Author + information
- Received May 7, 2008
- Revision received July 23, 2008
- Accepted September 12, 2008
- Published online January 1, 2009.
- Caroline Medi, MBBS, FRACP⁎,
- Gilles Montalescot, MD, PhD†,
- Andrzej Budaj, MD, PhD‡,
- Keith A.A. Fox, MB, ChB, FRCP§,
- José López-Sendón, MD, FACC∥,
- Gordon FitzGerald, PhD¶,
- David B. Brieger, MBBS, PhD, FRACP, FACC⁎,⁎ (, )
- GRACE Investigators
- ↵⁎Reprint requests and correspondence:
Dr. David B. Brieger, Concord Repatriation General Hospital, Coronary Care Unit, Level 3, Multi Building, Hospital Road, Concord, NSW Australia 2139
Objectives We investigated the relative benefit of reperfusion strategies in renal dysfunction and ST-segment elevation/left bundle branch block (STE/LBBB).
Background Few data are available informing the treatment of STE myocardial infarction in the presence of renal dysfunction.
Methods Patients (N = 12,532) from the GRACE (Global Registry of Acute Coronary Events) presenting with STE/LBBB were stratified by renal function and receipt of fibrinolysis, primary percutaneous coronary intervention (PCI), or neither.
Results As renal function declined, hospital mortality increased and reperfusion decreased (both p < 0.001). Compared with no reperfusion, primary PCI was associated with lower hospital mortality in patients with normal renal function (1.9% vs. 3.7%, p = 0.001, adjusted) but no reduction in those with renal dysfunction (14% vs. 15% for glomerular filtration rate [GFR] 30 to 59 ml/min/1.73 m2; 29% vs. 32% for GFR <30 ml/min/1.73 m2). Fibrinolysis was not associated with lower hospital mortality for normal (3.1% vs. 3.7%, p = NS) or low renal function (32% vs. 32%, p = NS) and with higher mortality with moderate renal dysfunction (adjusted odds ratio: 1.35, 95% confidence interval: 1.01 to 1.80). Primary PCI was associated with increased hospital bleeding and fibrinolysis with increased stroke in all patients. Among hospital survivors, primary PCI, but not fibrinolysis, was associated with lower mortality for moderate dysfunction. Both reperfusion strategies were associated with higher mortality for severe dysfunction.
Conclusions In STE/LBBB and renal dysfunction, mortality rates are high and reperfusion rates are lower. In moderate renal dysfunction, primary PCI is associated with mortality reduction at 6 months. Outcomes remain poor with severe renal dysfunction, despite receipt of reperfusion therapy.
- percutaneous coronary intervention
- renal dysfunction
- left bundle branch block
- ST-segment elevation
Renal dysfunction is associated with a significant increase in cardiovascular risk (1-6), and patients with a history of renal dysfunction who develop a myocardial infarction have a poor prognosis (2,7-9). Observational analyses have demonstrated improved survival in patients with renal dysfunction after revascularization in stable coronary disease (10); however, this group consistently manifests lower survival rates relative to those with intact kidney function (11,12). These patients constitute a disproportionately high proportion of individuals who die after myocardial infarction. Because patients with renal dysfunction are usually excluded from randomized trials of reperfusion therapies (13), there is a lack of both prospective evidence and guideline-based recommendations on the treatment of ST-segment elevation myocardial infarction (STEMI) in the presence of renal dysfunction.
The risks associated with reperfusion therapy are exaggerated in patients with renal dysfunction. After percutaneous coronary intervention (PCI), renal dysfunction has been associated with a reduction in procedural success and an increase in hospital and long-term adverse outcomes (14). In addition, patients with renal dysfunction might be relatively resistant to the beneficial effects of fibrinolysis and the risk of bleeding might be increased (15).
The impact of these unfavorable outcomes relative to the benefit of emergently re-establishing flow in patients with acute ST-segment elevation (STE)/left bundle branch block (LBBB) has not been well established. The aim of this study was to use data from the GRACE (Global Registry of Acute Coronary Events) to describe patterns of reperfusion as well as outcomes, after receipt or non-receipt of reperfusion therapy among a large cohort of patients with STE/LBBB and varying degrees of renal dysfunction.
The rationale and methods of the GRACE study have been previously published (16,17). A total of 121 hospitals located in 14 countries contributed data to this study from April 1999 to June 2007.
Data were collected by trained coordinators with a standardized case report form. Demographic characteristics, medical history, presenting symptoms, biochemical and electrocardiographic findings, treatment practices, and a variety of hospital outcome data were collected. In-hospital management of patients was left to the discretion of the investigating physicians. Standardized definitions for all patient-related variables and clinical diagnoses were used (16). Where required, study investigators received approval from their local hospital ethics or institutional review board for the conduct of this study. This analysis was restricted to patients presenting with STEMI/LBBB, defined by the presence of new STE >0.1 mV in 2 contiguous leads or new LBBB seen on the index or qualifying electrocardiogram. Patients with STEMI/LBBB who were transferred from hospitals not involved in the GRACE study were excluded. Patients with contraindications to either fibrinolytic therapy or PCI were also excluded.
