Optimal contrast agents for vascular imaging on computed tomography: Iodixanol versus iohexol

Harpreet K. Pannu, Richard E. Thompson, John Phelps, Carolyn A. Magee, Elliot K. Fishman

Research output: Contribution to journalArticlepeer-review

29 Scopus citations


Rationale and Objectives. Dimeric nonionic iodinated contrast has a lower osmolality than monomeric nonionic iodinated contrast but is available at lower iodine concentrations. Less dilution of intravascular fluid by influx from the extravascular space is proposed to occur with decreasing osmolality. The purpose of this study was to determine if a dimeric nonionic iso-osmolar contrast agent (iodixanol) gives equal vascular enhancement compared with a monomeric nonionic hyperosmolar contrast agent (iohexol). Materials and Methods. A dynamic single-level computed tomography (CT) scan was performed of the abdominal aorta of 12 sedated rabbits using a four-row multidetector CT scanner following injection of 1.5 mL contrast/kg body weight at 2 mL/sec. The rabbits were injected with the dimeric contrast agent iodixanol (Visipaque 320; Amersham Health) or the monomeric contrast agent iohexol (Omnipaque 350; Amersham Health). The order of the type of contrast media injected was randomized for each rabbit, and the interval between injections was 2 weeks. Using the 2.5-mm detectors, four contiguous 3-mm contrast-enhanced scans were obtained at a single level every 5 seconds for 120 seconds (total of 24 scans) with a kVp of 120, mA·s of 110, field of view of 106 mm, and soft tissue reconstruction algorithm. A single level was chosen to measure the attenuation of the abdominal aorta at 5-second intervals. The mean attenuation and standard deviation values were recorded for the whole aorta, for the central half of the vessel, and for the peripheral half of the vessel. A log-log transformation of the data was performed and regression analysis was done on the outcomes of interest (e.g., mean, standard deviation) on time for each region. Results. There was no statistically significant difference in mean attenuation for the whole aorta for iodixanol and iohexol (P = .918) even though the iodine content was 9.3% less with the dimeric iodixanol. The time-attenuation curve of iodixanol paralleled that of iohexol for all time points. The mean attenuation values of the central half of the aorta (P = .354) and peripheral half of the aorta (P = .758) were also not statistically different for the two contrast agents. Conclusion. The vascular attenuation provided by a 9.3% lower iodine concentration of iso-osmolar iodixanol is equal to that given by hyperosmolar iohexol. This suggests that there is less intravascular dilution of iso-osmolar contrast. The enhancement across the cross section of the vessel is also similar for both contrast agents. This suggests the vascular studies with iodixanol and iohexol are of equal quality even when a lower dose of iodine is given with iodixanol. It is relevant for patients with borderline or diminished renal function in whom less volume of contrast may be administered.

Original languageEnglish (US)
Pages (from-to)576-584
Number of pages9
JournalAcademic Radiology
Issue number5
StatePublished - May 2005
Externally publishedYes


  • Computed tomography
  • Iodixanol
  • Iohexol
  • Vascular imaging

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging


Dive into the research topics of 'Optimal contrast agents for vascular imaging on computed tomography: Iodixanol versus iohexol'. Together they form a unique fingerprint.

Cite this