Risk of Acute Kidney Injury After Intravenous Contrast Media Administration

Hinson, et al.  Annals of Emergency Medicine 2017. [paper]

Why I chose this study

On almost a daily basis, whether on Emergency or Internal Medicine, we all face a difficult question: “can this patient tolerate IV contrast administration for CT scan?” We are often guided in our decisions based on the patient’s current renal function.  Occasionally, a debate arises between the ordering physician and the radiologist whether or not the patient is likely to suffer from the dreaded contrast-induced nephropathy (CIN). While historical consensus is that CIN is a possible complication of IV contrast, little evidence with regards to modern-day IV contrast use is available to guide our practices.

I recently had a patient that suffered acute kidney injury (AKI) after CT with contrast administration to evaluate her abdominal complaints. The renal consultant provided a very thorough, albeit unhelpful, differential diagnosis, including CIN, septic ATN, multi-drug nephrotoxicity, and AIN.  While the patient’s renal function improved after addressing all of these possibilities, I was left questioning which insult was ultimately responsible for her AKI. Although this article does not help me answer this question, it sheds new light on a long-standing question (at no risk of AKI to the reader).

Background

Contrast has been cited as the third most common cause of iatrogenic acute kidney injury. The decision to use contrast in emergently necessary studies often hinges on the patient’s renal function. The fear of CIN is so prevalent that randomized clinical trials cannot be performed because they would be seen as unethical. Historical studies used a different contrast media and were flawed in other ways. Two retrospective studies using propensity-score analysis have recently been published and have conflicting results.

Research Question

Is IV contrast administration for computed tomography (CT) independently associated with increased risk for acute kidney injury and adverse clinical outcomes?

Study Design

Single-center, retrospective, propensity-matched cohort analysis performed in a large, US, urban, academic emergency department (ED) (average census of 62,179 visits per year). The intervention was CT scan with or without intravenous contrast administration.

Inclusion

  • Patients aged 18 years and older presenting to the ED who received a CT with or without contrast enhancement and a second control who received no contrast or imaging.
  • Initial serum creatinine level measured in the 8 hours before CT.
  • A second serum creatinine level measured 48 to 72 hours after CT.

Exclusion

  • Initial serum creatinine level less than 0.4 mg/dl (to minimize random laboratory error).
  • Initial serum creatinine greater than or equal to 4.0 mg/dl (already meeting partial criteria for severe acute kidney injury).
  • Insufficient serum creatinine level data.
  • History of renal transplant or ongoing or previous dialysis.
  • A CT scan or ED visit in the 6 months preceding the index ED visit (to minimize potential confounding residual effects of previously administered contrast media).
  • Contrast-enhanced CT performed within 72 hours of ED departure (to minimize potential group crossover).

Primary Outcome

Incidence of acute kidney injury using these criteria:

  • The most frequently published criteria for contrast-induced nephropathy:
    • Absolute increase in serum creatinine level greater than or equal to 0.5 mg/dl or greater than or equal to 25% increase over baseline serum creatinine level at 48 to 72 hours after imaging or, for non-CT patients, after initial serum creatinine level measurement.
  • Acute kidney injury as defined by the Acute Kidney Injury Network/Kidney Disease Improving Global Outcomes Guidelines:
    • Stage 1: Absolute increase in serum creatinine level greater than or equal to 0.3 mg/dl or a 1.5- to 1.9-fold increase over baseline serum creatinine level.
    • Stage 2: 2.0- to 2.9-fold increase over baseline serum creatinine level.
    • Stage 3: 3-fold increase over baseline serum creatinine level, increase to serum creatinine level greater than or equal to 4.0 mg/dl, or initiation of dialysis.