Patients were stratified into groups according to glomerular filtration rate (GFR), calculated by the Modification of Diet in Renal Disease method (18). Groups consisted of normal kidney function (GFR ≥60 ml/min/1.73 m2), moderate kidney function (GFR 30 to 59 ml/min/1.73 m2), and severe renal dysfunction (GFR <30 ml/min/1.73 m2). Each GFR group was analyzed according to hospital treatment received of reperfusion (either primary PCI or fibrinolysis) or no reperfusion, excluding patients who received both primary PCI and fibrinolysis during their hospital stay. Outcomes included mortality (both in hospital and at 6 months after discharge) and hospital major adverse events of cardiac failure/acute pulmonary edema, cardiac arrest/ventricular fibrillation, recurrent ischemia, recurrent myocardial infarction, stroke, and major bleeding (Online Appendix).
Dichotomous patient outcomes were compared across GFR groups with the Mantel-Haenszel linear trend test and across reperfusion groups with the chi-square test. For continuous variables, trend across GFR groups was assessed by linear regression.
Adjusted outcome odds ratios (ORs) were generated with multiple logistic regression. Hospital and 6-month mortality were adjusted for GRACE risk score (variables were age, Killip class, systolic blood pressure, ST-segment deviation, cardiac arrest on presentation, serum creatinine, positive initial cardiac markers, and pulse) (19). Six-month mortality (from hospital discharge to 6 months) was also adjusted for age, systolic blood pressure, creatinine concentration, positive initial markers, pulse, history of myocardial infarction, history of coronary heart failure, and ST-segment depression (20).
Renal dysfunction was seen in 3,450 (28%) of the 12,532 patients presenting with STEMI/LBBB, comprising 2,982 (24%) with moderate kidney function and 468 (3.7%) with severe renal dysfunction. Patients with renal dysfunction were sicker than those without renal disease (Table 1). They were older and had higher rates of prior myocardial infarction and congestive heart failure and a greater prevalence of cardiovascular risk factors. They were also more compromised on presentation, with a lower systolic blood pressure and higher Killip class.
Patients with renal dysfunction experienced worse outcomes (Fig. 1). Hospital mortality rates increased significantly as GFR decreased (p < 0.001), exceeding 30% in patients with severe renal dysfunction. Approximately one-third of patients with renal dysfunction experienced pulmonary edema or heart failure after infarction, 3 times the rate in patients with normal kidney function. The rate of cardiac arrest or ventricular fibrillation exceeded 20% in patients at the lowest stratum of kidney function. Patients with renal dysfunction demonstrated a significantly higher rate of major bleeding. Six-month mortality was notably higher across all treatment groups as GFR declined (p < 0.001).
Reperfusion and renal dysfunction
Despite their higher risk characteristics and poorer outcomes, patients with renal dysfunction were less likely to receive reperfusion therapy. Reperfusion was received by 62% of patients with normal renal function, by 48% with moderate function, and by 31% with severe dysfunction. Within each stratum of renal function, reperfusion therapy was typically offered to lower-risk patients versus those who did not receive reperfusion (Table 2). Patients who received fibrinolysis and proceeded to rescue angioplasty constituted 5% of the STEMI/LBBB population and did not differ significantly from the overall cohort. Impaired renal function was not associated with a greater likelihood of receiving rescue PCI. The proportion of patients in each GFR group comprising this population together with their median GRACE risk score and hospital death rates are presented in Table 3.
Renal dysfunction and impact on hospital mortality
In patients with normal renal function, primary PCI was associated with lower mortality rates than either fibrinolysis or no reperfusion (Table 4). Among patients with renal dysfunction, primary PCI was associated with numerically lower mortality rates than in patients not receiving reperfusion. Numbers with renal dysfunction were relatively small (especially severe renal dysfunction), and the differences were not statistically significant. Fibrinolysis was not associated with lower mortality than nonreperfusion among patients with normal renal function or severe renal dysfunction but, after adjustment, was associated with higher mortality among patients with moderate renal dysfunction.
Stroke rates and major bleeding
Fibrinolysis, but not primary PCI, was associated with an increased rate of stroke in all patients, significant for those with normal renal function (Table 5). Major bleeding occurred more frequently among all patients receiving reperfusion therapy. Differences in bleeding rates were statistically significant for patients with normal and moderate renal dysfunction undergoing PCI and for patients with normal renal function receiving fibrinolysis (Table 5).
