths include the use of standardized data collection efforts in a large multinational registry of patients hospitalized with an ACS and inclusion of patients hospitalized during relatively recent periods. The limitations of our study include its observational, nonrandomized nature and lack of information on the exact timing of development of CS. Furthermore, we did not collect information on patient's race/ethnicity and socioeconomic status and were unable to systematically characterize the severity of the ACS.
American Heart Journal Volume 163, Number 6
Conclusions The results of this large observational study provide insights into the characteristics, management practices, and short-term mortality of patients with ACS complicated by CS. Although the magnitude of CS appears to be slowly declining, it still develops at a relatively high rate after ACS. The hospital death rate among patients with this complication remains high despite advances in management practices and declining rates over time. It remains of considerable importance to examine contemporary trends in the magnitude and short-term outcomes associated with CS with serial reassessment of baseline demographic, clinical, and laboratory risk factors. Furthermore, it remains of necessity to monitor the ever-changing hemodynamic status of patients hospitalized with an ACS and to develop more refined risk stratification systems for identifying patients most likely to develop this devastating complication as well as those at high risk for dying from it.
References 1. Domanski MJ, Topol EJ. Cardiogenic shock: current understandings and future research directions. Am J Cardiol 1994;74:724-6. 2. Califf RM, Bengtson JR. Cardiogenic shock. N Engl J Med 1994;330: 1724-30. 3. Hollenberg SM, Kavinsky CJ, Parrillo JE. Cardiogenic shock. Ann Intern Med 1999;131:47-59. 4. Goldberg RJ, Samad NA, Yarzebski J, et al. Temporal trends in cardiogenic shock complicating acute myocardial infarction. N Engl J Med 1999;340:1162-8. 5. Rogers WJ, Canto JG, Lambrew CT, et al. Temporal trends in the treatment of over 1.5 million patients with myocardial infarction in the US from 1990 through 1999: the National Registry of Myocardial Infarction 1, 2 and 3. J Am Coll Cardiol 2000;36:2056-63. 6. Rationale and design of the GRACE (Global Registry of Acute Coronary Events) Project: a multinational registry of patients hospitalized with acute coronary syndromes. Am Heart J 2001;141:190-9. 7. Steg PG, Goldberg RJ, Gore JM, et al. Baseline characteristics, management practices, and in-hospital outcomes of patients hospitalized with acute coronary syndromes in the Global Registry of Acute Coronary Events (GRACE). Am J Cardiol 2002;90:358-63. 8. Cannon CP, Battler A, Brindis RG, et al. American College of Cardiology key data elements and definitions for measuring the clinical management and outcomes of patients with acute coronary syndromes. A report of the American College of Cardiology Task Force on Clinical Data Standards (Acute Coronary Syndromes Writing Committee). J Am Coll Cardiol 2001;38:2114-30. 9. Dauerman HL, Goldberg RJ, White K, et al. Revascularization, stenting, and outcomes of patients with acute myocardial infarction complicated by cardiogenic shock. Am J Cardiol 2002;90:838-42. 10. Hochman JS, Boland J, Sleeper LA, et al. Current spectrum of cardiogenic shock and effect of early revascularization on mortality. Results of an International Registry. SHOCK Registry Investigators. Circulation 1995;91:873-81. 11. Hochman JS, Sleeper LA, Webb JG, et al. Early revascularization in acute myocardial infarction complicated by cardiogenic shock. SHOCK Investigators. Should we emergently revascularize occluded coronaries for cardiogenic shock. N Engl J Med 1999;341:625-34.
Awad et al 971
12. Berger PB, Holmes Jr DR, Stebbins AL, et al. Impact of an aggressive invasive catheterization and revascularization strategy on mortality in patients with cardiogenic shock in the Global Utilization of Streptokinase and Tissue Plasminogen Activator for Occluded Coronary Arteries (GUSTO-I) trial. An observational study. Circulation 1997;96:122-7. 13. Holmes Jr DR, Califf RM, Van de Werf F, et al. Difference in countries' use of resources and clinical outcome for patients with cardiogenic shock after myocardial infarction: results from the GUSTO trial. Lancet 1997;349:75-8. 14. French JK, Armstrong PW, Cohen E, et al. Cardiogenic shock and heart failure post-percutaneous coronary intervention in ST-elevation myocardial infarction: observations from “Assessment of Pexelizumab in Acute Myocardial Infarction”. Am Heart J;162:89-97. 15. Goldberg RJ, Spencer FA, Gore JM, et al. Thirty-year trends (1975 to 2005) in the magnitude of, management of, and hospital death rates associated with cardiogenic shock in patients with acute myocardial infarction: a population-based perspective. Circulation 2009;119: 1211-9. 16. Hasdai D, Califf RM, Thompson TD, et al. Predictors of cardiogenic shock after thrombolytic therapy for acute myocardial infarction. J Am Coll Cardiol 2000;35:136-43. 17. Hochman JS, Sleeper LA, White HD, et al. One-year survival following early revascularization for cardiogenic shock. JAMA 2001; 285:190-2. 