Author + information
- Received March 28, 2014
- Accepted April 10, 2014
- Published online September 1, 2014.
- Niket Patel, MBBS, BSc∗,
- Giovanni Luigi De Maria, MD∗,
- George Kassimis, MD, MSc, PhD∗,
- Kazem Rahimi, DM, MPH∗,†,
- Derrick Bennett, PhD‡,
- Peter Ludman, MBBChir, MD§ and
- Adrian P. Banning, MBBS, MD∗∗ ()
- ∗Cardiology Department, Oxford Heart Centre, Oxford University Hospitals, Oxford, United Kingdom
- †George Institute for Global Health, University of Oxford, Oxford, United Kingdom
- ‡Clinical Trials Service Unit, University of Oxford, Oxford, United Kingdom
- §Department of Cardiology, University Hospital Birmingham, Birmingham, United Kingdom
- ↵∗Reprint requests and correspondence:
Prof. A.P. Banning, Oxford University Hospitals, Oxford Health Centre, Cardiology Department, Level 2, John Radcliffe Hospital, Headley Way, Oxford OX3 9DU, United Kingdom.
Objectives This study sought to evaluate in-hospital outcomes and 3-year mortality of patients presenting with unprotected left main stem occlusion (ULMSO) treated with primary percutaneous coronary intervention (PPCI).
Background Limited data exists about management and outcome following presentation with ULMSO.
Methods From January 1, 2007 to December 21, 2012, 446,257 PCI cases were recorded in the British Cardiovascular Intervention Society database of all PCI cases in England and Wales. Of those, 568 were patients having emergency PCI for ST-segment elevation infarction (0.6% of all PPCI) who presented with ULMSO (TIMI [Thrombolysis In Myocardial Infarction] flow grade 0/1 and stenosis >75%), and they were compared with 1,045 emergency patients treated with nonocclusive LMS disease. Follow-up was obtained through linkage with the Office of National Statistics.
Results Presentation with ULMSO, compared with nonocclusive LMS disease, was associated with a doubling in the likelihood of periprocedural shock (57.9% vs. 27.9%; p < 0.001) and/or intra-aortic balloon pump support (52.5% vs. 27.2%; p < 0.001). In-hospital (43.3% vs. 20.6%; p < 0.001), 1-year (52.8% vs. 32.4%; p < 0.001), and 3-year mortality (73.9% vs 52.3%, p < 0.001) rates were higher in patients with ULMSO, compared with patients presenting with a patent LMS, and were significantly influenced by the presence of cardiogenic shock. ULMSO and cardiogenic shock were independent predictors of 30-day (hazard ratio [HR]: 1.61 [95% confidence interval (CI): 1.07 to 2.41], p = 0.02, and HR: 5.43 [95% CI: 3.23 to 9.12], p<0.001, respectively) and 3-year all-cause mortality (HR: 1.52 [95% CI: 1.06 to 2.17], p = 0.02, and HR: 2.98 [95% CI: 1.99 to 4.49], p < 0.001, respectively).
Conclusions In patients undergoing PPCI for ULMSO, acute outcomes are poor and additional therapies are required to improve outcome. However, long-term outcomes for survivors of ULMSO are encouraging.
Presentation to the cardiac catheterization lab with myocardial infarction (MI) caused by unprotected left main stem occlusion (ULMSO) is very unusual. It is assumed that when it occurs acutely, it usually results in sudden cardiac death and that many of these patients never reach medical services and die in the community (1). Because of the widespread availability of primary percutaneous coronary intervention (PPCI) for acute MI, several small registries have investigated outcomes of patients undergoing emergency PPCI to left main stem (LMS) disease (2). However, only sporadic individual cases (3–7) and 5 small series (cumulatively n = 112) (8–12) have reported outcomes of patients suffering from occlusive or subocclusive LMS disease (8–12) from which there is limited 1-year mortality data available (total n = 100) (8,10–12).
These small studies, from several different populations, suggest that survival is poor (8–14); however, reliable contemporary information has not been collated previously, and this group has not been compared with outcomes from other emergency patients with MI. Given the paucity of data, management of this potentially catastrophic presentation poses a very significant challenge. We interrogated data from the British Cardiovascular Intervention Society’s (BCIS) national PCI audit with case-based mortality tracking for 3 years following intervention. Using these data, we hoped to inform clinical decision making in the absence of existing guidelines for emergency treatment.
