Background

Imaging is one of the most important diagnostic modalities that physicians utilize. In 2013 alone, over 70 million CT scans were performed. Contrast-enhanced imaging can aid in diagnosing certain pathology and improve image quality. There has historically been a concern for post-contrast acute kidney injury (AKI), which is generally considered an increase in creatinine or a decrease in glomerular filtration rate hours to days after contrast administration. The incidence has not been well determined due to differing definitions of AKI, differences in follow-up of patients after CT scans with contrast, and a lack of randomized control studies. The American College of Radiology believes the correlation between contrast administration and AKI may be better attributed to the patients’ antecedent pathology than to contrast. With more data and reviews, it is starting to come to light that the idea of “contrast-induced nephropathy” might be an overstated entity with an unfounded fear.

Clinical Question

Through a meta-analysis of observational reviews, is there evidence of an increased risk of acute kidney injury, need for renal replacement, and/or total mortality after contrast-enhanced CT versus noncontrast CT?

Design

Meta-Analysis and systematic review of 28 observational studies, which included over 100,000 patients.

Primary Results

Tests for heterogeneity (I2, p values statistically significant FOR heterogeneity)

Primary Outcome

  • Acute Kidney Injury
    • No significant association between contrast CT and outcome.
    • Overall: I2 = 65.1%, P <0.001, OR 0.938 (CI 0.825, 1.065)
    • Risk of AKI
      • with contrast: 7.2%
      • w/o contrast: 7.4%

Secondary Outcomes

  • Renal Replacement Therapy
    • No significant association between contrast CT and outcome.
    • Overall: I2 = 19.9%, P = 0.243, OR 0.825 (CI 0.587, 1.160)
    • Risk of RRT
      • with contrast: 0.6%
      • w/o contrast: 0.7%
  • Mortality
    • No significant association between contrast CT and outcome.
    • Overall: I2 = 39.8%, P = 0.102, OR 0.998 (CI 0.730, 1.362)
    • Risk of mortality
      • with contrast: 5.6%
      • w/o contrast: 5.9%

Strengths

  • Power of combined studies, including over 100,000 patients
  • Sub-group analysis, allowing to see there is no statistically significant difference within the patients from the ED groups.
  • Low likelihood of publication bias in this meta-analysis
  • Comprehensive literature search

Limitations

  • Moderate heterogeneity (I2 = 65%) within the studies for the primary outcome and 7 out of 28 articles with low risk of bias, remainder were moderate, serious or critical
  • Definitions for AKI varied between the studies
  • Only 18.5% had follow-up within 72 hours to evaluate for AKI
  • 2 largest studies (McDonald, Davenport) account for 38.7% of all included in analysis
  • The clinical significance of the AKI once it occurs remains unclear
  • Cannot give generalizability to intra-arterial procedures (including PCI)

Author's Conclusions

“In conclusion, our study found a lack of association between acute kidney injury and contrast-enhanced CT and no association with important patient-oriented and clinical outcomes, including the need for renal replacement therapy and mortality. The American College of Radiology ACR Manual on Contrast Media underscores this point and argues for a shift in language from contrast-induced nephropathy to post-contrast acute kidney injury, with the understanding that the acute kidney injury may be incidental rather than caused by the contrast. These findings are limited by the quality of included studies and by significant selection bias, including provider selection for contrast-enhanced CT. These observational data demonstrate that physician selection of patients to receive contrast-enhanced CT seems to add no additional risk of acute kidney injury, need for renal replacement therapy, or mortality. These findings are congruent with current assertions from the American College of Radiology.”

Our Conclusions

There is an issue of heterogeneity and influence of bias in this analysis, but overall an exceptional amount of patients included in this meta-analysis with data pointing towards no direct association between contrast and AKI. It is hard to signify a causal relationship and, for that, prospective data is needed. Based on a subgroup analysis of ED patients, there was no significant association between contrast and AKI. It is reasonable to believe that this previous dogma that contrast is nephrotoxic can start to fall out favor as more studies come out. It is also important to note that we still cannot come to the conclusion that this has any clinical significance. Studies for follow-up data and clinical significance need to also be further addressed. Overall, if a patient needs a contrast-enhanced scan, the need for that scan may outweigh any theoretical risk for an AKI, the precipitants and significance of which remains unclear.

Potential Impact To Current Practice

There is an inability to perform randomized controlled center trials on this outcome but further research, likely prospective, is necessary. Current practice guidelines vary by physician and hospital protocols. Whether it be prior hydration, preventive strategies or following creatinine, these may start to prove to be unnecessary. Likely, between discussions with our radiology colleagues, a new consensus will take shape and an old dogma can be put to rest.

Bottom Line

Evidence continues to mount that post-contrast AKI is likely not due to contrast, but rather caused by the underlying condition that prompted imaging.

Read More

ACR Statement on Post-Contrast AKI: American College of Radiology (PDF)

REBEL EM: Contrast Induced Nephropathy (CIN): Fact or Myth?

EM Lit of Note: Punching Holes in CIN