18. Ryan TJ. Early revascularization in cardiogenic shock—a positive view of a negative trial. N Engl J Med 1999;341:687-8. 19. Antman EM, Anbe DT, Armstrong PW, et al. ACC/AHA guidelines for the management of patients with ST-elevation myocardial infarction: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to Revise the 1999 Guidelines for the Management of Patients with Acute Myocardial Infarction). Circulation 2004;110:e82-292. 20. Curry SJ. Organizational interventions to encourage guideline implementation. Chest 2000;118(2 Suppl):40S-6S. 21. Effect of intravenous APSAC on mortality after acute myocardial infarction: preliminary report of a placebo-controlled clinical trial. AIMS Trial Study Group. Lancet 1988;1:545-9. 22. Lindholm MG, Boesgaard S, Thune JJ, et al. Percutaneous coronary intervention for acute MI does not prevent in-hospital development of cardiogenic shock compared to fibrinolysis. Eur J Heart Fail 2008;10: 668-74. 23. Gunnar RM, Bourdillon PD, Dixon DW, et al. ACC/AHA guidelines for the early management of patients with acute myocardial infarction. A report of the American College of Cardiology/American Heart Association Task Force on Assessment of Diagnostic and Therapeutic Cardiovascular Procedures (subcommittee to develop guidelines for the early management of patients with acute myocardial infarction). Circulation 1990;82:664-707. 24. Bengtson JR, Kaplan AJ, Pieper KS, et al. Prognosis in cardiogenic shock after acute myocardial infarction in the interventional era. J Am Coll Cardiol 1992;20:1482-9. 25. An international randomized trial comparing four thrombolytic strategies for acute myocardial infarction. The GUSTO investigators. N Engl J Med 1993;329:673-82. 26. Sanborn TA, Feldman T. Management strategies for cardiogenic shock. Curr Opin Cardiol 2004;19:608-12. 27. Davies CH. Revascularization for cardiogenic shock. QJM 2001;94: 57-67. 28. Babaev A, Frederick PD, Pasta DJ, et al. Trends in management and outcomes of patients with acute myocardial infarction complicated by cardiogenic shock. JAMA 2005;294:448-54.
Acute Ischemic Heart Disease
Cardiogenic shock complicating acute coronary syndromes: Insights from the Global Registry of Acute Coronary Events Hamza H. Awad, MD, PhD, a Frederick A. Anderson, Jr, PhD, a Joel M. Gore, MD, a Shaun G. Goodman, MD, MSc, b and Robert J. Goldberg, PhD a Worcester, MA; and Ontario, Canada
Introduction Despite advances in the management of patients with an acute coronary syndrome (ACS), cardiogenic shock (CS) remains the leading cause of death in these patients. The objective of this observational study was to describe the characteristics, management, and hospital outcomes of patients with an ACS complicated by CS. Our secondary study objective was to describe trends in the incidence and hospital case-fatality rates (CFRs) of CS and predictors of increased hospital mortality in these high-risk patients. Methods The population consisted of patients enrolled in the GRACE study between 1999 and 2007 who were hospitalized with an ACS. Results During the years under study, 2,992 patients (4.6%) developed CS. Patients with CS were more likely to be older, have a history of diabetes or atrial fibrillation, and present with a higher pulse rate or cardiac arrest. Cardiac catheterization was performed on 1,706 (57%) and in-hospital revascularization on 1,408 patients (47%) with CS. Patients with CS were less likely to receive evidence-based cardiac medications compared with patients who did not develop CS. The in-hospital CFR of patients with CS was 59.4%, compared with 2.3% in those who did not develop CS. Factors associated with an increased risk of dying in patients with CS included advanced age, diabetes mellitus, angina, and stroke. Adjusted incidence rates and hospital CFRs of CS showed modest declines over time. Conclusion Continued efforts are needed to reduce the incidence and CFRs of CS complicating ACS. (Am Heart J 2012;163:963-71.) Cardiogenic shock (CS) remains the most serious clinical complication and the leading cause of death for patients hospitalized with an acute coronary syndrome (ACS). 1,2 Despite recent studies suggesting possible declines in the risk of dying during hospitalization for patients with CS, which has been linked to advances in medical treatment, coronary revascularization techniques, and mechanical support, in-hospital case-fatality rates (CFRs) associated with CS remain high, exceeding 50%. 3 Moreover, despite the clinical importance of this complication, the incidence rates of CS in patients with an ACS have remained relatively constant over the past 30 years, averaging approximately 7%. 3-5 Using data from
a large multinational coronary disease registry, we describe the demographic and clinical characteristics of patients presenting with an ACS complicated by CS, the management of these high-risk patients, and their shortterm outcomes in comparison with patients who did not develop CS. A secondary study objective was to describe nearly decade-long (1999-2007) trends in the incidence and CFRs of CS complicating ACS and factors associated with an increased risk of dying in patients with CS.