Secondary Outcomes

  • Newly diagnosed chronic kidney disease within 6 months of the index ED visit
  • Initiation of dialysis within 6 months of the index ED visit
  • Renal transplantation within 6 months of the index ED visit

Analysis

  • The association between contrast media and AKI was first assessed with the test of proportions to compare incidence in patients who underwent contrast-enhanced CT with those who underwent unenhanced CT and with all patients who did not receive contrast media.
  • A multivariable logistic regression model was used after controlling for demographic variables and medical conditions previously reported to increase risk for developing AKI.
  • To reduce potential selection bias inherent to administration of contrast media, researchers also used propensity-score matching.
    • Given that the clinical decision to administer contrast media is guided by patient pathology and conditional patient-related factors that might contraindicate contrast media administration, the conditional patient factors included in the estimation of the propensity scores were sex, age, race, initial serum creatinine level or estimated glomerular filtration rate, crystalloid fluid administration, nephrotoxic medication administration, chronic comorbidities (diabetes mellitus, hypertension, HIV/AIDS, congestive heart failure, chronic kidney disease, and history of renal transplantation) and whether the patient was designated as requiring critical care.
    • Propensity-score matching was performed with default parameters (nearest neighbor of one, no caliper restriction, and sampling with replacement), and the average treatment effect was calculated.
  • All comparisons were made for the entire study population and for subgroups stratified by initial serum creatinine level and estimated glomerular filtration rate.
  • Matching for propensity score was performed by group for subgroup analyses.

Results

There were 17,934 ED visits for patients who underwent contrast-enhanced, unenhanced or no CT during a 5-year period (2009 to 2014).

  • 7,201 patients undergoing contrast-enhanced CT
  • 5,499 patients undergoing unenhanced CT
  • 5,234 patients with no imaging

Multivariable logistic regression modeling of the entire study population, with inclusion of predictor variables previously reported to affect the incidence of acute kidney injury, revealed no independent effect of contrast media on the probability of developing acute kidney injury.  Under the traditional definition of contrast-induced nephropathy and the Acute Kidney Injury Network criteria, the probability of developing acute kidney injury was respectively:

  • 10.6% and 6.8% in the contrast-enhanced CT group
  • 10.2% and 8.9% in the contrast-unenhanced CT group
  • 10.9%  and 8.1% in the non-CT group

Under the traditional definition contrast-enhanced CT patients were no more likely to develop acute kidney injury than all patients who did not receive contrast media (OR=1.01; 95% CI 0.92 to 1.12) or than patients who underwent unenhanced CT (OR=1.05; 95% CI 0.94 to 1.18). These results were consistent even when bias was accounted for in contrast media administration, using propensity-score matching (OR=0.99, 95% CI 0.98 to 1.00, and OR=1.00, 95% CI 0.98 to 1.01, respectively).

Using the Acute Kidney Injury Network criteria contrast media administration was associated with a decreased risk of developing acute kidney injury when directly compared to all patients who did not receive contrast media (OR=0.78; 95% confidence interval [CI] 0.70 to 0.88) and patients who underwent unenhanced CT (OR=0.75; 95% CI 0.66 to 0.85). This effect was not observed after propensity-score matching adjustment (OR=1.00, 95% CI 0.99 to 1.01, and OR=1.00, 95% CI 0.99 to 1.01, respectively).

Factors associated most strongly with an increased probability of acute kidney injury were increased age, administration of nephrotoxic medication(s), pre-existing diagnosis of congestive heart failure or chronic kidney disease, and hypoalbuminemia. Administration of intravenous crystalloids was associated with a lower probability of developing acute kidney injury.

There was also no association with contrast-enhanced CT scan and the development of chronic kidney disease, initiation of dialysis, or renal transplant within 6 months of the index ED visit.

An observation noted during the study: as initial estimated glomerular filtration rate declined, clinicians were less likely to order CT scans with contrast media enhancement and were nearly twice as likely to administer intravenous crystalloid fluids to patients undergoing contrast-enhanced CT than to patients undergoing unenhanced CT (24.3% and 14.2%, respectively).