The BCIS national audit of PCI is a prospective registry of all coronary intervention preformed in all interventional cardiology units within the United Kingdom since 2005 (117 institutions in 2011) (15). A total of 113 fields of clinical, procedural, and outcome data are collected (Online Appendix) centrally at the National Institute for Cardiovascular Outcomes Research at University College London (15). All-cause mortality data of the audit cohort in England and Wales is tracked by the Office of National Statistics.
The completeness and internal consistency of the data are assessed during submission by a set of validation rules (Online Appendix). If a major error is detected that might cause the upload to contaminate the reliability of the complete dataset, then the record is rejected for resubmission. Less serious inconsistencies are accepted, but an error log documents fatal and serious errors, allowing units to clean and correct their data locally.
There were 446,257 PCI records generated in England and Wales between January 1, 2007, and December 31, 2012. There were 105,216 procedures performed as an emergency (defined as treatment with primary PCI) on 102,057 patients. There were 2,125 patients identified as having undergone unprotected LMS PPCI (Figure 1). This included patients in whom the LMS was treated in isolation or in combination with other vessels. From the unprotected LMS PPCI cohort, information on TIMI (Thrombolysis In Myocardial Infarction) flow grade and lesion severity was available in 1,613 patients. There were 568 patients who presented with ULMSO (defined as pre-procedural LMS stenosis ≥75% and TIMI flow grade ≤1) and who were compared with 1,045 patients with nonocclusive pre-procedural LMS disease (defined as LMS stenosis <75% and/or TIMI flow grade ≥2). Patient groups were also divided based on the presence and absence of periprocedural cardiogenic shock.
Clinical endpoints and definitions
We report the following: 1) demographic, clinical, and procedural characteristics of patients with ULMSO; 2) 3-year all-cause mortality of the patients presenting with ULMSO stratified by the presence and absence of shock; 3) in-hospital rates of major adverse cardiac and cerebrovascular events (MACCE) defined as an accumulative composite of death, reinfarction or reintervention, in-hospital coronary artery bypass graft (CABG) and cerebrovascular events. Additionally, we aim to determine those factors that predict short- and long-term mortality.
Normally distributed continuous variables (e.g., age and body mass index) are presented as mean ± SD and were analyzed using the independent samples Student t test. Skewed continuous variables (e.g., length of stay) are presented as median (interquartile range) and were analyzed using the Mann-Whitney U test. Categorical data are presented as counts and proportions of valid cases from the database, and statistical comparisons were made using the chi-square statistic. All-cause mortality rates are presented as counts and percentages and as mortality plots with the number of subjects known to be at risk at each successive time point. Group differences are assessed using the log-rank test. A landmark analysis was then performed for 30-day and 1-year survivors. A multivariable Cox proportional hazards regression model was used to estimate the independent predictors of 30-day and 3-year mortality. Sequential univariate models were performed for biologically and clinically relevant covariates (age; sex; cardiovascular risk factors; renal dysfunction; history of previous MI, PCI, or CABG; symptom to balloon time; recent lysis; left ventricular [LV] dysfunction; periprocedural cardiogenic shock; occlusive LMS disease; multivessel disease; intra-aortic balloon pump support; glycoprotein (GP) IIb/IIIa inhibitor use; thrombus aspiration use; and the presence of post-procedural no-reflow). Covariates with a p value of <0.05 at univariate analysis were entered into a final model using a forward stepwise method. Proportional hazards assumptions were evaluated and met for both outcomes and hazard ratios (HR) are presented as HR (95% confidence interval [CI]). All p values are 2-sided, and a p value of <0.05 was considered to be statistically significant. All analyses were conducted in SPSS (version 20, SPSS, Inc., Chicago, Illinois).
During the investigation period, in England and Wales, 102,057 patients presented for emergency PPCI with a 30-day mortality of 5.9%. The absolute annual number of PPCI cases increased from 8,712 in 2007 to 23,632 in 2012, as did the proportion of PPCI relative to all interventions during the study period (8.7% to 23.6% per year). The rate of unprotected LMS PPCI also increased as a proportion of all PPCI from 2.0% to 2.2% per year. A total of 2,125 patients undergoing unprotected LMS PPCI were identified (2.1%). This cohort was compared with 97,974 patients undergoing non-LMS emergency PPCI. The 568 patients (0.6% of all PPCI) with ULMSO undergoing PPCI were then compared with the other 1,045 emergency PPCI patients with nonocclusive LMS disease.
