Choice of Contrast Medium in Patients With ImpairedRenal Function Undergoing Percutaneous

Coronary InterventionRainer Wessely, MD; Tobias Koppara, MD; Christian Bradaric, MD; Marc Vorpahl, MD;Siegmund Braun, MD; Stefanie Schulz, MD; Julinda Mehilli, MD; Albert Schomig, MD;

Adnan Kastrati, MD; for the Contrast Media and Nephrotoxicity Following CoronaryRevascularization by Angioplasty (CONTRAST) Trial investigators

Background—No clinical trial has yet focused on contrast-mediated nephrotoxicity in patients with chronic renal failureexclusively undergoing percutaneous coronary intervention (PCI). Therefore, the aim of this study was to compare theeffect of contemporary contrast media on nephrotoxicity in this high-risk patient population.

Methods and Results—This prospective, randomized, double-blind, comparative clinical trial randomly selected 939patients with chronic renal failure undergoing coronary angiography with potential PCI to receive either the iso-osmolarcontrast medium iodixanol or the low-osmolar contrast medium iomeprol. Of those 939 patients, 615 received diagnosticangiography only and were not included in the primary study analysis, but were followed up in a registry. Three hundredtwenty-four patients underwent PCI, of which one-half received iodixanol or iomeprol, respectively, and were includedin the primary study analysis. The primary end point was the peak increase in S-creatinine during hospitalization for PCI.Maximum increase in S-creatinine after PCI was lower than expected and thus impaired the power of the study. It wasnot significantly different between the 2 contrast groups (0.19�0.40 mg/dL for iodixanol and 0.21�0.34 mg/dL foriomeprol; P�0.53). Albeit contrast media–induced nephropathy rates were lower with iodixanol (22.2% compared with27.8% for iomeprol), this difference was not statistically different (P�0.25). Subgroup analysis suggested a favorableoutcome regarding nephrotoxicity in patients who received higher contrast volumes (�340 mL) in the iodixanol group(Pinteraction�0.016).

Conclusions—Routine use of iso-osmolar contrast medium is not associated with a significant reduction of nephrotoxicitycompared with low-osmolar contrast medium in patients with chronic renal failure undergoing PCI. However, a positiveeffect was seen in the iso-osmolar contrast group for patients receiving high amounts of contrast medium, which awaitsconfirmation of a specifically designed randomized clinical trial.

Clinical Trial Registration—clinicaltrials.gov Identifier: NCT00390585 (Circ Cardiovasc Intervent. 2009;2:430-437.)

Key Words: angioplasty � contrast media � kidney � chronic renal failure � contrast-induced nephropathy

Chronic kidney disease (CKD) carries an increased risk ofdeath, cardiovascular events, and hospitalization.1 Even

mild CKD is associated with worsened outcomes after myo-cardial infarction.2 Compared with the general population,patients with chronic renal failure are more likely to developcoronary artery disease (CAD)3 and exhibit an advanceddisease stage at the time of primary diagnosis or coronaryrevascularization.4 Heart failure is associated with chronicallyimpaired renal function irrespective of the presence of CAD.5

As is the case with coronary artery bypass grafting,6 the riskof percutaneous coronary intervention (PCI) is increased inpatients with renal dysfunction and associated with impaired

acute7 and long-term outcomes8 compared with the generalpopulation.

Clinical Perspective on p 437

Parenteral administration of iodized contrast medium iscompulsory for PCI. Because patients with CKD reveal ahigher frequency of diffuse disease and an increased rate ofmultivessel disease, applied contrast volumes are generallyhigher compared with regular patients.9

Although the incidence of contrast media–induced ne-phropathy (CIN) in patients with preserved renal function isvery low,10 the risk for patients with risk factors such as

Received April 23, 2009; accepted August 7, 2009.From the Deutsches Herzzentrum and 1. Medizinische Klinik, Klinikum rechts der Isar, Technische Universitat, Munich, Germany.Dr Wessely’s current address is Department of Cardiology and Angiology (R.W.), Bethesda-Johanniter-Klinikum, Duisburg, Germany.The online-only Data Supplement is available at http://circinterventions.ahajournals.org/cgi/content/full/CIRCINTERVENTIONS.109.874933/DC1.This work was presented as a late breaking clinical trial at the annual meeting of the American College of Cardiology in Chicago, Ill, 2008.Correspondence to Rainer Wessely, MD, Ev. Bethesda-Johanniter Klinikum, Kreuzacker 1-7, 47228 Duisburg, Germany. E-mail rwessely@web.de© 2009 American Heart Association, Inc.

Circ Cardiovasc Intervent is available at http://circinterventions.ahajournals.org DOI: 10.1161/CIRCINTERVENTIONS.109.874933

430 by guest on October 9, 2016http://circinterventions.ahajournals.org/Downloaded from by guest on October 9, 2016http://circinterventions.ahajournals.org/Downloaded from by guest on October 9, 2016http://circinterventions.ahajournals.org/Downloaded from by guest on October 9, 2016http://circinterventions.ahajournals.org/Downloaded from by guest on October 9, 2016http://circinterventions.ahajournals.org/Downloaded from by guest on October 9, 2016http://circinterventions.ahajournals.org/Downloaded from by guest on October 9, 2016http://circinterventions.ahajournals.org/Downloaded from

chronic renal failure and diabetes is considerably increasedand ranges between approximately 12% and 50%.11 Impor-tantly, it is elevated even in patients with moderate renaldysfunction.12 In addition to allergies, nephrotoxicity is amajor side effect of intravascular administration of contrastmedia and is one of the leading causes of hospital-acquiredrenal failure.13 CIN is associated with increased cost andimpaired acute and long-term outcome. Even mild impair-ment of renal function after PCI is independently associatedwith a significant increase in mortality after 5 years.14 Theneed for temporary or chronic renal replacement therapy as asequel of CIN is even associated with a sharp increase in1-year mortality.15 Therefore, prevention of CIN is warrantedfor improvement of renal and cardiovascular outcomes.16

Risk factors for the development of CIN include diabetesmellitus, urgent versus planned PCI, congestive heart failure,older age, hypertension, hypotension, and, most importantly,CKD.17 Non–patient-related risk factors comprise propertiesof contrast media such as high- versus low osmolality11,18 aswell as the volume of contrast medium delivered duringcoronary angiography.14

Increased severity of CAD in patients with CKD oftenrequires an elevated amount of contrast medium duringcomplex PCI sessions. This fact may even exacerbate thealready increased risk for CIN in patients presenting withchronically impaired renal function.