Among patients surviving to hospital discharge, receipt of reperfusion was associated with lower mortality for the overall STE/LBBB population. After stratifying for renal function, this benefit was seen among patients with normal renal function receiving fibrinolysis (adjusted OR: 0.65, 95% confidence interval: 0.43 to 1.00) and in those with moderate dysfunction receiving primary PCI (adjusted OR: 0.41, 95% confidence interval: 0.22 to 0.75). Mortality rates remained high for patients with severe renal dysfunction regardless of receipt of reperfusion, and reperfusion was associated with worse outcome. Comparison of adjusted ORs for mortality after primary PCI and fibrinolysis relative to no reperfusion for each GFR group is shown in Table 4.
This large multinational, multicenter analysis of over 12,000 patients presenting with STE/LBBB shows that patients with moderate renal dysfunction have an approximately 5-fold higher early mortality and that those with severe renal dysfunction have a mortality risk of approximately 10-fold that of patients with normal renal function. These data are consistent with other observational studies in patients with renal dysfunction (4,12,21,22). In our analysis, we found that reperfusion was offered to patients less frequently as renal function deteriorated. This risk-averse behavior is consistent with that described among physicians selecting patients with an acute coronary syndrome (ACS) for coronary angiography (23). The reasons for this are not clear and might reflect both a judgment that “active” treatment with reperfusion has an unfavorable risk/benefit profile in the context of a high prevalence of comorbidities; an increase in the proportion of patients with renal dysfunction who have a delayed or an atypical (24) presentation; a higher proportion who are ineligible for reperfusion therapy due to contraindications (24); and the absence of randomized controlled evidence in this population to support an aggressive approach.
In our nonrandomized cohort, reperfusion with fibrinolytic therapy was not associated with a mortality reduction in our population with renal impairment. Indeed, after adjustment, hospital mortality was greater for patients with moderate renal dysfunction receiving fibrinolysis. This was offset by a trend toward reduced mortality in this population at 6 months. Nonetheless, the lack of appreciable benefit of fibrinolysis in this cohort is an unsettling observation, because it is readily available in most clinical settings and is the most common form of reperfusion therapy. As expected, there was a numerically greater incidence of stroke among patients receiving fibrinolytic therapy regardless of renal function.
Consistent with randomized trial evidence, primary PCI was associated with the lowest hospital mortality rates among patients with normal renal function. There was no significant reduction in mortality after primary PCI observed among patients with renal dysfunction, although this might reflect a type II error because of the smaller number of patients with renal dysfunction. The absolute reduction of approximately 2% in mortality among renal dysfunction patients receiving primary PCI relative to those not offered reperfusion was comparable to that observed among patients with normal renal function.
As expected, any potential clinical benefit of primary PCI in patients with renal dysfunction was accompanied by an excess in complications. Percutaneous angiography or angioplasty carries an increased risk in the average patient with renal dysfunction, who is already at greater risk of stroke due to a high incidence of cerebrovascular disease and more advanced age, higher procedural risk due to frequent left ventricular dysfunction, and increased risk of major bleeding (25).
An increase in the likelihood of stroke after primary PCI was not observed in our cohort, but we did observe a significant increase in major bleeding. Recent studies highlight the prognostic importance of major bleeding in patients with ACS (26), and it is conceivable that bleeding events might have attenuated the mortality benefit that would otherwise have accompanied the rapid establishment of coronary artery patency in our cohort. For primary PCI to become a clearly superior treatment option in patients with renal dysfunction, strategies to reduce bleeding are required. This includes using a radial rather than femoral artery approach for coronary intervention to reduce the risk of local vascular complications (27), use of appropriately dose-adjusted anticoagulant therapies such as low-molecular-weight heparin (28,29), and the use of a direct thrombin inhibitor such as bivalirudin instead of intravenous unfractionated heparin during PCI (30,31).
Primary PCI was associated with lower mortality among moderate renal dysfunction patients at 6 months. This tallies with an earlier observation on revascularization in patients with coronary heart failure, where no in-hospital benefit was seen but a clear long-term benefit was evident for survivors to discharge (32). This also supports previous observations of reduced 2- and 3-year mortality rates (10,33) in patients with renal dysfunction and stable ischemic heart disease after revascularization.
