Is Endovascular Treatment Going to Put Open Surgery Out of Business?

History

Thomas Fogarty¹ developed the first balloon catheter for the semi-open procedure of peripheral arterial embolectomy, and Charles Dotter² was the first to use a balloon artery catheter inserted over a guidewire to dilate an atherosclerotic stenosis. Technological developments in catheter balloon manufacture provided Andreas Gruentzig³ with low-profile co-axial balloons that were strong enough to distend tough atherosclerotic plaques.

Mobin-Udin⁴ and Greenfield⁵ have also used the hybrid concept of open surgical access combined with the endovascular passage of a catheter delivery system. This enabled them to successfully deploy filters in the inferior vena cava, preventing pulmonary embolization from remote access sites.

Balloon Angioplasty and Stenting for Atherosclerotic Stenoses

Longitudinal plaque fissuring and over-distension of the arterial wall produced by angioplasty often result in rapid recoil and severe restenosis as a consequence of myointimal hyperplasia. The development of subintimal angioplasty allowed longer occlusions to be treated in the distal vessels of the lower limbs. However, not all centers have been able to achieve good results with this technique.⁶

The recognition of the many sub-optimal results achieved by simple balloon dilation led to the concept that restenosis can be reduced by inserting an internal support (stent) inside the dilated segment to hold the vessel open and smooth out the luminal surface. It was hoped that stents would also be used to cover any tears or dissections. A series of differently-configured expandable stents were developed and made from a variety of materials.⁷ These stents are placed through a catheter-based delivery system into the segment treated by angioplasty, where they self-expand (nitinol) or are distended by a balloon (e.g., stainless steel). The assumption that intravascular stents would dramatically improve the results of angioplasty was soon dispelled when in-stent restenosis became recognized as a significant problem.⁸ Some trials demonstrated that at least in the aorto-iliac segment, the routine placement of a stent had a marginal effect on vessel patency.⁹

Drug-eluting stents were developed to overcome the problem of in-stent restenosis by reducing myointimal hyperplasia. These biological stents have mainly been investigated in the coronary arteries, and several agents have been utilized, including sirolimus, paclitaxel and dexamethasone.^10,11 Sirolimus, used primarily in the prevention of renal transplant rejection, has also been studied in femoral occlusive disease.¹² The sirolimus-eluting or SMART stent (nitinol-expandable, sirolimus-coated; Cordis Endovascular, Warren, New Jersey) has been compared with bare metal stents in two trials: Sirolimus-Eluting versus Bare Nitinol Stent trial for Obstructive Superficial Femoral Artery Disease (SIROCCO I and II). These trials showed no long-term benefit in favor of sirolimus-eluting stents.^13,14

It has became apparent that balloon dilation with or without stenting is good at treating short stenoses or occlusions in large vessels (e.g., the iliac, superficial femoral and popliteal), but is far less effective at treating longer occlusions in small vessels.¹⁵ The Trans-Atlantic Inter-Society Consensus (TASC) has risk-stratified iliac vessel disease into 4 categories (A to D), A being short focal lesions, and D, complex longer lesions. The consensus conference recommended endovascular treatment for type A lesions and open surgery for type D. Recommendations for type B and C lesions were much less dogmatic. The overall outcome was better after surgery than endovascular treatment, provided there is good distal run-off. Surgical bypass of the coronary arteries is also better for multiple and/or extensive stenosis when compared with stenting, which seems to produce better results in focal disease.^16–18

The Role of Endovascular Approaches to Controlling Bleeding, Infracting Tumors and Closing Fistulae

Endovascular techniques have been used to allow coil and particulate embolization of arterio-venous fistulae, bleeding points from small vessels and as a means of infarcting large tumors by occluding their blood supply. Embolic occlusion of arterio-venous malformations is merely palliative and has to be endlessly repeated, unless combined with ablative destruction.^19–21 Endovascular stents have been placed over sites of arterial trauma to control hemorrhage and have also been used to close traumatic arterio-venous fistulae.²² This can be an extremely effective form of treatment in these circumstances, especially when surgical access is difficult and many high-pressure, thin-walled veins exist, making direct surgical approach unattractive.