Besides ionicity, contrast media are categorized accordingto their osmolality as so-called high-osmolar (osmolality�1000 mosm/kg), low-osmolar (600 to 1000 mosm/kg), andiso-osmolar (280 to 290 mosm/kg) contrast media.18,19 Now-adays, high-osmolar contrast media have been replaced bylow-osmolar and iso-osmolar contrast media because ofbetter tolerability, lower side effects, and, importantly, alower incidence of CIN.20,21 It is important to acknowledgethat this classification, albeit commonly used, might beconfusing because the so-called low-osmolar contrast mediaare in fact hyperosmolar compared with blood.

Conflicting data exist about the nephrotoxicity associatedwith the use of low-osmolar versus the use of iso-osmolarcontrast media in patients with impaired renal function.Although some clinical trials and a meta-analysis suggest alower incidence of CIN in patients receiving iso-osmolarcontrast medium,11,12,22 some other prospective studies23,24 donot confirm these findings. Importantly, no prospective ran-domized trial is currently available that compared the effec-tiveness of iso-osmolar to low-osmolar contrast media exclu-sively in patients undergoing PCI who are exposed to highvolumes of contrast medium.

Thus, the aim of the Contrast Media and NephrotoxicityFollowing Coronary Revascularization by Angioplasty(CONTRAST) trial was to compare the nephrotoxic effects ofthe iso-osmolar contrast medium iodixanol with the low-osmolar contrast medium iomeprol in patients with CKDundergoing PCI.

MethodsStudy PatientsPatients with stable or unstable angina pectoris or a positive stresstest undergoing coronary angiography with the possibility of PCI in

a native vessel or bypass graft were eligible for this study if theywere at least 18 years old and had impaired renal function with anestimated glomerular filtration rate (eGFR) of �60 mL/min/1.73 m2

as determined by the Modification of Diet in Renal Disease (MDRD)formula25 or S-creatinine of �1.5 mg/dL, measured within 24 hoursbefore coronary angiography. Criteria for exclusion were pregnancy,lactation, intravascular administration of iodine containing contrastmedium within the previous 7 days, known allergies or incompatibilitiesto contrast media, concurrent intake of nephrotoxic drugs within theprevious 7 days, renal transplantation, cardiogenic shock, end-stagerenal disease necessitating hemodialysis, and inability to give informedconsent to participate in the study. The study protocol was approved byan institutional ethics committee. All patients who were included in thetrial gave written informed consent for participation in this study.

Study ProtocolRandomization was performed with sealed, opaque envelopes con-taining a computer-generated random sequence. Patients were ran-domly assigned to receive 1 of the 2 contrast mediums: the nonionic,iso-osmolar, dimeric contrast medium iodixanol (320 mg of iodineper milliliter; 290 mosm/kg of water [Visipaque, GE HealthcareAmersham Buchler]) or the nonionic, low-osmolar, monomericcontrast medium iomeprol (350 mg of iodine per milliliter, 618mosm/kg of water [Imeron, Altana Pharma]). To ensure blinding,bottles of contrast media identical in volume were placed into anopaque retaining bracket so that invasive cardiologists were notaware of the type of contrast medium. Both contrast media weresupplied and kept at 37°C throughout the procedure. All patientsreceived intravenous hydration before contrast exposure. It wasrecommended to receive 1 mL NaCl 0.9%/kg body weight for 12hours before and after contrast exposure unless clinical or logisticcircumstances did not allow for this. Because of inconsistent resultsconcerning the role of acetylcysteine to prevent CIN, its use was notrecommended.16 Thus, no patient included in the study receivedacetylcysteine or other potential pharmacological agents to impact onthe incidence of CIN.

The primary end point of the study was the peak increase ofS-creatinine as a measure of contrast medium–induced nephrotoxic-ity during hospitalization for PCI. Specifically, S-creatinine wasassessed on a daily basis at least until the value after PCI was equalor less than the baseline value determined before PCI. Secondary endpoints were incidence of CIN (defined by a increase in creatinine�0.5 mg/dL or 25% of the initial value), severe CIN (defined by anincrease of S-creatinine by �1 mg/dL and/or dialysis), and durationof hospitalization. Major adverse cardiac events (death, myocardialinfarction, and target lesion revascularization) were assessed at 6months. The follow-up protocol included a phone interview at 30days and a 6-month visit.

Patients receiving diagnostic coronary angiography only werefollowed up separately for reasons of completeness in a registry.Follow-up of these patients included routine assessment of peakS-creatinine levels during hospitalization as well as a telephoneinterview at 6 months.

Statistical AnalysisThe peak increase of S-creatinine in patients who received low-osmolar contrast media varies considerably in previous studies thatincluded patients undergoing coronary angiography. This is exem-plified in 2 of the most recognized trials in the field. Although theCARE study reported a mean peak increase in S-creatinine of 0.07mg/dL,23 NEPHRIC accounted a mean peak increase in S-creatinineof 0.55 mg/dL.11 Because we expected a higher peak increase thanthe mean value of these numbers in our study that comprised patientswho were expected to receive higher volumes of contrast mediabecause of PCI, the sample size of the CONTRAST trial wascalculated assuming a mean increase in S-creatinine levels of 0.4mg/dL after PCI in the iomeprol group, at least 25% less increasewith iodixanol, a 2-sided � level of 0.05, and 90% power. Baselineand sequential S-creatinine measurements were determined in theidentical laboratory for clinical chemistry located in each hospital. Atotal sample size of 286 patients was calculated on the basis of these

Wessely et al Contrast Media–Induced Nephrotoxicity and PCI 431

by guest on October 9, 2016http://circinterventions.ahajournals.org/Downloaded from

assumptions. Because the performance of PCI could not be predictedbefore diagnostic coronary angiography, the design of the studyrequired randomization of a larger number of patients to allow for theinclusion of the needed number of patients who received PCI (Figure1). Because randomization was performed before diagnostic angiog-raphy, several adjustments in the group assignment ratio wereperformed to achieve an equal number of patients with PCI in eachof the study arms.

