Endoscopic Vein Harvest for Infrainguinal Vascular Reconstruction and Limb Salvage in Chronic Critical Limb Ischemia

Introduction

Autogenous greater saphenous vein is the preferred conduit for infrainguinal vascular reconstruction, and the most commonly utilized venous conduit for coronary artery bypass grafting. Harvesting for these procedures has traditionally utilized a longitudinal continuous saphenectomy incision. Wound problems can be significant with this approach and bridging techniques evolved to lessen this complication.^1–5 Endoscopic vein harvesting (EVH) was introduced in the 1990s and has now become the preferred method for greater saphenous vein harvesting in coronary artery bypass grafting.⁶ EVH for infrainguinal vascular reconstruction has not been as frequently reported. Nevertheless, there is compelling information that EVH for infrainguinal vascular reconstruction can be accomplished with a reduction in wound morbidity and yield satisfactory conduit performance.^7–12

Our experience with EVH for coronary artery bypass grafting has been favorable, paralleling that of others.^13,14 We expanded our application of this technique to infrainguinal vascular reconstruction after reviewing an initial limited experience with EVH, primarily in femoral to above-knee popliteal bypass grafting for activity limiting claudication. With this favorable experience, we incorporated EVH into infrainguinal vascular reconstruction in chronic critical limb ischemia (CLI). Intuitively, a minimally invasive approach in a patient population already in a compromised position for wound complications may preclude an additional risk. This report reviews this experience and details our observations with EVH in this challenging patient population.

Materials and Methods

All patients who underwent infrainguinal vascular reconstruction for activity-limiting claudication or limb salvage in chronic CLI between January 2004 and March 2006 were reviewed retrospectively from office and hospital records. All patients were evaluated by experienced interventionalists and a single surgeon, and all imaging studies were overread by a single interventionalist. No patients were referred for primary amputation until their evaluation was complete. Treatment plans were developed and often staged, requiring both percutaneous peripheral intervention (standard and cutting balloon angioplasty, stenting, laser or directional atherectomy, and cryoplasty) and open surgery (endarterectomy, bypass grafting).

Pre-operative evaluation included cardiac clearance with percutaneous coronary intervention (PCI) utilized when necessary for treatment of significant coronary artery disease (CAD) as opposed to coronary artery bypass grafting. Pulmonary function tests and serum creatinine were used to screen for pulmonary and renal co-morbidities, with pre-operative pulmonary rehabilitation and renoprotective measures instituted as indicated. Carotid duplex scanning identified hemodynamically-significant extracranial cerebrovascular disease with carotid endarterectomy performed when necessary. Carotid stenting was not utilized. All patients received pre-operative vein mapping of the greater and lesser saphenous vein systems. Operative approach was based on pre-operative imaging. This consisted of 64-detector computed tomography (CT) angiography with a specialized protocol to identify pedal outflow or conventional angiography, to include foot angiograms.

