Introduction

The past 25 years have been witness to a revolution in how vascular care is delivered. The majority of arterial interventions have converted from open surgery to minimally invasive percutaneous endovascular procedures. Herein, we review current endovascular therapy in multiple arterial beds with a focus on carotid artery occlusive disease, aortic pathologies, and lower extremity arterial occlusive disease. Indications for intervention, treatment examples, and outcomes analysis are presented. While not discussed below, endovascular therapy has also become first line in the treatment of coronary artery disease, chronic mesenteric arterial occlusive disease, superficial venous reflux, central vein occlusion, and acute venous thrombus intervention when indicated.

Carotid artery disease

Atherosclerosis of the carotid bifurcation accounts for 15–20 per cent of ischemic strokes via atheroemboli or in situ thrombosis. Carotid endarterectomy (CEA) is the surgical standard in treating carotid stenosis. Specifically, CEA has been found to have profound benefit in patients with hemispheric neurological symptoms attributable to a greater than 50% carotid stenosis [1]. CEA has a lesser, but still statistically significant, benefit for asymptomatic high grade carotid stenosis [2]. Because medical therapy has improved in stroke reduction the past 25 years, most vascular surgeons reserve CEA in asymptomatic patients to those with stenosis greater than 80% and a 5 year life expectancy. Endovascular treatment of carotid bifurcation disease has emerged as a less invasive alternative to CEA in select patients, particularly those patients thought to be high risk or have a hostile operative bed.

The first endovascular therapy to emerge in treating carotid disease was transfemoral stenting (TFS) with cerebral protection filters. This procedure involves passing catheters and sheaths via femoral arterial access through the aortic arch and into the common carotid artery (CCA), followed by crossing the stenosis with a low profile guidewire with an attached filter “basket.” Then the filter is opened to catch potential emboli. Angioplasty and stent are then performed over the wire, and then the filter is retrieved.

TFS is associated with a lower rate of periprocedural myocardial infarction compared to CEA [3, 4]; although the majority were mild non-Q wave troponin elevations [3, 4]. Periprocedural stroke rate with TFS, however, was twice that of CEA in randomized trials providing level 1 data [4,5,6]. The increased stroke risk of TFS relative to CEA is due to cerebral atheroembolization from the process of passing wires and sheaths through the aortic arch into the common carotid artery, as well as crossing the stenosis [5]. This stroke risk has limited the widespread use of TFS.

The transcarotid artery stenting with active carotid flow reversal (TCAR) procedure has emerged as newer technology to avoid atheroemboli. TCAR offers the benefit of being less invasive than CEA but without the increased stroke risk of TFS [7]. TCAR has reduced risk of cranial nerve injury, bleeding complications, neck swelling, and postoperative swallowing difficulties compared to CEA and has patency comparable to CEA at 3 years [7].

TCAR is performed via a small incision to expose the CCA. The CCA is then accessed with a short guide wire and sheath. A sheath is also inserted into one of the femoral veins. A flow reversal system is then activated effectively reversing flow in the CCA with blood passing through a filter to capture atherosclerotic debris and then returned to the femoral vein via a pump system (See Fig. 1). While flow reversal is active, a guidewire is then passed beyond the stenosis and into the distal internal carotid artery. Stent placement and angioplasty are then performed.

Fig. 1
figure 1

Transcarotid Revascularization (TCAR). Angiography confirms the presence of a significant proximal left internal carotid artery stenosis in a patient with recent left hemispheric stroke (a, black arrow). Post treatment angiogram reveals stenosis resolution after self expanding stent deployment using active carotid flow reversal (b, yellow arrow). Image c depicts the TCAR procedure performed via common carotid artery (CCA) cutdown, short sheath insertion in the CCA proximal to the stenosis, flow reversal system to filter any embolic debris to the femoral venous system, and subsequent angioplasty and stent placement. Sources: a, b original clinical practice, c: Silk Road Medical (reprint permission granted)

Full size image

In looking to the future, a significant shift to TCAR away from TFS and to a fairly good extent CEA in upcoming years is likely to occur for the above mentioned reasons. The future of all interventions for asymptomatic carotid disease will largely be based on the ongoing Carotid Revascularization Endarterectomy versus Stenting Trial 2 (CREST 2) trial comparing modern medical therapy to CEA and TFS in combination with best medical therapy as a method of stroke prevention [8].

