ADVERTISEMENT
Setting Contrast Thresholds in High-Risk Patients
What prompted you to study the association between contrast volume and acute kidney injury (AKI) (otherwise known as contrast-induced nephropathy [CIN])?
We had previously been using the maximal allowable contrast dose (MACD) and realized that there were a large number of patients who were developing AKI even when MACD was not exceeded. Since contrast media is excreted by the kidneys, it makes biological sense to dose patients based on renal function (as we currently do with all drugs that are excreted by the kidney). The logical next step was to assess the association between renal function-based contrast dose and the risk of AKI.
What has your research1,2 told you about the association between contrast volume and AKI?
It is actually a non-linear relationship, with the risk of AKI starting to increase when the contrast volume exceeds 2x the creatinine clearance and the risk increases significantly when the dose exceeds 3x the creatinine clearance. When even higher contrast doses are used (e.g., 5x creatinine clearance), the risk increases exponentially.
Which patients are most at risk for AKI and how should they be identified in the cath lab?
The patients at the highest risk of AKI are those with reduced renal function (abnormal glomerular filtration rate [GFR] or creatinine clearance) and those with hemodynamic instability. There are other factors such as diabetes, dehydration, high right atrial pressure, and congestive heart failure, that also increase the risk of AKI. We previously described the BMC2 risk model that uses a random forest to calculate the risk of AKI and need for dialysis (along with death and need for transfusion) using 18 routinely available clinical factors. These models are available as a web tool (https://bmc2.org/calculators/multi) and as a free app (https://www.scaipciriskapp.org/pci_welcome) supported by the Society for Cardiovascular Angiography and Interventions (SCAI), and are currently in use at many hospitals for pre-procedural risk stratification. Some hospitals, including ours, have embedded the risk models in the electronic medical record.
What is the hospital economic impact of contrast-induced AKI?
The occurrence of AKI has been estimated to add approximately $10,000-$12,000 to the in-hospital cost of a patient undergoing PCI.
Why has the contrast threshold become part of the Blue Cross Blue Shield of Michigan Cardiovascular Consortium (BMC2) best practice guidelines?
This is a simple tool that is easy to incorporate into clinical practice. Acknowledgement of the contrast threshold makes the entire cath lab staff cognizant of the importance of the contrast dosing to the patient and operators are able to stay under the contrast threshold. We have found that use of this ratio results in a reduction in the total amount of contrast use delivered to the patient and reduces the risk of AKI.
How is it currently being used across Michigan within the BMC2 Consortium?
The current BMC2 best practice protocols recommend the use of contrast thresholds (2x and 3x GFR) in procedure time outs and this is a common practice at most Michigan hospitals. The proportion of patients where the thresholds are exceeded is tracked on an ongoing basis, and this is included in both quarterly and annual BMC2 reports.
How is it currently being used across the United States?
I know of some hospitals that are using GFR-based dosing as part of their routine practice. However, I do not think it is a widespread practice outside Michigan and there is an opportunity to broadly apply this easy-to-implement quality improvement tool.
How should cath labs track their contrast use and track to an established threshold during procedures?
The first step is to routinely assess patient’s renal function and contrast thresholds as part of procedure time out. The total amount of contrast used and the proportion of patients where the contrast threshold is exceeded should be tracked as part of the routine quality improvement process. Pre procedural assessment of GFR and the amount of contrast volume used is an ACC/AHA/SCAI/AMA-PCPI/NCQA 2013 PCI Performance Measure. Some labs are using dedicated devices that have recently become available for this purpose.
How do you incorporate your cath lab team into establishing and managing to cardiovascular thresholds during procedures?
Quality requires team support and the prevention of AKI (and other complications) can only be successful if everyone in the catheterization laboratory is invested in continuous quality improvement. Our nurses will reconfirm the threshold before we start the procedure and the circulating technologist will warn us if we are getting close to the threshold.
What recommendations do you have to minimize contrast usage?
I routinely use 5 French (F) catheters and biplane angiography to minimize contrast in patients with moderate or high risk of AKI. It is prudent to forego LV grams/aortography in these procedures. In high-risk patients, we and others have performed procedures with minimal contrast volume. In the recent months, we have started using the Dyevert Plus System (Osprey Medical) in the appropriate patient population since it has been demonstrated to significantly reduce the total amount of contrast that is administered to the patient.
