Author + information
- Received April 13, 2018
- Revision received July 11, 2018
- Accepted July 17, 2018
- Published online November 19, 2018.
- Santiago Garcia, MDa,∗ (, )
- Deepak L. Bhatt, MD, MPHb,
- Martin Gallagher, MD, PhDc,
- Hani Jneid, MDd,
- James Kaufman, MDe,
- Paul M. Palevsky, MDf,
- Hongsheng Wu, PhDe,
- Steven D. Weisbord, MD, MScf,
- for the PRESERVE Trial Group
- aMinneapolis VA Healthcare System, University of Minnesota, Minneapolis Heart Institute, Minneapolis, Minnesota
- bVA Boston Healthcare System and Brigham and Women’s Hospital Heart & Vascular Center, Harvard Medical School, Boston, Massachusetts
- cUniversity of Sydney, Sydney, Australia
- dMichael E. DeBakey VA Medical Center and Baylor College of Medicine, Houston, Texas
- eVA Cooperative Studies Program Coordinating Center, VA Boston Healthcare System, Boston, Massachusetts
- fVA Pittsburgh Healthcare System and University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania
- ↵∗Address for correspondence:
Dr. Santiago Garcia, 920 East 28th Street, Suite 300, Minneapolis Heart Institute, Minneapolis, Minnesota 55407.
Objectives The aim of this study was to compare intravenous (IV) sodium bicarbonate with IV sodium chloride and oral acetylcysteine with placebo for the prevention of contrast-associated acute kidney injury (CAAKI) and intermediate-term adverse outcomes.
Background Data are conflicting on the optimal strategy to reduce CAAKI and related complications after percutaneous coronary intervention (PCI).
Methods The PRESERVE (Prevention of Serious Adverse Events Following Angiography) trial used a 2 × 2 factorial design to randomize 5,177 patients with stage III or IV chronic kidney disease undergoing angiography to IV 1.26% sodium bicarbonate or IV 0.9% sodium chloride and 5 days of oral acetylcysteine or placebo. A subgroup analysis was conducted of the efficacy of these interventions in patients who underwent PCI during the study angiographic examination. The primary endpoint was a composite of death, need for dialysis, or persistent kidney impairment at 90 days; CAAKI was a secondary endpoint.
Results A total of 1,161 PRESERVE patients (mean age 69 ± 8 years) underwent PCI. The median estimated glomerular filtration rate was 50.7 ml/min/1.73 m2 (interquartile range: 41.7 to 60.1 ml/min/1.73 m2), and 952 patients (82%) had diabetes mellitus. The primary endpoint occurred in 15 of 568 patients (2.6%) in the IV sodium bicarbonate group and 24 of 593 patients (4.0%) in the IV sodium chloride group (odds ratio: 0.64; 95% confidence interval: 0.33 to 1.24; p for interaction = 0.41) and in 23 of 598 patients (3.8%) in the acetylcysteine group and 16 of 563 patients (2.8%) in the placebo group (odds ratio: 1.37; 95% confidence interval: 0.71 to 2.62; p for interaction = 0.29). There were no significant between-group differences in the rates of CAAKI.
Conclusions Among patients with CKD undergoing PCI, there was no benefit of IV sodium bicarbonate over IV sodium chloride or of acetylcysteine over placebo for the prevention of CAAKI or intermediate-term adverse outcomes.
Approximately 7% of patients undergoing percutaneous coronary intervention (PCI) experience contrast-associated acute kidney injury (CAAKI), a complication strongly associated with adverse clinical outcomes, including death and increased hospital costs and lengths of stay (1–5). Current guidelines recommend pre-procedural assessment of risk for CAAKI prior to PCI and adequate preparatory intravascular volume expansion in high-risk patients (6). Two widely used strategies to prevent CAAKI in clinical practice include the use of intravenous (IV) sodium bicarbonate to induce urine alkalization and scavenging of reactive oxygen species through peri-procedural oral administration of acetylcysteine (7–9). However, results of clinical trials and meta-analyses of these interventions have yielded inconsistent results, and their use remains controversial (10–13).
