Predictors of Distal Embolization in Peripheral Percutaneous Interventions: A Report from a Large Peripheral Vascular Registry

ABSTRACT: Background. Distal embolization (DE) commonly occurs during peripheral percutaneous interventions (PPI) of the lower-extremity arterial vessels. In this study we evaluate the predictors of DE in a large cohort of patients undergoing PPI at our center. Methods. Patients who experienced clinically significant DE (requiring further mechanical or pharmacologic therapy as per operator judgment) were extracted from a peripheral vascular registry that prospectively tracks demographics, clinical, procedural and outcome variables on patients undergoing PPI at our medical center and compared these to patients in the same registry who did not experience DE. Univariate analysis was utilized to compare patients with and without DE. Logistic regression analysis was performed to determine the independent predictors of DE. Results. Of 577 patients, 14 (2.4%) experienced clinically significant DE. By univariate analysis, patients who experienced DE had longer lesion length (130.0 ± 123.35 mm vs. 90.05 ± 104.94 mm; p = 0.049), more severe angiographic pretreatment lesion stenosis (91.71% ± 14.76% vs. 85.65% ± 14.26%; p = 0.027), reduced pretreatment TIMI flow (1.21 ± 1.34 vs. 2.15 ± 1.1; p = 0.001), a higher rate of prior amputations (21.4% vs. 5.9%; p = 0.052), a higher prevalence of TASC-D lesions (59.3% vs. 29.6%; p = 0.002), more angiographic thrombus (35.7% vs. 6.4%; p = 0.001), and less frequency of chronic onset of symptoms on presentation (64.3% vs. 90.6%; p = 0.009). Logistic regression analysis showed that a prior history of amputation (odds ratio [OR] 3.56, 95% confidence interval [CI] 0.87–14.47; p = 0.08), presence of thrombus (OR 5.02, 95% CI 1.53–16.42; p = 0.008) and TASC-D lesions (OR 4.31, 95% CI 1.24–15.03; p = 0.022) were independent predictors of DE. Conclusion. Clinically significant DE requiring further mechanical or pharmacologic therapy occurs in approximately 2.4% of patients undergoing PPI. Patients with TASC-D lesions, angiographic thrombus and prior history of amputation are at high risk of DE. J INVASIVE CARDIOL 2009;21:628–631 Key words: angioplasty, peripheral arterial disease, embolic debris, complications, registry, predictors Distal embolization (DE) is described in the majority of patients undergoing percutaneous arterial intervention, irrespective of the vascular bed treated.1–7 In the lower extremities, DE occurs very frequently ranging from 50–100% of the time as detected by embolic filter protection or by Doppler studies.8 Clinically significant DE, defined as requiring further mechanical and/or pharmacological treatment, ranged in various studies from 0–11%8 and is likely to be significantly higher in patients with thrombotic lesions.9,10 The long-term outcome of untreated DE remains unknown, but it is likely that DE leading to no or slow flow in a major or a single tibial runoff will be treated when identified by the operator. Amputation has been reported with DE and was highest in patients with acute or subacute thrombotic lesions.9,10 Correlating the size of the debris with clinical or angiographic outcome is also a difficult task, although debris ≥ 2 mm is likely to obstruct a distal tibial vessel.7 Furthermore, very small debris that escapes the embolic filter (Methods This is a retrospective analysis of prospectively collected data from the cardiac catheterization laboratory at our medical center. Demographics, clinical and procedural variables (listed in Tables 1 and 2), as well as in-hospital outcomes including DE were collected on 577 patients (1,183 vessels) who underwent PPI from September 2004 to June 2007. All data were collected using electronic case-report forms and web-based data acquisition. All in-hospital adverse events were independently adjudicated by an experienced endovascular specialist. The study was approved by the institutional review board at our center. Procedures were coded as acute if a patient presented within 24 hours of symptom onset; subacute if presented within 1 month of symptom onset; and chronic if symptoms had been ongoing for > 1 month. An urgent procedure was performed within the same hospital stay; emergent within hours of presentation, and elective when the procedure was scheduled as outpatient. All lesions were classified according to the TransAtlantic InterSociety Consensus classification (original TASC-2000). Calcification in a vessel was graded subjectively by the interventionist as none-to-mild to moderate-to-severe. Patients were also classified on presentation based on the Rutherford-Baker classification (excluding acute limb ischemia) as limb ischemia (Classes IV–VI) or claudicants (Classes I–III). Although patients with thrombotic lesions were treated per operator discretion, the combination of lytic therapy and mechanical thrombectomy has been a favorable approach in our practice. Descriptive analysis was conducted on all variables using mean ± standard deviation (SD) for continuous variables and percentages for dichotomous variables. Univariate analysis was performed to compare patients who experienced clinically significant DE versus those who did not. Logistic regression (LR) analysis using backward elimination was used to determine the independent predictors of DE. Variables included in the LR model were age, amputation, lesion length, prestenosis severity, TASC type (ABC vs. D), thrombus, pre-thrombolysis in myocardial infarction (TIMI) flow, total occlusions (acute within 24 hours versus others). The term TIMI flow in peripheral arteries is used in this manuscript to indicate the following: normal/brisk (TIMI 3); slow, but fills the vessel completely (TIMI 2); and no flow (TIMI 0 and 1) analogous to the use of the term in describing coronary flow. Furthermore, in this cohort of patients, intervention was not performed for stump salvage, therefore in patients with a history of amputation, the described DE was in the treated nonamputated leg. Results Of 577 patients, 14 (2.4%) experienced clinically significant DE defined as requiring further pharmacologic or mechanical treatment according to operator judgment. By univariate analysis (Tables 1 and 2), patients who experienced DE had longer lesion length (130.0 ± 123.3 mm vs. 90.1 ± 104.9 mm; p = 0.05), more severe pretreatment lesion stenosis (91.7% ± 14.8% vs. 85.7% ± 14.3%; p = 0.027), reduced pretreatment TIMI flow (1.2 ± 1.3 vs. 2.2 ± 1.2; p = 0.001), a higher rate of prior amputations (21.4% vs. 5.9%; p = 0.052), a higher prevalence of TASC-D lesions (59.3% vs. 29.6%; p = 0.002), more angiographic thrombus (35.7% vs. 6.4%; p = 0.001), and less frequency of chronic onset of symptoms on presentation (64.3% vs. 90.6%; p = 0.009). Logistic regression analysis (Table 3) showed that a prior history of amputation (OR 3.56, 95% CI 0.87–14.47; p = 0.08), presence of thrombus (OR 5.02, 95% CI 1.53–16.4; p = 0.008) and TASC-D lesions (OR 4.31, 95% CI 1.24–15.03; p = 0.022) were independent predictors of DE. Discussion In this study, patients with a high risk of DE are those with a prior history of amputation, TASC-D lesions and the presence of angiographic thrombus. In this study, thrombotic lesions had 5.9 times greater odds of embolization than nonthrombotic