Acute kidney injury, contrast use vary by PCI operator

The findings of this study have important implications because the authors are revisiting an issue of contrast use that has not been looked at in much detail over the past decade. There has been interest in looking at contrast volume by creatinine clearance ratio and the impact of that ratio on outcomes, but this study is an important next step.

As with any study, particularly those using the NCDR, there are obvious advantages and significant limitations. The biggest advantage is the robust dataset. The NCDR has produced high-quality data for more than a decade, and the authors were able to include more than 1 million patients in their analysis.

One of the most interesting findings is the fact that the authors noted that as little as 75 mL of contrast was associated with an incremental risk in acute kidney injury. This is certainly a lower threshold than previously noted in other studies. In a commonly used scoring system (the Mehran Score), 100 mL is associated with a modest additive risk of AKI (the addition of 1 point of risk). Therefore, in light of the findings from this study, it looks like the emphasis is shifting toward lower contrast dosages that can cause AKI. It’s a provocative and interesting finding that even 75 mL of contrast is important.

The findings regarding the significant variation in contrast use were also notable. They were concordant with data from a Society for Cardiovascular Angiography and Interventions survey that showed a significant amount of variation in how we measure and handle contrast (Prasad A, et al. Catheter Cardiovasc Interv. 2016;doi:10.1002/ccd.26628).

There were also several comments and findings in the paper that are somewhat controversial. First, although the authors focus on contrast volume, they mention the modest benefit of hydration, which is a little debatable. Volume expansion before cardiac catheterization and contrast exposure is very important, especially in patients who are volume-contracted. Dilution of contrast material and increasing the rate of contrast excretion from the kidneys is important in minimizing the toxicity of dye.

Second, the authors attempt to examine lesion complexity. In registries, we tend to box patients into neat categories of lesion complexity, but there is a lot of subtlety within lesions. For example, left main PCI would be considered a complex lesion, but an ostial left main lesion might require one balloon inflation, one stent, one post-dilation and very little contrast dye. Similarly, for chronic total occlusion PCI, Brilakis and colleagues have found that patients can be divided into low-, medium- and high-contrast utilization groups and that patients in the high-contrast utilization group are the most troubling in terms of contrast nephropathy. The variation in use of contrast dye for CTOs is quite wide. Unfortunately, this can’t be deciphered simply from the NCDR dataset.

In addition, sheath size is very important in terms of the contrast dye being given, which was not adjusted for in this study. Larger catheters equal more contrast use. There has also been a lot of emphasis on the radial vs. femoral approach in the context of AKI. The data would suggest that people early in their experience with the radial approach use more contrast dye than with the femoral approach, but as their experience level improves, that difference disappears. In larger randomized controlled trials in which operators were not beginners, there is no clear signal that radial catheterization equates to renal harm. In fact, some data would suggest that the radial approach is associated with less contrast nephropathy. That difference between the radial and femoral approach was not really deciphered in this study.

The authors also did not mention the type of contrast dye, such as high- vs. low-osmolar contrast, used, which may be important. It’s controversial, but there are data that suggest that high-osmolar contrast may be beneficial in high-risk groups.

Finally, the variable of contrast volume itself can be problematic. Questions persist about the reliability of contrast volume as recorded in the NCDR or in any assessment. In one study, we found that we are not as good as we think at estimating the amount of contrast dye used (Prasad A, et al. J Invasive Cardiol. 2017;29:105-108). There is variability in how labs measure contrast volume. In some labs, physicians give the number, whereas in others, nurses or technologists do. Some also use marking of the contrast bottle before and after use to determine how much was used, so for the most part, it’s an estimate. Even the contrast bottles themselves have variable amounts of contrast dye in them, which can introduce error. In our study, we used technology for a contrast monitoring system that offered a more accurate way of accounting for contrast volume use, which is similar to what you would do for radiation and radiation exposure in the cath lab.

Generally, the fundamental problem is that you can’t randomize patients to receive high or low contrast. Even though we know contrast volume in the absolute sense or as a ratio of renal function influences outcomes, we have never been able to prospectively show that an intervention to decrease contrast dye may be beneficial to patients. The challenge is to design a study to really settle that issue. A prospective study to give operators tools to adjust their contrast use would be beneficial, as would an increased awareness in general of our contrast volumes to figure out when we are exceeding the thresholds that we know to be beneficial.