Our group recently completed the PRESERVE (Prevention of Serious Adverse Events Following Angiography) trial, which enrolled 4,993 patients with chronic kidney disease who were undergoing coronary or noncoronary angiography and used a 2 × 2 factorial design to compare IV sodium bicarbonate with IV sodium chloride and oral acetylcysteine with placebo for the prevention of 90-day death, need for dialysis, persistent decline in kidney function, and CAAKI (13). In the overall trial, we found no benefit of IV sodium bicarbonate compared with IV sodium chloride or of oral acetylcysteine compared with placebo for the prevention of any of these outcomes (14).
Notwithstanding these results in the full study population, patients undergoing PCI represent a subset at higher risk for adverse kidney outcomes because they receive higher volumes of contrast volume, a known risk factor for CAAKI (15), and have a high burden of comorbid conditions, including decompensated heart failure, acute coronary syndrome, and diabetes, which can exacerbate the deleterious effects of contrast media (16,17). In fact, subsequent to the publication of the PRESERVE trial results, some have questioned the generalizability of the findings of the overall trial to very high-risk patients, such as those undergoing PCI (18). In light of such questions and the importance of establishing the standard of care for the prevention of adverse kidney outcomes following PCI, we conducted a subgroup analysis of PRESERVE participants who underwent PCI to compare the efficacy of IV sodium bicarbonate with that of IV sodium chloride and of oral acetylcysteine compared with placebo in this high-risk patient group.
Trial design and interventions
PRESERVE was an international, double-blind, placebo- and comparator-controlled, randomized clinical trial with a 2 × 2 factorial design (NCT01467466). Patients were enrolled at 53 medical centers in the United States (35 Veterans Affairs sites), Australia (13 sites), Malaysia (3 sites), and New Zealand (2 sites). Study participants were randomized to receive IV 1.26% sodium bicarbonate or IV 0.9% sodium chloride and oral N-acetylcysteine or oral placebo using a centralized, computer-generated permuted-block plan stratified by site. The administration of IV fluids was based on protocol-specified ranges: 1 to 3 ml/kg/h over 1 to 12 h for a total of 3 to 12 ml/kg prior to angiography, 1 to 1.5 ml/kg/h during angiography/PCI, and 1 to 3 ml/kg/h over 2 to 12 h for a total of 6 to 12 ml/kg following angiography. A dose of 1,200 mg oral acetylcysteine or matching placebo was administered approximately 1 h before and 1 h after angiography and continued twice daily for 4 days, for a total of 10 doses.
The trial was funded by the Veterans Affairs Cooperative Studies Program and the National Health and Medical Research Council of Australia.
Inclusion and exclusion criteria
The PRESERVE trial included patients undergoing coronary or noncoronary angiography who were able to provide informed consent and who had baseline estimated glomerular filtration rates (eGFRs) of 15 to 44.9 ml/min/1.73 m2 body surface area or both an eGFR of 45 to 59.9 ml/min/1.73 m2 and diabetes. The eGFR used for screening patients was calculated using the MDRD (Modification of Diet in Renal Disease) equation, on the basis of the most recent serum creatinine value obtained as part of routine clinical care within 30 days before angiography. We excluded patients undergoing emergency angiography (e.g., those with ST-segment elevation myocardial infarction); those with unstable baseline levels of serum creatinine (≥25% variation within 3 days prior to angiography); decompensated heart failure requiring IV inotropic support, ultrafiltration, or intra-aortic balloon pump; and patients on dialysis or with baseline eGFRs <15 ml/min/1.73 m2. We also excluded patients with recent exposure to acetylcysteine or contrast media, those with allergies to acetylcysteine or contrast media, pregnant patients, and those participating in another interventional trial for which dual enrollment was not approved by the Veterans Affairs study sponsor. The present analysis is based on the subgroup of PRESERVE patients who underwent coronary angiography with PCI.
