Clinical Trial Update

Disrupt CAD III: Lithotripsy for Vessel Preparation in Calcified Coronary Arteries Prior to Stenting

Dean Kereiakes, MD, FACC, FSCAI

Medical Director of The Christ Hospital Heart and Vascular Center and the Christ Hospital Research Institute, Cincinnati, Ohio; Professor of Clinical Medicine, The Ohio State University; Co-Principal Investigator of Disrupt CAD III

Dean Kereiakes, MD, FACC, FSCAI

Medical Director of The Christ Hospital Heart and Vascular Center and the Christ Hospital Research Institute, Cincinnati, Ohio; Professor of Clinical Medicine, The Ohio State University; Co-Principal Investigator of Disrupt CAD III

What do we know about the presence of calcium in the coronary and peripheral arteries?

We have seen an increase in the presence and severity of calcium. The best data that I have looked at in aggregate show that in the coronary space, up to 25% of patients have at least moderate to severe calcium. In the iliac and femoral arteries, it is up to 50% of patients, which then becomes an issue for large-bore catheter access. Below-the-knee (BTK), more than half of these patients are diabetic and up to 75% have moderate to severe vascular calcium. Alongside the increase in frequency and severity of calcification, the need for large-bore catheter access has expanded. The miniaturization of technologies has resulted in an explosive increase in procedures such as endovascular aneurysm repair (EVAR), thoracic endovascular aneurysm repair (TEVAR), transcatheter aortic valve replacement (TAVR), and the use of mechanical, cardiac support devices like Impella (Abiomed). Importantly, transfemoral access is the access of choice for these procedures and is associated with the lowest incidence of complications. Yet 50% of these patients have moderate to severe calcium. It is a perfect storm. Atheroablative technologies are available, but have significant limitations, and now we have a novel technology: intravascular lithotripsy. I call it a novel technology packaged in a not-novel delivery system — a balloon — which is the most primitive technology we use as interventionalists. Its ease of use, as demonstrated in the Disrupt CAD III trial, is remarkable for a new technology. There really is little or no learning curve. 

Can you describe how intravascular lithotripsy is used? 

Lithotripsy is a technology we have used for 30 years to treat kidney stones. It is packaged in a fluid-filled angioplasty balloon that is inflated to a low atmospheric pressure of 4 atmospheres, enough to fill the lumen and coapt the endoluminal surface of the vessel. At 4 atmospheres, 10 pulses are delivered, with one pulse per second, each a microsecond in duration. The pulses transmit the equivalent of at least 50 atmospheres of pressure in a circumferential and transmural fashion. Intravascular lithotripsy does not suffer from the limitations of atheroablative technologies, i.e., wire bias, since IVL offers the circumferential, transmural delivery of therapy. After delivering the 10 pulses (in the coronary; in the periphery, it is 20 pulses by protocol), the balloon is inflated from 4 atmospheres to a nominal balloon filling pressure of 6 atmospheres. The balloon size is chosen to have a one-to-one ratio with the reference vessel diameter (RVD). If incomplete balloon expansion is still present at 6 atmospheres, we deflate the balloon, reperfuse the patient, and repeat the cycle. Eventually, full balloon inflation can be achieved at only 6 atmospheres of pressure. Think about successful low-pressure balloon inflation in heavily calcified arteries. In fact, the maximum pressure in the IVL balloon in the Disrupt CAD III trial was only 6 atmospheres, which is very uncommon in calcified vessels. When you actually look at the vessels we treated in the trial, it is shocking (pun intended). The target lesions in the trial had an average arc of 293˚ of calcium, had a thickness of calcium of 0.96 mm, and an average calcified segment length of 48 mm. There has never been a cohort like that in any new device IDE trial in history. 

Can you tell us more about Disrupt CAD III?