Conduct of the operation began with exposure of the distal target using standard techniques. EVH was then undertaken based on availability and location of the greater saphenous vein. Skin marking was not done. Commercially available EVH systems were used. Surgical physician assistants harvested all veins using standard techniques previously described.1^1,13 The initial EVH incision is longitudinal as opposed to transverse. Once the vein was harvested, it was assessed, irrigated, and gently pressurized with a heparinized blood-papaverine solution. Judicious attention to detail in preparing the vein (ligating branches, splicing segments) was important to optimize quality and quantity of conduit. The vein was used in a reversed fashion and valves were left intact. The vein was marked to facilitate proper orientation in the tunnel. The inflow target vessel was then exposed and tunnels were created as needed between the incisions, utilizing the bed of the harvested vein. The anatomic lie of the graft is decided by the distal target. For a distal anastomosis in the proximal one-third of the leg, the graft was routed behind the knee between the heads of gastrocnemius muscle. For a distal anastomosis of the mid and distal one-third of the leg, the posterior tibial artery or medial plantar artery, the graft was routed along the medial thigh and leg. For a distal anastomosis to the dorsalis pedal artery, the graft was coursed anteriorly over the distal leg and foot. Systemic heparinization with 7500 units of heparin was instituted and the patient was re-dosed after one hour based on activated clotting time (ACT). The distal anastomosis was then performed with continuous 7-0 polypropylene suture for tibial and pedal targets, and 6-0 polypropylene suture for popliteal targets. The suture line was left untied and the proximal anastomosis performed. Appropriate flushing was then accomplished and flow restored sequentially through the graft with distal flushing. Completion assessment was by arteriography initially and then Doppler interrogation, aided by the endoscopic camera system to visualize the entire tunneled graft for correct orientation and bleeding. Heparin was then fully reversed, and the tunnel gently “rolled” to evacuate accumulated blood. The wounds were closed in layers, except for incisions in the distal one-third of the leg and foot, which were closed with a single layer of interrupted nylon. Drains were not utilized unless the tunnel was remote from the graft, to avoid direct contact of the graft with a suction drain. Perioperative prophylactic antibiotics were a second-generation cephalosporin or vancomycin if there was a history of penicillin allergy, instituted immediately preop and continued for 48 hours.

Tissue debridement and/or limited minor amputations with wound closure (primarily, skin grafting and/or skin substitutes) are an important component of this treatment plan. Active infection was surgically addressed with debridement and drainage at the time of revascularization. Appropriate antibiotic therapy was instituted and continued post-operatively. In staged procedures with a percutaneous peripheral intervention component, active infection was surgically addressed with debridement and drainage approximately 48–72 hours before the percutaneous intervention. The open surgical component was undertaken after 2–3 weeks of aggressive wound care.

Low-molecular-weight Dextran was infused for 24 hours at 20 cc per hour. Clopidogrel, low-dose enteric-coated aspirin and statins were started the following day and continued indefinitely. Wound care was begun and maintained post-discharge until all wounds were healed. Physical therapy and rehabilitation therapy were initiated and continued post-discharge to facilitate mobility. Vascular surveillance was maintained using published Medicare guidelines for clinical follow-up and ultrasound-based testing.¹⁵ Abnormal findings suggestive of vein graft stenosis were investigated with CT angiography and intervened upon as indicated. Outpatient follow-up assessment was based on recommended reporting standards for lower extremity ischemia.¹⁶

Results

From January 2004 through March 2006, 65 patients underwent infrainguinal vascular reconstruction for severe activity-limiting claudication or chronic CLI and limb salvage. Of these patients, 26 utilized EVH for procurement of greater saphenous vein. There were 17 (65%) males, 9 (35%) females, 6 (23%) African-Americans, and 20 (77%) Caucasians. The average age was 69 years.

Indications for surgery were activity-limiting claudication in 3, rest pain in 5, ischemic necrosis in 10, and rest pain plus ischemic necrosis in 8. Associated co-morbidities were diabetes mellitus in 18, smoking in 18 (active in 8, remote in 10), hypercholesterolemia in 15, hypertension in 25, CAD in 14, end-stage renal disease on hemodialysis in 3, and chronic atrial fibrillation in 4.

Six of the procedures were re-operations in affected limbs that had previously undergone infrainguinal vascular reconstructive procedures; 16 of the procedures were performed in limbs that had previously undergone percutaneous peripheral intervention. Thus, 22 (85%) of the affected limbs had previously been intervened upon. All bypass procedures were elective and all were reconstructive. These distal revascularizations were of the following types: femoral–AK popliteal in 4; femoral–BK popliteal in 7; femoral–tibioperoneal trunk in 1; femoral–posterior tibial in 4; femoral–peroneal in 2; femoral–anterior tibial in 1; distal superficial femoral artery–anterior tibial in 1; distal SFA–dorsalis pedis in 1; AK popliteal–anterior tibial in 1; BK popliteal–dorsalis pedis in 2; BK popliteal–posterior tibial in 1; BK popliteal–medial plantar artery in 1. Of the proximal anastomoses, 22 (85%) were above the knee and 22 (85%) of the distal anastomoses were below the knee.

