- Email: [email protected]
Relation of Contrast-Induced Nephropathy to Long-Term Mortality After Percutaneous Coronary Intervention
Relation of Contrast-Induced Nephropathy to Long-Term Mortality After Percutaneous Coronary Intervention Mitsuru Abe, MDa,*, Takeshi Morimoto, MD, MPHb, Masaharu Akao, MDa, Yutaka Furukawa, MDc, Yoshihisa Nakagawa, MDd, Satoshi Shizuta, MDe, Natsuhiko Ehara, MDc, Ryoji Taniguchi, MDf, Takahiro Doi, MDe, Kei Nishiyama, MDg, Neiko Ozasa, MDe, Naritatsu Saito, MDe, Kozo Hoshino, MDh, Hirokazu Mitsuoka, MDi, Masanao Toma, MDf, Toshihiro Tamura, MDd, Yoshisumi Haruna, MDj, Toru Kita, MDc, and Takeshi Kimura, MDe There is little information on the effect of contrast-induced nephropathy (CIN) on longterm mortality after percutaneous coronary intervention in patients with or without chronic kidney disease (CKD). Of 4,371 patients who had paired serum creatinine (SCr) measurements before and after percutaneous coronary intervention and were discharged alive in the Coronary REvascularization Demonstrating Outcome Study in Kyoto registry, the incidence of CIN (an increase in SCr of ‡0.5 mg/dl from the baseline) was 5% in our study cohort. The rate of CIN in patients with CKD was 11%, although it was 2% without CKD (p <0.0001). During a median follow-up of 42.3 months after discharge, 374 patients (8.6%) died. After adjustment for prespecified confounders, CIN was significantly correlated with long-term mortality in the entire cohort (hazard ratio [HR] 2.26, 95% confidence interval [CI] 1.62 to 2.29, p <0.0001) and in patients with CKD (HR 2.62, 95% CI 1.91 to 3.57, p <0.0001) but not in patients without CKD (HR 1.23, 95% CI 0.47 to 2.62, p [ 0.6). Sensitivity analyses confirmed these results using the criteria defined as elevations of the SCr by ‡25% and 0.3 mg/dl from the baseline, respectively. In conclusion, CIN was significantly correlated with long-term mortality in patients with CKD but not in those without CKD. Ó 2014 Elsevier Inc. All rights reserved. (Am J Cardiol 2014;-:-e-) Recent studies have shown that contrast-induced nephropathy (CIN) is the third leading cause of all hospital-acquired renal insufficiency, accounting for 10%.1,2 The incidence of CIN in patients with chronic kidney disease (CKD) was reported to be greater compared with that in those without CKD.3e6 Although an association between CIN and death was previously suggested,4,5,7,8 it has not been fully evaluated whether the effect of CIN on long-term mortality is different between patients with CKD and those without CKD. In this report,
a Division of Cardiology, National Hospital Organization Kyoto Medical Center, Kyoto, Japan; bDivision of General Internal Medicine, Hyogo College of Medicine, Nishinomiya, Japan; cDepartment of Cardiovascular Medicine, Kobe City Medical Center General Hospital, Kobe, Japan; d Division of Cardiology, Tenri Hospital, Tenri, Japan; eDepartment of Cardiovascular Medicine, Graduate School of Medicine, Kyoto University, Kyoto, Japan; fDivision of Cardiology, Hyogo Prefectural Amagasaki Hospital, Amagasaki, Japan; gDepartment of Primary Care and Emergency Medicine, Kyoto University Hospital, Kyoto, Japan; hDivision of Cardiology, Nagai Hospital, Tsu, Japan; iDivision of Cardiology, Nara Hospital, Kinki University Faculty of Medicine, Ikoma, Japan; and jDivision of Cardiology, Hirakata Kohsai Hospital, Hirakata, Japan. Manuscript received February 24, 2014; revised manuscript received and accepted May 6, 2014. This work was supported by an educational grant from the Research Institute for Production Development, Kyoto, Japan. See page 6 for disclosure information. *Corresponding author: Tel: (þ81) 75-641-9161; fax: (þ81) 75-6434325. E-mail address: [email protected] (M. Abe).
0002-9149/14/$ - see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.amjcard.2014.05.009
we sought to clarify the incidence of CIN after percutaneous coronary intervention (PCI) and the effect of CIN on long-term mortality in patients with or without CKD using a large observational database of patients who underwent their first PCI in Japan. Methods The Coronary REvascularization Demonstrating Outcome Study in Kyoto (CREDO-Kyoto) is a multicenter registry in Japan that enrolled consecutive patients who underwent their first PCI or coronary artery bypass grafting from January 2000 to December 2002, excluding patients with acute myocardial infarction within a week before the index procedure. The relevant review board or ethics committees in all 30 participating centers approved the research protocol. The study design and patient enrollment of the CREDO-Kyoto registry were previously described in detail.9 Of the 9,877 patients who underwent PCI or coronary artery bypass grafting enrolled during the study period, we excluded 2,999 patients receiving coronary artery bypass grafting, 250 patients on hemodialysis, and 2,233 patients without paired serum creatinine (SCr) measurements before and after PCI. After excluding 24 patients who died during the index hospitalization, 4,371 patients finally made up the study population to assess the incidence of CIN and effect of CIN on long-term mortality (Figure 1). Demographic, angiographic, and procedural data were collected from hospital charts or databases in each center by independent clinical research coordinators, as previously www.ajconline.org
2
The American Journal of Cardiology (www.ajconline.org)
Figure 1. Patient flowchart for the present analysis.
