Don't Kill the Beans: The Specter of Contrast-Induced Nephropathy

Working in the Emergency Department, we often encounter patients with either pre-existing renal disease or an acute compromise of their renal function who also have a disease process necessitating a contrasted radiology study.  So what do we do with that patient with a creatinine of 1.8 who has a possible vascular dissection/traumatic injury/infection? What is the risk of contrast to that patient?  Should you compromise your diagnostic evaluation to avoid a harm to the patient's renal function?  Dr. Nick Ludmer, Dr Michael Miller, and Dr. Amanda Polsinelli recap 3 articles recently published looking into contrast induced nephropathy.  Take a listen to the podcast and read the blog post to get yourself acquainted with the current state of the literature.

Garfinkle MA, Stewart S, Basi R. Incidence of CT Contrast Agent–Induced Nephropathy: Toward a More Accurate Estimation. American Journal of Roentgenology 2015;204(6):1146–51. 

Recap by Nick Ludmer, MD PGY-3

This study was a retrospective cohort study, performed within a single health care system composed of three hospitals.  The primary outcomes were twofold:  the incidence of acute kidney injury following contrast enhanced CTs compared to non-contrasted studies, and the incidence of dialysis and kidney failure in contrast enhanced CTs compared to non-contrasted studies.  They then also subdivided the study population by underlying GFR in order to assess whether there was an increased incidence of AKI and dialysis in patients with varying degrees of underlying kidney disease.

Between 2006 and 2013, there were 2583 patients that met their inclusion criteria.  They found that there was no statistical difference overall between incidence of AKI in the contrasted study group vs. the non-contrasted study group.  They also found that there was no statistical difference between the need for dialysis following contrasted studies and non-contrasted studies.  When sub-divided by baseline GFR (which was determined by pre-study BMP), they found no statistical difference in incidence of AKI or the need for dialysis in any of the sub-groups after receiving either a contrasted or non-contrast study.

Overall, we felt this was a well designed study, and felt their approach to examining the effect of contrasted studies on patients with varying degrees of underlying kidney disease was well executed.  The results tend to mirror some of the larger retrospective cohorts performed in the radiology, nephrology, and now emergency medicine literature. 

One of the key drawbacks to this study, and many of the similar articles published recently on this topic, is that they are all retrospective cohort studies.  This is due to the fact that it would be unethical to prospectively randomize a patient to receive a sub-optimal study for their disease process.  As a result, there exists a large selection bias which may underestimate the true incidence of contrast induced nephropathy (AKI and dialysis secondary to IV contrast use). 

Another aspect of the study that is important to recognize, and the authors do a good job highlighting this in their own discussion section, is that the number of patients with a GFR of 15-29 and <15 were too low to draw any conclusions.  The confidence intervals for these groups were as high as 20-25, making it impossible to draw any conclusion for these groups.  Data for patients with a GFR > 30 however, was well powered enough to say that there was no statistical difference in incidence of AKI and need for dialysis between contrast enhanced and non-contrast studies.  

Our take away from this study is that for patients with patients with decent kidney function (GFR > 30) the incidence of AKI and dialysis is fairly low, and we can possibly be more liberal with our use of contrast in these patients.  However, the incidence is not 0, and there does appear to be a trend of increasing incidence with worsening underlying kidney function.  Though this was not statistically significant between contrasted and non-contrast studies, the n for patients with GFR < 30 was extremely low making it difficult to trust the results in this subgroup.  Therefore, we cannot ignore a patient’s kidney function completely, and we should still exercise caution when considering contrasted studies in patients with moderate to severe underlying kidney disease.

PhD JSHM, MS MREMM, MD DMF, et al. Risk of Acute Kidney Injury After Intravenous Contrast Media Administration. Annals of Emergency Medicine 2017;:1–14. 

Recap by Michael Miller, MD PGY-3

This was a single-center retrospective cohort analysis where the data was collected from 2009-2014 in an urban academic emergency department with an average yearly census of approximately 62,000 patients. The question addressed by this article was, “how often does acute kidney injury occur after enhanced and non-enhanced CT scans performed in the Emergency Department?”

The primary outcome was the incidence of acute kidney injury following a CT scan with intravenous contrast, wherein standard definitions of contrast induced nephropathy (CIN) and acute kidney injury (AKI) were used. Secondary outcomes included new chronic kidney disease, dialysis, and renal transplantation at 6 months. Secondary outcomes were determined using ICD-9 coding for procedures or diagnosis.