Methods The full details of the methods used in the GRACE study have been previously published. 6,7
Patient population a
b
From the University of Massachusetts Medical School, Worcester, MA, and Canadian Heart Research Centre, St. Michael's Hospital, University of Toronto, Toronto, Ontario, Canada. Submitted October 17, 2011; accepted March 12, 2012. Reprint requests: Robert J. Goldberg, PhD, Division of Epidemiology of Chronic Diseases and Vulnerable Populations, Department of Quantitative Health Sciences, University of Massachusetts Medical School, 55 Lake Avenue North, Worcester, MA 01655. E-mail: [email protected] 0002-8703/$ - see front matter © 2012, Mosby, Inc. All rights reserved. doi:10.1016/j.ahj.2012.03.003
This large multinational prospective study was designed to reflect an unselected population of patients with an ACS irrespective of geographic region. A total of 123 hospitals located in 14 countries in North and South America, Europe, Australia, and New Zealand have contributed data to this observational study. Adult patients (≥18 years old) admitted with a presumptive diagnosis of ACS to participating hospitals were potentially eligible. Patients with noncardiovascular causes for their clinical presentation were excluded.
American Heart Journal June 2012
964 Awad et al
Table I. Characteristics of patients with an ACS complicated by CS Patients with ACS complicated by CS
Characteristic Age (y) mean (SD) Age (y) mean (SD) b55 55-64 65-74 ≥75 Female, mean (SD) Medical history, mean (SD) Angina Atrial fibrillation Chronic heart failure Coronary artery disease Current smoking Diabetes mellitus Hypertension Myocardial infarction PCI CABG Renal insufficiency Stroke/TIA Clinical presentation, mean (SD) Pulse Systolic blood pressure Cardiac arrest STEMI Killip class on admission I II III IV
Total (n = 2992)
Presenting with CS on admission (n = 666)
Developing CS in-hospital after admission (n = 2326)
Patients with ACS not complicated by CS (n = 62 134)
n (%)
n (%)
n (%)
n (%)
71.01 (12.5)
69.8 (12.7)
71.4 (12.5)⁎
65.42 (13.2) †
368 (12.4) 529 (17.8) 817 (27.5) 1262 (42.4) 1143 (38.3)
93 (14.1) 140 (21.2) 177 (16.9) 249 (37.8) 249 (37.7)
275 (11.9) 388 (16.8) 640 (27.6) 1013 (43.7)⁎ 894 (38.6)
14 685 (23.8) 14 760 (23.8) 16 102 (26.0) 16 367 (26.5) † 20 100 (32.5) †
1346 (45.4) 263 (8.9) 516 (17.5) 665 (22.8) 1475 (50.1) 916 (31.0) 1768 (59.9) 837 (28.3) 309 (10.5) 272 (9.2) 325 (11.0) 324 (11.0)
232 (35.3) 56 (8.6) 115 (17.7) 138 (21.5) 324 (49.6) 166 (25.5) 364 (55.5) 193 (29.4) 73 (11.2) 57 (8.7) 49 (7.5) 48 (7.3)
1113 (48.2)⁎ 207 (9.0) 401 (17.4) 526 (23.1) 1150 (50.2) 750 (32.5)⁎ 1403 (61.1)⁎ 644 (27.8) 236 (10.2) 215 (9.3) 276 (12.0)⁎ 276 (12.0)⁎
32 608 (52.7) † 4690 (7.6) † 6090 (9.9) † 19 098 (31.3) † 35 762 (57.8) † 15 388 (24.9) † 38 386 (62.1) † 18 710 (30.3) † 11 045 (17.9) † 7746 (12.5) † 4600 (7.4) † 5070 (8.2) †
87.1 (28.3) 117.1 (37.9) 368 (12.5) 1909 (63.8)
86.0 (34.7) 94.9 (41.6) 193 (29.3) 414 (62.2)
87.4 (26.2) 123.1 (34.4)⁎ 175 (7.6)⁎ 1495 (64.3)
79.1 (20.8) † 142.5 (29.0) 933 (1.5) † 21 809 (35.1) †
1240 (42.5) 634 (21.7) 379 (13.0) 666 (22.8)
0 0 0 666 (100)
1240 (55.0) 634 (28.1) 379 (16.8) 0 (0)⁎
51 709 (85.1) 6982 (11.5) 2060 (3.4) None †
TIA, Transient ischemic attack. ⁎ P b .05 comparing patients presenting with CS with those developing it later during hospitalization. † P b .05 comparing patients with CS with patients without CS.
Prospective (warm pursuit) and retrospective (cold pursuit) surveillance approaches for identifying cases of ACS were used, as has been previously described. 6,7 Where required, study investigators received approval from their local hospital ethics or institutional review board. Standardized definitions of all patient-related variables, clinical diagnoses, and selected hospital complications and outcomes were based on the American College of Cardiology key data elements and definitions for measuring the clinical management and outcomes of patients hospitalized with ACS. 8 Patients were defined as having an ACS if they had symptoms typical of an ACS accompanied by at least 1 of the following: electrocardiographic changes consistent with ACS, serial increases in biochemical markers of cardiac necrosis (creatine kinase-MB fraction, creatine phosphokinase, or troponin), and documented coronary artery disease. Cardiogenic shock was defined as a systolic blood pressure of b80 mm Hg and congestive heart failure (Killip class IV) occurring at any time during the short-term hospitalization. 9 The specific timing of shock onset was not collected, but patients presenting
to the hospital with CS were differentiated from those developing shock in-hospital by Killip class on hospital presentation. Coronary revascularization was defined as the receipt of a percutaneous coronary intervention (PCI) or coronary artery bypass graft (CABG) at any time during the index hospitalization.