ones. Also, patients presenting with acute thrombotic occlusions (within 24 hours of symptom onset) showed a trend toward more DE than those presenting subacutely/chronically (Table 2). These data are consistent with previously published studies. Kasirajan et al9 reported a 2.3–9.8% embolization rate in patients with acute limb ischemia treated with the AngioJet thrombectomy (Medrad Interventional, Warrendale, Pennsylvania). Also, Wholey et al10 noted a 3.8% embolization rate in 235 patients (30% acute, 5% subacute and 59% chronic [> 3 months]) treated with thrombolysis alone in the lower extremity; two of which required amputation. In addition, in the DETHROMBOSIS registry, Shammas et al11 identified a 17% embolization rate in thrombus-containing vessels as seen on intravascular ultrasound when treated with combining power-pulse spray thrombolysis with mechanical thrombectomy. Finally, Siablis et al4 captured in an embolic filter protection device fresh thrombus, calcium, cholesterol and fibrin in 100% of patients with acute (treated with the AngioJet) or subacute (treated with angioplasty/stenting) thrombosis. In addition to thrombotic lesions, embolic filter protection was also used in the PROTECT registry7 in 40 patients (56 lesions) with long occlusions, moderate or high calcified vessels and irregular lesions. Macroembolization occurred in 55% of these patients and was highest in those treated with the SilverHawk device. Using continuous Doppler monitoring, Lam et al12 also reported DE after angioplasty, stenting, SilverHawk atherectomy and excimer laser therapy in the SFA, with the highest emboli signals seen with the laser. Debris size remains unknown, however, in this study. One patient underwent an amputation in their series. Our data does not address device-specific risk for DE, but TASC-D lesions, typically long and complex occlusions, are likely to embolize 3.7 times more often than TASC A–C lesions, therefore these long chronic occlusions are best treated under embolic filter protection. Recent data from Kaid et al13 noted that embolized macrodebris with SilverHawk atherectomy consisted predominantly of collagen with fibrosis, cholesterol and macrophages. Although calcium in a vessel has been presumed to be a risk for macroembolization, our data do not support this hypothesis. This is consistent with the analysis of the captured embolized debris by Kaid and colleagues that did not contain calcium. Of interest is the finding in our analysis that a prior history of amputation increases the odds of significant embolization by 4.1 times. In our experience, patients with a history of amputations tend to have several comorbidities and severe diffuse peripheral arterial disease. It is likely that the extent of diffuse disease in these patients increase the risk of DE. Conclusion In conclusion, DE occurs frequently in PPI and thrombotic lesions, long complex occlusions and patients with a prior history of amputation are associated with clinically significant DE requiring further therapy. DE does occur at a high frequency ranging from 20–100% when debris are captured in filters or detected by Doppler probe.8 Also, large debris > 2 mm do occur at a high frequency, ranging from 20% to > 90% depending on lesions and devices used.7,13 However, it appears that only a small fraction (2.4% in our study) of these debris require further treatment according to the operator’s judgment. There are no data at present to determine when DE needs to be treated and therefore operator judgment can vary substantially. Also, the long-term outcome of lower-extremity debris, treated or untreated, irrespective of size, is still unknown. It is clear, however, that macrodebris that result in slow or no flow in the distal lower extremity, particularly in patients with already compromised distal runoffs, will require further therapy. Patients with DE in our series did well following further treatment and none had an amputation. Other series have shown that amputation can result from DE8–10 and is highest in patients with acute or subacute thrombotic lesions.9–10 In our series, the overall procedural success was high and was not different than for those patients who did not experience DE. However, in a subset analysis of this registry that has been previously published,14 DE was shown to lead to a prolonged procedure time, more contrast use and more fluoroscopy and radiation exposure, all important to both operators and patients. Further research is needed to understand the short- and long-term impact of DE on patient outcomes and the role of EFP (embolic filter protection) in high-risk patients. Study limitations. This is a large retrospective analysis of a prospective registry, rather than a prospective data collection evaluating specifically the primary endpoint. Clinically significant DE does not necessarily reflect on the true embolization rate that has occurred in these patients. Since the treatment of DE was also left to the operator’s judgment, this could have led to an underestimation of significant DE. The main issue is the lack of consensus for the definition of “clinically significant” DE and when, how and in whom this should be treated. Operators at present might differ in their approach in defining and treating DE, and our definition is at best a reflection of our own bias in evaluating and treating DE. DE in our study is also a clinical definition and not based on actual debris captured in a filter or identified by a Doppler study. It is known that DE occurs at a much higher frequency than what we have seen as “clinically important.” The long-term outcomes of DE, and particularly the consequences of which tibials they affect (DP or AT), are unknown. Our prospective registry has not identified specifically which tibial vessel has been affected by DE. Furthermore, device-specific data have not been collected, and therefore this study can only comment on clinical, demographic and angiographic predictors of DE and not device-related DE. Specific devices (particularly SilverHawk atherectomy) can play an important role in DE, and how this would impact our findings is unknown. Finally, patients were stratified in the acute or subacute time frame, but it should be noted that many patients with acute embolic or thrombotic events might have presented outside this time frame. Thrombus is quite often underestimated angiographically and unless IVUS is performed, a thrombus is likely to be missed. Angiograms, however, are the standard in evaluating these patients and routine IVUS is not commonly performed in peripheral interventions. From the Midwest Cardiovascular Research Foundation, Davenport, Iowa. Presented in part as an abstract at Transcatheter Cardiovascular Therapeutics 2009, San Francisco, September 21–25. The authors report no conflicts of interest regarding the content herein. This study is supported by the Nicolas and Gail Shammas Research Fund at the Midwest Cardiovascular Research Foundation, Davenport, Iowa. Manuscript submitted May 20, 2009, provisional acceptance given June 23, 2009, final version accepted September 9, 2009. Address all correspondence to: Nicolas W. Shammas, MD, MS, Director, Midwest Cardiovascular Research Foundation, Cardiovascular Medicine, PC, 1236 E. Rusholme, Suite 300, Davenport, IA 52803. E-mail: shammas@mchsi.com

