Author + information
- Received May 14, 2012
- Revision received September 4, 2012
- Accepted October 11, 2012
- Published online April 1, 2013.
- Marije M. Vis, MD,
- Marcel A. Beijk, MD, PhD,
- Maik J. Grundeken, MD,
- Jan Baan Jr, MD, PhD,
- Karel T. Koch, MD, PhD,
- Joanna J. Wykrzykowska, MD, PhD,
- E. Karin Arkenbout, MD, PhD,
- Jan G.P. Tijssen, PhD,
- Robbert J. de Winter, MD, PhD,
- Jan J. Piek, MD, PhD and
- José P.S. Henriques, MD, PhD⁎ ()
- ↵⁎Reprint requests and correspondence:
Dr. José P. S. Henriques, Department of Cardiology, Academic Medical Center, University of Amsterdam, Meibergdreef 9, 1105 AZ, Amsterdam, the Netherlands
Objectives This study sought to evaluate 30-day all-cause mortality of patients treated with primary percutaneous coronary intervention (PCI) presenting with an acute myocardial infarction (AMI) due to an unprotected left main coronary artery (ULMCA) culprit lesion. In addition, an average estimated mortality rate was extrapolated from the available data.
Background There are limited data available on clinical outcome after primary PCI in patients presenting with AMI with unprotected left main as the infarct-related coronary artery.
Methods Medical literature databases were searched to identify cohort studies reporting on primary PCI for unprotected left main–related AMI. A total of 13 retrospective studies meeting all pre-specified criteria were included in the meta-analysis. No randomized trials were available. The primary endpoint for the meta-analysis was 30-day all-cause mortality.
Results This meta-analysis comprises a total of 977 patients, of which 252 (26%) presented in cardiogenic shock. Thirty-day all-cause mortality was evaluated using a forest plot analysis and showed higher event rates in patients presenting with cardiogenic shock among all subgroups. The average estimated 30-day all-cause mortality was 15% in patients presenting without signs of cardiogenic shock and 55% in patients presenting with cardiogenic shock (relative risk: 3.74, 95% confidence interval [CI]: 2.95 to 4.76, p < 0.001).
Conclusions In this large meta-analysis of patients treated with primary PCI for AMI due to an ULMCA culprit lesion, the 30-day all-cause mortality in patients presenting with shock is much higher than in patients not presenting with shock. The estimated all-cause mortality data may serve as a benchmark for future reference.
- cardiogenic shock
- myocardial infarction
- percutaneous coronary intervention
- unprotected left main coronary artery disease
A significant involvement of the left main coronary artery occurs in 4% to 7% of patients presenting with an acute myocardial infarction (AMI) (1,2). These critically ill patients frequently present with cardiogenic shock or cardiac arrest and are at high risk for in-hospital major cardiac adverse events (3,4). In nonurgent patients, coronary artery bypass grafting (CABG) has been the standard treatment for significant unprotected left main coronary artery (ULMCA) stenosis and is recommended as first choice of treatment by the American College of Cardiology/American Heart Association (5) and European Society of Cardiology (6) guidelines. In selected patients, however, several studies have reported increasingly good immediate and 1-year outcomes of ULMCA stenting comparable to the outcomes reported after CABG (7–11). Long-term follow-up of the SYNTAX (SYNergy Between PCI With TAXUS and Cardiac Surgery) trial showed that overall major adverse cardiac and cerebrovascular event rates were not significantly different between the percutaneous coronary intervention (PCI) group compared with the CABG group in the pre-specified left main coronary artery subgroup at 3 years (12). Although more PCI patients required reintervention, less stroke was observed compared with CABG-treated patients.
