Author + information
- Received May 25, 2004
- Revision received August 26, 2004
- Accepted September 14, 2004
- Published online December 21, 2004.
- Andrew J. Feiring, MD, FACC, FCSAI* (, )
- Amy A. Wesolowski, RN and
- Susan Lade, RN
- ↵*Reprint requests and correspondence:
Dr. Andrew J. Feiring, Director Cardiac and Vascular Intervention, Columbia-St. Mary's Medical Center, Suite 208, 2015 East Newport Avenue, Milwaukee, Wisconsin 53211
Objectives The objective of this study was an investigation of the safety and efficacy of primary below-knee stent-supported angioplasty (BKSSA) for restoring straight inline arterial flow in patients with critical limb ischemia (CLI) or lifestyle-limiting claudication (LLC).
Background Surgical tibial bypass for CLI and severe LLC is associated with significant morbidity, mortality, and graft failure, whereas percutaneous angioplasty is suboptimal.
Methods Below-knee stent-supported angioplasty was attempted in 82 patients (92 limbs) with either CLI (68%) or severe LLC (32%). Patients received daily aspirin, thienopyridine, and glycoprotein IIb/IIIa agents during the procedure. One-month major adverse events (MAEs) were defined as death, myocardial infarction, major unplanned amputation, need for surgical revascularization, or major bleeding. Clinical success was defined as improved resting ankle brachial index by ≥0.10, relief of resting pain, healing of ulceration or amputation, and improvement of claudication.
Results Mean age of patients was 74 ± 17 years. In 86 limbs, straight inline flow was restored to at least one tibial vessel. Technical success was 94% for de novo lesions and there were no MAEs. Ankle brachial indexes increased for all groups (CLI = 0.32 ± 0.13 to 0.9 ± 0.14 and LLC = 0.65 ± 0.09 to 0.95 ± 0.12; p ≤ 0.0001, pre vs. post). Relief of rest pain and healing of ulcerations and amputations were seen in 96% (47 of 49) of patients with CLI who underwent successful intervention.
Conclusions Below-knee stent-supported angioplasty for CLI and LLC improves ankle brachial indexes comparable to tibial bypass, heals amputations and ulcerations, relieves rest pain, and improves ambulation. Because BKSSA is associated with minimal MAE, it may hold promise as an alternative therapy for patients with CLI and LLC.
Critical limb ischemia (CLI) and severe lifestyle-limiting claudication (LLC) are often caused by flow-limiting stenosis or occlusion of a remaining below-knee artery. There is no optimal remedy for these patients. The focus of surgical therapy in patients with CLI is to either restore straight inline arterial blood flow to at least one tibial vessel or to amputate the limb when tissue loss or pain becomes intractable. Tibial bypass is underutilized in elderly and high-risk patients because of considerable procedural morbidity and mortality (1,2). Although bypass is usually reserved for patients with rest pain or tissue loss, it is infrequently recommended for claudicants. In these patients, the risk-benefit ratio is often unfavorable because of the attendant morbidity and mortality, as well as subsequent risk of graft failure and the need to conserve saphenous veins for future coronary bypass (2). Despite encouraging reports (3–6), the value of angioplasty for below-knee arterial CLI and LLC has not been widely accepted because of high restenosis rates, abrupt vessel closure, thrombus formation, perforation, embolization, and lack of demonstrable hemodynamic improvement (2,7–10).
In patients undergoing coronary intervention, the use of stents and antiplatelet antagonists has markedly reduced acute complications and symptomatic restenosis compared with balloon angioplasty. Extrapolating from this experience, we postulated that patients with CLI and severe LLC with jeopardized single-vessel runoff would benefit from similar interventional techniques, including primary below-knee stent-supported angioplasty (BKSSA) and aggressive platelet blockade.