Study Limitations

  • Single academic medical center ED with advanced radiology protocols and clinician practice patterns that may affect the overall incidence of acute kidney injury.
  • Majority of patients studied were admitted to the hospital.
  • Because patients requiring inpatient admission are more ill—and thus at potentially higher risk for developing acute kidney injury—than those discharged from the ED, results may overestimate the incidence of acute kidney injury in the general ED population.
  • Retrospective observational approach limited examination of comorbidities and outcomes to those recorded in the institutional electronic medical record, although electronic records for the study cohort were analyzed from 1993 through the end of 2014.
  • Although it is possible that patients included in the study sought treatment at other medical institutions for renal complications that were not recorded in the researcher’s electronic medical record, it is unlikely that this would occur in any particular patient cohort more than the others.
  • This analysis of clinical practice patterns was limited to the ED. Consequently, this analysis would not capture nephroprotective or nephrotoxic interventions performed after patients departed the ED, although researchers specifically excluded patients who received contrast-enhanced CT up to 72 hours after their ED departure.
  • While researchers attempted to minimize the bias associated with treatment assignment by using propensity score matching, this approach is limited by the inability to include all factors that might conceivably influence the clinical decision to administer contrast media.
  • All 3 patient groups analyzed were demographically similar, although
    • The unenhanced CT group was slightly older.
    • Patients in the contrast-enhanced CT group were less likely to have diabetes, congestive heart failure, or chronic kidney disease.
    • Initial serum creatinine values were similar across groups, although the contrast-enhanced CT group had a higher estimated glomerular filtration rate.

      The world’s first angiogram of an amputated hand from a dead body

Discussion

Historical studies of contrast-induced nephropathy have inferred causality from a temporal relationship between iodinated contrast media administration and occurrence of acute kidney injury.  The very definition of CIN, which relies on changes in renal function at an interval of 48-72 hours after contrast media administration, is based on this supposition. The majority of historical CIN studies were performed before the routine use of low- and iso-osmolar contrast media agents. Many were extrapolations from arterial angiography data and some did not include adequate control populations. This study demonstrates that for patients with initial serum creatinine levels less than 4.0 mg/dl, IV contrast administration is not associated with AKI, CKD, dialysis, or renal transplant.

In the ED, acute pathologies often independently affect the risk of developing AKI. The comparison of a population receiving IV contrast with two distinct control populations not receiving IV contrast media, in addition to the use of propensity-score-matching analysis, helps bypass these confounders, as well as helps to strengthen the findings of this study by minimizing selection bias inherent to retrospective analysis.

Nevertheless, given the study design, it is possible these conclusions inappropriately overstate the safety of IV contrast media administration. It is likely that nephroprotective measures are at least partially responsible for the observed lack of increased AKI incidence after contrast media administration, as suggested by the observation that clinicians were less likely to administer contrast media to patients with decreased baseline renal function and comorbid conditions associated with AKI and were more likely to administer IV fluids. A limitation of this statistical approach is the inability to account for all conceivable confounders, including clinician gestalt.

A well-controlled randomized prospective study is required to fully determine the contribution of IV contrast media to the development of AKI. These findings suggest that the randomization of patients to receive IV contrast, once considered ethically infeasible, is very likely safe (at least in patients with serum creatinine level less than 4.0 mg/dl). But consider a prospective RCT to study this research question. This would require that patients for whom contrast is apparently indicated be allocated randomly into a group either receiving or not receiving said indicated contrast. Alternatively, one could take healthy subjects not meeting indications for CT or advanced imaging at all, and expose one group to contrast while withholding it from the other. Good luck with the IRB approval for either of those.

I believe this study is very thorough in its analyses and conclusions even with its inherent limitations listed above, restriction to an ED setting, and the virtual impossibility to fully separate the effect of contrast media from the effects of the concurrent medical illness necessitating the contrast. Also, these findings suggest that the potential morbidity and mortality resulting from a failure to diagnose possibly life-threatening conditions outweigh any potential risk of contrast-induced nephropathy (again, in patients with serum creatinine level less than 4.0 mg/dl).

Also, propensity-score matching is itself an interesting feature worth discussion. The authors do admit to its limitations. Medpage Today has an interesting article on Propensity Scores, found here (tl;dr, he’s not a huge fan).

While we must remain judicious in our decisions to use not only contrast vs no contrast, one must consider the risk of acute kidney injury alongside the risk of missing a critical diagnosis. With the conclusions of this study and further emerging data on hand, I will now feel more prepared to advocate for my patient who would clearly benefit from contrast-enhancement on CT, despite what others might try to tout about historical beliefs.

 Will Brumley is a PGY2 EM/IM Resident