| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Circulation. 2003;108:2769.)
© 2003 American Heart Association, Inc.
Clinical Investigation and Reports |
From William Beaumont Hospital, Royal Oak, Mich (H.M.S., C.L.G., S.R.D.); the Cardiovascular Research Foundation, New York, NY (G.W.S., R.M., A.J.L., M.F.); Mid Carolina Cardiology, Charlotte, NC (D.A.C.); Hospital Gregorio Maranon, Madrid, Spain (E.G.); Duke Clinical Research Institute, Durham, NC (J.E.T.); Virginia Beach General Hospital, Virginia Beach, Va (J.J.G.); Moses Cone Memorial Hospital, Greensboro, NC (T.D.S.); Center for Cardiac and Vascular Research, Tacoma Park, Md (M.T.); and the University of Colorado Health Sciences Center, Denver (J.D.C.).
Correspondence to Gregg W. Stone, MD, The Cardiovascular Research Foundation, 55 E 59th St, 6th Floor, New York, NY 10022. E-mail gstone{at}crf.org
Received April 4, 2003; de novo received July 3, 2003; revision received September 11, 2003; accepted September 12, 2003.
| Abstract |
|---|
|
|
|---|
Methods and Results PCI was performed in 2082 AMI patients without shock presenting within 12 hours of symptom onset in a prospective, multicenter randomized trial. RI was defined as a calculated (Cockroft-Gault) creatinine clearance (CrCl)
60 mL/min. RI at baseline was present in 18% of patients. Compared with patients without RI, patients with RI were older and were more likely to be female; to have hypertension, peripheral vascular disease, or cerebrovascular disease; and to present in heart failure. Mortality was markedly increased in patients with versus without baseline RI both at 30 days (7.5% versus 0.8%, P<0.0001) and at 1 year (12.7% versus 2.4%, P<0.0001). Mortality rates increased incrementally for every 10-mL/min decrease in baseline CrCl. By multivariate analysis, reduced baseline CrCl was a powerful independent predictor of 30-day mortality (hazard ratio, 5.77; P<0.0001) and remained associated with reduced survival at 1 year (hazard ratio, 1.98; P=0.08). Hemorrhagic complications and transfusion requirements were also increased more than 2-fold in patients with RI, as were severe restenosis (diameter stenosis
70%; 20.6% versus 11.8%, P=0.024) and infarct artery reocclusion (14.7% versus 7.3%, P=0.02).
Conclusions Baseline RI in patients with AMI undergoing primary PCI is associated with a markedly increased risk of mortality, as well as bleeding and restenosis. Novel approaches are needed to improve the otherwise poor prognosis of patients with RI and AMI.
Key Words: angioplasty myocardial infarction kidney
| Introduction |
|---|
|
|
|---|
Management of acute myocardial infarction (AMI) in patients with RI is particularly problematic. RI has been shown to worsen the prognosis of patients with AMI treated medically.4,8,11,12 Although the existence of RI in patients undergoing percutaneous coronary intervention (PCI) in the non-AMI setting portends a poor prognosis,13 the outcomes of primary PCI in patients with AMI and RI have not been well characterized, because such patients are typically excluded from clinical trials.14,15 We therefore sought to evaluate the impact of RI in a large, contemporary randomized trial of patients undergoing primary PCI for AMI.
| Methods |
|---|
|
|
|---|
Renal Insufficiency
Baseline SCr levels were obtained before angiography in 1933 patients (93%). Estimated creatinine clearance (CrCl) was calculated by use of the Cockcroft-Gault formula: CrCl (mL/min)=[(140-age)xweight (kg)]/[SCr (mg/dL)x72], corrected in women by a factor of 0.85.16 A CrCl of
60 mL/min was chosen to define at least moderate RI.7 Outcomes as a function of RI were further assessed by examining CrCl strata of >60, 50 to 60, 40 to 49, 30 to 39, 20 to 29, and <20 mL/min. SCr levels were assessed at admission, 24 hours after PCI, and at discharge. Contrast-induced nephropathy (CIN) was defined as an SCr increase by >0.5 mg/dL within the index hospitalization (5.7±3.5 days).