Baseline demographic and clinical characteristics
Baseline demographic and clinical data for patients within the investigation cohorts are presented in Table 1. Patients undergoing PPCI treatment for unprotected LMS disease, when compared with those presenting with non-LMS MI, were >5 years older, more frequently women, and more likely to have traditional risk factors of coronary artery disease such as current smoking, diabetes, hypertension, hypercholesterolemia, and/or renal dysfunction (Table 1). Apart from hypertensive status and history of previous PCI, patients with ULMSO were well matched with those with nonocclusive LMS disease for age, sex, and cardiovascular risk factors (Table 1). The ULMSO group had a higher frequency of presentation with LV dysfunction (85.6% vs. 68.2%; p < 0.001) and was approximately twice as likely to present with periprocedural cardiogenic shock (57.9% vs. 27.9%; p < 0.001). The median symptom to PCI hospital and symptom to balloon intervals were significantly shorter in the ULMSO patients than they were in nonocclusive LMS disease patients (median: 2.1 [interquartile range (IQR): 1.3 to 4.2] vs. 2.6 [IQR: 1.4 to 6.0] h, p = 0.001, and median: 3.3 [IQR: 2.2 to 5.6] vs. 3.9 [IQR 2.4 to 7.5] h, p < 0.001, respectively).
Procedural characteristics and outcomes
Patients presenting for unprotected LMS PPCI, compared with the non-LMS PPCI cohort, had a higher frequency of femoral access use (68.4% vs. 53.7%; p < 0.001) and multivessel intervention (2.02 ± 0.88 vs. 1.10 ± 0.37; p < 0.001) (Table 2). Additionally, they more frequently required the use of mechanical ventilation (16.7% vs. 3.5%, respectively; p < 0.001) and circulatory support (57.1% vs. 7.1% respectively; p < 0.001). Following intervention, the median length of in-patient hospital stay was significantly longer for those having unprotected LMS PPCI (4.05 [IQR: 2.0 to 8.6] vs. 2.4 [IQR: 1.5 to 3.6] days; p < 0.001). In keeping with the higher rates of LV dysfunction and cardiogenic shock in the ULMSO group, the need for mechanical ventilation (23.2% vs. 12.5% respectively; p < 0.001) and circulatory support (87.5% vs. 40.9% respectively; p < 0.001) were significantly higher than for the nonocclusive LMS PPCI group. Furthermore, intervention was approximately 2× more likely to end with final no-reflow or slow flow in the ULMSO PPCI group than in the nonocclusive LMS PPCI group (7.5% vs. 3.1%, respectively; p < 0.001). The use of an aspiration catheter was significantly higher in the ULMSO PPCI group (38.7% vs. 23.0%, respectively; p < 0.001), whereas intravascular ultrasound utilization was one-half of that seen in the nonocclusive LMS PPCI group (11.0% vs. 20.7%, respectively; p < 0.001).
In-hospital clinical outcomes
Of all patients having unprotected LMS PPCI, the in-hospital mortality was more than 6-fold higher than that of the non-LMS PPCI cohort (26.2% vs. 4.1%, respectively; p < 0.001) (Table 3). Patients with ULMSO had in-hospital MACCE (43.3% vs. 20.6%; p < 0.001) and mortality rates (41.6% vs. 19.3%; p < 0.001) that were approximately 2× higher than those for patients with nonocclusive LMS disease. In all groups, the in-hospital MACCE rate was principally driven by in-hospital mortality (Table 3). In the ULMSO group, 9 patients (1.6%) underwent urgent or emergency CABG, 7 patients suffered nonfatal MI or underwent reintervention, and a further 2 patients had embolic strokes. Additionally, 30.6% (174) of all ULSMO patients were treated in PCI centers without on-site surgical cover. This was associated with a slightly higher, but not statistically significant, increase in in-hospital mortality (46.2% vs. 39.6%; p = 0.15).