Endovascular Repair of Aneurysms

The availability of stents encouraged Parodi²³ to develop the concept of an endovascular graft to exclude aneurismal sacs. Early homemade devices, which required many steps with complex wires and pulleys to achieve deployment, have given way to simpler manufactured devices that are much easier to deploy. The availability of these stent grafts, which can be delivered into the thoracic and abdominal aorta via a cut-down incision or transcutaneous puncture in the groin, has led to an explosion in their use to treat aneurysmal disease of the aorta and iliac arteries.

The multicenter trial Endovascular Aneurysm Repair (EVAR I),²⁴ which has compared stent grafting and open repair of elective abdominal aortic aneurysms, has shown a considerable early advantage in reducing 30-day mortality when stents are deployed (2% versus 4.5%). This early advantage, identified in the EVAR I and Dutch Randomized Endovascular Aneurysm Management (DREAM)2^5-27 trials, appears to be lost by a year.

Continuing problems of endoleaks into the excluded aneurismal sac, migration, kinking and disruption of the modular devices remain to be overcome, and lifetime surveillance must be maintained. During the first 4 years of follow-up, 41% of the patients stented required an additional intervention.²⁸ EVAR II,²⁹ which compared stenting to no intervention in those unfit for open aneurysm repair, showed that the mean hospital cost per patient in the intervention group over four years was £13,632 (approx. $24,720 USD) compared with £4,983 (approx. $9,219 USD) in the non-intervention group. There was no difference in the overall mortality of the two groups, with a high proportion of death within 1 year of inclusion in the study.

Many patients with complex aneurysm morphology cannot be treated by stent grafting. The open operation, which is known to be extremely durable, remains the treatment choice for relatively young and fit patients without serious co-morbidity. The role of stent grafting for the treatment of leaking aneurysms remains to be established. Initial controlled trial data comparing stenting and open operation for leaking aneurysms have shown no difference in mortality.³⁰

The development of early customized and fenestrated branch grafts for patients with suprarenal or thoracoabdominal aneurysms is still in its infancy. At present, the prohibitive cost of these devices must be tested in appropriately organized, randomized studies. The newer concept of combining extra anatomical bypasses with stent graft occlusion for thoracoabdominal aneurysms (hybrid procedure) is also in its infancy and requires formal prospective assessment.³¹

It is in the thoracic aorta that stent grafting is having its greatest impact, where a staggering improvement in mortality (25% versus 5–10%) compared to open surgery, combined with a reduction in the risk of paraplegia (> 5%) makes stent grafting a very attractive option. The problems of endoleak, stroke and graft migration remain, but the reduction in both mortality and paraplegia has made this the treatment of choice for localized thoracic aneurysms.^32,33 This is also true in the treatment of stable aortic transactions,³⁴ and stent grafting is rapidly becoming the treatment of choice for complicated or unstable patients with type B dissecting aneurysms.³⁵ The value of deploying stent grafts in all type B dissecting aneurysms still needs assessment, and trials are now underway to evaluate it.

Carotid Stenting

There is irrefutable evidence that surgery is beneficial for carotid disease in symptomatic patients with a stenosis of greater than 70%.³⁶ Not unsurprisingly, this was extended to the concept that carotid angioplasty and stenting would be equally beneficial while less invasive. The original trials that set out to test this hypothesis were seriously flawed and only reached equivalence because of appalling surgical results.³⁷ The Stenting and Angioplasty With Protection in Patients at High-Risk for Endarterectomy (SAPPHIRE)³⁸ trial suggested that stenting with cerebral protection in high-risk patients was as good as open surgery. The randomization of the trial was, however, limited, and it was overshadowed by a conflict of interest in one of the principal investigators.