Baseline descriptive statistics are presented as frequencies andpercentages for categorical variables and means�SD or median(interquartile range) for continuous variables. Intergroup compari-sons were assessed using Student t test (continuous data) and �2 orFisher exact test (where expected cell value was �5) for categoricalvariables. Survival and event-free status were assessed using themethods of Kaplan–Meier and the log-rank test. Statistical softwareS-PLUS, version 4.5 (S-PLUS, Insightful Corp, Seattle, Wash) wasused for all analyses.

ResultsFrom June 2006 to December 2007, a total of 324 patientswere enrolled in this trial: 162 patients received iodixanol and162 patients received iomeprol (Figure 1). Baseline charac-teristics, as depicted in Table 1, were equally distributedbetween the 2 groups. Compared with typical PCI popula-tions, patients with CKD included in the CONTRAST studyreflected a more complex study population with advancedage, a relatively high frequency of diabetes, and a highincidence of concomitant atherosclerotic risk factors such ashypertension and hyperlipidemia.

The high cardiac risk profile was reflected by the complex-ity of CAD with 91% of the patients who underwent PCIhaving multivessel disease and 78% of patients presentingwith complex lesions. Multiple lesions were treated in �40%of the cases (Table 2). Procedural characteristics for PCIpatients were similar and not statistically different in bothcontrast groups as shown in Table 3.

Renal function was similar for both contrast groups with aneGFR of 46.4�9.3 mL/min/1.73 m2 in the iodixanol groupand 47.1�9.0 mL/min/1.73 m2 in the iomeprol group,P�0.44 (Table 4). Distribution of eGFR is depicted in Figure2. Average hydration time before exposure to contrast me-dium was 6.3 hours for patients receiving iodixanol and 6.5hours for patients receiving iomeprol (P�0.59). Mean con-trast volume delivered during the entire procedure was notsignificantly different between the 2 groups (iodixanol,366�158 mL; iomeprol, 367�170 mL; P�0.93).

The mean change of S-creatinine post-PCI was not signif-icantly different between the groups and amounted to

0.19�0.40 mg/dL in the iodixanol group and 0.21�0.34mg/dL in the iomeprol group (P�0.53). This translated to a12.6�25.8% and 16.1�24.7% increase (P�0.20), respec-tively, compared with the index value. Both the rates of CIN(iodixanol, 22.2%; iomeprol, 27.7%; P�0.25) and severeCIN (iodixanol, 6.2%; iomeprol, 3.7%; P�0.30) and the needfor postinterventional hemodialysis were not significantlydifferent between the groups (Table 4; Figure 3). Thistranslated into a similar duration of hospitalization for PCI(iodixanol, 6.3�4.9 days; iomeprol, 6.5�4.4 days; P�0.59).

Figure 4 (A and B) illustrates the impact of iodixanol onthe primary end point in the prespecified PCI subgroupsdefined by the presence of diabetes, baseline renal functionand quantity of contrast medium used during the procedure.No significant interaction was found between the type ofcontrast medium and diabetes (Pinteraction�0.12) or eGFR(Pinteraction�0.14), respectively. Contrarily, there was a sig-nificant interaction between the type and quantity of contrastmedium used during the procedure (Pinteraction�0.016) in thesense that iodixanol was more favorable in terms of preven-tion of nephrotoxicity in patients who received higheramounts of contrast media (�340 mL) (Figure 4A).

Clinical follow-up was available for 99.4% of the studypopulation at 6 months (Table 5). No significant differenceswere detected between the 2 groups regarding 6-monthadverse outcomes. The curves of the composite major adverse

Figure 1. Study flow chart.

Table 1. Baseline and Clinical Characteristics of StudyPatients

CharacteristicIodixanol Group

(n�162)Iomeprol Group

(n�162) P

Age, y 75.0�9.1 73.2�9.1 0.08

Men 119 (73) 116 (72) 0.71

Diabetes mellitus 60 (37) 61 (38) 0.91

IDDM 23 (14) 22 (14) 0.87

History of smoking 61 (38) 49 (30) 0.16

Arterial hypertension 152 (94) 158 (98) 0.10

Hypercholesterolemia 153 (94) 146 (90) 0.15

Previous myocardial infarction 48 (30) 53 (33) 0.55

Previous aortocoronary bypasssurgery

42 (26) 31 (19) 0.14

Stable angina or silentischemia

131 (81) 141 (87) 0.24

NSTE ACS 22 (14) 16 (10) 0.24

ST-elevation myocardialinfarction

9 (6) 5 (3) 0.27

Ejection fraction, % 51.4�12.0 49.3�9.0 0.44

Medication at admission

ACE inhibitor 106 (65) 106 (65) 1.0

AT-1 blocker 29 (18) 36 (22) 0.33

�-blocker 125 (77) 139 (86) 0.05

Calcium channel blocker 24 (15) 28 (17) 0.54

Statin, n (%) 116 (72) 114 (70) 0.81

Data are presented as means�SD or n (%). IDDM indicates insulin-dependent diabetes mellitus; NSTE ACS, non–ST elevation acute coronarysyndrome; ACE, angiotensin-converting enzyme; AT angiotensin.

432 Circ Cardiovasc Intervent October 2009

by guest on October 9, 2016http://circinterventions.ahajournals.org/Downloaded from

cardiac event rates (death, myocardial infarction, or targetlesion revascularization) are shown in Figure 5. The trend infavor of iodixanol is mainly due to a lower rate of targetlesion revascularization in this group (5 patients versus 10patients in the iomeprol group).

For reasons of completeness, we also followed up patientsundergoing coronary angiography only in a supplementaryregistry. This registry included the 651 patients who receivedsolitary diagnostic coronary angiography (iodixanol, n�315;iomeprol, n�336; respectively). The data regarding thesepatients are provided online only (supplemental Tables Ithrough IV).

Baseline and clinical characteristics for these patients arerevealed in supplemental Table I. Compared with patientswho received PCI, those merely undergoing diagnostic cor-onary angiography had a less severe pattern of CAD (supple-mental Table II) as exemplified by a lower incidence of3-vessel disease (49% in the iodixanol group and 44% in theiomeprol group for patients who did not undergo PCI versus67% and 68%, respectively, for patients who received PCI).

