Post Contrast Acute Kidney Injury (PC-AKI): We Don’t Need Another Retrospective Observational Study

Background: Computed tomography pulmonary angiography (CTPA) is the current gold standard for diagnosing acute pulmonary embolism in the ED.  It has a high sensitivity, and specificity, is readily available, and can establish analternative diagnoses.  One issue with CTPA is that many hospital protocols create barriers for patients with chronic kidney disease or acute kidney injury (AKI) protocols in place from getting the necessary IV contrast.  There are several studies [2][3][4] that have evaluated the causal relationship between contrast exposure and nephrotoxicity. However, most of these studies are observational and retrospective in nature. The issue with retrospective studies is that they often cannot control for confounders and observational studies cannot give us causation, only association. We now have another retrospective observational study asking the same question, which has the inherent issues of previous studies.

What They Did:

  • Is the level of estimated glomerular filtration rate (eGFR) observed in the emergency department (ED) significantly associated with the occurrence of postcontrast acute kidney injury (PC-AKI) in patients undergoing CTPA
  • Single center retrospective observational study performed in Korea


  • Primary: PC-AKI (Defined as increase in serum creatinine concentration ≥0.3mg/dL or a ≥1.5 – 1.9 fold increase from baseline within 48 – 72hrs after contrast)
  • Secondary: Multivariate logistic regression analysis was used to confirm the effect of eGFR in the ED on the occurrence of PC-AKI after adjustment for confounding variables


  • All consecutive adult patients who underwent CTPA in the ED during the study period


  • End stage renal disease (ESRD)
  • No follow-up serum creatinine within 48hrs after contrast


  • 1300 patients underwent CTPA in the ED
  • 632 patients selected for the study after exclusion
    • eGFR <30mL/min/1.73m2 = 11 pts
    • eGFR 30 – 59mL/min/1.73m2 = 94 pts
    • eGFR 60 – 89mL/min/1.73m2 = 249 pts
    • eGFR ≥90mL/min/1.73m2 = 278pts
  • Incidence of PC-AKI = 6.49% (41/632 patients)
  • Incidence of PC-AKI Broken Down by eGFR:
    • eGFR <30mL/min/1.73m2 = 27.27%
    • eGFR 30 – 59mL/min/1.73m2 = 6.38%
    • eGFR 60 – 89mL/min/1.73m2 = 6.43%
    • eGFR ≥90mL/min/1.73m2 = 5.76%
    • No statistically significant association between eGFR and risk of PC-AKI observed
  • Among patients with PC-AKI only 5 (0.79%) required renal replacement therapy


  • Asks a question where there is continued active debate
  • Used univariate and multivariate analyses to limit confounders
  • Enrolled consecutive patients 


  • 639 out of 1300 patients excluded due to not having repeat serum creatine performed within 48 hours after contrast
  • Retrospective observational study making the study prone to unidentified confounders
  • Primary endpoint is not clinically important. A 0.3mg/dL increase in creatinine is not clinically important nor a patient oriented outcome
  • Single center study may not generalize to other practice settings
  • CT contrast protocol was a bolus infusion of 60ML of iodinated contrast agent (100mL/bottle; Xenetix 300) given IV at a rate of 4mL/s à The osmolarity of this agent is 695 mOsm/kg which may be higher than what we use in many institutions in the US
  • PC-AKI is not a patient oriented outcome, but the need for renal replacement therapy and mortality would be


  • There are other diagnoses in acutely ill patients that could be responsible for the AKI and not the contrast itself:
    • Chronic CHF
    • Multiple Myeloma
    • Leukemia
    • Poorly controlled diabetes mellitus (i.e. requiring insulin)
    • Sepsis/Septic Shock
  • This is important because the “risk factors associated with AKI are truly all over the map. Why would insulin usage be linked to AKI but diabetes mellitus itself not be? Also the level of consciousness being reduced as a risk factor in and of itself certainly does not cause AKI, but maybe being hemodynamically unstable (i.e. septic shock) could be the risk factor and doesn’t make logical sense.

Author Conclusion: “Our study findings could serve as useful reference for physicians who are concerned about performing computed tomography pulmonary angiography for fear of renal function deterioration.”

Clinical Take Home Point: This current study shows that the last eGFR prior to CTPA in patients with suspected acute pulmonary embolism in the ED was not associated with occurrence of PC-AKI, even if the eGFR was <30mL/min/1.73m2, but there was certainly a trend toward increased AKI in this patient population.  What we need now is not another retrospective observational study, and although a randomized clinical trial would be great, it is unlikely to ever happen.  We now have a huge amount of data saying it doesn’t matter what the kidney function of a patient is and we should start to change protocols to allow IV contrast in patients where critical diagnoses need to be made.


  1. Cho A et al. Postcontrast Acute Kidney Injury After Computed Tomography Pulmonary Angiography for Acute Pulmonary Embolism. JEM 2019. PMID: 31740158
  2. Hinson JS et al. Risk of Acute Kidney Injury After Intravenous Contrast Media Administration. Ann Emerg Med 2017; S0196 – 0644(16): 31388 -9. PMID: 28131489
  3. Hinson JS et al. Acute Kidney Injury Following contrast Media Administration in the Septic Patient: A Retrospective Propensity-Matched Analysis. J Crit Care 2019. PMID: 30798098
  4. Aycock RD et al. Acute Kidney Injury After Computed Tomography: A Meta-Analysis. Ann. Emerg Med. 2017 [Epub ahead of print]. PMID: 28811122

For More Thoughts on This Topic Checkout:

Post Peer Reviewed By: Anand Swaminathan, MD (Twitter: @EMSwami)

Cite this article as: Salim Rezaie, "Post Contrast Acute Kidney Injury (PC-AKI): We Don’t Need Another Retrospective Observational Study", REBEL EM blog, January 16, 2020. Available at:

Like this article?

Share on Facebook
Share on Twitter
Share on Linkdin
Share via Email

Want to support rebelem?