Clinical Insights into the Use of Embolic Protection Devices during Lower Extremity Peripheral Vascular Interventions

ABSTRACT: The use of embolic protection devices (EPD) during lower extremity peripheral vascular interventions (PVI) remains controversial. We examine the current data and present our Louisiana experience with discussions regarding the unresolved issues surrounding the use of EPD during lower extremity PVI. J INVASIVE CARDIOL 2009;21:418–422 Even though many experienced peripheral interventionalists are lowering their threshold for using embolic protection devices (EPD) during peripheral vascular interventions (PVI) for severe lower extremity peripheral arterial disease (PAD) and especially critical limb ischemia (CLI), the use of EPD remains controversial. As a cardiovascular surgeon, I find this controversy to be both intriguing and analogous to the controversy surrounding the justification of endovascular treatment of PAD over the last two decades where an overall lack of “hard” phase I data was the major challenge. There remain sparse “hard” phase I data even today by many accounts verifying the role of PVI in the treatment of PAD. Still, many of us continued to perform PVI because it was minimally invasive and was often “the right thing to do”, especially in the elderly high-risk PAD patient. Today, it is well accepted that most PAD can be treated with PVI even with sparse published phase I data. The plight of establishing EPD as a “standard of care” in lower extremity PVI today is somewhat analogous, but a procedure of this nature faces more challenges in the contemporary healthcare environment which is more cost, data and regulatory conscious and herein lies the crux of the controversy. What should not be controversial, however, is the tremendous clinical and therefore economic costs of CLI and amputations. Therefore, any “CLI tool or strategy” designed to improve CLI outcomes should be strongly considered, thoroughly investigated and given every benefit of the doubt, especially if it is found to be safe and feasible. The number of amputations performed yearly in the U.S. is estimated to be 160,000–180,000, with an estimated 10% yearly increase. In Europe, the estimate is 40,000 to 50,000 yearly for a combined estimate of > 220,000–240,000 amputations yearly.1–3 Since 1985, the U.S. amputation rate has increased from 19–30 per 100,000 persons each year with a 4- to 5-fold increase in the over 80-year age group to 140 per 100,000 person years.4,5 The population of patients who are over 80 years old is the single largest growing age group in the U.S.1 Within 1 year of being diagnosed with CLI, 40–50% of the now > 20 million U.S. diabetics will experience an amputation, and 20–25% will die.2,3,6 Overall, Existing Data Karnabatidis et al reported their experience using distal embolization in a prospective registry of 48 patients undergoing PVI on > 75% stenosis with a mean length of 52.2 mm.7 Fifty vessels were treated with nitinol filter baskets as protection. Procedural success was 98.3%, with one case of distal embolization. Particles of athermatous plaque and thrombi > 1.0 and > 3.0 mm were retrieved in 58% and 12% of their cases, respectively.7 In their univariate analysis, lesion length and diameter, acute thrombus, and chronic total occlusions (CTO) correlated with a higher embolization rate.7 Suri et al treated 10 superficial femoral (SFA) and popliteal arteries with the SilverHawk (ev3, Inc., Plymouth, Minnesota) atherectomy with EPD and observed macroembolization in 100%.8 Shammas et al, in the PROTECT registry, treated 40 patients, 56 vessels, 27% SFA CTOs with a treated length of 123.6 ± 102.0 mm, therefore representative of a group of patients with advanced disease.9 These authors reported a 55% incidence of macroembolization in the PTA ± stent group and 100% in the SilverHawk group. They defined clinically significant macrodebris as ≥ 2.0 mm in diameter with an overall incidence of 45% (37% in PTA/stent and 90.99% in the SilverHawk group). This underscores the high likelihood EPD will retrieve debris when utilized but further underscores our lack of even defining what is, or is not, clinically significant debris. We know distal embolization occurs from the pattern and number of “embolic hits” that occur during carotid PVI as measured by transcranial Doppler.10 Why should lower extremity (LE) PVI be any different? I strongly suspect it is not different. It is likely much worse in LE PVI than during carotid artery stenting when considering the plaque burden in LE arterial disease. The high incidence of Doppler embolic particle signals during each phase of carotid PVI has been well documented.10 Lam et al have recently reported a similar high incidence of “embolic hits” during various phases of LE PVI with the highest incidence being during atherectomy.11 Interestingly, in their conclusion, Lam et al did not support routine