The primary endpoint for the parent trial and the current subgroup analysis was a composite of death, need for dialysis, or a persistent increase of at least 50% in serum creatinine at 90 days after angiography. CAAKI, defined as an increase in serum creatinine of at least 25% or 0.5 mg/dl at 4 days following angiography, was a secondary endpoint.
To compare demographic, clinical, and procedural variables between patients who did and did not undergo PCI and between the sodium bicarbonate and sodium chloride groups and acetylcysteine and placebo groups among patients who underwent PCI, we used the Student's t-test for normally distributed continuous data, the Wilcoxon rank sum test for skewed continuous data, and the chi-square test for categorical variables. To assess whether the effect of treatment groups on the primary endpoint varied according to whether patients underwent PCI, we extended the multivariate logistic regression model used in the main trial (14) to test for the interaction between PCI and treatment assignment. Within the subgroup of patients who underwent PCI, we also conducted exploratory analyses to detect heterogeneity in the treatment effect of sodium bicarbonate compared with sodium chloride and acetylcysteine compared with placebo among pre-specified subgroups of interest: according to eGFR stratum, the presence of diabetes, albuminuria stratum, contrast volume, and country (United States vs. Australia, New Zealand, and Malaysia). All p values are 2-sided. SAS version 9.4 (SAS Institute, Cary, North Carolina) was used to conduct statistical analyses.
In the parent trial, a total of 4,993 patients were included in the final modified intention-to-treat analytic cohort, of whom 4,466 (89%) underwent coronary angiography. Of these 4,466 patients, 1,161 (26%) underwent PCI. Compared with the 3,305 patients who underwent coronary angiography without PCI, patients who underwent PCI received a higher median volume of contrast (160 ml vs. 75 ml; p < 0.001), with no other significant differences in clinical or procedural characteristics between these 2 groups (Table 1). These 1,161 patients who underwent PCI were randomly assigned as part of the parent trial to receive IV sodium bicarbonate (n = 568) or IV sodium chloride (n = 593) and acetylcysteine (n = 598) or placebo (n = 563) (Figure 1).
Baseline and procedural characteristics
The mean age of the 1,161 patients who underwent PCI was 69 ± 8 years, 952 (82%) had diabetes mellitus, and the median pre-angiography eGFR was 50 ml/min/1.73 m2 (interquartile range: 41 to 60 ml/min/1.73 m2). Overall, 1,056 PCI procedures (90%) were performed at the same time as the diagnostic coronary angiographic study, while 105 (10%) were performed as staged procedures. The median contrast volume administered was 160 ml (interquartile range: 115 to 220 ml), and iso-osmolal iodixanol (Visipaque, GE Healthcare, Little Chalfont, United Kingdom) was used in 646 patients (55%). Access site was transfemoral in 67% and radial in the remaining 33% of PCI cases. The mean left ventricular end-diastolic pressure was 17 ± 7 mm Hg. There were no significant differences in baseline clinical or procedural characteristics between the IV sodium bicarbonate and IV sodium chloride groups or between the acetylcysteine and placebo groups (Table 2). The median volume of trial IV fluid administered was 350 ml (interquartile range: 281 to 452 ml) prior to PCI, 160 ml (interquartile range: 110 to 220 ml) during PCI, and 590 ml (interquartile range: 490 to 687 ml) after PCI, with no differences in the total volume or duration of IV fluids across the comparator groups (Table 2).
