Methods

The Contrast Media and Nephrotoxicity Following Coronary Revascularization by Angioplasty for Acute Myocardial Infarction (CONTRAST-AMI) study was a prospective, multicenter, randomized, single-blind, parallel-group, noninferiority comparison of the incidence of CI-AKI and optimal tissue-level perfusion between iodixanol and iopromide in patients with STEMI who underwent primary PCI. The study was approved by the institutional ethics committees of each participating center and performed according to Good Clinical Practice standards and the principles of the Declaration of Helsinki and its subsequent amendments. Written informed consent was provided by all patients before enrollment in the study.

From January 2009 to April 2010, all consecutive patients admitted to the participating centers for STEMI who underwent primary PCI <12 hours (18 hours in case of cardiogenic shock) after the onset of symptoms were considered eligible. No upper age limit was used. Criteria for exclusion were pregnancy, lactation, administration of any investigational drug within the previous 30 days, intra-arterial or intravenous administration of iodinated contrast media from 7 days before to 72 hours after the administration of the study agents, intake of nephrotoxic medications from 24 hours before to 24 hours after the administration of the study agents, previous participation in this study, and inability to give informed consent to participate in the study.

Patients were randomized to receive iodixanol or iopromide as part of their catheterization procedures according to a computer-generated balanced randomization scheme that was provided to the study centers. Study agent was administered by intracoronary injection as necessary for each patient, and the total contrast medium volume administered was recorded. Because N-acetylcysteine may prevent CI-AKI with a dose-dependent effect in patients treated with primary PCI, improving hospital outcomes, it was used in all patients (1,200 mg intravenous diluted with 100 ml 5% glucose during the procedure and 1,200 mg orally twice daily for the next 48 hours after PCI). Immediately after intervention, all patients underwent hydration with intravenous isotonic saline (0.9%) at a rate of 1 ml/kg/hour for 12 hours (or 0.5 ml/kg/hour for 12 hours in cases of overt heart failure).

The decision to use intra-aortic balloon pumps, inotropic drugs, abciximab, β blockers, angiotensin-converting enzyme inhibitors, and diuretics was left to the discretion of interventional and coronary care unit cardiologists, as directed by international guidelines. Angiographic markers of epicardial flow and tissue-level perfusion were assessed on completion of diagnostic coronary angiography and shortly after PCI. Left ventricular function was evaluated by echocardiography in all patients <24 hours after admission and at discharge. Investigators reading angiograms and echocardiograms were blinded to the contrast medium randomization. All groups and individuals associated with the study remained blinded until the database was locked and the data analyzed.

Serum creatinine (sCr) was measured at baseline; on days 1, 2, and 3; and at discharge. Creatinine clearance was calculated by applying the Cockcroft-Gault formula to the sCr value.

Coronary angiography and PCI were performed according to standard clinical practice. Patients in the emergency department received a bolus of 5,000 U of unfractionated heparin, followed by additional intraprocedural boluses to maintain an activated clotting time of 200 to 250 seconds (≥300 seconds when abciximab was not used), a bolus of acetylsalicylic acid (500 mg intravenously or 300 mg orally), and an oral clopidogrel loading dose of 300 mg (600 mg if abciximab was not administered). The use of glycoprotein IIb/IIIa inhibitors as well as bare-metal or drug-eluting stents was left to the discretion of the interventionalist. Additional use of thrombectomy was recommended depending on thrombus in the infarct-related artery. All coronary angiograms were evaluated by 2 experienced readers who were unaware of the clinical status and type of the contrast medium used. Flow in the epicardial arteries was assessed before and after coronary intervention for Thrombolysis In Myocardial Infarction (TIMI) flow grade and corrected TIMI frame count by use of previously described methods. The TIMI myocardial perfusion grade (TMPG) was used to assess myocardial tissue-level perfusion. A “closed” microvasculature is defined as TMPG 0 or 1, with TMPG 2 or 3 representative of an “open” microvasculature. TMPG was assessed only in the area supplied by the culprit vessel.