In addition, patients who received solely coronary angiogra-phy had a marginally better renal function compared with PCIpatients with no significant difference between the iodixanolgroup (eGFR�49.2�13.0 mL/min/1.73 m2) and the iomeprolgroup (eGFR�49.1�10.9 mL/min/1.73 m2), P�0.84 (sup-plemental Table III). Contrast volumes were similar inpatients who received diagnostic coronary angiography only(iodixanol, 146�58 mL; iomeprol, 144�53 mL; P�0.57).Compared with PCI patients, those undergoing diagnosticcoronary angiography only displayed a lower incidence ofmaximal peak increase in S-creatinine after contrast exposure(iodixanol, 0.08�0.29 mg/dL; iomeprol, 0.06�0.27 mg/dL;P�0.47), CIN (iodixanol, 9.2%; iomeprol, 8.0%; P�0.59) aswell as severe CIN (iodixanol, 1.0%; iomeprol, 1.2%;P�0.77). The 6-month rate for myocardial infarction anddeath was similar in both groups (supplemental Table IV).

DiscussionThe CONTRAST study is the first trial that was conducted toexclusively assess the outcomes of patients with CKD under-going PCI with regard to contrast-mediated nephrotoxicity.The presence of chronic renal failure, its associated cardio-vascular risk factors, advanced CAD, and the PCI procedureitself put these patients at increased risk for contrast-mediatednephrotoxicity.17 On the other hand, the prevention of CINafter PCI has been shown to be associated with favorablein-hospital and long-term survival rates.14 Thus, there is ahigh clinical need to decrease the incidence of CIN.

Although there is general consensus that low-osmolarcontrast media are superior to high-osmolar contrast media, inparticular, in terms of nephrotoxicity,11,18 conflicting datahave been reported in clinical trials comparing low-osmolarwith iso-osmolar contrast media for prevention of nephrotox-icity in mixed patient populations receiving angiography butonly partially undergoing PCI.

Table 2. Baseline Angiographic Characteristics

CharacteristicIodixanol Group

(n�162)Iomeprol Group

(n�162) P

Diseased vessels 0.31

1-vessel disease 11 (7) 19 (12)

2-vessel disease 35 (22) 33 (20)

3-vessel disease 116 (72) 110 (68)

Target vessel 0.20

Bypass graft 16 (6) 6 (2)

Left main coronary artery 18 (7) 13 (5)

Left anterior descendingcoronary artery

93 (34) 83 (33)

Left circumflex coronary artery 78 (29) 73 (29)

Right coronary artery 65 (24) 75 (30)

Complex (type B2/C) lesions 210 (78) 194 (78) 0.96

No. lesions treated 1.7�0.9 1.6�0.8 0.19

Lesions treated �1 70 (43) 66 (41) 0.65

Chronic occlusions 14 (3) 24 (15) 0.05

Vessel size, mm 2.80�0.59 2.83�0.59 0.58

Lesion length, mm 14.9�9.9 14.0�8.7 0.28

Minimal lumen diameter beforeprocedure, mm

0.98�0.43 1.00�0.53 0.76

Diameter stenosis beforeprocedure, %

65.1�13.1 65.2�16.0 0.96

Data are presented as n (%) or mean�SD.

Table 3. Procedural Characteristics

CharacteristicIodixanol Group

(n�162)Iomeprol Group

(n�162) P

Maximal balloon pressure, atm 14.5�2.5 15.0�2.5 0.36

Balloon-to-vessel ratio 1.06�0.18 1.05�0.11 0.49

Length of stented segment, mm 25.4�12.3 24.5�12.1 0.41

No. stents per lesion 1.26�0.55 1.24�0.54 0.63

Data are presented as mean�SD.

Table 4. Pre- and Postprocedural Renal Function

CharacteristicIodixanol Group

(n�162)Iomeprol Group

(n�162) P

S-creatinine before PCI, mg/dL 1.36�0.51 1.37�0.33 0.74

S-Urea before PCI, mg/dL 55.6�23.7 57.8�27.4 0.44

eGFR, mL/min 46.4�9.3 47.1�9.0 0.44

Volume contrast mediadelivered, mL

366�158 367�170 0.93

Maximal S-creatinine after PCI,mg/dL

1.55�0.58 1.59�0.48 0.51

Maximal S-urea after PCI,mg/dL

59.4�30.8 61.2�30.0 0.59

Maximal rise in S-creatinine,mg/dL

0.19�0.40 0.21�0.34 0.53

Maximal rise in S-creatinine,% of prior PCI

12.6�25.8 16.1�24.7 0.20

Maximal rise in S-urea, mg/dL 3.7�20.5 3.4�18.9 0.88

Hemodialysis after PCI 3 (1.9) 1 (0.6) 0.31

CIN 36 (22) 45 (28) 0.25

Rise in S-creatinine �1 mg/dl 10 (6.2) 6 (3.7) 0.30

Data are presented as mean�SD or n (%).

Wessely et al Contrast Media–Induced Nephrotoxicity and PCI 433

by guest on October 9, 2016http://circinterventions.ahajournals.org/Downloaded from

Overall, we found no significant difference between themean peak increase of S-creatinine and CIN after PCIbetween the iso-osmolar iodixanol group and the low-osmolar iomeprol group. As expected, patients receivingdiagnostic coronary angiography only who were followed upoutside the clinical trial showed considerably lower levels ofnephrotoxicity, emphasizing the fact that the amount ofcontrast medium and the PCI procedure itself have a majorimpact on the increased rate of nephrotoxicity that wasobserved in the study population.

Patients randomly assigned to the CONTRAST study canbe considered a high-risk population in terms of renal andcardiovascular outcome. Because laboratory assessment ofrenal function by measurement of S-creatinine and eGFR wasmandatory before study enrollment, most patients had stableCAD. The high-risk profile of CONTRAST study patientsreceiving PCI is exemplarily reflected by advanced age withan average of 74.1 years, impaired ejection fraction with amean of 50.4%, and a high incidence of multivessel diseasethat affected �9 of 10 patients.

The high incidence of multivessel disease translated into aneed for increased contrast volumes during PCI. The averagecontrast volume delivered for diagnostic angiography imme-diately following coronary intervention was 367 mL, which isin good accordance with the COURT trial that studied theimpact of different contrast media on the incidence of majoradverse cardiac events after PCI.9 Other trials that included a

significant portion of patients solely undergoing coronaryangiography without PCI reported the use of lower amountsof contrast media, such as in NEPHRIC11 (�163 mL),RECOVER22 (�200 mL), and CARE23 (�135 mL).

