Speaker: Gary Roubin, MD, PhD
The CREST Trial was the largest, most rigorous and because of the way it was conducted, the most relevant investigation into the role of carotid stenting to date.
The National Institute of Neurological Diseases after critical review concluded that the study “Demonstrated that Endarterectomy and Stenting were equally efficacious methods of preventing stroke caused by carotid bifurcation stenoses.”
The primary endpoint was unequivocal but the components of this combined endpoint have been dissected by various groups to support different conclusions. Vascular surgeons have pointed to more minor strokes in the stenting group and cardiologists to more myocardial infarctions in the CEA cohort. The CREST Trial demonstrated remarkable safety from both procedures with a very low and similar major stroke and death rate. The small numbers of excess strokes in the stenting group were minor strokes and importantly further analyses of temporal trends have demonstrated this delta disappeared over the course of the study. Stenting stroke rates improved over time probably related to better selection of younger patients with more suitable anatomy for stenting.
If CREST was to restart in 2012, it is extremely unlikely that any difference whatsoever would be seen in comparing CEA and Stenting.
Importantly, minor strokes were not associated with a later excess mortality while a periprocedural MI was associated with death over the follow up period.
Quality of life analyses reflected the minor, non disabling nature of the small number of excess minor strokes. The comprehensive panel of SF36 mental and physical quality of life measures demonstrated no difference whatsoever between stenting and CEA.
Despite completing the study with first generation stents and embolic protection devices, the outcomes were gratifying. A critical FDA panel subsequently approved the extension of labeling for stenting use in standard risk CEA patients.
Now we await a considered response from CMS to acknowledge the demonstration of “reasonable safety and efficacy” and long awaited reimbursement for this patient friendly, percutaneous procedure.
We now are experiencing a curious push back from some in the neurological community and even some surgeons who argue that neither CEA nor stenting are needed in the treatment of asymptomatic patients.
· This—despite multiple trials that have demonstrated the superiority of revascularization and markedly improved revascularization results.
· This—despite no scientific evidence to support the equivalence of medical therapy in preventing stroke in carotid bifurcation disease.
The CREST data stand on their own merits. Looking at the survival curves for minor stroke versus myocardial infarction, as a surgeon who does a large volume of CEA and stenting, I am impressed with the benign outcome in the minor strokes and the bad outcomes associated with M.I. Again, as a surgeon, I doubt we can do anything to reduce the incidence of MI, but as a stent operator, I feel we can do much more to further reduce the incidence of stroke events.
For example, the neurological community has focused on the ICSS Trial sub-study that demonstrated a significant incidence of MRI-DWI defects after stenting. We don’t really know what DWI changes mean but the neurological community assumes they are bad. We see a 15-20% incidence with just a routine angiogram and they are probably just micro-bubbles causing these temporary defects. In the ICSS trials, the incidence of these lesions was 50% in the stent arm and 12% in the CEA arm. The conclusion was that stenting was “bad” but embolic protection devices were only used in 75% of the ICSS patients. Now we have new devices such as proximal occlusion balloons that have been shown to markedly reduce the incidence of these lesions. So, this is just one example of new stenting techniques that will reduce the incidence of stroke to even lower rates. There is also a lot of activity in the industry to make carotid stents covered with a fine, semi-permeable membrane that will reduce the chance of embolic debris from the procedure. With current devices and certainly the stent used in CREST, debris may be forced through the stent strut when you dilate.
So to me these are just two examples of things that will improve the stroke rate from the current 1% to 3% to near zero.
I am not sure what happened at the Medicare Coverage Advisory Meeting last week (January 2012) but basically the committee did not appear to focus on the CREST data.
One of the most important differentiating features of CREST compared to the European Trials was the emphasis on operator credentialing and ongoing training of operators over the 8 years of recruitment. This is evidenced by the improvement of stent outcomes over the time course of the trial.
I also want to point out that none of these trials place enough emphasis on cranial nerve injuries that are an exclusive and important complication of CEA. In CREST there was a 4.5% incidence of cranial nerve palsy in the CEA cohort, and 2% were still present at 6 months.
I firmly believe that CREST is a “game changer.” I spent a full day at the Medicare Coverage Advisory Committee last week (January 2012). Two things happened. The first was that the entire discussion was derailed by irrelevant discussion of the supposed value of medical therapy. As Ty said before, nobody appeared to be focused on the CREST data but was distracted by arguments about the value of medical management. A neurologist, Anne Abbott, took a large amount of time basically “trouncing” any type of revascularization therapy.
The critical consideration by the FDA and their approval of the devices for this indication is a better representation of where we stand today. Interdisciplinary factional disputes and politics aside, I believe CMS will want to expand the coverage for carotid stenting in some way.
I will say that although CREST is a “game changer”, it seems the cards may be stacked against stenting. Besides the cost issues that may be associated with expanded coverage, they are influenced by the self interest of the Society of Vascular Surgery that decided they were not going to support the data. If coverage is not expanded for stenting, I wonder if it is not unreasonable for other professional societies to conclude and pronounce that CEA should not be reimbursed in asymptomatic patients.
CMS would be interested in that. I actually think that given the dialogue on the day, it definitely could have gone that way.
It is fascinating to think about this. Given all of the level 1 scientific data supporting revascularization over medical therapy, in the United States today, CEA is the standard of care. Primary care physicians throughout the country recommend this for patients with severe stenoses. CMS is basically talking about erasing that standard of care or now an equivalent procedure from a reimbursement standpoint.
Let us all be clear about the evidence from CREST. In the younger patients, male patients and asymptomatic patients there was no significant difference in outcomes for stroke and death between stenting and CEA.