The primary end point was defined as the proportion of patients with relative increases in sCr ≥25% from baseline to 72 hours after study agent administration. The main secondary end point was the proportion of patients with TMPG 2 or 3 in the area supplied by the culprit vessel. Other secondary end points were a relative increase in sCr ≥50%; an absolute increase in sCr ≥0.5 or ≥1 mg/dl; the proportion of patients who required specific treatment for acute renal failure, who required dialysis, or who died of acute renal failure at 1 month; and the proportion of patients with major adverse cardiac events, defined as cardiac death, reinfarction, and rehospitalization for heart failure, at 1 month.

Data are expressed as mean ± SD for continuous variables and as absolute and relative frequencies for categorical variables. Continuous variables were compared between the 2 groups of patients using Student’s 2-sample t tests or Mann-Whitney U tests and categorical variables using chi-square tests or Fisher’s exact tests as appropriate. The primary end point was the comparison of the proportions of patients with sCr increases ≥25% from baseline to 72 hours after contrast agent application (CI-AKI). Iopromide was concluded to be not inferior to iodixanol if the upper limit of the 95% confidence interval for the difference in proportions with CI-AKI was <8% (prespecified noninferiority margin). In a second step, superiority was tested if noninferiority was shown. Superiority was concluded if the upper limit of the 95% confidence interval was <0%. This approach was repeated for all time points (24 hours, 48 hours, and maximum sCr). In addition, logistic regression analysis was applied to model CI-AKI at 72 hours, including the following risk factors: treatment group, age, gender, baseline sCr, volume of contrast agent, and the left ventricular ejection fraction at baseline and at discharge. Backward selection with the criterion of a 2-sided p value <0.05 was used to select the final model. Receiver-operating characteristic curve analysis was performed for significant factors to evaluate accuracy. Comparisons of creatinine levels and relative and absolute changes from baseline between the treatment groups were performed with baseline as the covariate. In case of unbalanced numbers of patients between the treatment groups with missing data for the primary end point, sensitivity analyses were conducted.

The main objective of this study was to evaluate whether the low-osmolar contrast medium is at least not inferior, or even superior, to the iso-osmolar contrast medium with regard to the prevention of CI-AKI. Given the lack of clear superiority of either agent in terms of the prevention of CI-AKI in previous studies, a 2-step design was chosen for the study. In the first step, noninferiority was tested, and in the second step, the superiority of the low-osmolar contrast medium to the iso-osmolar contrast medium was assessed. On the basis of International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use guidelines, the noninferiority margin aimed to be less than the largest clinically acceptable difference against the comparator and to preserve the effect of the comparator against a putative placebo comparison. To estimate the power and noninferiority margin at the time of study inception, data from the most pertinent related study was used. An 8% rate of CI-AKI was projected in the group receiving the iso-osmolar contrast medium. With a Δ value (or upper limit of the accepted difference) of 8%, the CI-AKI rate would be clinically relevant, if the iso-osmolar contrast medium group showed a rate of 8% and the low-osmolar contrast medium group a rate of <16%. This is still much better than CI-AKI rate without the use of N-acetylcysteine (33%) and not worse than low-osmolar contrast media with single-dose N-acetylcysteine, as shown by Marenzi et al (15%); moreover, the selected margin is lower compared to the variation observed in previous studies between high- and low-osmolar contrast media.

The sample size was thus based on the assumption of an 8% incidence of CI-AKI in the 2 treatment groups and a noninferiority margin of 8% for the difference in proportions between iopromide and iodixanol. With a level of significance of 0.025 (1 sided), enrollment of a total of 423 patients, accounting for a 5% dropout rate, was planned. The primary end point was tested at a 1-sided level of significance of 0.025, whereas all other statistical tests were 2 sided at a level of significance of 0.05. Statistical analyses were performed using SAS version 9.2 (SAS Institute Inc., Cary, North Carolina), and sample size estimation was performed with PASS 2005 (NCSS, Kaysville, Utah).