- Email: [email protected]
Clinical experience with intra-aortic balloon counterpulsation over 10 years: A retrospective cohort study of 459 patients
Resuscitation (2008) 77, 316—324
available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/resuscitation
CLINICAL PAPER
Clinical experience with intra-aortic balloon counterpulsation over 10 years: A retrospective cohort study of 459 patients Sheng-Nan Chang a,1, Juey-Jen Hwang a,b,1, Yih-Sharng Chen a,c, Jou-Wei Lin a,∗, Fu-Tien Chiang b a
Cardiovascular Center, National Taiwan University Hospital Yun-Lin Branch, Dou-Liou City, Yun-Lin, Taiwan Section of Cardiology, Department of Medicine, National Taiwan University Hospital, Taipei, Taiwan c Division of Cardiovascular Surgery, Department of Surgery, National Taiwan University Hospital and National Taiwan University College of Medicine, Taipei, Taiwan b
Received 23 October 2007; received in revised form 12 December 2007; accepted 6 January 2008
KEYWORDS Shock, Cardiogenic; Intra-aortic balloon pumping; Extracorporeal membrane oxygenation; Bypass; Coronary artery; Angioplasty, Transluminal, Percutaneous coronary
Summary Objective: The objective of this study was to identify prognostic predictors for the patients experiencing cardiogenic shock who required the institution of intra-aortic balloon counterpulsation (IABP). Design, setting, and patients: Patients with cardiogenic shock were retrieved from the clinical information system in National Taiwan University Hospital and classified according to their etiology: acute coronary syndrome (ACS), ST segment elevation myocardial infarction (STEMI), congestive heart failure (CHF), hemodynamic instability after post-coronary bypass graft operation (post-CABG) or after percutaneous intervention (post-PCI), and out-of-hospital cardiac arrest (OHCA) victims. Measurements: Kaplan—Meier curves and Cox regression model were applied to evaluate the factors associated with survival. Main results: A total of 459 patients were found to belong to one of six etiology categories between 1995 and 2004. The 30-day mortality was highest in the OHCA group, followed by the STEMI, CHF, ACS, post-PCI, and post-CABG groups in a decreasing frequency (log rank p < 0.001). Peak troponin I level was negatively associated with survival, and its effect largely paralleled with underlying etiology. Age and renal impairment were significant prognostic predictors for 30-day mortality (hazard ratio = 1.031, p < 0.001 and hazard ratio = 1.266, p < 0.001). Comparing
∗ Corresponding author at: Cardiovascular Center, National Taiwan University Hospital Yun-Lin Branch, 579 Yun-Lin Road, Section 2, Dou-Liou City, Yun-Lin 640, Taiwan. Tel.: +886 922 861953; fax: +886 5 5335373. E-mail addresses: [email protected], [email protected] (J.-W. Lin). 1 These authors contributed equally.
0300-9572/$ — see front matter © 2008 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.resuscitation.2008.01.016
IABP, cardiogenic shock, and prognostic predictors
317
to those manifested as OHCA who had the worst outcome, patients in the other etiology groups had significantly better survival. Conclusions: This study has illustrated that age, renal function, and etiology-related cardiac injury are predictors for in-hospital course and mortality in those who experienced cardiogenic shock with IABP. The optimal strategy for revascularization in this high-risk group needs further validation. © 2008 Elsevier Ireland Ltd. All rights reserved.
Introduction Cardiogenic shock is a state of inadequate tissue perfusion due to cardiac dysfunction, which complicates with several cardiovascular diseases such as ST-elevation myocardial infarction (STEMI), non-ST elevation myocardial infarction (NSTEMI), unstable angina (UA), congestive heart failure (CHF), complication of percutaneous coronary intervention (PCI), and post-coronary artery bypass graft (CABG). Mortality and morbidity rates of cardiogenic shock remain frustratingly high in the range between 50% and 80%.1 With the improvement in medical techniques and technologies, intra-aortic balloon counterpulsation (IABP) has become a unique tool to improve and support hemodynamic derangement associated with underlying cardiovascular disease.2 The goal of IABP, which inflates and displaces blood from the descending aorta during the diastolic phase, aims at decreasing cardiac workload, myocardial oxygen consumption, and after-load, and at improving coronary perfusion during diastole. Prophylactic IABP could provide circulatory support in case of pharmacologically uncontrollable chest pain, multivessel angioplasty, and severe left ventricular systolic dysfunction.3—6 Several common and easily available variables such as inotrope usage, cardiogenic shock, priority, left main stem disease, ejection fraction, re-do operation, and recent catheterization were used to assess the probability of needing an IABP prior to cardiac surgery.7 In 2004, AHA/ACC recommended IABP as Class I indication for cardiogenic shock in the management of acute myocardial infarction (AMI).8 AHA/ACC also gives Class IIa recommendation for IABP therapy in patients with severe malignant arrhythmias caused by underlying cardiac ischemia and refractory to medical therapy.2 Providing favorable hemodynamic effects, IABP is suggested as a bridging therapy to emergency revascularization, heart transplantation, and acute valvular repair.9,10 Therefore, it is of clinical interests to find prognostic predictors for survival in the patients with cardiogenic shock under IABP usage. This study was designed to construct a retrospective cohort in which the patients required hemodynamic support with IABP due to cardiac etiology. The objective was to find the factors associated with survival in the high-risk group and to illustrate the roles of IABP in various etiology groups.
Materials and methods Subjects and materials National Taiwan University Hospital (NTUH), founded in 1895, is one of the largest in southeastern Asia.11 A health
information system has been maintained by the Information Systems Office since 1982. For each admission, an electronic discharge note must be entered by attending physicians and residents.12 All the discharge notes from 1995 to 2004 were downloaded to MS ACCESS databases, and the narrative contents of more than 600,000 records were stored in separated fields.13 Keyword search was applied to find the context ‘‘intra-aortic balloon counterpulsation’’ or ‘‘IABP’’ in fields such as ‘‘Present Illness’’ and ‘‘In-Hospital Course.’’ Extracted medical records were then reviewed by a cardiologist to determine the eligibility of each patient. Inclusion criteria were subjects with cardiogenic shock who received IABP institution anytime during the admission. It required the use of inotropic agents to the maximal dose (e.g., dopamine 20 g/(kg min)) with blood pressure lower than 90/60 mmHg before IABP insertion or concurrent use of high-dose dopamine when a complication took place. Sepsis or anaphylaxis was excluded retrospectively at the cardiologist’s discretion. Underlying etiology could be further classified by the cardiologist into the following categories: (1) acute coronary syndrome (ACS) (UA or NSTEMI), (2) STEMI, (3) CHF with hemodynamic changes, (4) patients receiving IABP during CABG and maintaining it post-operatively (post-CABG), (5) patients experiencing complications during elective PCI (post-PCI), and (6) out-ofhospital cardiac arrest (OHCA). Etiology other than the above six categories was excluded from analysis because of infrequent occurrence. Examples were unexplained ventricular tachycardia/fibrillation, acute respiratory distress syndrome, myocarditis, and hemodynamic instability after valvular heart operation.
Technical issues IABP and extra-corporeal membrane oxygenation (ECMO) were applied according to the following principles. IABP is a catheter-based device with a 30—50 ml balloon sac to be chosen according to body height. If the patient was of body height over 182 cm, IABP catheter (Arrow International Inc.) with 50 ml balloon sac was chosen. If the patient was of body height below 162 cm, catheters (Arrow International Inc.) with 30 ml balloon sac were chosen. For the patients with body height between 162 and 182 cm, catheters (Arrow International Inc.) with 40 ml balloon sac were used. IABP was inserted into the femoral artery and further advanced to near the level of the left subclavian artery, with the balloon inflated with helium during diastole. The correct position of the catheter tip was evaluated by fluoroscopic guidance or adjusted after chest radiograph. Extra-corporeal membrane oxygenation was applied to the patients who manifested with catastrophic hemodynamics or to those who did not sustain a stable condition
318
S.-N. Chang et al.
after IABP. ECMO team responded to cardiologists’ call. As described previously,14 femoral cannulation was the first choice and distal limb perfusion was maintained. All circuits were primed with normal saline and heparin (2 U/ml), and blood was rewarmed to 37 ◦ C. Pump flow at 50—100 ml/(kg min) and high-dose catecholamine was applied to keep blood pressure above 60 mmHg.15 In the patients manifested with STEMI, it was attempted to perform primary PCI within the first 12 h after the event. The decision to undergo another PCI or CABG depended on patients’ clinical condition at attending cardiologists’ discretion. In the patients with UA/NSTEMI, the decision to receive early or conservative treatment and subsequent CABG therapy was made according to evolving myocardial ischemia refractory to optimal medical therapy, ongoing ischemia despite successful or failed PCI, complicated PCI, presence of left main stenosis and/or 3-vessel disease, and complex coronary anatomy.