Between January 2000 and January 2003, patients with CLI and LLC were evaluated. Patients with CLI were usually referred because of lack of adequate vascular conduit or surgical target, multiple co-morbidities, as a last attempt before amputation, or patient unwillingness to undergo surgery. Critical limb ischemia was defined as: 1) persistent, recurring rest pain requiring analgesia and an ankle systolic pressure ≤50 mm Hg and/or toe systolic pressure ≤30 mm Hg; and/or 2) ulceration, gangrene, or nonhealing wounds of the foot with ankle systolic pressure ≤50 mm Hg or toe systolic pressure ≤30 mm Hg (10). Patients with CLI are by definition categorized as Rutherford category IV to VI (12). Lifestyle-limiting claudication was defined as Rutherford category II to III associated with jeopardized single-vessel runoff or complete trifurcation vessel occlusion. Before intervention, patients had rest and exercise ankle brachial indexes (ABIs) recorded. Patients with noncompressible tibial vessels were excluded from analysis unless the postintervention ABIs could be appropriately measured.
The decision to intervene was made at the time of diagnostic angiography. Jeopardized runoff was defined as either complete trifurcation occlusion or a ≥50% stenosis of the remaining tibial vessel. Attempted revascularization was driven by clinical necessity irrespective of the degree of collateralization. Because all patients were symptomatic, the degree of collateralization was thought to be suboptimal. Patients with more proximal arterial inflow disease, in addition to below-knee disease, were included if upper-level revascularization was successful. Patients with total occlusions that could not be crossed with a wire were excluded from subsequent analysis. If both limbs were affected, the extremity with the most extensive tissue loss or the one that was most symptomatic was addressed first.
Patients with histories suggestive of active coronary ischemia were evaluated and underwent appropriate surgical or interventional myocardial revascularization before vascular intervention. Patients with renal insufficiency (creatinine ≥1.5 mg/dl) were adequately hydrated before and after the procedure. Starting in 2001, all patients with renal insufficiency received pre- and post-procedural mucomyst at 600 mg oral twice a day.
The contralateral retrograde approach was used in all but three patients. Interventions were performed through a 5-F, 40 to 45 cm sheath (Rabbe, Cook, Bloomington, Indiana, or Pinnacle, Boston Scientific, Maple Grove, Minnesota). If multilevel intervention was contemplated, a 6-F sheath was used. The goal was to maximize runoff and establish continuous inline flow in at least one below-knee vessel to the ankle. The anterior or posterior tibial arteries were preferred targets if available. Interventions were performed with 135-cm low-profile coronary balloon catheters. Total occlusions were crossed with 0.014-inch hydrophilic wires (Choice PT, Boston Scientific, or Shinobi, Cordis, Miami, Florida). Lesions were nominally predilated by one-half millimeter smaller than the reference vessel. Four lesions were resistant to high-pressure balloon inflations and subsequently underwent successful Rotablation (Boston Scientific) or Cutting Balloon (Boston Scientific) angioplasty. A variety of balloon-expandable and self-expanding coronary stents were used and deployed to obtain a −10% residual. Angiographic success was defined as dilation of all critical inflow lesions with a resulting residual stenosis of ≤20%. Technical success was defined as a residual stenosis of <20% after stent placement; Thrombolysis In Myocardial Infarction (TIMI) III antegrade flow; straight inline flow to the ankle in at least one tibial vessel; and absence of distal embolization, perforation, or need for unplanned corrective surgical intervention. All sheaths were either removed immediately or within 1 h of the procedure. Vascular closure devices were used when appropriate.
Before angiography all patients were started on 81 mg aspirin or clopidogrel 75 mg daily. For patients not taking clopidogrel, a loading dose of 450 mg was given before the procedure. After arterial access, patients received unfractionated heparin at 60 μ/kg. Activated clotting times were not monitored. After stent placement, abciximab (Reopro, Eli Lilly, Indianapolis, Indiana) (0.125/mg/kg) followed by an infusion at 10 μg/min for 12 h, or ebtifibatide (Integrilin, Cor Therapeutics, South San Francisco California) (180 U/kg bolus × 2) followed by an 18-h infusion) were administered. If the 4-h platelet count fell ≤100,000, then the infusion was discontinued. Patients were discharged on indefinite aspirin and clopidogrel 75 mg/day for six months or longer if financially feasible. Patients who were intolerant of clopidogrel received ticlopidine 250 mg twice daily.
