Purpose: To compare the incidence of contrast-induced nephropathy (CIN) for intravenous vs. intra-arterial administration of iodixanol, compared to non-administration.
Methods: We retrospectively identified 650 patients who had intravenous iodixanol-enhanced CT, 695 with intra-arterial iodixanol cardiac catheterization, 651 with unenhanced CT, and those who also had baseline and follow-up serum creatinine within 5 days of the exam. From the medical records, we recorded the gender, age, baseline and follow-up serum creatinine/eGFR; underlying renal injury risk factors; indication for imaging; contrast material administration volume, concentration, and route of administration; and use of pre-imaging prophylactic measures for CIN. Univariate and multivariate models were used to determine predictors of CIN.
Results: Baseline eGFR was lower for patients undergoing unenhanced CT than intravenous or intra-arterial patients (68 vs. 74.6 and 72.2, respectively, p < 0.01) and not different between intravenous and intra-arterial patients (p = 0.735). Simple logistic regression did not show a difference in the rate of CIN in patients who received intravenous vs. intra-arterial iodixanol (28 of 650, 4%, vs. 28 of 695, 4%, respectively, p = 0.798), nor a higher rate of CIN than seen with unenhanced CT (45 of 651, 7%, p = 0.99 and p = 0.98 by one-sided t test). Multivariate regression modeling showed that only elevated baseline creatinine or decreased eGFR and low hematocrit/hemoglobin were associated with CIN incidence (odds ratio 1.28 and 2.5; p < 0.023 and <0.006, respectively).
Conclusions: Elevation in serum creatinine due to intravenous and intra-arterial iodixanol administration is infrequent and is not more common than after unenhanced CT scans.
Keywords: CT; Cardiac catheterization; Contrast-induced nephropathy; Intra-arterial; Intravenous; Renal.