Endovascular aortic intervention

Endovascular infra-renal abdominal aortic aneurysm (AAA) repair (EVAR) was introduced by Parodi in 1991 [9]. The concept of endovascular aneurysm repair involves the endoluminal deployment of a covered stent that permits blood to flow through the stent while excluding flow into the aneurysm sac. This effectively depressurizes the aneurysm sac and reduces the risk of aneurysm rupture. Ideally all blood outside of the stent graft thromboses within the aneurysm sac. Originally performed via common femoral artery cutdown and primary repair, a shift towards entirely percutaneous EVAR has emerged in the past 5 years [10]. Current infrarenal stent grafts are modular bifurcated grafts which fixate a main body component just below the renal arteries and have individual iliac artery limbs that “pipe fit” into the main body component (see Fig. 2).

Fig. 2
figure 2

Infra-renal endovascular abdominal aortic aneurysm repair (EVAR). Aneurysm diameter of 6.7 cm noted in the infra-renal location (a). Completion angiography after insertion of a modular bifurcated stent graft with flow preservation to both renal arteries noted (yellow arrows b). Follow up CT angiogram (c) with no sign of endoleak

Full size image

A potential complication of endograft placement is development of an endoleak. An endoleak is defined when there is persistent blood flow in an aneurysm sac after deployment of an endograft. Endoleaks result in the need for secondary interventions in approximately 20% of patients after EVAR [11]. Endoleaks create the potential for aneurysms to continue to grow and rupture. For this reason, serial imaging of aortic aneurysms with either CT angiogram or duplex ultrasound is performed.

Endoleaks are classified by the source of the blood passing into the aneurysm sac and are summarized in Table 1 [12]. Types 1, 3, and 4 endoleaks require treatment as they involve systolic arterial pressurization of the aneurysm sac and therefore have a risk of aneurysm rupture. Type 2 endoleaks, the most common, are low pressure leaks at essentially venous pressure as blood is passing though capillary beds and small collaterals back into the aneurysm sac, and are safe to observe if the aneurysm does not grow in size and remains asymptomatic [13].

Table 1 Classification of types of endoleak following endovascular AAA repair
Full size table

Type 2 endoleaks associated with aneurysm sac growth on follow up imaging can be managed by a variety of techniques including: (1) embolization of feeding arteries via internal iliac or superior mesenteric artery catheterization [14]; (2) direct aneurysm sac puncture via translumbar techniques followed by sac embolization with coils or polymer [15]; (3) transcaval embolization by obtaining percutaneous central venous access to the inferior vena cava followed by endoluminal needle puncture into the aortic aneurysm sac and subsequent aneurysm sac embolization [16]; and (4) laparoscopic ligation of aortic side branches [12]. Rarely, type 2 endoleaks associated with aneurysm sac growth refractory to minimally invasive methods are treated with conversion to open AAA repair.

Type 1 endoleaks are managed by proximal or distal additional stent graft component insertion when there is a landing zone present. If there is no healthy segment of native aorta (proximally) or iliac artery (distally) to land an additional stent, then strategies include insertion of larger stents at the fixation sites, extension with fenestrated stent grafts, or open surgical conversion. Type 3 and 4 endoleaks are most easily managed by stent graft re-lining with additional endoluminal covered stent components [12].

Extensive randomized trials have been conducted comparing open AAA repair with EVAR. EVAR has the advantage of a lower 30 day mortality versus open surgical repair (1% vs 3%) [17,18,19,20]. This advantage, however, is lost at long term follow up with a slight advantage to open repair in overall survival at 5 and 10 years [18,19,20]. This loss of survival advantage is attributed to the often non-definitive nature of EVAR secondary to endoleaks and additional radiation exposure for EVAR patients. Regardless, the low 30 day mortality risk of EVAR has resulted in a shift towards EVAR over open repair. Currently the split is approximately 85% EVAR to 15% open repair in the United States [21].