use of EPD during SFA PVI on the basis of only 1 of 60 patients experiencing an angiographic and clinical demonstrable distal emboli in a CLI patient with single vessel runoff.11 They did not use an EPD, only Doppler signals and only 9/60 (15%) had long TASC D SFA disease or chronic total occlusions (CTO). Also 52/60 (86.6%) had 2- to 3-vessel runoff. Therefore I suspect that if these authors had treated a more complex CLI patient population with a greater number of CTOs and used EPD in all patients, they would have reported a much higher incidence of micro- and macroembolization. Louisiana Data and Experience Our Louisiana single site experience with filter EPD in 98 infrainguinal PVIs in primarily CLI patients began in July 2006.12 PVI treatments included: sole laser = 5, sole plaque excision (PE) = 3, laser PTA = 28, PE/PTA = 5, laser PTA/stent = 38, PE PTA/stent = 5, and PTA/stent = 14. The EPD delivery success rate was 96/98 (97.9%). There were no major EPD complications but 12/21 (57.1%) infrapopliteal filter deployments experienced vasospasm. Overall, particulate EPD debris was identified in 69/98 (70.4%) of cases (Figure 1, A–C). No EPD debris was found in 29/98 (29.5%) cases. Minor debris was identified in 50/98 (51.0%) and major debris in 19/98 (19.3%). Predictors of major debris included bypass grafts, CTO, PE, PTA, and PTA/stenting but minor debris occurred with all types of PVI including laser. Our EPD experience has lowered our threshold for EPD deployment, stimulated our interest in identifying the PVI patient population that would most likely benefit from PVI, and taught us several lessons regarding EPD use during PVI. EPD “tips and tricks” learned during our experience include: 1. EPD placement. Cross all SFA and bypass graft lesions, thrombosis and CTOs with a 0.035 mm tapered hydrophilic Quick-Cross (Spectranetics Corp., Colorado Springs, Colorado) catheter and your wire of choice. Since June 1, 2008, we have crossed all CTOs with the CROSSER (FlowCardia, Sunnyvale, California) high ultrasonic energy CTO crossing device followed by the wire – Quick-Cross exchange. We believe this central luminal crossing of the vessel will minimize the embolization risk of crossing the CTO in a non-central luminal or subadventitial plane utilizing traditional wire-J-wire techniques. This central luminal crossing would then facilitate EPD delivery and maximize whatever definitive PVI option the operator selects. We deliver the filter EPD through the Quick-Cross catheter delivering the filter > 10–15 mm distal to the most distal lesion. We have found the Spider filter system to be user friendly and easily delivered by this method saving 2–3 steps and exchanges. 2. Filter sizing. Never oversize the filter to the vessel. Match the filter to the exact vessel size or 1.0 mm less. This will minimize vessel injury and vasospasm. 3. Infrapopliteal EPD. These vessels are prone to spasm therefore a liberal use of intra-arterial nitroglycerin, verapamil (Abbott Laboratories, Abbott Park, Illinois) and antispasmodic agents are recommended with infrapopliteal EPD. The Spider EPD sizes range from 3–7 mm. Filter oversizing and movement must be avoided during infrapopliteal EPD to decrease complications. 4. Minimize EPD migration. It is mandatory to minimize EPD filter movement after deployment to avoid complications. Keep the filter in full fluoroscopy view during all PVI exchanges and manipulation. Failure to minimize EPD wire movement may increase the incidence of vasospasm and intimal injury. 5. Postprocedural filter angiogram. Obtain a detailed final cine magnified angiogram of the filter to identify any debris to strategize EPD capture. Partial filter capture is now recommended with the 0.035 mm Quick-Cross catheter on all cases when debris is identified on angiography. The horseshoe-shaded opaque marker on the proximal filter “mouth” facilitates partial capture by capturing only the marker within the Quick-Cross catheter during filter retrieval. The inner Quick-Cross catheter edge is hydrophilic and the diameter of the Quick-Cross catheter is larger than the existing Spider capture system which facilitates partial filter capture therefore less likelihood to extrude or debris through the filter pores during capture with a “full basket” (Figure 2A). 6. FilterWire support. The Spider filter wire is easily delivered and supportive enough to allow the majority of PVI’s including laser atherectomy, plaque excisional atherectomy and stenting. You must keep the filter wire wet with each exchange to minimize migration and facilitate the PVI. The Spider device is not compatible for use with the CSI orbital atherectomy system (ev3, Inc., Plymouth, Minnesota). 