EVH was accomplished in the ipsilateral leg in 21 patients. The contralateral leg was utilized in three patients and both legs in two patients. There were no conversions to open vein harvesting.

The reversed greater saphenous vein configuration was used in 25 patients and translocated greater saphenous vein in one case. Two vein grafts were constructed by splicing two segments of vein together for a composite graft. One adjunctive procedure was performed, a femoral endarterectomy with pericardial patch angioplasty, which was then incorporated into the proximal anastomosis.

Follow-up ranged from 30 days to 24 months, with an average of 8 months. There were no perioperative procedure-related deaths (< 30 days) and no procedure-related deaths in follow-up. Four late deaths (> 30 days) occurred during the course of follow-up (lung cancer – 1, heart failure – 2, bowel obstruction with multi-system organ failure – 1). Peri-operative procedure-related complications occurred in four patients. One restorative procedure was required on post-operative day 7, when the graft thrombosed and was thrombectomized and revised. This was the single translocated vein graft. This graft remained patent out to six months, with ulcer healing prior to the patient’s death of unrelated causes. One patient developed skin necrosis and ulceration over the EVH tunnel that required debridement, but healed with conservative wound care. One patient developed a methicillin-resistant staphylococcus aureus (MRSA) infection of his surgical wound and the underlying vein graft. Despite aggressive therapy, this ultimately required graft excision at 60 days for resolution. The graft was patent at the time of excision. One patient developed a superficial wound infection that was drained and resolved with local care; the graft remained patent. There were no cases of graft exposure due to overlying tissue necrosis.

Vascular surveillance identified four patients that required re-intervention to maintain patency. One patient experienced vein graft stenosis at three months, for which he underwent percutaneous transluminal balloon angioplasty (PTA) and stenting. This initially maintained assisted primary patency but required multiple re-interventions until graft closure at 10 months. This patient subsequently underwent above-knee amputation for a non-healing transmetatarsal amputation site. One patient experienced vein graft stenosis at six months, for which he underwent PTA and atherectomy, and maintained assisted primary patency. Another patient experienced vein graft stenosis at seven months, for which she underwent PTA and stenting, and maintained assisted primary patency. A fourth vein graft stenosis was identified at seven months, but the patient did not return for a percutaneous peripheral intervention that was advised. There were five major (above-knee or below-knee) amputations in this series. One occurred on post-operative day one for graft occlusion, irreversible ischemia, and no salvage options. A second amputation occurred at 30 days for intractable pain in the setting of a patent graft. A third amputation occurred at 60 days for graft occlusion, intractable pain, and no salvage options. A fourth amputation occurred at 90 days for non-healing ulceration and continued pain in the setting of a patent graft. A fifth amputation occurred at 10 months for progressive ischemic necrosis and rest pain after multiple peripheral interventions were required to maintain assisted primary patency.

Infectious complications occurred in three patients. Two of the infections were secondary to staphylococcus aureus. One was a superficial wound infection that resolved with antibiotics and local wound care; the graft remained patent. The second wound infection occurred in the setting of an occluded graft and retained prosthetic material from previous vascular reconstructive procedures. This limb was subsequently amputated. There was one MRSA infection that required excision of a patent graft for eradication.

Medical management post-operatively included anti-platelet agents and statins. All patients were able to be maintained on clopidogrel, 20 of the 26 patients were able to be maintained on aspirin, 15 of the 26 patients were able to be maintained on statin therapy, 4 patients received chronic anti-coagulation with coumadin and 10 patients continued cilostazol post-operatively. One patient was maintained on pentoxifylline post-operatively. The goals of pain relief and wound healing were accomplished in 21 patients (81%). Of these, 18 (86%) achieved markedly improved status according to Rutherford.¹⁶

Discussion

EVH allows acquisition of a suitable conduit from above the knee and below the knee with minimal incisions to heal in vascular-compromised patients at risk for wound complications. Our observation has been that it is more tedious to perform EVH below the knee in patients with chronic CLI than to perform EVH above the knee. Close attention to technical detail is required to avoid wound complications due to skin ischemia and hematoma formation in chronically ischemic limbs. The procedure is not risk-free, as evidenced by skin necrosis and tunnel hematoma in one EVH-related wound problem in our series. This is of concern because of the risk for graft exposure, which we did not experience in this series. We were able to obtain adequate conduit in a minimally invasive fashion in all patients without conversion to an open technique.