Statistical Analysis
Continuous variables are presented as medians and interquartile ranges and were compared by the Kruskal-Wallis nonparametric test. Categorical data were compared by the likelihood-ratio
2 test or Fishers exact test. Outcomes are summarized as Kaplan-Meier survival and compared by the log-rank test. Independent predictors of mortality were identified with Cox proportional hazards regression. Clinical, angiographic, treatment assignment, and discharge medication variables were selected by use of stepwise selection, entering all with a significant or borderline univariate association with mortality (P
0.10). Although age and sex are used to calculate the CrCl in the Cockcroft-Gault formula,16 they were not excluded from the multivariate analysis of mortality. A probability value of P<0.05 was defined as significant.
| Results |
|---|
|
|
|---|
60 mL/min was present in 350 patients (18%). Baseline characteristics of patients with and without RI are shown in Table 1. Patients with RI were older, more often female, less likely to smoke, and more likely to have hypertension and a history of peripheral vascular and cerebrovascular disease. Patients with RI also presented later after symptom onset and were more likely to have congestive heart failure on presentation and to have a lower baseline left ventricular ejection fraction.
|
Angiographic Results
Patients with RI were more likely to have triple-vessel coronary artery disease and slightly smaller infarct vessels than those without RI. Involvement of the left anterior descending coronary artery was similar in both groups, however. Restoration of TIMI-3 flow and final diameter stenosis achieved were also independent of baseline renal function. The median amount of contrast administered was 280 mL in patients with RI versus 300 mL in those with normal renal function (P=0.049).
Clinical Outcomes
Procedural success was reduced in patients with versus without RI (87.2% versus 92.0%, P=0.01). Medication use at baseline and at discharge appears in Table 2. Patients with RI were less likely to be treated with ß-blockers and statins at discharge, although they were more likely to receive ACE inhibitors, angiotensin receptor blockers, or calcium channel antagonists. The 30-day and 1-year mortality rates were markedly elevated in patients with RI. As seen in Table 3, patients with a reduced CrCl had a >9-fold increase in mortality at 30 days and a 5-fold increase in mortality at 1 year. Mortality increased incrementally for every 10-mL/min decline in baseline renal function (Figure 1). Both cardiac and noncardiac causes of death were more prevalent in patients with RI (Table 3). Most deaths were unrelated to dialysis. At 30 days and 1 year, 4 and 5 patients (0.19% and 0.24%) required dialysis, respectively, 3 of whom died. CIN developed in 4.6% of patients, being 3 times more prevalent in patients with baseline RI (Table 3). Early survival and late survival were markedly diminished in patients who developed CIN (Table 4).
|
|
|
|
By multivariate analysis, reduced baseline CrCl was a powerful independent predictor of 30-day mortality (Figure 2). At 1 year, RI remained associated with reduced 1-year survival. No interaction effect was present between either stents or abciximab and RI on mortality. The 30-day and 1-year MACE rates were also significantly increased in patients with RI compared with those without, principally because of the reduction in survival (Table 3). Patients with RI had notably higher rates of major hemorrhage, thrombocytopenia, and need for blood product transfusions (independent of abciximab). Among patients with RI, increased need for transfusion was independent of abciximab treatment. Although there were no significant differences in the rates of subacute thrombosis, ischemic TVR, or binary restenosis (diameter stenosis
50% at follow-up) in patients with versus without RI, severe restenosis (diameter stenosis
70% at follow-up) and infarct artery reocclusion were increased in patients with RI (Figure 3). Although increased rates of severe restenosis and reocclusion were independent of abciximab treatment in patients with and without RI, severe restenosis was lower with stenting versus PTCA in patients with (26.5% versus 15.1%, P=0.15) and without (17.6% versus 6.1%, P<0.0001) RI. Infarct artery reocclusion was reduced with stenting in patients without RI (10.9% versus 3.7%, P=0.002) but was not affected by stenting in patients with RI (14.3% versus 15.1%, P=0.91).