7. The “full basket”. If a filter becomes occluded (“full basket”) before the final angiogram, partially capture the filter and reposition a second new filter as it is difficult to “clean” a “full basket” (Figure 2A). This is not uncommon when performing complex PVI on a long SFA in-stent thrombosis or thrombosed bypass grafts. If sluggish flow is encountered any time during a contrast injection, visualize the filter and suspect embolization. We use GP IIb/IIIa inhibition in all PVI cases in which we suspect a large plaque-thrombus burden and the case is at high risk for distal embolization. 8. Adequate landing zone. A relatively disease free vessel of appropriate size must be identified. The mid-distal popliteal artery at least 5–10 mm above the infrapopliteal trifurcation is our ideal landing zone in the majority of cases. Our experience with DPD has revealed these “tools” to be safe and feasible and we currently consider EPD use in the following “high risk for embolization” clinical scenarios: 1. All bypass graft occlusions, thrombosis and any graft stenosis with a large plaque or thrombotic burden with an adequate landing zone; 2. All SFA stent occlusions, thrombosis and most cases of long SFA ISR with an adequate landing zone; 3. Patients with acute (Discussion The patient with lower extremity PAD, and especially CLI, is already predisposed to embolic risks during PVI. The CLI patients have multilevel disease with large pre-existing atherosclerotic plaque burdens, a likely high incidence of thrombus, high incidence of diabetes and hypercoagulability and the PVI occurs in a “low flow” environment as compared to PCI. These higher risk patients need optimal “proximal protection” strategies with optimal anticoagulation and antiplatelet pharmacotherapy coupled with optimal periprocedural techniques and potentially downstream “distal protection” with EPDs.13–15 While there are few published efficacy data supporting EPD use in LE PVI, a reasonable conclusion of the results of the multiple single-site results is that distal emboli are very common during LE PVI, EPD appear safe and at least feasible during LE PVI, and that more clinical data need to be accumulated to fully define the clinical entity of distal embolization and the true role of EPD during LE PVI. EPDs are considered “standard of care” during CAS and coronary artery bypass graft PCI. Traditional dogma would require multicenter, randomized data for EPD to become a “standard of care” during LE PVI. We would also agree with this, but somewhat reluctantly. I would strongly raise the question as to why EPD during carotid PVI became a “standard of care” without a wealth of “hard” Phase I data. I suspect the frequently made comment, “the brain is not as forgiving as the leg,” regarding embolic debris is at least very true and likely partially responsible for the clinical adoption of EPDs during carotid PPI. We would argue that the LE is not as clinically tolerant of embolic debris as perceived, especially the already compromised CLI limb. Analogous to a detailed neurologist exam and cerebral imaging post carotid PVI identifying a high incidence of periprocedural PVI embolic debris and clinical sequelae, we suspect a similar high likelihood of objective findings and clinical sequelae would be identified if the pedal microcirculation could be held to similar rigorous objective and clinical post-procedural assessments. Further complicating the issue, we have no “gold standard” technology analogous to cardiac nuclear scanning or cardiac MRI to assess pedal microcirculatory function. I question how many interventionalists (cardiologists, radiologists, and surgeons) perform a detailed physical exam and objective assessment of the foot pre- and post-PVI. It is likely that most contemporary interventionalists were not trained to access even the “minor” clinical signs and symptoms of distal microcirculatory embolic debris. Postprocedural clinical assessment is not an easy task and requires a major commitment by the interventionalist. Pain is a poor assessment for emboli as most CLI patients have diabetic neuropathy with an insensate foot and therefore have altered pain perception. Massive distal emboli are easily diagnosed but not so for microcirculatory debris that is often clinically asymptomatic, but which I strongly suggest are clinically relevant (Figure 2B). Distal tissue and skin changes oftentimes do not occur until 48–72 hours post PVI and therefore after discharge from the hospital (Figure 2C). We strongly suspect similar microcirculatory injury frequently occurs during LE PPI for CLI analogous to the cardiac microcirculatory injury and dysfunction during high-risk PCI or in post CABG vein grafts. We simply have no biomarkers or tools to accurately access the pedal microcirculation. When objective tools to access pedal microcirculatory function become available, I believe we will identify a very high incidence of distal embolic debris during LE PVI and be able to fully document their clinical significance. Several emerging technologies are being developed to access wound healing and for evaluating pedal