Measurements of predictors and outcomes Outcome measurement, i.e., survival or mortality at Days 7 and 30, was retrieved by the same cardiologist from electronic medical records. Survival time was started on the day the major cardiovascular event occurred, and terminated when a patient died. Patients surviving to discharge were regarded as censored at the discharge date.
Table 1
Demographic and hemodynamic data, including age, underling disease (hypertension and diabetes), vital signs, and treatment courses, were derived from electronic medical records and coded into an EXCEL spread sheet. Laboratory data were retrieved from the laboratory information system. Hemoglobin level, serum creatinine, peak cardiac enzyme (creatine kinase, MB form, and troponin I), and arterial blood gas were collected for further analysis.
Statistics Statistics was performed with SPSS 12.0 (SPSS Inc., Chicago, IL). Subjects were classified into six major categories: (1) ACS, (2) STEMI, (3) CHF, (4) post-CABG, (5) post-PCI, and (6) OHCA. Case number and age distribution were summarized by group. The proportion of subjects receiving ECMO or CABG and, consequently, 7- and 30-day mortality were reported accordingly. A Cox regression model was applied to find the hazard associated with 30-day survival after the initial event. Predictors, including age, sex, the presence of diabetes, hypertension, and smoking, peak cardiac enzyme level, serum creatinine level, hemoglobin, and etiology were added into the model in a forward stepwise approach. Hazard ratio (HR), 95% confidence interval, and p-value were reported. Cardiac enzyme was represented as serum troponin I (TnI) level greater or less than 50 mg/dl. Two dummy
Demographic data of the 459 patients in the retrospective cohort study
Age
67 ± 12 (n = 459)
Sex Men Women
110 (24.0%) 349 (76.0%)
Etiology Acute coronary syndrome ST segment elevation myocardial infarction Congestive heart failure Post-coronary bypass surgery Percutaneous coronary intervention complication Out-of-hospital cardiac arrest
174 (37.9%) 143 (31.2%) 21 (4.6%) 56 (12.2) 34 (7.4%) 31 (6.8%)
Underlying diseases Hypertension Diabetes mellitus Peak troponin I level >50 mg/dl <50 mg/dl unknown Hemodynamics after inotropic agents and IABP use Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg) Heart rate (min−1 ) Laboratory data Hemoglobin (g/dl) Hematocrit (%) Creatinine (mg/dl) pH Bicarbonate (mmol/l)
198 (43.1%) 156 (34.0%) 148 (32.2%) 133 (29.0%) 178 (38.8%) 120 ± 29 (n = 303) 67 ± 18 (n = 302) 87 ± 24 (n = 304) 12.1 ± 2.0 (n = 438) 35.7 ± 6.1 (n = 439) 1.96 ± 5.6 (n = 424) 7.35 ± 0.136 (n = 121) 19.3 ± 5.4 (n = 121)
ACS: acute coronary syndrome; STEMI: ST segment elevation myocardial infarction; CHF: congestive heart failure; CABG: coronary artery bypass graft; PCI: percutaneous coronary intervention; OHCA: out-of-hospital cardiac arrest; ECMO: extra-corporeal membranous oxygenation; SBP: SBP, DBP and HR: systolic blood pressure, diastolic blood pressure and heart rate after management with inotropic agents and IABP insertion.
27.0 35.0 33.3 3.6 5.9 74.2 28.5 16.7 25.2 9.5 1.8 2.9 54.8 18.7 ± ± ± ± ± ± ± 85 89 91 79 84 99 87 16 18 18 15 12 22 18 ± ± ± ± ± ± ± 69 63 63 75 74 53 67 174 143 21 56 34 31 459 ACS STEMI CHF Post-CABG Post-PCI OHCA Total
68 67 61 66 65 63 67
± ± ± ± ± ± ±
11 12 16 9 10 13 12
73:27 79:21 71:29 79:21 79:21 77:23 76:24
57.5 36.4 33.3 100.0 58.8 12.9 52.1
14.4 9.8 23.8 5.4 11.8 29.0 13.1
127 113 107 132 130 92 120
± ± ± ± ± ± ±
29 27 22 18 29 37 29
DBP SBP ECMO (%) CABG (%) Sex (M:F) (%) Age Case number Etiology
Table 2
Etiology of the patients experiencing cardiogenic shock requiring intra-aortic balloon counterpulsation
HR
20 27 25 15 26 29 24
7-Day mortality (%)
30-Day mortality (%)
IABP, cardiogenic shock, and prognostic predictors
319 variables were used to represent troponin I > 50, troponin I < 50, and troponin unknown. The survival among different etiologies was compared to that of the OHCA group. HR less than 1 indicated a proportional survival benefit over OHCA. A factor that was statistically significant (p < 0.05) remained in the survival regression model, and a final parsimonious multiple Cox regression model with statistically predictors was reported. Kaplan—Meier curves were plotted according to different etiology and peak cardiac enzyme level. Furthermore stratified by groups, survival curves were plotted according to whether a patient received ECMO during the admission. In addition, patients in the ACS and the STEMI groups were further stratified by their treatment strategy: (0) neither PCI nor CABG, (1) PCI but no CABG, (2) CABG but no PCI, and (3) both PCI and CABG. Log rank test was used to find survival differences.
Results In the period between 1995 and 2004, a total of 505 patients who experienced cardiogenic shock underwent IABP institution. Forty-six cases were excluded from further analysis due to infrequent etiology. Among the 459 indicated patients, there were three times more men than women. The average age was 66 ± 12 for men and 70 ± 11 for women, respectively (Table 1). Hemodynamic data prior to IABP insertion was largely unavailable from electronic chart review. Heart rate and blood pressure after management with inotropic agents and IABP insertion were reported (Tables 1 and 2). In the subsequent management of these patients, 13.1% received hemodynamic support by extra-corporeal membranous oxygenation and 52.1% of the patients ultimately underwent CABG (Table 2). Major complications directly related to IABP included only one severe low leg ischemia necessitating amputation and one hemorrhagic shock due to iatrogenic aortic trauma. The overall 7- and 30-day mortality was 18.7% and 28.5% in the 459 patients. The mortality rate varied significantly among various etiologies. The 7-day mortality was the highest in the OHCA group, following by the STEMI, ACS, CHF, post-PCI, and post-CABG groups in the decreasing frequency (log rank test p < 0.001). The 30-day mortality was also highest in the OHCA group, but the sequence was slightly different (STEMI > CHF > ACS > post-PCI > post-CABG) (log rank test p < 0.001, Figure 1). In addition to underlying etiology, the extent of cardiac damage was also shown to be associated with the outcome. Patients with the TnI level greater than 50 mg/dl had a higher 30-day mortality rate than those with TnI less than 50 mg/dl (log rank test p = 0.003, Figure 2). It was also shown that the extent of cardiac elevation paralleled with 30-day mortality in each etiology category, suggesting that the underlying pathophysiology of the heart determined the extent of cardiac damage and therefore the outcome (Figure 3). A Cox regression model identified that age, renal function, and underlying etiology were significant prognostic predictors for 30-day mortality in the patients with cardiogenic shock. Each year increase in age would lead to 1.031-fold of hazard (95% CI: 1.014—1.049, p < 0.001).
320
S.-N. Chang et al.
Figure 1 30-Day survival by etiology. ACS: acute coronary syndrome; STEMI: ST segment elevation myocardial infarction; CHF: congestive heart failure; post-CABG: post-coronary artery bypass graft; post-PCI: post-elective percutaneous coronary intervention; OHCA: out-of-hospital cardiac arrest.
