COVER STORY Contrast Selection in the Cardiac Catheterization Lab Does choice of contrast agent really matter? BY C. BRADLEY JONES, MD, AND JOSEPH D. BABB, MD, FACC, FSCAI ince the first selective injection of contrast media ionic) or by production of dimeric molecules. Numerous into the right coronary artery by F. Mason Sones, studies, including a meta-analysis of 31 trials, have shown MD, on October 30, 1958, there has been consid- low-osmolar contrast media (LOCM) to be associated erable refinement in contrast media. Multiple with significantly less CIN than high-osmolar contrast Sagents have been utilized and studied since the original media in patients with renal impairment and/or diabetes.2 data from Sones comparing meglumine iothalamate and There has been considerable refinement during the past sodium iothalamate to meglumine diatrizoate and sodi- decades from ionic high-osmolar, to nonionic low-osmo- um diatrizoate in 2,258 patients with and without cardiac lar, and finally to nonionic iso-osmolar contrast media disease.1 With the introduction of multiple newer-genera- (IOCM). High-osmolar agents have concentrations >2,000 tion contrast agents, the question arises, “Does contrast mOsm/kg H20, whereas LOCM are in the range of 600 agent selection matter?” To adequately address this ques- mOsm/kg H20 to 844 mOsm/kg H20. Blood has an osmo- tion, several issues must be considered. These include the lality of 290 mOsm/kg H20 and, therefore, an “isomolar” effect of various contrast agents on incidence of renal tox- agent has an osmolality the same as blood. LOCM that icity and contrast-induced nephropathy (CIN), prothrom- are nonionic agents include iohexol, iopamidol, iopentol, botic and anticoagulant properties in vivo that manifest iopromide, imeprol, iobitridol, and ioversol. The sole as major cardiac events (MACE) and, with the emerging LOCM ionic dimmer available is ioxaglate.3 The only cur- health care economic crisis, cost. rently available iso-osmolar agent is iodixanol. When considering an agent to visualize vascular struc- tures, there are many different properties to keep in mind; NEPHROTOXICITY the major properties are viscosity, ionic versus nonionic, Several studies have evaluated the incidence of CIN and osmolality. The osmolality depends on the number of comparing LOCM to ICOM (Table 1). Aspelin et al4 con- molecules that are present in the solution, which can be ducted a randomized, double-blind, prospective, multi- reduced by producing agents that do not dissociate (non- center study in 129 patients undergoing coronary or TABLE 1. MAJOR TRIALS OF NEPHROTOXICITY Study Agents Used Patient Population Endpoints Results Jo et al5 Iodixanol and ioxaglate 300 CrCl <60 mL/min Increase in serum creatine IOCM superior RECOVER ≥25% or ≥0.5 mg/dL Aspelin et al4 Iohexol and iodixanol 129 patients with dimeric Increase in serum creatine IOCM superior NEPHRIC molecules and SCr 1.5–3.5 ≥25% or ≥0.5 mg/dL Solomon et al7 Iopamidol and iodixanol 414 Serum creatine increase ≥0.5 No significant difference CARE Liss et al10 Iodixanol (iso-osmolar) and 57,000 Rehospitalization with renal LOCM superior Registry ioxaglate (low-osmolar) failure or hemodialysis 52 ICARDIAC INTERVENTIONS TODAYIJANUARY/FEBRUARY 2008 COVER STORY aortofemoral angiography comparing the nephrotoxic dol or iodixanol to high-risk patients, with or without dia- effects of iodixanol with those of iohexol. This study betes mellitus, and iopamidol was associated with a signif- (NEPHRIC) involved 129 patients with diabetes with icantly smaller mean increase in serum creatinine levels serum creatinine concentrations of 1.5 to 3.5 mg/dL. The when compared with iodixanol.7 Similarly, Barrett et al2 primary endpoint was the peak increase from baseline in compared the effects on renal function of iopamidol and the creatinine concentration during the first 3 days after iodixanol in a multicenter, double-blind, randomized trial angiography. The creatinine concentration increased sig- of 166 patients with chronic kidney disease defined as nificantly less in patients who received iodixanol. From serum creatinine ≥1.5 mg/dL and/or creatinine clearance day 0 to day 3, the mean peak increase in creatinine was ≤60 mL/min undergoing contrast-enhanced multidetec- 0.13 mg/dL in the iodixanol group and 0.55 mg/dL in the tor CT. An absolute increase ≥0.5 mg/dL in creatinine iohexol group (P=.001). These investigators concluded clearance was observed in none of the patients receiving that CIN may be less likely to develop in high-risk patients iopamidol-370 and in 2.6% (2/76) of patients receiving when iodixanol is used rather than a nonionic LOCM.4 iodixanol-320 (P=.2). These investigators concluded that Jo et al5 compared the renal tolerance of the LOCM the rate of CIN was similarly low in patients with stable ioxaglate (Hexabrix, Covidien, Hazelwood, MO) and iodix- moderate-to-severe chronic kidney disease after intra- anol (Visipaque, GE Healthcare, Waukesha, WI) in The venous administration of iopamidol-370 or iodixanol-320 Renal Toxicity Evaluation