microcirculatory function pre and post PVI. The OxyVu (Hypermed, Inc., Waltham, Massachusetts) uses medical hyperspectral imaging, a camera-based diagnostic tool that quantifies hyperspectral tissue oxygenation in diabetic foot ulcer wound healing, therefore has the potential to be an objective assessment of pedal microcirculatory disease and function. The Sensilase System (Väsamed, Inc., Eden Prairie, Minnesota) is a laser Doppler exam combining skin perfusion pressure (SPP) and pulse volume recording (PVR) to access the wound healing and access capillary circulation and help predict the level of amputation. Both technologies are under investigation and may show promise as objective physiologic assessments of pedal microcirculatory function and may help define the clinical role of distal embolization associated with PVI. There are several clinical trials being organized at this time. Critics of EPD use in LE PVI will cite the lack of Phase I data proving efficacy, lack of FDA-approved indications, economic costs, additional procedural steps, and potential complications as reasons today that EPD cannot be considered a “standard of care” during contemporary PVI. I would agree in general with each of those issues but contend that until these issues have been resolved, the lower extremity PAD patient and especially the CLI patient must be individualized and the potential risks versus benefits of EPD during high-risk PVI be considered. We would submit we cannot today accurately identify macroembolization, miss all microembolization and grossly underestimate the impact of “clinically significant” distal emboli. One may argue that if embolization can be identified, it can be easily and successfully treated with thrombolysis, GP IIb/IIIa agents, simple aspiration, laser or mechanical thrombectomy. We strongly submit that all of these “embolic rescue” strategies are not simple and add significantly to the clinical and economic costs to these very complex PVI cases that are already at higher risks for contrast-induced nephropathy, limb loss, and vascular access bleeding complications.15,16 We submit that a high-risk CLI and non-CLI patient population can be identified in which it would be better to prevent distal embolization than to wait and treat periprocedural distal macro/microembolization. It is unlikely EPD will become a “standard of care” for all LE PVI’s but it is likely a “higher risk” patient population can be identified who would benefit from EPD utilization. Our Louisiana experience has been favorable and EPD has provided many of our complex CLI patients with a protective “safety net” during complex limb salvage PVI where microembolic debris could have been catastrophic. Creative EPD strategies are now applied in high-risk patients who otherwise would have been offered only bypass surgical option. The filter EPD is rapidly becoming as important a tool in our “CLI toolbox” as any of the other emerging infrainguinal endovascular revascularization “tools” and has expanded our abilities to apply successful endovascular revascularization to an increasingly more complex population of CLI patients. Conclusion In conclusion, contemporary EPD use in the lower extremity PVI cannot be considered a “standard of care” as many issues surrounding efficacy, costs and risks must be verified by Phase I randomized data. Does this sound familiar and analogous to the initial applications of PVI for PAD of the 1990s? Many questions and answers regarding EPD remain elusive, especially to the clinical relevance of distal embolization in the lower extremity. I strongly suspect distal micro/macro embolization is not clinically irrelevant and is analogous to the clinical relevance identified during high-risk PCI. The emerging SensiLase and OxyVu technologies will help us answer many of the unanswered questions. It is time for industry, clinicians, and the FDA to devise a strategy to scientifically address the questions and unresolved efficacy issues remaining with EPD utilization in percutaneous lower extremity interventions as there are early reports and experiences providing safety and feasibility data warranting further randomized, multicenter investigations into the clinical efficacy of this promising strategy with the potential to improve outcomes in the high-risk lower extremity PAD and CLI patient population. From the *Louisiana Cardiovascular and Limb Salvage Center, Lafayette, Louisiana, and the §Cardiovascular Institute of the South, Lafayette, Louisiana. The authors report no conflicts of interest regarding the content herein. Manuscript submitted April 1, 2009, final version accepted April 15, 2009. Address for correspondence: David E. Allie, MD, Medical Director, Director of Cardiovascular and Endovascular Surgery, Louisiana Cardiovascular and Limb Salvage Center, APMC, 2730 Ambassador Caffery Pkwy, Suite 202-A, Lafayette, LA 70506. E-mail: david.allie@CV-LimbSalvage.com