What do you think about the technologies available to track and reduce contrast?
There is currently only one company, Osprey Medical, that markets devices specifically focused at contrast reduction and contrast monitoring. We use the DyeVert Plus in high-risk patients given the clinically significant reduction in contrast volume that has been previously demonstrated in controlled trials. Our experience appears to corroborate findings from those studies and the device does save significant contrast volume to the patient in each case. My hope is that there will be further innovation in this field and we will have a variety of tools that will help reduce the clinical challenges associated with contrast-induced kidney injury.
What role does hydration have in AKI prevention? What type of hydration protocol is recommended?
Hydration has been the cornerstone of CIN/AKI prevention for a long time. We start with oral hydration and then follow up with left ventricular end-diastolic pressure (LVEDP) hydration in the cath lab. The “Oral Hydration Guideline” on the following page shows the BMC2 hydration guideline (also available on our website at https://bmc2.org/system/files/private/best-practice-protocols-5- 20-14.pdf).
What about monitoring/determining a contrast threshold in ST-elevation myocardial infarction (STEMI) patients?
STEMI patients are a complicated issue. One, you often don’t have the luxury of hydrating them prior to the procedure, because the obvious priority is to get the artery open — but we almost always get the serum creatinine. If not before we start the procedure, often serum creatinine is available to us during the procedure. If you have access to serum creatinine prior to the procedure, you can take measures that will allow you to reduce the amount of contrast. For example, many operators including myself never do an LV gram if they know the patient’s creatinine is high and their creatinine clearance is low. You can then assess LVEDP at the end of the case and use that to determine how much hydration to use — we use the POSEIDON (Prevention of Contrast Renal Injury with Different Hydration Strategies) trial3 equation. It is supposed to be used in conjunction with pre procedural hydration, but in the case of a STEMI, we use the EDP and hydrate after the procedure based on that. There is a group doing a study with the DyeVert Plus System in STEMI, to see if that would help, and it is something where we would have to wait for the data.
DyeVert reduces contrast use, but also tracks how much contrast you are using. How do you measure when you are not using the device?
For now, we use the DyeVert Plus only in high-risk patients, so for the majority of our patients, we guesstimate the amount of contrast used, and there is always a little bit of wishful thinking that occurs. I do think we underestimate how much contrast we use in these cases, because we look at the bottle and think, I probably used 40 ml and I wasted [x] amount. When you measure it, the amount that you used is usually more than you thought. I think DyeVert Plus makes you more honest. You know you are being tracked and you are getting that injection-by-injection feedback with the device, so you might forego some injections. If your GFR is 40, we try and stay within 80ml of total contrast use. If we are not using the DyeVert Plus to show us where we are with contrast use (often we won’t have the device for somebody with a GFR of 40), then I will ask the team to let me know when I am approaching 80 ml, and the goal is to finish the case under 120 ml. Once you are aware, it is very easy to stay within that boundary.
Knowing you have a contrast limit, how much does it help to mentally plan out your procedure beforehand, in whatever time is available?
Once you are thinking about it, the procedure often becomes very straightforward in the sense that you can choose to use other strategies. There are some groups that have done PCIs even without using any contrast, if they have a diagnostic image on hand. I will use intravascular ultrasound (IVUS) particularly in high-risk patients, because IVUS can help me better size my stent, and you can use it pre and post stenting to check your result. Even after IVUS, I always like to have at least one picture that shows me that there is no perforation and good outflow. I have never been able to walk away from that, but you can almost always stay under the contrast limit even in the sickest patients.
What about expanding the contrast threshold into the peripheral realm?
Yes, peripheral procedures actually have many advantages in terms of contrast reduction abilities. You can do a lot of procedures without repeating multiple injections. You can use roadmapping and image overlay. When stenting iliacs, I will often use IVUS. You can do a peripheral case with very little contrast. You can also combine it with CO2 angiography. At our center, we will treat a fairly large number of patients whose creatinine clearance is very low, and with appropriate hydration and planning, we are able to do these cases and have a very low contrast-induced kidney injury rate. What we have found is that this ratio seems to hold its value, the cutoff of 3x especially, even in peripherals. Michael Grossman, one of my colleagues here, published a study showing that if you look at the risk of kidney injury in patients undergoing peripheral angiography, a cutoff of 3x the GFR is a very strong predictor. If you are below that cutoff with your contrast volume, your risk is low for kidney injury; if you exceed it, your risk is high.