The 1-year and 3-year all-cause mortality rates for patients undergoing unprotected LMS PPCI were 39.8% and 60.4%, respectively, compared with 9.8% and 20.7%, respectively, in those having non-LMS PPCI (p < 0.001) (Table 3). Mortality at both 1 and 3 years was adversely affected by the presence of periprocedural shock by approximately a factor of 2 in the unprotected LMS PPCI group (cardiogenic shock: 65.2% vs. no cardiogenic shock: 23.0%; p < 0.001 at 1 year; and cardiogenic shock: 80.7% vs. no cardiogenic shock: 43.5%; p < 0.001 at 3 years) and by a factor of >3 in the non-LMS PPCI group at 3 years (cardiogenic shock: 44.2% vs. no cardiogenic shock: 7.5%; p < 0.001 at 1 year; and cardiogenic shock: 61.4% vs. no cardiogenic shock: 17.4%; p < 0.001 at 3 years). The mortality plot in Figure 2A shows the mortality rate at 3-year follow-up for unprotected LMS PPCI and non-LMS PPCI groups stratified by the presence and absence of periprocedural cardiogenic shock and confirms a higher mortality in patients admitted with cardiogenic shock. Furthermore, the need for unprotected LMS intervention confers a further mortality risk irrespective of the presence or absence of cardiogenic shock (p < 0.001).
A similar trend was observed for 1- and 3-year all-cause mortality for patients undergoing ULMSO PPCI versus those having nonocclusive LMS PPCI (52.8% vs. 32.0%, p < 0.001 at 1 year; and 73.9% vs. 52.3%, p < 0.001 at 3 years). Furthermore, mortality at both 1 and 3 years were adversely affected by the presence of periprocedural shock in the ULMSO PPCI group (cardiogenic shock: 71.4% vs. no cardiogenic shock: 26.3%; p < 0.001 at 1 year; and cardiogenic shock: 86.1% vs. no cardiogenic shock: 51.4%; p < 0.001 at 3 years), and this affect was more pronounced in the nonocclusive LMS PPCI group (cardiogenic shock: 63.0% vs. no cardiogenic shock: 19.4%; p < 0.001 at 1 year; and cardiogenic shock: 79.2% vs. no cardiogenic shock: 37.7%; p < 0.001 at 3 years). Mortality plots in Figure 2B show the 3-year all-cause mortality in the ULMSO PPCI and nonocclusive LMS PPCI cohorts stratified by the presence and absence of periprocedural cardiogenic shock and confirms a higher mortality observed in patients admitted with cardiogenic shock and with a further adverse effect on mortality conferred by occlusive disease (p < 0.001). Furthermore, within the ULMSO cohort, there is a mortality benefit associated with intra-aortic balloon pump (IABP) use in those with periprocedural cardiogenic shock at 3 years (83.6% vs. 90.8%; p < 0.001) (Figure 3). In the absence of periprocedural cardiogenic shock, in this group, IABP use had a marginally lower mortality rate than for those in whom IABP was not used, but this difference was not statistically significant (53.1% vs. 48.8%; p = 0.47).
The use of aspiration catheters in those with ULMSO was not associated with a mortality benefit at 3 years (74.5% vs. 73.4%; p = 0.92); however, patients receiving GP IIb/IIIa inhibitors had a lower mortality rate than did those patients who were not treated with GP IIb/IIIa inhibitors (66.9% vs. 85.6%; p < 0.001).
Notably, mortality rates diverge early depending on LMS occlusion and cardiogenic shock status, and at 30 days, mortality rates are markedly discordant between the 4 subgroups (log rank p < 0.001) (Figure 4A). A significantly higher mortality was detected at 30 days in patients with ULMSO presenting with cardiogenic shock (74.8%) than in those with nonocclusive LMS PPCI with cardiogenic shock (63.8%), compared with those with UMLSO without cardiogenic shock (32.9%) and those with nonocclusive LMS PPCI without cardiogenic shock (19.0%). Landmark analysis of 30-day survivors, demonstrated a higher mortality in patients with cardiogenic shock on admission (32.4% vs. 12.8%, log rank p < 0.001) irrespective of the presence or absence of occlusive LMS disease (Figure 4B). Conversely, landmark analysis of 1-year survivors demonstrated similar mortality rates up to 3-year follow-up in all 4 subgroups (log rank p = 0.29) (Figure 4C).