We did not find a positive association between reperfusion and 6-month outcomes among patients with severe renal dysfunction. Indeed, a significant association with harm was identified. This should be interpreted with caution, given the small size of this cohort relative to those with greater retention of renal function. However, the striking mortality rate in this cohort was an important observation, and our data suggest that, in patients with severe renal dysfunction, coexisting serious morbidities might overwhelm the intermediate-term prognostic benefit of emergency reperfusion after acute myocardial infarction
Strengths and study limitations
The GRACE registry is a multicenter prospective study including unselected patients with ACS; thus it provides a wealth of information on treatments and outcomes in clinical practice. Whereas registry studies are subject to potential biases, they are valuable for examining real-world practice patterns, particularly among populations not well-represented in randomized clinical trials. In this analysis, patients with low GFR are rare (3.7% of STE/LBBB patients), with only 31% receiving reperfusion. However, an unselected cohort of 468 patients with low GFR represents a large patient sample relative to other studies. Nonetheless, our ability to generalize about the benefit or harm of reperfusion for these patients is limited. Although important information on treatment and outcomes of patients with renal dysfunction is available from GRACE, there are likely to be unrecognized biases, both in patient selection for reperfusion and the method of reperfusion. We have not reported the incidence of new-onset acute renal impairment requiring dialysis, because this information was not available from the registry. Additionally, the data were not controlled for prognostically important medications known to be underused in patients with renal dysfunction.
Renal dysfunction is common in patients presenting with STE/LBBB and is associated with a lower likelihood of emergency revascularization and significantly higher mortality and morbidity. After adjustment, primary PCI was associated with a lower long-term mortality in patients with moderate renal dysfunction, tempered by an increase in major hospital bleeding. Fibrinolysis was associated with increased hospital mortality in patients with moderate renal dysfunction but no excess in long-term mortality. Fibrinolysis was also associated with an increase in stroke in all GFR strata. Long-term mortality rates remain high in all patients with severe renal dysfunction. Our data suggest that randomized trials prospectively evaluating the relative benefits of method of reperfusion should actively include patients with renal dysfunction. In the meantime, efforts should be directed toward reducing major bleeding complications in renally impaired patients undergoing primary PCI.
The authors thank the physicians and nurses participating in GRACE and Sophie Rushton-Smith, PhD, for providing editorial assistance.
For a list of standardized definitions for GRACE, please see the online version of this article.
The Global Registry of Acute Coronary Events is supported by an unrestricted educational grant from Sanofi-Aventis (Paris, France) to the Center for Outcomes Research, University of Massachusetts Medical School, Worcester, Massachusetts. Dr. Montalescot has received support from Sanofi-Aventis. Dr. Budaj has received research grants and honoraria from Sanofi-Aventis, GlaxoSmithKline, AstraZeneca, and Boehringer Ingelheim and has been a consultant for and/or on the advisory board of Sanofi-Aventis and GlaxoSmithKline. Keith Fox has received research grants from Biosite, Bristol-Myers Squibb, Blue Cross Blue Shield of Michigan, Hewlett Foundation, Mardigian Fund, Pfizer, Sanofi-Aventis, and Varbedian Fund and has served as consultant to and/or on the advisory board of the National Institutes of Health, National Heart, Lung, and Blood Institute, Pfizer, Sanofi-Aventis, and the Robert Wood Johnson Foundation. Dr. López-Sendón has received modest research grants from Sanofi-Aventis, Pfizer, TIMI group, and Bristol-Myers Squibb and has received modest speaker honoraria from Sanofi, Bristol-Myers Squibb, Medtronic, Guidant, and Pfizer. Dr. Brieger has received support from the National Heart Foundation of Australia, Sanofi-Aventis, Eli Lilly, Astra Zeneca, and Schering Plough.
- Abbreviations and Acronyms
- acute coronary syndrome
- glomerular filtration rate
- left bundle branch block
- odds ratio
- percutaneous coronary intervention
- ST-segment elevation
- ST-segment elevation myocardial infarction
- Received May 7, 2008.
- Revision received July 23, 2008.
- Accepted September 12, 2008.
- American College of Cardiology Foundation
- Al Suwaidi J.,
- Reddan D.N.,
- Williams K.,
- et al.
- Meisinger C.,
- Doring A.,
- Lowel H.
- Hemmelgarn B.R.,
- Southern D.,
- Culleton B.F.,
- Mitchell L.B.,
- Knudtson M.L.,
- Ghali W.A.
- Szczech L.A.,
- Best P.J.,
- Crowley E.,
- et al.
- Rubenstein M.H.,
- Harrell L.C.,
- Sheynberg B.V.,
- Schunkert H.,
- Bazari H.,
- Palacios I.F.
- Eagle K.A.,
- Lim M.J.,
- Budaj A.,
- et al.
- McCullough P.A.,
- Soman S.S.,
- Shah S.S.,
- et al.
- Fox K.A.A.,
- Anderson F.A.J.,
- Dabbous O.H.,
- et al.
- Herzog C.A.,
- Littrell K.,
- Arko C.,
- Frederick P.D.,
- Blaney M.
- Spencer F.A.,
- Moscucci M.,
- Granger C.B.,
- et al.
- Agostoni P.,
- Biondi-Zoccai G.G.L.,
- Benedictis M.L.D.,
- et al.
- Manoukian S.V.,
- Feit F.,
- Mehran R.,
- et al.