|
|
| Discussion |
|---|
|
|
|---|
The marked negative impact of chronic RI on the prognosis of patients with cardiovascular disease has been underappreciated. In heart failure, RI has been found to be a stronger predictor of late mortality than reduced ejection fraction.5 In a large retrospective analysis of patients undergoing elective PCI, RI was found to have a negative prognostic impact similar to that of diabetes mellitus on cardiovascular morbidity and mortality.17 Mild RI may portend a risk of adverse outcomes and mortality similar to that of more severe renal impairment in patients undergoing PCI.13 The prognosis of patients with AMI treated medically is significantly diminished in patients with RI.4,11,12 On the basis of the results of the present study, patients with AMI undergoing primary PCI may now be added to the list of conditions for which baseline RI signifies a markedly worse prognosis.
The mechanisms by which chronic RI worsens late outcomes in patients with AMI undergoing primary PCI are multifactorial. We found that patients with RI were more often elderly and female, presented later, and were more likely to have a reduced left ventricular ejection fraction, all features that have been associated with increased mortality after primary PCI for AMI.14,18,19 Patients with RI presented with more diffuse atherosclerosis, as evidenced by more frequent peripheral vascular and cerebrovascular disease and triple-vessel coronary artery disease. Patients with RI were also less likely to be discharged on ß-blockers, which may have contributed to reduced survival. Undertreatment with statins and overtreatment with calcium channel blockers may also affect the late prognosis in patients with RI. Nonetheless, the impact of baseline RI on mortality was independent of age, sex, medication use, and other covariates when evaluated in a multivariable model. Unique metabolic abnormalities of chronic RI, including insulin resistance, dyslipidemia, homocystinemia, hyperuricemia, and increased atherosclerotic, thrombotic, and oxidative stress,2026 may contribute to the independent excess cardiovascular risk in these patients. In addition, the procedural success rate was lower in patients with RI owing to a higher rate of periprocedural complications, which may have contributed to their worse long-term prognosis.
Radiocontrast toxicity may also contribute to clinical deterioration after primary PCI in AMI. Depending on the definition used, CIN occurs in
1% to 15% of a general PCI population and in 20% to 40% of patients with preexisting RI.2729 The relatively low incidence of CIN in the present trial (4.6%) is probably related to the entry exclusion of patients with known severe RI, although CIN still developed in 9.7% of patients with mild to moderate baseline RI. The lack of routine daily creatinine measurements may also have led to an underestimation of the true incidence of CIN. Nonetheless, although dialysis was infrequently required in the present study, CIN was associated with a strikingly worse prognosis after PCI, similar to previous reports in elective PCI.2830 The 3-fold relative increase in CIN thus most likely contributed to the poor outcomes in patients with baseline RI.