Elevated serum creatinine level was also associated with a worse outcome (HR: 1.266 for each mg/dl, 95% CI: 1.142—1.403, p < 0.001). Comparing to those manifested as OHCA who had the worst outcome, patients in the other etiology groups had significantly better survival (HR: 0.259 in the ACS group, 0.352 in the STEMI group, 0.302 in the CHF group, 0.027 in the post-CABG group, and 0.045 in the post-PCI group) (Table 3). Serum TnI was no longer associated with 30-day survival when underlying etiology was adjusted. Other factors, such as sex, hemoglobin level, and the presence of diabetes and hypertension, were not significant predictors in the survival analysis. ECMO was usually applied to the patients with catastrophic hemodynamic events; therefore, its use also indicated a poor outcome. Classified by etiology, patients with ECMO had a worse 30-day outcome in the ACS group and the CHF group (log rank p = 0.0002 and 0.0241), but not in other groups (Figure 4). The difference in mortality might be not related to the procedure but to the fact that they were sicker. In those who presented with ACS, the patients were further stratified into the following four subgroups: (0) neither PCI nor CABG, (1) PCI but no CABG, (2) CABG but no PCI, and
Figure 3 Percentage of high troponin I (TnI > 50 mg/dl) and 30-day mortality by etiology.
Table 3 Cox regression model to identify parameters associated with 30-day mortality: hazard ratio, p-value, and 95% confidence interval (CI) Variables
Figure 2 30-Day survival by troponin I (TnI greater or less than 50 mg/dl).
Age Creatinine Etiology ACS STEMI CHF Post-CABG Post-PCI OHCA
Hazard ratio
p-Value
95% CI Lower
Upper
1.031 1.266
<0.001 <0.001
1.014 1.142
1.049 1.403
0.259 0.352 0.302 0.027 0.045 1
<0.001 <0.001 0.007 <0.001 0.003
0.147 0.198 0.126 0.006 0.006
0.455 0.624 0.725 0.117 0.339
IABP, cardiogenic shock, and prognostic predictors
321
Figure 4 ECMO and survival in the six etiology categories (solid line: without ECMO; dashed line: with ECMO), upper left: ACS group, upper middle: STEMI group, upper right: CHF group, lower left: post-CABG group, lower middle: post-PCI group and lower right: OHCA group.
Figure 5 Treatment strategy and survival in the ACS group (left) and in the STEMI group (right). PCI: percutaneous coronary intervention; CABG: coronary artery bypass graft; neither: neither PCI nor CABG; both: both PCI and CABG.
(3) both PCI and CABG. Survival was highest in subgroup 2, followed by subgroups 3, 1, and 0. However, there was only a trend but no statistical significance in the survival between subgroups 2 and 3 (log rank p = 0.328) and between subgroups 3 and 1 (p = 0.322). The direct comparison between the CABG-only patients (subgroup 2) and PCI-only ones (subgroup 1) favored CABG in a borderline statistical significance (p = 0.085). Patients receiving neither PCI nor CABG must have been derived from the most catastrophic subgroup and showed the worse prognosis (log rank p < 0.001) (Figure 5, Panel A). In those who experienced STEMI who had cardiogenic shock, patients receiving both PCI and then CABG had a trend of better survival than those with PCI only (log rank p = 0.170) while those who received neither had the worst outcome (p < 0.001). No subject received only CABG without primary PCI (Figure 5, Panel B).
Discussion Pathophysiology of cardiogenic shock has been described as a cascade of events initiated by myocardial dysfunction, which leads to both increasing left ventricular filling pressure and decreasing stroke volume.16 As the Should We Emergently Revascularize Occluded Coronaries for Cardiogenic Shock (SHOCK) trial showed, initial stabilization of cardiogenic shock patients due to acute myocardial infarction by using IABP (86% of the trial cases) might be associated with a 20% absolute risk reduction in mortality.17 Early revascularization by either primary PCI or CABG resulted in a non-significant trend in reducing the primary endpoint of 30-day mortality (46.7% vs. 56.0%, p = 0.11). Six months after intervention, the mortality was lower in the early intervention group (50.3% vs. 63.1%, p = 0.027).18 The effect sustained up to 1 year with 13% absolute
322 risk reduction and led to 67% improvement in 6-year survival.19 Previous studies have shown that cardiogenic shock could account for the vast majority of deaths in all types of myocardial ischemia.20 The outcome of acute myocardial infarction complicated with cardiogenic shock remains to be unsatisfactory with the mortality rate between 65% and 80% over the past two decades.21—25 Mortality at non-U.S. sites (68% and 64%) was higher than at U.S. sites (53% and 50%) in both GUSTO-I and GUSTO-III studies, respectively. Angioplasty, bypass surgery, and balloon pump rates were lower for non-U.S. patients.26 In our series (from 1995 to 2004) which accrued patients in a similar time span to the SHOCK registry (from 1994 to 2005), the 30-day mortality was only 35.0% in the STEMI group. Among different etiologies besides OHCA, patients with STEMI had the worse prognosis. Patients with ACS and CHF also had relatively poor outcome compared to those who experienced cardiovascular instability after elective PCI and CABG. As we have shown that the extent of cardiac injury (or troponin I level) largely paralleled with underlying etiology, this finding suggests that cardiac injury is usually more controllable and reversible in the closely monitored scenario (PCI and CABG) and the outcome is more favorable. In addition to underlying etiology, age and renal function are found to be significantly prognostic predictors for 30day mortality in the patients with cardiogenic shock. In the SHOCK trial, the group of patients aged ≥75 years did not appear to derive the mortality benefit from early revascularization versus initial medical stabilization that was seen in patients aged <75 years. This might be attributed to confounding factors such as left ventricular ejection fraction (LVEF) and jeopardy score.27 In our study, we demonstrated that older age could have increased the hazard even under IABP usage, comparable to the findings in the SHOCK trial. Poor renal function has been found to be a significantly negative prognostic predictor for 30-day mortality. As fluid overloads and central venous pressure increases, preload demand on the heart also increases and cardiac dysfunction worsens. Previous studies showed that anuria and azotemia were very poor prognostic features for cardiogenic shock28,29 ; as well our study demonstrated that worse renal function manifested with higher serum creatinine level was associated with a poor outcome. The dependence of ECMO to maintain survival could be regarded as IABP treatment failure. Therefore, another Cox regression model was applied to associate previously used predictors with ECMO-free survival. It turned out that hazard ratio was 1.013 (p = 0.05) for age and 1.007 (p = 0.478) for serum creatinine level. Compared to OHCA, HR was 0.321 (p < 0.001) for ACS, 0.364 (p < 0.001) for STEMI, 0.334 (p = 0.008) for CHF, 0.069 (p < 0.001) for post-CABG, and 0.132 for post-PCI (p < 0.001). Clinical implications were the same as what we reported, except that serum creatinine level became insignificant. The finding that creatinine was an independent predictor for long-term survival but not for ECMO-free survival could imply the importance of adequate renal function in maintaining physiological function during extensive life support. In a Sweden series of all patients suffering from OHCA between May 2003 and May 2005, the overall survival to discharge from hospital among the 508 patients was 8.5%.30 A
S.-N. Chang et al. 26% survival among the OHCA group in our series has demonstrated a potential beneficial effect of early IABP usage in the most catastrophic group. On the other hand, according to the Council of Teaching Hospital Affiliation in the American Hospital Association database, the overall in-hospital mortality rate for acute coronary syndrome was 4.8%, ranging from 4.5% at academic hospitals to 4.9% at nonacademic sites (p = 0.022). Cardiogenic shock was found among 2.5—3% of ACS patients.31 A relatively high mortality in patients with ACS (27%) in our series was possibly attributed to the reservation of IABP usage to the most severe subjects and to the delay of IABP insertion. It was a usual practice in our institute that the patients had to have cardiogenic shock with maximal inotropic agents before IABP was ever considered. This convention might be too conservative in the usage of IABP, and it is important to stress the timely insertion of IABP in the high-risk patients. Although randomized and non-randomized studies have consistently suggested that emergency coronary angiography and revascularization reduce the mortality of patients with cardiogenic shock,32—35 it remains as a crucial clinical question as to find the best treatment strategy in the patients who experience ACS or STEMI and sustain in cardiogenic shock. Since CABG was usually reserved to patients with less favorable conditions but resulted in a better survival trend, we could have under-estimated the actual benefit of surgical intervention. However, it is the coronary anatomy and the clinical status of the patient at the time which determines these groups and it is not a decision that clinicians can make as to which group a patient goes into. Therefore, it is not possible to draw a conclusion as to find the optimal treatment strategy: PCI, CABG, or both. One plausible approach is to extract comparable cases from both PCI and CABG groups by means of balanced matches (e.g., propensity-score based methods) when a randomized trial is not available yet.