The initial experience of carotid stenting and cerebral protection in our unit has produced some disappointing results (1 death, 1 stroke, 2 severe restenoses in the first 12 patients). Further trials are now underway to compare surgery with stenting and cerebral protection. International Carotid Stenting Study (ICSS) is a multi-center, randomized control trial which aims to overcome some of the problems of CAVATAS.³⁹ There is also a French trial, Endarterectomy versus Angioplasty in Patients with Symptomatic Severe Carotid Stenosis (EVA 3S)⁴⁰, as well as the Stent-Supported Percutaneous Angioplasty of the Carotid Artery versus Endarterectomy (SPACE) trial.^41,42 The Carotid Revascularization Endarterectomy versus Stenting (CREST) study, an American trial, has also recruited patients. The results of these trials will not be known for three to five years. Until these data are available, carotid stenting should only be selectively applied to special patients. All other suitable patients should be entered into one of the trials. Reports at the Society of Vascular Surgeons meeting in Chicago 2005 indicate that the risk of stroke and death following stenting remains double that of surgery.⁴³

Varicose Veins and Deep Vein Thrombosis

The endovascular explosion has also reached the venous system, where intraluminal devices that “cook” or “burn” the long saphenous veins can be induced via a catheter placed in the vein under ultrasound guidance. These devices have been introduced with little evidence of their efficacy and trials are now underway to try to assess these treatments against standard surgery and foam sclerotherapy.^44–47

Other devices aimed at dissolving or disrupting thrombus are still in development. Venous stents are useful to palliate malignant infiltration of large veins and in treating iliac vein compression syndrome.^48–52 Endoscopic varicose vein surgery appears to be a safe alternative to open surgery and may have a future role, but as yet there are no multi-center, randomized control trials with a reasonable period of follow-up looking at recurrence.^53,54 Subfascial endoscopic perforator vein surgery (SEPS) began in the 1980’s,⁵⁵ but it became more prominent in the management of chronic venous insufficiency in the 1990’s. The North American Subfascial Endoscopic Surgery (NASEPS⁾⁵⁶ registry was set up to evaluate the safety, feasibility and efficacy of SEPS for chronic venous insufficiency (CVI). It was a retrospective analysis looking at 151 patients in 17 centers in both the United States and Canada. The technique was found to be safe, with good rates of ulcer healing (88%). The mid-term results, however, showed that recurrence is still a significant problem.⁵⁷ Although some advocate SEPS as the treatment of choice for CVI, there is at present no level I evidence.⁵⁸

Valve repair for chronic venous insufficiency was developed by Kistner⁵⁹ and modified by Raju.6⁰ Subsequently, angioscopically-guided suture placement was reported by Gloviczki⁶¹ as a semi-open procedure. The use of these techniques remains to be properly evaluated.

Summary

At present, endovascular treatments are of value for:

1. Dilating localized arterial stenoses or short occlusions in peripheral arteries;

2. Placing caval filters;⁶²

3. Treating thoracic aneurysms and unstable aortic type B dissections and thoracic aortic transactions;

4. Embolizing arteriovenous fistulae;

5. Treating abdominal aortic aneurysms in patients who are otherwise unfit for an open repair, with the caveat that EVAR II has not shown a reduction in mortality in unfit patients treated by endovascular stent grafts.

The role of endovascular treatment remains to be established for long peripheral arterial occlusions, aortic aneurysms in fit patients, thoracoabdominal aneurysms, and in all patients who have a carotid cause for cerebrovascular symptoms. It is also important to consider the group of patients who develop restenosis or other complications related to endovascular treatment. There are a growing number of these patients who require challenging open surgery or further additional endovascular solutions to salvage the situation.

Vascular surgeons must properly inform patients of the risks and benefits of both open and endovascular treatments, in order that patients may come to a shared view on the best method of treatment for a particular condition. This should not be based on pecuniary reward or technological bias, but on clear, hard facts. Presently there is no way that stenting is going to put open vascular surgery out of business.

COMMENTARY

by Frank J. Criado

The article “Is Endovascular Treatment Going To Put Open Surgery Out of Business?” by Wadoodi, et al. is interesting and provocative. While the theme (and title) would seem to surround the question of whether endovascular technologies are going to put open standard surgery out of business, it is clear from the outset that the answer is going to be — predictably — on the negative. There will “always” be a need for standard surgical procedures — at least in the foreseeable future. But their role will be different and diminished.

I find myself in agreement with the general tone of the paper. However, I am left with the impression that the author failed to capture (or express) the enormous impact of the endovascular revolution. Profound and transformational are appropriate terms to describe it. The absence of level 1 evidence in many areas, notwithstanding innovative endovascular technologies, is here to stay, and nothing will ever be the same.