Because CONTRAST is a unique clinical trial regardingstudy population that comprised patients undergoing PCIonly, its outcome cannot be readily compared with that ofother trials. The incidence of CIN varies significantly be-tween numerous studies. Despite lower amounts of contrastmedia delivered, most trials reported CIN rates for low-osmolar contrast media comparable with those observed inthe CONTRAST trial;11,18,22,26–29 however, some studies havereported a lower incidence of CIN.23,30 Similarly, a CIN rate

Figure 2. Distribution of eGFR (mL/min/1.73m2) in the study population.

Figure 3. CIN associated with the use of iodixanol (black bars)or iomeprol (gray bars). The incidence of CIN as defined as aincrease of S-creatinine of �0.5 mg/dL or �25% of the valuebefore contrast exposure is shown on the left. The bars in thecenter display the incidence of severe CIN as defined as aincrease of S-creatinine of �1 mg/dL. The bars on the rightreveal the rate of dialysis that was required subsequent to PCI.

-0.4 -0.2 0.0 0.2 0.4

B

no diabetes

diabetes

NIDDM

IDDM

n=203

n=121

n=76

n=45

-0.4 -0.2 0.0 0.2 0.4

eGFR

≤ 50

> 50

contrastvolume

≤ 340

> 340

A

Δ S-creatinine

Iodixanol better Iomperol better

Figure 4. A, Subgroup analysis of the study population regard-ing nephrotoxicity associated with the grade of impairment ofrenal function and amount of contrast exposure. B, Comparisonof outcome according to the presence of diabetes. eGFR (mL/min/1.73 m2), amount of contrast volume delivered during coro-nary angiography, and PCI (mL). NIDDM, non–insulin-dependentdiabetes mellitus; IDDM, insulin-dependent diabetes mellitus.

434 Circ Cardiovasc Intervent October 2009

by guest on October 9, 2016http://circinterventions.ahajournals.org/Downloaded from

in the 2-digit percentage numbers was described for iodixanolin the ICON study31 with 16.2%; however, the CIN rate forthe identical contrast medium was only 3% in the NEPHRICtrial,11 which included exclusively patients with chronic renalfailure and concomitant diabetes undergoing coronary an-giography, but of which only �20% underwent PCI.

Overall, CIN rates reported in the CONTRAST trial seemhigher than in most other trials that assessed the nephrotox-icity associated with the use of contrast media in cardiovas-cular medicine. This can be readily explained by both thehigh-risk study population and the fact that all patients in thistrial underwent PCI that is associated with considerablyhigher volumes of contrast than mere diagnostic coronaryangiography. Presence of chronic renal failure and concom-itant PCI increase the risk for contrast-mediated nephrotox-icity considerably.17,32,33 Heterogeneity in study populationand delivered contrast volumes are most likely to accountprimarily for the divergent results among clinical contrasttrials available so far that did not require obligatory PCI forinclusion into the clinical study.

There is general consensus that intravenous hydration hasan important role for CIN prevention.16 However, no addi-tional pharmacological therapy is consistently associated withimproved outcomes after contrast exposure. Therefore, theCONTRAST study protocol required intravenous hydrationbefore and after contrast exposure but discouraged the use ofdrugs, such as acetylcysteine, fenoldopam, theophylline, tostrengthen clinical results. Thus, none of these drugs wasgiven to any of patients of this trial. In addition, nephrotoxicdrugs, such as metformin, aminoglycosides, or nonsteroidalanti-inflammatory drugs, had to be stopped at least 7 daysbefore cardiac catheterization.

In terms of cardiovascular end points, 6-month follow-uprevealed no differences regarding death and myocardialinfarction. However, there was a weak trend toward a lowerneed for target vessel revascularization in the iodixanol group(3.1% versus 6.2% for iomeprol; P�0.19) and overall majoradverse coronary events (8.6% versus 13.0%; P�0.21). Thisobservation can be regarded as hypothesis generating andneeds to be addressed in a dedicated clinical trial. In thiscontext, it should be acknowledged that the COURT trialreported a lower major adverse coronary event rate at 30 daysin patients who received iodixanol undergoing PCI.9

Formerly, the NEPHRIC study that included exclusivelydiabetic patients with impaired renal function found a signif-icantly lower incidence of nephrotoxicity associated withiodixanol use compared with a low-osmolar contrast medium

after coronary angiography. Therefore, we conducted a pre-specified subgroup analysis for diabetic patients. However, inthe diabetic patient population of the CONTRAST trial, wecould not detect a significant difference between the investi-gated contrast media.

Yet, patients who received higher amounts of contrastmedium than the median of 340 mL revealed less nephrotox-icity when they received iodixanol compared with iomeprol(Pinteraction�0.016). This suggests that the use of an iso-osmolar contrast medium might be a valid option in patientswith impaired renal function who are expected to receiveincreased contrast volumes for PCI.

Given the highly complex patient background, the in-creased contrast volumes needed for diagnostic and therapeu-tic coronary angiography, and the results of previous clinicaltrials that investigated the effect of lower contrast volumes instudy populations with only a fraction undergoing PCI, thedata of the CONTRAST study may hint to a ceiling effectregarding the efficacy of different contrast media for theprevention of contrast medium–related renal dysfunctionafter PCI. This is consistent with subgroup analysis thatsuggests that patients who receive a high amount of contrastvolume may still profit from iso-osmolar contrast media.These findings also implicate that contrast exposure shouldbe limited to a minimum in patients with chronic renal failureundergoing PCI and that physicians should be open to stagedprocedures, although they might be more demanding in termsof patient compliance and costs.

Limitations of the StudyAlthough this is the largest randomized trial examining theimpact of iso-osmolar contrast media compared with low-osmolar contrast media in patients with impaired renalfunction undergoing PCI on nephrotoxicity, the sample sizewas still limited to enable careful evaluation of clinical endpoints. CONTRAST used a common primary end point, theincrease in S-creatinine after exposure to contrast media.Although various studies have shown that there is a linkbetween the increase of S-creatinine as marker of renalfunction and clinical outcome, it should be emphasized thatS-creatinine is still a surrogate marker. In addition, interpret-

Figure 5. Kaplan–Meier curves showing the incidence of majoradverse coronary events during the 6-month follow-up.