Clinical end points
Clinical major adverse events (MAEs) were defined as death, stroke, myocardial infarction (MI), renal failure, retroperitoneal bleed, unplanned tibial/pedal bypass, major infection, compartment syndrome, acute renal failure, or need for procedure-related transfusion. Clinically important renal failure was defined as an elevation of post-procedure creatinine ≥0.5 mg. Clinical success was defined as continued relief of ischemic rest pain, reduction in severity of claudication, healing of ulceration, and freedom from unplanned surgical amputation or bypass surgery. Scheduled patient follow-up occurred at four weeks, six months, and one year. Rest and exercise ABIs were obtained before and four weeks after the procedure. Additional ABIs were obtained only if the patient's clinical situation deteriorated. At six months patients were graded on a “clinical improvement scale” where 0 represented entry clinical status, +1, +2, and +3 represented minimal, moderately, and marked improvements, and values −1, −2, and −3 represented mild, moderate, and marked worsening, respectively (13). Repeat angiography and target vessel revascularization were dictated by recurrent ulceration, rest pain, or worsening claudication.
Descriptive data are presented as the mean value ± SD. Mean data are presented as box-and-whisker graphs, where the central box represents the values from the lower to upper quartile (25th to 75th percentile), the middle line represents the median, and a line extends from the minimum to the maximum value, excluding “outside” values, which are displayed as separate points. Significant changes in ABI before and after intervention were evaluated by a paired Student ttest with p ≤ 0.05 indicating statistical significance. Because failure to cross a total occlusion had no influence on standard treatment, these patients were subsequently excluded from further analysis.
The clinical course of the 82 patients who underwent angiographic evaluation is described in Figure 1.In six patients with CLI (7%), there was failure to cross the total occlusion with a wire. No patient's condition was worsened by the attempted intervention. Four of these patients went on to below-knee amputation and two had successful distal bypass.
Seventy-six patients and 86 limbs underwent successful intervention. There was no significant difference between the mean age of claudicants and those with CLI. Baseline demographic data are presented in Table 1.Men outnumbered women by slightly <3:2. The mean age of patients was 74.2 ± 16.6 years with a range of 35 to 101 years. The average age of women patients was 4.5 years older than men. Demographic data are presented in Table 1. Before BKSSA, 12 patients (16%) underwent myocardial revascularization, three underwent coronary artery bypass surgery, and nine underwent multivessel coronary angioplasty. Below-knee stent-supported angioplasty was performed between 2 and 21 days after myocardial revascularization.
All infrageniculate lesions were either category III or IV (2). Technical success for all de novo lesions was 93% (86 of 92 limbs) and 94% (90 of 96) for all patients, including those with restenotic lesions. Once the wire crossed the lesion, technical success for restoring inline flow to the ankle was 100% for the 86 limbs (76 patients). There were no procedural- or one-month–related deaths, MIs, amputations, unplanned vascular surgeries, or worsening of lower limb ischemia. There was no evidence of embolization, thrombosis, perforation, or arterial spasm. There were no major access site complications or clinically relevant contrast nephropathy. One patient developed uncomplicated IIb/IIIa-related thrombocytopenia, which delayed discharge by three days. Ninety-three percent of all patients were discharged within 24 h of admission.
A negative residual was achieved for all stents placed. A total of 197 stents (mean = 2.3 ± 0.6 per patient) were implanted. Assortments of balloon expandable and self-expanding coronary stents were deployed (Table 2).Self-expanding stents were post-deployed at low pressure with a coronary balloon size 0.5 mm greater than the reference vessel. There was no appreciable difference noted regarding the deliverability or performance characteristics of the stents. The Hepacoat (Cordis) stent was the later stent of choice because of its theoretical antithrombotic properties. We preferred stents with closed-cell designs, unless side branch salvage was intended. Stent size ranged from 2.5 to 5.0 mm. The median size stent for the distal popliteal and tibioperoneal trunk was 4.0 mm, proximal anterior tibial/posterior tibial was 3.5 mm, and distal vessel or peroneal artery 2.5 to 3.0 mm. Concomitant superficial femoral and or above-knee popliteal artery interventions were performed in 27% of patients. Angiographic examples of critical limb ischemia and corresponding tibial interventions are shown in Figures 2 and 3.⇓⇓
The preintervention mean ABI for the entire cohort was 0.42 ± 0.13, which improved to 0.90 ± 0.14 at one month (p ≤ 0.0001) (Fig. 4).In patients with CLI, the mean ABI increased from 0.32 ± 0.13 to 0.9 ± 0.14 (p ≤ 0.0001) (Fig. 5).Those with claudication improved from 0.65 ± 0.09 to 0.95 ± 0.12, (p ≤ 0.001) (Fig. 6).At six months, only three patients felt that they had not improved, and no patients felt that they were worse off (Fig. 7).