Data collection The hospital records of patients with a validated ACS were abstracted for demographic and clinical data, complications occurring during hospitalization, electrocardiographic findings, and use of diagnostic procedures and therapeutic approaches. Data were stored and analyzed at the Center for Outcomes Research at the University of Massachusetts Medical School.
Data analysis Continuous variables were summarized by their means and SDs or medians, as appropriate. Categorical variables were summarized by counts and percentages. Univariate comparisons of patient characteristics, clinical presentation, treatments
American Heart Journal Volume 163, Number 6
Awad et al 965
Table II. Treatment of patients with an ACS complicated by CS Patients with ACS complicated by CS
Treatment Procedures Cardiac catheterization PCI Patients undergoing PCI (% stent) CABG IABP Pulmonary artery catheter Medications ACE inhibitors Angiotensin receptor blockers Aspirin β-Blockers Calcium-channel blockers Fibrinolysis GP IIb/IIIa inhibitor Nitrates Pressors/inotropes
Total (n = 2993)
Presenting with CS on admission (n = 666)
Developing CS in-hospital after admission (n = 2326)
Patients with ACS not complicated by CS (n = 62 134)
n (%)
n (%)
n (%)
n (%)
1706 (57.3) 1190 (40.1) 977 (88.0) 272 (9.2) 852 (29.0) 945 (32.0)
375 (56.6) 294 (44.3) 250 (90.6) 28 (4.2) 155 (23.9) 181 (27.7)
1331 (57.5) 896 (38.9)⁎ 727 (87.2) 244 (10.6)⁎ 697 (30.5)⁎ 764 (33.3)⁎
37 823 (61.4) † 23 622 (38.3) 21 309 (93.2) † 3107 (5.1) † 892 (1.5) † 2285 (3.7) †
1595 (53.9) 97 (3.3) 2512 (84.4) 1656 (56.1) 368 (12.6) 627 (21.2) 894 (30.4) 1916 (64.7) 2368 (79.8)
309 (46.8) 16 (2.4) 519 (78.5) 294 (44.8) 52 (8.0) 127 (19.4) 178 (37.1) 295 (45.2) 489 (74.0)
1286 (56.0)⁎ 81 (3.6) 1993 (86.1)⁎ 1362 (59.4)⁎ 316 (15.0)⁎
40 497 (65.7) † 3532 (5.8) † 58 130 (93.7) † 52 625 (85.2) † 14 439 (23.6) † 8003 (13.0) † 15 798 (25.7) † 49 875 (80.8) † 7030 (11.5) †
499 (21.7) 716 (31.3)⁎ 1620 (70.3)⁎ 1878 (81.5)⁎
ACE, Angiotensin-converting enzyme; GP, glycoprotein. ⁎ P b .5 comparing patients presenting with CS with those developing it later during hospitalization. † P b .5 comparing patients with CS with patients without CS.
prescribed, and hospital outcomes between patients with and without CS were carried out using the Wilcoxon rank sum or χ 2 test, as appropriate. Short-term survival rates were estimated using the KaplanMeier method, and log-rank tests were used for between group (CS present vs absent) comparisons. The short-term prognosis in each study year was examined by calculating in-hospital and 30-day postadmission CFRs. A logistic multivariable regression analysis was used to examine changes over time in the incidence rates of CS and in-hospital CFRs while controlling for several potentially confounding demographic (eg, age or sex) and clinical (eg, prior comorbidities or ACS type) factors of prognostic importance. These variables were considered as potential confounders based on our univariate results and the findings from previous research. The HosmerLemeshow goodness-of-fit test was used to examine the adequacy of our logistic regression models. Unadjusted and multivariable-adjusted Cox proportional hazards models were used to determine factors associated with an increased risk of dying in the hospital among patients who developed CS. Given the nonrandomized nature of the present study and the caveats and difficulties involved in the interpretation of any multivariable-adjusted estimates of association, we did not control for the hospital use of cardiac treatments in our regression analyses in which hospital or 30-day postadmission survival status was the key outcome. All analyses were performed with STATA version 11.0 (Stata Corporation, College Station, TX). No extramural funding was used to support this work. The authors are solely responsible for the design and conduct of this study, all study analyses, and drafting and editing of the paper.