Once EVAR became popularized, endovascular repair of thoracic aortic aneurysms (TEVAR) shortly followed [22]. TEVAR involves single tube grafts designed to fixate above and below an aneurysm in healthy segments of aorta to exclude flow into the sac. Covered stents also are used in the treatment of traumatic aortic transection and for treating aortic dissection with associated malperfusion or aneurysmal degeneration [22,23,24,25]. The goal in endovascular treatment of aortic dissection is to cover the most proximal aortic fenestration and apply radial force from within the true lumen of the aorta to compress and induce thrombosis of the false lumen. Dissection stents with a covered proximal stent graft and bare metal distal components have been designed to cover the proximal fenestration of the aorta while compressing the false lumen distally without compromising flow to intercostal and visceral branches [25].

TEVAR is routinely combined with open surgical cervical debranching procedures such as carotid-subclavian bypass or carotid-carotid bypass to achieve more proximal landing zones in the aortic arch. These bypasses are done in conjunction with branch origin ligation or embolization to prevent aortic arch branch type 2 endoleak [26]. Similarly, visceral artery open debranching can be performed with inflow arising off the iliac arterial system and bypassing to the individual renal and mesenteric arteries, branch origin ligation, and endovascular exclusion with standard TEVAR and EVAR stent graft components [27].

In the past decade, fenestrated endograft technology has emerged allowing expansion of endovascular repair of aortic pathology to include anatomical segments once not feasible. There are now endografts with fenestrations suiting branches of the aortic arch, visceral segment and internal iliac arteries [27,28,29,30,31]. Fenestrated grafts have “side holes” through which additional covered stents are passed into large aortic side branches such as the subclavian artery, celiac, superior mesenteric, and renal arteries. The main bodies of these grafts are deployed and then the target side branch arteries are then cannulated via smaller covered stents that are passed through the main body of the device and into the side branch. One end fixates within the branch vessel and the other end fixating within the main body aortic graft [28,29,30]. Fenestrated grafts exist in formats which are “off the shelf” designed to potentially treat the majority of anatomical variants for the visceral or aortic arch vessels. Custom made grafts are also manufactured in a factory setting and optimized to a given patient’s anatomy [28, 29].

Physician modification of an endograft component to create custom fenestrations is also widely reported (offlabel use of the device) [30, 32]. In this technique, a TEVAR stent is partially deployed on a sterile operating room back table. Fenestrations are then cut in the graft with specific locations and diameters to accommodate aortic side braches. Radio-opaque markers are sutured to the fenestrations to allow for identification under fluoroscopy. The graft is then “re-sheathed” and ultimately completely deployed in the aorta. This is followed by side branch cannulation with appropriately sized covered stents. This is well described for both visceral segments and the aortic arch [30, 32]. Figure 3 depicts treatment of a visceral segment aortic aneurysm treated with a physician modified endograft.

Fig. 3
figure 3

71 YO female with 6.5 cm visceral segment and infra-renal AAA (Images a, b). Chronic celiac artery occlusion was present as well as an atrophic right renal artery. She was treated with a physician modified endograft with custom made fenestrations for the superior mesenteric artery (Image c white arrow) and left renal artery (Image c yellow arrow). Image d is a CT reconstruction one year postoperatively showing a patent SMA (d, white arrow) and left renal artery (d yellow arrow). One year imaging revealed no endoleak and creatinine was noted to be normal

Full size image

The final method of treating complex aortic pathology by endovascular means involves parallel stents serving as effective periscopes or “snorkels.” In this technique, aortic side braches are cannulated, and covered stent grafts are deployed in them simultaneously to standard EVAR and TEVAR graft components. The grafts then lie in parallel. While this concept may seem to create a “square peg into round hole,” it has been found to effectively extend fixation sites for endografting with off the shelf, readily available implantables [33, 34]. This procedure has been popularized as access to factory made fenestrated devices has been limited by industry in the United States. Further, in acute aortic syndromes, there often does not exist the time necessary to obtain a fenestrated graft or the time to perform a physician modified endograft. Outcomes with parallel grafting have essentially matched that of fenestrated stent grafts [34].