Limitations Due to the retrospective design, information about further analysis of potential predictors in individual groups was largely unavailable. In the CHF group, potential predictors such as left ventricular function, degree of coronary atherosclerosis, presence of a biventricular pacemaker or ICD, may all be predictive in this group but this predictive ability is lost in diluting it with the other groups. Euroscore or Parsonnet predictors, which were found to be important prognostic factors in post-CABG groups, were not applied.7 Sufficient information should be collected before initiating a new prospective study with long-term follow-up.
Conclusions This study has shown that the mortality rate of cardiogenic shock is higher in patients with STEMI, ACS and CHF as compared to those who underwent complicated PCI or CABG. Age, worsening renal function, and the extent of etiologyrelated cardiac injury are major determinants of 30-day mortality. Although surgical intervention might be beneficial to the high-risk patients, the optimal revascularization strategy needs further validation.
IABP, cardiogenic shock, and prognostic predictors
Conflict of interest None.
Acknowledgements We would like to thank those doctors and health providers in National Taiwan University Hospital who provided high quality of care to the patients with cardiogenic shock. The assistance by Professor Feipei Lai, Miss Man-Hsiang Chang, and the Information Systems Office of NTUH is much appreciated.
References 1. Hollenberg SM. Cardiogenic shock. Crit Care Clin 2001;17(2): 391—410. 2. Santa-Cruz RA, Cohen MG, Ohman EM. Aortic counterpulsation: a review of the hemodynamic effects and indications for use. Catheter Cardiovasc Interv 2006;67(1):68—77. 3. Szatmary L, Marco J, Fajadet J, Caster L, Carrie D. Intra-aortic balloon counterpulsation and coronary angioplasty in high risk coronary heart patients. Acta Med Hung 1987;44(2/3):189—200. 4. Al-Joundi B, Kern MJ, Aguirre FV, Donohue TJ, Moore JA, Flynn MS. Interventional physiology. Part XVIII: Influence of intra-aortic balloon counterpulsation and collateral flow reversal during multivessel angioplasty. Cathet Cardiovasc Diagn 1995;35(2):149—58. 5. Kahn JK, Rutherford BD, McConahay DR, Johnson WL, Giorgi LV, Hartzler GO. Supported ‘‘high risk’’ coronary angioplasty using intraaortic balloon pump counterpulsation. J Am Coll Cardiol 1990;15(5):1151—5. 6. Briguori C, Sarais C, Pagnotta P, et al. Elective versus provisional intra-aortic balloon pumping in high-risk percutaneous transluminal coronary angioplasty. Am Heart J 2003;145(4):700—7. 7. Dunning J, Au JKK, Millner RWJ, Levine AJ. Derivation and validation of a clinical scoring system to predict the need for an intra-aortic balloon pump in patients undergoing adult cardiac surgery. Interact CardioVasc Thorac Surg 2003;2:639— 43. 8. Antman EM, 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(August 31 (9)):e82— 292. 9. Mueller H, Ayres SM, Conklin EF, et al. The effects of intra-aortic counterpulsation on cardiac performance and metabolism in shock associated with acute myocardial infarction. J Clin Invest 1971;50(9):1885—900. 10. Nanas JN, Moulopoulos SD. Counterpulsation: historical background, technical improvements, hemodynamic and metabolic effects. Cardiology 1994;84(3):156—67. 11. National Taiwan University Hospital (NTUH 2007). The medical center—–past, present, and future. Available at: http://ntuh. mc.ntu.edu.tw/english/html/about/index.htm [last retrieved on July4, 2007]. 12. NTUH Information Systems Office 2007 Available at: http://ntuh.mc.ntu.edu.tw/misplus/english homepage/brief introduction.htm [last retrieved on July 4, 2007]. 13. Chang SN, Lin JW, Liu SC, Hwang JJ. Measuring the process of quality of care for ST-segment elevation acute myocardial infarction through data-mining of the electronic discharge notes. J Eval Clin Pract 2008;14(1):116—20.
323 14. Chen YS, Wang MJ, Chou NK, et al. Rescue for acute myocarditis with shock by extracorporeal membrane oxygenation. Ann Thorac Surg 1999;68(6):2220—4. 15. Chen YS, Chao A, Yu HY, et al. Analysis and results of prolonged resuscitation in cardiac arrest patients rescued by extracorporeal membrane oxygenation. J Am Coll Cardiol 2003;41(2):197—203. 16. Hollenberg SM, Kavinsky CJ, Parrillo JE. Cardiogenic shock. Ann Intern Med 1999;131(1):47—59. 17. Sanborn TA, Sleeper LA, Bates ER, et al. Impact of thrombolysis, intra-aortic balloon pump counterpulsation, and their combination in cardiogenic shock complicating acute myocardial infarction: a report from the SHOCK Trial Registry. SHould we emergently revascularize Occluded Coronaries for cardiogenic shocK? J Am Coll Cardiol 2000;36(3 Suppl A):1123—9. 18. 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(9):625—34. 19. Hochman JS, et al. Early revascularization and long-term survival in cardiogenic shock complicating acute myocardial infarction. JAMA 2006;295:2511—5. 20. Iakobishvili Z, Behar S, Boyko V, Battler A, Hasdai D. Does current treatment of cardiogenic shock complicating the acute coronary syndromes comply with guidelines? Am Heart J 2005;149(1):98—103. 21. Goldberg RJ, Samad NA, Yarzebski J, Gurwitz J, Bigelow C, Gore JM. Temporal trends in cardiogenic shock complicating acute myocardial infarction. N Engl J Med 1999;340(15):1162—8. 22. Goldberg RJ, Gore JM, Thompson CA, Gurwitz JH. Recent magnitude of and temporal trends (1994—1997) in the incidence and hospital death rates of cardiogenic shock complicating acute myocardial infarction: the second national registry of myocardial infarction. Am Heart J 2001;141(1):65—72. 23. 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(8):838—42. 24. Hochman JS, Buller CE, Sleeper LA, et al. Cardiogenic shock complicating acute myocardial infarction—etiologies, management and outcome: a report from the SHOCK Trial Registry. SHould we emergently revascularize Occluded Coronaries for cardiogenic shocK? J Am Coll Cardiol 2000;36(September (3 Suppl A)):1063—70. 25. Hands ME, Rutherford JD, Muller JE, et al. The in-hospital development of cardiogenic shock after myocardial infarction: incidence, predictors of occurrence, outcome and prognostic factors. The MILIS Study Group. J Am Coll Cardiol 1989;14(1):40—6 [discussion 47—48]. 26. Menon V, Hochman JS, Stebbins A, et al. Lack of progress in cardiogenic shock: lessons from the GUSTO trials. Eur Heart J 2000;21(December (23)):1928—36. 27. Prasad A, Lennon RJ, Rihal CS, Berger PB, Holmes DR. Outcomes of elderly patients with cardiogenic shock treated with early percutaneous revascularization. Am Heart J 2004;147(6):1066—70. 28. Mielniczuk L, Mussivand T, Davies R, et al. Patient selection for left ventricular assist devices. Artif Organs 2004;28(2): 152—7. 29. Rao V, Oz MC, Flannery MA, Catanese KA, Argenziano M, Naka Y. Revised screening scale to predict survival after insertion of a left ventricular assist device. J Thorac Cardiovasc Surg 2003;125(4):855—62. 30. Axelsson C, Axelsson AB, Svensson L, Herlitz J, Characteristics. outcome among patients suffering from out-of-hospital cardiac arrest with the emphasis on availability for intervention trials. Resuscitation 2007;75(December (3)):460—8.
324 31. Patel MR, Chen AY, Roe MT, et al. A comparison of acute coronary syndrome care at academic and nonacademic hospitals. Am J Med 2007;120(January (1)):40—6. 32. Lee L, Bates ER, Pitt B, Walton JA, Laufer N, O’Neill WW. Percutaneous transluminal coronary angioplasty improves survival in acute myocardial infarction complicated by cardiogenic shock. Circulation 1988;78(6):1345—51. 33. Moosvi AR, Khaja F, Villanueva L, Gheorghiade M, Douthat L, Goldstein S. Early revascularization improves survival in cardiogenic shock complicating acute myocardial infarction. J Am Coll Cardiol 1992;19(5):907—14.