Results Baseline characteristics The study population consisted of 65,119 patients hospitalized with an ACS who were enrolled in GRACE between 1999 and 2007. Of these, 4.6% (n = 2,992) developed CS during their index hospitalization for an ACS. Of all patients with CS, 22.3% presented with CS on admission to the hospital, whereas the remainder of patients with CS developed it later during their index hospitalization. These proportions remained relatively unchanged during the years under study. Compared with patients without CS, patients who developed shock were more likely to be older (6 years on average), women, and to have a prior history of diabetes, atrial fibrillation, chronic heart failure, renal insufficiency, and stroke. Patients with CS were more likely to have a higher pulse rate, an ST-segment elevation myocardial infarction (STEMI), and to be in cardiac arrest compared with patients without CS at the time of hospital admission (Table I). Multivariableadjusted predictors of developing CS were older age (adjusted odds ratio [OR] per year 1.03, 95% CI 1.011.04), history of diabetes mellitus (OR 1.20, 95% CI 1.101.31) or atrial fibrillation (OR 1.34, 95% CI 1.16-1.54), and presenting with a higher pulse rate (OR 1.01, 95% CI 1.01-1.02) or cardiac arrest (OR 9.04, 95% CI 7.84-10.43). Multivariable predictors among patients with STEMI were similar to the predictors in the overall population, whereas additional predictors included a history of
American Heart Journal June 2012
966 Awad et al
Table III. Univariate and multivariate predictors of dying during hospitalization for patients with an ACS complicated by CS
Characteristic Age (y), mean (SD) Age (y) b55 55-64 65-74 ≥75 Female Medical history Angina Atrial fibrillation CAD PCI CABG Current smoking Diabetes mellitus Heart failure Hypertension Renal insufficiency Stroke/TIA Clinical presentation Presented with shock Cardiac arrest ST-segment elevation Non–ST-segment elevation
Died (n = 1775)
Survived (n = 1213)
n (%)
n (%)
P
73.9 (11.6)
66.8 (12.7)
b.01
139 (7.9) 239 (13.5) 460 (26) 929 (52.6) 733 (41.5)
229 (19) 289 (24) 355 (29.5) 331 (47.9) 409 (33.9)
b.01 b.01
872 (49.7) 188 (10.8) 419 (24.2) 177 (11.0) 170 (9.7) 763 (43.9) 591 (33.8) 356 (20.4) 1090 (62.5) 225 (12.8) 230 (13.2)
472 (39.0) 74 (6.2) 244 (20.6) 132 (10.2) 102 (8.5) 709 (58.9) 324 (26.9) 160 (13.3) 675 (56.0) 100 (8.3) 94 (7.8)
b.01 b.01 .02 .48 .24 b.01 b.01 b.01 b.01 b.01 b.01
322 (18.7) 228 (13.0) 1094 (61.7) 412 (23.2)
344 140 814 288
b.01 .28 b.01 .17
(29.0) (11.7) (67.1) (23.7)
Multivariable-adjusted hazards ratio (95% CI)
1 1.22 (09-1.6) 1.79 (1.4-2.4) 3.83 (3.0-5.0)
1.62 (1.3-1.8)
0.70 (0.6-0.8) 1.41 (1.2-1.6)
1.54 (1.1-1.9)
CAD, Coronary artery disease.
hypertension or myocardial infarction in patients with non–ST-segment elevation ACS.
cated by shock compared with patients without shock (Table II).
Treatment for patients with CS Although patients with ACS who developed CS were less likely to undergo cardiac catheterization (57.3% vs 61.4%), once they underwent angiography, they were more likely to undergo PCI (40.1% vs 38.3%) or CABG (9.2% vs 5.1%) compared with patients without shock. Stent use was significantly lower in patients with shock who underwent PCI compared with patients who did not develop CS (88.0% vs 93.2%) (Table II). Pulmonary artery catheterization and intraaortic balloon pumps [IABPs] as well as cardiac supportive drugs, including inotropes/pressors, were used significantly more often in patients with CS compared with patients without shock (Table II). Fibrinolytic therapy was administered more often to patients with shock compared with patients without CS (21.2% vs 13.0%). Patients with CS were less likely to have received angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, aspirin, β-blockers, calcium-channel blockers, low-molecular-weight heparin, and nitrates than patients without CS during their index hospitalization (Table II). Of note, only glycoprotein IIb/IIIa inhibitors and unfractionated heparin were used more frequently among ACS cases compli-
Short-term CFRs The crude in-hospital CFR for patients with CS was 59.4% compared with 2.3% in patients without shock (P b .001). Hospital mortality was markedly lower for patients who underwent coronary revascularization compared with those in whom a less aggressive approach was adopted (45.3% vs 72.0%, P b .001). Patients developing CS during their short-term hospitalization were at significantly higher risk for dying in-hospital compared with patients presenting to the hospital in CS (62.6% vs 48.4%, P b .001). In examining differences in possible prognostic factors in patients who died, as compared with those who survived CS, older patients and those with a history of diabetes, heart failure, hypertension, and renal insufficiency were more likely to die after developing CS than respective comparison groups (Table III). Multivariable-adjusted survival regression models showed that older age and a history of either angina, diabetes mellitus, or stroke were associated with an increased risk of dying among patients who developed CS. There were no differences in the predictors of death for patients with CS according to the presence or absence of ST-elevation in their baseline electrocardiogram.
American Heart Journal Volume 163, Number 6
Awad et al 967
Figure 1
Trends in hospital incidence rates of cardiogenic shock among patients with acute coronary syndromes.
Models that further included time of development of shock showed that patients with CS who survived their short-term hospitalization were significantly younger and were more likely to have presented to the hospital in CS. The overall crude 30-day CFR for patients with ACS complicated by CS was 59% compared with 2.3% for patients who did not develop CS during their index hospitalization (P b .001) (Table III). Significant univariate as well as multivariable-adjusted predictors of an increased risk of dying during the 30 days after development of ACS for patients with CS were similar to the predictors of in-hospital mortality.