Effect of study interventions
The primary composite endpoint occurred in 15 patients (2.6%) in the sodium bicarbonate group compared with 24 patients (4.0%) in the sodium chloride group (odds ratio: 0.64; 95% confidence interval [CI]: 0.33 to 1.24; p for interaction = 0.41) and in 23 patients (3.8%) in the acetylcysteine group compared with 16 patients (2.8%) in the placebo group (odds ratio: 1.37; 95% CI: 0.71 to 2.62; p for interaction = 0.29) (Figure 2, Table 3). The interaction between sodium bicarbonate and acetylcysteine with respect to the primary endpoint was not significant in the parent trial (p = 0.33).
CAAKI occurred in 64 patients (11.3%) in the sodium bicarbonate group and 71 patients (12.0%) in the sodium chloride group (odds ratio: 0.93; 95% CI: 0.65 to 1.34; p for interaction = 0.16) and in 69 patients (11.5%) in the acetylcysteine group compared with 71 patients (11.7%) in the placebo group (odds ratio: 0.98; 95% CI: 0.69 to 1.41; p for interaction = 0.49) (Figure 2, Table 3). The interaction between sodium bicarbonate and acetylcysteine with respect to the secondary endpoint was not significant in the parent trial (p = 0.46).
Subgroups of interest
There were no significant differences in the primary or secondary endpoints in any of the pre-specified subgroups for the comparison of sodium bicarbonate with sodium chloride or the comparison of acetylcysteine with placebo (Figure 3). In the eGFR stratum <45 ml/min/1.73 m2, there was a statistically nonsignificant lower rate of the primary endpoint with sodium bicarbonate (odds ratio: 0.44; 95% CI: 0.20 to 0.95; p for interaction = 0.07) without adjustment for multiple comparisons.
In this subgroup analysis of the PRESERVE trial among high-risk patients with chronic kidney disease who underwent PCI, we found that neither IV sodium bicarbonate (compared with IV sodium chloride) nor oral acetylcysteine (compared with placebo) was superior in reducing serious adverse 90-day events or CAAKI. These findings have important implications for current clinical practice.
The deleterious effects of iodinated contrast on kidney function are thought to be mediated by impaired renal hemodynamic status and direct toxic effects on tubular epithelial cells (19). There is evidence that reactive oxygen species play a role in the pathophysiology of CAAKI; hence urinary alkalization and the use of scavengers of reactive oxygen species have been the subject of intense preclinical and clinical research (20,21). Multiple previous trials and meta-analyses have compared the efficacy of IV sodium bicarbonate with that of IV sodium chloride and assessed the role of acetylcysteine for the prevention of adverse renal events in patients undergoing PCI, with inconsistent results (7–13). Important limitations of these previous studies include small sample sizes, inclusion of patients with normal or minimally impaired kidney function who were at low risk for developing CAAKI, and use of surrogate biochemical markers, most commonly small changes in serum creatinine, as primary endpoints. For example, in ACT (Acetylcysteine for Contrast-Induced Nephropathy Trial), more than 50% of patients had baseline eGFR values >60 ml/min/1.73 m2, the primary endpoint was a 25% increase in serum creatinine at 48 to 96 h, and a minority of patients (n = 665 [29%]) underwent PCI (9). In contrast, the PRESERVE trial enrolled patients who were at higher risk for renal complications after administration of contrast media (i.e., stage 3A chronic kidney disease with diabetes or stages 3B and 4 chronic kidney disease with or without diabetes) and used a primary endpoint comprising death, dialysis, and a persistent decline in kidney function at 90 days. Moreover, the present subgroup analysis of more than 1,150 patients, with more than 500 in each treatment arm, represents the largest trial of IV sodium bicarbonate and of acetylcysteine in patients undergoing PCI reported to date.