In patients presenting with AMI, primary PCI is considered the optimal reperfusion strategy (13). Whether PCI is the preferred therapy for significant involvement of ULMCA is still subject to debate. At present, the American Heart Association and American College of Cardiology guidelines are less clear about the optimal standard of care for patients presenting with an AMI due to a ULMCA culprit lesion (14). However, it is generally accepted to perform PCI in high-risk patients, such as patients in cardiogenic shock or cardiac arrest where CABG is associated with too high a mortality risk. Currently, there are only limited data on immediate percutaneous treatment for patients with an AMI due to a ULMCA culprit lesion. Especially for those complicated by cardiogenic shock, the left main shock syndrome, only small cohorts have been described.
We therefore performed a systematic review on the currently available literature and a meta-analysis on the treatment of PCI for an AMI due to a ULMCA culprit lesion. Patients were categorized according to initial clinical presentation of AMI with or without cardiogenic shock. Our second goal was to calculate an average mortality rate for this patient category from the available data, which may serve as a benchmark for future studies in this very high-risk patient group.
Inclusion criteria for the meta-analyses
To date, there are no randomized trials concerning primary PCI for left main coronary artery AMI. We therefore only included outcome data from cohort studies describing:
1. patients undergoing primary PCI for an AMI due to a ULMCA culprit lesion;
2. patients with reported 30-day all-cause mortality and;
3. a clear description of whether patients presented with or without cardiogenic shock and/or pre-procedure cardiac arrest with successful cardiopulmonary resuscitation.
Cardiogenic shock or pre-procedure cardiac arrest with successful cardiopulmonary resuscitation hereinafter is referred to as cardiogenic shock. All studies required that either in-hospital or 30-day mortality was available for at least 90% of the patients. In-hospital or 30-day mortality hereinafter is referred to as 30-day all-cause mortality.
We performed a MEDLINE (PubMed) literature search up to August 2011 for cohort studies describing the 30-day outcomes after primary PCI for unprotected left main AMI. Searches included the key words and corresponding Medical Subject Headings (MeSH) for unprotected left main coronary artery disease and AMI. All potentially relevant articles were independently reviewed by 2 investigators (M.M.V. and M.A.B.) to establish eligibility for the meta-analysis. In case of disagreement, this was resolved in consultation with a third reviewer (J.P.S.H.). Citations were screened at title/abstract level and retrieved as full reports. Non-English articles, case reports, and elective or non-urgent interventions were excluded.
The flow chart of the search strategy and selection of studies is depicted in Figure 1. We identified 21 nonrandomized cohort studies for inclusion in our meta-analysis. Eight studies were excluded because 30-day outcomes were only reported for the total study population and did not discriminate between patients with and without cardiogenic shock. One study was excluded because all patients with left main coronary AMI were treated with CABG. Finally, we included the results from a comparative cohort study of primary PCI for an AMI due to a ULMCA culprit lesion from our own research group (the Academic Medical Center [AMC] cohort). Therefore, a total of 13 cohort studies were included in our meta-analysis of primary PCI for an AMI due to a ULMCA culprit lesion (Table 1).
Data extraction and definitions
Pre-specified patient characteristics, the outcome, and the completeness of the follow-up data were independently extracted by 2 investigators (M.M.V. and M.A.B.). Data were grouped for cardiogenic shock at presentation and primary PCI. PCI treatment was clustered between balloon angioplasty (BA) and/or bare-metal stent (BMS), BMS and/or drug-eluting stent (DES), or DES alone.
Unprotected left main was considered to be angiographically documented stenosis >50% located in the left main coronary artery with no patent graft to the left anterior descending or circumflex coronary artery (15). AMI was defined as clinical and/or electrocardiographic signs of an AMI according to the universal definition of AMI (16). In addition, all patients underwent PCI for the left main after angiography. Cardiogenic shock was described in each individual study and was generally defined as systolic blood pressure <90 mm Hg for at least 30 min or the need for supportive measures to maintain a systolic blood pressure above 90 mm Hg and/or the clinical signs associated with end-organ hypoperfusion, such as cool extremities (or a urine output of <30 ml/h, and a heart rate of >60 beats/min) and/or patients presenting with a pre-procedure cardiac arrest with successful cardiopulmonary resuscitation (17).