In examining the association between in-hospital death rates and the shock index (SI) (SI = heart rate/systolic blood pressure), patients with SI b0.8 experienced an inhospital CFR of 54.9%, whereas patients with an SI ≥0.8 had a CFR of 64.5%. In addition, in categorizing patients according to SI quartiles, the corresponding in-hospital CFRs were 52.9%, 53.8%, 60.4%, and 67.4%, respectively.
Time trends in the incidence rates of CS The crude hospital incidence rates of CS decreased from 5.1% in 1999 to 3.6% in 2007. Trends in incidence
968 Awad et al
American Heart Journal June 2012
Figure 2
Trends in hospital case-fatality rates for acute coronary syndrome patients with and without cardiogenic shock.
rates by type of ACS and by time of development of shock are shown in Figure 1. The multivariable-adjusted risk of developing CS, adjusted for age, sex, medical history, clinical presentation, and length of hospital stay, showed significant declines over the years under study (annual OR 0.97, 95% CI 0.95-0.98).
Time trends in hospital CFRs The crude in-hospital CFRs among patients with CS declined from 62% in 1999 to 54% in 2005, with an increase in these death rates in 2007 (Figure 2). Trends in hospital CFRs by type of ACS and by time of development of shock are shown in Figure 3. In-hospital CFRs of patients with CS adjusted for age, sex, medical history, clinical presentation, length of hospital stay, and study site showed significant declines during the years under study (annual OR 0.94, 95% CI 0.90-0.99).
Discussion The results of our large multinational observational study provide insights into the magnitude, management, short-term mortality rates, and the characteristics of patients with an ACS likely to develop and survive CS. Encouragingly, our study showed declining short-term mortality and incidence rates related to CS during the years under study.
Baseline characteristics Identifying patients at increased risk for developing CS and providing these high-risk patients with urgent medical care is essential to decreasing the risk of developing this serious complication among patients
hospitalized with an ACS. Proper monitoring, risk stratification, and aggressive intervention have been associated with improved survival among patients developing CS. 10-13 Our study demonstrated that patients with ACS at high risk for developing CS were older, more likely to be female, diabetic, having chronic heart failure, and more likely to present to the hospital with STEMI and cardiac arrest. Similar to previously reported findings from studies in patients with an ACS, less than onequarter of all ACS patients with CS presented to the hospital with CS on admission. 14 Our results are similar to previously published findings carried out in different population settings including the observational Worcester Heart Attack Study and the GUSTO and SHOCK randomized trials. 10,15,16 These findings create a profile of individuals with an ACS who are at particularly increased risk for developing CS. Several of these high-risk individuals can be identified in advance and should be placed under increased hospital surveillance as well as considered for the receipt of effective aggressive treatment modalities, including coronary revascularization.
Treatment for patients with CS Previously published results from clinical trials and observational studies have demonstrated that early mechanical revascularization in patients with CS has been associated with lower mortality compared with initial medical stabilization (including IABP counterpulsation and fibrinolytic therapy) followed by late or no revascularization. 11,17 Based on these findings, the American College of Cardiology and the American Heart Association have elevated early mechanical revascularization for CS to a class I
American Heart Journal Volume 163, Number 6
Awad et al 969
Figure 3
Trends in hospital case-fatality rates of cardiogenic shock among patients with acute coronary syndromes.
recommendation for patients b75 years old 18 and a class IIA recommendation for those N75 years old who are considered to be suitable candidates. 19 However, the availability of clinical practice guidelines does not necessarily translate to changes in “real-world” practice. 20 The results of our study showed that N40% of patients with CS were N75 years old, a proportion that is consistent with the previously published literature. 4,10 Treatment for this high-risk group and appropriately selecting patients eligible for invasive management remain an important clinical challenge. Although patients with CS were less likely to undergo cardiac catheterization than patients who did not develop CS, once they were catheterized, they were more likely to
undergo PCI or CABG. This can likely be explained by the presence of more extensive myocardial damage, a greater frequency of multivessel underlying disease, and because of their clinically unstable status. Patients with shock were also more likely to receive fibrinolysis compared with patients who did not develop CS. Although the benefits of fibrinolysis in patients with CS are less established compared with patients with acute myocardial infarction, this reperfusion approach has been convincingly shown to reduce the risk of shock, 21 which is of considerable importance because most patients develop CS after hospital presentation (N6 hours). 9,10 The DANAMI-2 trial, which randomly assigned patients with
970 Awad et al
STEMI to either PCI or fibrinolysis, found no difference in the proportion of patients developing CS between the 2 comparison groups. 22 Patients with CS were also less likely to receive evidence-based medications compared with patients without CS. Although these medications are associated with better outcomes in patients with ACS, several of these therapies might aggravate hypotension in patients with CS. Inasmuch, these medications are often withheld until the patient is stabilized. 23 Because of the lack of information on the exact timing of development of CS in our study, we were unable to determine the relationship between the time of onset of CS and the prescribing of in-hospital medications or procedures. As expected, patients developing CS were more likely to receive supportive cardiac care. Future studies should continue to monitor the use of various treatment approaches in this high-risk patient population and identify patient groups less likely to be treated with evidence-based treatments.