described.9 Follow-up data were obtained from hospital charts or by contacting patients or referring physicians. In this study, we primarily defined CIN as an increase in the peak SCr concentration of 0.5 mg/dl after PCI from the SCr concentration before PCI. We also defined CIN-25% and CIN-0.3 as elevations in the SCr concentration 25% and 0.3 mg/dl after PCI, respectively, for sensitivity analyses. CKD was defined as an estimated glomerular filtration rate of <60 ml/min/1.73 m2 using the new equation in Japanese patients.10 The primary outcome of this study was all-cause mortality after discharge. We excluded patients with in-hospital death at index PCI because most patients who died during index hospitalization were severely ill on admission and/or after PCI, which predisposed them to acute deterioration of the kidney function related to several factors including CIN. Categorical variables are expressed as numbers and percentages and were compared using the chi-square test. Continuous variables of each group are presented with the mean SD and were compared using the Student t test or Wilcoxon rank-sum test based on their distributions. The cumulative incidence of all-cause death was estimated by the Kaplan-Meier method, and differences were assessed with the log-rank test. The adjusted relationships between CIN and death after discharge were evaluated using Cox proportional hazard models. In the CREDO-Kyoto registry, we previously identified 14 confounders correlated with death after discharge, including age 75 years, CKD, hemodialysis, history of heart failure, chronic obstructive lung disease, malignancy, anemia, peripheral vascular disease, stroke, left ventricular dysfunction, body mass index 25.0 kg/m2, diabetes with insulin, absence of statin use, and use of angiotensin-converting enzyme inhibitors.9 In this study, we excluded only 1 factor, hemodialysis, and adjusted the mortality with the remaining 13 confounders for analysis of the entire cohort. For the analysis of patients with or without CKD, we adjusted the mortality with the 12 confounders other than hemodialysis and CKD. The results of analyses are shown as the hazard ratio (HR) given with 95% confidence intervals (CIs) and p values. Results were considered significant with a p value of <0.05. We used JMP 8.0 (SAS Institute Inc., Cary, North Carolina) for all analyses. The investigators had full access to the data
and take responsibility for its integrity. All the investigators have read and approved the report as written. Results Baseline characteristics were significantly different between patients discharged alive and those who died in hospital (Supplementary Material). Patients discharged alive were younger and more likely to have a preserved ejection fraction and renal function and a current smoking habit. The in-hospital death group more often had a history of heart failure, diabetes, stroke, anemia, triple-vessel disease, proximal left anterior descending coronary artery disease, and total occlusion. Of the 255 patients who showed an increase in the peak SCr concentration of 0.5 mg/dl after PCI from the baseline SCr concentration, 237 patients (93%) were discharged alive. Of the 4,371 patients discharged alive in the present study, 1,650 patients had CKD. Baseline characteristics were significantly different between patients with CKD and those without CKD (Table 1). The CKD group included more elderly and high-risk patients. In our study population, the incidence of CIN was 5% in the entire cohort. The incidence of CIN was significantly greater in patients with CKD than in those without CKD (11% vs 2%, p <0.0001; Table 2). The baseline clinical characteristics of patients who developed CIN revealed a high prevalence of the high-risk factors, such as advanced age, heart failure, insulin-treated diabetes, CKD, and triple-vessel disease (Supplementary Material). During a median follow-up of 42.3 months (interquartile range 31.4 to 52.0) after discharge, 374 patients (8.6%) died. The unadjusted survival probability of patients with CIN was significantly less than that of patients without CIN in the entire cohort (Figure 2). In patients with CKD, the unadjusted survival rate of patients with CIN was significantly lower compared with that of those without CIN (Figure 2), whereas it was not significantly different in patients without CKD (Figure 2). After adjustment for the prespecified confounders, patients with CIN were significantly correlated with longterm mortality in the entire cohort (HR 2.26, 95% CI 1.62 to 2.29, p <0.0001). The association between CIN and
Coronary Artery Disease/Contrast-Induced Nephropathy and Mortality
3
Table 1 Baseline characteristics Variable
Age (years) 75 80 Female Body mass index (kg/m2) >25.0 Ejection fraction (%) 40% Heart failure Functional class 3/4 Prior myocardial infarction Atrial fibrillation Diabetes mellitus Insulin-treated Oral drug-treated Hemoglobin A1c (%) Hypertension Blood pressure (mm Hg) Systolic Diastolic Current smoker Stroke Peripheral vascular disease Chronic pulmonary disease Malignancy Baseline renal function Serum creatinine (mg/dL) Estimated glomerular filtration rate (mL/min/1.73 m2) <30 mL/min/1.73 m2 Anemia Emergency procedure Triple vessel coronary disease Left main coronary artery narrowing Proximal left anterior descending coronary artery narrowing Total coronary occlusion Treatment of 2 vessels Number of target vessels Medication at hospital discharge Statins Aspirin Thienopyridines Angiotensin-converting enzyme inhibitors Angiotensin-II receptor antagonists Beta-blockers Calcium channel blockers Nitrates
Entire Cohort (n ¼ 4371)
Chronic Kidney Disease
p Value*
Yes (n ¼ 1650)
No (n ¼ 2721)
71.6 8.7 630 (38%) 316 (19%) 563 (34%) 23.5 (21.4e25.7) 506 (31%) 60.7 14.2 144 (8.7%) 365 (22%) 123 (7.5%) 455 (28%) 169 (10%) 616 (37%) 166 (10%) 263 (16%) 7.1 1.4 1247 (76%)
65.2 10.4 542 (20%) 173 (6.4%) 775 (28%) 23.6 (21.8e25.6) 828 (30%) 63.9 12.5 118 (4.3%) 221 (8.1%) 76 (2.8%) 587 (22%) 135 (5.0%) 1010 (37%) 161 (5.9%) 461 (17%) 7.4 1.6 1789 (66%)
<0.0001 <0.0001 <0.0001 <0.0001 0.1 0.9 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 0.9 <0.0001 0.4 0.0005 <0.0001
137.3 22.2 75.8 13.3 1238 (28%) 664 (15%) 271 (6.2%) 118 (2.7%) 307 (7.0%)
138.8 23.6 75.9 13.4 349 (21%) 315 (19%) 150 (9.1%) 56 (3.4%) 137 (8.3%)
136.4 21.2 75.7 13.2 889 (33%) 349 (13%) 121 (4.5%) 62 (2.3%) 170 (6.3%)
0.0005 0.7 <0.0001 <0.0001 <0.0001 0.03 0.01
0.94 0.51 65.9 21.7 180 (4.1%) 959 (22%) 298 (6.8%) 996 (23%) 103 (2.4%) 1624 (37%) 1095 (25%) 1069 (24%) 1 (1e1) 1.27 0.53
1.26 0.70 45.7 11.7 180 (11%) 551 (33%) 98 (5.9%) 461 (28%) 42 (2.6%) 655 (40%) 463 (28%) 440 (27%) 1 (1e2) 1.30 0.54
0.74 0.14 78.2 16.5 0 (0%) 408 (15%) 200 (7.4%) 535 (20%) 61 (2.2%) 969 (36%) 632 (23%) 629 (23%) 1 (1e1) 1.26 0.52
<0.0001
67.7 10.3 1172 (27%) 489 (11%) 1338 (31%) 23.6 (21.6e25.6) 1334 (31%) 62.7 13.2 262 (6.0%) 586 (13%) 199 (4.6%) 1042 (24%) 304 (7.0%) 1626 (37%) 327 (7.5%) 724 (17%) 7.3 1.6 3036 (69%)
1423 3861 3400 1161 640 892 2476 3127
(33%) (88%) (78%) (27%) (15%) (20%) (57%) (72%)
518 1427 1253 487 289 362 936 1157
(31%) (86%) (76%) (30%) (18%) (22%) (57%) (70%)
905 2434 2147 674 351 530 1540 1970
(33%) (89%) (79%) (25%) (13%) (19%) (57%) (72%)
<0.0001 0.07 <0.0001 0.5 0.007 0.0004 0.008 0.009
0.2 0.003 0.02 0.0006 <0.0001 0.05 0.9 0.1
* Comparison between patients with and without CKD.