Chronic lower extremity peripheral artery disease

Endovascular therapy now encompasses over 75% of treatment for peripheral arterial disease (PAD) with open surgery typically reserved for long segment occlusions and femoral bifurcation disease [35,36,37,38,39,40]. Endovascular intervention for PAD has the advantage of being a minimally invasive, percutaneous procedure, which can be done on an outpatient basis. Open surgery has the advantage of increased long term patency of interventions, particularly for long segment occlusive disease. Open surgical intervention in comparison has a significantly increased 30 day morbidity [34,35,36,37,38]. Direct comparisons of endovascular versus open surgical revascularization for limb ischemia have revealed equivalent outcomes with respect to limb salvage and long term mortality [35, 37, 38]. The one caveat in randomized controlled trials has been that patients undergoing bypass after failed endovascular therapy have worse limb salvage outcomes than patients treated directly with surgical bypass at outset [38]. Endovascular techniques and open surgery are often combined to achieve in-line flow for a chronically ischemic foot. Most commonly, iliac or superficial femoral artery (SFA) angioplasty and stent are combined with common femoral artery endarterectomy or infrainguinal bypass [41, 42].

The primary endovascular techniques for treating chronic PAD are angioplasty, stent placement, and atherectomy. Angioplasty and stent placement are quite simple conceptually. Arterial access is obtained with Seldinger technique followed by insertion of an intra-arterial sheath which serves as a portal for guidewires, catheters, and interventional devices. Diagnostic angiography is performed followed by crossing stenoses or occlusions with guidewires and catheters. Once a lesion is crossed, confirmation of entry into the true lumen of the distal arterial tree is done with angiography. Angioplasty balloons and stents are then passed over the wire and deployed at the area of stenosis or occlusion, followed by completion angiography. Endovascular therapies are feasible in essentially all arterial beds, including pedal arteries below the malleoli [43].

Atherectomy serves as a method of mechanical atherosclerotic plaque removal either circumferentially (orbital catheters) or on one side of the catheter at a time (directional catheters). Atherectomy effectively debulks plaque, but supplemental angioplasty is usually required [44]. Atherectomy currently has a niche role in PAD, is high cost, and the benefit has been questioned in multiple robust analyses [39, 45].

In high risk candidates for open surgery, adjunctive endovascular devices are employed to treat long segment occlusions in a minimally invasive manner. Re-entry catheters are equipped with needles that facilitate puncturing back into the true lumen of an artery from the subintimal plane into which guidewires often pass while trying to cross long occlusions. Catheters designed to burrow into an atheromatous plaque and gain entry into an occluded segment also exist [46].

Outcomes of endovascular therapy guide the approach taken in different anatomic vessels. In treating iliac artery stenosis or occlusion, primary stent placement has been shown to have improved patency versus plain balloon angioplasty (POBA) [47, 48]. Further, polytetrafluoroethylene (PTFE) covered stent placement has been shown to have modest additional benefit to bare metal stenting in the iliac distribution [49]. However, covered stents have 3–4 times the cost of bare metal stents and require larger arterial access sheaths, making their use selective.

Analyzing outcomes of endovascular interventions in the femoral-popliteal and tibial segments include evaluating plain balloon angioplasty (POBA), drug eluting angioplasty, bare metal stent placement, covered stent placement, drug eluting bare metal stents, and atherectomy [50]. As a general guiding principle, the smaller the caliber of a given artery, the lower the long term patency rate. Iliac interventions have 5 year primary patency around 80% [51]. SFA interventions have varying patency with short segment stenosis treatment as high as 80% primary patency at 3 years versus just 50% 3 year primary patency for long segment occlusions [49]. Tibial arteries are of the smallest caliber and thus have the lowest patency with and an average rate of 50% at one year [49].