In-hospital and 30-day CFRs Cardiogenic shock remains the most frequent cause of in-hospital death among patients with an ACS. Historically, CFRs associated with the development of CS in previous studies have ranged from 50% to 80%. However, and despite the persistently high death rates associated with CS, the prognosis for patients hospitalized with CS has improved considerably over the past several decades. 4,11,17,24 Declines in hospital mortality rates have been associated with advances in supportive care and a more aggressive approach to coronary revascularization. 11,17 As expected, and consistent with the previous literature, patients with CS were significantly more likely to die in-hospital than patients who did not develop this complex hemodynamic complication. 14 Several randomized trials 11,17 and reports from observational studies suggest that PCI often results in improved short-term survival in patients with CS with survival contingent on the successful establishment of coronary reperfusion. 9 Uncontrolled observational studies have suggested an association between the receipt of CABG surgery and an improvement in short-term survival among patients with CS when they are treated soon after shock has developed. 15 However, we were unable to assess the role of these interventional procedures because we could not determine whether shock preceded or followed the use of these treatment strategies or determine the reasons why certain patients received these therapeutic regimens and others did not (confounding by indication). Consistent with the findings from previously published studies, older age, diabetes mellitus, and a history of stroke were associated with an increased risk of inhospital mortality among patients developing CS. How-
American Heart Journal June 2012
ever, developing CS during hospitalization was associated with the highest risk of dying during hospitalization among the possible predictors of in-hospital mortality among patients with CS. A recently published study from the APEX-AMI trial on patients with STEMI developing CS reported similar predictors of death among this highrisk population. 14 Future studies should explore factors associated with patient's changing hemodynamic status during hospitalization for an ACS and the risk of developing CS in this high-risk population.
Trends in the incidence and in-hospital CFRs of CS It is difficult to compare the actual incidence rates of CS in published studies because of varying population characteristics, definitions of CS, and whether patients developing shock in the prehospital setting were included. 4,25 The incidence of CS reported from previous studies has ranged from 3% to 15% 26,27 depending on the study population characteristics. The overall incidence rate of CS reported in our study (4.6%) falls within this range. Moreover, we demonstrated that the incidence rates of CS declined over the period under study, consistent with findings from previous studies examining trends in CS incidence rates during the 2000s. 15,28 Potential contributors to encouraging declines in the incidence rates of CS include the increased adoption of early revascularization of patients with ACS, increased use of evidence-based cardiac medications over time, and enhanced patient monitoring. The results of our study show a declining trend in the short-term mortality rates associated with CS between 1999 and 2006 with an increase in these death rates in 2007. Although the reasons for this increase in crude short-term CFRs during our most recent study year remain unexplained, declines in the multivariable-adjusted hospital death rates of patients with CS occurred during the years under study when other covariates of prognostic importance were adjusted for. It is likely that the increased and early use of aggressive treatment regimens and supportive cardiac care contributed to these observed declines in adjusted hospital CFRs. Study strengths and limitations Our study strengths include the use of standardized data collection efforts in a large multinational registry of patients hospitalized with an ACS and inclusion of patients hospitalized during relatively recent periods. The limitations of our study include its observational, nonrandomized nature and lack of information on the exact timing of development of CS. Furthermore, we did not collect information on patient's race/ethnicity and socioeconomic status and were unable to systematically characterize the severity of the ACS.
American Heart Journal Volume 163, Number 6
Conclusions The results of this large observational study provide insights into the characteristics, management practices, and short-term mortality of patients with ACS complicated by CS. Although the magnitude of CS appears to be slowly declining, it still develops at a relatively high rate after ACS. The hospital death rate among patients with this complication remains high despite advances in management practices and declining rates over time. It remains of considerable importance to examine contemporary trends in the magnitude and short-term outcomes associated with CS with serial reassessment of baseline demographic, clinical, and laboratory risk factors. Furthermore, it remains of necessity to monitor the ever-changing hemodynamic status of patients hospitalized with an ACS and to develop more refined risk stratification systems for identifying patients most likely to develop this devastating complication as well as those at high risk for dying from it.