death was apparent in patients with CKD (HR 2.62, 95% CI 1.91 to 3.57, p <0.0001) but not in those without CKD (HR 1.23, 95% CI 0.47 to 2.62, p ¼ 0.6); however, there was no significant interaction between CIN and CKD (interaction p ¼ 0.2; Table 3). The incidences of CIN-25% and CIN-0.3 in the entire cohort were 22% and 14%, respectively (Table 2). Baseline clinical characteristics are listed in Supplementary Material. Sensitivity analyses for long-term mortality using the definitions of CIN-25% and CIN-0.3 were consistent with the primary analysis. Unadjusted survival probabilities of patients in the CIN-25% and CIN-0.3 groups in the entire
Table 2 Incidence of contrast-induced nephropathy Definition of CIN
Entire Cohort (n ¼ 4371)
CIN CIN-25% CIN-0.3
237 (5%) 943 (22%) 604 (14%)
Chronic Kidney Disease Yes (n ¼ 1650)
No (n ¼ 2721)
181 (11%) 341 (21%) 362 (22%)
56 (2%) 602 (22%) 242 (9%)
p Value*
<0.0001 0.3 <0.0001
CIN, CIN-25%, and CIN-0.3 were defined as an elevation of SCr by 0.5 mg/dl, 25%, and 0.3 mg/dl from the baseline, respectively. CIN ¼ contrast-induced nephropathy. * Comparison between patients with and without CKD.
4
The American Journal of Cardiology (www.ajconline.org)
Figure 2. Unadjusted Kaplan-Meier survival curves according to CIN defined as an elevation of the SCr 0.5 mg/dl in the entire cohort (A) and in patients with CKD (B) or without CKD (C). Table 3 The adjusted effect of CIN on long-term mortality Definition of CIN
Entire Cohort (n ¼ 4371)
Chronic Kidney Disease Yes (n ¼ 1650)
CIN CIN-25% CIN-0.3
Interaction p Value
No (n ¼ 2721)
HR (95% CI)
p Value
HR (95% CI)
p Value
HR (95% CI)
p Value
2.26 (1.69e2.99) 1.39 (1.10e1.74) 1.78 (1.40e2.25)
<0.0001 0.005 <0.0001
2.62 (1.91e3.57) 1.68 (1.25e2.25) 2.03 (1.53e2.70)
<0.0001 0.0006 <0.0001
1.23 (0.47e2.62) 1.10 (0.75e1.59) 1.39 (0.85e2.16)
0.6 0.6 0.2
0.2 0.1 0.2
CI ¼ confidence interval; CIN ¼ contrast-induced nephropathy; HR ¼ hazard ratio.
cohort and in patients with CKD were significantly less compared with those in patients without CIN (Figures 3 and 4) but not in patients without CKD as of CIN-25% (Figure 3). Although unadjusted survival probabilities of patients in the CIN-0.3 group in patients without CKD were significantly less compared with those in patients without CIN (Figure 4), the difference was no longer significant after adjusting for confounders (Table 3). Discussion In our study population, the incidence of CIN after PCI was 5% in the entire cohort. The rate of CIN was 11% in
patients with CKD and 2% in patients without CKD (p <0.0001), although the reported incidence and clinical significance of CIN varied among studies because of differences in the definition of CIN and backgrounds of studied populations.11 CIN was typically defined as an absolute increase in the SCr concentration of 0.5 mg/dl12e14 or sometimes a relative increase in the SCr concentration of 25%15,16 after contrast exposure. Recently, the Acute Kidney Injury Network proposed uniform standards for diagnosing and classifying acute kidney injury, and the Acute Kidney Injury Network definition is an absolute increase in the SCr concentration by 0.3 mg/dl within 48 hours.17 In the present study, the incidence of CIN-25%
Coronary Artery Disease/Contrast-Induced Nephropathy and Mortality
5
Figure 3. Unadjusted Kaplan-Meier survival curves according to CIN-25% defined as an elevation of the SCr 25% in the entire cohort (A) and in patients with CKD (B) or without CKD (C).