Data synthesis and analysis
The primary endpoint for the meta-analysis was 30-day all-cause mortality. Results are presented as risk ratios for binary outcome measures with 95% confidence intervals (CIs). Binary outcomes from individual studies were combined respectively, with the random-effect models. We examined heterogeneity across studies by the Cochran's Q statistic and the I2 statistic. Potential publication bias was assessed by visual assessment of constructed funnel plots. Average estimated 30-day all-cause mortality of patients undergoing primary PCI for an AMI due to a ULMCA culprit lesion were calculated as relative risks (RR). Tests were 2-tailed, and a p value of < 0.05 was considered statistically significant. Review Manager (RevMan version 5.1, Nordic Cochrane Centre, Cochrane Collaboration, Copenhagen, Denmark) was used for statistical analysis.
Characteristics of the studies included in the meta-analysis
Searches included the key words and corresponding Medical Subject Headings (MeSH) unprotected left main disease and acute myocardial infarction. We found 406 studies on unprotected left main disease. After entering the second key word, 188 articles remained. After title screening, only 124 studies were selected for further evaluation. A total of 114 articles remained upon excluding non-English articles, case reports, elective or non-urgent interventions. These studies were screened for data on mortality and documentation and presence of cardiogenic shock. Finally, 13 studies (18–29), including our AMC cohort, met all criteria and were included in this meta-analysis (Fig. 1). The funnel plot shows a skewed distribution suggesting publication bias was present. Study characteristics are shown in Table 1. All but 3 were single-center registries. All PCIs were performed between 1990 and 2010. Of the 12 studies, only 3 studies included pre-procedure cardiac arrest patients with successful cardiopulmonary resuscitation: Bonello et al. (19) included 4 cardiac arrest patients, Yip et al. (29) included 6 cardiac arrest patients, Pedrazzini et al. (25) included 37 cardiac arrest patients. In our cohort, 15 patients with pre-procedure cardiac arrest with successful cardiopulmonary resuscitation were included, resulting in a total of 62 of 977 patients (6.3%). Patients with pre-procedure cardiac arrest with successful cardiopulmonary resuscitation and PCI performed were considered patients presenting with cardiogenic shock.
This meta-analysis comprises 977 patients. A total of 252 (26%) patients presented with cardiogenic shock. The observed 30-day all-cause mortality was 57% among patients presenting with cardiogenic shock and 11% among those without cardiogenic shock (risk ratio: 3.15, 95% CI: 1.90 to 5.23). A forest plot was performed from all individual studies presenting with 30-day all-cause mortality of patients with cardiogenic shock and patients without cardiogenic shock (Fig. 2). In 4 cohorts (N = 72, 7.4% of all patients), BA and/or bare-metal stenting was performed; in 7 cohorts (N = 869, 89% of all patients), including our AMC cohort, BMS and/or DES were used; and in 2 cohorts (N = 36, 3.7% of all patients), only DES were implanted (Fig. 2). In the BA and/or BMS subgroups, the observed 30-day all-cause mortality was 59% among patients admitted with cardiogenic shock and 21% among patients without cardiogenic shock (risk ratio 2.09, 95% CI: 0.81 to 5.42). In the BMS or DES subgroup, the observed 30-day all-cause mortality was 56% among patients with cardiogenic shock and 11% among patients without cardiogenic shock (risk ratio: 3.19, 95% CI: 1.70 to 5.97). In the DES subgroup, the observed 30-day all-cause mortality was 63% among patients with cardiogenic shock, whereas no patients died within 30 days when they presented without cardiogenic shock (risk ratio: 11.60, 95% CI: 1.66 to 80.97). In the BA and/or BMS groups and the DES group, there was no evidence of heterogeneity across the different trials. In the BMS and/or DES group, however, there was heterogeneity observed between the trials. Overall, there was a statistically significant heterogeneity across the 13 studies (I2 = 62%).