S.-N. Chang et al. 34. Eltchaninoff H, Simpfendorfer C, Franco I, Raymond RE, Casale PN, Whitlow PL. Early and 1-year survival rates in acute myocardial infarction complicated by cardiogenic shock: a retrospective study comparing coronary angioplasty with medical treatment. Am Heart J 1995;130(3 Pt 1):459—64. 35. Berger PB, Holmes Jr DR, Stebbins AL, Bates ER, Califf RM, Topol EJ. 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(1):122—7.
available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/resuscitation
CLINICAL PAPER
Clinical experience with intra-aortic balloon counterpulsation over 10 years: A retrospective cohort study of 459 patients Sheng-Nan Chang a,1, Juey-Jen Hwang a,b,1, Yih-Sharng Chen a,c, Jou-Wei Lin a,∗, Fu-Tien Chiang b a
Cardiovascular Center, National Taiwan University Hospital Yun-Lin Branch, Dou-Liou City, Yun-Lin, Taiwan Section of Cardiology, Department of Medicine, National Taiwan University Hospital, Taipei, Taiwan c Division of Cardiovascular Surgery, Department of Surgery, National Taiwan University Hospital and National Taiwan University College of Medicine, Taipei, Taiwan b
Received 23 October 2007; received in revised form 12 December 2007; accepted 6 January 2008
KEYWORDS Shock, Cardiogenic; Intra-aortic balloon pumping; Extracorporeal membrane oxygenation; Bypass; Coronary artery; Angioplasty, Transluminal, Percutaneous coronary
Summary Objective: The objective of this study was to identify prognostic predictors for the patients experiencing cardiogenic shock who required the institution of intra-aortic balloon counterpulsation (IABP). Design, setting, and patients: Patients with cardiogenic shock were retrieved from the clinical information system in National Taiwan University Hospital and classified according to their etiology: acute coronary syndrome (ACS), ST segment elevation myocardial infarction (STEMI), congestive heart failure (CHF), hemodynamic instability after post-coronary bypass graft operation (post-CABG) or after percutaneous intervention (post-PCI), and out-of-hospital cardiac arrest (OHCA) victims. Measurements: Kaplan—Meier curves and Cox regression model were applied to evaluate the factors associated with survival. Main results: A total of 459 patients were found to belong to one of six etiology categories between 1995 and 2004. The 30-day mortality was highest in the OHCA group, followed by the STEMI, CHF, ACS, post-PCI, and post-CABG groups in a decreasing frequency (log rank p < 0.001). Peak troponin I level was negatively associated with survival, and its effect largely paralleled with underlying etiology. Age and renal impairment were significant prognostic predictors for 30-day mortality (hazard ratio = 1.031, p < 0.001 and hazard ratio = 1.266, p < 0.001). Comparing
∗ Corresponding author at: Cardiovascular Center, National Taiwan University Hospital Yun-Lin Branch, 579 Yun-Lin Road, Section 2, Dou-Liou City, Yun-Lin 640, Taiwan. Tel.: +886 922 861953; fax: +886 5 5335373. E-mail addresses: [email protected], [email protected] (J.-W. Lin). 1 These authors contributed equally.
0300-9572/$ — see front matter © 2008 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.resuscitation.2008.01.016
IABP, cardiogenic shock, and prognostic predictors
317
to those manifested as OHCA who had the worst outcome, patients in the other etiology groups had significantly better survival. Conclusions: This study has illustrated that age, renal function, and etiology-related cardiac injury are predictors for in-hospital course and mortality in those who experienced cardiogenic shock with IABP. The optimal strategy for revascularization in this high-risk group needs further validation. © 2008 Elsevier Ireland Ltd. All rights reserved.
Introduction Cardiogenic shock is a state of inadequate tissue perfusion due to cardiac dysfunction, which complicates with several cardiovascular diseases such as ST-elevation myocardial infarction (STEMI), non-ST elevation myocardial infarction (NSTEMI), unstable angina (UA), congestive heart failure (CHF), complication of percutaneous coronary intervention (PCI), and post-coronary artery bypass graft (CABG). Mortality and morbidity rates of cardiogenic shock remain frustratingly high in the range between 50% and 80%.1 With the improvement in medical techniques and technologies, intra-aortic balloon counterpulsation (IABP) has become a unique tool to improve and support hemodynamic derangement associated with underlying cardiovascular disease.2 The goal of IABP, which inflates and displaces blood from the descending aorta during the diastolic phase, aims at decreasing cardiac workload, myocardial oxygen consumption, and after-load, and at improving coronary perfusion during diastole. Prophylactic IABP could provide circulatory support in case of pharmacologically uncontrollable chest pain, multivessel angioplasty, and severe left ventricular systolic dysfunction.3—6 Several common and easily available variables such as inotrope usage, cardiogenic shock, priority, left main stem disease, ejection fraction, re-do operation, and recent catheterization were used to assess the probability of needing an IABP prior to cardiac surgery.7 In 2004, AHA/ACC recommended IABP as Class I indication for cardiogenic shock in the management of acute myocardial infarction (AMI).8 AHA/ACC also gives Class IIa recommendation for IABP therapy in patients with severe malignant arrhythmias caused by underlying cardiac ischemia and refractory to medical therapy.2 Providing favorable hemodynamic effects, IABP is suggested as a bridging therapy to emergency revascularization, heart transplantation, and acute valvular repair.9,10 Therefore, it is of clinical interests to find prognostic predictors for survival in the patients with cardiogenic shock under IABP usage. This study was designed to construct a retrospective cohort in which the patients required hemodynamic support with IABP due to cardiac etiology. The objective was to find the factors associated with survival in the high-risk group and to illustrate the roles of IABP in various etiology groups.
Materials and methods Subjects and materials National Taiwan University Hospital (NTUH), founded in 1895, is one of the largest in southeastern Asia.11 A health
information system has been maintained by the Information Systems Office since 1982. For each admission, an electronic discharge note must be entered by attending physicians and residents.12 All the discharge notes from 1995 to 2004 were downloaded to MS ACCESS databases, and the narrative contents of more than 600,000 records were stored in separated fields.13 Keyword search was applied to find the context ‘‘intra-aortic balloon counterpulsation’’ or ‘‘IABP’’ in fields such as ‘‘Present Illness’’ and ‘‘In-Hospital Course.’’ Extracted medical records were then reviewed by a cardiologist to determine the eligibility of each patient. Inclusion criteria were subjects with cardiogenic shock who received IABP institution anytime during the admission. It required the use of inotropic agents to the maximal dose (e.g., dopamine 20 g/(kg min)) with blood pressure lower than 90/60 mmHg before IABP insertion or concurrent use of high-dose dopamine when a complication took place. Sepsis or anaphylaxis was excluded retrospectively at the cardiologist’s discretion. Underlying etiology could be further classified by the cardiologist into the following categories: (1) acute coronary syndrome (ACS) (UA or NSTEMI), (2) STEMI, (3) CHF with hemodynamic changes, (4) patients receiving IABP during CABG and maintaining it post-operatively (post-CABG), (5) patients experiencing complications during elective PCI (post-PCI), and (6) out-ofhospital cardiac arrest (OHCA). Etiology other than the above six categories was excluded from analysis because of infrequent occurrence. Examples were unexplained ventricular tachycardia/fibrillation, acute respiratory distress syndrome, myocarditis, and hemodynamic instability after valvular heart operation.
Technical issues IABP and extra-corporeal membrane oxygenation (ECMO) were applied according to the following principles. IABP is a catheter-based device with a 30—50 ml balloon sac to be chosen according to body height. If the patient was of body height over 182 cm, IABP catheter (Arrow International Inc.) with 50 ml balloon sac was chosen. If the patient was of body height below 162 cm, catheters (Arrow International Inc.) with 30 ml balloon sac were chosen. For the patients with body height between 162 and 182 cm, catheters (Arrow International Inc.) with 40 ml balloon sac were used. IABP was inserted into the femoral artery and further advanced to near the level of the left subclavian artery, with the balloon inflated with helium during diastole. The correct position of the catheter tip was evaluated by fluoroscopic guidance or adjusted after chest radiograph. Extra-corporeal membrane oxygenation was applied to the patients who manifested with catastrophic hemodynamics or to those who did not sustain a stable condition
318
S.-N. Chang et al.
after IABP. ECMO team responded to cardiologists’ call. As described previously,14 femoral cannulation was the first choice and distal limb perfusion was maintained. All circuits were primed with normal saline and heparin (2 U/ml), and blood was rewarmed to 37 ◦ C. Pump flow at 50—100 ml/(kg min) and high-dose catecholamine was applied to keep blood pressure above 60 mmHg.15 In the patients manifested with STEMI, it was attempted to perform primary PCI within the first 12 h after the event. The decision to undergo another PCI or CABG depended on patients’ clinical condition at attending cardiologists’ discretion. In the patients with UA/NSTEMI, the decision to receive early or conservative treatment and subsequent CABG therapy was made according to evolving myocardial ischemia refractory to optimal medical therapy, ongoing ischemia despite successful or failed PCI, complicated PCI, presence of left main stenosis and/or 3-vessel disease, and complex coronary anatomy.