References 1. Domanski MJ, Topol EJ. Cardiogenic shock: current understandings and future research directions. Am J Cardiol 1994;74:724-6. 2. Califf RM, Bengtson JR. Cardiogenic shock. N Engl J Med 1994;330: 1724-30. 3. Hollenberg SM, Kavinsky CJ, Parrillo JE. Cardiogenic shock. Ann Intern Med 1999;131:47-59. 4. Goldberg RJ, Samad NA, Yarzebski J, et al. Temporal trends in cardiogenic shock complicating acute myocardial infarction. N Engl J Med 1999;340:1162-8. 5. Rogers WJ, Canto JG, Lambrew CT, et al. Temporal trends in the treatment of over 1.5 million patients with myocardial infarction in the US from 1990 through 1999: the National Registry of Myocardial Infarction 1, 2 and 3. J Am Coll Cardiol 2000;36:2056-63. 6. Rationale and design of the GRACE (Global Registry of Acute Coronary Events) Project: a multinational registry of patients hospitalized with acute coronary syndromes. Am Heart J 2001;141:190-9. 7. Steg PG, Goldberg RJ, Gore JM, et al. Baseline characteristics, management practices, and in-hospital outcomes of patients hospitalized with acute coronary syndromes in the Global Registry of Acute Coronary Events (GRACE). Am J Cardiol 2002;90:358-63. 8. Cannon CP, Battler A, Brindis RG, et al. American College of Cardiology key data elements and definitions for measuring the clinical management and outcomes of patients with acute coronary syndromes. A report of the American College of Cardiology Task Force on Clinical Data Standards (Acute Coronary Syndromes Writing Committee). J Am Coll Cardiol 2001;38:2114-30. 9. Dauerman HL, Goldberg RJ, White K, et al. Revascularization, stenting, and outcomes of patients with acute myocardial infarction complicated by cardiogenic shock. Am J Cardiol 2002;90:838-42. 10. Hochman JS, Boland J, Sleeper LA, et al. Current spectrum of cardiogenic shock and effect of early revascularization on mortality. Results of an International Registry. SHOCK Registry Investigators. Circulation 1995;91:873-81. 11. Hochman JS, Sleeper LA, Webb JG, et al. Early revascularization in acute myocardial infarction complicated by cardiogenic shock. SHOCK Investigators. Should we emergently revascularize occluded coronaries for cardiogenic shock. N Engl J Med 1999;341:625-34.
Awad et al 971
12. Berger PB, Holmes Jr DR, Stebbins AL, et al. Impact of an aggressive invasive catheterization and revascularization strategy on mortality in patients with cardiogenic shock in the Global Utilization of Streptokinase and Tissue Plasminogen Activator for Occluded Coronary Arteries (GUSTO-I) trial. An observational study. Circulation 1997;96:122-7. 13. Holmes Jr DR, Califf RM, Van de Werf F, et al. Difference in countries' use of resources and clinical outcome for patients with cardiogenic shock after myocardial infarction: results from the GUSTO trial. Lancet 1997;349:75-8. 14. French JK, Armstrong PW, Cohen E, et al. Cardiogenic shock and heart failure post-percutaneous coronary intervention in ST-elevation myocardial infarction: observations from “Assessment of Pexelizumab in Acute Myocardial Infarction”. Am Heart J;162:89-97. 15. Goldberg RJ, Spencer FA, Gore JM, et al. Thirty-year trends (1975 to 2005) in the magnitude of, management of, and hospital death rates associated with cardiogenic shock in patients with acute myocardial infarction: a population-based perspective. Circulation 2009;119: 1211-9. 16. Hasdai D, Califf RM, Thompson TD, et al. Predictors of cardiogenic shock after thrombolytic therapy for acute myocardial infarction. J Am Coll Cardiol 2000;35:136-43. 17. Hochman JS, Sleeper LA, White HD, et al. One-year survival following early revascularization for cardiogenic shock. JAMA 2001; 285:190-2. 18. Ryan TJ. Early revascularization in cardiogenic shock—a positive view of a negative trial. N Engl J Med 1999;341:687-8. 19. Antman EM, Anbe DT, Armstrong PW, et al. ACC/AHA guidelines for the management of patients with ST-elevation myocardial infarction: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to Revise the 1999 Guidelines for the Management of Patients with Acute Myocardial Infarction). Circulation 2004;110:e82-292. 20. Curry SJ. Organizational interventions to encourage guideline implementation. Chest 2000;118(2 Suppl):40S-6S. 21. Effect of intravenous APSAC on mortality after acute myocardial infarction: preliminary report of a placebo-controlled clinical trial. AIMS Trial Study Group. Lancet 1988;1:545-9. 22. Lindholm MG, Boesgaard S, Thune JJ, et al. Percutaneous coronary intervention for acute MI does not prevent in-hospital development of cardiogenic shock compared to fibrinolysis. Eur J Heart Fail 2008;10: 668-74. 23. Gunnar RM, Bourdillon PD, Dixon DW, et al. ACC/AHA guidelines for the early management of patients with acute myocardial infarction. A report of the American College of Cardiology/American Heart Association Task Force on Assessment of Diagnostic and Therapeutic Cardiovascular Procedures (subcommittee to develop guidelines for the early management of patients with acute myocardial infarction). Circulation 1990;82:664-707. 24. Bengtson JR, Kaplan AJ, Pieper KS, et al. Prognosis in cardiogenic shock after acute myocardial infarction in the interventional era. J Am Coll Cardiol 1992;20:1482-9. 25. An international randomized trial comparing four thrombolytic strategies for acute myocardial infarction. The GUSTO investigators. N Engl J Med 1993;329:673-82. 26. Sanborn TA, Feldman T. Management strategies for cardiogenic shock. Curr Opin Cardiol 2004;19:608-12. 27. Davies CH. Revascularization for cardiogenic shock. QJM 2001;94: 57-67. 28. Babaev A, Frederick PD, Pasta DJ, et al. Trends in management and outcomes of patients with acute myocardial infarction complicated by cardiogenic shock. JAMA 2005;294:448-54.