Is carotid intervention recommended for all patients, or only those who display symptoms? Should revascularisation be done through surgery or using endovascular techniques? If the latter: what types of stent and embolic protection device (if any) should be used, and how should this service be organised? These were the issues handled by experts speaking at the annual meeting of the newly renamed Vascular Society (Bournemouth, UK; 23-25th November 2005). Vascular News dissects the issues and looks at the data that have added to the debate.
There is an interesting comparison to be made between the rates of intervention in the UK and other western countries, said Professor Nick Cheshire, St Mary’s Hospital (London, UK). The domestic tendency is to be conservative in terms of who is offered treatment. Often in the UK only patients displaying symptoms of carotid stenosis – and who therefore have a higher risk of stroke – are treated. In the US, for instance, intervention rates are a lot higher. According to data gathered by Dartmouth College, the mean intervention rate across the US in 2003 was 3.02 carotid endarterectomies per 1,000 Medicare enrolees (Medicare covers some 28 million people, or around 10% of the US population, aged more than 65 years old). “The only hard data were could get for the UK were from the Scottish Intercollegiate Guidelines Network, and they showed a 4- or 5-fold difference in intervention rate compared to the US,” Cheshire told Vascular News. “Even allowing for the fact that the Dartmouth figures are from government data and so may be as much as 10-20% out, this is a huge difference.
These figures do hide some great variations however. Cheshire showed that the surgery rate was as high as 4.5/1,000 Medicare patients in some States, but also went as low as 0.6 in Hawaii. In terms of surgeon workload, the average British vascular surgeon currently performs 14 operations per annum (Scotland 2000). Were UK intervention rates to rise to the US average, each of these surgeons would need to undertake an average of 54 operations per year, and if rates reached those seen in those states with the highest intervention then 78 operations per surgeon would be needed.
The UK figures looked no better when compared to European countries. Quoting both UK Department of Health and EVEM (European Endovascular Monitor) Panel data for Q3 2004-Q2 2005, Cheshire observed that there is a sizeable difference between the numbers of procedures quoted by each. The Department of Health reports around half the number of carotid endarterectomies as the EVEM data, but also records many more carotid artery stenting procedures. “I think there’s a glitch in the Department of Health data that suggest there were 1,000 carotid artery stenting procedures in the UK in 2003-04. That might be the right number for the total amount ever carried out UK, but not in a year,” he said. However, even using the larger number of carotid endarterectomies offered by the EVEM Panel for comparison, “the UK rate [of carotid intervention] is around a third of that of France or Italy (which have similar population sizes to the UK)”. He speculated that this could be “because we’re less aggressive in the types of patients we treat; the prevailing British attitude is that doctors are responsible for costs or perhaps there is a problem getting referrals quickly enough”.
Cheshire also presented evidence of the stroke rate, taken from a Finnish study that compared rates and trends in men and women from 51 industrialised and developing countries in different parts of the world from 1968 to 1994. From this it is clear that while the UK is at the lower end of the scale for incidence of stroke, it still has a rate around 50% greater than that of countries such as Switzerland, Canada, the US, France and Australia. “The evidence of the stroke rate is not good. Although the UK has improved its stroke rate since 1994 there remains more to be done; this year’s Department of Health focus on stroke is evidence of that. However we criticise the data – even if we accept that they have inaccuracies – the 1994 data are strong evidence that the UK system has room for improvement,” Cheshire said.
Regardless of the reason for the differences, “the gist is that probably, from the evidence, we’re not doing enough interventions in the UK compared to the rest of the world, particularly in the carotid artery. This is despite the fact that stroke is one of the most common causes of hospital admission in the UK and in the top five for bed usage”, he said.
Part of the problem, Cheshire proposed, is that the UK’s healthcare service is centrally funded and has a large, unyielding structure. This leads to entrenched working patterns and resistance to change, reflected in the UK’s relatively poor uptake of stent technology. “I’m a believer in stenting,” he said. “There are a lot of T. Rexes and Stegasauruses around who don’t believe the data on stenting. But the technology is improving and it is the way to go,” he concluded.
Patient device selection
Speaking next, Dr Marc Bosiers, AZ St-Blasius (Dendermonde, Belgium), highlighted the need for good patient device selection. He observed that there are many different types of stent and embolic protection device and that while they are equivalent in around 75% of situations, for the remaining 25% careful pre-screening is imperative to help decide which device suits the patient best.
Which embolic protection device?