It is a single arm, non-randomized clinical trial designed based on the precedent pivotal IDE trial for approval of orbital atherectomy, ORBIT II. ORBIT II designed performance goals from prior trials of calcified vessels to evaluate safety and effectiveness. The investigators really didn’t have a precedent atheroablative technology trial, so they constructed performance goals for safety and effectiveness, beat those performance goals in ORBIT II, and orbital atherectomy was approved by the FDA. ORBIT II was an ideal platform to adopt for this trial. We preferred not to randomize patients, because of technical procedural proficiency issues. Many younger physicians coming out of training don’t have the case volume experience of atherectomy necessary to be truly proficient, unless they come from a complex, high-risk, indicated procedure/patients (CHIP) program. So, as an alternative, we constructed performance goals with a single arm design for this trial. We sought to enroll a similar population, using similar endpoints and definitions as in ORBIT II, to provide a non-randomized comparison powered for non-inferiority to orbital atherectomy. The primary safety endpoint was freedom from major adverse cardiac events (MACE) at 30 days. MACE included death, target vessel myocardial infarction, and target vessel revascularization. The primary effectiveness endpoint was successful stent deployment with a <50% residual stenosis and no in-hospital MACE. The reason we picked <50% residual stenosis, although not a contemporary endpoint, is because that was the definition in ORBIT II. However, as a sensitivity analysis, we pre-specified an analysis of successful stent delivery using ≤30% residual stenosis evaluated by quantitative coronary analysis (QCA) and no in-hospital MACE. The definition of ≤30% residual stenosis is present in most of the drug-eluting stent trials done today and has been endorsed by the FDA. However, neither level of residual stenosis (<50% versus ≤30%) influenced the final conclusion, as the difference in endpoints was 92.4% versus 92.2%, respectively. We also created a central core laboratory for CK-MB, as many hospital labs no longer perform CK-MB. It was important to use the very same biomarker used in ORBIT II, with the same threshold (>3x upper limit of normal) for the diagnosis of periprocedural myocardial infarction. Some might argue that this definition is meaningless because the threshold is too low, but no matter how we cut the data, the conclusions are unchanged. For example, if we use the Society for Cardiovascular Angiography and Interventions (SCAI) definition for myocardial infarction, target vessel MI is only 2.6%, which further favors IVL. Because of the direct comparison with ORBIT II, we used the exact definition and threshold used in ORBIT II. Further, the non-inferiority margin was chosen based on FDA guidance. The enrollment criteria for Disrupt CAD III included patients with radiopacities on both sides of the vessel without contrast, at least 15 mm in length by angiography, or if intravascular imaging was done, there had to be at least a 270˚ arc of calcium. This is a “high bar”. For example, the calcium volume index (CVI) scoring system from Fujino et al3 uses an arc of calcium >180˚ as the most significant risk predictor for stent underexpansion. In this 4-point scoring system, calcium arc ≥180˚ gets two points, calcium length >5 mm is one point, and calcium thickness >0.5 mm adds one point. In target lesions with four points, the likelihood of stent underexpansion is very high. This scoring system was just incorporated into the SCAI consensus document4 for treatment of moderate to severely calcified vessels. We must acknowledge the importance of imaging in heavily calcified vessels to facilitate and optimize therapy, and emphasize that it is possible to predict which cases are most likely to have stent underexpansion, which is the number-one predictor of thrombosis and restenosis. We can predict underexpansion before we implant the stent, which is key. It’s always better to do this prospectively rather than retrospectively.

Will you be able to plan for IVL use then?

The IDE and FDA approval will not be tied to pre-procedural imaging, because most labs and operators don’t do routine imaging. Personally, I have been someone who didn’t do a lot of imaging previously, except with in-stent restenosis. In particular, I used intravascular ultrasound (IVUS), but this trial really opened our eyes. The use of optical coherence tomography (OCT) in DISRUPT CAD III establishes, as definitively as possible, the mechanism of action of OCT technology, which for me was probably the most insightful part of the entire trial. 

What were you looking for in the trial’s OCT sub study?

We looked at two things. One was diagnostic, documenting the presence and severity of calcium for diagnostic and prognostic purposes. Second, we wanted to demonstrate the mechanism of IVL action and assess its efficacy for optimizing stent deployment: what were our results after stenting and how do these results compare with other studies. Specific measures for stent optimization included area stenosis, minimum stent area (MSA), and the degree of stent expansion. It was important to establish exactly who these patients were by identifying the site of maximum calcium in the vessel. The site of maximum calcification was that site with the biggest calcium arc. If two sites had the same arc, for example, 290˚, then the “tie-breaker” was the maximum calcium thickness. Disrupt CAD III enrolled a very complex, severely diseased group of patients, with an average calcium arc of 293˚ and 0.96 millimeter in thickness. Further, the average calcified segment length of 48 mm is remarkable. In ORBIT II, the average calcified segment was only 28.6 mm. Although we set out to enroll similar patients, we actually got much worse patients with respect to target lesion complexity, and yet we achieved better clinical results. I want to hedge that statement by saying that it is not appropriate to do cross-trial comparisons and Disrupt CAD III wasn’t randomized. Yet the MACE rate at 30 days was 7.8% versus 10.4% in ORBIT II using the same endpoints and definitions. Following IVL, vessel perforation, abrupt closure, and no reflow were not observed. One patient had a micro perforation following subsequent stent deployment, but did not require pericardiocentesis. There was no pericardial effusion identified and no MACE to 30 days. A prolonged balloon inflation took care of it. This is a very different perforation than the type we often see with atherectomy, where it is usually not that simple.