Measurements of predictors and outcomes Outcome measurement, i.e., survival or mortality at Days 7 and 30, was retrieved by the same cardiologist from electronic medical records. Survival time was started on the day the major cardiovascular event occurred, and terminated when a patient died. Patients surviving to discharge were regarded as censored at the discharge date.
Table 1
Demographic and hemodynamic data, including age, underling disease (hypertension and diabetes), vital signs, and treatment courses, were derived from electronic medical records and coded into an EXCEL spread sheet. Laboratory data were retrieved from the laboratory information system. Hemoglobin level, serum creatinine, peak cardiac enzyme (creatine kinase, MB form, and troponin I), and arterial blood gas were collected for further analysis.
Statistics Statistics was performed with SPSS 12.0 (SPSS Inc., Chicago, IL). Subjects were classified into six major categories: (1) ACS, (2) STEMI, (3) CHF, (4) post-CABG, (5) post-PCI, and (6) OHCA. Case number and age distribution were summarized by group. The proportion of subjects receiving ECMO or CABG and, consequently, 7- and 30-day mortality were reported accordingly. A Cox regression model was applied to find the hazard associated with 30-day survival after the initial event. Predictors, including age, sex, the presence of diabetes, hypertension, and smoking, peak cardiac enzyme level, serum creatinine level, hemoglobin, and etiology were added into the model in a forward stepwise approach. Hazard ratio (HR), 95% confidence interval, and p-value were reported. Cardiac enzyme was represented as serum troponin I (TnI) level greater or less than 50 mg/dl. Two dummy
Demographic data of the 459 patients in the retrospective cohort study
Age
67 ± 12 (n = 459)
Sex Men Women
110 (24.0%) 349 (76.0%)
Etiology Acute coronary syndrome ST segment elevation myocardial infarction Congestive heart failure Post-coronary bypass surgery Percutaneous coronary intervention complication Out-of-hospital cardiac arrest
174 (37.9%) 143 (31.2%) 21 (4.6%) 56 (12.2) 34 (7.4%) 31 (6.8%)
Underlying diseases Hypertension Diabetes mellitus Peak troponin I level >50 mg/dl <50 mg/dl unknown Hemodynamics after inotropic agents and IABP use Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg) Heart rate (min−1 ) Laboratory data Hemoglobin (g/dl) Hematocrit (%) Creatinine (mg/dl) pH Bicarbonate (mmol/l)
198 (43.1%) 156 (34.0%) 148 (32.2%) 133 (29.0%) 178 (38.8%) 120 ± 29 (n = 303) 67 ± 18 (n = 302) 87 ± 24 (n = 304) 12.1 ± 2.0 (n = 438) 35.7 ± 6.1 (n = 439) 1.96 ± 5.6 (n = 424) 7.35 ± 0.136 (n = 121) 19.3 ± 5.4 (n = 121)
ACS: acute coronary syndrome; STEMI: ST segment elevation myocardial infarction; CHF: congestive heart failure; CABG: coronary artery bypass graft; PCI: percutaneous coronary intervention; OHCA: out-of-hospital cardiac arrest; ECMO: extra-corporeal membranous oxygenation; SBP: SBP, DBP and HR: systolic blood pressure, diastolic blood pressure and heart rate after management with inotropic agents and IABP insertion.
27.0 35.0 33.3 3.6 5.9 74.2 28.5 16.7 25.2 9.5 1.8 2.9 54.8 18.7 ± ± ± ± ± ± ± 85 89 91 79 84 99 87 16 18 18 15 12 22 18 ± ± ± ± ± ± ± 69 63 63 75 74 53 67 174 143 21 56 34 31 459 ACS STEMI CHF Post-CABG Post-PCI OHCA Total
68 67 61 66 65 63 67
± ± ± ± ± ± ±
11 12 16 9 10 13 12
73:27 79:21 71:29 79:21 79:21 77:23 76:24
57.5 36.4 33.3 100.0 58.8 12.9 52.1
14.4 9.8 23.8 5.4 11.8 29.0 13.1
127 113 107 132 130 92 120
± ± ± ± ± ± ±
29 27 22 18 29 37 29
DBP SBP ECMO (%) CABG (%) Sex (M:F) (%) Age Case number Etiology
Table 2
Etiology of the patients experiencing cardiogenic shock requiring intra-aortic balloon counterpulsation
HR
20 27 25 15 26 29 24
7-Day mortality (%)
30-Day mortality (%)
IABP, cardiogenic shock, and prognostic predictors
319 variables were used to represent troponin I > 50, troponin I < 50, and troponin unknown. The survival among different etiologies was compared to that of the OHCA group. HR less than 1 indicated a proportional survival benefit over OHCA. A factor that was statistically significant (p < 0.05) remained in the survival regression model, and a final parsimonious multiple Cox regression model with statistically predictors was reported. Kaplan—Meier curves were plotted according to different etiology and peak cardiac enzyme level. Furthermore stratified by groups, survival curves were plotted according to whether a patient received ECMO during the admission. In addition, patients in the ACS and the STEMI groups were further stratified by their treatment strategy: (0) neither PCI nor CABG, (1) PCI but no CABG, (2) CABG but no PCI, and (3) both PCI and CABG. Log rank test was used to find survival differences.
Results In the period between 1995 and 2004, a total of 505 patients who experienced cardiogenic shock underwent IABP institution. Forty-six cases were excluded from further analysis due to infrequent etiology. Among the 459 indicated patients, there were three times more men than women. The average age was 66 ± 12 for men and 70 ± 11 for women, respectively (Table 1). Hemodynamic data prior to IABP insertion was largely unavailable from electronic chart review. Heart rate and blood pressure after management with inotropic agents and IABP insertion were reported (Tables 1 and 2). In the subsequent management of these patients, 13.1% received hemodynamic support by extra-corporeal membranous oxygenation and 52.1% of the patients ultimately underwent CABG (Table 2). Major complications directly related to IABP included only one severe low leg ischemia necessitating amputation and one hemorrhagic shock due to iatrogenic aortic trauma. The overall 7- and 30-day mortality was 18.7% and 28.5% in the 459 patients. The mortality rate varied significantly among various etiologies. The 7-day mortality was the highest in the OHCA group, following by the STEMI, ACS, CHF, post-PCI, and post-CABG groups in the decreasing frequency (log rank test p < 0.001). The 30-day mortality was also highest in the OHCA group, but the sequence was slightly different (STEMI > CHF > ACS > post-PCI > post-CABG) (log rank test p < 0.001, Figure 1). In addition to underlying etiology, the extent of cardiac damage was also shown to be associated with the outcome. Patients with the TnI level greater than 50 mg/dl had a higher 30-day mortality rate than those with TnI less than 50 mg/dl (log rank test p = 0.003, Figure 2). It was also shown that the extent of cardiac elevation paralleled with 30-day mortality in each etiology category, suggesting that the underlying pathophysiology of the heart determined the extent of cardiac damage and therefore the outcome (Figure 3). A Cox regression model identified that age, renal function, and underlying etiology were significant prognostic predictors for 30-day mortality in the patients with cardiogenic shock. Each year increase in age would lead to 1.031-fold of hazard (95% CI: 1.014—1.049, p < 0.001).
320
S.-N. Chang et al.
Figure 1 30-Day survival by etiology. ACS: acute coronary syndrome; STEMI: ST segment elevation myocardial infarction; CHF: congestive heart failure; post-CABG: post-coronary artery bypass graft; post-PCI: post-elective percutaneous coronary intervention; OHCA: out-of-hospital cardiac arrest.