Angiographic restenosis rates have been shown to be dramatically higher in patients with end-stage renal disease undergoing balloon angioplasty, although possibly not stenting.3134 Although TVR rates were not affected by renal function in our study, significantly higher rates of severe restenosis and infarct artery reocclusion were present in patients with RI. Although abciximab treatment had no impact on restenosis or reocclusion in patients with RI, restenosis was reduced with stenting. Stenting did not reduce reocclusion in patients with RI, a finding that may have been masked by small numbers. A higher incidence of silent ischemia in patients with RI may explain the observed discrepancy between increased severe angiographic restenosis and lower symptom-driven TVR.35,36 In addition, the threshold for performing repeat angiography may be higher in patients with RI, and undetected disease progression may also have contributed to the increase in late mortality in these patients.17
Clinical Implications
Baseline RI is independently associated with strikingly reduced survival after mechanical reperfusion therapy for AMI. The impact of RI in this regard is similar to the risk of anterior infarction in the short term and also portends increased rates of major hemorrhagic complications, need for blood transfusion, and severe restenosis and infarct artery reocclusion that may not be clinically apparent. The high-risk nature of patients with even moderate RI must be appreciated and appropriate therapies initiated. Moreover, although ß-adrenergic blockers and ACE inhibitors improve survival in patients after AMI,37 these agents are often underused in patients with RI compared with those with normal renal function.4,38,39 All attempts must be made to prevent contrast nephropathy, including adequate hydration,40 minimizing contrast use,29 use of low-osmolar contrast,41 and possibly administration of N-acetylcysteine.42 Patients with baseline RI warrant close surveillance and intensive medical management, including tight control of diabetes, hypertension, and dyslipidemia20; dietary modification; and potentially frequent stress testing for early recognition of disease progression. Whether the incidence or prognostic implications of RI after fibrinolytic therapy are different from after primary PCI also deserves further study. Finally, novel approaches are required for patients with RI to favorably affect their otherwise poor prognosis.
| Acknowledgments |
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
C. R. Parikh, S. G. Coca, Y. Wang, F. A. Masoudi, and H. M. Krumholz Long-term Prognosis of Acute Kidney Injury After Acute Myocardial Infarction Arch Intern Med, May 12, 2008; 168(9): 987 - 995. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. A. McCullough Contrast-Induced Acute Kidney Injury J. Am. Coll. Cardiol., April 15, 2008; 51(15): 1419 - 1428. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. K. Inrig, U. D. Patel, L. P. Briley, L. She, B. S. Gillespie, J. D. Easton, E. J. Topol, and L. A. Szczech Mortality, kidney disease and cardiac procedures following acute coronary syndrome Nephrol. Dial. Transplant., March 1, 2008; 23(3): 934 - 940. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. Charytan, J. P. Forman, and D. E. Cutlip Risk of target lesion revascularization after coronary stenting in patients with and without chronic kidney disease Nephrol. Dial. Transplant., September 1, 2007; 22(9): 2578 - 2585. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. Jimenez-Quevedo, M. Sabate, D. J. Angiolillo, F. Alfonso, R. Hernandez-Antolin, M. SanMartin, J. A. Gomez-Hospital, C. Banuelos, J. Escaned, R. Moreno, et al. Long-term clinical benefit of sirolimus-eluting stent implantation in diabetic patients with de novo coronary stenoses: long-term results of the DIABETES trial Eur. Heart J., August 2, 2007; 28(16): 1946 - 1952. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. C. Keeley and L. D. Hillis Primary PCI for Myocardial Infarction with ST-Segment Elevation N. Engl. J. Med., January 4, 2007; 356(1): 47 - 54. [Full Text] [PDF] |
||||
![]() |
