This generally well-conducted review concluded that iodixanol was not associated with a reduced risk of contrast media-induced nephrotoxicity compared with low-osmolar contrast media, except in patients with intra-arterial administration and renal insufficiency. The authors' conclusions reflected the results of the review and should be reliable, but differences between the interventions were not fully explored.

To compare the toxicity to kidney cells (nephrotoxicity) of iso-osmolar iodixanol with non-ionic low-osmolar contrast media in randomised controlled trials (RCTs).

MEDLINE, EMBASE, BIOSIS Previews, Web of Science, Web of Knowledge, Science Citation Index Expanded, Current Contents Medizin, and the Cochrane Library were searched for articles from 1950 to 2007. Trial registers, abstracts from major scientific meetings from 2003 to 2007, and reference lists from relevant studies were also searched. Experts in the field and manufacturers were contacted for information on any relevant unpublished studies.

Randomised controlled trials (RCTs) comparing iso-osmolar iodixanol with one of the non-ionic low-osmolar contrast media, for diagnostic or therapeutic procedures, with intravenous or intra-arterial application, were included. For inclusion, trials had to assess the serum creatinine levels before and 24 to 72 hours after contrast medium administration or the incidence of a decrease in renal function of more than a threshold level.

The mean age ranged from 52 to 73 years and the percentage of women in each arm varied from 7% to 57%. The percentage of diabetic patients ranged from 2% to 100%. The dose of contrast media varied widely from 23.4g to 124.5g.

Two reviewers independently selected the trials and any discrepancies were resolved by consensus.

Two reviewers independently assessed the validity of each included trial based on the method of blinding, method of randomisation, allocation concealment, reporting of dropouts and withdrawals, number of randomised patients who were not included in the analysis, evidence of important baseline differences, specification of inclusion and exclusion criteria, intention-to-treat analysis, and power calculation. The five-point Jadad scale (zero to two points for method of randomisation, zero to two points for masking, and zero to one point for dropouts and withdrawals) was also used to derive an overall quality score.

For each trial, the number of patients with contrast media-induced nephrotoxicity, which was defined as more than a 25% increase in serum creatinine level or more than a 0.5mg per dL increase in serum creatinine level, and the mean peak change in serum creatinine level within three days of contrast media administration were extracted. If these data were not reported, the data that most closely resembled them were extracted. These data were used to calculate the relative risk of contrast media-induced nephrotoxicity and the difference in increase in serum creatinine between the trial groups.

Two reviewers extracted the data using a standardised form. Discrepancies were resolved through discussion.

Pooled relative risks and weighted mean differences, with 95% confidence intervals, were calculated using random-effects models. The Cochran Q and I² tests were used to assess heterogeneity. If heterogeneity was found, the influence of different patient and intervention characteristics on the pooled estimate was further explored in subgroup, one-way analysis of variance, and meta-regression analyses. Sensitivity analyses were conducted to assess the influence of including trials with closely related data, when no data was given for a particular definition of nephrotoxicity, unpublished trials, and weighting with quality scores. Publication bias was assessed in a graph and using the Begg and Egger tests.

Twenty-five trials (n=3,270 participants) were included and the median quality score was 3.5 (range 1 to 5).

There was little evidence that iodixanol reduced the risk of contrast medium-induced nephropathy compared with other low-osmolar contrast media, when using the definition an increase of at least 25% in serum creatinine level (RR 0.80, 95% CI 0.61 to 1.04; 16 RCTs). When a definition of at least 0.5mg per dL increase in serum creatinine level was used, there was still little evidence that iodixanol reduced the risk of contrast medium-induced nephropathy (RR 0.75, 95% CI 0.44 to 1.26; 16 RCTs). Heterogeneity was statistically significant (I²=44%, p=0.03). The increase in serum creatinine levels was not significantly different between iodixanol and low-osmolar contrast media (WMD 0.01mg/dL, 95% CI -0.01 to 0.03; 22 RCTs). Heterogeneity was statistically significant (I²=54%, p<0.01).

The type of low-osmolar contrast media, pre-existing renal insufficiency, and application route were important factors that contributed to the heterogeneity. Compared with iodixanol, the risk of contrast medium-induced nephropathy was significantly greater in patients, with pre-existing renal insufficiency, who were receiving iohexol intra-arterially (RR 0.38, 95% CI 0.21 to 0.68; three RCTs). The results of the subgroup analyses were presented.

No evidence of publication bias was found.

Iso-osmolar iodixanol was not associated with a reduced risk of contrast medium-induced nephropathy compared with low-osmolar contrast media, after intravenous application. In patients, with renal insufficiency, who were receiving intra-arterial application, low-osmolar iohexol was associated with a greater risk of contrast medium-induced nephropathy compared with iodixanol. No significant differences were found between iodixanol and other low-osmolar contrast media.

The review question was clear and was supported by clear inclusion criteria. The literature search was thorough and there were attempts to identify unpublished trials. The review process incorporated adequate steps to minimise bias and errors, such as independent and duplicate study selection, validity assessment, and data extraction. No language restrictions were applied, minimising the risk of language bias. Appropriate criteria were used to assess validity and the results were reported. The method of synthesis was appropriate and focused on predefined research objectives. The pooled estimate included trials that used different contrast media and it might have been helpful to explore this as a potential source of heterogeneity, using additional subgroup or sensitivity analyses.

This review was generally well-conducted and the conclusions should be reliable, but the impact of differences in the intervention on the outcomes was not fully explored.

Practice: The authors did not state any implications for practice.

Research: The authors stated that large multi-centre trials, using clinically relevant outcome measures, were required to assess whether nephrotoxicity was low with iodixanol compared with other low-osmolar contrast media, in high-risk patients after intra-arterial application.

No funding received.

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.