Clinical outcomes during the first year after BKSSA are shown in Table 3.During the first month there were no MAE. All patients completed the six-month and one-year clinical follow-up. Planned amputations occurred in eight patients. Six patients underwent toe amputations and two had transmetatarsal amputations. All amputations healed. Two patients with ulcerations initially healed and then recurred secondary to in-stent restenosis. Repeat angioplasty was unsuccessful and they ultimately underwent scheduled below-knee amputation. All patients with resting ischemic pain improved. These improvements were sustained over the first year. The one-year limb salvage rate for CLI was 47 of 49 limbs (96%) for those patients with initial technical success. When the data were expressed on an intention-to-treat basis, the one-year limb salvage rate for all patients undergoing angiographic evaluation was 87%. During the first year there were no deaths, MIs, or strokes. Between the 6th and 12th months, four patients underwent repeat angiographic evaluation. Two patients developed in-stent restenosis and two demonstrated progressive superficial femoral artery disease. All patients underwent successful percutaneous therapy. Contemporaneous angiography demonstrated no evidence of stent crush.
This study underscores the following unique observations:
Below-knee stent-support angioplasty is associated with a high rate of angiographic success, minimal MAE, and short hospital stay, even in the elderly with multiple co-morbidities.
Stent-supported below-knee percutaneous transluminal angioplasty (PTA) results in normalization of the one-month postintervention ABIs comparable to that seen after successful surgical tibial bypass (13).
During the first year after successful stenting, 95% of patients (82 of 86) demonstrated tissue healing, relief of resting ischemic pain, and improvement in claudication.
Using aspirin, thienopyridine, and glycoprotein IIb/IIIa inhibitor, we observed no evidence of abrupt stent occlusion or distal embolization.
Limitations of current therapy for CLI and below-knee LLC
Patients with CLI have a dismal clinical course. During the first year of diagnosis, mortality is 20% to 30%, and an equivalent number undergo amputations or suffer persistent pain (2). The surgical options of either amputation or tibial bypass have changed little since 1966 (14). Surgical complications include death (1.3% to 6%), MI (1.9% to 3.4%), wound infections (10% to 30%), leg edema (50% to 100%), and early graft failure rates (6% to 49%) requiring repeat surgery (2,15). When the conduit used is polytetrafluoroethylene, the five-year patency rates are as low as 33% (2). Ultimately, bypass surgery can only be offered to a minority of patients because of inadequate conduit, poor runoff, advanced age, and cardiac and/or other co-morbidities. Thus, definitive treatment is often withheld from those patients who would most benefit.
Angioplasty for CLI was reported nearly 40 years ago (16). Initial complications were high and success low, secondary to bulky catheters and guide wires, perforation, thrombosis, and vascular trauma related to antegrade puncture (2,5,6). Despite recent technologic advances, Hanna et al. (6) reported a 21% procedural complication rate, including embolization and thrombosis in patients with CLI undergoing balloon angioplasty, whereas Dorros et al. (3) demonstrated that PTA of below-knee arteries could be performed with relative safety and with satisfactory results. The prevailing surgical perspective regarding tibial angioplasty was articulated by Dormandy and Rutherford (2), who stated, “below-knee PTA should be considered in patients with limb-threatening ischemia without good surgical options or lack of conduit” and that “stents have not been adapted to use in the infrapopliteal vasculature.” Thus, the goal of this study was to address these perceived limitations using contemporary coronary interventional techniques.
Benefits of BKSSA for CLI
This is the first study to demonstrate the clinical and hemodynamic efficacy and safety of primary stent support below-knee angioplasty for CLI and severe LLC. Previously, only surgical bypass has been shown to improve hemodynamics measured by ABIs (13). In this study, the magnitude of ABI improvement was equivalent to that obtained with bypass surgery, yet it was accomplished with a fraction of surgery's morbidity and mortality. During the first year, improvement in distal perfusion pressure paralleled relief of resting pain, resolution of all but two ulcerations, and healing of all elective amputations. Despite the advanced age of our patients (1 of 3 ≥80 years), BBSSKA could be accomplished with minimal procedural risks. Most importantly, there was no clinical penalty for attempted endovascular intervention. If needed, repeat PTA could be attempted, and BKSSA did not appear to preclude future surgical options.