The three main categories of protection device are: distal occlusion, where a balloon is dilated in the internal carotid artery between the lesion and the brain; distal filter, where an ‘umbrella’ is opened in the internal carotid artery between the lesion and the brain; and proximal occlusion, where balloons are dilated in the common and external carotid arteries to bring about reversed flow or no flow.Bosiers continued, if the patient exhibits no or bad intracerebral circulation, it is mandatory to use a distal filter as it preserves the intracerebral flow during treatment. It is “doubtful” that temporary deflation of balloon devices (to replenish the brain’s oxygen supply) will work, as this can potentially increase the risk of embolisation. In other circumstances where access is difficult – if the iliac arteries are ‘tortuous’ or there is a difficult aortic arch anatomy for instance – then proximal occlusion will not be suitable as the catheters are often not flexible enough to safely pass the tortuous access vasculature. For patients with vulnerable lesions (tight filiform lesions, near occlusions and soft plaque, for example) then it is essential that you “avoid stressing the plaque”, so a proximal occlusion is the best technique to use as it offers protection before lesion passage.
If the artery is so twisted or is too close to cerebrum that there is no ‘distal landing zone’, that automatically rules out anything but the proximal occlusion intervention, he added. Similarly, for treatment of lesions that occur in both the internal carotid artery and intracerebrally, distal filters and occlusion interventions are not recommended.
Bosiers noted that stents come in both open- and closed-cell geometries. The former are only available in nitinol from Cordis (Johnson & Johnson), Guidant, ev3, Bard and Medtronic. In the latter group there is the choice between woven cobalt-chromium alloy type (Wallstent and Boston Scientific) and laser-cut nitinol stents (Xact, Abbott Vascular Intervention/NexStent and Boston Scientific).
Each type of stent has different functional properties, determined by its specific design and composition: nitinol or cobalt-chromium alloy; open- or closed-cell design; many or few struts between the stent rings; and straight, self-tapering, tapered-shouldered or tapered conical forms.
For calcified lesions, the ideal stent is one that has a high radial force to resist recoil and radial crush but which also provides its own moderate outward expansive force without excessive outward radial force. “Nitinol stents are known to have a higher outward force than stainless steel stents and are the preferred stents for use in very calcified carotid lesions,” said Bosiers. The closed-cell is the stent design with the highest radial force. However, in many cases, surgery remains the better option for severely calcified lesions, because there can often be significant residual stenosis following stenting.
Bosiers added, in cases where the patient has ‘vulnerable plaque’ that is likely to break off and result in embolisation, the ideal stent is one with a small free cell area, which will nearly always be a closed-cell stent, he observed, to provide maximum structural support while minimising the amount of plaque that is squeezed through the scaffolding.
If the lesion involves both the internal and common carotid arteries, the stent needs to cover the carotid bifurcation completely. But if the internal and common carotid arteries are of significantly different diameters, it is important to use a stent that can adapt to the anatomy of the vessel walls. A straight nitinol stent of either cell design that has many bridges will not be very adaptable; neither will a conical stent. However, it is the positioning of any stent that is the critical factor. Bosiers concluded that if accurately positioned, a stent with a shouldered tapering is best.
He also examined the situation when the vessels are tortuous and the stent has to be flexible to follow the natural curvature and not get kinked. For these cases, nitinol open-cell stents with few bridges have the required flexibility. Closed-cell braided stents can also be flexible, but must be very carefully positioned. However, he added a warning: “Don’t treat kink- or loop-stenoses endovascularly. Know when to quit: 20-30% of patients are better off with surgery.”
In general stenting is on the rise; the facts speak for themselves. The number of carotid endarterectomies carried out at AZ St-Blasius increased year on year until 1998, when it peaked at 170. Although the total number of cases requiring carotid intervention is still rising, since 1995 a growing proportion have been treated using stents: in 2004 there were 234 stenting procedures and only 54 carotid endarterectomies. In general, around 15-20% of patients are better suited to surgery, while a similar proportion are better served by stenting. The majority of patients can be effectively treated with either intervention, he told Vascular News. “The patient has a choice of either an operation that leaves a scar in the neck, or one with a small puncture wound in the groin, under local anaesthetic, that is minimally invasive. A lot of patients want whole body integrity.” Bosiers predicted that a department that covers a mix of interventions, both carotid endarterectomy and stenting, would handle the future of carotid artery work.
This view was confirmed by Professor Piergiorgio Cao (Perugia, Italy). In organising a stenting service, he recommended a hybrid operating room with a mobile C-arm in the operating room and a permanent angiosuite. The next generation of endovascular surgeons will be trained using virtual reality models, catheters and guidewires, proctored cases or ‘easy’ cases of restenosis with favourable anatomies. It is also vital that they gain experience of different types of cerebral protection device, including clamping and filters, and of different stent designs, he said.