and Comparison between for CT.8 This being stated, there may be differences in Visipaque (iodixanol) and Hexabrix (ioxaglate) in Patients effects of intravenous versus intra-arterial contrast effects with Renal Insufficiency Undergoing Coronary Angio- on nephrotoxicity. graphy (RECOVER) study. This was a prospective, ran- These findings were further supported at a 2006 oral domized trial of 300 patients with creatinine clearance of presentation of the Ionic Versus Nonionic Contrast to <60 mL/min. The primary endpoint was the incidence of Obviate Worsening of Nephropathy After Angioplasty in CIN defined as an increase in serum creatinine ≥25% or Chronic Renal Failure Patients (ICON) trial.9 This cohort of ≥0.5 mg/dL. The incidence of CIN in patients with severe 145 patients with renal insufficiency undergoing angiogra- renal impairment at baseline (CrCl <30 mL/min) or dia- phy was randomized to receive ioxaglate and iodixanol. All betes and in those receiving large doses (≥140 mL) of con- patients had chronic kidney disease with serum creatinine trast medium was also determined. The incidence of CIN measurements from 1.5 to 3 mg/dL. All subjects were well was determined to be significantly lower with iodixanol hydrated, receiving approximately 3.7 L of fluid. The use of (7.9%) than with ioxaglate (17%; P=.021). The incidence of N-acetyl-cysteine was left to the discretion of the investi- CIN was also significantly lower with iodixanol in patients gator. The primary endpoint was the peak increase in with severe renal impairment (P=.023) or concomitant serum creatinine out to day 3. Compared with ioxaglate, diabetes (P=.041), or in patients receiving ≥140 mL of con- iodixanol did not significantly reduce the increase in trast media (P=.038).5 These findings are further support- serum creatinine levels after coronary catheterization or ed by McCullough et al in a recent meta-analysis of 2,727 PCI. There was no significant difference between the two patients, indicating that the use of IOCM is associated agents at any time, nor was there any difference between with smaller increases in creatinine and lower rates of CIN the two agents when other definitions of contrast nephro- than LOCM, especially in patients with chronic kidney dis- pathy were used. Furthermore, in-hospital and 30-day out- ease and diabetes.6 comes did not differ between the two agents.9 The ICON In contrast, Solomon et al7 compared the incidence of trial has not been peer reviewed or published, however. CIN between the LOCM iopamidol with that of the Finally, Liss et al10 compared the Swedish Coronary IOCM iodixanol in the Cardiac Angiography in Renally Angiography and Angioplasty Registry with the Impaired Patients (CARE) study. This was a multicenter, Swedish Hospital Discharge Register to assess contrast randomized, double-blind trial that compared the renal media-induced renal failure. Data from 23 hospitals, tolerability of 414 high-risk patients (defined as having a including more than 57,000 patients, were analyzed. glomerular filtration rate of 20 to 59 mL/min) undergoing From 2000 to 2003, iodixanol (iso-osmolar) was used cardiac angiography or percutaneous coronary interven- in 45,485 patients, and ioxaglate (low-osmolar) was tion (PCI). The primary endpoint was defined as a serum used in 12,440 subjects. To include the earlier used creatinine increase ≥0.5 mg/dL over baseline at 2 to 5 contrast media iohexol (low-osmolar), analysis days. Mean postprocedure serum creatinine increases extended back to 1990 (86,334 patients). Renal toxici- were significantly less with iopamidol (P=.01). The investi- ty was defined by rehospitalization with renal failure gators determined that the rate of CIN is not statistically or by requiring dialysis. Using these definitions, the different after the intra-arterial administration of iopami- incidence of clinically significant renal failure was JANUARY/FEBRUARY 2008 I CARDIAC INTERVENTIONS TODAY I 53 COVER STORY greatest for patients receiving the IOCM iodixanol TABLE 2. MAJOR TRIALS OF MACE (1.7%). Ioxaglate-treated patients had a significantly Study Agents Used Endpoints lower renal failure incidence (0.8%; P<.001). The odds Davidson et al Iodixanol and Iodixanol reduction ratio for iodixanol-treated patients was significantly COURT11 ioxaglate in MACE higher than for ioxaglate (1 vs 0.48; P<.001). In subsets of either diabetic patients or patients with previous Harrison et al12 Iodixanol and Iodixanol reduction renal failure, the odds ratios for renal failure remained iopamidol in MACE greater in the iodixanol groups (P<.01). Hospitals Le Feurve et al13 Ioxaglate and iodix- Ioxaglate reduction switching contrast media to iodixanol experienced a anol in MACE doubling in clinically significant renal failure after car- Sutton et al14 Ioxaglate and iodix- Trend favoring diac procedures. Dialysis was required in 0.2% of anol ioxaglate patients receiving iodixanol, which was significantly higher (P<.01) than for ioxaglate-treated patients diac death was higher in the iodixonal cohort (5 vs 1; (0.1%).