During the first year there were no deaths or MIs and only two leg amputations. This is in contrast to a 30% death and amputation rate expected in traditionally treated CLI patients (17). The improved outcomes may have been secondary to the confluence of early coronary revascularization, avoidance of amputations and surgery, aggressive platelet inhibition, and universal use of lipid-lowering agents.
Stent-supported below-knee angioplasty for LLC
Claudication and CLI represent a continuum of the same atherosclerotic process. The expected rate of progression from claudication to amputation is reported as 1% per year, although more than 20% of claudicants will develop incapacitating LLC, seriously restricting the quality of life and independence of older patients (1,2,18,19). Surgical options for patients with severe LLC are limited, especially if venous conduit must be conserved for future coronary artery bypass. However, most importantly, there is the prospect that an unsuccessful tibial bypass may convert a “stable claudicant” into a limb salvage patient. For this same reason the use of PTA for below-knee claudicants has been restricted. This approach is supported by the Transatlantic Inter-Society Consensus document, which states, “infrapopliteal vessels are usually not treated unless there is critical acute or chronic limb ischemia” (2).
This is the first study to demonstrate that BKSSA can normalize ABI hemodynamics in patients with severe LLC. As in patients with CLI, improvement was accomplished with a minimum of MAE, and no patient was clinically worse off one year after intervention. Below-knee stent-supported angioplasty afforded many older patients the ability to improve their quality of life with minimal risk. Extrapolation of these data to patients with less severe claudication should be done cautiously. Our patients are a unique subset who have jeopardized single-vessel runoff and thus should be considered to be in a pre-CLI stage.
The endovascular methodology was standardized to minimize suboptimal interventional technique. We routinely used a retrograde contralateral approach, ≤6-F sheath size, sub 3.5-F balloons, 0.014-inch wires, high-pressure stent deployment, reduction of all stenosis ≤20%, maximization of distal runoff to ankle, and triple platelet inhibition. Applying this algorithm there were MAE. Using contralateral retrograde arterial access, we noticed little reduction in wire responsiveness, although an antegrade approach may have improved success for recanalizing total occlusions. The application of coronary stents to below-knee intervention is untested. However, both coronary and infragenoual arteries are medium-sized muscular arteries of similar dimensions. The median stent diameter deployed was 3.5 mm (range 2.5 to 5.0 mm), which is identical to the average-sized coronary stent. Because stents improve postintervention luminal cross-sectional area, prevent early recoil and negative remodeling, and tack up intimal flaps and dissections, there is little reason to suspect that this approach would not be effective for below-knee arterial interventions.
This was a single-center, single-operator, consecutive patient trial. Ankle brachial indexes were obtained before intervention to document hemodynamic impairment and at one month after intervention to measure improvement. Subsequent ABIs were obtained when there was a question of worsening clinical status. Therefore, one-year stent patency rate and ABI data are not reported. Nevertheless, the fate of patients who underwent BKSSA can be compared to the known natural history of patients with CLI. Patients undergoing successful intervention had a 96% freedom rate from major amputation compared with historic one-year amputation rates of 30% (2,15). Furthermore, the need for target vessel revascularization was only 8% compared with surgical revision rates as high as 49% (2). Published surgical data suggest that long-term patency of tibial bypass grafts for CLI may not be imperative as long as the duration of blood flow is enough to effect complete tissue healing (20). Therefore, although serial ABIs or angiographic follow-up would have been ideal, the fact that nearly all patients demonstrated sustained clinical improvement suggests that BKSSA may be a promising alternative to tibial bypass surgery.
The therapeutic goal for patients with LLC is to increase functionality, which can only be achieved by sustained hemodynamic improvement. Because all patients had jeopardized single-vessel runoff, occlusion of the sole remaining vessel would have resulted in worsening symptoms or CLI. During the first year, all patients with LLC improved and no patient regressed to baseline, suggesting continuous stent patency. Consequently, even without one-year ABIs, our data suggest that BKSSA for LCC has few of the limitations of tibial bypass surgery and provides an equivalent early benefit.