1. Casserly IP, Abou-Chebl A, Fathi RB, et al. Slow-flow phenomenon during carotid artery intervention with embolic protection devices: Predictors and clinical outcome. J Am Coll Cardiol 2005;46:1466–1472.

2. Edwards MS, Corriere MA, Craven TE, et al. Atheroembolism during percutaneous renal artery revascularization. J Vasc Surg 2007;46:55–61.

3. van Gaal WJ, Choudhury RP, Porto I, et al. Prediction of distal embolization during percutaneous coronary intervention in saphenous vein grafts. Am J Cardiol 2007;99:603–606.

4. Siablis D, Karnabatidis D, Katsanos K, et al. Outflow protection filters during percutaneous recanalization of lower extremities’ arterial occlusions: A pilot study. Eur J Radiol 2005;55:243–249.

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

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

7. 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.

8. Shammas NW. Optimal strategy in lower extremity peripheral arterial percutaneous interventions: An interventionalist’s perspective. Vascular Disease Management 2009;6:36–40

9. Kasirajan K, Haskal ZJ, Ouriel K. The use of mechanical thrombectomy devices in the management of acute peripheral arterial occlusive disease. J Vasc Interv Radiol 2001;12:405–411.

10. Wholey MH, Maynar MA, Wholey MH, et al. Comparison of thrombolytic therapy of lower-extremity acute, subacute, and chronic arterial occlusions. Cathet Cardiovasc Diagn 1998;44:159–169.

11. Shammas NW, Dippel EJ, Shammas G, et al. Dethrombosis of the lower extremity arteries using the power-pulse spray technique in patients with recent onset thrombotic occlusions: Results of the DETHROMBOSIS Registry. J Endovasc Ther 2008;15:570–579.

12. 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.

13. Kaid KA, Gopinathapillai R, Qian F, et al. Analysis of particulate debris after superficial femoral artery atherectomy. J Invasive Cardiol 2009;21:7–10

14. Shammas NW, Shammas, G, Dippel E, Jerin M. Intraprocedural outcomes following distal lower extremity embolization in patients undergoing peripheral percutaneous interventions. Vascular Disease Management 2009;6:58–61.