Independent predictors of 30-day and 3-year mortality
Multivariable Cox proportional hazards regression modeling showed that the following factors were independent predictors of 30-day mortality in those undergoing unprotected LMS PPCI: ULMSO (HR: 1.61 [95% CI: 1.07 to 2.41); p = 0.02); periprocedural cardiogenic shock (HR: 5.43 [95% CI: 3.23 to 9.12]; p < 0.001); age (HR: 1.02 [95% CI: 1.01 to 1.03]; p = 0.04); female sex (HR: 0.43 [95% CI: 0.27 to 0.69]; p = 0.001); acute LV dysfunction (HR: 2.37 [95% CI: 1.18 to 4.76]; p = 0.01); GP IIb/IIIa inhibitor use (HR: 0.63 [95% CI: 0.87 to 0.96]; p = 0.03); and post-procedural no-reflow (HR: 4.15 [95% CI: 2.20 to 7.81]; p < 0.001) (Figure 5A). At long-term follow-up (3 years), UMLSO (HR: 1.52 [95% CI: 1.06 to 2.17]; p = 0.02), per-procedural cardiogenic shock (HR: 2.98 [95% CI: 1.99 to 4.49]; p < 0.05 01), age (HR: 1.02 [95% CI: 1.01 to 1.03]; p = 0.03), female sex (HR: 0.64 [95% CI: 0.44 to 0.94]; p = 0.02), GP IIb/IIIa use (HR: 0.59 [95% CI: 0.41 to 0.85]; p = 0.004), and occurrence of post-procedural no-reflow (HR: 2.73 [95% CI: 1.56 to 4.79]; p = 0.001) were found to be independent predictors of mortality (Figure 5B).
Emergency presentation with occlusion of the left main coronary artery is arguably the most dramatic and catastrophic coronary event. The existing descriptive literature of patients and outcomes is limited to small single-center reports (8–12). We report the first data from a comprehensive national registry of patients undergoing ULMSO PPCI. Our data documents that the incidence of presenting with ULMSO and undergoing PPCI is 0.6% of all PPCI (approximately 15 cases/year/100,000 population). Despite treatment, in-hospital mortality was 42%; 1-year mortality was 53%; and 3-year mortality was 74%.
The current medical literature of patients presenting with ULMSO is based on approximately 150 cases including those reported as single case reports (3–12), and outcomes from the largest series are summarized in Table 4. These historical data can be contrasted with our presentation of contemporary and comprehensive U.K. interventional practice. In our series, almost all patients underwent coronary stenting (compared with only 56% to 80% of patients in previous reports [8,10–12]), and approximately two-thirds of the stents used were drug eluting. Our study shows that presentation with ULMSO is more likely in older patients, in patients with peripheral vascular disease, previous cerebrovascular accident, and/or renal failure relative to patients undergoing non-LMS PPCI. The lower prevalence of cardiogenic shock in our study compared with that of previous series (57.9%) may explain the lower use of IABP (52.5%) (8–12), but it may also be related to the availability of a wider range of other mechanical hemodynamic support devices (16). Reassuringly, the in-hospital mortality in our study is lower than that previously reported (43.3% vs. an average mortality in previous studies of 54%) (8–12) (Table 4), suggesting perhaps a positive impact on outcomes from contemporary therapies/strategies. However, 1-year mortality of this cohort remains comparable to previous series (52.8% vs. 44% to 70%, respectively) (8,10–12), as does the mortality in ULMSO PPCI patients with cardiogenic shock (71.4% vs. 71% to 79%, respectively), though the historical data are based on small numbers and subgroup analyses (10,13). To our knowledge, there are no long-term (3-year) outcome data for this cohort in the literature.
Our study shows that despite aggressive invasive treatment in-patient mortality remains very high. This information is valuable for responsible healthcare professionals, and patients’ families should be counseled accordingly. However, it is interesting to note that of 30-day survivors of unprotected LMS PPCI, mortality differences in the subsequent year appear to be principally driven by the presence or absence of cardiogenic shock and not primarily by occlusion status. Furthermore, irrespective of shock and occlusion status, no mortality difference was detected in 1-year survivors up to 3 years in these 4 subgroups. It is possible to speculate that this may be related to the presence of extensive myocardial stunning and ultimately myocardial recovery, which is likely to be particularly prevalent in this patient group.