Average estimated 30-day all-cause mortality
Thirty day all-cause mortality of patients among the different studies and the weight factor per study is shown in Table 2. The calculated average estimated 30-day all-cause mortality was 15% in patients without cardiogenic shock and 55% in patients with cardiogenic shock (RR: 3.74, 95% CI: 2.95 to 4.76, p < 0.001) (Table 3). A similar trend was seen in all subgroups (Table 3).
This is the first meta-analysis on primary PCI for an AMI due to a ULMCA culprit lesion with and without cardiogenic shock. Our main findings are that patients undergoing primary PCI for an AMI due to a ULMCA culprit lesion and presenting with cardiogenic shock have a high 30-day mortality compared with patients without cardiogenic shock, regardless of stent type. In addition, we calculated an average estimated 30-day all-cause mortality in patients with and without cardiogenic shock that may serve as a benchmark for future studies.
Primary PCI for an AMI due to a ULMCA culprit lesion is a rare procedure, frequently associated with adverse clinical outcomes. The incidence of AMI due to a ULMCA culprit lesion is reported to be 0.8% to 5.4% (30). Given the infrequent nature of this left main syndrome, our meta-analysis aimed to review the available data in literature. As expected, a large proportion of patients in this meta-analysis presented with cardiogenic shock (26%). It is likely that the average incidence of this left main syndrome may in fact be higher due to the various clinical conditions and medical decision making in a patient population with an expected very poor outcome. This meta-analysis provides a benchmark for clinical outcome in patients presenting with an AMI due to a ULMCA culprit lesion, which may serve for comparison, especially when patients are presenting with an already high risk for adverse clinical events, and for multiple reasons, physicians may refrain from taking these patients to the catheterization laboratory and further therapy. A striking difference in mortality was observed between the studies included in our meta-analysis. This may be partly explained by differences in definition of shock and patient selection. A trend towards higher mortality over time was observed, perhaps related to the widespread acceptance of primary PCI for more complex patients. In our center, primary PCI has evolved as the standard of care, and all ST-segment elevation myocardial infarction (STEMI) patients within our geographic referral area with or without cardiogenic shock, are transferred for primary PCI. This may explain the relatively high number of patients in our cohort, treated in critically ill conditions, such as cardiogenic shock or cardiac arrest. It is noteworthy that this meta-analysis showed that in patients not in cardiogenic shock who received a DES showed no mortality at 30 days; however, due to the small number of patients, future studies will have to confirm this finding.
In this meta-analysis, the use of balloon angioplasty, BMS or DES revealed similar clinical outcomes in patients presenting with cardiogenic shock. In patients without cardiogenic shock, treatment with DES seem to be associated with an improved 30-day all-cause mortality. However, this may be related to the time period the study was performed because the BA/BMS subgroup studies all included patients before 2004. Over the last years, practice has changed with improved PCI techniques, assist devices, use of thrombosuction devices, and antiplatelet therapy or additional medication. A hybrid approach of initial revascularization by primary PCI and elective surgery afterwards remains an alternative treatment option. Treating patients with cardiogenic shock or after cardiac arrest is one of the most challenging PCI procedures due to the nature of the clinical presentation while targeting a coronary lesion associated with a very large area of the myocardium at risk. These hemodynamically unstable patients have an extensive amount of ischemic myocardium. Compensatory systemic inflammatory stress syndrome and redistribution of blood from the vital organs can lead to multiorgan failure (31). Immediate mechanical hemodynamic support may prevent from further multiorgan failure. The recent CRISP AMI (Counterpulsation to Reduce Infarct Size Pre-PCI Acute Myocardial Infarction) trial did not show a beneficial effect of intra-aortic balloon pumping before PCI for STEMI in non-cardiogenic shock patients (32). However, the ongoing ISAR SHOCK 2 (Efficacy Study of LV Assist Device to Treat Patients With Cardiogenic Shock) trial will ultimately shed light on the effects of intra-aortic balloon pumping in STEMI patients presenting with cardiogenic shock and treated with PCI (33). Whether newer and more powerful circulatory assist devices (Impella/TandemHeart/ECMO) will result in better outcomes is the subject of ongoing clinical evaluation before widespread adoption (34).