Table 5. Clinical 6-Month Follow-Up

CharacteristicIodixanol Group

(n�162)Iomeprol Group

(n�162) P

Target lesion revascularization 5 (3.1) 10 (6.2) 0.19

Stent thrombosis 0 (0) 0 (0) N/A

Myocardial infarction 7 (4.3) 7 (4.3) �0.99

Death 6 (3.7) 7 (4.3) 0.78

MACE 14 (8.6) 21 (13.0) 0.21

Data are presented as n (%). MACE indicates major adverse coronary event.

Wessely et al Contrast Media–Induced Nephrotoxicity and PCI 435

by guest on October 9, 2016http://circinterventions.ahajournals.org/Downloaded from

ing the results of subgroup analyses should be considered ashypothesis generating. Although the trial was carefullyplanned according to the existing literature, the primary endpoint was lower than expected and thus weakened the powerof the trial.

In conclusion, the CONTRAST trial revealed that routineuse of iso-osmolar contrast medium is not associated with asignificantly lower degree of nephrotoxicity compared withthe use of low-osmolar contrast medium. However, certainsubgroups, such as patients who are expected to receive ahigh amount of contrast medium, may benefit from the use ofthe iso-osmolar contrast medium iodixanol. However, thisawaits confirmation by a dedicated clinical trial. Furtherstudies are needed to define the role of divergent contrastmedia in this or other high-risk patient subgroups.

AppendixStudy organization was as follows. Steering committee: A. Schomig(chairman), R. Wessely (principal investigator), and A. Kastrati.Participating centers: Deutsches Herzzentrum (Munich, Germany)and 1. Medizinische Klinik rechts der Isar, Technische Universitat(Munich, Germany). Data coordinating center: J. Mehilli (director),T. Koppara, C. Bradaric, M. Vorpahl, S. Schulz, H. Holle, F.Maimer-Rodrigues, A. Fabry, K. Hosl, B. Gließl, E. Turk, M.Dirlewanger, N. Sargon, and N. Rifatov. Clinical event adjudicationcommittee: J. Pache (chairman), N.V. Beckerath, C. Markwardt, andH. Bollwein. Angiographic core laboratory: R.A. Byrne, R. Iijima,O. Bruskina, S. Piniek, S. Hurt, and A. Bergbauer.

AcknowledgmentsThe authors thank the staffs of the cardiac catheterization laborato-ries of the Deutsches Herzzentrum Munchen and the Klinikum rechtsder Isar for their contributions to this study.

Sources of FundingThis work was supported by an independent research grant awardedfrom GE Healthcare Amersham Buchler.

DisclosuresDr Wessely has received lecture fees from GE Healthcare AmershamBuchler. The funding organization had no role in the design andconduct of the study; collection, management, analysis, and inter-pretation of the data; and preparation, review, or approval ofthe manuscript.

References1. Go AS, Chertow GM, Fan D, McCulloch CE, Hsu CY. Chronic kidney

disease and the risks of death, cardiovascular events, and hospitalization.N Engl J Med. 2004;351:1296–1305.

2. Anavekar NS, McMurray JJ, Velazquez EJ, Solomon SD, Kober L,Rouleau JL, White HD, Nordlander R, Maggioni A, Dickstein K, Zelen-kofske S, Leimberger JD, Califf RM, Pfeffer MA. Relation between renaldysfunction and cardiovascular outcomes after myocardial infarction.N Engl J Med. 2004;351:1285–1295.

3. Chen R, Kumar S, Timmis A, Feder G, Yaqoob MM, Hemingway H.Comparison of the relation between renal impairment, angiographiccoronary artery disease, and long-term mortality in women versus men.Am J Cardiol. 2006;97:630–632.

4. Lemos PA, Arampatzis CA, Hoye A, Daemen J, Ong AT, Saia F, van derGiessen WJ, McFadden EP, Sianos G, Smits PC, de Feyter P, Hofma SH,van Domburg RT, Serruys PW. Impact of baseline renal function onmortality after percutaneous coronary intervention with sirolimus-elutingstents or bare metal stents. Am J Cardiol. 2005;95:167–172.

5. Ezekowitz J, McAlister FA, Humphries KH, Norris CM, Tonelli M, GhaliWA, Knudtson ML. The association among renal insufficiency, pharma-cotherapy, and outcomes in 6,427 patients with heart failure and coronaryartery disease. J Am Coll Cardiol. 2004;44:1587–1592.

6. Hannan EL, Wu C, Bennett EV, Carlson RE, Culliford AT, Gold JP,Higgins RS, Isom OW, Smith CR, Jones RH. Risk stratification ofin-hospital mortality for coronary artery bypass graft surgery. J Am CollCardiol. 2006;47:661–668.

7. Osten MD, Ivanov J, Eichhofer J, Seidelin PH, Ross JR, Barolet A,Horlick EM, Ing D, Schwartz L, Mackie K, Dzavik V. Impact of renalinsufficiency on angiographic, procedural, and in-hospital outcomes fol-lowing percutaneous coronary intervention. Am J Cardiol. 2008;101:780–785.

8. Gruberg L, Dangas G, Mehran R, Mintz GS, Kent KM, Pichard AD,Satler LF, Lansky AJ, Stone GW, Leon MB. Clinical outcome followingpercutaneous coronary interventions in patients with chronic renal failure.Catheter Cardiovasc Interv. 2002;55:66–72.

9. Davidson CJ, Laskey WK, Hermiller JB, Harrison JK, Matthai W Jr,Vlietstra RE, Brinker JA, Kereiakes DJ, Muhlestein JB, Lansky A, PopmaJJ, Buchbinder M, Hirshfeld JW Jr. Randomized trial of contrast mediautilization in high-risk PTCA: the COURT trial. Circulation. 2000;101:2172–2177.

10. Lasser EC, Lyon SG, Berry CC. Reports on contrast media reactions:analysis of data from reports to the U.S. Food and Drug Administration.Radiology. 1997;203:605–610.

11. Aspelin P, Aubry P, Fransson S-G, Strasser R, Willenbrock R, Berg KJ;the NEPHRIC Study Investigators. Nephrotoxic effects in high-riskpatients undergoing angiography. N Engl J Med. 2003;348:491–499.

12. McCullough PA, Bertrand ME, Brinker JA, Stacul F. A meta-analysis ofthe renal safety of isosmolar iodixanol compared with low-osmolarcontrast media. J Am Coll Cardiol. 2006;48:692–699.

13. Nash K, Hafeez A, Hou S. Hospital-acquired renal insufficiency. Am JKidney Dis. 2002;39:930–936.

14. 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 percuta-neous coronary intervention. Circulation. 2002;105:2259–2264.