In treating long segment SFA occlusions, primary stent placement has been shown to have superior patency to POBA [50]. Similarly, a randomized prospective trial showed improved patency of SFA occlusions 3 cm or greater with PTFE covered stent placement relative to POBA [51]. Stenting of long segment SFA occlusions is fairly equivalent to PTFE conduit bypass but inferior to autologous saphenous vein bypass [52].

Paclitaxel eluting angioplasty balloons as well as paclitaxel eluting bare metal stents have been shown to improve patency in the femoral-popliteal segment relative to POBA in randomized prospective trials [47, 53]. These trials, however, do not compare drug eluting technology with bare metal stenting in the femoral-popliteal segment. In addition, while improved primary patency is shown, there is no advantage in limb salvage despite dramatically increased cost of drug eluting devices relative to POBA and bare metal stents [47, 53]. Most concerning is a recent meta-analysis which pooled randomized trials comparing drug eluting technology versus POBA and revealed a 1.8 times increased mortality at 5 years for patients receiving Paclitaxel [54]. Given this potential toxicity in combination with Paclitaxel providing no benefit regarding limb salvage, most vascular surgeons have moved away from using Paclitaxel eluting technologies.

In the tibial distribution, both drug eluting technologies and atherectomy have shown no benefit to POBA regarding patency or limb salvage [55,56,57,58]. Tibial angioplasty has nearly universally been found to have a one year primary patency around 50%, but offers limb salvage over 80% at one year in critical limb ischemia paitents [59,60,61,62,63]. Therefore, POBA is the endovascular treatment of choice for tibial artery occlusive disease (See Fig. 4) [39].

Fig. 4
figure 4

Endovascular treatment of multi-level atherosclerotic occlusive disease in a patient with digital gangrene. High grade mid superficial femoral artery stenosis (a, white arrow) treated with angioplasty with good resolution (b, white arrow). In addition complete occlusion of the mid right anterior tibial artery (c, yellow arrow) and tibioperoneal trunk (c, white arrow). Successful treatment of the tibioperoenal trunk (d, white arrow) and peroneal artery (d, yellow arrow) with 2.5 mm angioplasty to achieve in-line flow to the foot

Full size image

Regardless of the endovascular intervention performed, long term surveillance with serial arterial duplex and Doppler examinations with ankle brachial index and digital pressures is recommended [43, 64].

Acute limb ischemia

Acute limb ischemia (ALI) is a surgical emergency secondary to sudden thrombosis of an arterial tree feeding a limb. Both open surgical and endovascular revascularization options exist in the setting of ALI. Once ALI is diagnosed, emergent systemic anticoagulation is indicated.

Endovascular therapy for ALI is based on a combination of thrombolytic medication (Tissue Plasminogen Activator (TPA)) and percutaneous mechanical thrombectomy (PMT) devices [65]. PMT achieves an immediate flow channel through an occluded artery assuming a guidewire can be crossed through the thrombus. PMT is utilized via either pulse spray-suction, rotational force or thrombus aspiration technology [65]. PMT can be used in combination with thrombolytic therapy. If unable to cross the thrombus with a guidewire, thrombolysis without PMT can be attempted, but with less success than with a lytic catheter directly invested into the thrombus. Finally, after thrombolysis is completed, angioplasty and stenting of any residual stenosis from chronic atherosclerosis can be performed (See Fig. 5).

Fig. 5
figure 5

Acute arterial embolus to the right distal common iliac artery seen on image a. After 5 mg TPA bolus and percutaneous mechanical thrombectomy with angiojet device, successful resolution of the occlusion is seen in image b. The underlying external iliac stenosis would be subsequently treated with angioplasty

Full size image

Conclusion

Endovascular therapies are used in all vascular beds to treat the full spectrum of vascular pathologies. Aneurysm disease, atherosclerotic arterial occlusive disease, acute arterial and venous thrombosis, ongoing hemorrhage, and venous reflux are among the issues which can be addressed by endovascular means. The minimally invasive nature of endovascular treatments in what is largely a very co-morbid patient cohort is an attractive method of avoiding major procedural related morbidity and mortality.