1. Yost ML. Peripheral arterial disease: A report by The Sage Group. 2004; Vol. II.

2. Akbari CM. Diabetes and critical limb ischemia. Endovasc Today 2004;2:66–69.

3. Allie DE, Hebert CJ, Lirtzman MD, et al. Critical limb ischemia: A global epidemic. A critical analysis of current treatment unmasks the clinical and economic costs of CLI. Eur Interv 2005:1:75–84.

4. Dormandy JA, Ray S. The natural history of peripheral arterial disease. In: Tooke JE, Loew GD (eds). A Textbook of Vascular Medicine. London: Arnold. 1996, pp.162–175.

5. Hobson RW, Lynch TG, Padberg FTJ, et al. Results of revascularization and amputation in severe lower extremity ischemia: A five-year clinical experience. J Vasc Surg 1985:2:174–185.

6. Paaske WP, Lausten J. Femorodistal bypass grafting: Quality of life and socioeconomic aspects. Eur J Vasc Endovasc Surg 1995;10:226–230.

7. Karnabatidis D, Katsanos K, Kagadis GC, et al. Distal embolism during percutaneous revasculiarization of infra-aortic arterial occlusive disease: An underestimated phenomenon. J Endovasc Ther 2006;13:269–280.

8. Suri R, Wholey MH, Postoak D, et al. Distal embolic protection during femoropopliteal atherectomy. Catheter Cardiovasc Interv 2006;67:417–422.

9. Shammas NW, Dippel EJ, Coiner D, et al. Preventing lower extremity distal embolization using embolic filter protection: Results of the PROTECT registry. J Endovasc Ther 2008;15:270–276.

10. Antonius Carotid Endarterectomy, Angioplasty, and Stenting Study Group. Transcranial Doppler monitoring in angioplasty and stenting of the carotid bifurcation. J Endovasc Ther 2003;10:702–710.

11. Lam RC, Shah S, Faries PL, et al. Incidence and clinical significance of distal embolization during percutaneous interventions involving the superficial artery. J Vasc Surg 2007;46:1155–1159.

12. Allie DE, Hebert CJ, Patlola RR, Walker CM. Embolic protection devices in infrainguinal interventions for critical limb ischemia: A 24-month experience in 115 patients. Accepted abstract for oral presentation. Transcatheter Cardiovascular Therapeutics (TCT) 2008.Washington Convention Center, Washington, D.C., October 16, 2008.

13. Allie DE, Hebert CJ, Lirtzman MD, et al. Safety and feasibility report of combined direct thrombin and GP IIb/IIIa inhibition with bivalirudin and tirofiban in peripheral vascular disease intervention: “Treating critical limb ischemia like acute coronary syndrome.” J Invasive Cardiol 2005;17:427–432.

14. Allie DE, Hebert CJ, Lirtzman MD, et al. Combined glycoprotein IIb/IIIa and direct thrombin inhibition with eptifibatide and bivalirudin in the interventional treatment of critical limb ischemia: A safety and feasibility report. Vascular Disease Management 2006;3:368–375.

15. Allie DE. Patients with peripheral artery disease undergoing percutaneous peripheral interventions are at increased risk for complications without adequate anticoagulation. J Invasive Cardiol 2004;16(Suppl G):G18–G21.

16. Shammas NW, Lemke JH, Dippel EJ, et al. In-hospital complications of peripheral vascular interventions using unfractionated heparin as the primary anticoagulant. J Invasive Cardiol 2003;15:242–246.