15. McCullough PA, Wolyn R, Rocher LL, Levin RN, O’Neill WW. Acuterenal failure after coronary intervention: incidence, risk factors, andrelationship to mortality. Am J Med. 1997;103:368–375.

16. Pannu N, Wiebe N, Tonelli M. Prophylaxis strategies for contrast-inducednephropathy. JAMA. 2006;295:2765–2779.

17. Mehran R, Aymong ED, Nikolsky E, Lasic Z, Iakovou I, Fahy M, MintzGS, Lansky AJ, Moses JW, Stone GW. A simple risk score for predictionof contrast-induced nephropathy after percutaneous coronary inter-vention: development and initial validation. J Am Coll Cardiol. 2004;44:1393–1399.

18. Barrett BJ, Parfrey PS, Vavasour HM, McDonald J, Kent G, Hefferton D,O’Dea F, Stone E, Reddy R, McManamon PJ. Contrast nephropathy inpatients with impaired renal function: high versus low osmolar media.Kidney Int. 1992;41:1274–1279.

19. Parfrey PS, Barrett BJ. High-osmolality and low-osmolality contrastagents. N Engl J Med. 1992;327:204–205.

20. Barrett BJ, Carlisle EJ. Metaanalysis of the relative nephrotoxicity ofhigh- and low-osmolality iodinated contrast media. Radiology. 1993;188:171–178.

21. Barrett BJ, Parfrey PS, Vavasour HM, O’Dea F, Kent G, Stone E. Acomparison of nonionic, low-osmolality radiocontrast agents with ionic,high-osmolality agents during cardiac catheterization. N Engl J Med.1992;326:431–436.

22. Jo SH, Youn TJ, Koo BK, Park JS, Kang HJ, Cho YS, Chung WY, JooGW, Chae IH, Choi DJ, Oh BH, Lee MM, Park YB, Kim HS. Renaltoxicity evaluation and comparison between visipaque (iodixanol) andhexabrix (ioxaglate) in patients with renal insufficiency undergoingcoronary angiography: the RECOVER study: a randomized controlledtrial. J Am Coll Cardiol. 2006;48:924–930.

23. Solomon RJ, Natarajan MK, Doucet S, Sharma SK, Staniloae CS, KatholiRE, Gelormini JL, Labinaz M, Moreyra AE. Cardiac Angiography inRenally Impaired Patients (CARE) study: a randomized double-blind trialof contrast-induced nephropathy in patients with chronic kidney disease.Circulation. 2007;115:3189–3196.

24. Hardiek KJ, Katholi RE, Robbs RS, Katholi CE. Renal effects of contrastmedia in diabetic patients undergoing diagnostic or interventionalcoronary angiography. J Diabetes Complications. 2008;22:171–177.

25. Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D. A moreaccurate method to estimate glomerular filtration rate from serum creat-inine: a new prediction equation. Modification of Diet in Renal DiseaseStudy Group. Ann Intern Med. 1999;130:461–470.

436 Circ Cardiovasc Intervent October 2009

by guest on October 9, 2016http://circinterventions.ahajournals.org/Downloaded from

26. Rudnick MR, Goldfarb S, Wexler L, Ludbrook PA, Murphy MJ, HalpernEF, Hill JA, Winniford M, Cohen MB, VanFossen DB. Nephrotoxicity ofionic and nonionic contrast media in 1196 patients: a randomized trial.The Iohexol Cooperative Study. Kidney Int. 1995;47:254–261.

27. Taliercio CP, Vlietstra RE, Ilstrup DM, Burnett JC, Menke KK, StensrudSL, Holmes DR Jr. A randomized comparison of the nephrotoxicity ofiopamidol and diatrizoate in high risk patients undergoing cardiacangiography. J Am Coll Cardiol. 1991;17:384–390.

28. Manske CL, Sprafka JM, Strony JT, Wang Y. Contrast nephropathy inazotemic diabetic patients undergoing coronary angiography. Am J Med.1990;89:615–620.

29. Wang A, Holcslaw T, Bashore TM, Freed MI, Miller D, Rudnick MR,Szerlip H, Thames MD, Davidson CJ, Shusterman N, Schwab SJ. Exac-erbation of radiocontrast nephrotoxicity by endothelin receptorantagonism. Kidney Int. 2000;57:1675–1680.

30. Briguori C, Airoldi F, D’Andrea D, Bonizzoni E, Morici N, Focaccio A,Michev I, Montorfano M, Carlino M, Cosgrave J, Ricciardelli B,

Colombo A. Renal Insufficiency Following Contrast Media Adminis-tration Trial (REMEDIAL): a randomized comparison of 3 preventivestrategies. Circulation. 2007;115:1211–1217.

31. Mehran R, Nikolsky E, Kirtane AJ, Caixeta A, Wong SC, Teirstein PS,Downey WE, Batchelor WB, Casterella PJ, Kim YH, Fahy M, DangasGD. Ionic low-osmolar versus nonionic iso-osmolar contrast media toobviate worsening nephropathy after angioplasty in chronic renal failurepatients: the ICON (Ionic versus non-ionic Contrast to Obviate worseningNephropathy after angioplasty in chronic renal failure patients) study.J Am Coll Cardiol Cardiovasc Interv. 2009;2:415–421.

32. Pucelikova T, Dangas G, Mehran R. Contrast-induced nephropathy.Catheter Cardiovasc Interv. 2008;71:62–72.

33. Kahn JK, Rutherford BD, McConahay DR, Johnson WL, GiorgiLV, Shimshak TM, Hartzler GO. High-dose contrast agent adminis-tration during complex coronary angioplasty. Am Heart J. 1990;120:533–536.