A report on nearly 1 million patients undergoing PCI at 1,253 sites in the United States participating in the National Cardiovascular Data Registry Cath-PCI Registry demonstrated that the in-hospital mortality after PCI was 9.7% for patients who developed CAAKI and 34% in patients who required dialysis, compared with 0.5% for patients without CAAKI (1). Hence, strategies to prevent CAAKI, and reduce its attendant complications, are of great clinical importance. The risk for CAAKI is particularly high in patients who undergo PCI. This is evident in the present study, in which patients who underwent PCI received more than double the median volume of contrast (160 ml vs. 75 ml) and had a 30% higher incidence of CAAKI relative to participants in the PRESERVE trial who underwent coronary angiography without PCI. Unfortunately, even in this high-risk subset of patients, IV sodium bicarbonate was not superior to IV saline, and acetylcysteine was not effective for the prevention of CAAKI or patient-centered clinical events. These results are consistent with the overall trial and provide additional evidence to discontinue these widely used interventions in the cardiac catheterization laboratory.
Current American College of Cardiology, American Heart Association, and Society for Cardiovascular Angiography and Interventions PCI guidelines recommend that patients undergoing cardiac catheterization with contrast media receive adequate preparatory intravascular volume expansion with isotonic crystalloids (6). On the basis of the lack of clinical benefit of IV sodium bicarbonate compared with IV sodium chloride seen in the PRESERVE study, we believe this recommendation should be updated in favor of IV sodium chloride. Furthermore, the present study demonstrates that even in the highest risk patients who are undergoing coronary angiography with PCI, intravascular volume expansion with isotonic sodium chloride should be considered the standard of care for the prevention of adverse renal outcomes. In addition to its lack of superiority compared with IV sodium chloride, the use of IV 1.26% sodium bicarbonate in clinical practice is problematic because it requires formulation by local pharmacies, which introduces the potential for a delay in availability and compounding errors and is incompatible with other cardiac medications such as norepinephrine and dobutamine, which may require a separate IV infusion. In contrast, IV sodium chloride is readily available in hospital settings and does not have such incompatibilities with cardiac medications. These properties could simplify renal protection by ensuring timely administration of IV fluids prior to contrast exposure. Of note, the recently published AMACING (Maastricht Contrast-Induced Nephropathy Guideline) trial challenged the role of prophylactic IV fluid prior to administration of iodinated contrast media. This prospective, single-center, noninferiority randomized clinical trial of 660 patients with chronic kidney disease found that no IV fluid was noninferior to prophylactic IV isotonic saline for the prevention of CAAKI (2.6% vs. 2.7%, p = 0.47) (22). Notwithstanding this finding, AMACING had limited power because of substantial underenrollment relative to the target sample size (i.e., n = 1,300). Moreover, it included many patients who underwent procedures with IV rather than intra-arterial contrast and who had more preserved kidney function (i.e., eGFR 45 to 60 ml/min/1.73 m2) and were hence at much lower risk for kidney injury. Finally, just 14% of patients underwent interventional procedures. Consequently, the results of the AMACING trial are poorly generalizable to high-risk patients undergoing coronary interventions and do not substantiate the withholding of prophylactic IV isotonic fluid in this patient population.
First, the PRESERVE trial excluded some very high-risk patients (i.e., those with ST-segment elevation myocardial infarction and/or cardiogenic shock). Hence, our findings may not be generalizable to these patients. Second, the power of the present analysis was limited because of the smaller number of patients compared with the overall PRESERVE trial; however, our results are fully consistent with those of the overall trial, supporting the study conclusions. Finally, the study cohort was predominantly male; caution is required when extrapolating results to female patients.
Among patients with chronic kidney disease undergoing PCI, there was no benefit of IV sodium bicarbonate compared with IV sodium chloride or of oral acetylcysteine for the prevention of death, need for dialysis, or persistent decline in kidney function at 90 days or for the prevention of CAAKI. These results are consistent with the overall trial and support a discontinuation in the use of these interventions in the catheterization laboratory.
WHAT IS KNOWN? Periprocedural administration of acetylcysteine and IV sodium bicarbonate is commonly used to prevent CAAKI in clinical practice.