(21%) in patients with CKD was similar to that of CIN-25% (22%) in patients without CKD (p ¼ 0.3), suggesting that an absolute increase in the SCr should be regarded as CIN compared with the definition of a relative increase of 25%. With the definition of an absolute 0.3 mg/dl increase of the SCr, the incidence of CIN-0.3 in patients with CKD (22%) was significantly greater than that in patients without CKD (9%; p <0.0001). In our study cohort, CIN was significantly correlated with long-term mortality after adjustment in the entire cohort and patients with CKD. However, CIN was not associated with long-term mortality in patients without CKD. We confirmed these results by sensitivity analyses using the definitions of CIN-25% and CIN-0.3. Although Dangas et al5 previously reported that the prognostic impact of CIN and/or CIN-25% on 1-year mortality was marked in patients with baseline CKD (odds ratio 2.37, 95% CI 1.63 to 3.44, p <0.0001) and without baseline CKD (odds ratio 1.86, 95% CI 1.27 to 2.74, p ¼ 0.002), a stronger association between CIN and 1-year mortality was observed in patients with CKD. Because there was a difference in the definition of CIN, the background risk of patients, and the follow-up period compared with our study, we confirmed our results by sensitivity analyses using various definitions of CIN, as previously
mentioned, and we conducted analysis over a longer follow-up period. Many studies have addressed the question of whether CIN is the cause or just a marker of a high mortality rate.18 A decreased renal function has consistently been found as an independent risk factor for all-cause mortality.19 Therefore, the additional renal damage caused by contrast exposure might have a greater impact on mortality in patients with basal kidney dysfunction, and CIN would be the cause of a high mortality rate. Alternatively, CIN might be a more sensitive marker of death, especially in such a high-risk population with CKD. These issues should be investigated in future prospective studies. The present study has several limitations. First, we defined CIN depending on the difference between the baseline and the peak SCr concentration on index hospitalization. The time of collection of postintervention SCr was not provided because our study is a post hoc analysis of an existing database of the registry. Thus, we could not exclude the possibility that the SCr concentration reached a peak after discharge, which might underestimate the incidence of CIN. On the contrary, the SCr concentration increased as a consequence of the worsening general condition regardless of contrast exposure, which might overestimate the rate of CIN. Second, we included patients in a
6
The American Journal of Cardiology (www.ajconline.org)
Figure 4. Unadjusted Kaplan-Meier survival curves according to CIN-0.3 defined as an elevation of the SCr 0.3 mg/dl in the entire cohort (A) and in patients with CKD (B) or without CKD (C).
retrospective fashion. Therefore, all patients underwent PCI according to the attending physicians’ judgment, including consideration of the potential adverse influence of contrast exposure on the renal function. Such a situation undoubtedly resulted in selection bias. Third, data on the amount and type of contrast, hydration status, and potential nephrotoxic drugs such as nonsteroidal anti-inflammatory drugs were unavailable because of methodological limitations inherent in retrospective analyses. Fourth, the incidence of CIN in patients without CKD was low (2%), and only 56 patients developed CIN. Thus, the statistical power might be insufficient to evaluate the effect of CIN on long-term mortality in patients without CKD. Disclosures The authors have no conflicts of interest to disclose. Supplementary Data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j. amjcard.2014.05.009. 1. Nash K, Hafeez A, Hou S. Hospital-acquired renal insufficiency. Am J Kidney Dis 2002;39:930e936.
2. McCullough PA. Contrast-induced acute kidney injury. J Am Coll Cardiol 2008;51:1419e1428. 3. McCullough PA, Wolyn R, Rocher LL, Levin RN, O’Neill WW. Acute renal failure after coronary intervention: incidence, risk factors, and relationship to mortality. Am J Med 1997;103:368e375. 4. Nikolsky E, Mehran R, Turcot D, Aymong ED, Mintz GS, Lasic Z, Lansky AJ, Tsounias E, Moses JW, Stone GW, Leon MB, Dangas GD. Impact of chronic kidney disease on prognosis of patients with diabetes mellitus treated with percutaneous coronary intervention. Am J Cardiol 2004;94:300e305. 5. Dangas G, Iakovou I, Nikolsky E, Aymong ED, Mintz GS, Kipshidze NN, Lansky AJ, Moussa I, Stone GW, Moses JW, Leon MB, Mehran R. Contrast-induced nephropathy after percutaneous coronary interventions in relation to chronic kidney disease and hemodynamic variables. Am J Cardiol 2005;95:13e19. 6. Abe M, Kimura T, Morimoto T, Furukawa Y, Kita T. Incidence of and risk factors for contrast-induced nephropathy after cardiac catheterization in Japanese patients. Circ J 2009;73:1518e1522. 7. Rihal CS, Textor SC, Grill DE, Berger PB, Ting HH, Best PJ, Singh M, Bell MR, Barsness GW, Mathew V, Garratt KN, Holmes DR Jr. Incidence and prognostic importance of acute renal failure after percutaneous coronary intervention. Circulation 2002;105: 2259e2264. 8. James MT, Ghali WA, Knudtson ML, Ravani P, Tonelli M, Faris P, Pannu N, Manns BJ, Klarenbach SW, Hemmelgarn BR. Associations between acute kidney injury and cardiovascular and renal outcomes after coronary angiography. Circulation 2011;123:409e416. 9. Kimura T, Morimoto T, Furukawa Y, Nakagawa Y, Shizuta S, Ehara N, Taniguchi R, Doi T, Nishiyama K, Ozasa N, Saito N, Hoshino K, Mitsuoka H, Abe M, Toma M, Tamura T, Haruna Y, Imai Y, Teramukai S, Fukushima M, Kita T. Long-term outcomes of
Coronary Artery Disease/Contrast-Induced Nephropathy and Mortality
10.
11. 12.
13. 14.
coronary-artery bypass graft surgery versus percutaneous coronary intervention for multivessel coronary artery disease in the bare-metal stent era. Circulation 2008;118:S199eS209. Matsuo S, Imai E, Horio M, Yasuda Y, Tomita K, Nitta K, Yamagata K, Tomino Y, Yokoyama H, Hishida A. Revised equations for estimated GFR from serum creatinine in Japan. Am J Kidney Dis 2009;53: 982e992. Barrett BJ, Parfrey PS. Clinical practice. Preventing nephropathy induced by contrast medium. N Engl J Med 2006;354:379e386. Solomon R, Werner C, Mann D, D’Elia J, Silva P. Effects of saline, mannitol, and furosemide to prevent acute decreases in renal function induced by radiocontrast agents. N Engl J Med 1994;331: 1416e1420. Tepel M, van der Giet M, Schwarzfeld C, Laufer U, Liermann D, Zidek W. Prevention of radiographic-contrast-agent-induced reductions in renal function by acetylcysteine. N Engl J Med 2000;343:180e184. Recio-Mayoral A, Chaparro M, Prado B, Cozar R, Mendez I, Banerjee D, Kaski JC, Cubero J, Cruz JM. The reno-protective effect of hydration with sodium bicarbonate plus N-acetylcysteine in patients undergoing emergency percutaneous coronary intervention: the RENO Study. J Am Coll Cardiol 2007;49:1283e1288.