CLINICAL PERSPECTIVEImpaired renal function is frequently associated with coronary artery disease, which in turn has a negative impact on theprognosis of patients with chronic renal failure. Revascularization by percutaneous coronary intervention is an effectivetreatment of coronary artery stenosis. However, it requires the use of iodinated contrast medium that can lead to acute renalfailure, in particular in patients with underlying impaired renal function, termed contrast-induced nephropathy. Iodinatedcontrast media are grouped by their osmolality. Iso-osmolar contrast medium, such as iodixanol, resembles the osmolalityof blood, whereas all other commercially available contrast media, such as iomeprol, exhibit higher osmolality. Thefindings of the contrast study that randomly assigned patients undergoing percutaneous coronary intervention to receiveeither iodixanol or iomeprol suggest that iso-osmolar contrast medium does not exhibit significantly lower nephrotoxicitycompared with higher-osmolar contrast medium in unselected patients with chronic renal dysfunction. However, subgroupanalysis suggested decreased nephrotoxicity in patients who received high contrast volumes favoring iso-osmolar contrastmedium. Considering the results of the Contrast Media and Nephrotoxicity Following Coronary Revascularization byAngioplasty study as well as a recent meta-analysis and other randomized clinical trials that compared iso-osmolar withhigher-osmolar contrast medium in mixed patient cohorts undergoing coronary angiography, iso-osmolar contrast mediumshould be considered as a first-line option for use in this high-risk patient population.

Wessely et al Contrast Media–Induced Nephrotoxicity and PCI 437

by guest on October 9, 2016http://circinterventions.ahajournals.org/Downloaded from

Angioplasty (CONTRAST) Trial investigatorsfor the Contrast Media and Nephrotoxicity Following Coronary Revascularization by

Schulz, Julinda Mehilli, Albert Schömig and Adnan KastratiRainer Wessely, Tobias Koppara, Christian Bradaric, Marc Vorpahl, Siegmund Braun, Stefanie

Percutaneous Coronary InterventionChoice of Contrast Medium in Patients With Impaired Renal Function Undergoing

Print ISSN: 1941-7640. Online ISSN: 1941-7632 Copyright © 2009 American Heart Association, Inc. All rights reserved.

Avenue, Dallas, TX 75231is published by the American Heart Association, 7272 GreenvilleCirculation: Cardiovascular Interventions

doi: 10.1161/CIRCINTERVENTIONS.109.8749332009;2:430-437; originally published online September 22, 2009;Circ Cardiovasc Interv. 

http://circinterventions.ahajournals.org/content/2/5/430World Wide Web at:

The online version of this article, along with updated information and services, is located on the

http://circinterventions.ahajournals.org/content/suppl/2009/10/21/CIRCINTERVENTIONS.109.874933.DC1.htmlData Supplement (unedited) at:

  http://circinterventions.ahajournals.org//subscriptions/

is online at: Circulation: Cardiovascular Interventions Information about subscribing to Subscriptions: 

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

  document. Answer

Permissions and Rights Question andunder Services. Further information about this process is available in thepermission is being requested is located, click Request Permissions in the middle column of the Web pageClearance Center, not the Editorial Office. Once the online version of the published article for which

can be obtained via RightsLink, a service of the CopyrightCirculation: Cardiovascular Interventionsin Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:

by guest on October 9, 2016http://circinterventions.ahajournals.org/Downloaded from

 

 

 

Supplemental Material 

 

 

 

 

Supplemental material does exclusively consist of supplementary tables. 

Supplementary Table S1

Baseline and clinical characteristics of patients receiving diagnostic cardangiography

only

Characteristic Iodixanol group

(n = 315)

Iomeprol group

(n = 336) P

Age – yr 73.2 ± 9.2 73.1 ± 9.3 .88

Men – n, (%) 208 (66) 207 (62) .24

Diabetes mellitus – n, (%) 100(32) 99 (29) .53

IDDM – n, (%) 41 (13) 35 (10) .30

History of smoking – n, (%) 119 (38) 120 (36) .59

Arterial hypertension – n, (%) 298 (95) 306 (91) .08

Hypercholesterolemia – n, (%) 294 (93) 307 (91) .35

Prior myocardial infarction – n, (%) 103 (33) 98 (29) .33

Prior aortocoronary bypass surgery – n, (%) 59 (19) 59 (18) .70

Ejection fraction – (%) 50.8 ± 12.7 49.6 ± 14.1 .30

Medication on admission

ACE Inhibitor – n, (%) 204 (65) 226 (67) .50

AT-1 Blocker – n, (%) 111 (35) 109 (32) .45

ß Blocker – n, (%) 250 (79) 266 (79) .95

Calcium Channel Inhibitor – n, (%) 91 (29) 117(35) .10

Statin – n, (%) 264 (84) 275 (82) .51

Supplementary Table S 2

Baseline angiographic characteristics of patients receiving solely diagnostic

cardangiography and not undergoing PCI

Characteristic Iodixanol group

(n = 315)

Iomeprol group

(n = 336) P

Diseased vessels

- 1-vessel disease – n. (%)

- 2-vessel disease – n. (%)

- 3-vessel disease – n. (%)

45 (14)

60 (19)

155 (49)

52 (15)

56 (17)

149 (44)

.22

Supplementary Table S3

Renal function before and after contrast exposure in patients receiving solely diagnostic

cardangiography

Characteristic Iodixanol group

(n = 315)

Iomeprol group

(n = 336) P

S-Creatinine prior PCI – [mg/dl] 1.33 ± 0.41 1.29 ± 0.30 .18

S-Urea prior PCI – [mg/dl] 55.4 ± 27.7 53.6 ± 20.4 .34

eGFR – [ml/min*1.73m²] 49.2 ± 13.0 49.1 ± 10.9 .86

Volume contrast media delivered – [ml] 146 ± 58 144 ± 53 .57

Maximal S-Creatinine post PCI – [mg/dl] 1.41 ± 0.61 1.35 ± 0.43 .20

Maximal S-Urea post PCI – [mg/dl] 56.0 ± 26.5 54.1 ± 27.0 .32

Maximal rise in S-Creatinine – [mg/dl] 0.08 ± 0.29 0.06 ± 0.27 .47

Maximal rise in S-Creatinine – [% of prior PCI] 5.0 ± 15.9 4.9 ± 17.9 .95

Maximal rise in S-Urea – [mg/dl] 0.60 ± 15.0 0.51 ± 12.7 .94

Hemodialysis post PCI – n, (%) 0 (0) 0 (0) NA

CIN – n, (%) 29 (9.2) 27 (8.0) .59

Rise in S-Creatinine ≥ 1 mg/dl – n, (%) 3 (1.0) 4 (1.2) .77

Supplementary Table S4.

Clinical 6-months follow-up of patients receiving solely diagnostic cardangiography

Characteristic Iodixanol group

(n = 315)

Iomeprol group

(n = 336) P

Myocardial infarction – n, (%) 1 (0.3) 0 (0) .30

Death – n, (%) 18 (5.7) 16 (4.8) .59