This is a compromised group of patients at risk of limb loss, functional impairment, and cardiovascular morbidity and mortality. Risk factors, including male sex, age, ischemic necrosis and rest pain, diabetes mellitus, cigarette smoking, hypercholesterolemia, hypertension, coronary artery disease, previously intervened upon limbs, and distal leg bypasses make these patients a challenge to relieve ischemic pain and heal ischemic wounds in a safe and reproducible fashion. The “con” perspective suggests that a PAD population may not be improved in terms of functional status and quality of life.^17,18 Our “pro” approach to this challenging issue has been a program of total commitment to limb salvage.¹⁹ This implies the practitioner skill set, health care system resources, and technology to implement and maintain this type of program. With the rapid emergence of endovascular technology as it relates to the peripheral arterial circulation, coupled with traditional vascular surgical practice and procedures, there are now multiple treatment options for these patients short of primary amputation. Patient referral to an appropriate center should be made when necessary where a complete evaluation can be undertaken and treatment decisions made consistent with current guidelines. The recent publication of the American College of Cardiology/American Heart Association Guidelines for Management of Patients with Peripheral Arterial Disease provided all practitioners with remarkable access to a disease state that has largely been underrecognized and undertreated in the past.²⁰ An additional implication of the philosophy of total commitment to limb salvage is the concept that limb salvage involves more than a patent graft. This is supported by our experience of patients receiving amputation with patent grafts for continued pain and the ramifications/logistics of chronic wound care. Additionally, we, and others, have the experience of occluded grafts with healed wounds and resolved pain.^21,22 Thus, an open mind and reasonable expectations facilitate successful outcomes. Success is often evidenced by healed wounds, preserved ankle joints, and mobility shifting from wheelchair-bound to walker/cane-assisted or assisted ambulation to independent walking. A healed limb is not necessarily synonymous with graft patency.

Just as lower extremity PAD comes in a variety of anatomical configurations, our infrainguinal vascular reconstructions were a varied lot, dependent on the structural arterial anatomy. In this regard, imaging technology was vital to operative planning. With surprising frequency, distal target vessels could not be visualized with conventional angiography due to severe disease with minimal collateralization and slow filling. These arteriograms were often interpreted as “severe distal disease”, when in fact the imaging technology was merely inadequate. The 64-detector CT angiography has allowed complete visualization of the distal target vessels in the leg and foot. This facilitates efficient operative planning and avoids “blind” target vessel exploration with subsequent abandonment of the procedure if no suitable target is found. Additionally, CT angiography has allowed definitive imaging postoperatively when duplex scanning suggests further investigation is indicated for graft surveillance.

Conduit choice is clearly of great importance in infrainguinal vascular reconstructive surgery and autogenous greater saphenous vein is the preferred conduit. Secondary autogenous vein conduits include lesser saphenous vein and arm veins. Vein mapping was utilized throughout this experience, and although we were able to procure adequate greater saphenous vein, no lesser saphenous veins were found to be usable due to small size. Our experience with arm veins has been limited, but they remain a viable source of autogenous conduit when available. Alternative conduit choices for infrainguinal vascular reconstruction include PTFE with performed distal cuff and cryopreserved allograft. These alternative conduits were not utilized in this experience.