S. G. Coca, H. M. Krumholz, A. X. Garg, and C. R. Parikh Underrepresentation of renal disease in randomized controlled trials of cardiovascular disease. JAMA, September 20, 2006; 296(11): 1377 - 1384. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. Marenzi, E. Assanelli, I. Marana, G. Lauri, J. Campodonico, M. Grazi, M. De Metrio, S. Galli, F. Fabbiocchi, P. Montorsi, et al. N-acetylcysteine and contrast-induced nephropathy in primary angioplasty. N. Engl. J. Med., June 29, 2006; 354(26): 2773 - 2782. [Abstract] [Full Text] [PDF] |
||||
![]() |
I. Michishita and Z. Fujii A Novel Contrast Removal System From the Coronary Sinus Using an Adsorbing Column During Coronary Angiography in a Porcine Model J. Am. Coll. Cardiol., May 2, 2006; 47(9): 1866 - 1870. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. de Silva, N. P. Nikitin, K. K.A. Witte, A. S. Rigby, K. Goode, S. Bhandari, A. L. Clark, and J. G.F. Cleland Incidence of renal dysfunction over 6 months in patients with chronic heart failure due to left ventricular systolic dysfunction: contributing factors and relationship to prognosis Eur. Heart J., March 1, 2006; 27(5): 569 - 581. [Abstract] [Full Text] [PDF] |
||||
![]() |
A S. Mathis and J. J Gugger Percutaneous Coronary Intervention-Related Bleeding Risk Factors in Current Practice Ann. Pharmacother., October 1, 2005; 39(10): 1627 - 1632. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Vanholder, Z. Massy, A. Argiles, G. Spasovski, F. Verbeke, N. Lameire, and for the European Uremic Toxin Work Group (EUTox) Chronic kidney disease as cause of cardiovascular morbidity and mortality Nephrol. Dial. Transplant., June 1, 2005; 20(6): 1048 - 1056. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. W. Katzberg Contrast Medium-induced Nephrotoxicity: Which Pathway? Radiology, June 1, 2005; 235(3): 752 - 755. [Full Text] [PDF] |
||||
![]() |
A. Halkin, M. Singh, E. Nikolsky, C. L. Grines, J. E. Tcheng, E. Garcia, D. A. Cox, M. Turco, T. D. Stuckey, Y. Na, et al. Prediction of Mortality After Primary Percutaneous Coronary Intervention for Acute Myocardial Infarction: The CADILLAC Risk Score J. Am. Coll. Cardiol., May 3, 2005; 45(9): 1397 - 1405. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. Villeneuve and R. Sunderji Glycoprotein IIb/IIIa Inhibitors in Patients with End-Stage Renal Disease Ann. Pharmacother., April 1, 2005; 39(4): 732 - 735. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. Koenig, D. Twardella, H. Brenner, and D. Rothenbacher Plasma Concentrations of Cystatin C in Patients with Coronary Heart Disease and Risk for Secondary Cardiovascular Events: More than Simply a Marker of Glomerular Filtration Rate Clin. Chem., February 1, 2005; 51(2): 321 - 327. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. Knobler, T. Zornitzki, S. Vered, M. Oettinger, R. Levy, A. Caspi, D. Faraggi, and S. Livschitz Reduced glomerular filtration rate in asymptomatic diabetic patients: Predictor of increased risk for cardiac events independent of albuminuria J. Am. Coll. Cardiol., December 7, 2004; 44(11): 2142 - 2148. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. M. Gibson, R. L. Dumaine, E. V. Gelfand, S. A. Murphy, D. A. Morrow, S. D. Wiviott, R. P. Giugliano, C. P. Cannon, E. M. Antman, E. Braunwald, et al. Association of glomerular filtration rate on presentation with subsequent mortality in non-ST-segment elevation acute coronary syndrome; observations in 13307 patients in five TIMI trials Eur. Heart J., November 2, 2004; 25(22): 1998 - 2005. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. Marenzi, G. Lauri, E. Assanelli, J. Campodonico, M. De Metrio, I. Marana, M. Grazi, F. Veglia, and A. L. Bartorelli Contrast-induced nephropathy in patients undergoing primary angioplasty for acute myocardial infarction J. Am. Coll. Cardiol., November 2, 2004; 44(9): 1780 - 1785. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Gupta, Y. Birnbaum, and B. F. Uretsky The renal patient with coronary artery disease: Current concepts and dilemmas J. Am. Coll. Cardiol., October 6, 2004; 44(7): 1343 - 1353. [Abstract] [Full Text] [PDF] |
||||
![]() |
J.-H. Choi, K. L. Kim, W. Huh, B. Kim, J. Byun, W. Suh, J. Sung, E.-S. Jeon, H.-Y. Oh, and D.-K. Kim Decreased Number and Impaired Angiogenic Function of Endothelial Progenitor Cells in Patients With Chronic Renal Failure Arterioscler. Thromb. Vasc. Biol., July 1, 2004; 24(7): 1246 - 1252. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||