Systematic review and meta-analysis of frequency of acute kidney injury following intravenous contrast administration 

Trevor Bickford, Aaron Lane, Matt Vassar

Research output: Contribution to conferencePoster

Abstract

Purpose: To update a systematic review and meta-analysis of controlled studies examining the incidence of acute kidney injury (AKI) after exposure to intravenous contrast media compared to AKI after imaging without contrast media (control). The original meta-analysis was published in the Journal of the American College of Radiology in 2013 and is the basis for guidelines for the American College of Radiology and the American College of Family Physicians. Although this meta-analysis concluded there was no association between contrast exposure and AKI, most guidelines still provide recommendations to avoid contrast media under certain conditions for fear that it could theoretically cause AKI. The original meta-analysis called for additional studies to use a non-contrast control group, and the ACR guidelines call for more matched studies. This study attempts to collate the studies conducted on this issue in attempt to strengthen the evidence base behind clinical recommendations related to CIN.

Materials and Methods: EMBASE, Scopus, and Medline via PubMed's interface were searched using the same search criteria as the 2013 meta-analysis with the time period of interest changed from before 2011 to 2011 to present. Abstract screening was performed using the same inclusion criteria. The full text review was completed using the same exclusion criteria. The included studies were further screened by a clinician to ensure representation of clinical expertise. Data extraction was completed independently using a Google form and conflicts between researchers was resolved by a clinician. The same data points — baseline SCr or eGFR, contrast type, AKI definition, incidence of AKI, dialysis, and mortality — as the previous meta-analysis were recorded. Data extracted from the previous meta-analysis was combined with our updated data, and all data were analyzed and relative risks were calculated. Relative risk was calculated for incidence of AKI in both groups. A relative risk of less than 1.00 was indicative of a higher incidence of AKI in the non-contrast medium group.

Results: There were 8,358 studies identified. Of those, 34 (0.004%) were deemed to have met inclusion criteria; this represented 30,053,234 patients (1,731,241 receiving IV contrast and 28,321,993 not receiving contrast). There was 30 retrospective studies, 13 on which were propensity matched. There were only 4 prospective studies and 2 of those were propensity matched. The risk of AKI in the contrast medium group of the retrospective studies (RR= 0.9805; 95% confidence interval [CI]: 0.97, 0.99; p=0.0000000015) and prospective studies (RR= 1.009; 95% confidence interval [CI]: 0.81, 1.26; p= 0.934) was similar to the non-contrast medium group. The combination of retrospective and prospective studies with the previous meta-analysis findings also showed a similar risk (RR= 0.9806; 95% confidence interval [CI]: 0.97, 0.99; p= 0.0000000016) between contrast medium and non-contrast medium groups.

Conclusion: The meta-analysis from 2013 stated there was no difference in incidence of AKI between contrast exposure groups and control groups. Our data support this finding. Guidelines should be updated to reflect current research.
Original languageAmerican English
StatePublished - 4 Sep 2020
EventOklahoma State University Center for Health Sciences Research Day 2020 - Oklahoma State University Center for Health Sciences College of Osteopathic Medicine, Tulsa, United States
Duration: 27 Feb 202028 Feb 2020

Conference

ConferenceOklahoma State University Center for Health Sciences Research Day 2020
CountryUnited States
CityTulsa
Period27/02/2028/02/20

Fingerprint Dive into the research topics of 'Systematic review and meta-analysis of frequency of acute kidney injury following intravenous contrast administration '. Together they form a unique fingerprint.

Cite this