Elevated serum creatinine level was also associated with a worse outcome (HR: 1.266 for each mg/dl, 95% CI: 1.142—1.403, p < 0.001). Comparing to those manifested as OHCA who had the worst outcome, patients in the other etiology groups had significantly better survival (HR: 0.259 in the ACS group, 0.352 in the STEMI group, 0.302 in the CHF group, 0.027 in the post-CABG group, and 0.045 in the post-PCI group) (Table 3). Serum TnI was no longer associated with 30-day survival when underlying etiology was adjusted. Other factors, such as sex, hemoglobin level, and the presence of diabetes and hypertension, were not significant predictors in the survival analysis. ECMO was usually applied to the patients with catastrophic hemodynamic events; therefore, its use also indicated a poor outcome. Classified by etiology, patients with ECMO had a worse 30-day outcome in the ACS group and the CHF group (log rank p = 0.0002 and 0.0241), but not in other groups (Figure 4). The difference in mortality might be not related to the procedure but to the fact that they were sicker. In those who presented with ACS, the patients were further stratified into the following four subgroups: (0) neither PCI nor CABG, (1) PCI but no CABG, (2) CABG but no PCI, and
Figure 3 Percentage of high troponin I (TnI > 50 mg/dl) and 30-day mortality by etiology.
Table 3 Cox regression model to identify parameters associated with 30-day mortality: hazard ratio, p-value, and 95% confidence interval (CI) Variables
Figure 2 30-Day survival by troponin I (TnI greater or less than 50 mg/dl).
Age Creatinine Etiology ACS STEMI CHF Post-CABG Post-PCI OHCA
Hazard ratio
p-Value
95% CI Lower
Upper
1.031 1.266
<0.001 <0.001
1.014 1.142
1.049 1.403
0.259 0.352 0.302 0.027 0.045 1
<0.001 <0.001 0.007 <0.001 0.003
0.147 0.198 0.126 0.006 0.006
0.455 0.624 0.725 0.117 0.339
IABP, cardiogenic shock, and prognostic predictors
321
Figure 4 ECMO and survival in the six etiology categories (solid line: without ECMO; dashed line: with ECMO), upper left: ACS group, upper middle: STEMI group, upper right: CHF group, lower left: post-CABG group, lower middle: post-PCI group and lower right: OHCA group.
Figure 5 Treatment strategy and survival in the ACS group (left) and in the STEMI group (right). PCI: percutaneous coronary intervention; CABG: coronary artery bypass graft; neither: neither PCI nor CABG; both: both PCI and CABG.
(3) both PCI and CABG. Survival was highest in subgroup 2, followed by subgroups 3, 1, and 0. However, there was only a trend but no statistical significance in the survival between subgroups 2 and 3 (log rank p = 0.328) and between subgroups 3 and 1 (p = 0.322). The direct comparison between the CABG-only patients (subgroup 2) and PCI-only ones (subgroup 1) favored CABG in a borderline statistical significance (p = 0.085). Patients receiving neither PCI nor CABG must have been derived from the most catastrophic subgroup and showed the worse prognosis (log rank p < 0.001) (Figure 5, Panel A). In those who experienced STEMI who had cardiogenic shock, patients receiving both PCI and then CABG had a trend of better survival than those with PCI only (log rank p = 0.170) while those who received neither had the worst outcome (p < 0.001). No subject received only CABG without primary PCI (Figure 5, Panel B).
Discussion Pathophysiology of cardiogenic shock has been described as a cascade of events initiated by myocardial dysfunction, which leads to both increasing left ventricular filling pressure and decreasing stroke volume.16 As the Should We Emergently Revascularize Occluded Coronaries for Cardiogenic Shock (SHOCK) trial showed, initial stabilization of cardiogenic shock patients due to acute myocardial infarction by using IABP (86% of the trial cases) might be associated with a 20% absolute risk reduction in mortality.17 Early revascularization by either primary PCI or CABG resulted in a non-significant trend in reducing the primary endpoint of 30-day mortality (46.7% vs. 56.0%, p = 0.11). Six months after intervention, the mortality was lower in the early intervention group (50.3% vs. 63.1%, p = 0.027).18 The effect sustained up to 1 year with 13% absolute
322 risk reduction and led to 67% improvement in 6-year survival.19 Previous studies have shown that cardiogenic shock could account for the vast majority of deaths in all types of myocardial ischemia.20 The outcome of acute myocardial infarction complicated with cardiogenic shock remains to be unsatisfactory with the mortality rate between 65% and 80% over the past two decades.21—25 Mortality at non-U.S. sites (68% and 64%) was higher than at U.S. sites (53% and 50%) in both GUSTO-I and GUSTO-III studies, respectively. Angioplasty, bypass surgery, and balloon pump rates were lower for non-U.S. patients.26 In our series (from 1995 to 2004) which accrued patients in a similar time span to the SHOCK registry (from 1994 to 2005), the 30-day mortality was only 35.0% in the STEMI group. Among different etiologies besides OHCA, patients with STEMI had the worse prognosis. Patients with ACS and CHF also had relatively poor outcome compared to those who experienced cardiovascular instability after elective PCI and CABG. As we have shown that the extent of cardiac injury (or troponin I level) largely paralleled with underlying etiology, this finding suggests that cardiac injury is usually more controllable and reversible in the closely monitored scenario (PCI and CABG) and the outcome is more favorable. In addition to underlying etiology, age and renal function are found to be significantly prognostic predictors for 30day mortality in the patients with cardiogenic shock. In the SHOCK trial, the group of patients aged ≥75 years did not appear to derive the mortality benefit from early revascularization versus initial medical stabilization that was seen in patients aged <75 years. This might be attributed to confounding factors such as left ventricular ejection fraction (LVEF) and jeopardy score.27 In our study, we demonstrated that older age could have increased the hazard even under IABP usage, comparable to the findings in the SHOCK trial. Poor renal function has been found to be a significantly negative prognostic predictor for 30-day mortality. As fluid overloads and central venous pressure increases, preload demand on the heart also increases and cardiac dysfunction worsens. Previous studies showed that anuria and azotemia were very poor prognostic features for cardiogenic shock28,29 ; as well our study demonstrated that worse renal function manifested with higher serum creatinine level was associated with a poor outcome. The dependence of ECMO to maintain survival could be regarded as IABP treatment failure. Therefore, another Cox regression model was applied to associate previously used predictors with ECMO-free survival. It turned out that hazard ratio was 1.013 (p = 0.05) for age and 1.007 (p = 0.478) for serum creatinine level. Compared to OHCA, HR was 0.321 (p < 0.001) for ACS, 0.364 (p < 0.001) for STEMI, 0.334 (p = 0.008) for CHF, 0.069 (p < 0.001) for post-CABG, and 0.132 for post-PCI (p < 0.001). Clinical implications were the same as what we reported, except that serum creatinine level became insignificant. The finding that creatinine was an independent predictor for long-term survival but not for ECMO-free survival could imply the importance of adequate renal function in maintaining physiological function during extensive life support. In a Sweden series of all patients suffering from OHCA between May 2003 and May 2005, the overall survival to discharge from hospital among the 508 patients was 8.5%.30 A
S.-N. Chang et al. 26% survival among the OHCA group in our series has demonstrated a potential beneficial effect of early IABP usage in the most catastrophic group. On the other hand, according to the Council of Teaching Hospital Affiliation in the American Hospital Association database, the overall in-hospital mortality rate for acute coronary syndrome was 4.8%, ranging from 4.5% at academic hospitals to 4.9% at nonacademic sites (p = 0.022). Cardiogenic shock was found among 2.5—3% of ACS patients.31 A relatively high mortality in patients with ACS (27%) in our series was possibly attributed to the reservation of IABP usage to the most severe subjects and to the delay of IABP insertion. It was a usual practice in our institute that the patients had to have cardiogenic shock with maximal inotropic agents before IABP was ever considered. This convention might be too conservative in the usage of IABP, and it is important to stress the timely insertion of IABP in the high-risk patients. Although randomized and non-randomized studies have consistently suggested that emergency coronary angiography and revascularization reduce the mortality of patients with cardiogenic shock,32—35 it remains as a crucial clinical question as to find the best treatment strategy in the patients who experience ACS or STEMI and sustain in cardiogenic shock. Since CABG was usually reserved to patients with less favorable conditions but resulted in a better survival trend, we could have under-estimated the actual benefit of surgical intervention. However, it is the coronary anatomy and the clinical status of the patient at the time which determines these groups and it is not a decision that clinicians can make as to which group a patient goes into. Therefore, it is not possible to draw a conclusion as to find the optimal treatment strategy: PCI, CABG, or both. One plausible approach is to extract comparable cases from both PCI and CABG groups by means of balanced matches (e.g., propensity-score based methods) when a randomized trial is not available yet.
Limitations Due to the retrospective design, information about further analysis of potential predictors in individual groups was largely unavailable. In the CHF group, potential predictors such as left ventricular function, degree of coronary atherosclerosis, presence of a biventricular pacemaker or ICD, may all be predictive in this group but this predictive ability is lost in diluting it with the other groups. Euroscore or Parsonnet predictors, which were found to be important prognostic factors in post-CABG groups, were not applied.7 Sufficient information should be collected before initiating a new prospective study with long-term follow-up.