Medical therapy post-operatively is important to the overall success of this limb salvage program. We found that we were able to maintain anti-platelet therapy with clopidogrel and low-dose aspirin fairly easily despite issues with the cost of clopidogrel. We began to focus on statin use during the period of this review and feel that it provides an added beneficial effect.²³ We now make every effort to institute or continue their use in these patients. We did not track utilization of ACE inhibitors or beta-blockers, but feel that these medications are also important in the overall medical management of this patient population.

It is important to utilize graft surveillance, as evidenced by our finding of vein graft stenoses in four patients. Three of these grafts were salvaged with percutaneous peripheral intervention and two remained patent. It is noteworthy that the vein graft stenoses all occurred in the proximal one-third of the vein graft. In the reversed position, this corresponds to the most distal portion of the harvested greater saphenous vein, and due to its relatively smaller size and exposure to the effects of chronic ischemia, may well be prone to early fibrointimal hyperplasia with resultant vein graft stenosis. It has been our experience that this situation responds favorably to percutaneous peripheral intervention.

The goal of limb salvage was achieved in 21 (81%) patients and 18 (86%) of these were markedly improved. There were 5 amputations, 2 of which occurred with patent grafts and 3 of which occurred with occluded grafts. Despite the hope that above-knee and below-knee amputees will acquire appropriate prostheses and progress to independent ambulation, this is often not the case. With a multi-disciplinary approach to care, many of these patients can achieve limb salvage and progress to independent ambulation. This status can then be maintained with proper follow-up and surveillance. Our results have reinforced our philosophy of total commitment to limb salvage.

This study is an observational review, with a small number of patients. Nevertheless, we were able to document the feasibility of utilizing EVH in infrainguinal vascular reconstruction in chronic CLI. Additionally, we observed less postoperative incisional pain with EVH than with a traditional open technique. The ability to procure adequate quantity and quality of autogenous greater saphenous vein that performs satisfactorily as arterial conduit, the lesser impact of harvest site wound healing, and the added benefits of a preformed graft bed, and the ability for endoscopic graft inspection all make EVH an attractive technology for this patient population. Our findings of 85% of limbs previously intervened upon (surgical or percutaneous) and 85% of distal anastomoses below the knee reflect the state of practice patterns and patient populations in regard to chronic CLI.

The main drawback to EVH in our experience has been a prolongation of the procedure by 45–60 minutes. Our EVH is performed by surgical physician assistants who have several years’ experience with this technology in harvesting greater saphenous vein for coronary artery bypass grafting. This factor mitigated a “learning curve,” but did not obviate the added time. It is our opinion that there are anatomic and physiologic factors that make EVH for coronary artery bypass grafting different from EVH for infrainguinal vascular reconstruction. In EVH for coronary bypass conduit, greater saphenous vein is typically harvested from the thigh. It is therefore contained in an anatomic environment with a greater tissue mass and is more “tolerant and forgiving.” In contrast, EVH for leg bypass conduit utilizes greater saphenous vein from both above and below the knee. Below the knee, there is less tissue mass, and thus less margin for success without complications. Additionally, it is our impression that the physiologic environment of the lower extremity in the chronic CLI patient is different from that in the coronary artery bypass graft patient. With a chronically ischemic limb and the added deleterious effects of active open wounds, possible infection (overt or smoldering) and multiple previous surgical or percutaneous procedures, the underlying pathophysiology impacts the soft tissue of the leg in such a way as to cause edema and an inflammatory response. This, in turn, may cause tissue planes to be indistinct and fibrotic, necessitating a slower and more controlled EVH approach to avoid undue damage to the vein as well as overlying skin.

Conclusion

As both surgical and percutaneous vascular intervention for chronic CLI continues to rapidly grow and evolve, the goal of achieving limb salvage with its attendant improvement in quality of life and functionality must be kept in focus. These clinical scenarios are problematic, with a high incidence of aggravating co-morbidities. Treatment plans are often multi-dimensional, with each component depending on the success of the other. Endoscopic vein harvest offers the surgical component a less invasive approach that appears to lessen wound complications in this compromised group of patients. Attention to technical detail and knowledge of conduit anatomy facilitate successful outcomes.