A multivariable logistic regression model and between groups odds ratio with and without propensity score matching was used to test for an independent association between contrast administration and the primary/secondary outcomes. To minimize bias, the author used two distinct control populations that did not receive contrast (CT scan without contrast group and a no CT scan at all group) then analyzed large numbers of patients in all subgroups of baseline renal function.

For the primary outcome, contrast administration was not associated with an increased incidence of acute kidney injury (CIN criteria OR = 0.96, 95% CI 0.85-1.08 and AKI criteria OR = 1.00, 95% CI 0.87-1.16.) This was true for all subgroup analysis regardless of baseline renal function and whether comparisons were made directly or after propensity matching.

For secondary outcome, contrast administration was not associated with increased incidence of chronic kidney disease, dialysis, or renal transplant at 6 months. Other findings included: clinicians were less likely to prescribe contrast to patients with decreased renal function and more likely to prescribe intravenous fluids if contrast was administered.

Overall, our journal club group felt this was a well-designed study with an important reinforcement of our current practice patterns. To expand on this, the consensus was that while AKI/CIN from contrast enhanced CT scans certainly occurs - it probably occurs less frequently then older articles would suggest so in the face of potentially life-threatening pathology (aortic dissection, pulmonary embolism, embolic stroke etc) benefit likely outweighs risk. Our group also discussed potential future studies that could be spun from this article including evaluating code stroke CTA head and neck with IV contrast as this patient population receives IV contrast irrespective of underlying kidney function and analyzing this information could further add to the collection of data regarding AKI/CIN incidence following contrast enhanced CT scans.

MSc ECN, MD RJR, MD PJN, et al. Prophylactic hydration to protect renal function from intravascular iodinated contrast material in patients at high risk of contrast-induced nephropathy (AMACING): a prospective, randomised, phase 3, controlled, open-label, non-inferiority trial. The Lancet 2017;389(10076):1312–22.

Recap by Amanda Polsinelli, MD PGY-3

The AMACING trial is a single center, randomized, prospective, parallel-group, open-label non-inferiority study comparing pre-hydration for those at high risk for CIN with no pre-hydration. The inclusion criteria for this study were in line with current European guidelines of those at high risk of CIN. Excluded from the study were those with a very low eFGR of less than 30, those on dialysis and those in the ICU. Also, importantly, this test was done on those referred in for either a contrasted scan or vascular interventional study, and not on those presenting to the emergency department.

The primary end points were incidence of CIN as defined by an increase in creatinine by 25% within 5 days of the study and cost-effectiveness. The secondary end points were major adverse events.

Ultimately, 328 patients were randomly assigned to the hydration group and 332 patients were assigned to the no-hydration group. A mean of 1637 mL of normal saline was used for hydration and low-osmolar iodinated contrast was used in every patient. Creatinine was followed up at 2-6 days post-contrast and at 26-35 days post-contrast.

The incidence of CIN was 2.7% in the hydration group and 2.6% in the no-hydration group with the upper limit of the confidence interval being below the cut off for non-inferiority of 2.1%. Additionally, it was found that there was significant cost-effectiveness in the no-hydration group particularly when it came to hospital stay charges. There were no major renal events and no admissions to ICU in either group. There were 13 patients, or 4%, in the hydration group who experienced complications related to hydration and no events in the no-hydration group.

Ultimately, the study found that no pre-hydration prophylactic treatment is non-inferior to pre-hydration and there is a cost benefit with no-hydration as related to hospital stay. Finally, pre-hydration therapy was associated with additional risks of fluid overload and arrhythmias.

The strength of this study is the prospective randomized design comparing our most common pre-treatment with no-pretreatment. The disadvantages to this study are primarily that it is a single center study done with non-emergent imaging. Hypotension or other physiologic stresses on our population in the emergency department may have dynamic effects on CIN that will be missed in this study. Additionally, this study used an eGFR cutoff of 30 for safety considerations so this cannot be applied to those with that degree of renal failure.

This is probably not a game changing study for us in the emergency department given these limitations, but it may lend some level of reassurance when we obtain contrasted studies in those with CKD for can’t-miss pathologies such as dissection. 

Graphics by Jeffery Hill, MD MEd Licensed under Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.