Can you talk about what happens to calcium after treatment with IVL and after use of an atheroablative technology?

Atheroablative technologies pulverize calcium into microparticles that are less than the size of a red blood cell. Much of this particulate debris is at the 5 micron level and is absorbed by the reticuloendothelial system. However, we know that the phenomenon of slow flow/no-reflow is associated with atheroablation and is, at least in part, related to embolic plaque material and possibly thrombus as well. This can lead to downstream, microvascular obstruction. I always pre-treat with nicardipine before doing atheroablative procedures to prevent slow-flow/no-reflow. IVL doesn’t have this problem. Patients regularly ask me, where does the calcium go? And if you crack it, what happens? The answer is, it is embedded in the vessel wall and is fractured in situ (doesn’t break off) by IVL. IVL-induced calcium fracture improves vessel compliance. Following IVL, there is a significant increase in the minimal lumen diameter (MLD) with only a 6-atmosphere inflation.2 When the calcium is fractured, not only is transmural compliance increased, but remarkably, fibro-elastic recoil is reduced. Every interventionalist experiences recoil following balloon angioplasty. With IVL, you will be amazed at how many times you see a stent-like result following a low atmosphere pressure inflation. It is a real advantage of this technology, along with the lack of wire bias. With atheroablative technologies, the only way to modify deep calcium is if there is extreme wire bias, which creates an eccentric rut or trough, or by using a bigger burr. Bigger burrs are associated with an increased incidence of perforation or dissection, which means worse outcomes. Most of us will use the smallest burr necessary to alter the compliance and get full balloon expansion. With intravascular lithotripsy, you match the balloon size to the RVD, and utilize a lower atmosphere inflation pressure. It is a very different experience.

What should we focus on regarding intravascular lithotripsy based on the results of Disrupt CAD III? 

IVL’s ease of use and safety, and its mechanism of action, which was circumferential, longitudinal, multiplanar calcium fracture. Stent implantation is associated with fracture expansion, and is the mechanism of increased transmural compliance. Imagine using only a 6 atmosphere inflation in patients with an average calcium arc of 293˚ and an average calcium thickness of 0.96 mm. With IVL, we were able to achieve full balloon inflation at 6 atmospheres pressure and achieve 102% average stent expansion in the zone of maximum calcium. Based on this, one would predict not only a great acute result, but excellent late results as well, so I am excited about the one-year follow-up results. I think we are going to find that when you get average stent expansion >100% and a minimum stent area of 6.5 mm2, that these measures will predict excellent long-term clinical results for patients.

What do you envision for the future with this technology?

We find calcium not just in the coronaries, but throughout the vasculature. Operators will gravitate to IVL due to its ease of use. In fact, we tracked the first case at each one of the 47 international sites in the trial (roll-in patients), compared to the next 384 cases. Remarkably, performance parameters such as balloon crossing, procedure success, and freedom from MACE at 30 days were all basically the same. That is a remarkable statement regarding IVL ease of use. This technology uses a balloon, the most primitive technology from the time vascular intervention began in the days of Andreas Gruentzig in the 1970s. IVL offers predictability, reliability, safety, and ease of use. On top of all that, it is very effective. 

Disclosure: Dr. Kereiakes reports he is a consultant to Shockwave Medical.

Dean Kereiakes, MD, can be contacted at:

  1. Kereiakes D. Intravascular lithotripsy for treatment of severely calcified coronary artery disease: the Disrupt CAD III study. Lecture presented at Transcatheter Cardiovascular Therapeutics (TCT) 2020; October 15, 2020. Available online at Accessed December 8, 2020.
  2. Hill JM, Kereiakes DJ, Shlofmitz RA, Klein AJ, Riley RF, Price MJ, Herrmann HC, Bachinsky W, Waksman R, Stone GW; Disrupt CAD III Investigators. Intravascular lithotripsy for treatment of severely calcified coronary artery disease. J Am Coll Cardiol. 2020 Dec 1;76(22):2635-2646. doi: 10.1016/j.jacc.2020.09.603. 
  3. Fujino A, Mintz GS, Matsumura M, et al. A new optical coherence tomography-based calcium scoring system to predict stent underexpansion. EuroIntervention. 2018 Apr 6;13(18):e2182-e2189. doi: 10.4244/EIJ-D-17-00962. 
  4. Riley RF, Henry TD, Mahmud E, Kirtane AJ, et al. SCAI position statement on optimal percutaneous coronary interventional therapy for complex coronary artery disease. Catheter Cardiovasc Interv. 2020 Aug;96(2):346-362. doi: 10.1002/ccd.28994.