7
15. Marenzi G, Marana I, Lauri G, Assanelli E, Grazi M, Campodonico J, Trabattoni D, Fabbiocchi F, Montorsi P, Bartorelli AL. The prevention of radiocontrast-agent-induced nephropathy by hemofiltration. N Engl J Med 2003;349:1333e1340. 16. Merten GJ, Burgess WP, Gray LV, Holleman JH, Roush TS, Kowalchuk GJ, Bersin RM, Van Moore A, Simonton CA 3rd, Rittase RA, Norton HJ, Kennedy TP. Prevention of contrast-induced nephropathy with sodium bicarbonate: a randomized controlled trial. JAMA 2004;291:2328e2334. 17. Mehta RL, Kellum JA, Shah SV, Molitoris BA, Ronco C, Warnock DG, Levin A. Acute Kidney Injury Network: report of an initiative to improve outcomes in acute kidney injury. Crit Care 2007;11:R31. 18. Rudnick M, Feldman H. Contrast-induced nephropathy: what are the true clinical consequences? Clin J Am Soc Nephrol 2008;3:263e272. 19. Sarnak MJ, Levey AS, Schoolwerth AC, Coresh J, Culleton B, Hamm LL, McCullough PA, Kasiske BL, Kelepouris E, Klag MJ, Parfrey P, Pfeffer M, Raij L, Spinosa DJ, Wilson PW. Kidney disease as a risk factor for development of cardiovascular disease: a statement from the American Heart Association Councils on Kidney in Cardiovascular Disease, High Blood Pressure Research, Clinical Cardiology, and Epidemiology and Prevention. Circulation 2003;108:2154e2169.
a Division of Cardiology, National Hospital Organization Kyoto Medical Center, Kyoto, Japan; bDivision of General Internal Medicine, Hyogo College of Medicine, Nishinomiya, Japan; cDepartment of Cardiovascular Medicine, Kobe City Medical Center General Hospital, Kobe, Japan; d Division of Cardiology, Tenri Hospital, Tenri, Japan; eDepartment of Cardiovascular Medicine, Graduate School of Medicine, Kyoto University, Kyoto, Japan; fDivision of Cardiology, Hyogo Prefectural Amagasaki Hospital, Amagasaki, Japan; gDepartment of Primary Care and Emergency Medicine, Kyoto University Hospital, Kyoto, Japan; hDivision of Cardiology, Nagai Hospital, Tsu, Japan; iDivision of Cardiology, Nara Hospital, Kinki University Faculty of Medicine, Ikoma, Japan; and jDivision of Cardiology, Hirakata Kohsai Hospital, Hirakata, Japan. Manuscript received February 24, 2014; revised manuscript received and accepted May 6, 2014. This work was supported by an educational grant from the Research Institute for Production Development, Kyoto, Japan. See page 6 for disclosure information. *Corresponding author: Tel: (þ81) 75-641-9161; fax: (þ81) 75-6434325. E-mail address: [email protected] (M. Abe).
0002-9149/14/$ - see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.amjcard.2014.05.009
we sought to clarify the incidence of CIN after percutaneous coronary intervention (PCI) and the effect of CIN on long-term mortality in patients with or without CKD using a large observational database of patients who underwent their first PCI in Japan. Methods The Coronary REvascularization Demonstrating Outcome Study in Kyoto (CREDO-Kyoto) is a multicenter registry in Japan that enrolled consecutive patients who underwent their first PCI or coronary artery bypass grafting from January 2000 to December 2002, excluding patients with acute myocardial infarction within a week before the index procedure. The relevant review board or ethics committees in all 30 participating centers approved the research protocol. The study design and patient enrollment of the CREDO-Kyoto registry were previously described in detail.9 Of the 9,877 patients who underwent PCI or coronary artery bypass grafting enrolled during the study period, we excluded 2,999 patients receiving coronary artery bypass grafting, 250 patients on hemodialysis, and 2,233 patients without paired serum creatinine (SCr) measurements before and after PCI. After excluding 24 patients who died during the index hospitalization, 4,371 patients finally made up the study population to assess the incidence of CIN and effect of CIN on long-term mortality (Figure 1). Demographic, angiographic, and procedural data were collected from hospital charts or databases in each center by independent clinical research coordinators, as previously www.ajconline.org
2
The American Journal of Cardiology (www.ajconline.org)
Figure 1. Patient flowchart for the present analysis.
described.9 Follow-up data were obtained from hospital charts or by contacting patients or referring physicians. In this study, we primarily defined CIN as an increase in the peak SCr concentration of 0.5 mg/dl after PCI from the SCr concentration before PCI. We also defined CIN-25% and CIN-0.3 as elevations in the SCr concentration 25% and 0.3 mg/dl after PCI, respectively, for sensitivity analyses. CKD was defined as an estimated glomerular filtration rate of <60 ml/min/1.73 m2 using the new equation in Japanese patients.10 The primary outcome of this study was all-cause mortality after discharge. We excluded patients with in-hospital death at index PCI because most patients who died during index hospitalization were severely ill on admission and/or after PCI, which predisposed them to acute deterioration of the kidney function related to several factors including CIN. Categorical variables are expressed as numbers and percentages and were compared using the chi-square test. Continuous variables of each group are presented with the mean SD and were compared using the Student t test or Wilcoxon rank-sum test based on their distributions. The cumulative incidence of all-cause death was estimated by the Kaplan-Meier method, and differences were assessed with the log-rank test. The adjusted relationships between CIN and death after discharge were evaluated using Cox proportional hazard models. In the CREDO-Kyoto registry, we previously identified 14 confounders correlated with death after discharge, including age 75 years, CKD, hemodialysis, history of heart failure, chronic obstructive lung disease, malignancy, anemia, peripheral vascular disease, stroke, left ventricular dysfunction, body mass index 25.0 kg/m2, diabetes with insulin, absence of statin use, and use of angiotensin-converting enzyme inhibitors.9 In this study, we excluded only 1 factor, hemodialysis, and adjusted the mortality with the remaining 13 confounders for analysis of the entire cohort. For the analysis of patients with or without CKD, we adjusted the mortality with the 12 confounders other than hemodialysis and CKD. The results of analyses are shown as the hazard ratio (HR) given with 95% confidence intervals (CIs) and p values. Results were considered significant with a p value of <0.05. We used JMP 8.0 (SAS Institute Inc., Cary, North Carolina) for all analyses. The investigators had full access to the data