Conclusions This study has shown that the mortality rate of cardiogenic shock is higher in patients with STEMI, ACS and CHF as compared to those who underwent complicated PCI or CABG. Age, worsening renal function, and the extent of etiologyrelated cardiac injury are major determinants of 30-day mortality. Although surgical intervention might be beneficial to the high-risk patients, the optimal revascularization strategy needs further validation.
IABP, cardiogenic shock, and prognostic predictors
Conflict of interest None.
Acknowledgements We would like to thank those doctors and health providers in National Taiwan University Hospital who provided high quality of care to the patients with cardiogenic shock. The assistance by Professor Feipei Lai, Miss Man-Hsiang Chang, and the Information Systems Office of NTUH is much appreciated.
References 1. Hollenberg SM. Cardiogenic shock. Crit Care Clin 2001;17(2): 391—410. 2. Santa-Cruz RA, Cohen MG, Ohman EM. Aortic counterpulsation: a review of the hemodynamic effects and indications for use. Catheter Cardiovasc Interv 2006;67(1):68—77. 3. Szatmary L, Marco J, Fajadet J, Caster L, Carrie D. Intra-aortic balloon counterpulsation and coronary angioplasty in high risk coronary heart patients. Acta Med Hung 1987;44(2/3):189—200. 4. Al-Joundi B, Kern MJ, Aguirre FV, Donohue TJ, Moore JA, Flynn MS. Interventional physiology. Part XVIII: Influence of intra-aortic balloon counterpulsation and collateral flow reversal during multivessel angioplasty. Cathet Cardiovasc Diagn 1995;35(2):149—58. 5. Kahn JK, Rutherford BD, McConahay DR, Johnson WL, Giorgi LV, Hartzler GO. Supported ‘‘high risk’’ coronary angioplasty using intraaortic balloon pump counterpulsation. J Am Coll Cardiol 1990;15(5):1151—5. 6. Briguori C, Sarais C, Pagnotta P, et al. Elective versus provisional intra-aortic balloon pumping in high-risk percutaneous transluminal coronary angioplasty. Am Heart J 2003;145(4):700—7. 7. Dunning J, Au JKK, Millner RWJ, Levine AJ. Derivation and validation of a clinical scoring system to predict the need for an intra-aortic balloon pump in patients undergoing adult cardiac surgery. Interact CardioVasc Thorac Surg 2003;2:639— 43. 8. Antman EM, 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(August 31 (9)):e82— 292. 9. Mueller H, Ayres SM, Conklin EF, et al. The effects of intra-aortic counterpulsation on cardiac performance and metabolism in shock associated with acute myocardial infarction. J Clin Invest 1971;50(9):1885—900. 10. Nanas JN, Moulopoulos SD. Counterpulsation: historical background, technical improvements, hemodynamic and metabolic effects. Cardiology 1994;84(3):156—67. 11. National Taiwan University Hospital (NTUH 2007). The medical center—–past, present, and future. Available at: http://ntuh. mc.ntu.edu.tw/english/html/about/index.htm [last retrieved on July4, 2007]. 12. NTUH Information Systems Office 2007 Available at: http://ntuh.mc.ntu.edu.tw/misplus/english homepage/brief introduction.htm [last retrieved on July 4, 2007]. 13. Chang SN, Lin JW, Liu SC, Hwang JJ. Measuring the process of quality of care for ST-segment elevation acute myocardial infarction through data-mining of the electronic discharge notes. J Eval Clin Pract 2008;14(1):116—20.
323 14. Chen YS, Wang MJ, Chou NK, et al. Rescue for acute myocarditis with shock by extracorporeal membrane oxygenation. Ann Thorac Surg 1999;68(6):2220—4. 15. Chen YS, Chao A, Yu HY, et al. Analysis and results of prolonged resuscitation in cardiac arrest patients rescued by extracorporeal membrane oxygenation. J Am Coll Cardiol 2003;41(2):197—203. 16. Hollenberg SM, Kavinsky CJ, Parrillo JE. Cardiogenic shock. Ann Intern Med 1999;131(1):47—59. 17. Sanborn TA, Sleeper LA, Bates ER, et al. Impact of thrombolysis, intra-aortic balloon pump counterpulsation, and their combination in cardiogenic shock complicating acute myocardial infarction: a report from the SHOCK Trial Registry. SHould we emergently revascularize Occluded Coronaries for cardiogenic shocK? J Am Coll Cardiol 2000;36(3 Suppl A):1123—9. 18. 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(9):625—34. 19. Hochman JS, et al. Early revascularization and long-term survival in cardiogenic shock complicating acute myocardial infarction. JAMA 2006;295:2511—5. 20. Iakobishvili Z, Behar S, Boyko V, Battler A, Hasdai D. Does current treatment of cardiogenic shock complicating the acute coronary syndromes comply with guidelines? Am Heart J 2005;149(1):98—103. 21. Goldberg RJ, Samad NA, Yarzebski J, Gurwitz J, Bigelow C, Gore JM. Temporal trends in cardiogenic shock complicating acute myocardial infarction. N Engl J Med 1999;340(15):1162—8. 22. Goldberg RJ, Gore JM, Thompson CA, Gurwitz JH. Recent magnitude of and temporal trends (1994—1997) in the incidence and hospital death rates of cardiogenic shock complicating acute myocardial infarction: the second national registry of myocardial infarction. Am Heart J 2001;141(1):65—72. 23. 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(8):838—42. 24. Hochman JS, Buller CE, Sleeper LA, et al. Cardiogenic shock complicating acute myocardial infarction—etiologies, management and outcome: a report from the SHOCK Trial Registry. SHould we emergently revascularize Occluded Coronaries for cardiogenic shocK? J Am Coll Cardiol 2000;36(September (3 Suppl A)):1063—70. 25. Hands ME, Rutherford JD, Muller JE, et al. The in-hospital development of cardiogenic shock after myocardial infarction: incidence, predictors of occurrence, outcome and prognostic factors. The MILIS Study Group. J Am Coll Cardiol 1989;14(1):40—6 [discussion 47—48]. 26. Menon V, Hochman JS, Stebbins A, et al. Lack of progress in cardiogenic shock: lessons from the GUSTO trials. Eur Heart J 2000;21(December (23)):1928—36. 27. Prasad A, Lennon RJ, Rihal CS, Berger PB, Holmes DR. Outcomes of elderly patients with cardiogenic shock treated with early percutaneous revascularization. Am Heart J 2004;147(6):1066—70. 28. Mielniczuk L, Mussivand T, Davies R, et al. Patient selection for left ventricular assist devices. Artif Organs 2004;28(2): 152—7. 29. Rao V, Oz MC, Flannery MA, Catanese KA, Argenziano M, Naka Y. Revised screening scale to predict survival after insertion of a left ventricular assist device. J Thorac Cardiovasc Surg 2003;125(4):855—62. 30. Axelsson C, Axelsson AB, Svensson L, Herlitz J, Characteristics. outcome among patients suffering from out-of-hospital cardiac arrest with the emphasis on availability for intervention trials. Resuscitation 2007;75(December (3)):460—8.
324 31. Patel MR, Chen AY, Roe MT, et al. A comparison of acute coronary syndrome care at academic and nonacademic hospitals. Am J Med 2007;120(January (1)):40—6. 32. Lee L, Bates ER, Pitt B, Walton JA, Laufer N, O’Neill WW. Percutaneous transluminal coronary angioplasty improves survival in acute myocardial infarction complicated by cardiogenic shock. Circulation 1988;78(6):1345—51. 33. Moosvi AR, Khaja F, Villanueva L, Gheorghiade M, Douthat L, Goldstein S. Early revascularization improves survival in cardiogenic shock complicating acute myocardial infarction. J Am Coll Cardiol 1992;19(5):907—14.
S.-N. Chang et al. 34. Eltchaninoff H, Simpfendorfer C, Franco I, Raymond RE, Casale PN, Whitlow PL. Early and 1-year survival rates in acute myocardial infarction complicated by cardiogenic shock: a retrospective study comparing coronary angioplasty with medical treatment. Am Heart J 1995;130(3 Pt 1):459—64. 35. Berger PB, Holmes Jr DR, Stebbins AL, Bates ER, Califf RM, Topol EJ. 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(1):122—7.