Can you tell us more about the Blue Cross Blue Shield of Michigan Cardiovascular Consortium (BMC2) Registry and how it is similar/different from the NCDR CathPCI Registry?
BMC2 is a quality improvement collaborative of all hospitals that perform PCI in the State of Michigan. It is funded by Blue Cross Blue Shield of Michigan to help improve quality of care of patients undergoing percutaneous coronary intervention (PCI). We collect data using both the American College of Cardiology’s National Cardiovascular Data Registry (NCDR) platform and additional data using the BMC2 website. The registry is locally run and there is strong physician involvement in defining the quality agenda. There is a clearly defined physician champion at each site who is responsible for enhancing the safety and quality of PCI at their institution. The registry works as a learning network with hospitals sharing best practices and focusing on collaborative quality improvement. There are detailed reports that not only provide the usual rates and proportions, but specifically highlight where an institution and an operator is an outlier. There is a cross-institutional peer review process whereby blinded case reviews are performed on randomly selected cases by operators at another institution. The goal of this exercise is to help improve the safety, quality, and appropriateness of PCI.
Who participates in the BMC2 Registry?
All non-federal hospitals in the state that perform PCI participate in BMC2.
What is your role with the BMC2 Registry?
I am the project director and have the privilege of representing a large group of dedicated colleagues who are all focused on improving quality of interventional care.
How representative/translatable are these data (BMC2) to the broader U.S. population?
The results of our work are directly translatable to the broader U.S. population. BMC2 has a robust data collection process with ongoing education of the coordinators and annual (or more frequent) audit of all institutions. This ensures that the data quality is much better than in any other PCI registry. We have data on all the patients and thus our findings are broadly generalizable.
Could talk more about your experience working within the BMC2 framework?
BMC2 is unique. It started 2 decades ago with 5 hospitals, and was funded by Blue Cross Blue Shield of Michigan with the only goal being quality improvement. They began with the question, can you take people who are competitors for the same business, and get them to work together to figure out how to improve quality, safety, and appropriateness, create benchmarks, and then look at each other’s practices, and learn from each other? Today BMC2 covers the entire state. It is a very collaborative group with regular meetings. People use this as an opportunity to bring up issues that they are having at their hospital. Others will say, we had this problem and we did x, y, z, to solve this problem. That then leads to sharing the best practice, which people then start to use across their health system. I’ll give you an example. One issue we are very interested in is contrast nephropathy. We have a small hospital in the Beaumont Health System, Beaumont Troy, and they have been pioneers in making sure patients are very well hydrated, with both oral hydration and IV hydration, prior to their procedure. People have adopted these protocols at their own hospitals. As a result, we have seen a decline in the number of patients who come to the cath lab without getting hydrated and a decline in contrast nephropathy. There are physicians in BMC2 who are passionate about certain quality issues. We take what they learn and implement, and share it across the entire collaborative. Recently we focused on appropriateness and implemented a peer review system that started last year. Cases are selected at random from all hospitals and are sent in a blinded fashion onto a server. Reviewers from different hospitals are advised to look at the cases and then the blinded reviews are provided back to the original hospitals. It has been very well received. Sometimes, operators working in one hospital adapt a similar style. An external review offers answer to questions such as, am I doing the right thing? Should I be doing things differently? Am I using the right pharmacotherapy? Some of our large hospital systems want to use the same platform to review even a larger number of their cases internally, with the idea of using peer review to drive their own internal quality improvement. We are very excited about that. It is done in a way that is all about collaboration and improving quality. We discussed public reporting but moved away from it, based on the feedback we were receiving from our internal group. The concern raised was that once you start reporting data publicly, then people start focusing on the data in a way to make it look good, versus being more honest and open about it. Internally, these data are dissected in an open fashion. We present the data at a meeting in February every year and look at 4 or 5 different complications. Hospitals that are doing well are invited to present as well as those hospitals that are not doing well. For example, if mortality is more than expected, people will present, be very open about it, and say, we took this patient with metastatic lung cancer and he died from hemoptysis, and in hindsight, we probably shouldn’t have offered him the intervention. It allows everyone to reflect on what happened in the hospital that didn’t have a good outcome, but other people are able to say, we had this issue a few years ago and here is how we tackled it. It leads to a very positive dialogue and helps to improve quality all around.