There are currently no specific guidelines for the management of acute ULMSO. In the absence of any reports of conservative medical management or trial data addressing the relative benefit of PCI versus CABG, we propose that the goals of therapy in the acute setting are 3-fold. First, we recommend prompt restoration of left coronary artery myocardial perfusion. CABG can be considered (13,17) but the fulminant presentation of these cases may make it impractical unless a cardiac theater and its team are immediately available. There are no comparative data, and it is unlikely that there will ever be a trial. Attempting to re-establish epicardial coronary blood flow by thrombus aspiration is a logical preliminary step, as this may confirm the anatomy and the location of the guidewire. Reconsideration of revascularization options is probably appropriate at this stage as our study demonstrates that coronary stenting has a high risk of no-reflow, in part due to a relatively large plaque and/or thrombotic volume (18,19), as well as a reflection of extensive myocardial ischemia and necrosis. In support of this, we report a detrimental effect on long-term survival associated with no-reflow (HR: 2.73; p = 0.001) and evidence of a mortality benefit associated with the use of GP IIb/IIIa inhibitors (HR: 0.59; p = 0.004 in those with unprotected LMS; and p < 0.001 in those with ULMSO).
Second, our data emphasize the likelihood of requiring hemodynamic support during the acute management of ULMSO in view of a high shock rate (almost 60%). The majority of previous ULMSO series report almost universal use of IABP and precede recent data that have questioned the utility of IABP, including the IABP-SHOCK II (Intra-Aortic Balloon Counterpulsation in Acute Myocardial Infarction Complicated by Cardiogenic Shock) trial (20), which suggested no benefit in PPCI patients with cardiogenic shock. However, the BCIS-I (Balloon-Pump Assisted Coronary Intervention Study) trial (21) did demonstrate mortality benefit in those with a high jeopardy score and severe left ventricular dysfunction undergoing elective PCI. Jeopardy scores confirm that LMS intervention is associated with a very large area of myocardium at risk (21–24), and the majority of the LV is likely to be hibernating/stunned as a consequence of unprotected LMS occlusion (25). In our institutional experience of ULMSO (3,4), pulmonary edema may not be present on arrival in hospital, but it occurs shortly thereafter mandating use of mechanical hemodynamic support early during the procedure. IABP use is recommended by EACTS/ESC (26) and ACCF/AHA/SCAI guidelines (Class I and Class IIa recommendations, respectively) (27). Our data supports IABP use in ULMSO, particularly in those with periprocedural cardiogenic shock. However, this device does not provide comparable LV support when compared with newer, more efficient percutaneous hemodynamic support devices including Impella (Abiomed, Inc., Danvers, Massachusetts), TandemHeart (CardiacAssist, Inc., Pittsburgh, Pennsylvania), and extracorporeal membrane oxygenation (28), which may explain the very high mortality rate despite IABP’s use and therefore these newer devices may have an important role in the future. Percutaneous techniques to allow rapid establishment of extracorporeal membrane oxygenation would appear to be beneficial, but the increase in afterload may prove to be detrimental to LV function. In principle, devices such as Impella, which increase cardiac output without increasing myocardial work (28), might be more suited for use in this condition, which has a high initial mortality but a reasonable long-term prognosis.
Third, supplementary inotropic and ventilator support was required 14- and 6-fold, respectively, more often in patients with ULMSO than in patients undergoing non-LMS PPCI. Therefore, we recommend early/immediate anesthetic support at presentation not only for help in acute ventilatory/gas exchange stabilization of the patient, but also for early advice regarding a coordinated management plan on the intensive care unit. Cardiogenic shock is common among patients undergoing ULMSO PPCI, and in this group, its associated mortality is very high (61.4% in-hospital mortality). Optimization of treatment on intensive or coronary care units for such patients is of paramount importance as death in patients with cardiogenic shock occurs not only as a consequence of pump failure, but also as a secondary consequence to multiorgan failure and systemic inflammatory response syndrome (29,30). Pharmacological inotropes can improve observed hemodynamics, but use of these agents should be carefully considered as they could increase myocardial ischemia and infarct size as well as having detrimental effects on peripheral organ perfusion, which may limit long-term survival (16).