and take responsibility for its integrity. All the investigators have read and approved the report as written. Results Baseline characteristics were significantly different between patients discharged alive and those who died in hospital (Supplementary Material). Patients discharged alive were younger and more likely to have a preserved ejection fraction and renal function and a current smoking habit. The in-hospital death group more often had a history of heart failure, diabetes, stroke, anemia, triple-vessel disease, proximal left anterior descending coronary artery disease, and total occlusion. Of the 255 patients who showed an increase in the peak SCr concentration of 0.5 mg/dl after PCI from the baseline SCr concentration, 237 patients (93%) were discharged alive. Of the 4,371 patients discharged alive in the present study, 1,650 patients had CKD. Baseline characteristics were significantly different between patients with CKD and those without CKD (Table 1). The CKD group included more elderly and high-risk patients. In our study population, the incidence of CIN was 5% in the entire cohort. The incidence of CIN was significantly greater in patients with CKD than in those without CKD (11% vs 2%, p <0.0001; Table 2). The baseline clinical characteristics of patients who developed CIN revealed a high prevalence of the high-risk factors, such as advanced age, heart failure, insulin-treated diabetes, CKD, and triple-vessel disease (Supplementary Material). During a median follow-up of 42.3 months (interquartile range 31.4 to 52.0) after discharge, 374 patients (8.6%) died. The unadjusted survival probability of patients with CIN was significantly less than that of patients without CIN in the entire cohort (Figure 2). In patients with CKD, the unadjusted survival rate of patients with CIN was significantly lower compared with that of those without CIN (Figure 2), whereas it was not significantly different in patients without CKD (Figure 2). After adjustment for the prespecified confounders, patients with CIN were significantly correlated with longterm mortality in the entire cohort (HR 2.26, 95% CI 1.62 to 2.29, p <0.0001). The association between CIN and
Coronary Artery Disease/Contrast-Induced Nephropathy and Mortality
3
Table 1 Baseline characteristics Variable
Age (years) 75 80 Female Body mass index (kg/m2) >25.0 Ejection fraction (%) 40% Heart failure Functional class 3/4 Prior myocardial infarction Atrial fibrillation Diabetes mellitus Insulin-treated Oral drug-treated Hemoglobin A1c (%) Hypertension Blood pressure (mm Hg) Systolic Diastolic Current smoker Stroke Peripheral vascular disease Chronic pulmonary disease Malignancy Baseline renal function Serum creatinine (mg/dL) Estimated glomerular filtration rate (mL/min/1.73 m2) <30 mL/min/1.73 m2 Anemia Emergency procedure Triple vessel coronary disease Left main coronary artery narrowing Proximal left anterior descending coronary artery narrowing Total coronary occlusion Treatment of 2 vessels Number of target vessels Medication at hospital discharge Statins Aspirin Thienopyridines Angiotensin-converting enzyme inhibitors Angiotensin-II receptor antagonists Beta-blockers Calcium channel blockers Nitrates
Entire Cohort (n ¼ 4371)
Chronic Kidney Disease
p Value*
Yes (n ¼ 1650)
No (n ¼ 2721)
71.6 8.7 630 (38%) 316 (19%) 563 (34%) 23.5 (21.4e25.7) 506 (31%) 60.7 14.2 144 (8.7%) 365 (22%) 123 (7.5%) 455 (28%) 169 (10%) 616 (37%) 166 (10%) 263 (16%) 7.1 1.4 1247 (76%)
65.2 10.4 542 (20%) 173 (6.4%) 775 (28%) 23.6 (21.8e25.6) 828 (30%) 63.9 12.5 118 (4.3%) 221 (8.1%) 76 (2.8%) 587 (22%) 135 (5.0%) 1010 (37%) 161 (5.9%) 461 (17%) 7.4 1.6 1789 (66%)
<0.0001 <0.0001 <0.0001 <0.0001 0.1 0.9 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 0.9 <0.0001 0.4 0.0005 <0.0001
137.3 22.2 75.8 13.3 1238 (28%) 664 (15%) 271 (6.2%) 118 (2.7%) 307 (7.0%)
138.8 23.6 75.9 13.4 349 (21%) 315 (19%) 150 (9.1%) 56 (3.4%) 137 (8.3%)
136.4 21.2 75.7 13.2 889 (33%) 349 (13%) 121 (4.5%) 62 (2.3%) 170 (6.3%)
0.0005 0.7 <0.0001 <0.0001 <0.0001 0.03 0.01
0.94 0.51 65.9 21.7 180 (4.1%) 959 (22%) 298 (6.8%) 996 (23%) 103 (2.4%) 1624 (37%) 1095 (25%) 1069 (24%) 1 (1e1) 1.27 0.53
1.26 0.70 45.7 11.7 180 (11%) 551 (33%) 98 (5.9%) 461 (28%) 42 (2.6%) 655 (40%) 463 (28%) 440 (27%) 1 (1e2) 1.30 0.54
0.74 0.14 78.2 16.5 0 (0%) 408 (15%) 200 (7.4%) 535 (20%) 61 (2.2%) 969 (36%) 632 (23%) 629 (23%) 1 (1e1) 1.26 0.52
<0.0001
67.7 10.3 1172 (27%) 489 (11%) 1338 (31%) 23.6 (21.6e25.6) 1334 (31%) 62.7 13.2 262 (6.0%) 586 (13%) 199 (4.6%) 1042 (24%) 304 (7.0%) 1626 (37%) 327 (7.5%) 724 (17%) 7.3 1.6 3036 (69%)
1423 3861 3400 1161 640 892 2476 3127
(33%) (88%) (78%) (27%) (15%) (20%) (57%) (72%)
518 1427 1253 487 289 362 936 1157
(31%) (86%) (76%) (30%) (18%) (22%) (57%) (70%)
905 2434 2147 674 351 530 1540 1970
(33%) (89%) (79%) (25%) (13%) (19%) (57%) (72%)
<0.0001 0.07 <0.0001 0.5 0.007 0.0004 0.008 0.009
0.2 0.003 0.02 0.0006 <0.0001 0.05 0.9 0.1
* Comparison between patients with and without CKD.