Can you tell us about your lab and hospitals?
I work out of two hospitals: the University of Michigan cath lab and the VA. We are a busy cath lab with 8 interventional cardiologists. All of us do coronaries, but we have tried to develop specialization for some procedures: we have 3 interventionalists who do structural interventions, 2 who do chronic total occlusions (CTOs), and 4 who do peripheral procedures. Between the two systems, we do about 1000 PCIs a year. We do about 350 structural interventions a year and we just did our 1000th transcatheter aortic valve replacement (TAVR).
References
- Gurm HS, Dixon SR, Smith DE, et al. Renal function-based contrast dosing to define safe limits of radiographic contrast media in patients undergoing percutaneous coronary interventions. J Am Coll Cardiol. 2011 Aug 23; 58(9): 907-914. doi: 10.1016/j.jacc.2011.05.023.
- Gurm HS, Seth M, Mehran R, et al. Impact of contrast dose reduction on incidence of acute kidney injury (AKI) among patients undergoing PCI: a modeling study. J Invasive Cardiol. 2016 Apr; 28(4): 142-146.
- Brar SS, Aharonian V, Mansukhani P, et al. Haemodynamic-guided fluid administration for the prevention of contrast-induced acute kidney injury: the POSEIDON randomised controlled trial. Lancet. 2014; 383: 1814-1823.
- Prolonged fasting prior to coronary angiography is unnecessary. Different institutions have different rules for fasting prior to conscious sedation. The possibility of emergency surgery and need for general anesthesia (although exceedingly rare) is the main reason that patients are told to fast from midnight. In patients with normal gastric emptying, current guidelines recommend that patients not partake of fatty food for 8 hours, light food and milk for 6 hours, and clear liquids for 2 hours. Clear liquids include water, carbonated beverages, black coffee, or clear tea.1 Alcohol should specifically be avoided.
- Oral hydration with tap water should be encouraged before and after the procedure. Studies that have demonstrated a benefit have required the patients to drink up to 2 hours prior to the procedure. Further, studies that have permitted free fluids have not demonstrated as much benefit as those that mandated a certain volume of water intake. This is probably due to the fact that patients do not feel the desire to drink enough to have a brisk urinary output. While there are no data that have compared different oral hydration strategies, 500 ml of water started 4 hours prior and stopped 2 hours before the procedure, or 1 ml/kg/hour orally started 6-12 hours before the procedure and continued up to 2 hours before the procedure are two regimens that have been demonstrated to be as effective as intravenous normal saline.2 Another trial required patients to take oral sodium chloride for 2 days prior to the procedure.
- Patients should be told to avoid a fatty meal for 8 hours and solids and milk for 6 hours, but encouraged to drink water or a sodium-rich fluid (such as Gatorade) for up to 2 hours before the procedure. Patients should be specifically encouraged to drink half liter of water between 4-2 hours prior to the procedure. An easy instruction would be to ask patients to drink 1 glass of water (or sodium-containing clear fluid) every hour up to 2 hours before the procedure.
- Free fluid intake may not be appropriate in patients with delayed gastric emptying, or those with decompensated heart failure or clinical evidence of volume overload. However, free fluids can be used in almost all other patients.
- Institutions should continue to use intravenous hydration in addition to oral hydration.
- American Society of Anesthesiologists Committee. Practice guidelines for preoperative fasting and the use of pharmacologic agents to reduce the risk of pulmonary aspiration: application to healthy patients undergoing elective procedures: an updated report by the American Society of Anesthesiologists Committee on Standards and Practice Parameters. Anesthesiology. 2011; 114: 495-511.
- Agarwal SK, Mohareb S, Patel A, et al. Systematic oral hydration with water is similar to parenteral hydration for prevention of contrast-induced nephropathy: an updated meta-analysis of randomised clinical data. Open Heart. 2015 Oct 5; 2(1): e000317. doi: 10.1136/openhrt-2015-000317.
- Brar SS, Aharonian V, Mansukhani P, et al. Haemodynamic-guided fluid administration for the prevention of contrast-induced acute kidney injury: the POSEIDON randomised controlled trial. Lancet. 2014; 383: 1814-1823.