Our study data is from a national PCI database, and although it has standardized, validated, and comprehensive data, the limitations of the retrospective analysis of a registry require acknowledgment. Additionally, this registry does not have data pertaining to those that died prior to any attempted PCI (including those labeled with sudden cardiac death in the community), or those patients treated conservatively. There is no data regarding those referred for emergency CABG without initial PCI. Therefore, a true representation of the incidence of this condition cannot be made. Accurate data are lacking regarding cardiac enzyme titers and other serological parameters such as creatinine (renal failure was defined as creatinine >200 μmol/l), and the database lacks precise anatomical or lesion-specific information including lesion site (e.g., involvement of the primary bifurcation, ostial lesions, SYNTAX [Synergy Between Percutaneous Coronary Intervention With Taxus and Cardiac Surgery] score) that may offer insight into outcomes dependent on complexity of intervention.
The strongest predictors of 3-year mortality in patients undergoing PPCI to an unprotected LMS are periprocedural cardiogenic shock, post-procedural no-reflow, and occlusive LMS disease. The in-hospital, 1-year, and 3-year mortality rates for patients presenting to the catheterization laboratory with acute MI due to ULMSO are 41.6%, 52.8%, and 73.9%, respectively. For those with periprocedural cardiogenic shock, mortality rates are 61.4%, 71.4%, and 86.1%, respectively. This registry provides a resource to inform survivors of the long-term outcomes of patients presenting to the catheterization lab with an acute MI due to ULMSO and may serve for future reference to optimize treatment strategies of this high-risk patient group.
The authors would like to thank the BCIS, the National Institute for Cardiovascular Outcomes Research, and the Office of National Statistics for collecting and sharing the data presented. Additionally, they acknowledge all the hospitals in England and Wales for their contribution of data.
Dr. Rahimi is funded by a Career Development Fellowship from the National Institute for Health Research. Dr. Banning is partially funded by the National Institute for Health Research Oxford Biomedical Research Centre; has received an unrestricted research funding grant from Boston Scientific; and has received speaker fees/honorarium from Medtronic and Abbott Vascular. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- British Cardiovascular Intervention Society
- coronary artery bypass graft
- confidence interval
- hazard ratio
- intra-aortic balloon pump
- interquartile range
- left main stem
- left ventricle
- major adverse cardiac and cerebrovascular events
- myocardial infarction
- percutaneous coronary intervention
- primary percutaneous coronary intervention
- Thrombolysis In Myocardial Infarction
- unprotected left main stem occlusion
- Received March 28, 2014.
- Accepted April 10, 2014.
- American College of Cardiology Foundation
- Fabre A.,
- Sheppard M.N.
- Vis M.M.,
- Beijk M.A.,
- Grundeken M.J.,
- et al.
- Banning A.P.
- Van Gaal W.J.,
- Jennings B.R.,
- Banning A.P.
- Aygül N.,
- Aygül M.U.,
- Özdemir K.,
- Altunkeser B.B.
- Friedenberger J.,
- Thiele H.
- Grundeken M.J.,
- Vis M.M.,
- Beijk M.A.,
- et al.
- Puricel S.,
- Adorjan P.,
- Oberhänsli M.,
- et al.
- Ludman P.F.
- Ouweneel D.M.,
- Henriques J.P.
- Montalescot G.,
- Brieger D.,
- Eagle K.A.,
- et al.,
- for the GRACE Investigators
- Tanaka A.,
- Kawarabayashi T.,
- Nishibori Y.,
- et al.
- Thiele H.,
- Zeymer U.,
- Neumann F.J.,
- et al.,
- for the IABP-SHOCK II Trial Investigators
- Perera D.,
- Stables R.,
- Clayton T.,
- et al.,
- for the BCIS-1 Investigators
- Califf R.M.,
- Phillips H.R. 3rd.,
- Hindman M.C.,
- et al.
- Alderman E.,
- Stadius M.
- Wijns W.,
- Kolh P.,
- Danchin N.,
- et al.,
- for the Task Force on Myocardial Revascularization of the European Society of Cardiology and the European Association for Cardio-Thoracic Surgery, European Association for Percutaneous Cardiovascular Interventions
- O’Gara P.T.,
- Kushner F.G.,
- Ascheim D.D.,
- et al.
- Prondzinsky R.,
- Lemm H.,
- Swyter M.,
- et al.
- Hochman J.