death was apparent in patients with CKD (HR 2.62, 95% CI 1.91 to 3.57, p <0.0001) but not in those without CKD (HR 1.23, 95% CI 0.47 to 2.62, p ¼ 0.6); however, there was no significant interaction between CIN and CKD (interaction p ¼ 0.2; Table 3). The incidences of CIN-25% and CIN-0.3 in the entire cohort were 22% and 14%, respectively (Table 2). Baseline clinical characteristics are listed in Supplementary Material. Sensitivity analyses for long-term mortality using the definitions of CIN-25% and CIN-0.3 were consistent with the primary analysis. Unadjusted survival probabilities of patients in the CIN-25% and CIN-0.3 groups in the entire
Table 2 Incidence of contrast-induced nephropathy Definition of CIN
Entire Cohort (n ¼ 4371)
CIN CIN-25% CIN-0.3
237 (5%) 943 (22%) 604 (14%)
Chronic Kidney Disease Yes (n ¼ 1650)
No (n ¼ 2721)
181 (11%) 341 (21%) 362 (22%)
56 (2%) 602 (22%) 242 (9%)
p Value*
<0.0001 0.3 <0.0001
CIN, CIN-25%, and CIN-0.3 were defined as an elevation of SCr by 0.5 mg/dl, 25%, and 0.3 mg/dl from the baseline, respectively. CIN ¼ contrast-induced nephropathy. * Comparison between patients with and without CKD.
4
The American Journal of Cardiology (www.ajconline.org)
Figure 2. Unadjusted Kaplan-Meier survival curves according to CIN defined as an elevation of the SCr 0.5 mg/dl in the entire cohort (A) and in patients with CKD (B) or without CKD (C). Table 3 The adjusted effect of CIN on long-term mortality Definition of CIN
Entire Cohort (n ¼ 4371)
Chronic Kidney Disease Yes (n ¼ 1650)
CIN CIN-25% CIN-0.3
Interaction p Value
No (n ¼ 2721)
HR (95% CI)
p Value
HR (95% CI)
p Value
HR (95% CI)
p Value
2.26 (1.69e2.99) 1.39 (1.10e1.74) 1.78 (1.40e2.25)
<0.0001 0.005 <0.0001
2.62 (1.91e3.57) 1.68 (1.25e2.25) 2.03 (1.53e2.70)
<0.0001 0.0006 <0.0001
1.23 (0.47e2.62) 1.10 (0.75e1.59) 1.39 (0.85e2.16)
0.6 0.6 0.2
0.2 0.1 0.2
CI ¼ confidence interval; CIN ¼ contrast-induced nephropathy; HR ¼ hazard ratio.
cohort and in patients with CKD were significantly less compared with those in patients without CIN (Figures 3 and 4) but not in patients without CKD as of CIN-25% (Figure 3). Although unadjusted survival probabilities of patients in the CIN-0.3 group in patients without CKD were significantly less compared with those in patients without CIN (Figure 4), the difference was no longer significant after adjusting for confounders (Table 3). Discussion In our study population, the incidence of CIN after PCI was 5% in the entire cohort. The rate of CIN was 11% in
patients with CKD and 2% in patients without CKD (p <0.0001), although the reported incidence and clinical significance of CIN varied among studies because of differences in the definition of CIN and backgrounds of studied populations.11 CIN was typically defined as an absolute increase in the SCr concentration of 0.5 mg/dl12e14 or sometimes a relative increase in the SCr concentration of 25%15,16 after contrast exposure. Recently, the Acute Kidney Injury Network proposed uniform standards for diagnosing and classifying acute kidney injury, and the Acute Kidney Injury Network definition is an absolute increase in the SCr concentration by 0.3 mg/dl within 48 hours.17 In the present study, the incidence of CIN-25%
Coronary Artery Disease/Contrast-Induced Nephropathy and Mortality
5
Figure 3. Unadjusted Kaplan-Meier survival curves according to CIN-25% defined as an elevation of the SCr 25% in the entire cohort (A) and in patients with CKD (B) or without CKD (C).
(21%) in patients with CKD was similar to that of CIN-25% (22%) in patients without CKD (p ¼ 0.3), suggesting that an absolute increase in the SCr should be regarded as CIN compared with the definition of a relative increase of 25%. With the definition of an absolute 0.3 mg/dl increase of the SCr, the incidence of CIN-0.3 in patients with CKD (22%) was significantly greater than that in patients without CKD (9%; p <0.0001). In our study cohort, CIN was significantly correlated with long-term mortality after adjustment in the entire cohort and patients with CKD. However, CIN was not associated with long-term mortality in patients without CKD. We confirmed these results by sensitivity analyses using the definitions of CIN-25% and CIN-0.3. Although Dangas et al5 previously reported that the prognostic impact of CIN and/or CIN-25% on 1-year mortality was marked in patients with baseline CKD (odds ratio 2.37, 95% CI 1.63 to 3.44, p <0.0001) and without baseline CKD (odds ratio 1.86, 95% CI 1.27 to 2.74, p ¼ 0.002), a stronger association between CIN and 1-year mortality was observed in patients with CKD. Because there was a difference in the definition of CIN, the background risk of patients, and the follow-up period compared with our study, we confirmed our results by sensitivity analyses using various definitions of CIN, as previously
mentioned, and we conducted analysis over a longer follow-up period. Many studies have addressed the question of whether CIN is the cause or just a marker of a high mortality rate.18 A decreased renal function has consistently been found as an independent risk factor for all-cause mortality.19 Therefore, the additional renal damage caused by contrast exposure might have a greater impact on mortality in patients with basal kidney dysfunction, and CIN would be the cause of a high mortality rate. Alternatively, CIN might be a more sensitive marker of death, especially in such a high-risk population with CKD. These issues should be investigated in future prospective studies. The present study has several limitations. First, we defined CIN depending on the difference between the baseline and the peak SCr concentration on index hospitalization. The time of collection of postintervention SCr was not provided because our study is a post hoc analysis of an existing database of the registry. Thus, we could not exclude the possibility that the SCr concentration reached a peak after discharge, which might underestimate the incidence of CIN. On the contrary, the SCr concentration increased as a consequence of the worsening general condition regardless of contrast exposure, which might overestimate the rate of CIN. Second, we included patients in a
6
The American Journal of Cardiology (www.ajconline.org)
Figure 4. Unadjusted Kaplan-Meier survival curves according to CIN-0.3 defined as an elevation of the SCr 0.3 mg/dl in the entire cohort (A) and in patients with CKD (B) or without CKD (C).
retrospective fashion. Therefore, all patients underwent PCI according to the attending physicians’ judgment, including consideration of the potential adverse influence of contrast exposure on the renal function. Such a situation undoubtedly resulted in selection bias. Third, data on the amount and type of contrast, hydration status, and potential nephrotoxic drugs such as nonsteroidal anti-inflammatory drugs were unavailable because of methodological limitations inherent in retrospective analyses. Fourth, the incidence of CIN in patients without CKD was low (2%), and only 56 patients developed CIN. Thus, the statistical power might be insufficient to evaluate the effect of CIN on long-term mortality in patients without CKD. Disclosures The authors have no conflicts of interest to disclose. Supplementary Data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j. amjcard.2014.05.009. 1. Nash K, Hafeez A, Hou S. Hospital-acquired renal insufficiency. Am J Kidney Dis 2002;39:930e936.