This meta-analysis of primary PCI for AMI due to a ULMCA culprit lesion provides a benchmark that may serve for future reference, to optimize treatment of this high-risk patient group with, for example, left ventricular assist devices or hybrid revascularization.
All studies included in this meta-analysis are retrospective, reflecting outcomes after clinical decision making by treating physicians, including taking patients to the catheterization laboratory in very poor condition and the usage of different modes of PCI (BA and different types of intracoronary stents) as well as different types of hemodynamic support devices. In addition, these data are also hampered by clinical decision making in patients that undergo CABG after immediate coronary angiography, as they only describe patients that actually underwent PCI. Moreover, selection bias based on the time differences between CABG or PCI is inevitable. From the studies included in this meta-analysis, no discrimination could be made between patient groups treated solely with BA, BMS, or DES. Detailed descriptions of cause of death were not available, specifically for patients with pre-procedure cardiac arrest with successful cardiopulmonary resuscitation, because those patients might have a worse outcome. Therefore, the primary endpoint of this meta-analysis was 30-day all-cause mortality. Finally, in 4 of the 13 studies included, the angiographic data on the severity of coronary artery disease were not available.
This is the first meta-analysis, to our knowledge, on primary PCI for AMI due to a ULMCA culprit lesion. The observed 30-day all-cause mortality rate was higher in patients presenting with cardiogenic shock compared with patients without cardiogenic shock. The results were not affected by stent type. In addition, for benchmark purposes, we calculated an average estimated 30-day all-cause mortality of 55% for patients with cardiogenic shock and 15% for patients without cardiogenic shock. Table 2
Prof. Dr. Piek is on the medical advisory board of Abbott Vascular; and is a consultant for Miracor. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose. The first two authors contributed equally to this work.
- Abbreviations and Acronyms
- Academic Medical Center
- acute myocardial infarction
- balloon angioplasty
- bare-metal stent(s)
- coronary artery bypass grafting
- confidence interval
- drug-eluting stent(s)
- percutaneous coronary intervention
- relative risk
- ST-segment elevation myocardial infarction
- unprotected left main coronary artery
- Received May 14, 2012.
- Revision received September 4, 2012.
- Accepted October 11, 2012.
- American College of Cardiology Foundation
- Spiecker M.,
- Erbel R.,
- Rupprecht H.J.,
- Meyer J.
- Patel M.R.,
- Bailey S.R.,
- Bonow R.O.,
- et al.
- Silber S.,
- Albertsson P.,
- Avilés F.F.,
- et al.
- Buszman P.E.,
- Kiesz S.R.,
- Bochenek A.,
- et al.
- Chieffo A.,
- Morici N.,
- Maisano F.,
- et al.
- Kappetein A.P.,
- Feldman T.E.,
- Mack M.J.,
- et al.
- White H.D.,
- Assmann S.F.,
- Sanborn T.A.,
- et al.
- Kushner F.G.,
- Hand M.,
- Smith S.C. Jr..,
- et al.
- Puricel S.,
- Adorjan P.,
- Oberhänsli M.,
- et al.
- Thygesen K.,
- Alpert J.S.,
- White H.D.,
- et al.
- Montalescot G.,
- Brieger D.,
- Eagle K.A.,
- et al.
- Pedrazzini G.B.,
- Radovanovic D.,
- Vassalli G.,
- et al.
- Baek J.Y.,
- Seo S.M.,
- Seung K.B.,
- et al.
- Hochman J.S.
- Seyfarth M.,
- Sibbing D.,
- Bauer I.,
- et al.