Monitoring is a big part of today’s carotid stenting service. There should be full intra-procedural monitoring with externally audited results and analysis of all adverse events. Talking about his own institution’s results, Cao highlighted the fact that of the 1,392 carotid procedures performed at Perugia between 2001 and 2005, 909 were carotid endarterectomies while 483 used stents, although – like Bosiers’ trends – more and more patients are receiving endovascular treatments.
In the last couple of years, stenting has become the more common procedure and during this time the major stroke/death rate has fallen from 3% to 1%. The crucial times when most major strokes take place are either when the catheter is inserted or the guidewire is crossing the lesion. The three main predictors of whether a patient will succumb to stroke are if they are aged over 80, have diabetes or display ipsilateral symptoms. Cao added that those patients with diseased arteries, an aortic arch, severe proximal carotid lesions or a mobile thrombus are also unsuitable for stenting. Therefore, he concluded, proper material and patient selection can greatly enhance the neurological outcome after stenting. Furthermore, to be properly trained, an interventionalist should see a wider variety of patients than those that are strictly required to earn your credentials. Finally, echoing Bosiers, he said: “Know when to quit!”
Carotid endarterectomy versus stenting evidence
Rounding up the evidence for the two procedures was Trevor Cleveland, Sheffield Vascular Institute (UK). He highlighted 2004’s Cochrane Review, which examined randomised trials comparing endovascular treatment for carotid artery stenosis with surgery. The review covered five trials, Leicester (1998), Wallstent, Kentucky and CAVATAS (all 2001), and SAPPHIRE (2004). At the time of the review, only Kentucky and CAVATAS were completed (608 patients); Leicester and Wallstent had been stopped prematurely (242 patients) while SAPPHIRE only had 30-day data available for 307 patients. The Cochrane Review concluded that there were no significant differences between endovascular and surgical treatment of carotid artery stenosis in terms of either odds of death or stroke at 30 days or prevention of stroke or death. For the risk of cranial neuropathy, it appears that endovascular treatments are significantly better. However, “The results are not absolutely clear cut,” Cleveland told Vascular News. That is because there was “substantial heterogeneity” among the trials. Furthermore, the fact that two of them were terminated early will give a bias against endovascular procedures.
The trouble with assessing a young field such as carotid artery stenting is that “the technology available has exploded”, he observed. “There are now all kinds of stents and protection systems, and it’s difficult to pin them down for study… Over the last 10-15 years, adverse rates [for stenting] have fallen, but is that because the technology has progressed or because we’ve learnt how to do it better and are better able to judge whom we should do it to?”
What is clear from the Cochrane Review is that more randomised controlled trials are needed. There are three currently ongoing: EVA-3S in France, SPACE (Germany) and ICSS (aka ‘CAVATAS II’; mainly UK). “These have been designed so that their endpoints are poolable,” Cleveland explained.
An interesting issue with choosing which type of carotid intervention is appropriate is deciding who will carry out the procedure. Surgeons such as Bosiers have already branched out into endovascular work and can offer either treatment. Cao ended his presentation by saying: “Only a vascular surgeon with endovascular skills can offer both treatment options tailored to the individual patient.” This type of multi-skilling is more common on the continent and in the US than the UK, Cleveland explained. “I’m not sure I agree with it. I believe that patients want specialists who have done hundreds of each type of procedure, rather than someone who has maybe done a lot of different procedures a few times.” However, the flip side is that this lack of interchangeability may mean that specialist surgeons feel threatened by the rise of stenting. “This is why we need to address such issues as a whole vascular community,” he stressed. “Then we can genuinely offer the best options carried out by the best people.” The Sheffield Vascular Institute is rare in that it has a Vascular Unit – of which Cleveland is Clinical Director – that covers all vascular procedures, rather than separating the disciplines out. “We have a team of neurologists, vascular surgeons, radiologists, etc. who work together, meeting weekly to discuss individual case studies and examine patient data. Then we can decide as a team on the best course of treatment.”
The evidence is mounting in the stenting versus surgery debate, but is far from conclusive. “It might turn out that one is better than the other,” Cleveland mused, but on the whole it is likely that with more evidence we will simply be able to widen the group of patients who can potentially get treatment for carotid stenosis. One thing however is obvious: expert opinion is softening towards stenting. Bosiers concluded: “I was amazed [at the reception to my talk]. If I’d given the same talk three years ago, the audience would probably have shot me!”