WHAT IS NEW? In the present subgroup analysis of more than 1,150 patients with chronic kidney disease undergoing PCI, we found that neither IV sodium bicarbonate (compared with IV sodium chloride) nor oral acetylcysteine (compared with placebo) was superior in reducing serious adverse 90-day events or CAAKI.
WHAT IS NEXT? Our results support a discontinuation in the use of these interventions in the catheterization laboratory. Intravascular volume expansion with isotonic sodium chloride should be considered the standard of care for the prevention of adverse renal outcomes.
The contents of this article do not represent the views of the U.S. Department of Veterans Affairs or U.S. government. This trial was funded by the VA Cooperative Studies Program and the National Health and Medical Research Council of Australia. Dr. Garcia has received grant support from Edwards Lifesciences; and consulting fees from Medtronic, Boston Scientific, Osprey Medical, and Surmodics. Dr. Bhatt has received grant support from Amarin, AstraZeneca, Bristol-Myers Squibb, Eisai, Ethicon, Medtronic, Sanofi, The Medicines Company, Roche, Pfizer, Forest Laboratories/AstraZeneca, Ischemix, Amgen, Lilly, Chiesi, and Ironwood; has participated in an unfunded research collaboration with FlowCo, PLx Pharma, Takeda, and Merck; has received fees for serving on data monitoring committees and the operations committee, publications committee, and steering committee of the Population Health Research Institute; has received fees for serving as editor-in-chief of the Harvard Heart Letter from Belvoir Publications; has received fees for serving as chief medical editor of Cardiology Today’s Intervention from Slack Publications; has received fees for serving on steering committees for continuing medical education from WebMD; has received advisory board fees from Elsevier; has served on advisory boards for Medscape Cardiology, Regado Biosciences, and Cardax; has received fees for serving as editor-in-chief of the Journal of Invasive Cardiology from HMP Communications; has served as deputy editor for Clinical Cardiology; has received fees for serving as guest editor and associate editor for the Journal of the American College of Cardiology; and has served as site coinvestigator for St. Jude Medical, Biotronik, and Boston Scientific. Dr. Gallagher has served on a study steering committee for Baxter Australia. Dr. Palevsky has received consulting fees and advisory committee fees from Durect; consulting fees from HealthSpan Dx; fees for serving as a member of a data and safety monitoring board from Baxter; fees for serving as a member of an endpoint adjudication committee from GE Healthcare; and consulting fees and advisory fees from Novartis. Dr. Weisbord has received consulting fees and advisory fees from Durect. All other authors have reported that they have no relationships relevant to the contents of this article to disclose.
- Abbreviations and Acronyms
- contrast-associated acute kidney injury
- confidence interval
- chronic kidney disease
- estimated glomerular filtration rate
- percutaneous coronary intervention
- Received April 13, 2018.
- Revision received July 11, 2018.
- Accepted July 17, 2018.
- Tsai T.T.,
- Patel U.D.,
- Chang T.I.,
- et al.
- Gruberg L.,
- Mintz G.S.,
- Mehran R.,
- et al.
- Chertow G.M.,
- Burdick E.,
- Honour M.,
- Bonventre J.V.,
- Bates D.W.
- Levine G.N.,
- Bates E.R.,
- Blankenship J.C.,
- et al.
- ACT Investigators
- Zhang B.,
- Liang L.,
- Chen W.,
- Liang C.,
- Zhang S.
- Kang X.,
- Hu D.Y.,
- Li C.B.,
- Ai Z.S.,
- Peng A.
- Weisbord S.D.,
- Gallagher M.,
- Kaufman J.,
- et al.
- Weisbord S.D.,
- Gallagher M.,
- Jneid H.,
- et al.
- Gurm H.S.,
- Dixon S.R.,
- Smith D.E.,
- et al.
- Recio-Mayoral A.,
- Chaparro M.,
- Prado B.,
- et al.
- Solomon R.
- Nijssen E.C.,
- Rennenberg R.J.,
- Nelemans P.J.,
- et al.