The possibility of stent crush cannot be excluded, although those returning for repeat intervention revealed no evidence of stent deformation. Furthermore, because all patients had jeopardized single-vessel runoff, mechanical compromise of their remaining tibial vessel would readily manifest itself as failure to heal ulceration, recurrence of rest pain, or worsening claudication. With the exception of the distal popliteal artery and anterior and posterior tibial arteries, all crural vessels are circumferentially protected by the anterior and posterior muscle compartments. When stenting these areas, we favor the use of self-expanding coronary nitinol stents (i.e., Radius, Boston Scientific).
The expense of implanting multiple stents can be considerable, although potential cost savings may be realized by shorter hospitalizations and reduction in surgeries, amputation, and extended rehabilitation. Moreover, with the advent of drug-eluting stents (DES), the price of bare-metal stents continues to decline. Our preliminary data suggest that clinically significant in-stent restenosis is ≤10%. However, because the need for repeat angiography was driven by worsening symptoms, we have no data on binary restenosis rates. It is hoped that issues of acute gain and late loss will be addressed by industry-supported trials. Although it is premature to speculate how DES will fit into this treatment paradigm, it is reasonable to postulate that DES should inhibit crural neointimal hyperplasia similar to coronary vessels, and thereby reduce restenosis in high-risk patients with diabetes, small-diameter vessels, and long lesions (21). To date we have treated 14 patients with high-risk variables for restenosis (not included in this analysis) with Cypher (Cordis) DES without adverse events.
Although we encountered no evidence of stent thromboses, the implication of abrupt stent occlusion is of paramount concern in patients with single-vessel runoff. For this reason we employed the same effective pharmacologic algorithm used to prevent abrupt closure after coronary stenting. This is the first study to report on the use of multilevel platelet blockade for infragenoual endovascular revascularization. The use of clopidogrel and aspirin was predicated on studies showing a significant reduction in stroke death and MI in patients with vascular disease (22), as well as extensive coronary interventional data demonstrating reduction in MAE after coronary stenting (23). Finally, IIb/IIIa agents were empirically added because of the well-known effect of these agents for reducing the incidence of early thrombotic complications after angioplasty (24). Thus, although the efficacy of these agents has been well established for coronary intervention, there are no specific data that support their effectiveness in lower extremity intervention. The effectiveness of these agents, either alone or in combination, will need to be evaluated in controlled trials.
This study demonstrates the early and midterm safety and efficacy of BKSSA both for CLI and severe LLC. Percutaneous intervention with primary stent-supported intervention offers immediate and consistent endoluminal reconstruction with near normalization of ABIs. These improvements in hemodynamics were associated with healing of ulceration, elective amputations, and ability to ambulate. Because primary stent-supported endovascular therapy is effective, has few adverse effects, and does not preclude future tibial bypass surgery, it may be considered as an option for patients with CLI and severe LLC with jeopardized single-vessel runoff. The relative benefits of adjunctive antiplatelet therapy, DES, and preintervention coronary artery revascularization need further study. (11)
- Abbreviations and acronyms
- ankle brachial index
- below-knee stent-supported angioplasty
- critical limb ischemia
- drug-eluting stent
- lifestyle-limiting claudication
- major adverse events
- myocardial infarction
- percutaneous transluminal angioplasty
- Received May 25, 2004.
- Revision received August 26, 2004.
- Accepted September 14, 2004.
- American College of Cardiology Foundation
- ↵Whittemore AD, Belkin M. Infrainguinal bypass. In: Rutherford, JB, editor. Vascular Surgery 2000. Philadelphia, PA: W.B. Saunders Co., 2000:998–1018.
- Dormandy J.A.,
- Rutherford R.B.
- Dorros G.,
- Jaff M.R.,
- Dorros A.M.,
- Mathiak L.M.,
- He T.
- Holm J.,
- Arfvidsson B.,
- Jivegard L.,
- Lundholm K.,
- Schersten T.,
- et al.
- Hanna G.P.,
- Fujise K.,
- Kjellgren O.,
- et al.
- ↵Schneider PA, Rutherford RB. Endovascular interventions in the management of chronic lower extremity ischemia. In: Rutherford JB, editor. Vascular Surgery 2000. 5th edition. Philadelphia, PA: W.B. Saunders Co., 1058–9.
- Dotter C.T.,
- Judkins M.P.