2. McCullough PA. Contrast-induced acute kidney injury. J Am Coll Cardiol 2008;51:1419e1428. 3. McCullough PA, Wolyn R, Rocher LL, Levin RN, O’Neill WW. Acute renal failure after coronary intervention: incidence, risk factors, and relationship to mortality. Am J Med 1997;103:368e375. 4. Nikolsky E, Mehran R, Turcot D, Aymong ED, Mintz GS, Lasic Z, Lansky AJ, Tsounias E, Moses JW, Stone GW, Leon MB, Dangas GD. Impact of chronic kidney disease on prognosis of patients with diabetes mellitus treated with percutaneous coronary intervention. Am J Cardiol 2004;94:300e305. 5. Dangas G, Iakovou I, Nikolsky E, Aymong ED, Mintz GS, Kipshidze NN, Lansky AJ, Moussa I, Stone GW, Moses JW, Leon MB, Mehran R. Contrast-induced nephropathy after percutaneous coronary interventions in relation to chronic kidney disease and hemodynamic variables. Am J Cardiol 2005;95:13e19. 6. Abe M, Kimura T, Morimoto T, Furukawa Y, Kita T. Incidence of and risk factors for contrast-induced nephropathy after cardiac catheterization in Japanese patients. Circ J 2009;73:1518e1522. 7. Rihal CS, Textor SC, Grill DE, Berger PB, Ting HH, Best PJ, Singh M, Bell MR, Barsness GW, Mathew V, Garratt KN, Holmes DR Jr. Incidence and prognostic importance of acute renal failure after percutaneous coronary intervention. Circulation 2002;105: 2259e2264. 8. James MT, Ghali WA, Knudtson ML, Ravani P, Tonelli M, Faris P, Pannu N, Manns BJ, Klarenbach SW, Hemmelgarn BR. Associations between acute kidney injury and cardiovascular and renal outcomes after coronary angiography. Circulation 2011;123:409e416. 9. Kimura T, Morimoto T, Furukawa Y, Nakagawa Y, Shizuta S, Ehara N, Taniguchi R, Doi T, Nishiyama K, Ozasa N, Saito N, Hoshino K, Mitsuoka H, Abe M, Toma M, Tamura T, Haruna Y, Imai Y, Teramukai S, Fukushima M, Kita T. Long-term outcomes of
Coronary Artery Disease/Contrast-Induced Nephropathy and Mortality
10.
11. 12.
13. 14.
coronary-artery bypass graft surgery versus percutaneous coronary intervention for multivessel coronary artery disease in the bare-metal stent era. Circulation 2008;118:S199eS209. Matsuo S, Imai E, Horio M, Yasuda Y, Tomita K, Nitta K, Yamagata K, Tomino Y, Yokoyama H, Hishida A. Revised equations for estimated GFR from serum creatinine in Japan. Am J Kidney Dis 2009;53: 982e992. Barrett BJ, Parfrey PS. Clinical practice. Preventing nephropathy induced by contrast medium. N Engl J Med 2006;354:379e386. Solomon R, Werner C, Mann D, D’Elia J, Silva P. Effects of saline, mannitol, and furosemide to prevent acute decreases in renal function induced by radiocontrast agents. N Engl J Med 1994;331: 1416e1420. Tepel M, van der Giet M, Schwarzfeld C, Laufer U, Liermann D, Zidek W. Prevention of radiographic-contrast-agent-induced reductions in renal function by acetylcysteine. N Engl J Med 2000;343:180e184. Recio-Mayoral A, Chaparro M, Prado B, Cozar R, Mendez I, Banerjee D, Kaski JC, Cubero J, Cruz JM. The reno-protective effect of hydration with sodium bicarbonate plus N-acetylcysteine in patients undergoing emergency percutaneous coronary intervention: the RENO Study. J Am Coll Cardiol 2007;49:1283e1288.
7
15. Marenzi G, Marana I, Lauri G, Assanelli E, Grazi M, Campodonico J, Trabattoni D, Fabbiocchi F, Montorsi P, Bartorelli AL. The prevention of radiocontrast-agent-induced nephropathy by hemofiltration. N Engl J Med 2003;349:1333e1340. 16. Merten GJ, Burgess WP, Gray LV, Holleman JH, Roush TS, Kowalchuk GJ, Bersin RM, Van Moore A, Simonton CA 3rd, Rittase RA, Norton HJ, Kennedy TP. Prevention of contrast-induced nephropathy with sodium bicarbonate: a randomized controlled trial. JAMA 2004;291:2328e2334. 17. Mehta RL, Kellum JA, Shah SV, Molitoris BA, Ronco C, Warnock DG, Levin A. Acute Kidney Injury Network: report of an initiative to improve outcomes in acute kidney injury. Crit Care 2007;11:R31. 18. Rudnick M, Feldman H. Contrast-induced nephropathy: what are the true clinical consequences? Clin J Am Soc Nephrol 2008;3:263e272. 19. Sarnak MJ, Levey AS, Schoolwerth AC, Coresh J, Culleton B, Hamm LL, McCullough PA, Kasiske BL, Kelepouris E, Klag MJ, Parfrey P, Pfeffer M, Raij L, Spinosa DJ, Wilson PW. Kidney disease as a risk factor for development of cardiovascular disease: a statement from the American Heart Association Councils on Kidney in Cardiovascular Disease, High Blood Pressure Research, Clinical Cardiology, and Epidemiology and Prevention. Circulation 2003;108:2154e2169.